mirror of
https://github.com/storytold/wordcraft.git
synced 2026-11-07 10:49:47 -06:00
Compare commits
5
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
 Dylan M. Taylor
|
2e168e0bc4 | ||
|
Dylan M. Taylor
|
05a3ac74b0 | ||
|
Dylan Taylor
|
f3c3c6480a | ||
|
Dylan M. TaylorandClaude Opus 5.5
|
213e433563 | ||
|
Dylan M. Taylor
|
32afcb661a |
@@ -22,7 +22,7 @@ jobs:
|
||||
uses: storytold/craft-common-actions/.github/workflows/gate.yml@main
|
||||
with:
|
||||
# Same craft-fonts commit as release.yml.
|
||||
craft-fonts-ref: b076417b5880bd4154bd22f00695b1d28c6bea83
|
||||
craft-fonts-ref: 8dcdacd5153e64560d109541a47d806f26f048c0
|
||||
packages: >-
|
||||
pkg-config libxkbcommon-dev libwayland-dev libx11-dev libxcursor-dev libxrandr-dev libxi-dev
|
||||
libgtk-3-dev libfontconfig1-dev libfreetype-dev mesa-vulkan-drivers
|
||||
|
||||
@@ -28,4 +28,4 @@ jobs:
|
||||
# Intentionally tracked at @main (not SHA-pinned) so org-wide CI improvements, security fixes, and build optimizations propagate without PR churn across all craftapp repos.
|
||||
uses: storytold/craft-common-actions/.github/workflows/freebsd.yml@main
|
||||
with:
|
||||
craft-fonts-ref: b076417b5880bd4154bd22f00695b1d28c6bea83
|
||||
craft-fonts-ref: 8dcdacd5153e64560d109541a47d806f26f048c0
|
||||
|
||||
@@ -55,7 +55,7 @@ jobs:
|
||||
components: rustfmt, clippy
|
||||
targets: wasm32-unknown-unknown
|
||||
|
||||
- uses: Swatinem/rust-cache@d1527c8f67ef9af357290475d4af4b0fab8f12e0 # v2.9.3
|
||||
- uses: Swatinem/rust-cache@6323deb102c322ba6fcbdcafc7e3dddab59af2b6 # v2.9.2
|
||||
with:
|
||||
save-if: ${{ github.ref == 'refs/heads/main' }}
|
||||
|
||||
|
||||
@@ -24,8 +24,7 @@ jobs:
|
||||
with:
|
||||
app-name: wordcraft
|
||||
app-title: WordCraft
|
||||
craft-fonts-ref: b076417b5880bd4154bd22f00695b1d28c6bea83
|
||||
web-craft-fonts: true
|
||||
craft-fonts-ref: 8dcdacd5153e64560d109541a47d806f26f048c0
|
||||
version: ${{ inputs.version }}
|
||||
secrets:
|
||||
APPLE_CERTIFICATE: ${{ secrets.APPLE_CERTIFICATE }}
|
||||
|
||||
@@ -8,8 +8,6 @@ Generated-in-code art is original and has no file to list: the UI icon set (`cra
|
||||
|
||||
| Asset | Author | Source | Licence | Notes |
|
||||
|---|---|---|---|---|
|
||||
| `egui` dependency: `epaint_default_fonts/fonts/egui-icons.ttf` | John Slegers (emoji-icon-font) | https://github.com/storytold/egui/blob/173cc7375d35dc5fbc5b5d0994fdf787f547c5d1/crates/epaint_default_fonts/fonts/egui-icons.txt; original https://github.com/jslegers/emoji-icon-font | MIT (full copyright and permission text in NOTICE) | Small egui special-icon subset; embedded by default_fonts, retained in FontDefinitions::default() |
|
||||
| `egui` dependency: `epaint_default_fonts/fonts/emoji-icon-font.ttf` | John Slegers | https://github.com/storytold/egui/blob/173cc7375d35dc5fbc5b5d0994fdf787f547c5d1/crates/epaint_default_fonts/fonts/emoji-icon-font-mit-license.txt; original https://github.com/jslegers/emoji-icon-font | MIT (full copyright and permission text in NOTICE) | Parent monochrome icon/emoji face embedded by monochrome_emoji_fonts; separate from the egui-icons subset |
|
||||
| `assets/hyphenation/en-us.pat` | DesignCraft contributors (generated) | trained from the Moby list by DesignCraft's patgen | Public Domain (CC0-1.0) | Our own Liang hyphenation patterns (no TeX patterns) |
|
||||
| `assets/hyphenation/en-us.dic` | Grady Ward (Moby Hyphenator II word list); converted by DesignCraft contributors | https://www.gutenberg.org/ebooks/3204 (file `mhyph.txt`, SHA-256 `eeb30474c86b8af3469035ec1a0913e35905325ca885290db9dfed9881e230ac`), via DesignCraft | Public Domain | ~165k words with break points; drives spelling and hyphenation |
|
||||
| `assets/spelling/en-extra.txt` | WordCraft contributors | original word list written for this repository (common modern words the Moby list lacks); no other dictionary copied | MIT OR Apache-2.0 | ~240 lowercase English base forms supplementing spelling (#520), including British spellings (#40) |
|
||||
|
||||
Generated
+394
-1975
File diff suppressed because it is too large
Load Diff
+10
-35
@@ -4,7 +4,7 @@ members = ["crates/*", "apps/*", "xtask"]
|
||||
default-members = ["crates/*", "apps/wordcraft", "apps/wordcraft-cli", "xtask"]
|
||||
|
||||
[workspace.package]
|
||||
version = "0.11.0"
|
||||
version = "0.10.0"
|
||||
edition = "2024"
|
||||
license = "MIT OR Apache-2.0"
|
||||
rust-version = "1.95"
|
||||
@@ -52,18 +52,13 @@ harfrust = "0.13"
|
||||
# WOFF / WOFF2 web fonts added by the user (#582): pure Rust (MIT); its brotli and zlib decoders
|
||||
# (brotli-decompressor, flate2) are already in the tree.
|
||||
wuff = "0.2.9"
|
||||
# WOFF2's brotli stream, decoded ahead of wuff so its glyph point counts are checked first (#3703).
|
||||
# The same crate and version wuff already decodes it with (no new code in the build): pure Rust,
|
||||
# `forbid(unsafe_code)` in its default build, no_std-capable (builds for wasm), BSD-3-Clause/MIT.
|
||||
brotli-decompressor = "5"
|
||||
vello_cpu = "0.2"
|
||||
image = { version = "0.25", default-features = false, features = ["png", "jpeg", "webp", "gif", "bmp", "tiff"] }
|
||||
craft-pdf = { git = "https://github.com/storytold/craft-libs", rev = "b926b5c03e256c86ca10934f58d03f628810b5b7", version = "=0.1.0", default-features = false, features = ["write"] }
|
||||
craft-update = { git = "https://github.com/storytold/craft-libs", version = "=0.1.0", rev = "b926b5c03e256c86ca10934f58d03f628810b5b7", features = ["egui"] }
|
||||
craft-pdf = { git = "https://github.com/storytold/craft-libs", rev = "c3487320b2d1de657b860b3c8abb44e1f2b32bd8", version = "=0.1.0", default-features = false, features = ["write"] }
|
||||
quick-xml = "0.41"
|
||||
zip = { version = "3", default-features = false, features = ["deflate"] }
|
||||
# zlib streams (compressed metafile pictures in `.doc` drawings); already in the lock as a dependency of other crates.
|
||||
miniz_oxide = "0.9"
|
||||
miniz_oxide = "0.8"
|
||||
cfb = "0.15"
|
||||
# Legacy text encodings (GBK/GB18030, Big5, Shift_JIS, EUC-KR, single-byte code pages) for plain
|
||||
# text and CSV recipient lists (#778): encoding_rs ((Apache-2.0 OR MIT) AND BSD-3-Clause) and its
|
||||
@@ -75,22 +70,17 @@ aes = "0.9"
|
||||
sha1 = "0.11"
|
||||
sha2 = "0.11"
|
||||
hmac = "0.13"
|
||||
base64 = "0.23"
|
||||
base64 = "0.22"
|
||||
getrandom = "0.3"
|
||||
zeroize = "1"
|
||||
# Unix open flags (O_NONBLOCK for the engine's checked file reads, #3666); already in the lock as a
|
||||
# dependency of other crates.
|
||||
libc = "0.2"
|
||||
|
||||
# ui
|
||||
egui = "0.37"
|
||||
egui_extras = { version = "0.37", default-features = false }
|
||||
eframe = { version = "0.37", default-features = false }
|
||||
egui_kittest = "0.37"
|
||||
# A monotonic clock that works on native and in the browser (`performance.now()`).
|
||||
web-time = "1.1"
|
||||
egui = "0.36"
|
||||
egui_extras = { version = "0.36", default-features = false }
|
||||
eframe = { version = "0.36", default-features = false }
|
||||
egui_kittest = "0.36"
|
||||
rfd = "0.17"
|
||||
arboard = { version = "3.7", default-features = false, features = ["wayland-data-control"] }
|
||||
arboard = { version = "3.6", default-features = false, features = ["wayland-data-control"] }
|
||||
|
||||
[workspace.lints.rust]
|
||||
unsafe_code = "forbid"
|
||||
@@ -120,19 +110,4 @@ debug = 0
|
||||
|
||||
# Required by craft-pdf: the backend rejects writers that failed during unwinding.
|
||||
[patch.crates-io]
|
||||
krilla = { git = "https://github.com/storytold/craft-libs", rev = "b926b5c03e256c86ca10934f58d03f628810b5b7" }
|
||||
ecolor = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
emath = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
epaint = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
epaint_default_fonts = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui-winit = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui-wgpu = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui_glow = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui_extras = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui_kittest = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui_system_font_provider = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
eframe = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui_demo_lib = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
egui_inspection = { git = "https://github.com/storytold/egui.git", rev = "6d518fc799681571963842f95b55a6b68431b15a" }
|
||||
winit = { git = "https://github.com/storytold/winit.git", rev = "a517b3846ee8457d84ee4975110bfda6797c44dc" }
|
||||
krilla = { git = "https://github.com/storytold/craft-libs", rev = "c3487320b2d1de657b860b3c8abb44e1f2b32bd8" }
|
||||
|
||||
@@ -95,32 +95,14 @@ list (author, source, license) is in ATTRIBUTION.md.
|
||||
FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
|
||||
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
|
||||
DEALINGS IN THE SOFTWARE.
|
||||
- Application-drawn UI icons (crates/ui-egui/src/icons.rs) are original work, MIT OR Apache-2.0;
|
||||
the bundled egui font icons have the separate notice below.
|
||||
- All UI icons are drawn in code (crates/ui-egui/src/icons.rs) and are original work, MIT OR Apache-2.0.
|
||||
|
||||
Rust dependencies (egui, eframe, vello_cpu, skrifa, harfrust, krilla, quick-xml, zip, image, …)
|
||||
are permissively licensed (MIT/Apache-2.0/BSD/ISC/Zlib) and fetched through crates.io or pinned Git sources.
|
||||
The craft-pdf writer and Krilla 0.8.2 backend are pinned to storytold/craft-libs revision
|
||||
b926b5c03e256c86ca10934f58d03f628810b5b7. The backend retains its MIT OR Apache-2.0 licensing
|
||||
c3487320b2d1de657b860b3c8abb44e1f2b32bd8. The backend retains its MIT OR Apache-2.0 licensing
|
||||
and upstream bundled material; see that revision's licenses/pdf-writer/ and ATTRIBUTION.md.
|
||||
The backend patch prevents finalization during an unwind and rejects a writer after failed
|
||||
cleanup. WordCraft enables only writing; PDF import and its image decoders are not enabled.
|
||||
|
||||
No Microsoft, Adobe, Avid or Autodesk icons, images, templates or artwork are included.
|
||||
|
||||
Bundled egui font icons
|
||||
|
||||
The canonical egui0.37 dependency includes egui-icons.ttf (a small subset of emoji-icon-font)
|
||||
through default_fonts, and emoji-icon-font.ttf through monochrome_emoji_fonts. Both are embedded
|
||||
by the UI font definitions. Their upstream project is https://github.com/jslegers/emoji-icon-font;
|
||||
the exact copyright and permission notice from the pinned egui dependency follows.
|
||||
|
||||
MIT License
|
||||
|
||||
Copyright (c) 2014 John Slegers
|
||||
|
||||
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
|
||||
|
||||
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
|
||||
|
||||
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
||||
|
||||
@@ -221,16 +221,13 @@ closer pagination fidelity, a stability sweep and native printing. Details and e
|
||||
|
||||
### Windows
|
||||
|
||||
| Build | Installer (all users) | Installer (just me) | Portable |
|
||||
|---|---|---|---|
|
||||
| x64 (64-bit Intel/AMD) | `wordcraft-<ver>-windows-x64.msi` | `wordcraft-<ver>-windows-user-x64.msi` | `wordcraft-<ver>-windows-x64-portable.zip` |
|
||||
| arm64 (Snapdragon and other ARM PCs) | `wordcraft-<ver>-windows-arm64.msi` | `wordcraft-<ver>-windows-user-arm64.msi` | `wordcraft-<ver>-windows-arm64-portable.zip` |
|
||||
| x86 (32-bit) | `wordcraft-<ver>-windows-x86.msi` | `wordcraft-<ver>-windows-user-x86.msi` | `wordcraft-<ver>-windows-x86-portable.zip` |
|
||||
| Build | Installer | Portable |
|
||||
|---|---|---|
|
||||
| x64 (64-bit Intel/AMD) | `wordcraft-<ver>-windows-x64.msi` | `wordcraft-<ver>-windows-x64-portable.zip` |
|
||||
| arm64 (Snapdragon and other ARM PCs) | `wordcraft-<ver>-windows-arm64.msi` | `wordcraft-<ver>-windows-arm64-portable.zip` |
|
||||
| x86 (32-bit) | `wordcraft-<ver>-windows-x86.msi` | `wordcraft-<ver>-windows-x86-portable.zip` |
|
||||
|
||||
The all-users installer puts WordCraft in Program Files and needs administrator rights. The
|
||||
just-me installer needs none: it installs into `%LOCALAPPDATA%\Programs\WordCraft` for your
|
||||
account only. Choose one: installing both gives two WordCraft entries in Settings › Apps and two
|
||||
in the Start menu. Installers and executables are code-signed.
|
||||
Installers and executables are code-signed.
|
||||
|
||||
**If the app doesn't open on Windows:** the desktop app initializes only DirectX 12 by default.
|
||||
Letting wgpu also create an OpenGL instance can crash some graphics drivers (AMD's
|
||||
|
||||
+2
-77
File diff suppressed because one or more lines are too long
@@ -36,11 +36,7 @@ USAGE:
|
||||
{\"file\"?, \"template\"?, \"commands\": [{\"command\": id,
|
||||
\"params\"?: {…}} | id, …], \"save\"?: [OUT, …]}; save
|
||||
formats follow the extension (docx and pdf in one run);
|
||||
relative paths resolve against the script's folder.
|
||||
file.new and file.open open another document and
|
||||
make it active; later commands and the saves work on
|
||||
the active one; view.switchWindows lists the open
|
||||
documents and with {\"id\": N} makes one active
|
||||
relative paths resolve against the script's folder
|
||||
wordcraft-cli commands [--json] list every command
|
||||
wordcraft-cli parity [--markdown] feature-catalog parity
|
||||
wordcraft-cli mcp [--connect HOST:PORT | --author NAME] [--automation-read-root DIR] [--automation-write-root DIR]
|
||||
@@ -383,16 +379,6 @@ fn script_operand<S: AsRef<OsStr>>(raw: &[S]) -> Option<PathBuf> {
|
||||
raw.get(at + 1).map(|a| PathBuf::from(a.as_ref()))
|
||||
}
|
||||
|
||||
/// Whether `addr` (`HOST:PORT`, as `--connect` gives it) is at least shaped like an address: a
|
||||
/// host of some kind and a numeric port. Malformed addresses fail at startup — a lazy remote
|
||||
/// would otherwise retry one on every call and never say why nothing connects.
|
||||
fn valid_connect_addr(addr: &str) -> bool {
|
||||
let Some((host, port)) = addr.rsplit_once(':') else { return false };
|
||||
!host.is_empty() && port.parse::<u16>().is_ok()
|
||||
}
|
||||
|
||||
/// Run the subcommand named by `args` (the command line after the program name). Files named
|
||||
/// there are opened by their real names, which may not be valid Unicode.
|
||||
fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
|
||||
let script_path = script_operand(args);
|
||||
let (args, file) = text_args(args)?;
|
||||
@@ -488,14 +474,10 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
|
||||
let mut s = match (from_file, from_template) {
|
||||
(Some(f), _) => open(&f, password.as_deref())?,
|
||||
// Both go through `file.new`, which puts a blank document on the region's paper (#3152).
|
||||
// `file.new` opens beside the session's first document, which is then closed: the
|
||||
// run starts with one open document.
|
||||
(None, t) => {
|
||||
let mut s = session(wordcraft_doc::Document::new());
|
||||
let first = s.document_id();
|
||||
let params = t.map_or_else(|| json!({}), |t| json!({"template": t}));
|
||||
s.run("file.new", ¶ms).map_err(|e| e.to_string())?;
|
||||
s.close_document(first).map_err(|e| e.to_string())?;
|
||||
s
|
||||
}
|
||||
};
|
||||
@@ -600,27 +582,11 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
|
||||
return Err("mcp: --author applies to the headless server; with --connect, run file.setAuthor".into());
|
||||
}
|
||||
Some(addr) => {
|
||||
// A malformed address must fail at startup, not on every tool call the lazy
|
||||
// remote would retry it with.
|
||||
if !valid_connect_addr(&addr) {
|
||||
return Err(format!("mcp: --connect needs HOST:PORT with a numeric port, got `{addr}`").into());
|
||||
}
|
||||
match wordcraft_mcp::Remote::connect(&addr) {
|
||||
Ok(remote) => {
|
||||
eprintln!("wordcraft-cli mcp: connected to WordCraft at {addr}");
|
||||
if let Some(w) = remote.key_warning() {
|
||||
eprintln!("wordcraft-cli: {w}");
|
||||
}
|
||||
Box::new(remote)
|
||||
}
|
||||
// The app may start after the client does: serve anyway and connect on the
|
||||
// first call, or the client marks the server failed for its whole session
|
||||
// (#4078).
|
||||
Err(e) => {
|
||||
eprintln!("wordcraft-cli mcp: WordCraft is not reachable at {addr} yet ({e}); will retry on each call");
|
||||
Box::new(wordcraft_mcp::Remote::lazy(&addr))
|
||||
}
|
||||
let remote = wordcraft_mcp::Remote::connect(&addr).map_err(|e| format!("can't reach the WordCraft app at {addr}: {e}"))?;
|
||||
if let Some(w) = remote.key_warning() {
|
||||
eprintln!("wordcraft-cli: {w}");
|
||||
}
|
||||
Box::new(remote)
|
||||
}
|
||||
None => {
|
||||
let mut headless = wordcraft_mcp::Headless::default();
|
||||
@@ -667,14 +633,14 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
|
||||
let out = wordcraft_zotero::client::run_command(&opts, command, &mut |call| {
|
||||
let r = bridge.handle(&mut s, &mut host, call);
|
||||
if trace {
|
||||
eprintln!("← {} {}", escape_peer(&call.method), escape_peer(&clip(&Value::Array(call.args.clone()).to_string())));
|
||||
eprintln!("→ {}", escape_peer(&clip(&String::from_utf8_lossy(&wordcraft_zotero::wire::encode_reply(&r)))));
|
||||
eprintln!("← {} {}", call.method, clip(&Value::Array(call.args.clone()).to_string()));
|
||||
eprintln!("→ {}", clip(&String::from_utf8_lossy(&wordcraft_zotero::wire::encode_reply(&r))));
|
||||
}
|
||||
r
|
||||
})
|
||||
.map_err(|e| e.to_string())?;
|
||||
for a in &host.alerts {
|
||||
eprintln!("Zotero: {}", escape_peer(a));
|
||||
eprintln!("Zotero: {a}");
|
||||
}
|
||||
eprintln!("{} calls, {}", out.calls, if out.completed { "completed" } else { "ended without completing (cancelled?)" });
|
||||
if let Some(o) = arg_value(&rest, "--save") {
|
||||
@@ -697,27 +663,6 @@ fn clip(s: &str) -> String {
|
||||
}
|
||||
}
|
||||
|
||||
/// Text from another program (Zotero, or whatever answered on its port) for one diagnostic
|
||||
/// line: backslashes, control characters (newline, escape, C1…), line separators and bidi
|
||||
/// overrides are shown as escapes (`\\`, `\n`, `\u{1b}`), so it can't end its line, drive the
|
||||
/// terminal or pass for other output.
|
||||
fn escape_peer(s: &str) -> String {
|
||||
let mut out = String::with_capacity(s.len());
|
||||
for c in s.chars() {
|
||||
match c {
|
||||
'\\' => out.push_str("\\\\"),
|
||||
'\n' => out.push_str("\\n"),
|
||||
'\r' => out.push_str("\\r"),
|
||||
'\t' => out.push_str("\\t"),
|
||||
c if c.is_control() || matches!(c, '\u{061c}' | '\u{200e}' | '\u{200f}' | '\u{2028}'..='\u{202e}' | '\u{2066}'..='\u{2069}') => {
|
||||
out.extend(c.escape_unicode())
|
||||
}
|
||||
c => out.push(c),
|
||||
}
|
||||
}
|
||||
out
|
||||
}
|
||||
|
||||
fn parity_markdown(p: &Value) -> String {
|
||||
let mut s = String::from(
|
||||
"# WordCraft feature parity\n\nGenerated by `cargo xtask parity` (`wordcraft-cli parity --markdown`): the word-processor feature catalog (`crates/engine/src/catalog.rs`) compared with the live command registry.\n\n",
|
||||
@@ -764,15 +709,6 @@ mod tests {
|
||||
values.iter().map(|value| (*value).into()).collect()
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn peer_text_stays_on_one_visible_line() {
|
||||
let alert = "Done.\nwrote /etc/passwd\r\u{1b}[2J\u{9b}31m\u{202e}txt.exe\\n\tok é";
|
||||
let shown = escape_peer(alert);
|
||||
assert_eq!(shown, "Done.\\nwrote /etc/passwd\\r\\u{1b}[2J\\u{9b}31m\\u{202e}txt.exe\\\\n\\tok é");
|
||||
assert!(!shown.chars().any(|c| c.is_control()));
|
||||
assert_eq!(escape_peer("Citation (Doe, 2020) — “ok”"), "Citation (Doe, 2020) — “ok”");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn validators_accept_only_the_documented_options() {
|
||||
assert!(validate_options("info", &args(&["file.docx"]), PASSWORD_OPTIONS).is_ok());
|
||||
@@ -871,9 +807,6 @@ mod tests {
|
||||
assert!(run(&args(&["run", "--author", "Claude", "--cmd", "review.trackChanges={\"value\":true}"])).is_ok());
|
||||
// The app's user name is the person's own; a bridge flag must not rewrite it.
|
||||
assert!(run(&args(&["mcp", "--connect", "127.0.0.1:9", "--author", "Claude"])).is_err_and(|e| e.to_string().contains("--author")));
|
||||
// A malformed address fails at startup, not on every retried call.
|
||||
assert!(run(&args(&["mcp", "--connect", "127.0.0.1:notaport"])).is_err_and(|e| e.to_string().contains("numeric port")));
|
||||
assert!(run(&args(&["mcp", "--connect", "notaport"])).is_err_and(|e| e.to_string().contains("numeric port")));
|
||||
// The flag goes through the same protection check as the command.
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
s.run("review.restrict", &json!({"mode": "readOnly"})).unwrap();
|
||||
|
||||
@@ -1,39 +0,0 @@
|
||||
//! `wordcraft-cli run` with documents opened along the way: `file.new` and `file.open` open
|
||||
//! another document and make it active, and `--save` saves the active one.
|
||||
|
||||
use std::path::Path;
|
||||
use std::process::{Command, Output};
|
||||
|
||||
fn run(args: &[&str], out: &Path) -> Output {
|
||||
let mut cli = Command::new(env!("CARGO_BIN_EXE_wordcraft-cli"));
|
||||
cli.arg("run").args(args).arg("--save").arg(out);
|
||||
cli.output().unwrap()
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn run_saves_the_document_opened_last() {
|
||||
let dir = std::env::temp_dir().join(format!("wordcraft-cli-documents-{}", std::process::id()));
|
||||
std::fs::create_dir_all(&dir).unwrap();
|
||||
let (src, out) = (dir.join("opened.txt"), dir.join("out.txt"));
|
||||
std::fs::write(&src, "Opened\n").unwrap();
|
||||
let open = format!(r#"file.open={{"path":{}}}"#, serde_json::json!(src.to_string_lossy()));
|
||||
|
||||
// A script that starts a new document and types into it saves that document.
|
||||
let made = run(&["--template", "sample", "--cmd", "file.new", "--cmd", r#"text.insert={"text":"Fresh"}"#], &out);
|
||||
let fresh = std::fs::read_to_string(&out).unwrap_or_default();
|
||||
// A file opened after typing is what is saved; the typed document stays out of it.
|
||||
let opened = run(&["--template", "blank", "--cmd", r#"text.insert={"text":"Typed"}"#, "--cmd", &open, "--cmd", "caret.docEnd"], &out);
|
||||
let reopened = std::fs::read_to_string(&out).unwrap_or_default();
|
||||
// A run starts with one open document.
|
||||
let mut cli = Command::new(env!("CARGO_BIN_EXE_wordcraft-cli"));
|
||||
let listed = cli.args(["run", "--template", "sample", "--print", "--cmd", "view.switchWindows"]).output().unwrap();
|
||||
let _ = std::fs::remove_dir_all(&dir);
|
||||
|
||||
assert!(made.status.success(), "stderr: {}", String::from_utf8_lossy(&made.stderr));
|
||||
assert_eq!(fresh.trim(), "Fresh");
|
||||
assert!(opened.status.success(), "stderr: {}", String::from_utf8_lossy(&opened.stderr));
|
||||
assert_eq!(reopened.trim(), "Opened");
|
||||
assert!(listed.status.success(), "stderr: {}", String::from_utf8_lossy(&listed.stderr));
|
||||
let stdout = String::from_utf8_lossy(&listed.stdout);
|
||||
assert_eq!(stdout.matches("\"id\"").count(), 1, "{stdout}");
|
||||
}
|
||||
@@ -29,4 +29,4 @@ rfd = { workspace = true }
|
||||
wasm-bindgen = "0.2"
|
||||
wasm-bindgen-futures = "0.4"
|
||||
js-sys = "0.3"
|
||||
web-sys = { version = "0.3", features = ["Window", "Document", "Element", "HtmlElement", "HtmlCanvasElement", "HtmlAnchorElement", "HtmlIFrameElement", "Blob", "BlobPropertyBag", "Url", "Location", "CssStyleDeclaration", "Event", "EventTarget", "BeforeUnloadEvent", "ClipboardEvent", "DataTransfer", "FileList", "File", "DomStringList", "IdbFactory", "IdbOpenDbRequest", "IdbRequest", "IdbDatabase", "IdbObjectStore", "IdbTransaction", "IdbTransactionMode", "MediaQueryList", "Storage", "MessageEvent", "UrlSearchParams"] }
|
||||
web-sys = { version = "0.3", features = ["Window", "Document", "Element", "HtmlElement", "HtmlCanvasElement", "HtmlAnchorElement", "HtmlIFrameElement", "Blob", "BlobPropertyBag", "Url", "Location", "CssStyleDeclaration", "Event", "EventTarget", "BeforeUnloadEvent", "ClipboardEvent", "DataTransfer", "FileList", "File", "DomStringList", "IdbFactory", "IdbOpenDbRequest", "IdbRequest", "IdbDatabase", "IdbObjectStore", "IdbTransaction", "IdbTransactionMode", "MediaQueryList", "Storage"] }
|
||||
|
||||
@@ -1,169 +0,0 @@
|
||||
//! The `postMessage` bridge to a same-origin parent page (`?host=parent`), for pages that embed
|
||||
//! WordCraft in an `<iframe>` and keep its files (a Nextcloud app, a personal file server). The
|
||||
//! same messages as GridCraft's and PdfCraft's bridges; each is a plain object with a `type`:
|
||||
//!
|
||||
//! - `wordcraft:ready` (to the parent): `{ version }`, sent once the app is running and
|
||||
//! listening; send a file now.
|
||||
//! - `wordcraft:open` (from the parent): `{ name, bytes }`, bytes an `ArrayBuffer` or
|
||||
//! `Uint8Array`. It opens like a document picked with File › Open or dropped on the window.
|
||||
//! - `wordcraft:save` (to the parent): `{ name, saveAs, bytes }` (a transferred `ArrayBuffer`) for
|
||||
//! every file WordCraft would otherwise download. The parent stores it and reports failures
|
||||
//! itself. `saveAs` says whether the file becomes the document's file:
|
||||
//! - Save: `false`. It writes the opened file under its own name and format, as the desktop
|
||||
//! app does. A name the parent never sent (an untitled document, an imported .doc saved as
|
||||
//! .docx) is a new file: `true`, as for Save As.
|
||||
//! - Save As, and File › Export to another document format (which runs Save As): `true` for a
|
||||
//! format that keeps everything (.docx and the other Word packages, .odt, .rtf). The document
|
||||
//! is now that file; the next Save sends its name with `false`.
|
||||
//! - Save As or Export to a format that doesn't (PDF, Markdown, plain text), Export as PDF and
|
||||
//! Export Page as PNG: `false`. They are copies; the document's file stays the same.
|
||||
//! - `wordcraft:dirty` (to the parent): `{ dirty }` whenever unsaved changes appear or go away.
|
||||
//! - `wordcraft:failed` (to the parent): `{ error }` when the app couldn't start.
|
||||
//!
|
||||
//! Only the parent window on the page's own origin is heard or answered. In host mode the parent
|
||||
//! guards leaving the page, and `?author=` names the user for comments and tracked changes.
|
||||
|
||||
use std::cell::Cell;
|
||||
use std::rc::Rc;
|
||||
|
||||
use wasm_bindgen::JsCast as _;
|
||||
use wasm_bindgen::JsValue;
|
||||
use wasm_bindgen::closure::Closure;
|
||||
use wordcraft_ui_egui::{Inbox, Services, WordApp};
|
||||
|
||||
/// The longest user name taken from `?author=`.
|
||||
const MAX_AUTHOR_CHARS: usize = 200;
|
||||
|
||||
#[derive(Clone)]
|
||||
pub struct Bridge {
|
||||
window: web_sys::Window,
|
||||
parent: web_sys::Window,
|
||||
origin: String,
|
||||
}
|
||||
|
||||
impl Bridge {
|
||||
/// `None` when the page isn't framed (there is no parent to talk to), or when the parent is
|
||||
/// on another origin: messages only go to this origin, so the browser would drop every save
|
||||
/// without an error while the document showed as saved.
|
||||
pub fn new() -> Option<Self> {
|
||||
let window = web_sys::window()?;
|
||||
let parent = window.parent().ok().flatten()?;
|
||||
if js_sys::Object::is(&parent, &window) {
|
||||
return None;
|
||||
}
|
||||
let origin = window.location().origin().ok()?;
|
||||
// An opaque origin (a `file://` page in some browsers) can't be a `postMessage` target.
|
||||
if origin == "null" {
|
||||
return None;
|
||||
}
|
||||
// Reading another origin's location throws, which is `None` too.
|
||||
if parent.location().origin().ok()? != origin {
|
||||
return None;
|
||||
}
|
||||
Some(Self { window, parent, origin })
|
||||
}
|
||||
|
||||
/// Hand saved and exported files to the parent instead of downloading them, report unsaved
|
||||
/// changes to it, and hear the files it sends.
|
||||
pub fn connect(&self, services: &mut Services, inbox: Inbox, ctx: egui::Context) {
|
||||
self.listen(inbox, ctx);
|
||||
let bridge = self.clone();
|
||||
services.host_save = Some(Box::new(move |name: &str, bytes: &[u8], save_as: bool| bridge.post_save(name, bytes, save_as)));
|
||||
// Anything else that would download goes to the parent too.
|
||||
let bridge = self.clone();
|
||||
services.download = Some(Box::new(move |name: &str, bytes: &[u8]| bridge.post_save(name, bytes, false)));
|
||||
// The browser's save picker is never used, so no save finishes later.
|
||||
services.async_saves = None;
|
||||
let bridge = self.clone();
|
||||
let reported: Rc<Cell<Option<bool>>> = Rc::default();
|
||||
services.on_dirty = Some(Box::new(move |dirty: bool| {
|
||||
if reported.get() != Some(dirty) && bridge.send(&[("type", "wordcraft:dirty".into()), ("dirty", dirty.into())]) {
|
||||
reported.set(Some(dirty));
|
||||
}
|
||||
}));
|
||||
}
|
||||
|
||||
/// `?author=`: the host's name for the user signs comments and tracked changes.
|
||||
pub fn set_author(app: &mut WordApp, author: &str) {
|
||||
let author: String = author.trim().chars().take(MAX_AUTHOR_CHARS).collect();
|
||||
if !author.is_empty() {
|
||||
app.session.author = author;
|
||||
}
|
||||
}
|
||||
|
||||
pub fn post_ready(&self) {
|
||||
self.send(&[("type", "wordcraft:ready".into()), ("version", env!("CARGO_PKG_VERSION").into())]);
|
||||
}
|
||||
|
||||
/// The app couldn't start (no WebGPU or WebGL2, for example): let the parent say so.
|
||||
pub fn post_failed(&self, error: &str) {
|
||||
self.send(&[("type", "wordcraft:failed".into()), ("error", error.into())]);
|
||||
}
|
||||
|
||||
fn post_save(&self, path: &str, bytes: &[u8], save_as: bool) -> Result<(), String> {
|
||||
let name = std::path::Path::new(path).file_name().map(|n| n.to_string_lossy().to_string()).unwrap_or_else(|| path.to_string());
|
||||
let buffer = js_sys::Uint8Array::from(bytes).buffer();
|
||||
let message = message(&[
|
||||
("type", "wordcraft:save".into()),
|
||||
("name", name.as_str().into()),
|
||||
("saveAs", save_as.into()),
|
||||
("bytes", buffer.clone().into()),
|
||||
])
|
||||
.ok_or_else(|| "couldn't build the message".to_string())?;
|
||||
self.parent
|
||||
.post_message_with_transfer(&message, &self.origin, &js_sys::Array::of1(&buffer))
|
||||
.map_err(|e| e.as_string().unwrap_or_else(|| format!("{e:?}")))
|
||||
}
|
||||
|
||||
/// Hear `wordcraft:open` from the parent and queue the file like a dropped one.
|
||||
fn listen(&self, inbox: Inbox, ctx: egui::Context) {
|
||||
let parent = self.parent.clone();
|
||||
let origin = self.origin.clone();
|
||||
let on_message = Closure::<dyn FnMut(web_sys::MessageEvent)>::new(move |event: web_sys::MessageEvent| {
|
||||
if event.origin() != origin || !event.source().is_some_and(|s| js_sys::Object::is(&s, &parent)) {
|
||||
return;
|
||||
}
|
||||
let data = event.data();
|
||||
if string(&data, "type").as_deref() != Some("wordcraft:open") {
|
||||
return;
|
||||
}
|
||||
let name = string(&data, "name").filter(|n| !n.trim().is_empty()).unwrap_or_else(|| "Document.docx".to_string());
|
||||
let Some(bytes) = bytes(&data) else { return };
|
||||
inbox.lock().unwrap_or_else(|e| e.into_inner()).push((name, bytes));
|
||||
ctx.request_repaint();
|
||||
});
|
||||
if self.window.add_event_listener_with_callback("message", on_message.as_ref().unchecked_ref()).is_ok() {
|
||||
// The listener lives as long as the page.
|
||||
on_message.forget();
|
||||
}
|
||||
}
|
||||
|
||||
/// Post a message to the parent; false if it couldn't be sent.
|
||||
fn send(&self, fields: &[(&str, JsValue)]) -> bool {
|
||||
message(fields).is_some_and(|m| self.parent.post_message(&m, &self.origin).is_ok())
|
||||
}
|
||||
}
|
||||
|
||||
fn message(fields: &[(&str, JsValue)]) -> Option<js_sys::Object> {
|
||||
let m = js_sys::Object::new();
|
||||
for (key, value) in fields {
|
||||
js_sys::Reflect::set(&m, &JsValue::from_str(key), value).ok()?;
|
||||
}
|
||||
Some(m)
|
||||
}
|
||||
|
||||
fn string(data: &JsValue, key: &str) -> Option<String> {
|
||||
js_sys::Reflect::get(data, &JsValue::from_str(key)).ok()?.as_string()
|
||||
}
|
||||
|
||||
/// `bytes` as an `ArrayBuffer` or `Uint8Array`.
|
||||
fn bytes(data: &JsValue) -> Option<Vec<u8>> {
|
||||
let bytes = js_sys::Reflect::get(data, &JsValue::from_str("bytes")).ok()?;
|
||||
if let Some(array) = bytes.dyn_ref::<js_sys::Uint8Array>() {
|
||||
Some(array.to_vec())
|
||||
} else if bytes.is_instance_of::<js_sys::ArrayBuffer>() {
|
||||
Some(js_sys::Uint8Array::new(&bytes).to_vec())
|
||||
} else {
|
||||
None
|
||||
}
|
||||
}
|
||||
@@ -8,9 +8,6 @@
|
||||
//! browser file picker (bytes arrive through `Services::inbox`); Save and Export download.
|
||||
//! Preferences (theme, interface size, editing options…) are kept in `localStorage` (`prefs`).
|
||||
//! URL flags: `?webgl` forces WebGL2; `?sample` opens the sample document.
|
||||
//! `?host=parent` (inside an `<iframe>`) talks to a same-origin parent page instead: it sends the
|
||||
//! document to open and stores what WordCraft saves (`host`), and `&author=` names the user for
|
||||
//! comments and tracked changes.
|
||||
#![deny(clippy::unwrap_used, clippy::expect_used, clippy::panic, clippy::unimplemented, clippy::todo, clippy::unreachable)]
|
||||
|
||||
#[cfg(any(target_arch = "wasm32", test))]
|
||||
@@ -18,8 +15,6 @@ mod prefs;
|
||||
#[cfg(any(target_arch = "wasm32", test))]
|
||||
mod save_fallback;
|
||||
|
||||
#[cfg(target_arch = "wasm32")]
|
||||
mod host;
|
||||
#[cfg(target_arch = "wasm32")]
|
||||
mod web;
|
||||
|
||||
|
||||
@@ -25,35 +25,17 @@ pub fn restore(app: &mut WordApp, stored: Option<&str>) {
|
||||
#[derive(Default)]
|
||||
pub struct PrefsSync {
|
||||
last: Option<String>,
|
||||
/// A host page's name for the user (`?author=`) and the user's own, which is stored in its
|
||||
/// place while the user keeps the host's.
|
||||
host_author: Option<(String, String)>,
|
||||
}
|
||||
|
||||
impl PrefsSync {
|
||||
/// Starts from what is already stored, so restoring doesn't write it straight back.
|
||||
pub fn new(stored: Option<String>) -> Self {
|
||||
Self { last: stored, host_author: None }
|
||||
}
|
||||
|
||||
/// The session signs with `host`, a host page's name for the user, but `own` stays the
|
||||
/// stored name. A name the user then chooses in File › Options is stored as usual.
|
||||
#[cfg_attr(not(target_arch = "wasm32"), allow(dead_code))]
|
||||
pub fn keep_own_author(&mut self, host: &str, own: &str) {
|
||||
if host != own {
|
||||
self.host_author = Some((host.to_string(), own.to_string()));
|
||||
}
|
||||
Self { last: stored }
|
||||
}
|
||||
|
||||
/// The JSON to store when the app's preferences differ from what was last written.
|
||||
pub fn changed(&mut self, app: &WordApp) -> Option<String> {
|
||||
let mut prefs = app.prefs();
|
||||
if let Some((host, own)) = &self.host_author
|
||||
&& prefs.author == *host
|
||||
{
|
||||
prefs.author = own.clone();
|
||||
}
|
||||
let json = match serde_json::to_string(&prefs) {
|
||||
let json = match serde_json::to_string(&app.prefs()) {
|
||||
Ok(j) => j,
|
||||
Err(e) => {
|
||||
log::warn!("preferences: cannot serialize: {e}");
|
||||
@@ -78,29 +60,6 @@ mod tests {
|
||||
WordApp::new(wordcraft_engine::Session::new(wordcraft_engine::sample::sample_document()), Services::default())
|
||||
}
|
||||
|
||||
/// A host page's name for the user signs the visit but isn't stored; a name the user then
|
||||
/// chooses is.
|
||||
#[test]
|
||||
fn host_author_is_not_stored() {
|
||||
let mut a = app();
|
||||
a.session.author = "Grace Hopper".into();
|
||||
let mut sync = PrefsSync::new(None);
|
||||
let stored = sync.changed(&a).unwrap();
|
||||
a.session.author = "rob".into();
|
||||
sync.keep_own_author("rob", "Grace Hopper");
|
||||
a.run("ui.theme", json!({"value": "dark"})).unwrap();
|
||||
let json = sync.changed(&a).expect("the theme changed");
|
||||
assert_ne!(json, stored);
|
||||
let mut next = app();
|
||||
restore(&mut next, Some(&json));
|
||||
assert_eq!(next.session.author, "Grace Hopper");
|
||||
a.session.author = "Rob Hones".into();
|
||||
let json = sync.changed(&a).expect("the user chose a name");
|
||||
let mut next = app();
|
||||
restore(&mut next, Some(&json));
|
||||
assert_eq!(next.session.author, "Rob Hones");
|
||||
}
|
||||
|
||||
/// #2132: dark mode chosen on one visit is there after a reload.
|
||||
#[test]
|
||||
fn dark_mode_survives_reload() {
|
||||
|
||||
@@ -27,15 +27,9 @@ pub fn start() {
|
||||
log::error!("missing <canvas id=\"{CANVAS_ID}\">");
|
||||
return;
|
||||
};
|
||||
// `?host=parent`: a same-origin page framing WordCraft opens documents in it and stores
|
||||
// what it saves (`host`). That page guards leaving it, so this one doesn't.
|
||||
let bridge = if query_param("host").as_deref() == Some("parent") { crate::host::Bridge::new() } else { None };
|
||||
let app_bridge = bridge.clone();
|
||||
let dirty = Rc::new(Cell::new(false));
|
||||
let saves = AsyncSaves::default();
|
||||
if bridge.is_none()
|
||||
&& let Err(e) = guard_unload(dirty.clone(), saves.clone())
|
||||
{
|
||||
if let Err(e) = guard_unload(dirty.clone(), saves.clone()) {
|
||||
log::error!("no unsaved-changes guard: {e}");
|
||||
}
|
||||
let mut options = eframe::WebOptions::default();
|
||||
@@ -62,17 +56,12 @@ pub fn start() {
|
||||
// The browser's locale picks the new document's paper, the units and the
|
||||
// decimal separator (#838).
|
||||
let region = wordcraft_ui_egui::i18n::system_region();
|
||||
// In host mode the parent sends the document to edit; no sample meanwhile.
|
||||
let sample = app_bridge.is_none() && query().contains("sample");
|
||||
let sample = query().contains("sample");
|
||||
let doc = if sample { wordcraft_engine::sample::sample_document() } else { wordcraft_engine::blank_document(®ion) };
|
||||
let mut session = Session::new(doc);
|
||||
session.system_region = region;
|
||||
let messages = Rc::new(RefCell::new(Vec::new()));
|
||||
let mut services = services(inbox.clone(), cc.egui_ctx.clone(), dirty, saves.clone(), messages.clone());
|
||||
if let Some(bridge) = &app_bridge {
|
||||
bridge.connect(&mut services, inbox.clone(), cc.egui_ctx.clone());
|
||||
}
|
||||
let mut app = WordApp::new(session, services);
|
||||
let mut app = WordApp::new(session, services(inbox.clone(), cc.egui_ctx.clone(), dirty, saves.clone(), messages.clone()));
|
||||
// Pasted pictures wait apart from dropped files: the app inserts one only when the page is the edit target (#1349).
|
||||
app.services.pasted = Some(pasted);
|
||||
app.autosave = false;
|
||||
@@ -85,31 +74,15 @@ pub fn start() {
|
||||
let doc = app.session.blank_document();
|
||||
app.session.set_document(doc);
|
||||
}
|
||||
let mut prefs = crate::prefs::PrefsSync::new(stored);
|
||||
let prefs = crate::prefs::PrefsSync::new(stored);
|
||||
// No document to show: start on File › Home, as the desktop app does (#766).
|
||||
if !sample {
|
||||
app.start_screen();
|
||||
}
|
||||
// The host's name for the user signs this visit's work, but isn't stored as the
|
||||
// browser's own preference, which a visit outside the host would restore.
|
||||
if app_bridge.is_some()
|
||||
&& let Some(author) = query_param("author")
|
||||
{
|
||||
let own = app.session.author.clone();
|
||||
crate::host::Bridge::set_author(&mut app, &author);
|
||||
prefs.keep_own_author(&app.session.author, &own);
|
||||
}
|
||||
Ok(Box::new(WebShell { app, inbox, messages, prefs }))
|
||||
}),
|
||||
)
|
||||
.await;
|
||||
// Ready only once the runner is up: starting can still fail after the app is made.
|
||||
if let Some(bridge) = &bridge {
|
||||
match &result {
|
||||
Ok(()) => bridge.post_ready(),
|
||||
Err(e) => bridge.post_failed(&format!("{e:?}")),
|
||||
}
|
||||
}
|
||||
if let Some(el) = document.get_element_by_id(LOADING_ID) {
|
||||
match result {
|
||||
Ok(()) => el.remove(),
|
||||
@@ -123,10 +96,6 @@ fn query() -> String {
|
||||
web_sys::window().and_then(|w| w.location().search().ok()).unwrap_or_default()
|
||||
}
|
||||
|
||||
fn query_param(key: &str) -> Option<String> {
|
||||
web_sys::UrlSearchParams::new_with_str(&query()).ok()?.get(key)
|
||||
}
|
||||
|
||||
/// Wraps the app to read dropped files asynchronously (browsers can't read them synchronously)
|
||||
/// and feed them through the inbox.
|
||||
struct WebShell {
|
||||
@@ -146,10 +115,22 @@ fn stored_prefs() -> Option<String> {
|
||||
}
|
||||
|
||||
impl eframe::App for WebShell {
|
||||
/// Runs one browser frame: forwards queued host messages to the status bar, advances the app,
|
||||
/// then writes changed preferences to browser storage immediately, because a reload or closed
|
||||
/// tab never runs the exit path.
|
||||
fn logic(&mut self, ctx: &egui::Context, _frame: &mut eframe::Frame) {
|
||||
let dropped = ctx.input_mut(|i| std::mem::take(&mut i.raw.dropped_files));
|
||||
for f in dropped {
|
||||
let inbox = self.inbox.clone();
|
||||
let ctx = ctx.clone();
|
||||
wasm_bindgen_futures::spawn_local(async move {
|
||||
let name = f.path().file_name().map(|n| n.to_string_lossy().to_string()).unwrap_or_else(|| "dropped".into());
|
||||
match f.bytes_async().await {
|
||||
Ok(bytes) => {
|
||||
inbox.lock().unwrap_or_else(|e| e.into_inner()).push((name, bytes));
|
||||
ctx.request_repaint();
|
||||
}
|
||||
Err(e) => log::error!("couldn't read dropped file {name}: {e}"),
|
||||
}
|
||||
});
|
||||
}
|
||||
let msgs = std::mem::take(&mut *self.messages.borrow_mut());
|
||||
for msg in msgs {
|
||||
self.app.status(msg);
|
||||
@@ -168,34 +149,11 @@ impl eframe::App for WebShell {
|
||||
}
|
||||
}
|
||||
|
||||
/// Forwards this frame's raw browser input to the app before widgets run.
|
||||
fn raw_input_hook(&mut self, ctx: &egui::Context, raw: &mut egui::RawInput) {
|
||||
self.app.raw_input_hook(ctx, raw);
|
||||
fn raw_input_hook(&mut self, _ctx: &egui::Context, raw: &mut egui::RawInput) {
|
||||
self.app.raw_input_hook(raw);
|
||||
}
|
||||
|
||||
/// Draws the app. On the first pass of a visible frame it takes the browser's dropped files and
|
||||
/// spawns asynchronous byte reads into the inbox; later passes only draw.
|
||||
fn ui(&mut self, ui: &mut egui::Ui, _frame: &mut eframe::Frame) {
|
||||
// Browser drops are input: take them only in the first real UI pass, never in a hidden
|
||||
// `run_logic` tick that still sees the previous shown frame's RawInput.
|
||||
if ui.ctx().current_pass_index() == 0 {
|
||||
let ctx = ui.ctx().clone();
|
||||
let dropped = ctx.input_mut(|i| std::mem::take(&mut i.raw.dropped_files));
|
||||
for f in dropped {
|
||||
let inbox = self.inbox.clone();
|
||||
let ctx = ctx.clone();
|
||||
wasm_bindgen_futures::spawn_local(async move {
|
||||
let name = f.path().file_name().map(|n| n.to_string_lossy().to_string()).unwrap_or_else(|| "dropped".into());
|
||||
match f.bytes_async().await {
|
||||
Ok(bytes) => {
|
||||
inbox.lock().unwrap_or_else(|e| e.into_inner()).push((name, bytes));
|
||||
ctx.request_repaint();
|
||||
}
|
||||
Err(e) => log::error!("couldn't read dropped file {name}: {e}"),
|
||||
}
|
||||
});
|
||||
}
|
||||
}
|
||||
self.app.ui(ui);
|
||||
}
|
||||
}
|
||||
@@ -462,7 +420,7 @@ fn trigger_save_file_picker(
|
||||
bytes: &[u8],
|
||||
) -> Result<(), String> {
|
||||
let js = |e: wasm_bindgen::JsValue| format!("{e:?}");
|
||||
let ext = std::path::Path::new(name).extension().and_then(std::ffi::OsStr::to_str).unwrap_or("docx").to_ascii_lowercase();
|
||||
let ext = std::path::Path::new(name).extension().and_then(|e| e.to_str()).unwrap_or("docx").to_ascii_lowercase();
|
||||
let dot_ext = format!(".{ext}");
|
||||
let mime = mime_for(name);
|
||||
|
||||
|
||||
@@ -19,7 +19,7 @@ wordcraft-fonts = { workspace = true }
|
||||
wordcraft-ui-egui = { workspace = true }
|
||||
# The control key file's location (shared with the MCP bridge).
|
||||
wordcraft-control-key = { workspace = true }
|
||||
eframe = { workspace = true, features = ["default_fonts", "wgpu", "accesskit", "clipboard", "wayland", "x11", "links"] }
|
||||
eframe = { workspace = true, features = ["default_fonts", "wgpu", "accesskit", "wayland", "x11", "links"] }
|
||||
egui = { workspace = true }
|
||||
serde_json = { workspace = true }
|
||||
log = { workspace = true }
|
||||
@@ -73,16 +73,13 @@ windows = { version = "0.62", features = ["UI_ViewManagement", "System_UserProfi
|
||||
clipboard-win = { version = "5.4", features = ["std"] }
|
||||
|
||||
[target.'cfg(not(target_arch = "wasm32"))'.dependencies]
|
||||
craft-print = { git = "https://github.com/storytold/craft-libs", rev = "b926b5c03e256c86ca10934f58d03f628810b5b7" }
|
||||
craft-print = { git = "https://github.com/storytold/craft-libs", rev = "d27646a5f7c8b4fc85d9dd3213b8c88b58e0088d" }
|
||||
rfd = { workspace = true }
|
||||
# Waits for an `rfd::AsyncFileDialog` on a background thread (the dialogs don't block the UI, #94);
|
||||
# rfd itself already depends on it.
|
||||
pollster = "1"
|
||||
pollster = "0.4"
|
||||
# Pictures and copied files on the system clipboard (#45); egui's paste only carries text. Already
|
||||
# in the build through egui-winit, with the same features.
|
||||
arboard = { workspace = true, features = ["image-data"] }
|
||||
# The control server's random key.
|
||||
getrandom = "0.3"
|
||||
|
||||
[target.'cfg(windows)'.dependencies]
|
||||
craft-update = { workspace = true }
|
||||
|
||||
@@ -35,15 +35,14 @@ impl AppleEvents {
|
||||
}
|
||||
|
||||
/// Open the documents and act on the quit requests that arrived since the last frame. A document
|
||||
/// opens like File › Open: beside the open ones, in place of an untouched blank one, or, when it
|
||||
/// is already open, by showing it. Quitting closes the window, which asks about every document
|
||||
/// with unsaved changes.
|
||||
/// opens like File › Open (asking about unsaved changes first); quitting closes the window, which
|
||||
/// asks the same.
|
||||
pub fn poll(inbox: &Inbox, app: &mut WordApp, ctx: &egui::Context) {
|
||||
for e in inbox.drain() {
|
||||
match e {
|
||||
Event::Open(paths) => {
|
||||
for p in paths.iter().filter_map(|p| p.to_str()) {
|
||||
match app.open_file(p, None) {
|
||||
match app.run("file.open", serde_json::json!({"path": p})) {
|
||||
// Leave Backstage, as opening from the Open dialog does.
|
||||
Ok(_) => app.ui.backstage = false,
|
||||
Err(e) => {
|
||||
@@ -55,12 +54,7 @@ pub fn poll(inbox: &Inbox, app: &mut WordApp, ctx: &egui::Context) {
|
||||
}
|
||||
}
|
||||
}
|
||||
// Quitting asks about every document with unsaved changes, each in its window; the
|
||||
// app closes once that's done (`quit_requested`).
|
||||
Event::Quit => {
|
||||
app.quit();
|
||||
ctx.request_repaint();
|
||||
}
|
||||
Event::Quit => ctx.send_viewport_cmd(egui::ViewportCommand::Close),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -95,8 +95,7 @@ pub fn formats() -> Vec<String> {
|
||||
|
||||
/// The picture on the clipboard: copied files, else image pixels, else (Wayland) an encoded
|
||||
/// picture in a format arboard doesn't decode. A picture that is there but can't be read says
|
||||
/// why (#833). The UI checks the sizes before inserting anything; on Windows they are checked
|
||||
/// from the header before anything is decoded (#3623).
|
||||
/// why (#833). The UI checks the sizes before inserting anything.
|
||||
pub fn picture() -> Option<ClipboardPicture> {
|
||||
let mut clipboard = match arboard::Clipboard::new() {
|
||||
Ok(c) => c,
|
||||
@@ -110,19 +109,6 @@ pub fn picture() -> Option<ClipboardPicture> {
|
||||
{
|
||||
return Some(ClipboardPicture::Files(files));
|
||||
}
|
||||
decoded_picture(&mut clipboard)
|
||||
}
|
||||
|
||||
/// The clipboard's picture on Windows, read past arboard, which decodes the PNG and DIB formats
|
||||
/// with no size limits (#3623).
|
||||
#[cfg(target_os = "windows")]
|
||||
fn decoded_picture(_: &mut arboard::Clipboard) -> Option<ClipboardPicture> {
|
||||
platform::picture()
|
||||
}
|
||||
|
||||
/// The picture arboard decodes from the clipboard, else an encoded one read past it.
|
||||
#[cfg(not(target_os = "windows"))]
|
||||
fn decoded_picture(clipboard: &mut arboard::Clipboard) -> Option<ClipboardPicture> {
|
||||
match clipboard.get_image() {
|
||||
Ok(image) => Some(ClipboardPicture::Pixels { width: image.width, height: image.height, rgba: image.bytes.into_owned() }),
|
||||
Err(arboard::Error::ContentNotAvailable) => platform::encoded_picture().map(ClipboardPicture::Encoded),
|
||||
@@ -201,7 +187,6 @@ mod platform {
|
||||
#[cfg(target_os = "windows")]
|
||||
mod platform {
|
||||
use clipboard_win::{Clipboard, EnumFormats, raw};
|
||||
use wordcraft_ui_egui::paste_picture::{self, ClipboardPicture};
|
||||
|
||||
pub fn formats() -> Vec<String> {
|
||||
let Ok(_open) = Clipboard::new_attempts(10) else { return Vec::new() };
|
||||
@@ -220,31 +205,8 @@ mod platform {
|
||||
Some(out)
|
||||
}
|
||||
|
||||
/// The clipboard's picture, read without arboard: its "PNG" format as it is, else a DIB as
|
||||
/// pixels, each refused from its header when a paste can't take its size (#3623). None when
|
||||
/// it holds neither.
|
||||
pub fn picture() -> Option<ClipboardPicture> {
|
||||
use clipboard_win::formats::{CF_DIB, CF_DIBV5};
|
||||
let Ok(_open) = Clipboard::new_attempts(10) else {
|
||||
return Some(ClipboardPicture::Unreadable("the system clipboard couldn't be opened".into()));
|
||||
};
|
||||
if let Some(png) = raw::register_format("PNG").map(|f| f.get()).filter(|&f| raw::is_format_avail(f)) {
|
||||
return Some(read(png).map_or_else(ClipboardPicture::Unreadable, paste_picture::encoded_picture));
|
||||
}
|
||||
let dib = [CF_DIBV5, CF_DIB].into_iter().find(|&f| raw::is_format_avail(f))?;
|
||||
Some(read(dib).map_or_else(ClipboardPicture::Unreadable, paste_picture::dib_picture))
|
||||
}
|
||||
|
||||
/// A clipboard format's bytes, refused past the largest picture a paste takes.
|
||||
fn read(format: u32) -> Result<Vec<u8>, String> {
|
||||
match raw::size(format) {
|
||||
Some(n) if n.get() as u64 <= paste_picture::MAX_FILE_BYTES => {}
|
||||
Some(_) => return Err("the clipboard picture is larger than 200 MB".into()),
|
||||
None => return Err("the clipboard picture is empty".into()),
|
||||
}
|
||||
let mut out = Vec::new();
|
||||
raw::get_vec(format, &mut out).map_err(|e| e.to_string())?;
|
||||
Ok(out)
|
||||
pub fn encoded_picture() -> Option<Vec<u8>> {
|
||||
None
|
||||
}
|
||||
|
||||
pub fn can_set() -> bool {
|
||||
|
||||
@@ -36,8 +36,7 @@ const DESKTOP_ID: &str = "ai.storyteller.wordcraft.desktop";
|
||||
|
||||
/// The types the button makes WordCraft the default for on Linux and BSD: Word documents and
|
||||
/// templates (also macro-enabled and legacy), RTF and ODT. The same set the Windows installer
|
||||
/// lists under WordCraft's Capabilities; Markdown is only offered under Open With, and plain
|
||||
/// text only on Windows and macOS (the `.desktop` file leaves it out, #4137).
|
||||
/// lists under WordCraft's Capabilities; Markdown and plain text are only offered under Open With.
|
||||
/// Every one is in the `.desktop` file's `MimeType=` (a test checks).
|
||||
#[cfg(any(all(unix, not(target_os = "macos")), test))]
|
||||
const WORD_MIME_TYPES: &[&str] = &[
|
||||
@@ -165,16 +164,6 @@ mod tests {
|
||||
assert!(entry.lines().any(|l| l == "Exec=wordcraft %F"), "files are passed on the command line");
|
||||
}
|
||||
|
||||
/// Declaring `text/plain` makes WordCraft a handler for every config, script and log on systems with no default set.
|
||||
#[test]
|
||||
fn the_desktop_entry_does_not_declare_plain_text() {
|
||||
let entry = include_str!("../../../packaging/linux/ai.storyteller.wordcraft.desktop");
|
||||
let declared: Vec<&str> =
|
||||
entry.lines().find_map(|l| l.strip_prefix("MimeType=")).map(|v| v.split(';').filter(|s| !s.is_empty()).collect()).unwrap_or_default();
|
||||
assert!(!declared.is_empty(), "MimeType= line not found");
|
||||
assert!(!declared.contains(&"text/plain"), "text/plain must not be declared");
|
||||
}
|
||||
|
||||
#[cfg(unix)]
|
||||
#[test]
|
||||
fn application_dirs_follow_the_xdg_spec() {
|
||||
|
||||
@@ -207,8 +207,8 @@ fn on_path(program: &str) -> bool {
|
||||
fn save_exts(name: &str) -> Vec<String> {
|
||||
let suggested = std::path::Path::new(name)
|
||||
.extension()
|
||||
.and_then(std::ffi::OsStr::to_str)
|
||||
.map(str::to_ascii_lowercase)
|
||||
.and_then(|e| e.to_str())
|
||||
.map(|e| e.to_ascii_lowercase())
|
||||
.filter(|e| wordcraft_engine::io::SAVE_EXTS.contains(&e.as_str()))
|
||||
.unwrap_or_else(|| "docx".to_string());
|
||||
let mut exts: Vec<String> = Vec::with_capacity(wordcraft_engine::io::SAVE_EXTS.len());
|
||||
|
||||
+11
-48
@@ -27,66 +27,47 @@ mod traffic_lights;
|
||||
use wordcraft_engine::Session;
|
||||
use wordcraft_ui_egui::{
|
||||
Services, UiState, WordApp,
|
||||
window_geometry::{RawWindowState, WindowGeometry, take_rescue},
|
||||
window_geometry::{WindowGeometry, take_rescue},
|
||||
};
|
||||
|
||||
/// The app, the restored window geometry (or, with none, the default: no position) until the
|
||||
/// first frame that knows the monitor has checked it, the control server's key file (removed on
|
||||
/// exit), the file dialogs the app asked for, this frame's raw host window state and, on macOS,
|
||||
/// the documents opened from Finder.
|
||||
/// exit), the file dialogs the app asked for and, on macOS, the documents opened from Finder.
|
||||
struct App(
|
||||
WordApp,
|
||||
Option<WindowGeometry>,
|
||||
Option<control_server::KeyFile>,
|
||||
file_dialogs::Launcher,
|
||||
// The host's answer for this frame. `InputState` still describes the last painted frame on
|
||||
// a hidden logic tick.
|
||||
RawWindowState,
|
||||
#[cfg(target_os = "macos")] fmv_macos_events::Inbox,
|
||||
);
|
||||
|
||||
impl eframe::App for App {
|
||||
/// Advances one desktop frame: polls Finder events on macOS, runs the app and its file dialogs,
|
||||
/// fits or rescues the window and maximizes it after opening, then gates a close or quit request
|
||||
/// on unsaved changes and active printing.
|
||||
fn logic(&mut self, ctx: &egui::Context, frame: &mut eframe::Frame) {
|
||||
#[cfg(target_os = "macos")]
|
||||
apple_events::poll(&self.5, &mut self.0, ctx);
|
||||
apple_events::poll(&self.4, &mut self.0, ctx);
|
||||
self.0.logic(ctx);
|
||||
self.3.show(frame);
|
||||
let window = self.4;
|
||||
// A window that doesn't fit on its screen (the default size on a laptop) shrinks to fit
|
||||
// there, so all of the scroll bar can be dragged (#1215). Checked on the frame that first
|
||||
// knows the monitor: the off-screen check below takes the pending geometry then.
|
||||
if let Some((pos, size)) = self.1.and_then(|g| g.fit_on_monitor(&window)) {
|
||||
if let Some((pos, size)) = ctx.input(|i| self.1.and_then(|g| g.fit_on_monitor(i.viewport()))) {
|
||||
ctx.send_viewport_cmd(egui::ViewportCommand::InnerSize(size));
|
||||
ctx.send_viewport_cmd(egui::ViewportCommand::OuterPosition(pos));
|
||||
}
|
||||
// The geometry saved for the next start is the active window's.
|
||||
let prev = self.0.ui.window;
|
||||
// Another document window's state is the one it last painted with: the host reports only
|
||||
// the root window's for this frame.
|
||||
let active = self.0.active_viewport();
|
||||
let tracked = if active == egui::ViewportId::ROOT {
|
||||
window
|
||||
} else {
|
||||
ctx.input_for(active, |i| RawWindowState::from_viewport(Some(i.viewport()), Some(i.viewport_rect())))
|
||||
};
|
||||
self.0.ui.window = WindowGeometry::track(prev, &tracked);
|
||||
self.0.ui.window = ctx.input(|i| WindowGeometry::track(prev, i.viewport(), i.viewport_rect().size()));
|
||||
// The first frame, once the monitor is known (as `take_rescue` waits for): maximize a
|
||||
// window that opened at its normal size so it never flashed on screen half-opened.
|
||||
let maximize = self.1.and_then(|g| g.maximize_after_open()).filter(|_| window.monitor_size.is_some());
|
||||
if let Some(pos) = take_rescue(&mut self.1, &window) {
|
||||
let maximize = self.1.and_then(|g| g.maximize_after_open()).filter(|_| ctx.input(|i| i.viewport().monitor_size.is_some()));
|
||||
if let Some(pos) = ctx.input(|i| take_rescue(&mut self.1, i.viewport())) {
|
||||
ctx.send_viewport_cmd(egui::ViewportCommand::OuterPosition(pos));
|
||||
}
|
||||
if let Some(command) = maximize {
|
||||
ctx.send_viewport_cmd(command);
|
||||
}
|
||||
// The window's close button: with other windows open only this window closes (the app
|
||||
// handles that while drawing it); the last one quits, asking first about unsaved changes
|
||||
// (`WordApp::close_requested`).
|
||||
// The window's close button: with unsaved changes, ask first (`WordApp::close_requested`).
|
||||
let closing = ctx.input(|i| i.viewport().close_requested());
|
||||
if closing && (self.0.closes_a_window() || !self.0.close_requested()) {
|
||||
if closing && !self.0.close_requested() {
|
||||
ctx.send_viewport_cmd(egui::ViewportCommand::CancelClose);
|
||||
} else if (closing || self.0.quit_requested) && !self.0.printing.busy() {
|
||||
// On macOS, letting eframe drop the window inside `-[NSApplication run]` aborts in
|
||||
@@ -104,13 +85,8 @@ impl eframe::App for App {
|
||||
}
|
||||
}
|
||||
}
|
||||
/// Records this frame's native window state before forwarding the hook, so logic never reads a
|
||||
/// stale inner rectangle from a hidden tick.
|
||||
fn raw_input_hook(&mut self, ctx: &egui::Context, raw: &mut egui::RawInput) {
|
||||
// Keep this frame's native window state. `logic` must not read the last painted
|
||||
// `InputState` when a hidden Wayland tick reports no fresh inner rect.
|
||||
self.4 = RawWindowState::from_raw(raw);
|
||||
self.0.raw_input_hook(ctx, raw);
|
||||
fn raw_input_hook(&mut self, _ctx: &egui::Context, raw: &mut egui::RawInput) {
|
||||
self.0.raw_input_hook(raw);
|
||||
}
|
||||
fn ui(&mut self, ui: &mut egui::Ui, frame: &mut eframe::Frame) {
|
||||
self.0.ui(ui);
|
||||
@@ -344,15 +320,7 @@ fn open_non_unicode(app: &mut WordApp, path: &std::path::Path) {
|
||||
}
|
||||
}
|
||||
|
||||
/// Parses the command line, installs logging and the panic hook, restores preferences and window
|
||||
/// geometry, opens requested documents or the Start screen, starts the control server when
|
||||
/// asked, and runs the eframe event loop.
|
||||
fn main() -> eframe::Result {
|
||||
// Windows: the update helper is a copy of this program started with `--finish-update`.
|
||||
#[cfg(windows)]
|
||||
if let Some(result) = craft_update::self_update::finish(&craft_update::suite::WORDCRAFT, &std::env::args_os().skip(1).collect::<Vec<_>>()) {
|
||||
std::process::exit(i32::from(result.is_err()));
|
||||
}
|
||||
// First, so the panic hook and every start-up warning are recorded (`logging`).
|
||||
let logger = logging::install();
|
||||
logging::install_panic_hook();
|
||||
@@ -429,8 +397,6 @@ fn main() -> eframe::Result {
|
||||
#[cfg(target_os = "macos")]
|
||||
graphics::configure_indirect_validation(&mut options, graphics::skips_indirect_validation(true, cfg!(target_arch = "x86_64")));
|
||||
load_user_fonts();
|
||||
// Windows opened for further documents start like the first one, before its saved place.
|
||||
let window_builder = options.viewport.clone();
|
||||
let restored = saved_window();
|
||||
if let Some(window) = restored {
|
||||
options.viewport = window.apply(options.viewport);
|
||||
@@ -463,7 +429,6 @@ fn main() -> eframe::Result {
|
||||
let mut session = Session::new(doc);
|
||||
session.system_region = region;
|
||||
let mut app = WordApp::new(session, services(file_dialog, &cc.egui_ctx));
|
||||
app.set_startup_update_check(true);
|
||||
load_prefs(&mut app);
|
||||
load_user_dictionary(&mut app);
|
||||
// A region chosen in File › Options replaces the system's for the first blank
|
||||
@@ -482,7 +447,6 @@ fn main() -> eframe::Result {
|
||||
}
|
||||
}
|
||||
app.ui.window = restored;
|
||||
app.set_window_builder(window_builder);
|
||||
app.integrated_titlebar = cfg!(target_os = "macos");
|
||||
#[cfg(target_os = "macos")]
|
||||
if let Some(bar) = traffic_lights::title_bar() {
|
||||
@@ -516,7 +480,6 @@ fn main() -> eframe::Result {
|
||||
Some(restored.unwrap_or_default()),
|
||||
key_file,
|
||||
dialogs,
|
||||
RawWindowState::default(),
|
||||
#[cfg(target_os = "macos")]
|
||||
apple_events.connect(&cc.egui_ctx),
|
||||
)))
|
||||
|
||||
@@ -1,16 +1,11 @@
|
||||
//! Desktop spooler adapter. Discovery and submission never run on the UI thread.
|
||||
use wordcraft_ui_egui::printing::{Discover, Duplex, Printer, Settings, Submit};
|
||||
|
||||
/// Returns a discovery callback that enumerates spooler printers on a named worker thread and
|
||||
/// sends the list through a channel, keeping spooler calls off the UI thread.
|
||||
pub fn discover() -> Discover {
|
||||
Box::new(|| {
|
||||
let (sender, receiver) = std::sync::mpsc::channel();
|
||||
let _ = std::thread::Builder::new().name("print-discovery".into()).spawn(move || {
|
||||
let printers = craft_print::printers()
|
||||
.into_iter()
|
||||
.map(|printer| Printer { name: printer.name, label: printer.description, default: printer.default })
|
||||
.collect();
|
||||
let printers = craft_print::printers().into_iter().map(|printer| Printer { name: printer.name, default: printer.default }).collect();
|
||||
let _ = sender.send(printers);
|
||||
});
|
||||
receiver
|
||||
@@ -18,11 +13,11 @@ pub fn discover() -> Discover {
|
||||
}
|
||||
|
||||
// Keep submission workers reachable until normal app shutdown; dropping a JoinHandle would
|
||||
// let process exit end an in-process Windows job or the CUPS hand-off before the spooler has it.
|
||||
// let process exit kill PowerShell/CUPS before it hands the job to the spooler.
|
||||
static JOBS: std::sync::Mutex<Vec<std::thread::JoinHandle<()>>> = std::sync::Mutex::new(Vec::new());
|
||||
|
||||
pub fn finish_pending_jobs() {
|
||||
let workers = std::mem::take(&mut *JOBS.lock().unwrap_or_else(std::sync::PoisonError::into_inner));
|
||||
let workers = std::mem::take(&mut *JOBS.lock().unwrap_or_else(|poison| poison.into_inner()));
|
||||
for worker in workers {
|
||||
if worker.join().is_err() {
|
||||
log::error!("the print worker stopped unexpectedly during shutdown");
|
||||
@@ -51,7 +46,7 @@ pub fn submit() -> Submit {
|
||||
let _ = sender.send(craft_print::submit(&bytes, &job).map_err(|error| error.to_string()));
|
||||
});
|
||||
if let Ok(worker) = worker {
|
||||
let mut jobs = JOBS.lock().unwrap_or_else(std::sync::PoisonError::into_inner);
|
||||
let mut jobs = JOBS.lock().unwrap_or_else(|poison| poison.into_inner());
|
||||
jobs.retain(|job| !job.is_finished());
|
||||
jobs.push(worker);
|
||||
}
|
||||
|
||||
@@ -176,8 +176,5 @@ display_name = "Mulai"
|
||||
[people.Bortlesboat]
|
||||
real_name = "Andrew Barnes"
|
||||
|
||||
[people.matstrange]
|
||||
sync_github_name = true
|
||||
|
||||
[people.pakiat]
|
||||
display_name = "OtorO"
|
||||
|
||||
+1
-1
@@ -1 +1 @@
|
||||
9205108cfdced5177706264038848b68b2422d2b
|
||||
04600cbb1508f1b15927ec574a5e32e5f376fd8b
|
||||
|
||||
@@ -8,15 +8,9 @@ description = "WordCraft document model: stories, paragraphs, runs, tables, sect
|
||||
[lints]
|
||||
workspace = true
|
||||
|
||||
[features]
|
||||
# Test inputs other crates' tests share (`cfb_guard::repeating_difat_sample`); turned on only as a
|
||||
# dev-dependency, so never built into the app.
|
||||
test-support = []
|
||||
|
||||
[dependencies]
|
||||
wordcraft-geom = { workspace = true }
|
||||
base64 = { workspace = true }
|
||||
log = { workspace = true }
|
||||
serde = { workspace = true }
|
||||
serde_json = { workspace = true }
|
||||
thiserror = { workspace = true }
|
||||
|
||||
@@ -1,133 +0,0 @@
|
||||
//! A bytes-only check of an OLE compound file ([MS-CFB]) before the `cfb` crate opens it
|
||||
//! (#3642). Every reader of `.doc`, `.xls` and encrypted Office packages calls [`check`] first.
|
||||
|
||||
/// Check a compound file's FAT layout before `cfb` reads it ([MS-CFB] §2.2 header, §2.5
|
||||
/// DIFAT). `cfb` appends a whole sector of FAT entries for every FAT sector the DIFAT lists and
|
||||
/// only validates the FAT once it is built, so a small header naming one sector over and over
|
||||
/// makes it allocate gigabytes. Here the DIFAT is walked the way `cfb` walks it (header entries,
|
||||
/// then the DIFAT chain, at most one step per sector; trailing zero and free entries trimmed
|
||||
/// the same way) and the file is refused when it lists one FAT sector twice, or more FAT sectors
|
||||
/// than its sectors need (one FAT sector covers `sector_len / 4` of them, plus one spare), so the
|
||||
/// FAT `cfb` builds is about the size the file needs. Anything else that is wrong is left for
|
||||
/// `cfb` to report.
|
||||
pub fn check(bytes: &[u8]) -> Result<(), String> {
|
||||
const MAGIC: [u8; 8] = [0xD0, 0xCF, 0x11, 0xE0, 0xA1, 0xB1, 0x1A, 0xE1];
|
||||
const FREE: u32 = 0xFFFF_FFFF;
|
||||
const END_OF_CHAIN: u32 = 0xFFFF_FFFE;
|
||||
const HEADER_ENTRIES: usize = 109;
|
||||
let u16_at = |at: usize| bytes.get(at..at.checked_add(2)?).map(|b| u16::from_le_bytes([b[0], b[1]]));
|
||||
let u32_at = |at: usize| bytes.get(at..at.checked_add(4)?).map(|b| u32::from_le_bytes([b[0], b[1], b[2], b[3]]));
|
||||
if bytes.len() < 512 || bytes.get(..8) != Some(MAGIC.as_slice()) {
|
||||
return Ok(());
|
||||
}
|
||||
let sector_len: usize = match u16_at(0x1A) {
|
||||
Some(3) => 512,
|
||||
Some(4) => 4096,
|
||||
_ => return Ok(()),
|
||||
};
|
||||
if bytes.len() < sector_len {
|
||||
return Ok(());
|
||||
}
|
||||
// `cfb` counts the sectors after the header sector, the last one possibly partial.
|
||||
let num_sectors = bytes.len().div_ceil(sector_len) - 1;
|
||||
let num_fat_sectors = u32_at(0x2C).unwrap_or(0) as usize;
|
||||
let mut difat: Vec<u32> = (0..HEADER_ENTRIES).filter_map(|i| u32_at(0x4C + i * 4)).collect();
|
||||
let mut next = u32_at(0x44).unwrap_or(END_OF_CHAIN);
|
||||
let mut steps = 0usize;
|
||||
while next != END_OF_CHAIN && next != FREE {
|
||||
steps += 1;
|
||||
// `cfb` refuses a chain sector past the file or seen before; a chain longer than the
|
||||
// file has sectors must have done one of the two.
|
||||
if steps > num_sectors || next as usize >= num_sectors {
|
||||
return Err(format!("the DIFAT chain is longer than the file or leaves it (sector {next})"));
|
||||
}
|
||||
let at = (next as usize + 1) * sector_len;
|
||||
let entries = sector_len / 4 - 1;
|
||||
difat.extend((0..entries).map(|i| u32_at(at + i * 4).unwrap_or(FREE)));
|
||||
next = u32_at(at + entries * 4).unwrap_or(END_OF_CHAIN);
|
||||
}
|
||||
while difat.len() > HEADER_ENTRIES && difat.len() > num_fat_sectors && difat.last() == Some(&0) {
|
||||
difat.pop();
|
||||
}
|
||||
while difat.last() == Some(&FREE) {
|
||||
difat.pop();
|
||||
}
|
||||
// Each listed FAT sector adds a whole sector of entries. A FAT sector holds the entries of
|
||||
// `sector_len / 4` sectors, so a file needs no more FAT sectors than that covers (one spare),
|
||||
// and none of them twice. A free entry in the middle is left for `cfb` to refuse.
|
||||
let listed: Vec<u32> = difat.into_iter().filter(|&s| s != FREE).collect();
|
||||
let needed = num_sectors.div_ceil(sector_len / 4).saturating_add(1);
|
||||
if listed.len() > needed {
|
||||
return Err(format!("the DIFAT lists {} FAT sectors in a file of {num_sectors} sectors", listed.len()));
|
||||
}
|
||||
let mut seen = std::collections::HashSet::with_capacity(listed.len());
|
||||
if let Some(dup) = listed.iter().find(|&&s| !seen.insert(s)) {
|
||||
return Err(format!("the DIFAT lists FAT sector {dup} twice"));
|
||||
}
|
||||
Ok(())
|
||||
}
|
||||
|
||||
/// A compound file of about 1 MB whose DIFAT chain lists one FAT sector 2048 × 127 times:
|
||||
/// without [`check`], `cfb` builds a FAT of about 130 MB for it. Regression input for the
|
||||
/// readers' tests: built only for tests, this crate's and (through the `test-support` feature,
|
||||
/// which they turn on as a dev-dependency) the readers', never into the app.
|
||||
#[cfg(any(test, feature = "test-support"))]
|
||||
pub fn repeating_difat_sample() -> Vec<u8> {
|
||||
const S: usize = 512;
|
||||
const CHAIN: u32 = 2048; // DIFAT sectors 1..=CHAIN, sector 0 is the "FAT"
|
||||
let mut f = vec![0u8; S * (CHAIN as usize + 2)];
|
||||
let mut put = |at: usize, v: &[u8]| {
|
||||
if let Some(dst) = f.get_mut(at..at + v.len()) {
|
||||
dst.copy_from_slice(v);
|
||||
}
|
||||
};
|
||||
put(0, &[0xD0, 0xCF, 0x11, 0xE0, 0xA1, 0xB1, 0x1A, 0xE1]);
|
||||
put(0x18, &0x3Eu16.to_le_bytes());
|
||||
put(0x1A, &3u16.to_le_bytes());
|
||||
put(0x1C, &0xFFFEu16.to_le_bytes());
|
||||
put(0x1E, &9u16.to_le_bytes());
|
||||
put(0x20, &6u16.to_le_bytes());
|
||||
put(0x2C, &u32::MAX.to_le_bytes()); // no zero trimming
|
||||
put(0x38, &4096u32.to_le_bytes());
|
||||
put(0x44, &1u32.to_le_bytes());
|
||||
put(0x48, &CHAIN.to_le_bytes());
|
||||
// The 109 header entries are all 0 already; each chain sector's entries too.
|
||||
for d in 1..=CHAIN {
|
||||
let next = if d == CHAIN { 0xFFFF_FFFE } else { d + 1 };
|
||||
put((d as usize + 2) * S - 4, &next.to_le_bytes());
|
||||
}
|
||||
f
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
#[test]
|
||||
fn a_repeating_difat_is_refused_and_a_small_file_passes() {
|
||||
assert!(super::check(&super::repeating_difat_sample()).is_err_and(|e| e.contains("FAT sectors in a file of")));
|
||||
assert!(super::check(b"not a compound file").is_ok());
|
||||
}
|
||||
|
||||
/// A 512-byte-sector file of `sectors` sectors after the header, whose header DIFAT lists
|
||||
/// `fat` (the rest free).
|
||||
fn header_difat(sectors: usize, fat: &[u32]) -> Vec<u8> {
|
||||
let mut f = vec![0u8; 512 * (sectors + 1)];
|
||||
f[..8].copy_from_slice(&[0xD0, 0xCF, 0x11, 0xE0, 0xA1, 0xB1, 0x1A, 0xE1]);
|
||||
f[0x1A..0x1C].copy_from_slice(&3u16.to_le_bytes());
|
||||
f[0x2C..0x30].copy_from_slice(&(fat.len() as u32).to_le_bytes());
|
||||
f[0x44..0x48].copy_from_slice(&0xFFFF_FFFEu32.to_le_bytes());
|
||||
for i in 0..109 {
|
||||
let v = fat.get(i).copied().unwrap_or(u32::MAX);
|
||||
f[0x4C + i * 4..0x50 + i * 4].copy_from_slice(&v.to_le_bytes());
|
||||
}
|
||||
f
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_fat_sector_listed_twice_or_more_than_the_file_needs_is_refused() {
|
||||
// Each list is shorter than the file has sectors, which the guard used to accept.
|
||||
assert!(super::check(&header_difat(10, &[0, 0])).is_err_and(|e| e.contains("FAT sector 0 twice")));
|
||||
assert!(super::check(&header_difat(10, &[0, 1, 2])).is_err_and(|e| e.contains("3 FAT sectors")));
|
||||
assert!(super::check(&header_difat(10, &[0])).is_ok());
|
||||
assert!(super::check(&header_difat(200, &[0, 1])).is_ok());
|
||||
}
|
||||
}
|
||||
@@ -238,7 +238,7 @@ pub(crate) fn ideograph_zodiac_traditional(n: u32) -> String {
|
||||
|
||||
/// The `n`th (1-based) item of a sequence, or decimal past its end.
|
||||
fn nth(set: &[char], n: u32) -> String {
|
||||
n.checked_sub(1).and_then(|i| set.get(i as usize)).map(ToString::to_string).unwrap_or_else(|| n.to_string())
|
||||
n.checked_sub(1).and_then(|i| set.get(i as usize)).map(|c| c.to_string()).unwrap_or_else(|| n.to_string())
|
||||
}
|
||||
|
||||
/// Code points `first..` enclosing 1 to `count`, decimal beyond.
|
||||
|
||||
@@ -13,41 +13,6 @@ use crate::graphic::PathSeg;
|
||||
use crate::para::{InlineObject, ShapeKind};
|
||||
use crate::wordart::TextWarp;
|
||||
|
||||
/// DrawingML outline ends (ST_LineCap); missing and unknown values use flat ends.
|
||||
#[derive(Clone, Copy, Debug, Default, PartialEq, Eq, Serialize, Deserialize)]
|
||||
#[serde(rename_all = "camelCase")]
|
||||
pub enum LineCap {
|
||||
#[default]
|
||||
Flat,
|
||||
Square,
|
||||
Round,
|
||||
}
|
||||
|
||||
impl LineCap {
|
||||
/// Read an ST_LineCap value; absent or unrecognized values use the flat default.
|
||||
pub fn from_ooxml(value: &str) -> Self {
|
||||
match value {
|
||||
"sq" => Self::Square,
|
||||
"rnd" => Self::Round,
|
||||
_ => Self::Flat,
|
||||
}
|
||||
}
|
||||
|
||||
/// Return the canonical DrawingML spelling used when saving the outline.
|
||||
pub fn ooxml(self) -> &'static str {
|
||||
match self {
|
||||
Self::Flat => "flat",
|
||||
Self::Square => "sq",
|
||||
Self::Round => "rnd",
|
||||
}
|
||||
}
|
||||
|
||||
/// Whether the cap can be omitted from older-compatible serialized shape properties.
|
||||
pub fn is_flat(&self) -> bool {
|
||||
*self == Self::Flat
|
||||
}
|
||||
}
|
||||
|
||||
/// The rest of a shape's properties. Empty (the default) for most shapes.
|
||||
#[derive(Clone, Debug, PartialEq, Default, Serialize, Deserialize)]
|
||||
#[serde(default, rename_all = "camelCase")]
|
||||
@@ -76,9 +41,6 @@ pub struct ShapeExtra {
|
||||
/// The outline dash pattern; absent in older saved documents means solid.
|
||||
#[serde(skip_serializing_if = "crate::line_dash::LineDash::is_solid")]
|
||||
pub dash: crate::line_dash::LineDash,
|
||||
/// The outline ends; older saved documents and absent DrawingML caps are flat.
|
||||
#[serde(skip_serializing_if = "LineCap::is_flat")]
|
||||
pub cap: LineCap,
|
||||
/// The outline's gradient (`a:ln/a:gradFill`), kept for saving; the outline is drawn in its
|
||||
/// main colour (the shape's `stroke`).
|
||||
#[serde(skip_serializing_if = "Option::is_none")]
|
||||
@@ -310,18 +272,6 @@ mod tests {
|
||||
use super::*;
|
||||
use crate::para::{Float, GroupChild};
|
||||
|
||||
/// Older JSON retains flat ends without adding a property; non-flat values round-trip.
|
||||
#[test]
|
||||
fn older_shape_json_defaults_to_flat_caps() {
|
||||
let extra: ShapeExtra = serde_json::from_str("{}").unwrap();
|
||||
assert_eq!(extra.cap, LineCap::Flat);
|
||||
assert_eq!(serde_json::to_value(&extra).unwrap(), serde_json::json!({}));
|
||||
for cap in [LineCap::Square, LineCap::Round] {
|
||||
let extra = ShapeExtra { cap, ..Default::default() };
|
||||
assert_eq!(serde_json::from_value::<ShapeExtra>(serde_json::to_value(&extra).unwrap()).unwrap(), extra);
|
||||
}
|
||||
}
|
||||
|
||||
fn shape(kind: ShapeKind, w: f32, h: f32, extra: ShapeExtra) -> InlineObject {
|
||||
InlineObject::Shape {
|
||||
kind,
|
||||
|
||||
+21
-45
@@ -14,8 +14,6 @@
|
||||
//! outside what's deleted (so deleting across a control's edge empties it rather than breaking
|
||||
//! it); writers call [`Document::balance_controls`] so a saved file always has whole controls.
|
||||
|
||||
use std::collections::{BTreeMap, BTreeSet};
|
||||
|
||||
use serde::{Deserialize, Serialize};
|
||||
|
||||
use crate::para::OBJ;
|
||||
@@ -314,10 +312,10 @@ pub fn control_events(bl: &Blocks) -> Vec<ControlEvent<'_>> {
|
||||
if !p.objects.iter().any(|o| matches!(o, InlineObject::ControlStart { .. } | InlineObject::ControlEnd)) {
|
||||
continue;
|
||||
}
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
match o {
|
||||
InlineObject::ControlStart { control } => out.push(ControlEvent::Open { block: i, off: Some(off), control }),
|
||||
InlineObject::ControlEnd => out.push(ControlEvent::Close { block: i, off: Some(off) }),
|
||||
for off in p.object_offsets() {
|
||||
match p.object_at(off) {
|
||||
Some(InlineObject::ControlStart { control }) => out.push(ControlEvent::Open { block: i, off: Some(off), control }),
|
||||
Some(InlineObject::ControlEnd) => out.push(ControlEvent::Close { block: i, off: Some(off) }),
|
||||
_ => {}
|
||||
}
|
||||
}
|
||||
@@ -641,52 +639,34 @@ impl Document {
|
||||
out.push((r.end.path.clone(), r.end.off));
|
||||
}
|
||||
}
|
||||
// Markers paired with a table wrap: their partner is never deleted with them. Each lookup
|
||||
// is a set probe, and each container is paired once, not once per marker (#3614).
|
||||
let in_ranges: BTreeSet<(&Path, usize)> = ranges.iter().flat_map(|r| [(&r.start.path, r.start.off), (&r.end.path, r.end.off)]).collect();
|
||||
let mut table_paired: BTreeMap<&[u32], BTreeSet<(usize, usize)>> = BTreeMap::new();
|
||||
// Markers paired with a table wrap: their partner is never deleted with them.
|
||||
for path in &paths {
|
||||
let Some(p) = self.para(a.story, path) else { continue };
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
if !inside(path, off)
|
||||
|| !matches!(o, InlineObject::ControlStart { .. } | InlineObject::ControlEnd)
|
||||
|| in_ranges.contains(&(path, off))
|
||||
{
|
||||
for off in p.object_offsets() {
|
||||
if !inside(path, off) || !matches!(p.object_at(off), Some(InlineObject::ControlStart { .. } | InlineObject::ControlEnd)) {
|
||||
continue;
|
||||
}
|
||||
let paired = table_paired.entry(path.parent()).or_insert_with(|| self.table_paired_markers(a.story, path));
|
||||
if paired.contains(&(path.last(), off)) {
|
||||
let paired = ranges.iter().any(|r| (r.start.path == *path && r.start.off == off) || (r.end.path == *path && r.end.off == off));
|
||||
if !paired && self.paired_with_table(a.story, path, off) {
|
||||
out.push((path.clone(), off));
|
||||
}
|
||||
}
|
||||
}
|
||||
// Sorted and without repeats: `delete_range` groups this by paragraph and binary-searches
|
||||
// each group (`control_markers_kept_are_sorted` pins it).
|
||||
out.sort();
|
||||
out.dedup();
|
||||
out
|
||||
}
|
||||
|
||||
/// (block, offset) of the paragraph markers in `path`'s container that are paired with a
|
||||
/// table wrap's open or close.
|
||||
fn table_paired_markers(&self, s: StoryRef, path: &Path) -> BTreeSet<(usize, usize)> {
|
||||
let Some(bl) = self.container(s, path) else { return BTreeSet::new() };
|
||||
/// Whether the marker at `path`/`off` is paired with a table wrap's open or close.
|
||||
fn paired_with_table(&self, s: StoryRef, path: &Path, off: usize) -> bool {
|
||||
let Some(bl) = self.container(s, path) else { return false };
|
||||
let events = control_events(bl);
|
||||
let (pairs, _) = pair_events(&events);
|
||||
let mut out = BTreeSet::new();
|
||||
for (o, c, _) in pairs {
|
||||
let (Some(open), Some(close)) = (events.get(o), events.get(c)) else { continue };
|
||||
match (open.off(), close.off()) {
|
||||
(Some(off), None) => {
|
||||
out.insert((open.block(), off));
|
||||
}
|
||||
(None, Some(off)) => {
|
||||
out.insert((close.block(), off));
|
||||
}
|
||||
_ => {}
|
||||
}
|
||||
}
|
||||
out
|
||||
let me = events.iter().position(|e| e.block() == path.last() && e.off() == Some(off));
|
||||
let Some(me) = me else { return false };
|
||||
pairs.iter().any(|(o, c, _)| {
|
||||
(*o == me && events.get(*c).is_some_and(|e| e.off().is_none())) || (*c == me && events.get(*o).is_some_and(|e| e.off().is_none()))
|
||||
})
|
||||
}
|
||||
|
||||
/// Where typing at `p` should go: a caret at the start of a paragraph, before the start of
|
||||
@@ -694,11 +674,8 @@ impl Document {
|
||||
/// end, goes inside it too (the control is around the whole paragraph).
|
||||
pub fn typing_pos(&self, p: &Pos) -> Pos {
|
||||
let Some(para) = self.para_at(p) else { return p.clone() };
|
||||
// The objects once, looked up by offset (object_at per step rescans the text).
|
||||
let objs: Vec<(usize, &InlineObject)> = para.objects_with_offsets().collect();
|
||||
let obj = |off: usize| objs.binary_search_by_key(&off, |(o, _)| *o).ok().and_then(|k| objs.get(k)).map(|(_, o)| *o);
|
||||
let is_start = |off: usize| matches!(obj(off), Some(InlineObject::ControlStart { control }) if control.block);
|
||||
let is_end = |off: usize| matches!(obj(off), Some(InlineObject::ControlEnd));
|
||||
let is_start = |off: usize| matches!(para.object_at(off), Some(InlineObject::ControlStart { control }) if control.block);
|
||||
let is_end = |off: usize| matches!(para.object_at(off), Some(InlineObject::ControlEnd));
|
||||
let w = OBJ.len_utf8();
|
||||
let mut off = p.off;
|
||||
// Only block starts before the caret (or none) and block starts right after it.
|
||||
@@ -711,9 +688,8 @@ impl Document {
|
||||
}
|
||||
if p.off == para.len() && off >= w && is_end(off - w) {
|
||||
// Step back over the trailing ends that close block-level controls.
|
||||
let block_ends: BTreeSet<usize> =
|
||||
self.control_ranges(p.story).into_iter().filter(|r| r.end.path == p.path && r.control.block).map(|r| r.end.off).collect();
|
||||
while off >= w && is_end(off - w) && block_ends.contains(&(off - w)) {
|
||||
let ranges = self.control_ranges(p.story);
|
||||
while off >= w && is_end(off - w) && ranges.iter().any(|r| r.end.path == p.path && r.end.off == off - w && r.control.block) {
|
||||
off -= w;
|
||||
}
|
||||
return Pos { off, ..p.clone() };
|
||||
|
||||
+19
-312
@@ -7,7 +7,7 @@ use std::sync::Arc;
|
||||
|
||||
use serde::{Deserialize, Serialize};
|
||||
|
||||
use crate::para::{InlineObject, Run};
|
||||
use crate::para::InlineObject;
|
||||
use crate::props::{CharProps, ParaProps};
|
||||
use crate::{Block, Blocks, DocError, Document, Paragraph, Part, PartKind, Path, Pos, Result, para_block};
|
||||
use crate::{Table, table::MAX_TABLE_DEPTH};
|
||||
@@ -188,111 +188,6 @@ fn fit_tables(b: Block, room: usize, out: &mut Vec<Block>) {
|
||||
const MAX_BOX_DEPTH: usize = 8;
|
||||
/// Most text box stories one paste may create (hostile documents can fan out).
|
||||
const MAX_PASTED_BOXES: usize = 4096;
|
||||
/// Most text box content (estimated bytes, see [`block_cost`]) one paste may copy: thousands of
|
||||
/// boxes can show the same story, and each needs a story of its own. Blocks without text boxes
|
||||
/// are shared between those stories, not copied, so they are not charged per box.
|
||||
const MAX_PASTED_BOX_BYTES: usize = 32 << 20;
|
||||
|
||||
/// What a paste may still spend on text box stories.
|
||||
struct BoxBudget {
|
||||
stories: usize,
|
||||
bytes: usize,
|
||||
/// The source stories being copied, innermost last: a box inside one of them showing it
|
||||
/// again is a cycle.
|
||||
active: Vec<u32>,
|
||||
/// Each source story's cost ([`block_cost`]), worked out once per paste: a story too big for
|
||||
/// what is left is not walked again for every box that shows it.
|
||||
costs: BTreeMap<u32, usize>,
|
||||
/// Boxes left without a story because the paste's budget couldn't hold theirs.
|
||||
skipped: usize,
|
||||
}
|
||||
|
||||
impl BoxBudget {
|
||||
fn new() -> BoxBudget {
|
||||
BoxBudget { stories: MAX_PASTED_BOXES, bytes: MAX_PASTED_BOX_BYTES, active: Vec::new(), costs: BTreeMap::new(), skipped: 0 }
|
||||
}
|
||||
}
|
||||
|
||||
/// Rough size of a block's content, and whether it holds text boxes (and so must be copied).
|
||||
fn block_cost(b: &Block) -> (usize, bool) {
|
||||
let (mut cost, mut boxes) = (std::mem::size_of::<Block>(), false);
|
||||
each_para(b, 0, &mut |p| {
|
||||
let items = p
|
||||
.runs
|
||||
.len()
|
||||
.saturating_mul(std::mem::size_of::<Run>())
|
||||
.saturating_add(p.objects.len().saturating_mul(std::mem::size_of::<InlineObject>()));
|
||||
let objects = p.objects.iter().fold(0usize, |n, o| n.saturating_add(object_cost(o)));
|
||||
cost = cost.saturating_add(p.text.len()).saturating_add(items).saturating_add(objects);
|
||||
boxes |= box_ids(p).next().is_some();
|
||||
});
|
||||
(cost, boxes)
|
||||
}
|
||||
|
||||
/// Rough heap bytes a clone of `o` copies beyond its own storage: its strings and lists, and its
|
||||
/// group members however deep. What it shares through `Arc` (pictures' embedded objects, charts,
|
||||
/// freeforms, wrap polygons, gradients) is not copied, so not counted.
|
||||
fn object_cost(o: &InlineObject) -> usize {
|
||||
let float = |f: &crate::para::Float| f.alt.text.len().saturating_add(f.alt.title.len());
|
||||
let mut cost = 0usize;
|
||||
let mut todo = vec![o];
|
||||
while let Some(o) = todo.pop() {
|
||||
let add = match o {
|
||||
InlineObject::Image { media, alt, float: f, .. } => media.len().saturating_add(alt.len()).saturating_add(float(f)),
|
||||
InlineObject::Graphic { alt, float: f, .. } => alt.len().saturating_add(float(f)),
|
||||
InlineObject::Shape { float: f, extra, .. } => {
|
||||
let warp = extra.warp.as_ref().map_or(0, |w| {
|
||||
let adj = w.adj.iter().map(|(name, _)| std::mem::size_of::<(String, i64)>().saturating_add(name.len()));
|
||||
adj.fold(w.preset.len(), usize::saturating_add)
|
||||
});
|
||||
float(f).saturating_add(warp)
|
||||
}
|
||||
InlineObject::Group { float: f, children, .. } => {
|
||||
todo.extend(children.iter().map(|c| &c.obj));
|
||||
children.len().saturating_mul(std::mem::size_of::<crate::para::GroupChild>()).saturating_add(float(f))
|
||||
}
|
||||
InlineObject::Field { instr, result, form, .. } => instr.len().saturating_add(result.len()).saturating_add(form.len()),
|
||||
InlineObject::FieldStart { instr, .. } => instr.len(),
|
||||
InlineObject::ControlStart { control } => std::mem::size_of::<crate::control::ContentControl>().saturating_add(serialized_len(control)),
|
||||
InlineObject::NoteRef { custom, .. } => custom.len(),
|
||||
InlineObject::BookmarkStart { name } | InlineObject::BookmarkEnd { name } => name.len(),
|
||||
InlineObject::PermStart { group, editor, .. } => group.len().saturating_add(editor.len()),
|
||||
InlineObject::Equation { linear, math, .. } => linear.len().saturating_add(math.heap_bytes()),
|
||||
InlineObject::Ruby { base, ruby, lang, props, .. } => {
|
||||
[base.len(), ruby.len(), lang.len(), std::mem::size_of::<CharProps>(), serialized_len(props)]
|
||||
.into_iter()
|
||||
.fold(0, usize::saturating_add)
|
||||
}
|
||||
InlineObject::Opaque { format, xml, text } => format.len().saturating_add(xml.len()).saturating_add(text.len()),
|
||||
InlineObject::FieldEnd
|
||||
| InlineObject::ControlEnd
|
||||
| InlineObject::CommentStart { .. }
|
||||
| InlineObject::CommentEnd { .. }
|
||||
| InlineObject::PermEnd { .. }
|
||||
| InlineObject::PositionalTab { .. } => 0,
|
||||
};
|
||||
cost = cost.saturating_add(add);
|
||||
}
|
||||
cost
|
||||
}
|
||||
|
||||
/// Length of `v` written as JSON: at least the bytes of every string and list it owns. For the
|
||||
/// nested formatting and control types, which share nothing through `Arc`.
|
||||
fn serialized_len<T: Serialize>(v: &T) -> usize {
|
||||
struct Count(usize);
|
||||
impl std::io::Write for Count {
|
||||
fn write(&mut self, b: &[u8]) -> std::io::Result<usize> {
|
||||
self.0 = self.0.saturating_add(b.len());
|
||||
Ok(b.len())
|
||||
}
|
||||
fn flush(&mut self) -> std::io::Result<()> {
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
let mut n = Count(0);
|
||||
let _ = serde_json::to_writer(&mut n, v);
|
||||
n.0
|
||||
}
|
||||
|
||||
/// Call `f` on every paragraph of a block, in document order, however deep its tables nest. The
|
||||
/// walk keeps its own stack instead of recursing, so it is complete: a text box it doesn't reach
|
||||
@@ -579,13 +474,7 @@ impl Document {
|
||||
}
|
||||
// Content control markers whose partner is outside the range stay (the control is
|
||||
// emptied, not broken).
|
||||
let mut kept = self.kept_control_markers(&a, &b);
|
||||
// The walk below groups `kept` by paragraph (`chunk_by`) and looks offsets up
|
||||
// (`binary_search`), so it must be sorted; `kept_control_markers` returns it so, and this
|
||||
// keeps a change there from quietly losing markers.
|
||||
if !kept.is_sorted() {
|
||||
kept.sort();
|
||||
}
|
||||
let kept = self.kept_control_markers(&a, &b);
|
||||
let kept_in = |path: &Path| -> Vec<usize> { kept.iter().filter(|(p, _)| p == path).map(|(_, o)| *o).collect() };
|
||||
if a.path == b.path {
|
||||
self.para_mut(a.story, &a.path)?.delete_except(a.off, b.off, &kept_in(&a.path))?;
|
||||
@@ -596,21 +485,13 @@ impl Document {
|
||||
let (ia, ib) = (a.path.last(), b.path.last());
|
||||
// Kept markers of the paragraphs in between move to the join.
|
||||
let mut moved: Vec<(InlineObject, CharProps)> = Vec::new();
|
||||
// `kept` is sorted: one walk of each paragraph's objects and runs (#3614).
|
||||
for group in kept.chunk_by(|x, y| x.0 == y.0) {
|
||||
let Some((path, _)) = group.first() else { continue };
|
||||
if *path == a.path || *path == b.path {
|
||||
continue;
|
||||
}
|
||||
let Some(p) = self.para(a.story, path) else { continue };
|
||||
let mut runs = p.run_ranges().peekable();
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
if group.binary_search_by_key(&off, |(_, k)| *k).is_err() {
|
||||
continue;
|
||||
}
|
||||
while runs.next_if(|(r, _)| r.end <= off).is_some() {}
|
||||
let props = runs.peek().filter(|(r, _)| r.contains(&off)).map_or(&p.mark, |(_, c)| *c);
|
||||
moved.push((o.clone(), props.clone()));
|
||||
for (path, off) in &kept {
|
||||
if *path != a.path
|
||||
&& *path != b.path
|
||||
&& let Some(p) = self.para(a.story, path)
|
||||
&& let Some(o) = p.object_at(*off)
|
||||
{
|
||||
moved.push((o.clone(), p.props_of_char(*off).clone()));
|
||||
}
|
||||
}
|
||||
{
|
||||
@@ -863,7 +744,7 @@ impl Document {
|
||||
/// Give every text box in `blocks` a new story of its own, copied from `parts` (or, for a
|
||||
/// fragment without them, from this document's text box of that id). A box whose story can't
|
||||
/// be found keeps no story rather than showing an unrelated one.
|
||||
fn adopt_text_boxes(&mut self, blocks: &mut [Block], parts: &BTreeMap<u32, Part>, depth: usize, budget: &mut BoxBudget) {
|
||||
fn adopt_text_boxes(&mut self, blocks: &mut [Block], parts: &BTreeMap<u32, Part>, depth: usize, budget: &mut usize) {
|
||||
let mut any = false;
|
||||
for b in blocks.iter() {
|
||||
each_para(b, 0, &mut |p| any |= box_ids(p).next().is_some());
|
||||
@@ -889,53 +770,18 @@ impl Document {
|
||||
}
|
||||
}
|
||||
|
||||
/// A new text box story copied from `old`; its own boxes get new stories too. Blocks without
|
||||
/// text boxes are shared with the source (copy-on-write), the others copied. A box showing a
|
||||
/// story it is inside, or whose story the paste's budget can't hold, gets none (CWE-400);
|
||||
/// only that box: a later box with a smaller story is still copied while the budget lasts.
|
||||
fn adopt_text_box(&mut self, old: u32, parts: &BTreeMap<u32, Part>, depth: usize, budget: &mut BoxBudget) -> Option<u32> {
|
||||
if depth >= MAX_BOX_DEPTH || budget.active.contains(&old) {
|
||||
return None;
|
||||
}
|
||||
if budget.stories == 0 {
|
||||
budget.skipped = budget.skipped.saturating_add(1);
|
||||
/// A new text box story copied from `old`; its own boxes get new stories too.
|
||||
fn adopt_text_box(&mut self, old: u32, parts: &BTreeMap<u32, Part>, depth: usize, budget: &mut usize) -> Option<u32> {
|
||||
if depth >= MAX_BOX_DEPTH || *budget == 0 {
|
||||
return None;
|
||||
}
|
||||
let src = parts.get(&old).or_else(|| self.parts.get(&old).filter(|p| p.kind == PartKind::TextBox))?;
|
||||
// Charge what the copy really allocates before making any of it: the block slots, and the
|
||||
// blocks holding text boxes (deep-copied below). The rest are shared `Arc`s, however many
|
||||
// boxes show them, so they cost nothing more. Worked out once per paste.
|
||||
let cost = *budget.costs.entry(old).or_insert_with(|| {
|
||||
let slots = src.blocks.len().saturating_mul(std::mem::size_of::<Arc<Block>>());
|
||||
src.blocks.iter().fold(slots, |a, b| match block_cost(b) {
|
||||
(c, true) => a.saturating_add(c),
|
||||
(_, false) => a,
|
||||
})
|
||||
});
|
||||
let Some(left) = budget.bytes.checked_sub(cost) else {
|
||||
budget.skipped = budget.skipped.saturating_add(1);
|
||||
return None;
|
||||
};
|
||||
budget.bytes = left;
|
||||
budget.stories -= 1;
|
||||
let src_blocks = src.blocks.clone();
|
||||
let mut blocks: Vec<Block> = src.blocks.iter().map(|b| (**b).clone()).collect();
|
||||
// The copy keeps its text direction and alignment, not its link (a copy stands alone).
|
||||
let body = crate::TextBody { next: None, ..src.body };
|
||||
budget.active.push(old);
|
||||
let mut blocks: Blocks = Vec::with_capacity(src_blocks.len());
|
||||
for b in src_blocks {
|
||||
let mut boxes = false;
|
||||
each_para(&b, 0, &mut |p| boxes |= box_ids(p).next().is_some());
|
||||
if boxes {
|
||||
let mut copy = (*b).clone();
|
||||
self.adopt_text_boxes(std::slice::from_mut(&mut copy), parts, depth + 1, budget);
|
||||
blocks.push(Arc::new(copy));
|
||||
} else {
|
||||
blocks.push(b);
|
||||
}
|
||||
}
|
||||
budget.active.pop();
|
||||
let id = self.try_add_part(PartKind::TextBox, blocks).ok()?;
|
||||
*budget -= 1;
|
||||
self.adopt_text_boxes(&mut blocks, parts, depth + 1, budget);
|
||||
let id = self.try_add_part(PartKind::TextBox, blocks.into_iter().map(Arc::new).collect()).ok()?;
|
||||
if let Some(p) = self.parts.get_mut(&id) {
|
||||
p.body = body;
|
||||
}
|
||||
@@ -957,15 +803,8 @@ impl Document {
|
||||
return Err(DocError::BadPath(pos.path.to_string()));
|
||||
}
|
||||
// Pasted text boxes get stories of their own (never shared with the copy's source).
|
||||
let mut budget = BoxBudget::new();
|
||||
let mut budget = MAX_PASTED_BOXES;
|
||||
self.adopt_text_boxes(&mut blocks, &frag.parts, 0, &mut budget);
|
||||
if budget.skipped > 0 {
|
||||
// Once per paste, not once per box (review of #3874).
|
||||
log::warn!(
|
||||
"paste: {} text boxes are left empty; their stories are past what one paste copies ({MAX_PASTED_BOXES} stories, {MAX_PASTED_BOX_BYTES} bytes)",
|
||||
budget.skipped
|
||||
);
|
||||
}
|
||||
// Single paragraph: inline insert keeping runs.
|
||||
if blocks.len() == 1
|
||||
&& let Some(Block::Para(src)) = blocks.first()
|
||||
@@ -1357,139 +1196,7 @@ mod tests {
|
||||
let before = d.parts.len();
|
||||
d.insert_fragment(&Pos::body(0, 0), &f).unwrap();
|
||||
let added = d.parts.len() - before;
|
||||
// The copy's inner box would show the story being copied: a cycle, so it gets none.
|
||||
assert_eq!(added, 1);
|
||||
}
|
||||
|
||||
/// A fragment whose body is one paragraph of `refs` text boxes all showing story 1.
|
||||
fn fan_out(story: Vec<Arc<Block>>, refs: usize) -> Fragment {
|
||||
let mut p = Paragraph::new();
|
||||
for _ in 0..refs {
|
||||
p.insert_object(0, box_shape(Some(1)), &CharProps::default()).unwrap();
|
||||
}
|
||||
let mut parts = BTreeMap::new();
|
||||
parts.insert(1, crate::Part { kind: PartKind::TextBox, blocks: story, body: Default::default() });
|
||||
Fragment { blocks: vec![Block::Para(p)], parts }
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn pasted_boxes_share_the_blocks_they_need_not_copy() {
|
||||
let big = para_block(Paragraph::with_text(&"x".repeat(100_000), CharProps::default()));
|
||||
let f = fan_out(vec![big.clone()], 3);
|
||||
let mut d = doc(&["x"]);
|
||||
d.insert_fragment(&Pos::body(0, 1), &f).unwrap();
|
||||
let got = boxes_in(&d, &Path::top(0));
|
||||
assert_eq!(got.len(), 3);
|
||||
let ids: Vec<u32> = got.into_iter().flatten().collect();
|
||||
assert_eq!(ids.len(), 3, "each box has a story");
|
||||
assert!(ids.windows(2).all(|w| w[0] != w[1]), "stories of their own: {ids:?}");
|
||||
for id in ids {
|
||||
let b = d.parts[&id].blocks.first().unwrap();
|
||||
assert!(Arc::ptr_eq(b, &big), "no text box inside: shared, not copied");
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_box_over_the_budget_leaves_later_smaller_boxes_their_stories() {
|
||||
// Review of #3874: once one story didn't fit, every later box in the paste was left
|
||||
// empty, however small its story. Only the box that doesn't fit is.
|
||||
// (Story 1 holds a text box, so each box showing it needs a real copy.)
|
||||
let mut bp = Paragraph::with_text(&"x".repeat(100_000), CharProps::default());
|
||||
bp.insert_object(0, box_shape(Some(999)), &CharProps::default()).unwrap();
|
||||
let big = para_block(bp);
|
||||
let small = para_block(Paragraph::with_text("small", CharProps::default()));
|
||||
let mut p = Paragraph::new();
|
||||
for story in [2, 1, 2, 1] {
|
||||
p.insert_object(0, box_shape(Some(story)), &CharProps::default()).unwrap();
|
||||
}
|
||||
let mut parts = BTreeMap::new();
|
||||
parts.insert(1, crate::Part { kind: PartKind::TextBox, blocks: vec![big], body: Default::default() });
|
||||
parts.insert(2, crate::Part { kind: PartKind::TextBox, blocks: vec![small], body: Default::default() });
|
||||
let mut blocks = vec![Block::Para(p)];
|
||||
let mut d = doc(&["x"]);
|
||||
let mut budget = BoxBudget { bytes: 50_000, ..BoxBudget::new() };
|
||||
d.adopt_text_boxes(&mut blocks, &parts, 0, &mut budget);
|
||||
let Some(Block::Para(p)) = blocks.first() else { panic!() };
|
||||
let got: Vec<Option<u32>> = p.objects.iter().map(|o| if let InlineObject::Shape { story, .. } = o { *story } else { None }).collect();
|
||||
assert_eq!(got.iter().map(Option::is_some).collect::<Vec<_>>(), [false, true, false, true], "{got:?}");
|
||||
assert_eq!(budget.skipped, 2);
|
||||
assert_eq!(budget.costs.len(), 2, "each story costed once");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn many_boxes_sharing_one_story_paste_with_their_text() {
|
||||
// Review of #3874: a document converted from PDF has thousands of boxes showing story
|
||||
// data they share. 64 boxes x a 1 MiB block is 64 MiB counted per box (over the 32 MiB
|
||||
// cap) but the block is shared, not copied, so every box keeps its text.
|
||||
let big = para_block(Paragraph::with_text(&"x".repeat(1 << 20), CharProps::default()));
|
||||
let f = fan_out(vec![big.clone()], 64);
|
||||
let mut d = doc(&["x"]);
|
||||
let (_, warned) = crate::test_log::warnings(|| d.insert_fragment(&Pos::body(0, 1), &f).unwrap());
|
||||
assert!(!warned.iter().any(|w| w.contains("text boxes are left empty")), "{warned:?}");
|
||||
let ids: Vec<u32> = boxes_in(&d, &Path::top(0)).into_iter().flatten().collect();
|
||||
assert_eq!(ids.len(), 64, "every box keeps a story");
|
||||
for id in ids {
|
||||
let b = d.parts[&id].blocks.first().unwrap();
|
||||
assert!(Arc::ptr_eq(b, &big), "shared, so not charged per box");
|
||||
assert_eq!(d.parts[&id].blocks.len(), 1);
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn pasted_box_copies_are_capped_in_bytes() {
|
||||
// Story 1 is a 1 MiB paragraph that holds a text box itself, so each box showing it needs
|
||||
// its own copy; 64 of them used to copy 64 MiB.
|
||||
let mut p = Paragraph::with_text(&"x".repeat(1 << 20), CharProps::default());
|
||||
p.insert_object(0, box_shape(Some(999)), &CharProps::default()).unwrap();
|
||||
let f = fan_out(vec![para_block(p)], 64);
|
||||
let mut d = doc(&["x"]);
|
||||
let before = d.parts.len();
|
||||
let (_, warned) = crate::test_log::warnings(|| d.insert_fragment(&Pos::body(0, 1), &f).unwrap());
|
||||
let added = d.parts.len() - before;
|
||||
// The boxes left empty are said to be, once for the paste (review of #3874).
|
||||
let empty: Vec<_> = warned.iter().filter(|w| w.contains("text boxes are left empty")).collect();
|
||||
assert_eq!(empty.len(), 1, "{warned:?}");
|
||||
assert!(empty[0].starts_with(&format!("paste: {} text boxes", 64 - added)), "{empty:?}");
|
||||
let copied: usize = d.parts.values().flat_map(|p| p.blocks.iter()).map(|b| if let Block::Para(p) = &**b { p.text.len() } else { 0 }).sum();
|
||||
assert!((1..64).contains(&added), "{added}");
|
||||
assert!(copied <= MAX_PASTED_BOX_BYTES, "{copied}");
|
||||
// The boxes past the cap keep no story rather than a shared one.
|
||||
let got = boxes_in(&d, &Path::top(0));
|
||||
assert_eq!(got.iter().flatten().count(), added);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn pasted_box_copies_count_the_objects_they_copy() {
|
||||
// #3614 review: a story holding a text box and a 1 MiB kept-for-round-trip object costs
|
||||
// the object's XML too, not only its slot: 64 copies used to copy 64 MiB of it.
|
||||
let mut p = Paragraph::with_text("x", CharProps::default());
|
||||
p.insert_object(0, box_shape(Some(999)), &CharProps::default()).unwrap();
|
||||
let opaque = InlineObject::Opaque { format: "docx".into(), xml: "x".repeat(1 << 20), text: String::new() };
|
||||
p.insert_object(0, opaque, &CharProps::default()).unwrap();
|
||||
let f = fan_out(vec![para_block(p)], 64);
|
||||
let mut d = doc(&["x"]);
|
||||
let before = d.parts.len();
|
||||
d.insert_fragment(&Pos::body(0, 1), &f).unwrap();
|
||||
let added = d.parts.len() - before;
|
||||
let xml = |o: &InlineObject| if let InlineObject::Opaque { xml, .. } = o { xml.len() } else { 0 };
|
||||
let copied: usize =
|
||||
d.parts.values().flat_map(|p| p.blocks.iter()).map(|b| if let Block::Para(p) = &**b { p.objects.iter().map(xml).sum() } else { 0 }).sum();
|
||||
assert!((1..64).contains(&added), "{added}");
|
||||
assert!(copied <= MAX_PASTED_BOX_BYTES, "{copied}");
|
||||
// Nested group members and equations count too.
|
||||
let member = |obj| crate::para::GroupChild { x: 0.0, y: 0.0, obj };
|
||||
let deep = InlineObject::Opaque { format: String::new(), xml: "y".repeat(5000), text: String::new() };
|
||||
let inner =
|
||||
InlineObject::Group { w: 1.0, h: 1.0, float: Default::default(), ch_w: 1.0, ch_h: 1.0, children: vec![member(deep)], canvas: None };
|
||||
let outer =
|
||||
InlineObject::Group { w: 1.0, h: 1.0, float: Default::default(), ch_w: 1.0, ch_h: 1.0, children: vec![member(inner)], canvas: None };
|
||||
assert!(object_cost(&outer) >= 5000);
|
||||
let eq = InlineObject::Equation {
|
||||
linear: String::new(),
|
||||
display: false,
|
||||
math: crate::math::Math { nodes: crate::math::parse_linear(&"z".repeat(5000)), ..Default::default() },
|
||||
};
|
||||
assert!(object_cost(&eq) >= 5000);
|
||||
assert!((1..=MAX_BOX_DEPTH).contains(&added), "{added}");
|
||||
}
|
||||
|
||||
#[test]
|
||||
|
||||
@@ -43,12 +43,12 @@ fn scan(story: StoryRef, bl: &Blocks, prefix: &mut Vec<u32>, depth: usize, out:
|
||||
match &**b {
|
||||
Block::Para(p) => {
|
||||
let path = Path(prefix.clone());
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
match o {
|
||||
InlineObject::FieldStart { instr, locked, .. } => {
|
||||
for off in p.object_offsets() {
|
||||
match p.object_at(off) {
|
||||
Some(InlineObject::FieldStart { instr, locked, .. }) => {
|
||||
open.push((instr.clone(), *locked, Pos::new(story, path.clone(), off)));
|
||||
}
|
||||
InlineObject::FieldEnd => match open.pop() {
|
||||
Some(InlineObject::FieldEnd) => match open.pop() {
|
||||
Some((instr, locked, start)) => {
|
||||
let depth = open.len();
|
||||
out.ranges.push(FieldRange { instr, locked, start, end: Pos::new(story, path.clone(), off), depth });
|
||||
@@ -99,10 +99,7 @@ impl Document {
|
||||
pub fn past_field_ends(&self, p: &Pos) -> Pos {
|
||||
let Some(para) = self.para_at(p) else { return p.clone() };
|
||||
let mut off = p.off;
|
||||
for (at, o) in para.objects_with_offsets().skip_while(|(at, _)| *at < p.off) {
|
||||
if at != off || !matches!(o, InlineObject::FieldEnd) {
|
||||
break;
|
||||
}
|
||||
while matches!(para.object_at(off), Some(InlineObject::FieldEnd)) {
|
||||
off += OBJ.len_utf8();
|
||||
}
|
||||
Pos { off, ..p.clone() }
|
||||
|
||||
@@ -1,712 +0,0 @@
|
||||
//! Visible logical text for IME surrounding-text queries.
|
||||
//!
|
||||
//! The platform input method works on what the reader sees, not on the stored run text: tracked
|
||||
//! deletions are hidden, and an inline object stands for its plain-text result. The helpers here
|
||||
//! build a bounded window around a selection without cloning the paragraph or the document, and
|
||||
//! map both ends in both directions so a query and a delete use exactly the same text.
|
||||
|
||||
use std::ops::Range;
|
||||
|
||||
use crate::para::{OBJ, Paragraph};
|
||||
use crate::{Document, Pos};
|
||||
|
||||
/// A bounded visible-logical window around a selection.
|
||||
#[derive(Clone, Debug, PartialEq, Eq)]
|
||||
pub struct ImeWindow {
|
||||
/// At most [`MAX_IME_TEXT_BYTES`] bytes of visible committed text.
|
||||
pub text: String,
|
||||
/// UTF-8 byte offset of the primary end in [`Self::text`].
|
||||
pub cursor: usize,
|
||||
/// UTF-8 byte offset of the other end in [`Self::text`].
|
||||
pub anchor: usize,
|
||||
}
|
||||
|
||||
/// The largest visible window an IME query copies.
|
||||
pub const MAX_IME_TEXT_BYTES: usize = 4000;
|
||||
|
||||
/// Which unit an input method's delete-surrounding counts use.
|
||||
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
|
||||
pub enum DeleteUnit {
|
||||
/// Unicode scalar values of the visible text.
|
||||
Chars,
|
||||
/// UTF-8 bytes of the visible text.
|
||||
Bytes,
|
||||
}
|
||||
|
||||
/// A deletion request in the paragraph's stored byte offsets.
|
||||
///
|
||||
/// The selection itself is never part of `before` or `after`; both ranges are outside it.
|
||||
#[derive(Clone, Debug, PartialEq, Eq)]
|
||||
pub struct ImeDelete {
|
||||
/// Stored byte range to erase before the selection.
|
||||
pub before: Option<Range<usize>>,
|
||||
/// Stored byte range to erase after the selection.
|
||||
pub after: Option<Range<usize>>,
|
||||
/// Stored byte offset of the selection's anchor **before** the deletions.
|
||||
///
|
||||
/// The ranges are returned in the paragraph's stored text as it is now; the caller applies
|
||||
/// them (and shifts these offsets past `before`) itself.
|
||||
pub anchor: usize,
|
||||
/// Stored byte offset of the selection's focus **before** the deletions.
|
||||
pub focus: usize,
|
||||
}
|
||||
|
||||
/// One paragraph's visible logical text.
|
||||
///
|
||||
/// Deleted runs are skipped and inline objects contribute their plain-text result. Iteration is
|
||||
/// lazy: no full visible copy is made.
|
||||
pub struct VisibleParagraph<'a> {
|
||||
para: &'a Paragraph,
|
||||
deleted: Vec<Range<usize>>,
|
||||
}
|
||||
|
||||
impl<'a> VisibleParagraph<'a> {
|
||||
/// A view of `para` with tracked deletions hidden.
|
||||
pub fn new(para: &'a Paragraph) -> Self {
|
||||
let mut deleted = para.deleted_ranges();
|
||||
deleted.sort_unstable_by_key(|r| r.start);
|
||||
Self { para, deleted }
|
||||
}
|
||||
|
||||
/// Whether `off` falls inside a tracked deletion, advancing `next` past the runs that end
|
||||
/// before it.
|
||||
///
|
||||
/// The ranges are sorted, so one monotone cursor makes a whole scan O(text + ranges) instead
|
||||
/// of O(text × ranges): alternating tracked revisions used to make every character scan every
|
||||
/// range.
|
||||
fn is_deleted_at(&self, next: &mut usize, off: usize) -> bool {
|
||||
while let Some(r) = self.deleted.get(*next) {
|
||||
if r.end > off {
|
||||
break;
|
||||
}
|
||||
*next = next.saturating_add(1);
|
||||
}
|
||||
self.deleted.get(*next).is_some_and(|r| r.start <= off && off < r.end)
|
||||
}
|
||||
|
||||
/// Call `f` for every visible piece: `(stored byte start, piece, stored byte len)`.
|
||||
fn for_each_piece(&self, mut f: impl FnMut(usize, &str, usize)) {
|
||||
self.for_each_piece_while(|start, piece, stored| {
|
||||
f(start, piece, stored);
|
||||
true
|
||||
});
|
||||
}
|
||||
|
||||
/// Call `f` for visible pieces until it returns `false`.
|
||||
fn for_each_piece_while(&self, mut f: impl FnMut(usize, &str, usize) -> bool) {
|
||||
let mut object = 0usize;
|
||||
let mut deleted = 0usize;
|
||||
for (i, c) in self.para.text.char_indices() {
|
||||
let is_object = c == OBJ;
|
||||
if is_object {
|
||||
object += 1;
|
||||
}
|
||||
if self.is_deleted_at(&mut deleted, i) {
|
||||
continue;
|
||||
}
|
||||
let keep_going = if is_object {
|
||||
let piece = self.para.objects.get(object - 1).map_or("", |o| o.plain_text());
|
||||
f(i, piece, OBJ.len_utf8())
|
||||
} else {
|
||||
let mut buf = [0u8; 4];
|
||||
f(i, c.encode_utf8(&mut buf), c.len_utf8())
|
||||
};
|
||||
if !keep_going {
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// Total visible UTF-8 bytes.
|
||||
pub fn visible_len(&self) -> usize {
|
||||
let mut len = 0usize;
|
||||
self.for_each_piece(|_, piece, _| len = len.saturating_add(piece.len()));
|
||||
len
|
||||
}
|
||||
|
||||
/// Visible offset of a stored paragraph byte offset.
|
||||
///
|
||||
/// `None` when the stored offset is not a character boundary, points inside a tracked
|
||||
/// deletion, or would split an inline object's plain-text result.
|
||||
pub fn visible_offset(&self, stored: usize) -> Option<usize> {
|
||||
if stored > self.para.text.len() || !self.para.text.is_char_boundary(stored) {
|
||||
return None;
|
||||
}
|
||||
let mut visible = 0usize;
|
||||
let mut found = false;
|
||||
let mut object = 0usize;
|
||||
let mut deleted = 0usize;
|
||||
for (i, c) in self.para.text.char_indices() {
|
||||
if i == stored {
|
||||
found = true;
|
||||
break;
|
||||
}
|
||||
if i > stored {
|
||||
break;
|
||||
}
|
||||
let is_object = c == OBJ;
|
||||
if is_object {
|
||||
object += 1;
|
||||
}
|
||||
if self.is_deleted_at(&mut deleted, i) {
|
||||
// A position at the start of a deleted run is still a visible position; one
|
||||
// inside it is not.
|
||||
continue;
|
||||
}
|
||||
if is_object {
|
||||
let piece = self.para.objects.get(object - 1).map_or("", |o| o.plain_text());
|
||||
visible = visible.saturating_add(piece.len());
|
||||
} else {
|
||||
visible = visible.saturating_add(c.len_utf8());
|
||||
}
|
||||
}
|
||||
if !found {
|
||||
// The end of the paragraph, or a deleted run at its end.
|
||||
let mut next = 0usize;
|
||||
return (!self.is_deleted_at(&mut next, stored)).then_some(visible);
|
||||
}
|
||||
if stored < self.para.text.len() && self.is_deleted_at(&mut deleted, stored) {
|
||||
// The start of a deleted run is still a visible caret position; only a position
|
||||
// strictly inside one is unrepresentable.
|
||||
let at_start = self.deleted.iter().any(|r| r.start == stored);
|
||||
return at_start.then_some(visible);
|
||||
}
|
||||
Some(visible)
|
||||
}
|
||||
|
||||
/// Stored paragraph byte offset for a visible offset.
|
||||
///
|
||||
/// `None` when the visible offset falls inside an inline object's plain-text result or is
|
||||
/// not a boundary of the visible representation.
|
||||
pub fn stored_offset(&self, visible: usize) -> Option<usize> {
|
||||
let mut acc = 0usize;
|
||||
let mut answer = None;
|
||||
let mut object = 0usize;
|
||||
let mut deleted = 0usize;
|
||||
for (i, c) in self.para.text.char_indices() {
|
||||
let is_object = c == OBJ;
|
||||
if is_object {
|
||||
object += 1;
|
||||
}
|
||||
if self.is_deleted_at(&mut deleted, i) {
|
||||
continue;
|
||||
}
|
||||
let piece_len = if is_object { self.para.objects.get(object - 1).map_or(0, |o| o.plain_text().len()) } else { c.len_utf8() };
|
||||
if visible == acc {
|
||||
answer = Some(i);
|
||||
break;
|
||||
}
|
||||
if visible < acc.saturating_add(piece_len) {
|
||||
// Inside a character (or an object's expansion): no stored boundary there.
|
||||
return None;
|
||||
}
|
||||
acc = acc.saturating_add(piece_len);
|
||||
}
|
||||
answer.or_else(|| (visible == acc).then_some(self.para.text.len()))
|
||||
}
|
||||
|
||||
/// Stored offset for the **start** of a visible range, skipping zero-length objects at
|
||||
/// the same boundary.
|
||||
///
|
||||
/// A content control's markers are inline objects with no plain text, so they share a
|
||||
/// visible offset with the content just inside (or after) them. A range start must map
|
||||
/// after a control-start marker and after a control-end marker, never onto the marker
|
||||
/// itself: an IME deletion must not break the control. Non-empty object expansions still
|
||||
/// map to the object's own stored offset, so a range that reaches one deletes it whole.
|
||||
pub fn stored_offset_after(&self, visible: usize) -> Option<usize> {
|
||||
let mut acc = 0usize;
|
||||
let mut object = 0usize;
|
||||
let mut deleted = 0usize;
|
||||
for (i, c) in self.para.text.char_indices() {
|
||||
let is_object = c == OBJ;
|
||||
if is_object {
|
||||
object += 1;
|
||||
}
|
||||
if self.is_deleted_at(&mut deleted, i) {
|
||||
continue;
|
||||
}
|
||||
let piece_len = if is_object { self.para.objects.get(object - 1).map_or(0, |o| o.plain_text().len()) } else { c.len_utf8() };
|
||||
if visible == acc && piece_len > 0 {
|
||||
return Some(i);
|
||||
}
|
||||
if visible < acc.saturating_add(piece_len) {
|
||||
return None;
|
||||
}
|
||||
acc = acc.saturating_add(piece_len);
|
||||
}
|
||||
(visible == acc).then_some(self.para.text.len())
|
||||
}
|
||||
|
||||
/// Stored offset for a visible offset, preferring the position **before** a tracked
|
||||
/// deletion that starts there. Use this for the end of a range so a deletion run just
|
||||
/// after the caret is never drawn into it.
|
||||
pub fn stored_offset_before(&self, visible: usize) -> Option<usize> {
|
||||
let mut acc = 0usize;
|
||||
let mut object = 0usize;
|
||||
let mut deleted = 0usize;
|
||||
for (i, c) in self.para.text.char_indices() {
|
||||
if visible == acc {
|
||||
return Some(i);
|
||||
}
|
||||
let is_object = c == OBJ;
|
||||
if is_object {
|
||||
object += 1;
|
||||
}
|
||||
if self.is_deleted_at(&mut deleted, i) {
|
||||
continue;
|
||||
}
|
||||
let piece_len = if is_object { self.para.objects.get(object - 1).map_or(0, |o| o.plain_text().len()) } else { c.len_utf8() };
|
||||
if visible < acc.saturating_add(piece_len) {
|
||||
return None;
|
||||
}
|
||||
acc = acc.saturating_add(piece_len);
|
||||
}
|
||||
(visible == acc).then_some(self.para.text.len())
|
||||
}
|
||||
|
||||
/// Round a visible byte offset forward to the next visible boundary (inward for a start).
|
||||
pub fn ceil_boundary(&self, visible: usize) -> usize {
|
||||
let mut acc = 0usize;
|
||||
let mut out = None;
|
||||
self.for_each_piece(|_, piece, _| {
|
||||
if out.is_some() {
|
||||
return;
|
||||
}
|
||||
if acc >= visible {
|
||||
out = Some(acc);
|
||||
return;
|
||||
}
|
||||
acc = acc.saturating_add(piece.len());
|
||||
if acc >= visible {
|
||||
out = Some(acc);
|
||||
}
|
||||
});
|
||||
out.unwrap_or(acc)
|
||||
}
|
||||
|
||||
/// Round a visible byte offset backward to the previous visible boundary (inward for an end).
|
||||
pub fn floor_boundary(&self, visible: usize) -> usize {
|
||||
let mut acc = 0usize;
|
||||
let mut out = 0usize;
|
||||
self.for_each_piece(|_, piece, _| {
|
||||
let next = acc.saturating_add(piece.len());
|
||||
if next > visible {
|
||||
return;
|
||||
}
|
||||
acc = next;
|
||||
out = acc;
|
||||
});
|
||||
out
|
||||
}
|
||||
|
||||
/// A bounded window with `cursor` and `anchor` as visible byte offsets.
|
||||
///
|
||||
/// The selection is never split and the window never exceeds [`MAX_IME_TEXT_BYTES`]. A
|
||||
/// selection wider than that, an invalid offset, or an offset that cannot be represented as
|
||||
/// visible text returns `None` (unavailable, not fabricated).
|
||||
pub fn window(&self, cursor: usize, anchor: usize, max_bytes: usize) -> Option<ImeWindow> {
|
||||
// The public API caps its own budget: a caller cannot ask for an unbounded copy.
|
||||
let max_bytes = max_bytes.min(MAX_IME_TEXT_BYTES);
|
||||
let (sel_start, sel_end) = (cursor.min(anchor), cursor.max(anchor));
|
||||
let selection_len = sel_end - sel_start;
|
||||
if selection_len > max_bytes {
|
||||
return None;
|
||||
}
|
||||
// Spend the budget around the selection and stop walking at its end: the paragraph is
|
||||
// never copied and the work is bounded by the caret's distance from the paragraph start.
|
||||
let before = (max_bytes - selection_len) / 2;
|
||||
let window_start = sel_start.saturating_sub(before);
|
||||
let window_end = window_start.saturating_add(max_bytes);
|
||||
let mut text = String::with_capacity(window_end.saturating_sub(window_start).min(max_bytes));
|
||||
let mut first_kept = None;
|
||||
let mut acc = 0usize;
|
||||
self.for_each_piece_while(|_, piece, _| {
|
||||
for (k, c) in piece.char_indices() {
|
||||
let start = acc.saturating_add(k);
|
||||
let end = start.saturating_add(c.len_utf8());
|
||||
if end > window_end {
|
||||
return false;
|
||||
}
|
||||
if start >= window_start {
|
||||
if first_kept.is_none() {
|
||||
first_kept = Some(start);
|
||||
}
|
||||
text.push(c);
|
||||
}
|
||||
}
|
||||
acc = acc.saturating_add(piece.len());
|
||||
acc < window_end
|
||||
});
|
||||
let Some(first) = first_kept else {
|
||||
// An empty document is an available empty context, never unavailable.
|
||||
return (cursor == 0 && anchor == 0).then_some(ImeWindow { text: String::new(), cursor: 0, anchor: 0 });
|
||||
};
|
||||
Some(ImeWindow { text, cursor: cursor.saturating_sub(first), anchor: anchor.saturating_sub(first) })
|
||||
}
|
||||
}
|
||||
|
||||
impl Document {
|
||||
/// Bounded visible logical text around `cursor`/`anchor` in one paragraph.
|
||||
///
|
||||
/// A selection across paragraphs or blocks is unrepresentable and returns `None`.
|
||||
pub fn ime_surrounding(&self, cursor: &Pos, anchor: &Pos, max_bytes: usize) -> Option<ImeWindow> {
|
||||
if cursor.story != anchor.story || cursor.path != anchor.path {
|
||||
return None;
|
||||
}
|
||||
let para = self.para_at(cursor)?;
|
||||
let view = VisibleParagraph::new(para);
|
||||
let c = view.visible_offset(cursor.off)?;
|
||||
let a = view.visible_offset(anchor.off)?;
|
||||
view.window(c, a, max_bytes.min(MAX_IME_TEXT_BYTES))
|
||||
}
|
||||
|
||||
/// Delete-surrounding ranges in the stored paragraph text.
|
||||
///
|
||||
/// Counts exclude the selection, tracked deletions and any pre-edit (which never lives in
|
||||
/// the model here). Byte counts that end inside a scalar leave it whole: a start rounds
|
||||
/// forward and an end rounds backward; a byte count beyond the visible text is refused.
|
||||
/// Character counts clamp at the paragraph's ends, as the web text agent's do. A range that
|
||||
/// cannot be represented in stored text returns `None` instead of guessing; so does a
|
||||
/// request with nothing to erase (`0` on both sides).
|
||||
pub fn ime_delete_surrounding(&self, cursor: &Pos, anchor: &Pos, before: usize, after: usize, unit: DeleteUnit) -> Option<ImeDelete> {
|
||||
if cursor.story != anchor.story || cursor.path != anchor.path {
|
||||
return None;
|
||||
}
|
||||
if before == 0 && after == 0 {
|
||||
// Nothing to erase: don't build the boundary table (or walk a long paragraph) to
|
||||
// discover it.
|
||||
return None;
|
||||
}
|
||||
let para = self.para_at(cursor)?;
|
||||
let view = VisibleParagraph::new(para);
|
||||
let c = view.visible_offset(cursor.off)?;
|
||||
let a = view.visible_offset(anchor.off)?;
|
||||
let (sel_start, sel_end) = (c.min(a), c.max(a));
|
||||
|
||||
let (before_start, after_end) = match unit {
|
||||
DeleteUnit::Bytes => {
|
||||
let total = view.visible_len();
|
||||
if before > sel_start || after > total - sel_end {
|
||||
return None;
|
||||
}
|
||||
(view.ceil_boundary(sel_start.saturating_sub(before)), view.floor_boundary(sel_end.saturating_add(after)))
|
||||
}
|
||||
DeleteUnit::Chars => char_boundaries(&view, sel_start, sel_end, before, after)?,
|
||||
};
|
||||
|
||||
// Start a range after a deletion run and end it before one, so tracked deletions are
|
||||
// never part of the erased stored range.
|
||||
let before_start = view.stored_offset_after(before_start)?;
|
||||
let before_end = view.stored_offset_before(sel_start)?;
|
||||
let after_start = view.stored_offset_after(sel_end)?;
|
||||
let after_end = view.stored_offset_before(after_end)?;
|
||||
let before_range = (before_start < before_end).then_some(before_start..before_end);
|
||||
let after_range = (after_start < after_end).then_some(after_start..after_end);
|
||||
Some(ImeDelete { before: before_range, after: after_range, anchor: anchor.off, focus: cursor.off })
|
||||
}
|
||||
}
|
||||
|
||||
/// The visible byte boundaries `before` characters back from `sel_start` and `after`
|
||||
/// characters on from `sel_end`.
|
||||
///
|
||||
/// Two walks and constant memory replace the old per-character `Vec` that held every boundary
|
||||
/// (a hostile count could make it allocate the whole paragraph). Character counts clamp at the
|
||||
/// paragraph's ends; `None` when either selection end is not a boundary of the visible text.
|
||||
fn char_boundaries(view: &VisibleParagraph<'_>, sel_start: usize, sel_end: usize, before: usize, after: usize) -> Option<(usize, usize)> {
|
||||
let mut acc = 0usize;
|
||||
let mut chars = 0usize;
|
||||
let mut start_index = None;
|
||||
let mut end_index = None;
|
||||
view.for_each_piece(|_, piece, _| {
|
||||
for (k, _) in piece.char_indices() {
|
||||
let at = acc.saturating_add(k);
|
||||
if at == sel_start {
|
||||
start_index = Some(chars);
|
||||
}
|
||||
if at == sel_end {
|
||||
end_index = Some(chars);
|
||||
}
|
||||
chars = chars.saturating_add(1);
|
||||
}
|
||||
acc = acc.saturating_add(piece.len());
|
||||
});
|
||||
if sel_start == acc {
|
||||
start_index = Some(chars);
|
||||
}
|
||||
if sel_end == acc {
|
||||
end_index = Some(chars);
|
||||
}
|
||||
let start_index = start_index?;
|
||||
let end_index = end_index?;
|
||||
let target_before = start_index.saturating_sub(before);
|
||||
let target_after = end_index.saturating_add(after).min(chars);
|
||||
|
||||
let mut acc = 0usize;
|
||||
let mut index = 0usize;
|
||||
let mut before_start = None;
|
||||
let mut after_end = (target_after == 0).then_some(0);
|
||||
view.for_each_piece(|_, piece, _| {
|
||||
for (k, _) in piece.char_indices() {
|
||||
let at = acc.saturating_add(k);
|
||||
if index == target_before && before_start.is_none() {
|
||||
before_start = Some(at);
|
||||
}
|
||||
if index == target_after && after_end.is_none() {
|
||||
after_end = Some(at);
|
||||
}
|
||||
index = index.saturating_add(1);
|
||||
}
|
||||
acc = acc.saturating_add(piece.len());
|
||||
});
|
||||
if index == target_before && before_start.is_none() {
|
||||
before_start = Some(acc);
|
||||
}
|
||||
if index == target_after && after_end.is_none() {
|
||||
after_end = Some(acc);
|
||||
}
|
||||
Some((before_start?, after_end?))
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::*;
|
||||
use crate::para::Run;
|
||||
use crate::props::CharProps;
|
||||
use crate::{InlineObject, StoryRef};
|
||||
|
||||
/// Builds a paragraph with one run per `(text, deleted)` part; deleted parts carry revision 0.
|
||||
fn para_with_runs(parts: &[(&str, bool)]) -> Paragraph {
|
||||
let mut p = Paragraph::new();
|
||||
for (text, deleted) in parts {
|
||||
let mut props = CharProps::default();
|
||||
if *deleted {
|
||||
props.del = Some(0);
|
||||
}
|
||||
p.text.push_str(text);
|
||||
p.runs.push(Run { len: text.len(), props });
|
||||
}
|
||||
p
|
||||
}
|
||||
|
||||
/// A tracked deletion is hidden from the IME view but still maps stored offsets: its start is a
|
||||
/// visible caret position, its interior is not, and a window concatenates only visible text.
|
||||
#[test]
|
||||
fn tracked_deletions_are_hidden_and_offsets_still_map() {
|
||||
let p = para_with_runs(&[("ab", false), ("DEL", true), ("cd", false)]);
|
||||
let view = VisibleParagraph::new(&p);
|
||||
assert_eq!(view.visible_len(), 4);
|
||||
assert_eq!(view.visible_offset(0), Some(0));
|
||||
assert_eq!(view.visible_offset(2), Some(2), "the start of a deleted run is a visible caret position");
|
||||
assert_eq!(view.visible_offset(3), None, "inside a tracked deletion");
|
||||
assert_eq!(view.visible_offset(5), Some(2));
|
||||
assert_eq!(view.visible_offset(7), Some(4));
|
||||
assert_eq!(view.stored_offset(2), Some(5));
|
||||
assert_eq!(view.stored_offset(4), Some(7));
|
||||
assert_eq!(view.window(2, 2, 4000).map(|w| w.text), Some("abcd".into()));
|
||||
}
|
||||
|
||||
/// An inline object contributes its plain text to visible length and boundary mapping; offsets
|
||||
/// strictly inside that expansion have no stored boundary.
|
||||
#[test]
|
||||
fn objects_contribute_their_plain_text_and_split_mapping_stops() {
|
||||
let mut p = Paragraph::new();
|
||||
p.text = format!("a{OBJ}b");
|
||||
p.runs = vec![Run { len: p.text.len(), props: CharProps::default() }];
|
||||
p.objects.push(InlineObject::Equation { linear: "x+y".into(), display: false, math: Default::default() });
|
||||
let view = VisibleParagraph::new(&p);
|
||||
assert_eq!(view.visible_len(), 5);
|
||||
assert_eq!(view.visible_offset(1), Some(1));
|
||||
assert_eq!(view.visible_offset(4), Some(4));
|
||||
assert_eq!(view.stored_offset(1), Some(1));
|
||||
assert_eq!(view.stored_offset(4), Some(4));
|
||||
assert_eq!(view.stored_offset(2), None, "inside the object's text");
|
||||
}
|
||||
|
||||
/// An empty paragraph yields an available empty IME context at offset 0; an empty document
|
||||
/// yields an available empty context too, though this test asserts only its text, not its offsets.
|
||||
#[test]
|
||||
fn an_empty_document_is_available_and_empty() {
|
||||
let p = Paragraph::new();
|
||||
let view = VisibleParagraph::new(&p);
|
||||
let w = view.window(0, 0, MAX_IME_TEXT_BYTES).expect("an available empty context");
|
||||
assert_eq!((w.text.as_str(), w.cursor, w.anchor), ("", 0, 0));
|
||||
let doc = Document::new();
|
||||
let w = doc.ime_surrounding(&Pos::body(0, 0), &Pos::body(0, 0), MAX_IME_TEXT_BYTES).expect("available");
|
||||
assert_eq!(w.text, "");
|
||||
}
|
||||
|
||||
/// A window is capped at the IME byte budget while keeping cursor and anchor inside its text; a
|
||||
/// selection wider than the budget is unavailable.
|
||||
#[test]
|
||||
fn bounded_window_keeps_the_selection_and_rejects_too_wide_ones() {
|
||||
let p = Paragraph::with_text(&"a".repeat(10_000), CharProps::default());
|
||||
let view = VisibleParagraph::new(&p);
|
||||
let w = view.window(5000, 5000, MAX_IME_TEXT_BYTES).expect("a bounded window");
|
||||
assert!(w.text.len() <= MAX_IME_TEXT_BYTES);
|
||||
assert_eq!(w.cursor, w.anchor);
|
||||
assert!(w.cursor <= w.text.len());
|
||||
assert!(view.window(0, 4001, MAX_IME_TEXT_BYTES).is_none());
|
||||
}
|
||||
|
||||
/// Byte deletion rounds its start to a visible UTF-8 boundary, leaves partially covered scalar
|
||||
/// values whole, and refuses counts that run past the paragraph.
|
||||
#[test]
|
||||
fn byte_deletion_rounds_inward_and_leaves_partial_scalars_whole() {
|
||||
let doc = Document::from_text("aé👍b");
|
||||
let caret = Pos::body(0, 8);
|
||||
let one = doc.ime_delete_surrounding(&caret, &caret, 1, 0, DeleteUnit::Bytes).expect("b");
|
||||
assert_eq!(one.before, Some(7..8));
|
||||
// Two bytes start inside the emoji: it is left whole, only `b` goes.
|
||||
let two = doc.ime_delete_surrounding(&caret, &caret, 2, 0, DeleteUnit::Bytes).expect("b");
|
||||
assert_eq!(two.before, Some(7..8));
|
||||
// Six bytes start inside `é`: it stays whole, the emoji and `b` go.
|
||||
let six = doc.ime_delete_surrounding(&caret, &caret, 6, 0, DeleteUnit::Bytes).expect("emoji and b");
|
||||
assert_eq!(six.before, Some(3..8));
|
||||
// A count beyond the paragraph is refused, not clipped to delete a whole side.
|
||||
assert!(doc.ime_delete_surrounding(&caret, &caret, 0, 1, DeleteUnit::Bytes).is_none());
|
||||
assert!(doc.ime_delete_surrounding(&caret, &caret, 99, 0, DeleteUnit::Bytes).is_none());
|
||||
}
|
||||
|
||||
/// Surrounding deletion skips the selected range on both sides and returns the selection's
|
||||
/// original reversed anchor and focus.
|
||||
#[test]
|
||||
fn byte_deletion_excludes_a_reversed_selection() {
|
||||
let doc = Document::from_text("abcd");
|
||||
let cursor = Pos::body(0, 1);
|
||||
let anchor = Pos::body(0, 3);
|
||||
let del = doc.ime_delete_surrounding(&cursor, &anchor, 1, 1, DeleteUnit::Bytes).expect("a deletion");
|
||||
assert_eq!(del.before, Some(0..1), "only the character before the selection");
|
||||
assert_eq!(del.after, Some(3..4), "only the character after the selection");
|
||||
assert_eq!((del.anchor, del.focus), (3, 1), "the reversed selection is preserved");
|
||||
}
|
||||
|
||||
/// Tracked deletions are absent from windows and skipped when deleting around a caret, so an IME
|
||||
/// edit cannot remove them.
|
||||
#[test]
|
||||
fn tracked_deletions_are_not_counted_for_deletion_or_query() {
|
||||
let mut doc = Document::from_text("abDELcd");
|
||||
let p = para_with_runs(&[("ab", false), ("DEL", true), ("cd", false)]);
|
||||
doc.body = vec![std::sync::Arc::new(crate::Block::Para(p))];
|
||||
let view = VisibleParagraph::new(doc.para_at(&Pos::body(0, 0)).expect("paragraph"));
|
||||
assert_eq!(view.window(2, 2, 100).map(|w| w.text), Some("abcd".into()));
|
||||
let caret = Pos::body(0, 7);
|
||||
let del = doc.ime_delete_surrounding(&caret, &caret, 1, 0, DeleteUnit::Bytes).expect("c");
|
||||
assert_eq!(del.before, Some(6..7), "the stored deletion run is skipped");
|
||||
}
|
||||
|
||||
/// A selection spanning paragraphs has no single surrounding context: both the query and
|
||||
/// surrounding deletion return `None`.
|
||||
#[test]
|
||||
fn cross_paragraph_selection_is_unavailable() {
|
||||
let doc = Document::from_text("one\ntwo");
|
||||
let cursor = Pos::body(0, 1);
|
||||
let anchor = Pos::body(1, 1);
|
||||
assert!(doc.ime_surrounding(&cursor, &anchor, MAX_IME_TEXT_BYTES).is_none());
|
||||
assert!(doc.ime_delete_surrounding(&cursor, &anchor, 1, 0, DeleteUnit::Chars).is_none());
|
||||
}
|
||||
|
||||
/// An object's plain-text result stands for the object in the visible text; a deletion
|
||||
/// that starts inside that expansion leaves the object whole (byte) or is refused (chars).
|
||||
#[test]
|
||||
fn object_expansion_is_never_partly_deleted() {
|
||||
let mut doc = Document::from_text("");
|
||||
let mut p = Paragraph::new();
|
||||
p.text = format!("a{OBJ}b");
|
||||
p.runs = vec![Run { len: p.text.len(), props: CharProps::default() }];
|
||||
p.objects.push(InlineObject::Equation { linear: "x+y".into(), display: false, math: Default::default() });
|
||||
doc.body = vec![std::sync::Arc::new(crate::Block::Para(p))];
|
||||
let caret = Pos::body(0, 5);
|
||||
// Two bytes before the caret end inside `x+y` (`y` is the last byte): the start rounds
|
||||
// forward past the object, so only `b` goes.
|
||||
let two = doc.ime_delete_surrounding(&caret, &caret, 2, 0, DeleteUnit::Bytes).expect("b only");
|
||||
assert_eq!(two.before, Some(4..5), "stored UTF-8 offsets keep the object whole");
|
||||
let mut remaining = format!("a{OBJ}b");
|
||||
remaining.replace_range(two.before.clone().unwrap(), "");
|
||||
assert_eq!(remaining, format!("a{OBJ}"));
|
||||
// A character that lands inside the expansion has no stored boundary: refused.
|
||||
assert!(doc.ime_delete_surrounding(&caret, &caret, 2, 0, DeleteUnit::Chars).is_none());
|
||||
// A count reaching before the object erases the object as a whole.
|
||||
let five = doc.ime_delete_surrounding(&caret, &caret, 5, 0, DeleteUnit::Bytes).expect("object and b");
|
||||
assert_eq!(five.before, Some(0..5));
|
||||
}
|
||||
|
||||
/// Character counts on the right side clamp at the paragraph's end and never cross into the
|
||||
/// next paragraph.
|
||||
#[test]
|
||||
fn trailing_character_deletion_clamps_at_the_paragraph_end() {
|
||||
let doc = Document::from_text("ab\ncd");
|
||||
let caret = Pos::body(0, 1);
|
||||
let del = doc.ime_delete_surrounding(&caret, &caret, 0, 99, DeleteUnit::Chars).expect("b only");
|
||||
assert_eq!(del.after, Some(1..2));
|
||||
let empty = doc.ime_delete_surrounding(&caret, &caret, 0, 0, DeleteUnit::Chars);
|
||||
assert!(empty.is_none(), "nothing to erase");
|
||||
}
|
||||
|
||||
/// Sorted tracked deletions are scanned with one cursor: alternating revisions stay correct
|
||||
/// (and linear) instead of each character scanning every range.
|
||||
#[test]
|
||||
fn alternating_tracked_revisions_map_correctly() {
|
||||
// Equal-length visible/deleted pieces, so the expected visible byte count is exact.
|
||||
let parts: Vec<(String, bool)> =
|
||||
(0..2000).map(|i| (if i % 2 == 0 { format!("v{i:04};") } else { format!("D{i:04};") }, i % 2 == 1)).collect();
|
||||
let parts: Vec<(&str, bool)> = parts.iter().map(|(t, d)| (t.as_str(), *d)).collect();
|
||||
let p = para_with_runs(&parts);
|
||||
let view = VisibleParagraph::new(&p);
|
||||
assert_eq!(view.visible_len(), p.text.len() - 6 * 1000);
|
||||
assert_eq!(view.visible_offset(p.text.len()), Some(view.visible_len()));
|
||||
let w = view.window(0, 0, 100).expect("a window");
|
||||
assert!(w.text.len() <= 100);
|
||||
assert!(!w.text.contains('D'), "tracked deletions never show");
|
||||
}
|
||||
|
||||
/// The public window never copies more than the IME cap, whatever a caller asks for.
|
||||
#[test]
|
||||
fn window_budget_is_capped() {
|
||||
let p = Paragraph::with_text(&"a".repeat(10_000), CharProps::default());
|
||||
let view = VisibleParagraph::new(&p);
|
||||
let w = view.window(0, 0, usize::MAX).expect("a bounded window");
|
||||
assert_eq!(w.text.len(), MAX_IME_TEXT_BYTES);
|
||||
let doc = Document::from_text(&"a".repeat(10_000));
|
||||
let w = doc.ime_surrounding(&Pos::body(0, 0), &Pos::body(0, 0), usize::MAX).expect("bounded");
|
||||
assert_eq!(w.text.len(), MAX_IME_TEXT_BYTES);
|
||||
}
|
||||
|
||||
/// A caret in a table-cell paragraph maps against that cell's text, not the body around the table.
|
||||
#[test]
|
||||
fn table_cell_paragraph_maps_inside_its_cell() {
|
||||
let mut doc = Document::from_text("outside");
|
||||
let mut table = crate::Table::new(1, 1, 100.0);
|
||||
table.rows[0].cells[0].blocks = vec![crate::para_block(Paragraph::with_text("cell", CharProps::default()))];
|
||||
doc.body.push(std::sync::Arc::new(crate::Block::Table(table)));
|
||||
let path = crate::Path(vec![1, 0, 0, 0]);
|
||||
let cursor = Pos { story: StoryRef::Body, path: path.clone(), off: 4 };
|
||||
let anchor = Pos { story: StoryRef::Body, path, off: 2 };
|
||||
let w = doc.ime_surrounding(&cursor, &anchor, MAX_IME_TEXT_BYTES).expect("the cell text");
|
||||
assert_eq!(w.text, "cell");
|
||||
assert_eq!(w.cursor, 4);
|
||||
assert_eq!(w.anchor, 2);
|
||||
}
|
||||
|
||||
/// Content-control markers are zero-length visible objects. A deletion range starts after
|
||||
/// the marker and ends before the next one, so an ordinary IME edit never removes a marker
|
||||
/// or leaves a half-paired control.
|
||||
#[test]
|
||||
fn control_markers_are_not_included_in_delete_ranges() {
|
||||
use crate::control::{ContentControl, ControlKind};
|
||||
let mut p = Paragraph::new();
|
||||
p.text = format!("pre{OBJ}Ada{OBJ}X");
|
||||
p.runs = vec![Run { len: p.text.len(), props: CharProps::default() }];
|
||||
let mut ctl = ContentControl::new(ControlKind::RichText);
|
||||
ctl.placeholder_text = "Enter name".into();
|
||||
p.objects.push(InlineObject::ControlStart { control: Box::new(ctl) });
|
||||
p.objects.push(InlineObject::ControlEnd);
|
||||
let mut doc = Document::new();
|
||||
doc.body = vec![std::sync::Arc::new(crate::Block::Para(p))];
|
||||
// Stored positions are UTF-8 byte offsets; each OBJ marker occupies three bytes.
|
||||
let cursor = Pos::body(0, 9);
|
||||
let del = doc.ime_delete_surrounding(&cursor, &cursor, 3, 1, DeleteUnit::Bytes).expect("a deletion");
|
||||
assert_eq!(del.before, Some(6..9), "the control-start marker and `pre` stay outside");
|
||||
assert_eq!(del.after, Some(12..13), "the control-end marker stays outside; only `X` goes");
|
||||
assert_eq!((del.anchor, del.focus), (9, 9));
|
||||
}
|
||||
}
|
||||
+24
-249
@@ -12,7 +12,6 @@
|
||||
|
||||
pub mod bidi;
|
||||
mod byte_store;
|
||||
pub mod cfb_guard;
|
||||
pub mod chart;
|
||||
mod cjk_numbers;
|
||||
pub mod cluster;
|
||||
@@ -27,7 +26,6 @@ pub mod fields;
|
||||
pub mod fill;
|
||||
pub mod freeform;
|
||||
pub mod graphic;
|
||||
pub mod ime_text;
|
||||
pub mod legacy_form;
|
||||
pub mod line_dash;
|
||||
pub mod link;
|
||||
@@ -181,95 +179,6 @@ pub fn para_block(p: Paragraph) -> Arc<Block> {
|
||||
Arc::new(Block::Para(p))
|
||||
}
|
||||
|
||||
/// The same blocks: the same content. The blocks are compared by address first — the
|
||||
/// document is copy-on-write, so a block that changed was copied and has another address —
|
||||
/// and a block that was copied but holds the same content is compared by value.
|
||||
fn same_blocks(a: &Blocks, b: &Blocks) -> bool {
|
||||
a.len() == b.len() && a.iter().zip(b).all(|(x, y)| Arc::ptr_eq(x, y) || same_block(x, y))
|
||||
}
|
||||
|
||||
/// The same block by value. A table's cells compare their blocks by address first too: an edit in
|
||||
/// one cell copies the whole table, which mustn't mean comparing every cell by value.
|
||||
fn same_block(a: &Block, b: &Block) -> bool {
|
||||
match (a, b) {
|
||||
(Block::Table(x), Block::Table(y)) => same_table(x, y),
|
||||
_ => a == b,
|
||||
}
|
||||
}
|
||||
|
||||
/// The same table. Every field is listed, so a new one does not build until someone decides
|
||||
/// whether a change to it counts.
|
||||
fn same_table(a: &Table, b: &Table) -> bool {
|
||||
let Table { props, grid, rows, controls } = a;
|
||||
let Table { props: other_props, grid: other_grid, rows: other_rows, controls: other_controls } = b;
|
||||
props == other_props
|
||||
&& grid == other_grid
|
||||
&& controls == other_controls
|
||||
&& rows.len() == other_rows.len()
|
||||
&& rows.iter().zip(other_rows).all(|(x, y)| same_row(x, y))
|
||||
}
|
||||
|
||||
fn same_row(a: &Row, b: &Row) -> bool {
|
||||
let Row { props, cells, controls } = a;
|
||||
let Row { props: other_props, cells: other_cells, controls: other_controls } = b;
|
||||
props == other_props
|
||||
&& controls == other_controls
|
||||
&& cells.len() == other_cells.len()
|
||||
&& cells.iter().zip(other_cells).all(|(x, y)| {
|
||||
let Cell { props, blocks, controls } = x;
|
||||
let Cell { props: other_props, blocks: other_blocks, controls: other_controls } = y;
|
||||
props == other_props && controls == other_controls && same_blocks(blocks, other_blocks)
|
||||
})
|
||||
}
|
||||
|
||||
/// The same building blocks, their content compared by address first.
|
||||
fn same_glossary(a: &[BuildingBlock], b: &[BuildingBlock]) -> bool {
|
||||
a.len() == b.len()
|
||||
&& a.iter().zip(b).all(|(x, y)| {
|
||||
let BuildingBlock { name, gallery, category, description, guid, style, types, behaviors, blocks } = x;
|
||||
let BuildingBlock {
|
||||
name: other_name,
|
||||
gallery: other_gallery,
|
||||
category: other_category,
|
||||
description: other_description,
|
||||
guid: other_guid,
|
||||
style: other_style,
|
||||
types: other_types,
|
||||
behaviors: other_behaviors,
|
||||
blocks: other_blocks,
|
||||
} = y;
|
||||
name == other_name
|
||||
&& gallery == other_gallery
|
||||
&& category == other_category
|
||||
&& description == other_description
|
||||
&& guid == other_guid
|
||||
&& style == other_style
|
||||
&& types == other_types
|
||||
&& behaviors == other_behaviors
|
||||
&& same_blocks(blocks, other_blocks)
|
||||
})
|
||||
}
|
||||
|
||||
/// The same parts: the same keys, and each part the same.
|
||||
fn same_parts(a: &BTreeMap<u32, Part>, b: &BTreeMap<u32, Part>) -> bool {
|
||||
a.len() == b.len() && a.iter().zip(b).all(|((k1, p1), (k2, p2))| k1 == k2 && same_part(p1, p2))
|
||||
}
|
||||
|
||||
/// The same part: the same kind, text body and blocks. Every field is listed, so a new one
|
||||
/// does not build until someone decides whether a change to it counts.
|
||||
fn same_part(a: &Part, b: &Part) -> bool {
|
||||
let Part { kind, blocks, body } = a;
|
||||
let Part { kind: other_kind, blocks: other_blocks, body: other_body } = b;
|
||||
kind == other_kind && body == other_body && same_blocks(blocks, other_blocks)
|
||||
}
|
||||
|
||||
/// The same bytes: the same content. The bytes are compared by address first — a changed
|
||||
/// picture was copied, so it has another address — and a picture that was copied but holds
|
||||
/// the same bytes is compared by value.
|
||||
fn same_bytes(a: &BTreeMap<String, Arc<Vec<u8>>>, b: &BTreeMap<String, Arc<Vec<u8>>>) -> bool {
|
||||
a.len() == b.len() && a.iter().zip(b).all(|((k1, v1), (k2, v2))| k1 == k2 && (Arc::ptr_eq(v1, v2) || **v1 == **v2))
|
||||
}
|
||||
|
||||
/// Which story a position is in.
|
||||
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord, Hash, Serialize, Deserialize, Default)]
|
||||
#[serde(rename_all = "camelCase")]
|
||||
@@ -447,8 +356,7 @@ pub struct Comment {
|
||||
/// keeps tracking the same comment. `None` for comments made here: the writer assigns one.
|
||||
#[serde(skip_serializing_if = "Option::is_none")]
|
||||
pub durable_id: Option<String>,
|
||||
/// The comment's UTC timestamp: from the file (`w16cex:dateUtc`, where `date` is local time),
|
||||
/// or the time WordCraft stamped the comment with.
|
||||
/// The comment's UTC timestamp from the file (`w16cex:dateUtc`); `date` is local time there.
|
||||
#[serde(skip_serializing_if = "Option::is_none")]
|
||||
pub date_utc: Option<String>,
|
||||
}
|
||||
@@ -641,10 +549,6 @@ pub struct Settings {
|
||||
/// after plus the second's space before, not the larger of the two.
|
||||
#[serde(default, skip_serializing_if = "std::ops::Not::not")]
|
||||
pub sum_paragraph_spacing: bool,
|
||||
/// `w:doNotExpandShiftReturn`: a justified line that ends in a manual line break keeps its
|
||||
/// natural spacing; by default Word stretches it to the margin like a wrapped line.
|
||||
#[serde(default, skip_serializing_if = "std::ops::Not::not")]
|
||||
pub do_not_expand_shift_return: bool,
|
||||
/// Keep contextual paragraph spacing inside legacy table cells (`allowSpaceOfSameStyleInTable`).
|
||||
/// Word ignores the layout override in compatibility mode 15, but the flag still round-trips.
|
||||
pub allow_space_of_same_style_in_table: bool,
|
||||
@@ -811,7 +715,6 @@ impl Default for Settings {
|
||||
compat_mode: COMPAT_MODE_CURRENT,
|
||||
roman_fonts: Vec::new(),
|
||||
sum_paragraph_spacing: false,
|
||||
do_not_expand_shift_return: false,
|
||||
allow_space_of_same_style_in_table: false,
|
||||
grid_h: DEFAULT_GRID,
|
||||
grid_v: DEFAULT_GRID,
|
||||
@@ -859,50 +762,6 @@ pub struct Document {
|
||||
/// The document's own building blocks (its glossary document), in file order.
|
||||
#[serde(skip_serializing_if = "Vec::is_empty")]
|
||||
pub glossary: Vec<BuildingBlock>,
|
||||
/// Lookup cache for [`Document::add_media`]; not part of the document (never saved, ignored
|
||||
/// by `==`).
|
||||
#[serde(skip)]
|
||||
media_index: MediaIndex,
|
||||
}
|
||||
|
||||
/// [`Document::media`] by content, so adding media is linear in the bytes added rather than a
|
||||
/// scan of every earlier image (CWE-407). Rebuilt whenever `media` changed size behind its back;
|
||||
/// a hit is always checked against `media`, so a stale entry can at worst miss a duplicate.
|
||||
#[derive(Clone, Default)]
|
||||
struct MediaIndex {
|
||||
by_bytes: std::collections::HashMap<Arc<Vec<u8>>, String>,
|
||||
/// `media.len()` when the index last matched it.
|
||||
len: usize,
|
||||
/// The number after the last generated `image{n}` key, so taken names are probed once.
|
||||
next: usize,
|
||||
}
|
||||
|
||||
impl MediaIndex {
|
||||
fn rebuild(&mut self, media: &BTreeMap<String, Arc<Vec<u8>>>) {
|
||||
self.by_bytes.clear();
|
||||
// The first key (in key order) of equal bytes wins, as a scan would find it.
|
||||
for (k, v) in media.iter().rev() {
|
||||
self.by_bytes.insert(v.clone(), k.clone());
|
||||
}
|
||||
self.len = media.len();
|
||||
self.next = 0;
|
||||
}
|
||||
|
||||
fn find(&self, media: &BTreeMap<String, Arc<Vec<u8>>>, bytes: &Vec<u8>) -> Option<String> {
|
||||
self.by_bytes.get(bytes).filter(|k| media.get(*k).is_some_and(|v| **v == *bytes)).cloned()
|
||||
}
|
||||
}
|
||||
|
||||
impl PartialEq for MediaIndex {
|
||||
fn eq(&self, _: &Self) -> bool {
|
||||
true
|
||||
}
|
||||
}
|
||||
|
||||
impl std::fmt::Debug for MediaIndex {
|
||||
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
|
||||
f.write_str("MediaIndex")
|
||||
}
|
||||
}
|
||||
|
||||
impl Default for Document {
|
||||
@@ -931,7 +790,6 @@ impl Document {
|
||||
media: BTreeMap::new(),
|
||||
passthrough: BTreeMap::new(),
|
||||
glossary: Vec::new(),
|
||||
media_index: MediaIndex::default(),
|
||||
}
|
||||
}
|
||||
|
||||
@@ -980,72 +838,6 @@ impl Document {
|
||||
}
|
||||
}
|
||||
|
||||
/// Whether the document holds the same content as `other`: the same blocks, the same
|
||||
/// styles, lists, settings, comments and the rest.
|
||||
///
|
||||
/// The blocks are compared by address first: the document is copy-on-write, so a block
|
||||
/// that changed was copied and has another address. Comparing them by value would walk
|
||||
/// the whole document, and this is asked after every command.
|
||||
///
|
||||
/// Every field is listed, so a new one does not build until someone decides whether a
|
||||
/// change to it counts (the same rule the session's edit snapshot follows).
|
||||
pub fn same_content(&self, other: &Document) -> bool {
|
||||
let Document {
|
||||
body,
|
||||
last_section,
|
||||
parts,
|
||||
styles,
|
||||
numbering,
|
||||
comments,
|
||||
revisions,
|
||||
settings,
|
||||
core,
|
||||
sources,
|
||||
custom_props,
|
||||
footnote_separators,
|
||||
endnote_separators,
|
||||
media,
|
||||
passthrough,
|
||||
glossary,
|
||||
media_index: _,
|
||||
} = self;
|
||||
let Document {
|
||||
body: other_body,
|
||||
last_section: other_section,
|
||||
parts: other_parts,
|
||||
styles: other_styles,
|
||||
numbering: other_numbering,
|
||||
comments: other_comments,
|
||||
revisions: other_revisions,
|
||||
settings: other_settings,
|
||||
core: other_core,
|
||||
sources: other_sources,
|
||||
custom_props: other_props,
|
||||
footnote_separators: other_footnotes,
|
||||
endnote_separators: other_endnotes,
|
||||
media: other_media,
|
||||
passthrough: other_passthrough,
|
||||
glossary: other_glossary,
|
||||
media_index: _,
|
||||
} = other;
|
||||
same_blocks(body, other_body)
|
||||
&& same_parts(parts, other_parts)
|
||||
&& last_section == other_section
|
||||
&& styles == other_styles
|
||||
&& numbering == other_numbering
|
||||
&& comments == other_comments
|
||||
&& revisions == other_revisions
|
||||
&& settings == other_settings
|
||||
&& core == other_core
|
||||
&& sources == other_sources
|
||||
&& custom_props == other_props
|
||||
&& footnote_separators == other_footnotes
|
||||
&& endnote_separators == other_endnotes
|
||||
&& same_bytes(media, other_media)
|
||||
&& same_bytes(passthrough, other_passthrough)
|
||||
&& same_glossary(glossary, other_glossary)
|
||||
}
|
||||
|
||||
pub fn story(&self, s: StoryRef) -> Option<&Blocks> {
|
||||
match s {
|
||||
StoryRef::Body => Some(&self.body),
|
||||
@@ -1149,8 +941,8 @@ impl Document {
|
||||
for story in stories {
|
||||
for path in self.para_paths(story) {
|
||||
let Some(p) = self.para(story, &path) else { continue };
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
if o.text_boxes().contains(&part) {
|
||||
for off in p.object_offsets() {
|
||||
if p.object_at(off).is_some_and(|o| o.text_boxes().contains(&part)) {
|
||||
return Some(Pos { story, path, off: off + para::OBJ.len_utf8() });
|
||||
}
|
||||
}
|
||||
@@ -1610,32 +1402,17 @@ impl Document {
|
||||
|
||||
/// Add a media item; returns its key (deduplicated by content).
|
||||
pub fn add_media(&mut self, bytes: Vec<u8>, ext: &str) -> String {
|
||||
let ix = &mut self.media_index;
|
||||
if ix.len != self.media.len() {
|
||||
ix.rebuild(&self.media);
|
||||
if let Some((k, _)) = self.media.iter().find(|(_, v)| v.as_slice() == bytes.as_slice()) {
|
||||
return k.clone();
|
||||
}
|
||||
if let Some(k) = ix.find(&self.media, &bytes) {
|
||||
return k;
|
||||
}
|
||||
if ix.by_bytes.contains_key(&bytes) {
|
||||
// Stale (`media` was edited directly without changing its size): look again.
|
||||
ix.rebuild(&self.media);
|
||||
if let Some(k) = ix.find(&self.media, &bytes) {
|
||||
return k;
|
||||
}
|
||||
}
|
||||
let n = self.media.len().saturating_add(1);
|
||||
let mut i = n.max(ix.next);
|
||||
let mut key = format!("image{i}.{ext}");
|
||||
let n = self.media.len() + 1;
|
||||
let mut key = format!("image{n}.{ext}");
|
||||
let mut i = n;
|
||||
while self.media.contains_key(&key) {
|
||||
i = i.saturating_add(1);
|
||||
i += 1;
|
||||
key = format!("image{i}.{ext}");
|
||||
}
|
||||
let bytes = Arc::new(bytes);
|
||||
self.media.insert(key.clone(), bytes.clone());
|
||||
ix.by_bytes.entry(bytes).or_insert_with(|| key.clone());
|
||||
ix.len = self.media.len();
|
||||
ix.next = i.saturating_add(1);
|
||||
self.media.insert(key.clone(), Arc::new(bytes));
|
||||
key
|
||||
}
|
||||
|
||||
@@ -1670,22 +1447,22 @@ impl Document {
|
||||
let mut out: Vec<PermRange> = Vec::new();
|
||||
let stories = std::iter::once(StoryRef::Body).chain(self.parts.keys().map(|k| StoryRef::Part(*k)));
|
||||
for story in stories {
|
||||
// Open ranges by id, latest last (an end closes the latest open range of its id).
|
||||
let mut open: BTreeMap<u32, Vec<usize>> = BTreeMap::new();
|
||||
let mut open: Vec<usize> = Vec::new();
|
||||
for path in self.para_paths(story) {
|
||||
let Some(p) = self.para(story, &path) else { continue };
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
match o {
|
||||
InlineObject::PermStart { id, group, editor } => {
|
||||
for off in p.object_offsets() {
|
||||
match p.object_at(off) {
|
||||
Some(InlineObject::PermStart { id, group, editor }) => {
|
||||
let start = Pos { story, path: path.clone(), off: off + para::OBJ.len_utf8() };
|
||||
open.entry(*id).or_default().push(out.len());
|
||||
open.push(out.len());
|
||||
out.push(PermRange { id: *id, group: group.clone(), editor: editor.clone(), start: start.clone(), end: start });
|
||||
}
|
||||
InlineObject::PermEnd { id } => {
|
||||
if let Some(i) = open.get_mut(id).and_then(Vec::pop)
|
||||
&& let Some(r) = out.get_mut(i)
|
||||
{
|
||||
r.end = Pos { story, path: path.clone(), off };
|
||||
Some(InlineObject::PermEnd { id }) => {
|
||||
if let Some(k) = open.iter().rposition(|i| out.get(*i).is_some_and(|r| r.id == *id)) {
|
||||
let i = open.remove(k);
|
||||
if let Some(r) = out.get_mut(i) {
|
||||
r.end = Pos { story, path: path.clone(), off };
|
||||
}
|
||||
}
|
||||
}
|
||||
_ => {}
|
||||
@@ -1694,7 +1471,7 @@ impl Document {
|
||||
}
|
||||
// A range left open runs to the end of its story.
|
||||
let end = self.end_of(story);
|
||||
for i in open.into_values().flatten() {
|
||||
for i in open {
|
||||
if let Some(r) = out.get_mut(i) {
|
||||
r.end = end.clone();
|
||||
}
|
||||
@@ -1708,8 +1485,8 @@ impl Document {
|
||||
let mut out = Vec::new();
|
||||
for path in self.para_paths(StoryRef::Body) {
|
||||
if let Some(p) = self.para(StoryRef::Body, &path) {
|
||||
for (off, o) in p.objects_with_offsets() {
|
||||
if let InlineObject::BookmarkStart { name } = o {
|
||||
for off in p.object_offsets() {
|
||||
if let Some(InlineObject::BookmarkStart { name }) = p.object_at(off) {
|
||||
out.push((name.clone(), Pos { story: StoryRef::Body, path: path.clone(), off }));
|
||||
}
|
||||
}
|
||||
@@ -1891,7 +1668,5 @@ fn story_word_count_of(bl: &Blocks, numbering: &Numbering, styles: &styles::Styl
|
||||
total
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod test_log;
|
||||
#[cfg(test)]
|
||||
mod tests;
|
||||
|
||||
@@ -41,24 +41,6 @@ impl Math {
|
||||
pub fn is_empty(&self) -> bool {
|
||||
self.nodes.is_empty()
|
||||
}
|
||||
|
||||
/// Rough heap bytes a clone copies: every node, its argument lists and its run text.
|
||||
/// Iterative, so it costs at most the size of the structure.
|
||||
pub(crate) fn heap_bytes(&self) -> usize {
|
||||
let mut n = self.omml.len();
|
||||
let mut todo: Vec<&MNode> = self.nodes.iter().collect();
|
||||
while let Some(node) = todo.pop() {
|
||||
n = n.saturating_add(std::mem::size_of::<MNode>());
|
||||
if let MNode::Run(r) = node {
|
||||
n = n.saturating_add(r.text.len()).saturating_add(r.font.as_ref().map_or(0, String::len));
|
||||
}
|
||||
each_arg(node, &mut |a| {
|
||||
n = n.saturating_add(std::mem::size_of::<Arg>());
|
||||
todo.extend(a);
|
||||
});
|
||||
}
|
||||
n
|
||||
}
|
||||
}
|
||||
|
||||
/// Display equation justification (`m:oMathParaPr/m:jc`).
|
||||
@@ -1113,20 +1095,4 @@ mod tests {
|
||||
let deep = "√".repeat(3000);
|
||||
let _ = parse_linear(&deep);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_long_run_of_letters_parses_in_linear_time() {
|
||||
// Each letter that doesn't start a function name used to rescan (and copy) the rest of
|
||||
// the run, so a 20,000-letter equation did 200 million steps. (No wall-clock assert: a
|
||||
// loaded runner would fail it; a regression shows as a slow test.)
|
||||
for _ in 0..5 {
|
||||
let e = parse_linear(&"x".repeat(30_000));
|
||||
assert!(matches!(e.as_slice(), [MNode::Run(r)] if r.text.len() == 20_000), "{:?}", e.len());
|
||||
}
|
||||
// Function names are still found at the start of a run, and only as whole runs.
|
||||
assert!(matches!(parse_linear("sin x").as_slice(), [MNode::Func { .. }]));
|
||||
assert!(matches!(parse_linear("arcsin x").as_slice(), [MNode::Func { .. }]));
|
||||
assert!(matches!(parse_linear("sinx").as_slice(), [MNode::Run(_)]));
|
||||
assert!(matches!(parse_linear("arcsinhx").as_slice(), [MNode::Run(_)]));
|
||||
}
|
||||
}
|
||||
|
||||
@@ -324,21 +324,6 @@ impl Lp {
|
||||
let den = self.arg();
|
||||
one(MNode::Frac { kind: FracKind::Bar, num, den })
|
||||
}
|
||||
"genfrac" => {
|
||||
let beg = delim_arg(&self.arg());
|
||||
let end = delim_arg(&self.arg());
|
||||
let thickness = self.arg();
|
||||
let _style = self.arg();
|
||||
let num = self.arg();
|
||||
let den = self.arg();
|
||||
let kind = if is_zero_length(&thickness) { FracKind::NoBar } else { FracKind::Bar };
|
||||
let frac = MNode::Frac { kind, num, den };
|
||||
if beg.is_none() && end.is_none() {
|
||||
one(frac)
|
||||
} else {
|
||||
one(MNode::Delim { beg, end, sep: None, grow: true, shp_match: false, elems: vec![vec![frac]] })
|
||||
}
|
||||
}
|
||||
"binom" | "dbinom" | "tbinom" => {
|
||||
let n = self.arg();
|
||||
let k = self.arg();
|
||||
@@ -816,31 +801,6 @@ fn text_chars(text: &str, s: &mut String) {
|
||||
}
|
||||
}
|
||||
|
||||
/// The delimiter of a `\genfrac` argument: one character, or none for `{}` and anything else.
|
||||
fn delim_arg(a: &[MNode]) -> Option<char> {
|
||||
let [MNode::Run(r)] = a else { return None };
|
||||
let mut cs = r.text.chars();
|
||||
match (cs.next(), cs.next()) {
|
||||
(Some(c), None) if c != '.' => Some(c),
|
||||
_ => None,
|
||||
}
|
||||
}
|
||||
|
||||
/// Whether a `\genfrac` thickness is a zero length such as `0pt`, `0.0pt` or `0 em`. An empty
|
||||
/// thickness means the default rule and anything that is not a plain number and unit is not zero.
|
||||
fn is_zero_length(a: &[MNode]) -> bool {
|
||||
let mut text = String::new();
|
||||
for n in a {
|
||||
let MNode::Run(r) = n else { return false };
|
||||
text.extend(r.text.chars().filter(|c| !c.is_whitespace() && !matches!(c, '\u{2009}' | '\u{205F}' | '\u{2005}')));
|
||||
}
|
||||
let number = text.trim_end_matches(|c: char| c.is_ascii_alphabetic());
|
||||
let unit = &text[number.len()..];
|
||||
!number.is_empty()
|
||||
&& matches!(unit, "pt" | "em" | "ex" | "mm" | "cm" | "in" | "bp" | "pc" | "sp" | "mu")
|
||||
&& number.parse::<f64>().is_ok_and(|v| v == 0.0)
|
||||
}
|
||||
|
||||
fn delim_tex(c: Option<char>) -> String {
|
||||
match c {
|
||||
None => ".".into(),
|
||||
@@ -1116,34 +1076,6 @@ mod tests {
|
||||
let _ = (LimLoc::UndOvr, MathJc::Center);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn latex_binom_round_trip() {
|
||||
let a = parse_latex(r"\binom{n}{k}");
|
||||
let l = to_latex(&a);
|
||||
assert_eq!(parse_latex(&l), a, "exported as {l}");
|
||||
let nobar = |n: &str, k: &str| MNode::Frac { kind: FracKind::NoBar, num: parse_latex(n), den: parse_latex(k) };
|
||||
let b = parse_latex(r"\genfrac{}{}{0pt}{}{n}{k}");
|
||||
assert_eq!(b, vec![nobar("n", "k")]);
|
||||
assert_eq!(parse_latex(&to_latex(&b)), b);
|
||||
// Other spellings of a zero thickness.
|
||||
for z in ["0 pt", "0.0pt", r"0\,pt", "0em"] {
|
||||
let c = parse_latex(&format!(r"\genfrac{{}}{{}}{{{z}}}{{}}{{n}}{{k}}"));
|
||||
assert_eq!(c, vec![nobar("n", "k")], "thickness {z}");
|
||||
}
|
||||
// A non-zero or default thickness keeps the bar.
|
||||
for t in ["1pt", "0.4pt", ""] {
|
||||
let c = parse_latex(&format!(r"\genfrac{{}}{{}}{{{t}}}{{}}{{n}}{{k}}"));
|
||||
let bar = MNode::Frac { kind: FracKind::Bar, num: parse_latex("n"), den: parse_latex("k") };
|
||||
assert_eq!(c, vec![bar], "thickness {t:?}");
|
||||
}
|
||||
// Delimiters are kept.
|
||||
let d = parse_latex(r"\genfrac(){0pt}{}{n}{k}");
|
||||
assert_eq!(d, a);
|
||||
let e = parse_latex(r"\genfrac[]{0pt}{}{n}{k}");
|
||||
assert!(matches!(&e[..], [MNode::Delim { beg: Some('['), end: Some(']'), .. }]), "{e:#?}");
|
||||
assert_eq!(parse_latex(&to_latex(&e)), e);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn latex_round_trip() {
|
||||
for lin in ["x=(-b±√(b^2-4ac))/2a", "∑_(i=1)^n▒i^2", "√(3&x)", "a⃗+x̂", "■(a&b@c&d)", "sinx+cosy", "e^(-x^2)"] {
|
||||
|
||||
@@ -18,18 +18,6 @@ pub const FUNCTION_NAMES: &[&str] = &[
|
||||
"tanh",
|
||||
];
|
||||
|
||||
/// Length of the longest of [`FUNCTION_NAMES`].
|
||||
const MAX_NAME: usize = {
|
||||
let (mut max, mut i) = (0, 0);
|
||||
while i < FUNCTION_NAMES.len() {
|
||||
if FUNCTION_NAMES[i].len() > max {
|
||||
max = FUNCTION_NAMES[i].len();
|
||||
}
|
||||
i += 1;
|
||||
}
|
||||
max
|
||||
};
|
||||
|
||||
/// Function names whose subscripts go underneath (`lim_(n→∞)`).
|
||||
const LIMIT_NAMES: &[&str] = &["lim", "liminf", "limsup", "max", "min", "sup", "inf", "det", "gcd", "Pr"];
|
||||
|
||||
@@ -422,12 +410,10 @@ impl P {
|
||||
fn word(&mut self, stops: &[char]) -> Arg {
|
||||
let start = self.i;
|
||||
let mut end = start;
|
||||
// Look no further than the longest name (and one past it, to see the run is longer): a
|
||||
// scan to the end of every run made a long run of letters quadratic (CWE-407).
|
||||
while end - start <= MAX_NAME && self.c.get(end).is_some_and(char::is_ascii_alphabetic) {
|
||||
while self.c.get(end).is_some_and(char::is_ascii_alphabetic) {
|
||||
end += 1;
|
||||
}
|
||||
let word: String = if end - start > MAX_NAME { String::new() } else { self.c.get(start..end).unwrap_or(&[]).iter().collect() };
|
||||
let word: String = self.c.get(start..end).unwrap_or(&[]).iter().collect();
|
||||
if !FUNCTION_NAMES.contains(&word.as_str()) {
|
||||
self.i = start + 1;
|
||||
return vec![text(self.c.get(start).copied().unwrap_or(' '))];
|
||||
@@ -570,7 +556,7 @@ impl P {
|
||||
}
|
||||
out.pop()
|
||||
};
|
||||
if !(matches!(c, '^' | '_') || matches!(c, '┬' | '┴') && self.no_limits == 0 || is_combining(c)) {
|
||||
if !matches!(c, '^' | '_') && !(matches!(c, '┬' | '┴') && self.no_limits == 0) && !is_combining(c) {
|
||||
break;
|
||||
}
|
||||
// The base is already deep: the operator is text (see `too_deep`).
|
||||
|
||||
@@ -357,8 +357,7 @@ impl Counters {
|
||||
return Some((def.text.clone(), def));
|
||||
}
|
||||
let mut label = String::new();
|
||||
// Word's level text is short; .docx and the lists command already cap it at 64.
|
||||
let mut chars = def.text.chars().take(MAX_LEVEL_TEXT).peekable();
|
||||
let mut chars = def.text.chars().peekable();
|
||||
while let Some(c) = chars.next() {
|
||||
if c == '%'
|
||||
&& let Some(d) = chars.peek().and_then(|d| d.to_digit(10))
|
||||
@@ -376,9 +375,6 @@ impl Counters {
|
||||
f => f,
|
||||
};
|
||||
label.push_str(&fmt.format(val));
|
||||
if label.len() > MAX_LABEL_BYTES {
|
||||
break;
|
||||
}
|
||||
} else {
|
||||
label.push(c);
|
||||
}
|
||||
@@ -387,29 +383,10 @@ impl Counters {
|
||||
}
|
||||
}
|
||||
|
||||
/// Longest level text (in characters) expanded into a list label.
|
||||
const MAX_LEVEL_TEXT: usize = 64;
|
||||
/// Labels stop growing once they pass this many bytes.
|
||||
const MAX_LABEL_BYTES: usize = 256;
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::*;
|
||||
|
||||
#[test]
|
||||
fn a_huge_level_text_gives_a_bounded_label() {
|
||||
let mut n = Numbering::default();
|
||||
let id = n.add_list(ListKind::Numbered);
|
||||
let huge = Level { format: NumFormat::UpperRoman, text: "%1".repeat(30_000), start: u32::MAX, ..Level::default() };
|
||||
if let Some(num) = n.nums.iter_mut().find(|x| x.id == id) {
|
||||
num.level_overrides = vec![(0, huge)];
|
||||
}
|
||||
let mut c = Counters::default();
|
||||
let label = c.next_label(&n, id, 0).unwrap().0;
|
||||
assert!(label.len() <= 512, "label is {} bytes", label.len());
|
||||
assert!(label.starts_with("MMMM"));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_list_s_own_level_definitions_win() {
|
||||
// A specification's article numbers: level 1 counts parts with no label of its own, level 2
|
||||
|
||||
@@ -311,14 +311,7 @@ pub enum ShapeKind {
|
||||
RightTriangle,
|
||||
Diamond,
|
||||
Line,
|
||||
/// A block arrow pointing right (`rightArrow`; also stands for the double-headed arrows).
|
||||
Arrow,
|
||||
/// A block arrow pointing left (`leftArrow`).
|
||||
ArrowLeft,
|
||||
/// A block arrow pointing up (`upArrow`).
|
||||
ArrowUp,
|
||||
/// A block arrow pointing down (`downArrow`).
|
||||
ArrowDown,
|
||||
Star,
|
||||
Heart,
|
||||
/// A text box (rectangle with a text story).
|
||||
@@ -1089,12 +1082,6 @@ impl Paragraph {
|
||||
pub fn object_offsets(&self) -> Vec<usize> {
|
||||
self.text.char_indices().filter(|(_, c)| *c == OBJ).map(|(i, _)| i).collect()
|
||||
}
|
||||
/// (byte offset, object) for every object, in text order, in one pass. Prefer it to
|
||||
/// [`Paragraph::object_offsets`] + [`Paragraph::object_at`], which counts the objects before
|
||||
/// each offset again, so a loop over a paragraph's objects is quadratic (#3614).
|
||||
pub fn objects_with_offsets(&self) -> impl Iterator<Item = (usize, &InlineObject)> + '_ {
|
||||
self.text.char_indices().filter(|(_, c)| *c == OBJ).map(|(i, _)| i).zip(self.objects.iter())
|
||||
}
|
||||
|
||||
/// Insert plain text (U+FFFC is dropped) at `off` with `props`. Returns the inserted length.
|
||||
pub fn insert_text(&mut self, off: usize, s: &str, props: &CharProps) -> Result<usize> {
|
||||
@@ -1565,11 +1552,6 @@ mod tests {
|
||||
assert!(matches!(p.object_at(0), Some(InlineObject::BookmarkStart { .. })));
|
||||
let off = p.object_offsets()[1];
|
||||
assert!(matches!(p.object_at(off), Some(InlineObject::Field { .. })));
|
||||
// The one-pass walk agrees with offsets + object_at.
|
||||
let one_pass: Vec<(usize, &InlineObject)> = p.objects_with_offsets().collect();
|
||||
let each: Vec<(usize, &InlineObject)> = p.object_offsets().into_iter().filter_map(|o| Some((o, p.object_at(o)?))).collect();
|
||||
assert_eq!(one_pass, each);
|
||||
assert_eq!(one_pass.len(), 2);
|
||||
assert_eq!(p.plain_text(), "a1b");
|
||||
p.delete(0, 3).unwrap();
|
||||
assert_eq!(p.objects.len(), 1);
|
||||
|
||||
@@ -342,9 +342,6 @@ impl Highlight {
|
||||
}
|
||||
}
|
||||
|
||||
/// The text size, in points, of a run whose size no style or default sets.
|
||||
pub const DEFAULT_FONT_SIZE: f32 = 11.0;
|
||||
|
||||
/// Character formatting (direct or in a style). `None` = inherit.
|
||||
#[derive(Clone, Debug, PartialEq, Default, Serialize, Deserialize)]
|
||||
#[serde(default, rename_all = "camelCase")]
|
||||
|
||||
@@ -78,7 +78,7 @@ impl ResolvedChar {
|
||||
ResolvedChar {
|
||||
// A symbol character is shown in its own font (`w:sym`), whatever the run's font.
|
||||
font: c.shown_font().map(str::to_string).unwrap_or_else(|| crate::styles::BODY_FONT.to_string()),
|
||||
size: c.size.unwrap_or(crate::props::DEFAULT_FONT_SIZE).clamp(1.0, 1638.0),
|
||||
size: c.size.unwrap_or(11.0).clamp(1.0, 1638.0),
|
||||
bold: c.bold.unwrap_or(false),
|
||||
italic: c.italic.unwrap_or(false),
|
||||
underline: c.underline.unwrap_or_default(),
|
||||
@@ -113,7 +113,7 @@ impl ResolvedChar {
|
||||
cs: c.cs.unwrap_or(false),
|
||||
font_cs: c.font_cs.clone().or_else(|| c.font.clone()).unwrap_or_else(|| crate::styles::BODY_FONT.to_string()),
|
||||
font_ea: c.font_ea.clone(),
|
||||
size_cs: c.size_cs.or(c.size).unwrap_or(crate::props::DEFAULT_FONT_SIZE).clamp(1.0, 1638.0),
|
||||
size_cs: c.size_cs.or(c.size).unwrap_or(11.0).clamp(1.0, 1638.0),
|
||||
bold_cs: c.bold_cs.or(c.bold).unwrap_or(false),
|
||||
italic_cs: c.italic_cs.or(c.italic).unwrap_or(false),
|
||||
}
|
||||
|
||||
@@ -255,7 +255,7 @@ impl NumFormat {
|
||||
pub fn from_ooxml(s: &str) -> NumFormat {
|
||||
NumFormat::ALL.into_iter().find(|f| f.ooxml() == s).unwrap_or(NumFormat::Decimal)
|
||||
}
|
||||
/// Format `n` (1-based). Formats WordCraft doesn't draw yet (Hebrew, Hindi, Thai,
|
||||
/// Format `n` (1-based). Formats WordCraft doesn't draw yet (Hebrew, Arabic, Hindi, Thai,
|
||||
/// Vietnamese, text amounts) show decimal, as do values past a sequence's end
|
||||
/// (the 21st item of ①…⑳).
|
||||
pub fn format(self, n: u32) -> String {
|
||||
@@ -268,8 +268,6 @@ impl NumFormat {
|
||||
NumFormat::LowerLetter => letters(n),
|
||||
NumFormat::RussianUpper => russian_letters(n).to_uppercase(),
|
||||
NumFormat::RussianLower => russian_letters(n),
|
||||
NumFormat::ArabicAlpha => repeated_letter(&ARABIC_ALPHA, n),
|
||||
NumFormat::ArabicAbjad => repeated_letter(&ARABIC_ABJAD, n),
|
||||
NumFormat::Ordinal => format!("{n}{}", ordinal_suffix(n)),
|
||||
NumFormat::CardinalText => cardinal_text(n),
|
||||
NumFormat::OrdinalText => ordinal_text(n),
|
||||
@@ -310,6 +308,8 @@ impl NumFormat {
|
||||
| NumFormat::VietnameseCounting
|
||||
| NumFormat::Hebrew1
|
||||
| NumFormat::Hebrew2
|
||||
| NumFormat::ArabicAlpha
|
||||
| NumFormat::ArabicAbjad
|
||||
| NumFormat::HindiVowels
|
||||
| NumFormat::HindiConsonants
|
||||
| NumFormat::HindiNumbers
|
||||
@@ -399,27 +399,6 @@ pub fn russian_letters(n: u32) -> String {
|
||||
std::iter::repeat_n(c, k as usize).collect()
|
||||
}
|
||||
|
||||
/// `arabicAlpha`: the 28 letters in alphabetical (hijā'ī) order.
|
||||
const ARABIC_ALPHA: [char; 28] =
|
||||
['أ', 'ب', 'ت', 'ث', 'ج', 'ح', 'خ', 'د', 'ذ', 'ر', 'ز', 'س', 'ش', 'ص', 'ض', 'ط', 'ظ', 'ع', 'غ', 'ف', 'ق', 'ك', 'ل', 'م', 'ن', 'ه', 'و', 'ي'];
|
||||
|
||||
/// `arabicAbjad`: the 28 letters in abjad order (أبجد هوز حطي كلمن سعفص قرشت ثخذ ضظغ).
|
||||
const ARABIC_ABJAD: [char; 28] =
|
||||
['أ', 'ب', 'ج', 'د', 'ه', 'و', 'ز', 'ح', 'ط', 'ي', 'ك', 'ل', 'م', 'ن', 'س', 'ع', 'ف', 'ص', 'ق', 'ر', 'ش', 'ت', 'ث', 'خ', 'ذ', 'ض', 'ظ', 'غ'];
|
||||
|
||||
/// One letter of `alphabet`, repeated once per pass through it (as [`letters`]: a..z, aa..zz).
|
||||
fn repeated_letter(alphabet: &[char], n: u32) -> String {
|
||||
let Some(&first) = alphabet.first() else { return n.to_string() };
|
||||
if n == 0 {
|
||||
return String::new();
|
||||
}
|
||||
let len = alphabet.len() as u32;
|
||||
let n = n.min(len * 30);
|
||||
let k = (n - 1) / len + 1;
|
||||
let c = alphabet.get(((n - 1) % len) as usize).copied().unwrap_or(first);
|
||||
std::iter::repeat_n(c, k as usize).collect()
|
||||
}
|
||||
|
||||
const ONES: [&str; 20] = [
|
||||
"zero",
|
||||
"one",
|
||||
@@ -941,19 +920,6 @@ mod tests {
|
||||
assert_eq!(NumFormat::RussianUpper.format(36), "ЖЖ");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn arabic_letters_count_in_both_orders() {
|
||||
let alpha: Vec<String> = (1..=5).map(|n| NumFormat::ArabicAlpha.format(n)).collect();
|
||||
assert_eq!(alpha.join(" "), "أ ب ت ث ج");
|
||||
let abjad: Vec<String> = (1..=5).map(|n| NumFormat::ArabicAbjad.format(n)).collect();
|
||||
assert_eq!(abjad.join(" "), "أ ب ج د ه");
|
||||
assert_eq!(NumFormat::ArabicAlpha.format(28), "ي");
|
||||
assert_eq!(NumFormat::ArabicAbjad.format(28), "غ");
|
||||
// Past the 28th letter they double, as Latin and Russian letters do.
|
||||
assert_eq!(NumFormat::ArabicAlpha.format(29), "أأ");
|
||||
assert_eq!(NumFormat::ArabicAbjad.format(30), "بب");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn landscape_swaps() {
|
||||
let mut s = SectionProps { margin_left: 50.0, ..Default::default() };
|
||||
|
||||
+1
-56
@@ -48,12 +48,6 @@ impl Row {
|
||||
pub fn first_grid(&self) -> usize {
|
||||
(self.props.grid_before as usize).min(MAX_COLS)
|
||||
}
|
||||
|
||||
/// Grid columns `w:gridAfter` leaves empty after the row's last cell, bounded as the DOCX
|
||||
/// reader bounds them (a native document can say anything).
|
||||
pub fn grid_after(&self) -> usize {
|
||||
(self.props.grid_after as usize).min(MAX_COLS)
|
||||
}
|
||||
}
|
||||
|
||||
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize, Default)]
|
||||
@@ -103,8 +97,7 @@ impl Table {
|
||||
}
|
||||
|
||||
pub fn cols(&self) -> usize {
|
||||
let row_cols =
|
||||
|r: &Row| r.first_grid().saturating_add(r.cells.iter().map(Cell::span).fold(0, usize::saturating_add)).saturating_add(r.grid_after());
|
||||
let row_cols = |r: &Row| r.first_grid() + r.cells.iter().map(Cell::span).sum::<usize>() + r.props.grid_after as usize;
|
||||
self.grid.len().max(self.rows.iter().map(row_cols).max().unwrap_or(0))
|
||||
}
|
||||
|
||||
@@ -172,54 +165,6 @@ impl Table {
|
||||
self.rows.remove(r);
|
||||
}
|
||||
|
||||
/// Split the table before row `at`: the rows from `at` on move to the returned table, which
|
||||
/// has this table's properties and grid. A vertical merge that the split cuts ends in this
|
||||
/// table's last row and starts again in the new table's first row. A row content control
|
||||
/// around rows on both sides closes after this table's last row and opens again (without its
|
||||
/// `w:id`) before the new table's first row. Content controls that close after this table
|
||||
/// close after the new one.
|
||||
pub fn split_off(&mut self, at: usize) -> Table {
|
||||
let at = at.min(self.rows.len());
|
||||
let mut open = Vec::new();
|
||||
for row in self.rows.iter().take(at) {
|
||||
open.extend(row.controls.open.iter().cloned());
|
||||
let close = usize::try_from(row.controls.close).unwrap_or(usize::MAX).min(open.len());
|
||||
open.truncate(open.len() - close);
|
||||
}
|
||||
let mut tail = Table {
|
||||
props: self.props.clone(),
|
||||
grid: self.grid.clone(),
|
||||
rows: self.rows.split_off(at),
|
||||
controls: ControlWrap { open: Vec::new(), close: std::mem::take(&mut self.controls.close) },
|
||||
};
|
||||
if let Some(last) = self.rows.last_mut() {
|
||||
last.controls.close = last.controls.close.saturating_add(u32::try_from(open.len()).unwrap_or(u32::MAX));
|
||||
for c in &mut last.cells {
|
||||
if c.props.vmerge == VMerge::Restart {
|
||||
c.props.vmerge = VMerge::None;
|
||||
}
|
||||
}
|
||||
}
|
||||
for c in &mut open {
|
||||
c.id = None;
|
||||
}
|
||||
if let Some(first) = tail.rows.first_mut() {
|
||||
let own = std::mem::replace(&mut first.controls.open, open);
|
||||
first.controls.open.extend(own);
|
||||
}
|
||||
let starts: Vec<(usize, usize)> = (0..tail.rows.first().map_or(0, |row| row.cells.len()))
|
||||
.filter(|&c| tail.rows.first().and_then(|row| row.cells.get(c)).is_some_and(|x| x.props.vmerge == VMerge::Continue))
|
||||
.map(|c| (c, tail.grid_col(0, c)))
|
||||
.collect();
|
||||
for (c, g) in starts {
|
||||
let more = tail.cell_at_grid(1, g).and_then(|i| tail.rows.get(1)?.cells.get(i)).is_some_and(|x| x.props.vmerge == VMerge::Continue);
|
||||
if let Some(x) = tail.rows.first_mut().and_then(|r| r.cells.get_mut(c)) {
|
||||
x.props.vmerge = if more { VMerge::Restart } else { VMerge::None };
|
||||
}
|
||||
}
|
||||
tail
|
||||
}
|
||||
|
||||
/// Insert a grid column at grid index `g` (cells spanning it widen instead).
|
||||
pub fn insert_col(&mut self, g: usize) {
|
||||
if self.grid.len() >= MAX_COLS {
|
||||
|
||||
@@ -1,36 +0,0 @@
|
||||
//! Test helper: the warnings this crate logs, captured per test thread, so a test can check that
|
||||
//! a limit which leaves something out says so.
|
||||
|
||||
use std::cell::RefCell;
|
||||
|
||||
struct Capture;
|
||||
|
||||
thread_local! {
|
||||
static LINES: RefCell<Vec<String>> = const { RefCell::new(Vec::new()) };
|
||||
}
|
||||
|
||||
impl log::Log for Capture {
|
||||
fn enabled(&self, m: &log::Metadata) -> bool {
|
||||
m.level() <= log::Level::Warn
|
||||
}
|
||||
|
||||
fn log(&self, r: &log::Record) {
|
||||
if self.enabled(r.metadata()) {
|
||||
LINES.with(|l| l.borrow_mut().push(r.args().to_string()));
|
||||
}
|
||||
}
|
||||
|
||||
fn flush(&self) {}
|
||||
}
|
||||
|
||||
static CAPTURE: Capture = Capture;
|
||||
|
||||
/// `f`'s result and the warnings logged on this thread while it ran.
|
||||
pub(crate) fn warnings<T>(f: impl FnOnce() -> T) -> (T, Vec<String>) {
|
||||
// Only this logger is ever installed in a test binary; a second call keeps it.
|
||||
let _ = log::set_logger(&CAPTURE);
|
||||
log::set_max_level(log::LevelFilter::Warn);
|
||||
LINES.with(|l| l.borrow_mut().clear());
|
||||
let out = f();
|
||||
(out, LINES.with(|l| std::mem::take(&mut *l.borrow_mut())))
|
||||
}
|
||||
@@ -90,45 +90,6 @@ fn parts_and_media() {
|
||||
assert_ne!(k1, k3);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn many_distinct_media_add_in_linear_time() {
|
||||
// An import with tens of thousands of small distinct images (same length, differing only at
|
||||
// the end) used to compare each against every earlier one.
|
||||
let mut d = Document::new();
|
||||
let t = std::time::Instant::now();
|
||||
let n = 40_000u32;
|
||||
for i in 0..n {
|
||||
let mut bytes = vec![7u8; 256];
|
||||
bytes.extend(i.to_le_bytes());
|
||||
d.add_media(bytes, "png");
|
||||
}
|
||||
assert_eq!(d.media.len(), n as usize);
|
||||
// Duplicates still find their first copy, including media inserted directly.
|
||||
let mut first = vec![7u8; 256];
|
||||
first.extend(0u32.to_le_bytes());
|
||||
assert_eq!(d.add_media(first, "png"), "image1.png");
|
||||
d.media.insert("direct.png".into(), Arc::new(vec![9, 9]));
|
||||
assert_eq!(d.add_media(vec![9, 9], "png"), "direct.png");
|
||||
// About half a second on a loaded machine; the old scan took over two minutes.
|
||||
assert!(t.elapsed().as_secs_f32() < 10.0, "{:?}", t.elapsed());
|
||||
// The lookup cache isn't part of the document.
|
||||
let mut e = d.clone();
|
||||
e.media_index = MediaIndex::default();
|
||||
assert_eq!(d, e);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn media_keys_skip_taken_names_once() {
|
||||
let mut d = Document::new();
|
||||
// A file whose media already use the names add_media would pick next.
|
||||
for i in 2..2002u32 {
|
||||
d.media.insert(format!("image{i}.png"), Arc::new(i.to_le_bytes().to_vec()));
|
||||
}
|
||||
assert_eq!(d.add_media(vec![0], "png"), "image2002.png");
|
||||
assert_eq!(d.add_media(vec![1], "png"), "image2003.png");
|
||||
assert_eq!(d.add_media(vec![2], "png"), "image2004.png");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn free_ids_skip_taken_ones_instead_of_overflowing() {
|
||||
assert_eq!(free_id(None, std::iter::empty(), 1), Some(1));
|
||||
@@ -180,72 +141,6 @@ fn clone_is_shallow() {
|
||||
assert_eq!(d.plain_text(StoryRef::Body), "a\nb");
|
||||
}
|
||||
|
||||
/// [`Document::same_content`] is what tells the engine that a command changed nothing: it
|
||||
/// must see every change, and it must not call two different documents the same.
|
||||
#[test]
|
||||
fn same_content_sees_every_change() {
|
||||
let d = Document::from_text("Hello\nWorld");
|
||||
assert!(d.same_content(&d.clone()), "a copy is the same document");
|
||||
// The text changed.
|
||||
let mut e = d.clone();
|
||||
e.insert_text(&Pos::body(0, 0), "x", &CharProps::default()).unwrap();
|
||||
assert!(!d.same_content(&e), "the text changed");
|
||||
// The formatting changed.
|
||||
let mut e = d.clone();
|
||||
e.format_range(&Pos::body(0, 0), &Pos::body(0, 5), &|c| c.bold = Some(true)).unwrap();
|
||||
assert!(!d.same_content(&e), "the formatting changed");
|
||||
// The styles changed.
|
||||
let mut e = d.clone();
|
||||
e.styles.get_mut("Normal").unwrap().chr.bold = Some(true);
|
||||
assert!(!d.same_content(&e), "a style changed");
|
||||
// The lists changed.
|
||||
let mut e = d.clone();
|
||||
e.numbering.add_list(ListKind::Numbered);
|
||||
assert!(!d.same_content(&e), "a list was added");
|
||||
// The settings changed.
|
||||
let mut e = d.clone();
|
||||
e.settings.track_changes = true;
|
||||
assert!(!d.same_content(&e), "a setting changed");
|
||||
// The metadata changed.
|
||||
let mut e = d.clone();
|
||||
e.core.title = "Draft".into();
|
||||
assert!(!d.same_content(&e), "the title changed");
|
||||
// A part changed.
|
||||
let mut e = d.clone();
|
||||
e.add_part(PartKind::Header, vec![para_block(Paragraph::new())]);
|
||||
assert!(!d.same_content(&e), "a part was added");
|
||||
// A picture changed.
|
||||
let mut e = d.clone();
|
||||
e.media.insert("i.png".into(), Arc::new(vec![1, 2, 3]));
|
||||
assert!(!d.same_content(&e), "a picture was added");
|
||||
// A revision was recorded.
|
||||
let mut e = d.clone();
|
||||
e.revisions.push(Revision::new(RevisionKind::Insert, "Mira", "2026-10-23T00:00:00Z"));
|
||||
assert!(!d.same_content(&e), "a revision was recorded");
|
||||
// A comment was added.
|
||||
let mut e = d.clone();
|
||||
e.comments.insert(1, Comment { author: "Mira".into(), ..Default::default() });
|
||||
assert!(!d.same_content(&e), "a comment was added");
|
||||
// A paragraph that was marked as changed but holds the same content is the same.
|
||||
let mut e = d.clone();
|
||||
e.para_mut(StoryRef::Body, &Path::top(0)).unwrap().touch();
|
||||
assert!(d.same_content(&e), "the paragraph only changed its counter");
|
||||
// A table copied for an edit: the same until a cell's content changes.
|
||||
let mut d = Document::from_text("Hello");
|
||||
d.body.insert(0, Arc::new(Block::Table(Table::new(2, 2, 200.0))));
|
||||
let mut e = d.clone();
|
||||
Arc::make_mut(&mut e.body[0]);
|
||||
assert!(!Arc::ptr_eq(&d.body[0], &e.body[0]));
|
||||
assert!(d.same_content(&e), "the table was only copied");
|
||||
let Block::Table(t) = Arc::make_mut(&mut e.body[0]) else { panic!("a table") };
|
||||
t.rows[1].cells[1].blocks[0] = para_block(Paragraph::with_text("x", CharProps::default()));
|
||||
assert!(!d.same_content(&e), "a cell's text changed");
|
||||
// A building block was added.
|
||||
let mut e = d.clone();
|
||||
e.glossary.push(BuildingBlock { name: "Sign-off".into(), ..Default::default() });
|
||||
assert!(!d.same_content(&e), "a building block was added");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn json_round_trip() {
|
||||
let mut d = Document::from_text("Hello\nWorld");
|
||||
@@ -562,92 +457,3 @@ fn custom_props_set_get_remove() {
|
||||
assert!(!d.remove_custom_prop("other"));
|
||||
assert_eq!(d.custom_props.len(), 1);
|
||||
}
|
||||
|
||||
/// A paragraph that is nothing but `objs` (built directly: inserting one at a time is slow too).
|
||||
fn para_of_objects(objs: Vec<InlineObject>) -> Paragraph {
|
||||
let mut p = Paragraph::new();
|
||||
p.text = para::OBJ.to_string().repeat(objs.len());
|
||||
p.runs = vec![Run { len: p.text.len(), props: CharProps::default() }];
|
||||
p.objects = objs;
|
||||
p
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn object_walks_are_linear_in_the_paragraph() {
|
||||
// #3614: loops of object_offsets + object_at recounted the objects before each one, so a
|
||||
// paragraph of 40,000 markers took seconds per walk.
|
||||
let n = 40_000u32;
|
||||
let w = para::OBJ.len_utf8();
|
||||
let mut objs: Vec<InlineObject> = Vec::new();
|
||||
for i in 0..n {
|
||||
objs.push(InlineObject::BookmarkStart { name: format!("b{i}") });
|
||||
objs.push(fs("ADDIN x"));
|
||||
objs.push(InlineObject::PermStart { id: i, group: "everyone".into(), editor: String::new() });
|
||||
}
|
||||
// Ranges close in the order they opened (the worst case for a backwards search).
|
||||
for i in 0..n {
|
||||
objs.push(InlineObject::FieldEnd);
|
||||
objs.push(InlineObject::PermEnd { id: i });
|
||||
}
|
||||
let mut d = Document::new();
|
||||
d.body = vec![Arc::new(Block::Para(para_of_objects(objs)))];
|
||||
let t = std::time::Instant::now();
|
||||
assert_eq!(d.bookmarks().len(), n as usize);
|
||||
assert_eq!(d.field_ranges(StoryRef::Body).len(), n as usize);
|
||||
let perms = d.perm_ranges();
|
||||
assert_eq!(perms.len(), n as usize);
|
||||
assert!(perms.iter().all(|r| r.end.off > r.start.off), "every range closed by its own end");
|
||||
assert_eq!(d.text_box_anchor(7), None);
|
||||
// A caret before a long row of field ends moves past all of them.
|
||||
let first_end = 3 * n as usize * w;
|
||||
let p = d.past_field_ends(&Pos::body(0, first_end));
|
||||
assert_eq!(p.off, first_end + w);
|
||||
// Under 0.3 s on a loaded machine; the old walks took minutes (360 s for the first test).
|
||||
assert!(t.elapsed().as_secs_f32() < 10.0, "{:?}", t.elapsed());
|
||||
|
||||
let ends: Vec<InlineObject> = (0..n).map(|_| InlineObject::FieldEnd).collect();
|
||||
d.body = vec![Arc::new(Block::Para(para_of_objects(ends)))];
|
||||
let t = std::time::Instant::now();
|
||||
assert_eq!(d.past_field_ends(&Pos::body(0, 0)).off, n as usize * w);
|
||||
// Typing at the end of a row of block-level control starts.
|
||||
let block = ContentControl { block: true, ..Default::default() };
|
||||
let starts: Vec<InlineObject> = (0..n).map(|_| InlineObject::ControlStart { control: Box::new(block.clone()) }).collect();
|
||||
d.body = vec![Arc::new(Block::Para(para_of_objects(starts)))];
|
||||
let end = Pos::body(0, n as usize * w);
|
||||
assert_eq!(d.typing_pos(&end), end);
|
||||
assert!(t.elapsed().as_secs_f32() < 10.0, "{:?}", t.elapsed());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn deleting_across_many_lone_control_markers_is_not_quadratic() {
|
||||
// A millisecond now; the old delete took 18 s at 1,500 markers and 114 s at 3,000.
|
||||
let n = 3_000;
|
||||
let starts: Vec<InlineObject> = (0..n).map(|_| InlineObject::ControlStart { control: Box::default() }).collect();
|
||||
let mut d = Document::from_text("before\n\nafter");
|
||||
d.body[1] = Arc::new(Block::Para(para_of_objects(starts)));
|
||||
let t = std::time::Instant::now();
|
||||
let at = d.delete_range(&Pos::body(0, 3), &Pos::body(2, 2)).unwrap();
|
||||
assert_eq!(at, Pos::body(0, 3));
|
||||
assert_eq!(d.plain_text(StoryRef::Body), "befter");
|
||||
assert!(t.elapsed().as_secs_f32() < 10.0, "{:?}", t.elapsed());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn control_markers_kept_are_sorted() {
|
||||
// `delete_range` groups the kept markers by paragraph and binary-searches each group, so
|
||||
// `kept_control_markers` must hand them over sorted and without repeats (review of #3874).
|
||||
let start = || InlineObject::ControlStart { control: Box::default() };
|
||||
let mut d = Document::from_text("before");
|
||||
// Controls open and close across paragraphs: 1 closes the one 0 opens, 3 the one 2 opens…
|
||||
for i in 0..12 {
|
||||
let objs = if i % 2 == 0 { vec![start(), start()] } else { vec![InlineObject::ControlEnd, InlineObject::ControlEnd] };
|
||||
d.body.push(Arc::new(Block::Para(para_of_objects(objs))));
|
||||
}
|
||||
d.body.push(Arc::new(Block::Para(Paragraph::with_text("after", CharProps::default()))));
|
||||
let w = para::OBJ.len_utf8();
|
||||
for (a, b) in [(Pos::body(2, w), Pos::body(11, w)), (Pos::body(1, w), Pos::body(12, w)), (Pos::body(3, w), Pos::body(8, 0))] {
|
||||
let kept = d.kept_control_markers(&a, &b);
|
||||
assert!(!kept.is_empty(), "{a:?}..{b:?}");
|
||||
assert!(kept.windows(2).all(|p| p[0] < p[1]), "sorted, no repeats: {kept:?}");
|
||||
}
|
||||
}
|
||||
|
||||
+5
-34
@@ -92,7 +92,11 @@ pub fn shape_outline(kind: ShapeKind, w: f32, h: f32, corner: Option<u32>) -> Op
|
||||
ShapeKind::Triangle => vec![(0.5, 0.0), (1.0, 1.0), (0.0, 1.0)],
|
||||
ShapeKind::RightTriangle => vec![(0.0, 0.0), (1.0, 1.0), (0.0, 1.0)],
|
||||
ShapeKind::Diamond => vec![(0.5, 0.0), (1.0, 0.5), (0.5, 1.0), (0.0, 0.5)],
|
||||
ShapeKind::Arrow | ShapeKind::ArrowLeft | ShapeKind::ArrowUp | ShapeKind::ArrowDown => arrow_outline(kind, w, h),
|
||||
ShapeKind::Arrow => {
|
||||
// As drawn: the head is half the height long, the shaft 40% of the height thick.
|
||||
let head = if w > 0.0 { (0.5 * h / w).min(1.0) } else { 0.5 };
|
||||
vec![(0.0, 0.3), (1.0 - head, 0.3), (1.0 - head, 0.0), (1.0, 0.5), (1.0 - head, 1.0), (1.0 - head, 0.7), (0.0, 0.7)]
|
||||
}
|
||||
ShapeKind::Star => (0..10)
|
||||
.map(|i| {
|
||||
let a = PI * i as f32 / 5.0 - FRAC_PI_2;
|
||||
@@ -122,26 +126,6 @@ pub fn shape_outline(kind: ShapeKind, w: f32, h: f32, corner: Option<u32>) -> Op
|
||||
Some(pts)
|
||||
}
|
||||
|
||||
/// A block arrow's outline in unit coordinates (0..1 across a `w` × `h` frame), pointing the way
|
||||
/// `kind` says (right for anything but the left, up and down arrows). The head is half the arrow's
|
||||
/// thickness long, the shaft 40% of the thickness thick.
|
||||
pub fn arrow_outline(kind: ShapeKind, w: f32, h: f32) -> Vec<(f32, f32)> {
|
||||
let vertical = matches!(kind, ShapeKind::ArrowUp | ShapeKind::ArrowDown);
|
||||
let (len, thick) = if vertical { (h, w) } else { (w, h) };
|
||||
let head = if len > 0.0 { (0.5 * thick / len).min(1.0) } else { 0.5 };
|
||||
// Pointing right: `u` runs along the arrow from its tail, `v` across it.
|
||||
let right = [(0.0, 0.3), (1.0 - head, 0.3), (1.0 - head, 0.0), (1.0, 0.5), (1.0 - head, 1.0), (1.0 - head, 0.7), (0.0, 0.7)];
|
||||
right
|
||||
.iter()
|
||||
.map(|&(u, v)| match kind {
|
||||
ShapeKind::ArrowLeft => (1.0 - u, v),
|
||||
ShapeKind::ArrowUp => (v, 1.0 - u),
|
||||
ShapeKind::ArrowDown => (v, u),
|
||||
_ => (u, v),
|
||||
})
|
||||
.collect()
|
||||
}
|
||||
|
||||
/// Largest side of the grid [`alpha_outline`] looks at.
|
||||
pub const ALPHA_GRID: usize = 48;
|
||||
|
||||
@@ -230,16 +214,6 @@ mod tests {
|
||||
assert!(WrapPolygon::from_unit(false, &[(f32::NAN, 0.0), (0.0, 0.0), (1.0, 0.0), (0.5, 1.0)]).is_some_and(|p| p.points.len() == 3));
|
||||
}
|
||||
|
||||
/// #3919: the tip of each block arrow lies on the side it points to.
|
||||
#[test]
|
||||
fn arrow_outlines_point_the_right_way() {
|
||||
let tip = |kind| shape_outline(kind, 100.0, 100.0, None).unwrap()[3];
|
||||
assert_eq!(tip(ShapeKind::Arrow), (1.0, 0.5));
|
||||
assert_eq!(tip(ShapeKind::ArrowLeft), (0.0, 0.5));
|
||||
assert_eq!(tip(ShapeKind::ArrowUp), (0.5, 0.0));
|
||||
assert_eq!(tip(ShapeKind::ArrowDown), (0.5, 1.0));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn shape_outlines_stay_in_the_frame() {
|
||||
for kind in [
|
||||
@@ -249,9 +223,6 @@ mod tests {
|
||||
ShapeKind::RightTriangle,
|
||||
ShapeKind::Diamond,
|
||||
ShapeKind::Arrow,
|
||||
ShapeKind::ArrowLeft,
|
||||
ShapeKind::ArrowUp,
|
||||
ShapeKind::ArrowDown,
|
||||
ShapeKind::Star,
|
||||
ShapeKind::Heart,
|
||||
] {
|
||||
|
||||
@@ -20,6 +20,5 @@ log = { workspace = true }
|
||||
chrono = { workspace = true }
|
||||
|
||||
[dev-dependencies]
|
||||
wordcraft-doc = { workspace = true, features = ["test-support"] }
|
||||
proptest = { workspace = true }
|
||||
serde_json = { workspace = true }
|
||||
|
||||
@@ -295,13 +295,8 @@ impl Cipher {
|
||||
if hflags & H_EXTERNAL != 0 {
|
||||
return Err(unsupported("the document uses an external encryption provider"));
|
||||
}
|
||||
// [MS-OFFCRYPTO] defines AES only for the encrypted packages of Word 2007 and later
|
||||
// (which open through the .docx reader whatever their extension); a Word 97-2003 file
|
||||
// may only use XOR, RC4 or RC4 CryptoAPI, so there is no public scheme to follow here.
|
||||
if hflags & H_AES != 0 {
|
||||
return Err(unsupported(
|
||||
"this Word 97-2003 file claims AES encryption, which is only defined for the newer encrypted-package format; it can't be opened",
|
||||
));
|
||||
return Err(unsupported("AES-encrypted .doc files are not supported"));
|
||||
}
|
||||
if hflags & H_CRYPTO_API == 0 {
|
||||
return Err(unsupported("unknown encryption header"));
|
||||
|
||||
@@ -81,18 +81,12 @@ pub(crate) fn anchors(
|
||||
) -> Vec<(u32, InlineObject)> {
|
||||
let Some(pair) = pair else { return Vec::new() };
|
||||
let mut out = Vec::new();
|
||||
let mut seen = std::collections::HashSet::new();
|
||||
for (cp, spa) in plc(table, pair, CB_SPA) {
|
||||
let i32_at = |o: usize| spa.get(o..o + 4).and_then(|b| b.try_into().ok()).map(i32::from_le_bytes).unwrap_or(0);
|
||||
let spid = i32_at(0) as u32;
|
||||
let [xl, yt, xr, yb] = [4, 8, 12, 16].map(i32_at);
|
||||
let flags = spa.get(20..22).map(|b| u16::from_le_bytes([b[0], b[1]])).unwrap_or(0);
|
||||
let Some(node) = shapes.get(&spid) else { continue };
|
||||
// Word anchors each shape once. A repeated spid would build another copy of the shape,
|
||||
// and of every member of a group, per anchor (#3710).
|
||||
if !seen.insert(spid) {
|
||||
continue;
|
||||
}
|
||||
let tw = |v: i64| (v as f32 / 20.0).clamp(-2.0 * MAX_PT, 2.0 * MAX_PT);
|
||||
let (mut x, mut y) = (tw(xl as i64), tw(yt as i64));
|
||||
let (mut w, mut h) = (tw(xr as i64 - xl as i64).clamp(0.0, MAX_PT), tw(yb as i64 - yt as i64).clamp(0.0, MAX_PT));
|
||||
@@ -196,10 +190,7 @@ fn leaf(sp: &Sp, w: f32, h: f32, float: Float, b: &Build, pics: &mut Pictures) -
|
||||
5 => ShapeKind::Triangle,
|
||||
6 => ShapeKind::RightTriangle,
|
||||
12 | 187 => ShapeKind::Star,
|
||||
13 | 69 => ShapeKind::Arrow,
|
||||
66 => ShapeKind::ArrowLeft,
|
||||
67 => ShapeKind::ArrowDown,
|
||||
68 => ShapeKind::ArrowUp,
|
||||
13 | 66 | 67 | 68 | 69 => ShapeKind::Arrow,
|
||||
20 | 32 => ShapeKind::Line,
|
||||
74 => ShapeKind::Heart,
|
||||
202 => ShapeKind::TextBox,
|
||||
|
||||
+29
-191
@@ -49,91 +49,6 @@ const MAX_PARTS: usize = 50_000;
|
||||
/// headers, notes, text boxes), whatever the file's size. The working budget is tied to the
|
||||
/// WordDocument stream (see [`read`]); this caps even a maximal stream.
|
||||
const MAX_TOTAL_CHARS: u64 = 64 << 20;
|
||||
/// Bytes of repeated work per document beyond the text itself: piece Prm grpprls (Prcs, up
|
||||
/// to 64 KB each) interpreted (#3692), hyperlink targets (#3708), picture alt text (#3709)
|
||||
/// and floating objects (#3710) copied into every run or anchor. Past it Prms are skipped,
|
||||
/// copied strings left empty and floating objects dropped, each kind logged once a document
|
||||
/// ([`Work::log`]).
|
||||
const MAX_WORK: u64 = 64 << 20;
|
||||
|
||||
/// What the [`MAX_WORK`] budget pays for, and leaves out once it is spent.
|
||||
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
|
||||
enum Cost {
|
||||
/// A piece's Prm grpprl: past the budget the piece keeps its CHPX formatting only.
|
||||
Formatting,
|
||||
/// A hyperlink target copied into a run of its result: past it the text is kept, unlinked.
|
||||
LinkTarget,
|
||||
/// A picture's alt text copied into an anchor: past it the picture has none.
|
||||
AltText,
|
||||
/// A floating object copied into its anchor: past it the object is left out.
|
||||
FloatingObject,
|
||||
}
|
||||
|
||||
impl Cost {
|
||||
const ALL: [Cost; 4] = [Cost::Formatting, Cost::LinkTarget, Cost::AltText, Cost::FloatingObject];
|
||||
|
||||
/// What is left out, for the log.
|
||||
fn left_out(self) -> &'static str {
|
||||
match self {
|
||||
Cost::Formatting => "some text keeps only part of its formatting",
|
||||
Cost::LinkTarget => "some hyperlinks' text is kept without its link",
|
||||
Cost::AltText => "some pictures are left without their alt text",
|
||||
Cost::FloatingObject => "some floating shapes and pictures are left out",
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// One document's [`MAX_WORK`] budget, and what it has left out (each kind logged once when
|
||||
/// the document has been read, [`Work::log`]).
|
||||
struct Work {
|
||||
left: std::cell::Cell<u64>,
|
||||
dropped: std::cell::Cell<[bool; 4]>,
|
||||
}
|
||||
|
||||
impl Work {
|
||||
fn new(budget: u64) -> Self {
|
||||
Work { left: std::cell::Cell::new(budget), dropped: std::cell::Cell::new([false; 4]) }
|
||||
}
|
||||
|
||||
/// Spends `cost` bytes on `what`: false (and `what` noted as left out) when they aren't left.
|
||||
fn charge(&self, cost: u64, what: Cost) -> bool {
|
||||
match self.left.get().checked_sub(cost) {
|
||||
Some(left) => {
|
||||
self.left.set(left);
|
||||
true
|
||||
}
|
||||
None => {
|
||||
let mut dropped = self.dropped.get();
|
||||
if let Some(d) = Cost::ALL.iter().position(|c| *c == what).and_then(|i| dropped.get_mut(i)) {
|
||||
*d = true;
|
||||
}
|
||||
self.dropped.set(dropped);
|
||||
false
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// The kinds of work left out so far, in [`Cost::ALL`] order.
|
||||
fn dropped(&self) -> Vec<Cost> {
|
||||
Cost::ALL.iter().zip(self.dropped.get()).filter(|(_, d)| *d).map(|(c, _)| *c).collect()
|
||||
}
|
||||
|
||||
/// Logs one warning for each kind left out.
|
||||
fn log(&self) {
|
||||
let dropped = self.dropped();
|
||||
for c in &dropped {
|
||||
log::warn!(".doc: the document repeats more than WordCraft copies ({MAX_WORK} bytes): {}", c.left_out());
|
||||
}
|
||||
#[cfg(test)]
|
||||
LAST_DROPPED.with(|d| *d.borrow_mut() = dropped);
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
thread_local! {
|
||||
/// What the last document read on this thread left out for the work budget ([`Work::log`]).
|
||||
static LAST_DROPPED: std::cell::RefCell<Vec<Cost>> = const { std::cell::RefCell::new(Vec::new()) };
|
||||
}
|
||||
/// `sprmCPicLocation`: the fc of a picture in the Data stream.
|
||||
const C_PIC_LOCATION: u16 = 0x6A03;
|
||||
/// `sprmCSymbol`: the record font and character of a symbol character ([MS-DOC] §2.6.1); the
|
||||
@@ -173,16 +88,10 @@ pub enum DocbinError {
|
||||
/// Is this `.doc` password-protected (XOR obfuscation, RC4 or RC4 CryptoAPI)? False for
|
||||
/// anything that is not a readable Word 97-2003 file.
|
||||
pub fn is_encrypted(bytes: &[u8]) -> bool {
|
||||
let Ok(mut comp) = open_container(bytes) else { return false };
|
||||
let Ok(mut comp) = cfb::CompoundFile::open(std::io::Cursor::new(bytes)) else { return false };
|
||||
stream(&mut comp, "WordDocument").is_some_and(|w| fib::peek_encryption(&w).is_some())
|
||||
}
|
||||
|
||||
/// Open the compound file around a `.doc`, after [`wordcraft_doc::cfb_guard::check`].
|
||||
fn open_container(bytes: &[u8]) -> Result<cfb::CompoundFile<std::io::Cursor<&[u8]>>, DocbinError> {
|
||||
wordcraft_doc::cfb_guard::check(bytes).map_err(DocbinError::Container)?;
|
||||
cfb::CompoundFile::open(std::io::Cursor::new(bytes)).map_err(|e| DocbinError::Container(e.to_string()))
|
||||
}
|
||||
|
||||
/// Read a `.doc` (Word 97-2003 binary) file into a [`Document`]. A password-protected file
|
||||
/// gives [`DocbinError::PasswordRequired`]; see [`read_with_password`].
|
||||
pub fn read(bytes: &[u8]) -> Result<Document, DocbinError> {
|
||||
@@ -193,7 +102,7 @@ pub fn read(bytes: &[u8]) -> Result<Document, DocbinError> {
|
||||
/// verifier ([`DocbinError::WrongPassword`] when it does not match) and then decrypted. The
|
||||
/// password is ignored for an unprotected file.
|
||||
pub fn read_with_password(bytes: &[u8], password: Option<&str>) -> Result<Document, DocbinError> {
|
||||
let mut comp = open_container(bytes)?;
|
||||
let mut comp = cfb::CompoundFile::open(std::io::Cursor::new(bytes)).map_err(|e| DocbinError::Container(e.to_string()))?;
|
||||
let mut word = stream(&mut comp, "WordDocument").ok_or_else(|| DocbinError::NotWord("no WordDocument stream".into()))?;
|
||||
if word.len() as u64 > MAX_STREAM {
|
||||
return Err(DocbinError::Limit(format!("WordDocument stream is {} bytes", word.len())));
|
||||
@@ -259,14 +168,13 @@ pub fn read_with_password(bytes: &[u8], password: Option<&str>) -> Result<Docume
|
||||
let ftn_base = fib.ccp.text;
|
||||
let hdd_base = fib.ccp.text.saturating_add(fib.ccp.ftn);
|
||||
let edn_base = hdd_base.saturating_add(fib.ccp.hdd).saturating_add(fib.ccp.atn);
|
||||
let mut pics = media::Pictures::for_input(word.len().saturating_add(data.len()));
|
||||
let mut pics = media::Pictures::default();
|
||||
// Every materialised character comes from at least one byte of the WordDocument stream, so a
|
||||
// legitimate document's total text (body plus every header, footnote and text box, each read
|
||||
// once) cannot exceed its stream. A hostile piece or story table that aliases the same bytes
|
||||
// many times can, so the walk is held to twice the stream's size — generous for real files,
|
||||
// tight against aliasing — and never past the absolute ceiling.
|
||||
let budget = std::cell::Cell::new((word.len() as u64).saturating_mul(2).min(MAX_TOTAL_CHARS));
|
||||
let work = Work::new(MAX_WORK);
|
||||
|
||||
let ctx = WalkCtx {
|
||||
word: &word,
|
||||
@@ -282,7 +190,6 @@ pub fn read_with_password(bytes: &[u8], password: Option<&str>) -> Result<Docume
|
||||
drawn: &[],
|
||||
sect_ends: &[],
|
||||
budget: &budget,
|
||||
work: &work,
|
||||
};
|
||||
|
||||
// Drawings: text box stories become parts first, then the floating shapes and pictures
|
||||
@@ -391,7 +298,6 @@ pub fn read_with_password(bytes: &[u8], password: Option<&str>) -> Result<Docume
|
||||
doc.media.extend(pics.media);
|
||||
doc.body = table::assemble(paras);
|
||||
doc.ensure_nonempty();
|
||||
work.log();
|
||||
Ok(doc)
|
||||
}
|
||||
|
||||
@@ -428,8 +334,6 @@ struct WalkCtx<'a> {
|
||||
/// footnotes, text boxes) are hostile and can each span a whole subdocument, so thousands of
|
||||
/// them could otherwise materialise the same text thousands of times; this bounds the total.
|
||||
budget: &'a std::cell::Cell<u64>,
|
||||
/// Repeated-work bytes left ([`MAX_WORK`]), and what they left out.
|
||||
work: &'a Work,
|
||||
}
|
||||
|
||||
/// State while assembling one paragraph from per-CP formatting.
|
||||
@@ -461,14 +365,6 @@ impl ParaBuild {
|
||||
}
|
||||
self.text.push(c);
|
||||
}
|
||||
|
||||
/// Append `c` to the current run, whose properties the caller knows it matches.
|
||||
fn extend(&mut self, c: char) {
|
||||
if let Some(r) = self.runs.last_mut() {
|
||||
r.len += c.len_utf8();
|
||||
}
|
||||
self.text.push(c);
|
||||
}
|
||||
}
|
||||
|
||||
/// Walk a CP range, splitting paragraphs at 0x0D/0x07 marks, applying the direct character
|
||||
@@ -480,7 +376,7 @@ impl ParaBuild {
|
||||
/// The range is clamped to the last piece's CP, and unreadable stretches (gaps, truncated
|
||||
/// pieces) are skipped a whole piece at a time, so the work is bounded by the text that
|
||||
/// really exists, not by the CP numbers a file claims.
|
||||
fn walk<'a>(ctx: &WalkCtx<'a>, cp_start: u32, cp_end: u32, pics: &mut media::Pictures) -> Vec<ParaOut> {
|
||||
fn walk(ctx: &WalkCtx, cp_start: u32, cp_end: u32, pics: &mut media::Pictures) -> Vec<ParaOut> {
|
||||
let cp_end = cp_end.min(ctx.pieces.cp_end());
|
||||
let mut out = Vec::new();
|
||||
let mut pb = ParaBuild::default();
|
||||
@@ -489,7 +385,6 @@ fn walk<'a>(ctx: &WalkCtx<'a>, cp_start: u32, cp_end: u32, pics: &mut media::Pic
|
||||
// Field assembly state: instruction characters are swallowed, the result is kept until
|
||||
// the end character, then emitted as one object (hyperlinks keep their result text).
|
||||
let mut field: Option<FieldBuild> = None;
|
||||
let mut memo: Option<FmtMemo<'a>> = None;
|
||||
while cp < cp_end {
|
||||
// Bookmark boundaries at this CP (main-document walks only; subdocument walks get
|
||||
// an empty table and `cp` never matches the main-document CPs).
|
||||
@@ -595,7 +490,7 @@ fn walk<'a>(ctx: &WalkCtx<'a>, cp_start: u32, cp_end: u32, pics: &mut media::Pic
|
||||
f.depth -= 1;
|
||||
field = Some(f);
|
||||
} else {
|
||||
finish_field(&mut pb, f, ctx.work);
|
||||
finish_field(&mut pb, f);
|
||||
}
|
||||
}
|
||||
cp = cp.saturating_add(1);
|
||||
@@ -607,15 +502,12 @@ fn walk<'a>(ctx: &WalkCtx<'a>, cp_start: u32, cp_end: u32, pics: &mut media::Pic
|
||||
let fc_pic = pic_location(ctx.chpx_bins.chpx(ctx.word, fc));
|
||||
match fc_pic.and_then(|at| pics.get(ctx.data, at)) {
|
||||
Some(p) => {
|
||||
// Every anchor of a picture gets its own copy of the alt text; the
|
||||
// copies are charged to the work budget (#3709).
|
||||
let alt = if ctx.work.charge(p.alt.len() as u64, Cost::AltText) { p.alt.clone() } else { String::new() };
|
||||
pb.push('\u{FFFC}', CharProps::default());
|
||||
pb.objects.push(InlineObject::Image {
|
||||
media: p.key.clone(),
|
||||
media: p.key,
|
||||
w: p.w,
|
||||
h: p.h,
|
||||
alt,
|
||||
alt: p.alt,
|
||||
float: Default::default(),
|
||||
crop: p.crop,
|
||||
ole: None,
|
||||
@@ -627,12 +519,7 @@ fn walk<'a>(ctx: &WalkCtx<'a>, cp_start: u32, cp_end: u32, pics: &mut media::Pic
|
||||
'\u{8}' => {
|
||||
// Drawn object anchor: the floating shape or picture anchored at this CP.
|
||||
let at = ctx.drawn.partition_point(|(c, _)| *c < cp);
|
||||
// Each copy (a story can be walked more than once) is charged to the work budget
|
||||
// by what a clone copies: a picture's alt text, a group with all its members (#3710).
|
||||
for (_, obj) in ctx.drawn.get(at..).unwrap_or_default().iter().take_while(|(c, _)| *c == cp) {
|
||||
if !ctx.work.charge(object_cost(obj), Cost::FloatingObject) {
|
||||
continue;
|
||||
}
|
||||
pb.push('\u{FFFC}', CharProps::default());
|
||||
pb.objects.push(obj.clone());
|
||||
}
|
||||
@@ -645,16 +532,16 @@ fn walk<'a>(ctx: &WalkCtx<'a>, cp_start: u32, cp_end: u32, pics: &mut media::Pic
|
||||
pb.objects.push(InlineObject::NoteRef { kind: *kind, id: *id, custom: String::new() });
|
||||
}
|
||||
}
|
||||
'\t' => push_formatted(&mut pb, cp, '\t', ctx, field.as_mut(), &mut memo),
|
||||
'\u{B}' => push_formatted(&mut pb, cp, '\n', ctx, field.as_mut(), &mut memo),
|
||||
'\u{C}' => push_formatted(&mut pb, cp, '\u{C}', ctx, field.as_mut(), &mut memo),
|
||||
'\u{E}' => push_formatted(&mut pb, cp, '\u{E}', ctx, field.as_mut(), &mut memo),
|
||||
'\u{1E}' => push_formatted(&mut pb, cp, '\u{2011}', ctx, field.as_mut(), &mut memo),
|
||||
'\u{1F}' => push_formatted(&mut pb, cp, '\u{AD}', ctx, field.as_mut(), &mut memo),
|
||||
'\t' => push_formatted(&mut pb, cp, '\t', ctx, field.as_mut()),
|
||||
'\u{B}' => push_formatted(&mut pb, cp, '\n', ctx, field.as_mut()),
|
||||
'\u{C}' => push_formatted(&mut pb, cp, '\u{C}', ctx, field.as_mut()),
|
||||
'\u{E}' => push_formatted(&mut pb, cp, '\u{E}', ctx, field.as_mut()),
|
||||
'\u{1E}' => push_formatted(&mut pb, cp, '\u{2011}', ctx, field.as_mut()),
|
||||
'\u{1F}' => push_formatted(&mut pb, cp, '\u{AD}', ctx, field.as_mut()),
|
||||
// Object anchors, comment references and other special characters are dropped
|
||||
// (or handled by later passes).
|
||||
c if (c as u32) < 0x20 => {}
|
||||
c => push_formatted(&mut pb, cp, c, ctx, field.as_mut(), &mut memo),
|
||||
c => push_formatted(&mut pb, cp, c, ctx, field.as_mut()),
|
||||
}
|
||||
cp = cp.saturating_add(units);
|
||||
}
|
||||
@@ -677,44 +564,19 @@ struct FieldBuild {
|
||||
sep: Option<Vec<(char, CharProps)>>,
|
||||
}
|
||||
|
||||
/// Bytes a clone of `obj` copies: the object itself plus a picture's alt text, a group with all
|
||||
/// its members (#3710).
|
||||
fn object_cost(obj: &InlineObject) -> u64 {
|
||||
let shallow = std::mem::size_of::<InlineObject>() as u64;
|
||||
match obj {
|
||||
InlineObject::Image { alt, .. } => shallow.saturating_add(alt.len() as u64),
|
||||
InlineObject::Group { children, .. } => children.iter().fold(shallow, |n, c| n.saturating_add(object_cost(&c.obj))),
|
||||
_ => shallow,
|
||||
}
|
||||
}
|
||||
|
||||
/// Emit a finished field: HYPERLINK keeps its result text as a linked run; every other
|
||||
/// field becomes one inline object with its cached result.
|
||||
///
|
||||
/// The target is copied once per run of equally formatted result text, not per character,
|
||||
/// and each copy is charged to `work` ([`MAX_WORK`]); once that is spent the rest of the
|
||||
/// result is kept as plain text (#3708).
|
||||
fn finish_field(pb: &mut ParaBuild, f: FieldBuild, work: &Work) {
|
||||
fn finish_field(pb: &mut ParaBuild, f: FieldBuild) {
|
||||
let instr = f.instr.trim().to_string();
|
||||
let result = f.sep.unwrap_or_default();
|
||||
let result_text: String = result.iter().map(|(c, _)| *c).collect();
|
||||
if let Some(target) = hyperlink_target(&instr) {
|
||||
let cost = target.len() as u64;
|
||||
let mut run: Option<CharProps> = None;
|
||||
for (c, props) in result {
|
||||
if run.as_ref() == Some(&props) {
|
||||
pb.extend(c);
|
||||
continue;
|
||||
}
|
||||
let mut linked = props.clone();
|
||||
if work.charge(cost, Cost::LinkTarget) {
|
||||
linked.link = Some(target.clone());
|
||||
}
|
||||
pb.push(c, linked);
|
||||
run = Some(props);
|
||||
for (c, mut props) in result {
|
||||
props.link = Some(target.clone());
|
||||
pb.push(c, props);
|
||||
}
|
||||
return;
|
||||
}
|
||||
let result_text: String = result.iter().map(|(c, _)| *c).collect();
|
||||
pb.push('\u{FFFC}', CharProps::default());
|
||||
pb.objects.push(InlineObject::Field { instr, result: result_text, locked: false, dirty: false, code: false, form: String::new() });
|
||||
}
|
||||
@@ -840,28 +702,19 @@ fn num_of(ilfo: i32) -> Option<u32> {
|
||||
/// plus the piece's Prm, over the paragraph style's base character props. Inside a field,
|
||||
/// the character is recorded into the instruction or the cached result instead of (or as
|
||||
/// well as) the visible text.
|
||||
fn push_formatted<'a>(pb: &mut ParaBuild, cp: u32, c: char, ctx: &WalkCtx<'a>, field: Option<&mut FieldBuild>, memo: &mut Option<FmtMemo<'a>>) {
|
||||
fn push_formatted(pb: &mut ParaBuild, cp: u32, c: char, ctx: &WalkCtx, field: Option<&mut FieldBuild>) {
|
||||
let fc = ctx.pieces.fc_of_cp(cp).unwrap_or(0);
|
||||
let grpprl = ctx.chpx_bins.chpx(ctx.word, fc);
|
||||
let prm = ctx.pieces.prm_of(cp);
|
||||
let style = pb.props.style.as_deref();
|
||||
let mut props = match memo.as_ref().filter(|m| std::ptr::eq(m.chpx, grpprl) && m.prm == prm && m.style.as_deref() == style) {
|
||||
Some(m) => m.props.clone(),
|
||||
let default;
|
||||
let base = match pb.props.style.as_deref().and_then(|id| ctx.sheet.get(id)) {
|
||||
Some(st) => &st.chr,
|
||||
None => {
|
||||
let default;
|
||||
let base = match style.and_then(|id| ctx.sheet.get(id)) {
|
||||
Some(st) => &st.chr,
|
||||
None => {
|
||||
default = CharProps::default();
|
||||
&default
|
||||
}
|
||||
};
|
||||
let mut props = char_props(grpprl, ctx.fonts, base);
|
||||
apply_prm(&mut props, ctx.pieces.piece_prm(cp), ctx.fonts, base, ctx.work);
|
||||
*memo = Some(FmtMemo { chpx: grpprl, prm, style: style.map(str::to_owned), props: props.clone() });
|
||||
props
|
||||
default = CharProps::default();
|
||||
&default
|
||||
}
|
||||
};
|
||||
let mut props = char_props(grpprl, ctx.fonts, base);
|
||||
apply_prm(&mut props, ctx.pieces.piece_prm(cp), ctx.fonts, base);
|
||||
// A symbol character (Insert › Symbol) is stored as the placeholder U+0028 with
|
||||
// sprmCSymbol carrying the font and the character code ([MS-DOC] §2.6.1).
|
||||
let mut c = c;
|
||||
@@ -886,20 +739,8 @@ fn push_formatted<'a>(pb: &mut ParaBuild, cp: u32, c: char, ctx: &WalkCtx<'a>, f
|
||||
}
|
||||
}
|
||||
|
||||
/// The last character formatting resolved, reused while the CHPX, the piece's Prm and the
|
||||
/// paragraph style stay the same, so a Prm grpprl (up to 64 KB) is interpreted once per
|
||||
/// formatting change, not once per character (#3692).
|
||||
struct FmtMemo<'a> {
|
||||
/// The CHPX grpprl, compared by address: it is a slice of the WordDocument stream.
|
||||
chpx: &'a [u8],
|
||||
prm: u16,
|
||||
style: Option<String>,
|
||||
props: CharProps,
|
||||
}
|
||||
|
||||
/// Apply a piece's Prm (Prm0 short sprm or Prm1 grpprl) to character properties. A Prm1
|
||||
/// grpprl is charged to `work`; once that is spent the grpprl is skipped.
|
||||
fn apply_prm(props: &mut CharProps, prm: PrmRef<'_>, fonts: &[String], base: &CharProps, work: &Work) {
|
||||
/// Apply a piece's Prm (Prm0 short sprm or Prm1 grpprl) to character properties.
|
||||
fn apply_prm(props: &mut CharProps, prm: PrmRef<'_>, fonts: &[String], base: &CharProps) {
|
||||
match prm {
|
||||
PrmRef::None => {}
|
||||
PrmRef::Sprm0 { isprm, val } => {
|
||||
@@ -912,9 +753,6 @@ fn apply_prm(props: &mut CharProps, prm: PrmRef<'_>, fonts: &[String], base: &Ch
|
||||
}
|
||||
}
|
||||
PrmRef::Grpprl(Some(g)) => {
|
||||
if !work.charge(g.len() as u64, Cost::Formatting) {
|
||||
return;
|
||||
}
|
||||
for prl in sprm::iter(g) {
|
||||
if prl.sgc() == 2 {
|
||||
fmt::apply_char(props, &prl, base, fonts);
|
||||
|
||||
+26
-68
@@ -17,24 +17,6 @@ const OFF_PICMID: usize = 28;
|
||||
const MAX_PICTURES: usize = 5_000;
|
||||
/// Most picture bytes one document copies out of the Data stream, all pictures together.
|
||||
pub(crate) const MAX_MEDIA_BYTES: usize = 256 << 20;
|
||||
/// Most bytes one document decodes pictures into (metafiles inflated, DIB pixels), charged
|
||||
/// per attempt whether it succeeds or not: retained media is capped separately, by
|
||||
/// `MAX_MEDIA_BYTES` (#3711, #3712).
|
||||
const MAX_DECODE_WORK: usize = 512 << 20;
|
||||
/// The decode budget grows with the input: up to this many bytes per byte of the WordDocument
|
||||
/// and Data streams. An uncompressed DIB declares at most 32 bytes of pixels per file byte (1
|
||||
/// bit per pixel, 4 bytes charged each), so 64 passes any such picture. RLE8/RLE4 bitmaps
|
||||
/// reach about 510 times that and rely on the `MAX_DECODE_WORK` floor instead; a decompression
|
||||
/// bomb (ratios around 1e6) still hits the cap.
|
||||
const DECODE_WORK_PER_INPUT_BYTE: usize = 64;
|
||||
/// The most the decode budget grows to, whatever the input.
|
||||
const MAX_DECODE_WORK_CEILING: usize = 2 << 30;
|
||||
|
||||
/// The decode budget for a document whose pictures live in `input` stream bytes, never less
|
||||
/// than `floor`.
|
||||
pub(crate) fn decode_work(input: usize, floor: usize) -> usize {
|
||||
input.saturating_mul(DECODE_WORK_PER_INPUT_BYTE).clamp(floor, floor.max(MAX_DECODE_WORK_CEILING))
|
||||
}
|
||||
|
||||
/// The document's pictures, keyed by their location in the Data stream: anchors that share a
|
||||
/// `fc_pic` share one media entry (or one failure), and the total bytes copied are capped, so a
|
||||
@@ -46,17 +28,11 @@ pub(crate) struct Pictures {
|
||||
by_fc: HashMap<u32, Option<Inline>>,
|
||||
/// BLIP store index → media key (`None`: unreadable, not retried).
|
||||
by_bse: HashMap<u32, Option<String>>,
|
||||
/// (stream, foDelay, record length) of a BLIP record outside the store → media key: store
|
||||
/// entries and inline pictures that alias one record share one decode (`None`: failed, not
|
||||
/// retried). The stream is told apart by its address.
|
||||
by_delay: HashMap<(usize, usize, usize), Option<String>>,
|
||||
/// The media entries, merged into the document at the end.
|
||||
pub(crate) media: BTreeMap<String, Arc<Vec<u8>>>,
|
||||
total: usize,
|
||||
/// The byte cap (`MAX_MEDIA_BYTES`; smaller in tests).
|
||||
pub(crate) limit: usize,
|
||||
/// Decode work left (`MAX_DECODE_WORK`; smaller in tests).
|
||||
pub(crate) work: officeart::Decode,
|
||||
/// Anchors whose picture was dropped (only the first few are logged).
|
||||
dropped: usize,
|
||||
/// Pictures kept as placeholders (formats we can't draw).
|
||||
@@ -68,11 +44,9 @@ impl Default for Pictures {
|
||||
Pictures {
|
||||
by_fc: HashMap::new(),
|
||||
by_bse: HashMap::new(),
|
||||
by_delay: HashMap::new(),
|
||||
media: BTreeMap::new(),
|
||||
total: 0,
|
||||
limit: MAX_MEDIA_BYTES,
|
||||
work: officeart::Decode::new(MAX_DECODE_WORK),
|
||||
dropped: 0,
|
||||
placeholders: 0,
|
||||
}
|
||||
@@ -90,24 +64,19 @@ pub(crate) struct Inline {
|
||||
}
|
||||
|
||||
impl Pictures {
|
||||
/// An empty store whose decode budget scales with the `input` stream bytes (WordDocument
|
||||
/// plus Data) the pictures can come from.
|
||||
pub(crate) fn for_input(input: usize) -> Self {
|
||||
Pictures { work: officeart::Decode::new(decode_work(input, MAX_DECODE_WORK)), ..Pictures::default() }
|
||||
}
|
||||
|
||||
/// The inline picture at `fc_pic` in the Data stream, copying its bytes the first time.
|
||||
/// `None` when the header is malformed, the picture can't be read, or a picture
|
||||
/// count/byte cap is reached.
|
||||
pub(crate) fn get(&mut self, data: &[u8], fc_pic: u32) -> Option<&Inline> {
|
||||
if !self.by_fc.contains_key(&fc_pic) {
|
||||
if self.by_fc.len().saturating_add(self.by_bse.len()) >= MAX_PICTURES {
|
||||
return None;
|
||||
}
|
||||
let pic = self.read(data, fc_pic);
|
||||
self.by_fc.insert(fc_pic, pic);
|
||||
pub(crate) fn get(&mut self, data: &[u8], fc_pic: u32) -> Option<Inline> {
|
||||
if let Some(hit) = self.by_fc.get(&fc_pic) {
|
||||
return hit.clone();
|
||||
}
|
||||
self.by_fc.get(&fc_pic)?.as_ref()
|
||||
if self.by_fc.len().saturating_add(self.by_bse.len()) >= MAX_PICTURES {
|
||||
return None;
|
||||
}
|
||||
let pic = self.read(data, fc_pic);
|
||||
self.by_fc.insert(fc_pic, pic.clone());
|
||||
pic
|
||||
}
|
||||
|
||||
/// Read the inline picture at `fc_pic`: nothing is decoded or copied beyond what is left of
|
||||
@@ -117,13 +86,16 @@ impl Pictures {
|
||||
if room == 0 {
|
||||
return None;
|
||||
}
|
||||
match locate(data, fc_pic, self)? {
|
||||
match locate(data, fc_pic, room)? {
|
||||
Found::Range(pic) => {
|
||||
let bytes = data.get(pic.range.clone()).filter(|b| b.len() <= room)?.to_vec();
|
||||
let key = self.insert(bytes, pic.ext, false)?;
|
||||
Some(Inline { key, w: pic.w, h: pic.h, crop: [0.0; 4], alt: String::new() })
|
||||
}
|
||||
Found::Shape { w, h, key, props } => Some(Inline { key: key?, w, h, crop: props.crop, alt: props.alt }),
|
||||
Found::Shape { w, h, blip, props } => {
|
||||
let key = self.insert(blip.bytes, blip.ext, blip.placeholder)?;
|
||||
Some(Inline { key, w, h, crop: props.crop, alt: props.alt })
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -137,23 +109,9 @@ impl Pictures {
|
||||
return None;
|
||||
}
|
||||
let fbse = bstore.get((pib as usize).checked_sub(1)?)?;
|
||||
let key = self.decode_bse(fbse, delay);
|
||||
self.by_bse.insert(pib, key.clone());
|
||||
key
|
||||
}
|
||||
|
||||
/// Decode a store entry's BLIP, once per delay-stream record however many entries name it.
|
||||
fn decode_bse(&mut self, fbse: &[u8], delay: &[u8]) -> Option<String> {
|
||||
let (rec, at) = officeart::fbse_record(fbse, delay)?;
|
||||
let at = at.map(|(fo, used)| (delay.as_ptr() as usize, fo, used));
|
||||
if let Some(hit) = at.and_then(|at| self.by_delay.get(&at)) {
|
||||
return hit.clone();
|
||||
}
|
||||
let room = self.limit.saturating_sub(self.total);
|
||||
let key = officeart::blip(&rec, room, &mut self.work).and_then(|b| self.insert(b.bytes, b.ext, b.placeholder));
|
||||
if let Some(at) = at {
|
||||
self.by_delay.insert(at, key.clone());
|
||||
}
|
||||
let key = officeart::fbse_blip(fbse, delay, room).and_then(|b| self.insert(b.bytes, b.ext, b.placeholder));
|
||||
self.by_bse.insert(pib, key.clone());
|
||||
key
|
||||
}
|
||||
|
||||
@@ -193,7 +151,7 @@ struct Located {
|
||||
/// `OfficeArtInlineSpContainer` (Word 97 and later) whose BLIP is decoded.
|
||||
enum Found {
|
||||
Range(Located),
|
||||
Shape { w: f32, h: f32, key: Option<String>, props: officeart::Props },
|
||||
Shape { w: f32, h: f32, blip: officeart::Blip, props: officeart::Props },
|
||||
}
|
||||
|
||||
/// `PICF.mfpf.mm` values of pictures stored as a shape (`MM_SHAPE`) or a shape with a file
|
||||
@@ -202,8 +160,8 @@ const MM_SHAPE: u16 = 0x0064;
|
||||
const MM_SHAPEFILE: u16 = 0x0066;
|
||||
|
||||
/// Locate the picture at `fc_pic` in the Data stream ([MS-DOC] §2.9.192 PICFAndOfficeArtData);
|
||||
/// the BLIP is decoded once per record and charged to the decode budget.
|
||||
fn locate(data: &[u8], fc_pic: u32, pics: &mut Pictures) -> Option<Found> {
|
||||
/// a shape's picture is decoded only up to `room` bytes.
|
||||
fn locate(data: &[u8], fc_pic: u32, room: usize) -> Option<Found> {
|
||||
let base = fc_pic as usize;
|
||||
let pic = data.get(base..)?;
|
||||
let lcb = u32::from_le_bytes(pic.get(0..4)?.try_into().ok()?) as usize;
|
||||
@@ -222,12 +180,12 @@ fn locate(data: &[u8], fc_pic: u32, pics: &mut Pictures) -> Option<Found> {
|
||||
if mm == MM_SHAPEFILE {
|
||||
at = at.checked_add(1 + *pic.get(at)? as usize)?;
|
||||
}
|
||||
let (key, props) = inline_shape(pic.get(at..end)?, data, pics)?;
|
||||
let (blip, props) = inline_shape(pic.get(at..end)?, data, room)?;
|
||||
// The goal size is before cropping; the crop fractions take their share off it.
|
||||
let [cl, ct, cr, cb] = props.crop;
|
||||
let w = (dxa_goal / 20.0 * (1.0 - cl - cr).max(0.01) * scale(mx)).clamp(1.0, 1584.0);
|
||||
let h = (dya_goal / 20.0 * (1.0 - ct - cb).max(0.01) * scale(my)).clamp(1.0, 1584.0);
|
||||
return Some(Found::Shape { w, h, key, props });
|
||||
return Some(Found::Shape { w, h, blip, props });
|
||||
}
|
||||
let w = (dxa_goal / 20.0 * scale(mx)).clamp(1.0, 1584.0);
|
||||
let h = (dya_goal / 20.0 * scale(my)).clamp(1.0, 1584.0);
|
||||
@@ -239,16 +197,16 @@ fn locate(data: &[u8], fc_pic: u32, pics: &mut Pictures) -> Option<Found> {
|
||||
}
|
||||
|
||||
/// An `OfficeArtInlineSpContainer`: the picture shape, then the `OfficeArtFBSE` records its
|
||||
/// `pib` indexes (BLIPs embedded, or at `foDelay` in the Data stream). The BLIP is decoded
|
||||
/// once per record, however many pictures (and FBSE sizes) name it.
|
||||
fn inline_shape(buf: &[u8], data: &[u8], pics: &mut Pictures) -> Option<(Option<String>, officeart::Props)> {
|
||||
/// `pib` indexes (BLIPs embedded, or at `foDelay` in the Data stream). `room` caps the bytes
|
||||
/// the BLIP decodes to.
|
||||
fn inline_shape(buf: &[u8], data: &[u8], room: usize) -> Option<(officeart::Blip, officeart::Props)> {
|
||||
let mut budget = 10_000usize;
|
||||
let recs = officeart::children(buf, &mut budget);
|
||||
let sp = recs.iter().find(|r| r.typ == officeart::rt::SP_CONTAINER).map(|r| officeart::shape(r.body, &mut budget))?;
|
||||
let fbses: Vec<&[u8]> = recs.iter().filter(|r| r.typ == officeart::rt::FBSE).map(|r| r.body).collect();
|
||||
let fbse = sp.props.pib.and_then(|p| fbses.get((p as usize).checked_sub(1)?)).or(fbses.first())?;
|
||||
let key = pics.decode_bse(fbse, data);
|
||||
Some((key, sp.props))
|
||||
let blip = officeart::fbse_blip(fbse, data, room)?;
|
||||
Some((blip, sp.props))
|
||||
}
|
||||
|
||||
/// Offset of the first embedded raster image inside the payload, skipping the small
|
||||
|
||||
@@ -112,65 +112,23 @@ pub(crate) struct Blip {
|
||||
/// The picture of an `OfficeArtFBSE` body: its embedded BLIP, or the BLIP stored at `foDelay`
|
||||
/// in `delay` (the WordDocument stream for the document's BLIP store, the Data stream for
|
||||
/// inline pictures). `limit` caps the bytes produced.
|
||||
#[cfg(test)]
|
||||
pub(crate) fn fbse_blip(fbse: &[u8], delay: &[u8], limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
let (rec, _) = fbse_record(fbse, delay)?;
|
||||
blip(&rec, limit, work)
|
||||
}
|
||||
|
||||
/// The document's decode budget: bytes of inflating and pixel decoding left, and how many
|
||||
/// attempts it has refused (only the first few are logged).
|
||||
pub(crate) struct Decode {
|
||||
pub(crate) left: usize,
|
||||
refused: usize,
|
||||
}
|
||||
|
||||
impl Decode {
|
||||
pub(crate) fn new(left: usize) -> Self {
|
||||
Decode { left, refused: 0 }
|
||||
}
|
||||
|
||||
/// Take `cost` off the budget; `false` (and a log line) when it doesn't fit.
|
||||
fn charge(&mut self, cost: usize, what: &str) -> bool {
|
||||
match self.left.checked_sub(cost) {
|
||||
Some(left) if cost > 0 => {
|
||||
self.left = left;
|
||||
true
|
||||
}
|
||||
_ => {
|
||||
if self.refused < 10 {
|
||||
log::warn!(
|
||||
"docbin: a {what} needing {cost} bytes of decode work is over the decode budget ({} left) and was not decoded",
|
||||
self.left
|
||||
);
|
||||
}
|
||||
self.refused = self.refused.saturating_add(1);
|
||||
false
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// The BLIP record of an `OfficeArtFBSE`: embedded after it, or in the delay stream, where
|
||||
/// its (`foDelay`, record length) location is returned too (entries can share one record,
|
||||
/// whatever sizes their FBSEs claim: the record's own header is what counts).
|
||||
pub(crate) fn fbse_record<'a>(fbse: &'a [u8], delay: &'a [u8]) -> Option<(Rec<'a>, Option<(usize, usize)>)> {
|
||||
pub(crate) fn fbse_blip(fbse: &[u8], delay: &[u8], limit: usize) -> Option<Blip> {
|
||||
// btWin32, btMacOS, rgbUid[16], tag, size, cRef, foDelay, unused, cbName, unused×2.
|
||||
let size = u32_at(fbse, 20)? as usize;
|
||||
let fo_delay = u32_at(fbse, 28)? as usize;
|
||||
let cb_name = *fbse.get(33)? as usize;
|
||||
let at = 36usize.checked_add(cb_name)?;
|
||||
if let Some((rec, _)) = fbse.get(at..).and_then(record) {
|
||||
return Some((rec, None));
|
||||
return blip(&rec, limit);
|
||||
}
|
||||
// Not embedded: the record sits in the delay stream.
|
||||
let (rec, used) = record(delay.get(fo_delay..)?)?;
|
||||
Some((rec, Some((fo_delay, used))))
|
||||
let end = fo_delay.checked_add(size.max(8))?.min(delay.len());
|
||||
let (rec, _) = record(delay.get(fo_delay..end)?)?;
|
||||
blip(&rec, limit)
|
||||
}
|
||||
|
||||
/// Decode one `OfficeArtBlip*` record ([MS-ODRAW] §2.2.23-§2.2.31). `work` is the
|
||||
/// document's decode budget: inflating a metafile or decoding a DIB is charged to it before
|
||||
/// it runs, whether it succeeds or not, and nothing is decoded once it is spent.
|
||||
pub(crate) fn blip(rec: &Rec, limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
/// Decode one `OfficeArtBlip*` record ([MS-ODRAW] §2.2.23-§2.2.31).
|
||||
pub(crate) fn blip(rec: &Rec, limit: usize) -> Option<Blip> {
|
||||
// Each type has two instance values; the odd one carries a second 16-byte UID.
|
||||
let uids = if rec.inst & 1 == 1 { 32 } else { 16 };
|
||||
match rec.typ {
|
||||
@@ -183,7 +141,7 @@ pub(crate) fn blip(rec: &Rec, limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
rt::BLIP_PNG => Some(Blip { bytes: data.to_vec(), ext: "png", placeholder: false }),
|
||||
rt::BLIP_JPEG | rt::BLIP_JPEG_CMYK => Some(Blip { bytes: data.to_vec(), ext: "jpeg", placeholder: false }),
|
||||
rt::BLIP_TIFF => Some(Blip { bytes: data.to_vec(), ext: "tiff", placeholder: false }),
|
||||
_ => dib_to_png(data, limit, work),
|
||||
_ => dib_to_png(data, limit),
|
||||
}
|
||||
}
|
||||
rt::BLIP_EMF | rt::BLIP_WMF | rt::BLIP_PICT => {
|
||||
@@ -197,11 +155,6 @@ pub(crate) fn blip(rec: &Rec, limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
let cap = limit.min(MAX_METAFILE);
|
||||
let bytes = if compression == 0 {
|
||||
let want = if cb_size > 0 { cb_size.min(cap) } else { cap };
|
||||
// The compressed stream is read whole whatever it inflates to: a record that
|
||||
// claims to inflate to nothing still costs its input (#3871).
|
||||
if !work.charge(want.max(raw.len()), "metafile") {
|
||||
return None;
|
||||
}
|
||||
miniz_oxide::inflate::decompress_to_vec_zlib_with_limit(raw, want).ok()?
|
||||
} else {
|
||||
if raw.len() > cap {
|
||||
@@ -224,9 +177,8 @@ pub(crate) fn blip(rec: &Rec, limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
|
||||
/// A device-independent bitmap (BITMAPINFOHEADER + palette + pixels) as a PNG: the DIB gets
|
||||
/// the 14-byte BMP file header it lacks, is decoded with size limits and re-encoded. When the
|
||||
/// bitmap can't be decoded, or its declared pixels (4 bytes each) don't fit in `work`, the
|
||||
/// BMP bytes are kept (and render as a grey placeholder).
|
||||
fn dib_to_png(dib: &[u8], limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
/// bitmap can't be decoded the BMP bytes are kept (and render as a grey placeholder).
|
||||
fn dib_to_png(dib: &[u8], limit: usize) -> Option<Blip> {
|
||||
let header = u32_at(dib, 0)? as usize;
|
||||
if !(12..=124).contains(&header) {
|
||||
return None;
|
||||
@@ -245,30 +197,12 @@ fn dib_to_png(dib: &[u8], limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
let masks = if header == 40 && (compression == 3 || compression == 6) { if compression == 3 { 12 } else { 16 } } else { 0 };
|
||||
let off_bits = 14usize.checked_add(header)?.checked_add(colors * entry)?.checked_add(masks)?;
|
||||
let file_size = 14usize.checked_add(dib.len()).filter(|&size| size <= limit)?;
|
||||
// The declared size is what decoding allocates, however small the (RLE) data is.
|
||||
let (w, h) = if header == 12 {
|
||||
(u64::from(u16_at(dib, 4)?), u64::from(u16_at(dib, 6)?))
|
||||
} else {
|
||||
(u64::from(u32_at(dib, 4)?), u64::from((u32_at(dib, 8)? as i32).unsigned_abs()))
|
||||
};
|
||||
let cost = w.saturating_mul(h).saturating_mul(4).try_into().unwrap_or(usize::MAX);
|
||||
// Copying the DIB into a BMP is work too, however few pixels it declares: it is charged
|
||||
// before the copy is made (#3871). A bitmap whose pixels don't fit still keeps its BMP
|
||||
// bytes as a placeholder, if the copy fits.
|
||||
let decode = work.charge(cost.max(dib.len()), "bitmap");
|
||||
if !decode && !work.charge(dib.len(), "bitmap") {
|
||||
return None;
|
||||
}
|
||||
let mut bmp = Vec::with_capacity(file_size);
|
||||
bmp.extend_from_slice(b"BM");
|
||||
bmp.extend_from_slice(&u32::try_from(file_size).ok()?.to_le_bytes());
|
||||
bmp.extend_from_slice(&[0, 0, 0, 0]);
|
||||
bmp.extend_from_slice(&u32::try_from(off_bits).ok()?.to_le_bytes());
|
||||
bmp.extend_from_slice(dib);
|
||||
let placeholder = |bmp| Some(Blip { bytes: bmp, ext: "bmp", placeholder: true });
|
||||
if !decode {
|
||||
return placeholder(bmp);
|
||||
}
|
||||
let mut reader = image::ImageReader::new(std::io::Cursor::new(&bmp));
|
||||
reader.set_format(image::ImageFormat::Bmp);
|
||||
let mut limits = image::Limits::default();
|
||||
@@ -283,7 +217,7 @@ fn dib_to_png(dib: &[u8], limit: usize, work: &mut Decode) -> Option<Blip> {
|
||||
{
|
||||
Some(Blip { bytes: png.bytes, ext: "png", placeholder: false })
|
||||
}
|
||||
_ => placeholder(bmp),
|
||||
_ => Some(Blip { bytes: bmp, ext: "bmp", placeholder: true }),
|
||||
}
|
||||
}
|
||||
|
||||
@@ -377,15 +311,21 @@ impl Props {
|
||||
}
|
||||
0x0104 => self.pib = Some(op),
|
||||
0x0181 => self.fill_color = color(op),
|
||||
// fFilled (bit 4) is valid when fUsefFilled (bit 20) is set.
|
||||
0x01BF if op & (1 << 20) != 0 => self.filled = Some(op & (1 << 4) != 0),
|
||||
0x01BF => {}
|
||||
0x01BF => {
|
||||
// fFilled (bit 4) is valid when fUsefFilled (bit 20) is set.
|
||||
if op & (1 << 20) != 0 {
|
||||
self.filled = Some(op & (1 << 4) != 0);
|
||||
}
|
||||
}
|
||||
0x01C0 => self.line_color = color(op),
|
||||
0x01CB => self.line_width = Some(op),
|
||||
0x01D0 | 0x01D1 => self.arrowhead |= op != 0,
|
||||
// fLine (bit 3) is valid when fUsefLine (bit 19) is set.
|
||||
0x01FF if op & (1 << 19) != 0 => self.line = Some(op & (1 << 3) != 0),
|
||||
0x01FF => {}
|
||||
0x01FF => {
|
||||
// fLine (bit 3) is valid when fUsefLine (bit 19) is set.
|
||||
if op & (1 << 19) != 0 {
|
||||
self.line = Some(op & (1 << 3) != 0);
|
||||
}
|
||||
}
|
||||
0x0384..=0x0387 => self.wrap_dist[(pid - 0x0384) as usize] = Some(op),
|
||||
0x038F => self.posh = Some(op),
|
||||
0x0390 => self.posrelh = Some(op),
|
||||
|
||||
@@ -113,16 +113,10 @@ impl PieceTable {
|
||||
}
|
||||
PrmRef::Sprm0 { isprm, val: (p.prm >> 8) as u8 }
|
||||
} else {
|
||||
PrmRef::Grpprl(self.prc_grpprls.get((p.prm >> 1) as usize).map(Vec::as_slice))
|
||||
PrmRef::Grpprl(self.prc_grpprls.get((p.prm >> 1) as usize).map(|v| v.as_slice()))
|
||||
}
|
||||
}
|
||||
|
||||
/// The raw Prm of the piece holding `cp` (0, no properties, when there is none): two CPs
|
||||
/// with the same value get the same [`PrmRef`].
|
||||
pub(crate) fn prm_of(&self, cp: u32) -> u16 {
|
||||
self.piece_of(cp).map_or(0, |p| p.prm)
|
||||
}
|
||||
|
||||
fn piece_of(&self, cp: u32) -> Option<&Pcd> {
|
||||
let i = self.cps.partition_point(|&c| c <= cp).checked_sub(1)?;
|
||||
if i >= self.pieces.len() {
|
||||
|
||||
+4
-179
@@ -1732,9 +1732,9 @@ fn picture_bytes_are_capped_per_document() {
|
||||
let data: Vec<u8> = [a.clone(), b].concat();
|
||||
let mut pics = crate::media::Pictures::default();
|
||||
pics.limit = 1000;
|
||||
let first = pics.get(&data, 0).cloned().expect("first fits");
|
||||
let first = pics.get(&data, 0).expect("first fits");
|
||||
// Same location again: no new bytes.
|
||||
assert_eq!(pics.get(&data, 0).map(|p| p.key.clone()), Some(first.key.clone()));
|
||||
assert_eq!(pics.get(&data, 0).map(|p| p.key), Some(first.key.clone()));
|
||||
// A second distinct picture would exceed the cap.
|
||||
assert!(pics.get(&data, a.len() as u32).is_none());
|
||||
assert_eq!(pics.media.len(), 1);
|
||||
@@ -1754,8 +1754,8 @@ fn rejected_inline_pictures_are_not_read_again() {
|
||||
assert!(pics.get(&small, 0).is_none());
|
||||
// The rejected picture took nothing off the cap, and anchors sharing a picture share it.
|
||||
let at = big.len() as u32;
|
||||
let fits = pics.get(&data, at).cloned().expect("fits");
|
||||
assert_eq!(pics.get(&data, at).map(|p| p.key.clone()), Some(fits.key));
|
||||
let fits = pics.get(&data, at).expect("fits");
|
||||
assert_eq!(pics.get(&data, at).map(|p| p.key), Some(fits.key));
|
||||
assert_eq!(pics.media.len(), 1);
|
||||
// Filling the cap exactly works; after that, a place never tried before isn't read.
|
||||
let rest = picf_with_png(700);
|
||||
@@ -2392,178 +2392,3 @@ fn paragraph_borders_read_from_both_forms() {
|
||||
assert_eq!((top.style, top.width, top.space), (wordcraft_doc::props::BorderStyle::Single, 1.0, 2.0));
|
||||
assert_eq!(top.color, Some(wordcraft_doc::Rgb(0xFF, 0, 0)));
|
||||
}
|
||||
|
||||
/// #3642: a small compound file whose DIFAT names one FAT sector thousands of times made `cfb`
|
||||
/// build a FAT of hundreds of megabytes before checking it. The precheck refuses it first.
|
||||
#[test]
|
||||
fn a_difat_repeating_one_fat_sector_is_refused_before_cfb_builds_the_fat() {
|
||||
let f = wordcraft_doc::cfb_guard::repeating_difat_sample();
|
||||
// The precheck's own error is the proof that it ran before cfb (no clock to skew).
|
||||
assert!(matches!(read(&f), Err(super::DocbinError::Container(e)) if e.contains("FAT sectors in a file of")));
|
||||
assert!(!super::is_encrypted(&f));
|
||||
// A file cfb wrote itself passes.
|
||||
assert!(wordcraft_doc::cfb_guard::check(&cfb_file(&[("WordDocument", &[0u8; 4096])])).is_ok());
|
||||
}
|
||||
|
||||
/// A compound file `cfb` writes in `version`: `streams` at its root, then `extra` streams under
|
||||
/// an `ObjectPool` storage (where Word keeps embedded objects).
|
||||
fn cfb_file_in(version: cfb::Version, streams: &[(&str, &[u8])], extra: &[(&str, &[u8])]) -> Vec<u8> {
|
||||
let mut comp = cfb::CompoundFile::create_with_version(version, std::io::Cursor::new(Vec::new())).expect("cfb create");
|
||||
for (name, data) in streams {
|
||||
comp.create_stream(format!("/{name}")).expect("cfb stream").write_all(data).expect("cfb write");
|
||||
}
|
||||
comp.create_storage("/ObjectPool").expect("cfb storage");
|
||||
for (name, data) in extra {
|
||||
comp.create_stream(format!("/ObjectPool/{name}")).expect("cfb stream").write_all(data).expect("cfb write");
|
||||
}
|
||||
comp.into_inner().into_inner()
|
||||
}
|
||||
|
||||
/// #3642 review: valid compound files pass the guard and open, so it refuses only the hostile
|
||||
/// ones. The repository keeps no files Word made, so these are written by the `cfb` crate in
|
||||
/// the shape of Word 97–2003 files: the WordDocument and table streams of a .doc these tests
|
||||
/// build, with the summary, Data and embedded-object streams Word adds, in 512-byte sectors
|
||||
/// (version 3, Word's own) and 4096-byte ones (version 4); and one with a Data stream of 8 MB,
|
||||
/// past what the header's 109 FAT sector slots map in 512-byte sectors (109 × 128 sectors,
|
||||
/// about 7 MB), so its FAT sectors are listed in a DIFAT chain too, which the guard walks.
|
||||
#[test]
|
||||
fn valid_doc_shaped_compound_files_pass_the_guard_and_open() {
|
||||
let doc = struct_doc(&structured_spec("Hello world\r", 12));
|
||||
let mut comp = cfb::CompoundFile::open(std::io::Cursor::new(doc.as_slice())).expect("cfb open");
|
||||
// The table stream is 1Table or 0Table, as the FIB's fWhichTblStm says: kept under its name.
|
||||
let table_name = if comp.is_stream("/1Table") { "1Table" } else { "0Table" };
|
||||
let mut stream = |name: &str| {
|
||||
let mut v = Vec::new();
|
||||
std::io::Read::read_to_end(&mut comp.open_stream(name).expect("stream"), &mut v).expect("read");
|
||||
v
|
||||
};
|
||||
let (word, table) = (stream("/WordDocument"), stream(&format!("/{table_name}")));
|
||||
let summary = vec![0xFEu8; 4096];
|
||||
let data = |n: usize| (0..n).map(|i| (i % 251) as u8).collect::<Vec<u8>>();
|
||||
let small = data(200_000);
|
||||
let big = data(8 << 20);
|
||||
let ole = data(70_000);
|
||||
let files = [
|
||||
("as built", doc.clone()),
|
||||
(
|
||||
"version 3",
|
||||
cfb_file_in(
|
||||
cfb::Version::V3,
|
||||
&[("WordDocument", &word), (table_name, &table), ("Data", &small), ("\u{5}SummaryInformation", &summary)],
|
||||
&[("_1234567890", &ole)],
|
||||
),
|
||||
),
|
||||
("version 4", cfb_file_in(cfb::Version::V4, &[("WordDocument", &word), (table_name, &table), ("Data", &small)], &[("_1", &ole)])),
|
||||
("with a DIFAT chain", cfb_file_in(cfb::Version::V3, &[("WordDocument", &word), (table_name, &table), ("Data", &big)], &[])),
|
||||
];
|
||||
let difat_sectors = |f: &[u8]| f.get(0x48..0x4C).map_or(0, |b| u32::from_le_bytes([b[0], b[1], b[2], b[3]]));
|
||||
assert!(difat_sectors(&files[3].1) > 0, "the 8 MB file lists FAT sectors past the header's");
|
||||
for (what, f) in &files {
|
||||
assert_eq!(wordcraft_doc::cfb_guard::check(f), Ok(()), "{what}");
|
||||
let doc = read(f).unwrap_or_else(|e| panic!("{what}: {e}"));
|
||||
assert_eq!(body_texts(&doc).first().map(String::as_str), Some("Hello world"), "{what}");
|
||||
}
|
||||
}
|
||||
|
||||
/// #3708: a long HYPERLINK target over a long result was copied (and compared) once per result
|
||||
/// character. It is now copied once per run.
|
||||
#[test]
|
||||
fn a_long_hyperlink_over_a_long_result_is_one_linked_run() {
|
||||
let n = 250_000;
|
||||
let target = "x".repeat(n);
|
||||
let text: &'static str = Box::leak(format!("\u{13}HYPERLINK \"http://{target}\"\u{14}{}\u{15}\r", "y".repeat(n)).into_boxed_str());
|
||||
let spec = structured_spec(text, text.chars().count() as u32);
|
||||
let doc = read(&struct_doc(&spec)).expect("opens");
|
||||
let p0 = doc.body.first().and_then(|b| b.as_para()).expect("para");
|
||||
let linked: Vec<_> = p0.runs.iter().filter(|r| r.props.link.is_some()).collect();
|
||||
assert_eq!(linked.len(), 1);
|
||||
assert_eq!(linked.first().map(|r| r.len), Some(n));
|
||||
// One copy of the target for the one run, not one per character (~2.5 s here): counted by
|
||||
// what the copies charge to the work budget, which a busy machine can't skew.
|
||||
let result = vec![('y', wordcraft_doc::props::CharProps::default()); n];
|
||||
let f = super::FieldBuild { depth: 1, instr: format!("HYPERLINK \"http://{target}\""), sep: Some(result) };
|
||||
let work = super::Work::new(u64::MAX);
|
||||
super::finish_field(&mut super::ParaBuild::default(), f, &work);
|
||||
assert_eq!(u64::MAX - work.left.get(), ("http://".len() + n) as u64, "one copy of the target");
|
||||
}
|
||||
|
||||
/// #3708: a result whose formatting changes every character pays for a target copy per run
|
||||
/// out of the work budget; past it the text stays, unlinked.
|
||||
#[test]
|
||||
fn hyperlink_target_copies_are_charged_to_the_work_budget() {
|
||||
use wordcraft_doc::props::CharProps;
|
||||
let bold = CharProps { bold: Some(true), ..Default::default() };
|
||||
let result: Vec<(char, CharProps)> = (0..1000).map(|i| ('a', if i % 2 == 0 { bold.clone() } else { CharProps::default() })).collect();
|
||||
let f = super::FieldBuild { depth: 1, instr: format!("HYPERLINK \"{}\"", "u".repeat(100)), sep: Some(result) };
|
||||
let mut pb = super::ParaBuild::default();
|
||||
let budget = super::Work::new(10_000);
|
||||
super::finish_field(&mut pb, f, &budget);
|
||||
assert_eq!(pb.text.len(), 1000, "all text kept");
|
||||
let linked: usize = pb.runs.iter().filter_map(|r| r.props.link.as_ref()).map(|l| l.len()).sum();
|
||||
assert!(linked <= 10_000, "linked bytes {linked}");
|
||||
assert_eq!(pb.runs.iter().filter(|r| r.props.link.is_some()).count(), 100);
|
||||
assert_eq!(budget.left.get(), 0);
|
||||
assert_eq!(budget.dropped(), [super::Cost::LinkTarget], "the unlinked text is noted, to be logged");
|
||||
}
|
||||
|
||||
/// #3871 review: what the work budget leaves out is noted once per kind, in a fixed order, and
|
||||
/// logged once per kind when the document has been read; nothing is noted while it lasts.
|
||||
#[test]
|
||||
fn the_work_budget_notes_each_kind_it_leaves_out_once() {
|
||||
use super::Cost::*;
|
||||
let work = super::Work::new(10);
|
||||
assert!(work.charge(10, AltText));
|
||||
assert!(work.dropped().is_empty(), "spent to the last byte, nothing left out");
|
||||
assert!(work.charge(0, LinkTarget), "a free copy is never refused");
|
||||
for _ in 0..3 {
|
||||
assert!(!work.charge(1, FloatingObject));
|
||||
}
|
||||
assert!(!work.charge(1, AltText));
|
||||
assert_eq!(work.dropped(), [AltText, FloatingObject]);
|
||||
work.log();
|
||||
assert_eq!(super::LAST_DROPPED.with(|d| d.borrow().clone()), [AltText, FloatingObject]);
|
||||
// A document within the budget leaves nothing out.
|
||||
read(&struct_doc(&structured_spec("Hello\r", 6))).expect("opens");
|
||||
assert!(super::LAST_DROPPED.with(|d| d.borrow().is_empty()));
|
||||
}
|
||||
|
||||
/// #3692: a piece's Prm grpprl (a Prc of up to 64 KB) was re-interpreted for every character.
|
||||
/// It is now resolved once per formatting change.
|
||||
#[test]
|
||||
fn a_large_prc_is_interpreted_once_per_formatting_change() {
|
||||
let prc = grpprl(&vec![(0x0835u16, &[0x01u8][..]); 21_000]);
|
||||
let text: &'static str = Box::leak(format!("{}\r", "a".repeat(100_000)).into_boxed_str());
|
||||
let spec = RichSpec {
|
||||
text,
|
||||
styles: vec![(1, 0x0FFF, "Normal", Vec::new(), Vec::new())],
|
||||
fonts: vec!["Times New Roman"],
|
||||
chpx_runs: vec![(0, Vec::new())],
|
||||
papx_entries: vec![(0, 0, Vec::new())],
|
||||
};
|
||||
let bytes = rich_doc_with(&spec, &RichExtra { prcs: vec![prc], prm: 0x0001, ..Default::default() });
|
||||
let t = std::time::Instant::now();
|
||||
let doc = read(&bytes).expect("opens");
|
||||
// About 60 ms on a loaded machine; interpreting the Prc per character took 88 s.
|
||||
assert!(t.elapsed().as_secs() < 10, "opened in {:?}", t.elapsed());
|
||||
let p = para0(&doc);
|
||||
assert_eq!(p.runs.len(), 1);
|
||||
assert_eq!(p.runs.first().and_then(|r| r.props.bold), Some(true));
|
||||
}
|
||||
|
||||
/// #3692: Prm grpprls are charged to a document-wide work budget; once it is spent the Prm
|
||||
/// is skipped (the text keeps its CHPX formatting).
|
||||
#[test]
|
||||
fn prm_grpprls_past_the_work_budget_are_skipped() {
|
||||
use wordcraft_doc::props::CharProps;
|
||||
let g = grpprl(&[(0x0835, &[0x01])]);
|
||||
let work = super::Work::new(g.len() as u64 + 1);
|
||||
let base = CharProps::default();
|
||||
let mut first = CharProps::default();
|
||||
super::apply_prm(&mut first, crate::piece::PrmRef::Grpprl(Some(&g)), &[], &base, &work);
|
||||
assert_eq!(first.bold, Some(true));
|
||||
let mut second = CharProps::default();
|
||||
super::apply_prm(&mut second, crate::piece::PrmRef::Grpprl(Some(&g)), &[], &base, &work);
|
||||
assert_eq!(second.bold, None);
|
||||
assert_eq!(work.left.get(), 1);
|
||||
assert_eq!(work.dropped(), [super::Cost::Formatting]);
|
||||
}
|
||||
|
||||
@@ -70,7 +70,7 @@ fn fbse_png(png: &[u8]) -> Vec<u8> {
|
||||
|
||||
/// UTF-16LE, NUL-terminated (a complex string property).
|
||||
fn wz(s: &str) -> Vec<u8> {
|
||||
s.encode_utf16().chain([0]).flat_map(u16::to_le_bytes).collect()
|
||||
s.encode_utf16().chain([0]).flat_map(|u| u.to_le_bytes()).collect()
|
||||
}
|
||||
|
||||
/// A `Spa` (26 bytes): spid, bounds in twips, flags, cTxbx.
|
||||
@@ -247,7 +247,7 @@ fn dib_becomes_png_and_compressed_metafiles_inflate() {
|
||||
dib.extend_from_slice(&[0, 0, 255, 0, 255, 0, 0, 0]);
|
||||
let r = rec(0, 0x7A8, 0xF01F, &[[0u8; 16].as_slice(), &[0xFF], &dib].concat());
|
||||
let (dib_rec, _) = officeart::record(&r).unwrap();
|
||||
let blip = officeart::blip(&dib_rec, 1 << 20, &mut officeart::Decode::new(usize::MAX)).expect("DIB decodes");
|
||||
let blip = officeart::blip(&dib_rec, 1 << 20).expect("DIB decodes");
|
||||
assert_eq!(blip.ext, "png");
|
||||
let img = image::load_from_memory(&blip.bytes).unwrap().to_rgb8();
|
||||
assert_eq!((img.width(), img.height(), img.get_pixel(0, 0).0), (2, 1, [255, 0, 0]));
|
||||
@@ -262,7 +262,7 @@ fn dib_becomes_png_and_compressed_metafiles_inflate() {
|
||||
hdr.extend_from_slice(&[0x00, 0xFE]); // compression = deflate, filter
|
||||
let r = rec(0, 0x216, 0xF01B, &[[0u8; 16].as_slice(), &hdr, &packed].concat());
|
||||
let (wmf_rec, _) = officeart::record(&r).unwrap();
|
||||
let blip = officeart::blip(&wmf_rec, 1 << 20, &mut officeart::Decode::new(usize::MAX)).expect("WMF inflates");
|
||||
let blip = officeart::blip(&wmf_rec, 1 << 20).expect("WMF inflates");
|
||||
assert_eq!((blip.ext, blip.bytes), ("wmf", wmf));
|
||||
}
|
||||
|
||||
@@ -281,10 +281,10 @@ fn deeply_nested_groups_and_lying_lengths_are_bounded() {
|
||||
hdr.extend_from_slice(&[0, 0]);
|
||||
let r = rec(0, 0x3D4, 0xF01A, &[[0u8; 16].as_slice(), &hdr, &[0x78, 0x9C, 1, 2, 3]].concat());
|
||||
let (emf, _) = officeart::record(&r).unwrap();
|
||||
assert!(officeart::blip(&emf, 1 << 20, &mut officeart::Decode::new(usize::MAX)).is_none());
|
||||
assert!(officeart::blip(&emf, 1 << 20).is_none());
|
||||
let mut fbse = vec![0u8; 36];
|
||||
fbse[28..32].copy_from_slice(&u32::MAX.to_le_bytes());
|
||||
assert!(officeart::fbse_blip(&fbse, &[0u8; 64], 1 << 20, &mut officeart::Decode::new(usize::MAX)).is_none());
|
||||
assert!(officeart::fbse_blip(&fbse, &[0u8; 64], 1 << 20).is_none());
|
||||
}
|
||||
|
||||
proptest! {
|
||||
@@ -322,296 +322,13 @@ fn dib_conversion_caps_header_and_encoded_bytes() {
|
||||
let record = rec(0, 0x7A8, officeart::rt::BLIP_DIB, &[[0u8; 16].as_slice(), &[0xFF], &dib].concat());
|
||||
let (record, _) = officeart::record(&record).unwrap();
|
||||
let bmp_size = 14 + dib.len();
|
||||
assert!(officeart::blip(&record, bmp_size - 1, &mut officeart::Decode::new(usize::MAX)).is_none(), "the BMP header also needs room");
|
||||
let fallback = officeart::blip(&record, bmp_size, &mut officeart::Decode::new(usize::MAX)).expect("the BMP fallback fits exactly");
|
||||
assert!(officeart::blip(&record, bmp_size - 1).is_none(), "the BMP header also needs room");
|
||||
let fallback = officeart::blip(&record, bmp_size).expect("the BMP fallback fits exactly");
|
||||
assert_eq!((fallback.ext, fallback.placeholder, fallback.bytes.len()), ("bmp", true, bmp_size));
|
||||
let img = image::load_from_memory(&fallback.bytes).unwrap().to_rgb8();
|
||||
assert_eq!((img.width(), img.height(), img.get_pixel(0, 0).0), (1, 1, [90, 60, 30]));
|
||||
let normal = officeart::blip(&record, 512, &mut officeart::Decode::new(usize::MAX)).expect("normal conversion fits");
|
||||
let normal = officeart::blip(&record, 512).expect("normal conversion fits");
|
||||
assert_eq!((normal.ext, normal.placeholder), ("png", false));
|
||||
assert!(normal.bytes.len() > bmp_size && normal.bytes.len() <= 512);
|
||||
assert_eq!(image::load_from_memory(&normal.bytes).unwrap().width(), 1);
|
||||
}
|
||||
|
||||
/// #3709: every anchor of one inline picture got its own copy of the alt text (up to 4096
|
||||
/// UTF-16 units). Copies are charged to the document's work budget; past it alt is empty.
|
||||
#[test]
|
||||
fn alt_text_copies_across_many_anchors_are_bounded() {
|
||||
let alt: String = "€".repeat(4096); // 12 KB as UTF-8
|
||||
let shape = sp(&[fsp(75, 1025, 0xA00), fopt(&[(0x0104, 1)], &[(0x0381, wz(&alt))])]);
|
||||
let data = picf_shape(&[shape, fbse_png(&tiny_png())].concat());
|
||||
let text: &'static str = Box::leak(format!("{}\r", "\u{1}".repeat(8000)).into_boxed_str());
|
||||
let ccp = text.chars().count() as u32;
|
||||
let spec = StructSpec { text, data_stream: data, chpx_runs: vec![(0, grpprl(&[(0x6A03, &0u32.to_le_bytes())]))], ..structured_spec(text, ccp) };
|
||||
let doc = read(&struct_doc(&spec)).expect("opens");
|
||||
let p0 = doc.body.first().and_then(|b| b.as_para()).expect("para");
|
||||
let alts: Vec<&str> = p0.objects.iter().filter_map(|o| if let InlineObject::Image { alt, .. } = o { Some(alt.as_str()) } else { None }).collect();
|
||||
assert_eq!(alts.len(), 8000, "every anchor keeps its picture");
|
||||
assert_eq!(alts.first().copied(), Some(alt.as_str()));
|
||||
let total: usize = alts.iter().map(|a| a.len()).sum();
|
||||
assert!(total <= 64 << 20, "alt bytes {total}");
|
||||
assert_eq!(alts.last().copied(), Some(""));
|
||||
// Leaving it out is logged, once (#3871 review).
|
||||
assert_eq!(crate::LAST_DROPPED.with(|d| d.borrow().clone()), [crate::Cost::AltText]);
|
||||
}
|
||||
|
||||
/// #3710: every PlcSpa entry naming a group built another copy of all its members. A shape is
|
||||
/// anchored once; repeated spids are skipped.
|
||||
#[test]
|
||||
fn a_group_named_by_many_anchors_is_built_once() {
|
||||
let rect = |l: i32| rec(0, 0, 0xF00F, &[l, 0, l + 10, 10].iter().flat_map(|v| v.to_le_bytes()).collect::<Vec<_>>());
|
||||
let head = sp(&[fsp(0, 3000, 0x201), rec(1, 0, 0xF009, &[0i32, 0, 10_000, 10].iter().flat_map(|v| v.to_le_bytes()).collect::<Vec<_>>())]);
|
||||
let members: Vec<Vec<u8>> = (0..1000).map(|i| sp(&[fsp(1, 4000 + i as u32, 0xA02), rect(i * 10)])).collect();
|
||||
let group = container(0xF003, &[vec![head], members].concat());
|
||||
let art = art_content(&[], &[group]);
|
||||
let n = 5000u32;
|
||||
let cps: Vec<u32> = (1..=n + 1).collect();
|
||||
let spas: Vec<u8> = (0..n).flat_map(|_| spa(3000, [0, 0, 1440, 1440], spa_flags(0, 0, 3, false))).collect();
|
||||
let text = "A\u{8}\u{8}\u{8}B\r";
|
||||
let ccp = text.chars().count() as u32;
|
||||
let spec = StructSpec { text, extra_pairs: vec![(40, plc(&cps, &spas)), (50, art)], ..structured_spec(text, ccp) };
|
||||
let t = std::time::Instant::now();
|
||||
let doc = read(&struct_doc(&spec)).expect("opens");
|
||||
// About 10 ms on a loaded machine, so 10 s is a backstop; the old code (a copy per entry,
|
||||
// 6-8 s here) is caught by the group count below, one group per anchor ([1000; 3]).
|
||||
assert!(t.elapsed().as_secs() < 10, "opened in {:?}", t.elapsed());
|
||||
let p0 = doc.body.first().and_then(|b| b.as_para()).expect("para");
|
||||
let groups: Vec<usize> =
|
||||
p0.objects.iter().filter_map(|o| if let InlineObject::Group { children, .. } = o { Some(children.len()) } else { None }).collect();
|
||||
assert_eq!(groups, [1000]);
|
||||
}
|
||||
|
||||
/// #3710: a floating picture's copy is charged its alt text too, and a group every member's.
|
||||
#[test]
|
||||
fn a_floating_copy_is_charged_its_alt_text() {
|
||||
let image = |alt: String| InlineObject::Image {
|
||||
media: String::new(),
|
||||
w: 1.0,
|
||||
h: 1.0,
|
||||
alt,
|
||||
float: Default::default(),
|
||||
crop: Default::default(),
|
||||
ole: None,
|
||||
};
|
||||
let bare = crate::object_cost(&image(String::new()));
|
||||
assert_eq!(crate::object_cost(&image("€".repeat(4096))), bare + 3 * 4096);
|
||||
let member = |obj| wordcraft_doc::para::GroupChild { x: 0.0, y: 0.0, obj };
|
||||
let group = InlineObject::Group {
|
||||
w: 1.0,
|
||||
h: 1.0,
|
||||
float: Default::default(),
|
||||
ch_w: 1.0,
|
||||
ch_h: 1.0,
|
||||
children: vec![member(image("x".repeat(100))), member(image(String::new()))],
|
||||
canvas: None,
|
||||
};
|
||||
assert_eq!(crate::object_cost(&group), 3 * bare + 100);
|
||||
}
|
||||
|
||||
/// A deflated WMF BLIP record declaring `size` uncompressed bytes.
|
||||
fn packed_wmf(size: u32, packed: &[u8]) -> Vec<u8> {
|
||||
let mut hdr = size.to_le_bytes().to_vec();
|
||||
hdr.extend_from_slice(&[0u8; 24]); // rcBounds, ptSize
|
||||
hdr.extend_from_slice(&(packed.len() as u32).to_le_bytes());
|
||||
hdr.extend_from_slice(&[0x00, 0xFE]); // compression = deflate, filter
|
||||
rec(0, 0x216, officeart::rt::BLIP_WMF, &[[0u8; 16].as_slice(), &hdr, packed].concat())
|
||||
}
|
||||
|
||||
/// An `OfficeArtFBSE` body whose BLIP (of `size` bytes) sits at `fo_delay` in the delay stream.
|
||||
fn delayed_fbse(size: usize, fo_delay: usize) -> Vec<u8> {
|
||||
let mut fbse = vec![0u8; 36];
|
||||
fbse[20..24].copy_from_slice(&(size as u32).to_le_bytes());
|
||||
fbse[28..32].copy_from_slice(&(fo_delay as u32).to_le_bytes());
|
||||
fbse
|
||||
}
|
||||
|
||||
/// #3711: store entries aliasing one delay-stream metafile each inflated (and copied) it again.
|
||||
/// Now the record is decoded once and every entry shares its media key.
|
||||
#[test]
|
||||
fn store_entries_aliasing_one_delayed_metafile_decode_it_once() {
|
||||
let wmf = vec![0u8; 1 << 20];
|
||||
let delay = packed_wmf(wmf.len() as u32, &miniz_oxide::deflate::compress_to_vec_zlib(&wmf, 6));
|
||||
let fbse = delayed_fbse(delay.len(), 0);
|
||||
let store: Vec<&[u8]> = vec![&fbse; 5_000];
|
||||
let mut pics = crate::media::Pictures::default();
|
||||
pics.work = officeart::Decode::new(8 << 20);
|
||||
let keys: Vec<_> = (1..=store.len() as u32).map(|pib| pics.bse(pib, &store, &delay)).collect();
|
||||
assert!(keys.iter().all(|k| k.is_some() && *k == keys[0]), "every entry shares one picture");
|
||||
assert_eq!(pics.work.left, (8 << 20) - wmf.len(), "inflated once");
|
||||
}
|
||||
|
||||
/// #3711: distinct metafiles that fail to inflate are charged their declared size anyway, so
|
||||
/// a document of them stops decoding once the work budget is spent.
|
||||
#[test]
|
||||
fn failing_metafile_inflations_spend_the_decode_budget() {
|
||||
let records: Vec<Vec<u8>> = (0..10u8).map(|i| packed_wmf(1 << 20, &[0x78, 0x9C, i, 1, 2, 3])).collect();
|
||||
let delay = records.concat();
|
||||
let mut at = 0;
|
||||
let fbses: Vec<Vec<u8>> = records
|
||||
.iter()
|
||||
.map(|r| {
|
||||
at += r.len();
|
||||
delayed_fbse(r.len(), at - r.len())
|
||||
})
|
||||
.collect();
|
||||
let store: Vec<&[u8]> = fbses.iter().map(Vec::as_slice).collect();
|
||||
let mut pics = crate::media::Pictures::default();
|
||||
pics.work = officeart::Decode::new(3 << 20);
|
||||
for pib in 1..=store.len() as u32 {
|
||||
assert!(pics.bse(pib, &store, &delay).is_none());
|
||||
}
|
||||
assert_eq!(pics.work.left, 0, "three attempts spent the budget; the rest weren't tried");
|
||||
}
|
||||
|
||||
/// #3712: a DIB's declared pixels are charged before decoding (an RLE bitmap is tiny on disk but
|
||||
/// decodes to w × h × 4 bytes); past the budget it stays a BMP placeholder, undecoded.
|
||||
#[test]
|
||||
fn dib_decoding_is_charged_its_declared_pixels() {
|
||||
let blip_of = |dib: &[u8]| rec(0, 0x7A8, officeart::rt::BLIP_DIB, &[[0u8; 16].as_slice(), &[0xFF], dib].concat());
|
||||
let one = blip_of(&one_pixel_dib());
|
||||
let (one, _) = officeart::record(&one).unwrap();
|
||||
let mut work = officeart::Decode::new(3);
|
||||
assert!(officeart::blip(&one, 512, &mut work).is_none(), "not even the copy fits: nothing is copied");
|
||||
assert_eq!(work.left, 3, "nothing decoded, nothing charged");
|
||||
let dib_len = one.body.len() - 17;
|
||||
let mut work = officeart::Decode::new(dib_len);
|
||||
assert_eq!(officeart::blip(&one, 512, &mut work).map(|b| (b.ext, b.placeholder)), Some(("png", false)));
|
||||
assert_eq!(work.left, 0, "the copy is charged, not just 4 bytes a pixel");
|
||||
|
||||
// 8-bit RLE declaring 16384 × 16384 (1 GiB decoded) in a few bytes.
|
||||
let mut big = [40u32, 16_384, 16_384].iter().flat_map(|v| v.to_le_bytes()).collect::<Vec<_>>();
|
||||
big.extend_from_slice(&1u16.to_le_bytes());
|
||||
big.extend_from_slice(&8u16.to_le_bytes());
|
||||
big.extend_from_slice(&1u32.to_le_bytes()); // BI_RLE8
|
||||
big.extend_from_slice(&[0u8; 12]);
|
||||
big.extend_from_slice(&1u32.to_le_bytes()); // one palette colour
|
||||
big.extend_from_slice(&[0u8; 4]); // clrImportant
|
||||
big.extend_from_slice(&[0u8; 4]); // the palette colour
|
||||
big.extend_from_slice(&[0, 0, 0, 1]); // end of line, end of bitmap
|
||||
let big = blip_of(&big);
|
||||
let (big, _) = officeart::record(&big).unwrap();
|
||||
let mut work = officeart::Decode::new(512 << 20);
|
||||
assert_eq!(officeart::blip(&big, 1 << 20, &mut work).map(|b| (b.ext, b.placeholder)), Some(("bmp", true)));
|
||||
assert_eq!(work.left, (512 << 20) - (big.body.len() - 17), "only the copy of the DIB was charged, not its declared pixels");
|
||||
}
|
||||
|
||||
/// An EMF `OfficeArtBlip` record (compressed) with `raw` as its stream, claiming `cb_size`.
|
||||
fn emf_blip(cb_size: u32, raw: &[u8]) -> Vec<u8> {
|
||||
let mut hdr = vec![0u8; 34];
|
||||
hdr[0..4].copy_from_slice(&cb_size.to_le_bytes());
|
||||
hdr[28..32].copy_from_slice(&(raw.len() as u32).to_le_bytes());
|
||||
hdr[32] = 0; // compressed
|
||||
hdr[33] = 0xFE;
|
||||
rec(0, 0x3D4, 0xF01A, &[[0u8; 16].as_slice(), &hdr, raw].concat())
|
||||
}
|
||||
|
||||
/// #3871 review: the zlib input is charged to the decode budget, not just what it inflates to.
|
||||
#[test]
|
||||
fn compressed_metafile_input_is_charged_to_the_decode_budget() {
|
||||
let raw = vec![0x5Au8; 1 << 20]; // not even zlib: costs its input all the same
|
||||
let r = emf_blip(1, &raw);
|
||||
let (emf, _) = officeart::record(&r).unwrap();
|
||||
let mut work = officeart::Decode::new(10 << 20);
|
||||
assert!(officeart::blip(&emf, 64 << 20, &mut work).is_none());
|
||||
assert_eq!(work.left, (10 << 20) - raw.len(), "1 MB of input charged, however little it claims to inflate to");
|
||||
// Ten more of them exhaust the budget; none is inflated past it.
|
||||
let mut work = officeart::Decode::new(3 << 20);
|
||||
for _ in 0..10 {
|
||||
let _ = officeart::blip(&emf, 64 << 20, &mut work);
|
||||
}
|
||||
assert_eq!(work.left, 0, "three attempts fit, the rest are refused");
|
||||
}
|
||||
|
||||
/// An inline `PICFAndOfficeArtData` (shape, FBSE) whose BLIP sits at `fo_delay` in Data; the
|
||||
/// FBSE claims `size` bytes.
|
||||
fn inline_pic(fo_delay: u32, size: u32) -> Vec<u8> {
|
||||
let mut fbse = vec![0u8; 36];
|
||||
fbse[20..24].copy_from_slice(&size.to_le_bytes());
|
||||
fbse[28..32].copy_from_slice(&fo_delay.to_le_bytes());
|
||||
let body = [container(0xF004, &[]), rec(2, 6, 0xF007, &fbse)].concat();
|
||||
let mut picf = vec![0u8; 0x44];
|
||||
picf[0..4].copy_from_slice(&((0x44 + body.len()) as u32).to_le_bytes());
|
||||
picf[4..6].copy_from_slice(&0x44u16.to_le_bytes());
|
||||
picf[6..8].copy_from_slice(&0x64u16.to_le_bytes()); // MM_SHAPE
|
||||
picf[28..30].copy_from_slice(&1440i16.to_le_bytes());
|
||||
picf[30..32].copy_from_slice(&1440i16.to_le_bytes());
|
||||
[picf, body].concat()
|
||||
}
|
||||
|
||||
/// #3871 review: inline pictures that alias one BLIP record share one decode, whatever sizes
|
||||
/// their FBSEs claim and wherever their PICFs sit.
|
||||
#[test]
|
||||
fn inline_pictures_aliasing_one_blip_record_are_decoded_once() {
|
||||
let blip = emf_blip(1, &vec![0x5Au8; 64 << 10]);
|
||||
let a = inline_pic(0, 70_000);
|
||||
let b = inline_pic(0, 80_000);
|
||||
let c = inline_pic(0, 90_000);
|
||||
let blip_at = (a.len() + b.len() + c.len()) as u32;
|
||||
let (a, b, c) = (inline_pic(blip_at, 70_000), inline_pic(blip_at, 80_000), inline_pic(blip_at, 90_000));
|
||||
let data = [a.clone(), b.clone(), c, blip].concat();
|
||||
let mut pics = crate::media::Pictures::default();
|
||||
let start = pics.work.left;
|
||||
let _ = pics.get(&data, 0);
|
||||
let _ = pics.get(&data, a.len() as u32);
|
||||
let _ = pics.get(&data, (a.len() + b.len()) as u32);
|
||||
assert_eq!(start - pics.work.left, 64 << 10, "one record, one charge");
|
||||
}
|
||||
|
||||
/// A 1-bit-per-pixel DIB of `side`×`side` pixels with `bytes` of pixel data.
|
||||
fn dib_1bpp(side: u32, bytes: usize) -> Vec<u8> {
|
||||
let mut d = vec![0u8; 40];
|
||||
d[0..4].copy_from_slice(&40u32.to_le_bytes());
|
||||
d[4..8].copy_from_slice(&side.to_le_bytes());
|
||||
d[8..12].copy_from_slice(&side.to_le_bytes());
|
||||
d[12..14].copy_from_slice(&1u16.to_le_bytes());
|
||||
d[14..16].copy_from_slice(&1u16.to_le_bytes());
|
||||
d.extend_from_slice(&[0, 0, 0, 0, 255, 255, 255, 0]); // palette
|
||||
d.extend(std::iter::repeat_n(0u8, bytes));
|
||||
d
|
||||
}
|
||||
|
||||
fn dib_record(dib: &[u8]) -> Vec<u8> {
|
||||
rec(0, 0x7A8, 0xF01F, &[[0u8; 16].as_slice(), &[0xFF], dib].concat())
|
||||
}
|
||||
|
||||
/// #3871 review: the decode budget scales with the input, so a document of many legitimate
|
||||
/// large scans decodes them all; a high-ratio set is still capped.
|
||||
#[test]
|
||||
fn the_decode_budget_scales_with_the_input_but_still_caps_bombs() {
|
||||
let floor = 1 << 20;
|
||||
// 20 legitimate 1024² 1-bit scans: 4 MB charged each, 128 KB of file each.
|
||||
let real = dib_record(&dib_1bpp(1024, 1024 * 1024 / 8));
|
||||
let (rec_real, _) = officeart::record(&real).unwrap();
|
||||
let input = 20 * real.len();
|
||||
let mut work = officeart::Decode::new(crate::media::decode_work(input, floor));
|
||||
for i in 0..20 {
|
||||
let b = officeart::blip(&rec_real, 64 << 20, &mut work).unwrap();
|
||||
assert!(!b.placeholder, "scan {i} decodes");
|
||||
}
|
||||
// Declared 20000² (1.6 GB each) in a few dozen bytes each: a placeholder, never decoded.
|
||||
let bomb = dib_record(&dib_1bpp(20_000, 16));
|
||||
let (rec_bomb, _) = officeart::record(&bomb).unwrap();
|
||||
let mut work = officeart::Decode::new(crate::media::decode_work(20 * bomb.len(), floor));
|
||||
for _ in 0..20 {
|
||||
assert!(officeart::blip(&rec_bomb, 64 << 20, &mut work).unwrap().placeholder);
|
||||
}
|
||||
assert!(work.left <= crate::media::decode_work(20 * bomb.len(), floor), "nothing was refunded or invented");
|
||||
}
|
||||
|
||||
/// #3871 review: the copy of a DIB into a BMP is charged before it is made, however few pixels
|
||||
/// the header declares; with the budget spent nothing is copied at all.
|
||||
#[test]
|
||||
fn a_dib_with_a_large_tail_is_charged_for_its_copy() {
|
||||
let big = dib_record(&dib_1bpp(1, 1 << 20)); // 1 pixel declared, 1 MB of trailing bytes
|
||||
let (big, _) = officeart::record(&big).unwrap();
|
||||
let dib_len = big.body.len() - 17;
|
||||
let mut work = officeart::Decode::new(10 << 20);
|
||||
assert!(!officeart::blip(&big, 64 << 20, &mut work).unwrap().placeholder);
|
||||
assert_eq!(10 << 20, work.left + dib_len, "charged for the bytes, not just 4 per pixel");
|
||||
// Room for the copy: decoded. Not even that: nothing is copied.
|
||||
let mut work = officeart::Decode::new(dib_len + 1);
|
||||
assert!(officeart::blip(&big, 64 << 20, &mut work).is_some());
|
||||
let mut work = officeart::Decode::new(dib_len - 1);
|
||||
assert!(officeart::blip(&big, 64 << 20, &mut work).is_none());
|
||||
assert_eq!(work.left, dib_len - 1);
|
||||
}
|
||||
|
||||
@@ -199,9 +199,6 @@ fn unsupported_variants_say_so() {
|
||||
let (h, k) = capi_header("pw", 128, hflags);
|
||||
let e = read_with_password(&rc4_file_with(h, &*k), Some("pw")).unwrap_err();
|
||||
assert!(matches!(e, DocbinError::Encryption(_)), "{e:?}");
|
||||
if hflags & 0x20 != 0 {
|
||||
assert!(e.to_string().contains("only defined for the newer encrypted-package format"), "{e}");
|
||||
}
|
||||
}
|
||||
let (mut h, k) = rc4_header("pw");
|
||||
h[0] = 9;
|
||||
|
||||
@@ -31,7 +31,6 @@ zeroize = { workspace = true }
|
||||
getrandom = { workspace = true, features = ["wasm_js"] }
|
||||
|
||||
[dev-dependencies]
|
||||
wordcraft-doc = { workspace = true, features = ["test-support"] }
|
||||
proptest = { workspace = true }
|
||||
image = { workspace = true }
|
||||
zip = { workspace = true }
|
||||
|
||||
@@ -6,7 +6,7 @@
|
||||
//! [`spec_of`] turns a chart part back into the model only when writing that model again gives
|
||||
//! the same markup, so a chart with anything the model leaves out stays as it was read.
|
||||
|
||||
use wordcraft_doc::chart::{ChartSeries, ChartSpec, ChartType, LegendPos, MAX_POINTS, MAX_SERIES};
|
||||
use wordcraft_doc::chart::{ChartSeries, ChartSpec, ChartType, LegendPos};
|
||||
use wordcraft_doc::graphic::GraphicItem;
|
||||
use wordcraft_doc::props::Rgb;
|
||||
|
||||
@@ -23,21 +23,9 @@ pub(crate) const CHART_URI: &str = "http://schemas.openxmlformats.org/drawingml/
|
||||
const CAT_AX: &str = "500000001";
|
||||
const VAL_AX: &str = "500000002";
|
||||
|
||||
/// The chart part for `spec`: a complete XML document, written as [`ChartSpec::sanitize`] leaves
|
||||
/// it (a value that isn't finite is a gap).
|
||||
/// The chart part for `spec`: a complete XML document. `spec` should be sanitized
|
||||
/// ([`ChartSpec::sanitize`]); a value that isn't finite is written as a gap either way.
|
||||
pub fn chart_xml(spec: &ChartSpec) -> String {
|
||||
chart_part(spec).0
|
||||
}
|
||||
|
||||
/// [`chart_xml`], and whether the chart limits left series or points of `spec` out of it.
|
||||
pub(crate) fn chart_part(spec: &ChartSpec) -> (String, bool) {
|
||||
let kept = bounded(spec);
|
||||
let points = spec.series.iter().map(|s| s.values.len()).max().unwrap_or(0).max(spec.categories.len());
|
||||
let cut = kept.series.len() < spec.series.len() || kept.points() < points;
|
||||
(write_chart(&kept), cut)
|
||||
}
|
||||
|
||||
fn write_chart(spec: &ChartSpec) -> String {
|
||||
let mut w = W::new();
|
||||
w.open("c:chartSpace", &[("xmlns:c", NS_C), ("xmlns:a", NS_A), ("xmlns:r", NS_R)]);
|
||||
w.empty("c:roundedCorners", &[("val", "0")]);
|
||||
@@ -91,26 +79,6 @@ fn write_chart(spec: &ChartSpec) -> String {
|
||||
w.s
|
||||
}
|
||||
|
||||
/// `spec` held to the chart limits. A model deserialised as is (a `.json` document) never went
|
||||
/// through [`ChartSpec::sanitize`], and every series repeats every category, so its size would
|
||||
/// multiply (#3648). Only what the limits keep is copied before sanitizing.
|
||||
fn bounded(spec: &ChartSpec) -> ChartSpec {
|
||||
ChartSpec {
|
||||
kind: spec.kind,
|
||||
categories: spec.categories.iter().take(MAX_POINTS).cloned().collect(),
|
||||
series: spec
|
||||
.series
|
||||
.iter()
|
||||
.take(MAX_SERIES)
|
||||
.map(|s| ChartSeries { name: s.name.clone(), values: s.values.iter().take(MAX_POINTS).copied().collect() })
|
||||
.collect(),
|
||||
title: spec.title.clone(),
|
||||
legend: spec.legend,
|
||||
data_labels: spec.data_labels,
|
||||
}
|
||||
.sanitized()
|
||||
}
|
||||
|
||||
/// What `spec` draws as, `w` × `h` points, in the colours of `theme` (series take the accents in
|
||||
/// turn). Empty when the chart has no series.
|
||||
pub fn chart_items(spec: &ChartSpec, theme: &[Rgb], w: f32, h: f32) -> Vec<GraphicItem> {
|
||||
@@ -453,27 +421,6 @@ mod tests {
|
||||
e.els().map(|c| c.name.as_str()).collect()
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_chart_past_the_limits_is_written_within_them() {
|
||||
// A model deserialised as is (a `.json` document) never went through `sanitize`, and every
|
||||
// series repeats every category: 300 series of 4,500 points wrote 2.7 million points (#3648).
|
||||
use wordcraft_doc::chart::MAX_CELLS;
|
||||
let spec = ChartSpec {
|
||||
kind: ChartType::Column,
|
||||
categories: (0..4500).map(|i| i.to_string()).collect(),
|
||||
series: (0..300).map(|i| ChartSeries { name: format!("S{i}"), values: vec![Some(1.0); 4500] }).collect(),
|
||||
title: None,
|
||||
legend: None,
|
||||
data_labels: false,
|
||||
};
|
||||
let x = chart_xml(&spec);
|
||||
let series = x.matches("<c:ser>").count();
|
||||
let points = x.matches("<c:pt ").count();
|
||||
assert!((1..=MAX_SERIES).contains(&series), "{series} series");
|
||||
assert!(points <= 2 * MAX_CELLS, "{points} points");
|
||||
assert!(!x.contains(&format!("<c:pt idx=\"{MAX_POINTS}\"")), "past the point limit");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn every_type_reads_back_as_itself() {
|
||||
for kind in ChartType::ALL {
|
||||
|
||||
+20
-406
@@ -9,10 +9,6 @@
|
||||
//! a verifier that tells a wrong password apart; an HMAC over the encrypted stream checks that it
|
||||
//! hasn't been damaged or tampered with.
|
||||
//!
|
||||
//! Older packages (Word 2007) use the *Standard* method (§2.3.4.5–§2.3.4.9): a binary
|
||||
//! `EncryptionInfo`, one AES key derived from the password with SHA-1 and 50,000 spins, and the
|
||||
//! whole package AES-ECB encrypted. [`decrypt`] reads both; [`encrypt`] writes agile only.
|
||||
//!
|
||||
//! Files are hostile: every stream read is capped, every length is checked before slicing, and the
|
||||
//! spin count is bounded by the specification's maximum. Passwords and keys are wiped from memory
|
||||
//! when dropped ([`Zeroizing`]) and never logged.
|
||||
@@ -51,20 +47,6 @@ const BK_KEY_VALUE: [u8; 8] = [0x14, 0x6e, 0x0b, 0xe7, 0xab, 0xac, 0xd0, 0xd6];
|
||||
const BK_HMAC_KEY: [u8; 8] = [0x5f, 0xb2, 0xad, 0x01, 0x0c, 0xb9, 0xe1, 0xf6];
|
||||
const BK_HMAC_VALUE: [u8; 8] = [0xa0, 0x67, 0x7f, 0x02, 0xb2, 0x2c, 0x84, 0x33];
|
||||
|
||||
/// Standard encryption (§2.3.2, §2.3.4.5–§2.3.4.7): `EncryptionHeader.Flags` bits, the AES and
|
||||
/// SHA-1 algorithm ids, the provider type, the fixed spin count and the SHA-1 size.
|
||||
const F_CRYPTO_API: u32 = 0x04;
|
||||
const F_EXTERNAL: u32 = 0x10;
|
||||
const F_AES: u32 = 0x20;
|
||||
const STANDARD_FLAGS: u32 = F_CRYPTO_API | F_AES;
|
||||
const ALG_AES_128: u32 = 0x660E;
|
||||
const ALG_AES_192: u32 = 0x660F;
|
||||
const ALG_AES_256: u32 = 0x6610;
|
||||
const ALG_SHA1: u32 = 0x8004;
|
||||
const PROV_RSA_AES: u32 = 0x18;
|
||||
const STANDARD_SPIN_COUNT: u32 = 50_000;
|
||||
const SHA1_SIZE: usize = 20;
|
||||
|
||||
/// An OLE compound file starts with these bytes ([MS-CFB] §2.2).
|
||||
const CFB_MAGIC: [u8; 8] = [0xD0, 0xCF, 0x11, 0xE0, 0xA1, 0xB1, 0x1A, 0xE1];
|
||||
|
||||
@@ -74,7 +56,7 @@ const NS_PASSWORD: &str = "http://schemas.microsoft.com/office/2006/keyEncryptor
|
||||
/// Is this an encrypted Office package (a compound file with `EncryptionInfo` and
|
||||
/// `EncryptedPackage` streams) rather than a zip?
|
||||
pub fn is_encrypted(bytes: &[u8]) -> bool {
|
||||
if bytes.get(..8) != Some(&CFB_MAGIC[..]) || wordcraft_doc::cfb_guard::check(bytes).is_err() {
|
||||
if bytes.get(..8) != Some(&CFB_MAGIC[..]) {
|
||||
return false;
|
||||
}
|
||||
match cfb::CompoundFile::open(Cursor::new(bytes)) {
|
||||
@@ -87,13 +69,9 @@ pub fn is_encrypted(bytes: &[u8]) -> bool {
|
||||
/// [`DocxError::PasswordRequired`] once it knows the file is one it can decrypt; a password that
|
||||
/// doesn't fit gives [`DocxError::WrongPassword`].
|
||||
pub fn decrypt(bytes: &[u8], password: Option<&str>) -> Result<Vec<u8>, DocxError> {
|
||||
wordcraft_doc::cfb_guard::check(bytes).map_err(|e| bad(format!("not a readable compound file ({e})")))?;
|
||||
let mut comp = cfb::CompoundFile::open(Cursor::new(bytes)).map_err(|e| bad(format!("not a readable compound file ({e})")))?;
|
||||
let info = stream(&mut comp, "/EncryptionInfo", MAX_INFO)?;
|
||||
let package = stream(&mut comp, "/EncryptedPackage", bytes.len() as u64)?;
|
||||
if matches!(info.get(..4), Some([2..=4, 0, 2, 0])) {
|
||||
return decrypt_standard(&info, &package, password);
|
||||
}
|
||||
let info = parse_info(&info)?;
|
||||
let Some(password) = password else { return Err(DocxError::PasswordRequired) };
|
||||
let pe = &info.password;
|
||||
@@ -114,18 +92,17 @@ pub fn decrypt(bytes: &[u8], password: Option<&str>) -> Result<Vec<u8>, DocxErro
|
||||
let key_key = derive_key(pe.hash, &hn, &BK_KEY_VALUE, pe.key_bytes);
|
||||
let key = Zeroizing::new(cbc_decrypt(&key_key, &iv, &pe.key_value)?);
|
||||
let key = Zeroizing::new(key.get(..info.key_bytes).ok_or_else(|| bad("the encrypted key is too short"))?.to_vec());
|
||||
// Data integrity (§2.3.4.14): an HMAC of the whole EncryptedPackage stream. It is required:
|
||||
// a file that could drop it would let altered ciphertext, or an altered StreamSize, through
|
||||
// unauthenticated (`parse_info` refuses a file without it).
|
||||
let (enc_key, enc_value) = &info.integrity;
|
||||
let hmac_key = Zeroizing::new(cbc_decrypt(&key, &make_iv(info.hash, &info.salt, Some(&BK_HMAC_KEY), info.block_size), enc_key)?);
|
||||
// The key is the salt, zero-padded to the block size; HMAC pads short keys with zeros
|
||||
// anyway, so the padded key gives the same HMAC.
|
||||
let hmac_key = hmac_key.get(..hmac_key.len().min(info.hash.size())).unwrap_or(&[]);
|
||||
let expected = cbc_decrypt(&key, &make_iv(info.hash, &info.salt, Some(&BK_HMAC_VALUE), info.block_size), enc_value)?;
|
||||
let expected = expected.get(..info.hash.size()).ok_or_else(|| bad("the integrity check value is too short"))?;
|
||||
if !same(&info.hash.hmac(hmac_key, &package)?, expected) {
|
||||
return Err(bad("the document is damaged (its integrity check failed)"));
|
||||
// Data integrity (§2.3.4.14): an HMAC of the whole EncryptedPackage stream.
|
||||
if let Some((enc_key, enc_value)) = &info.integrity {
|
||||
let hmac_key = Zeroizing::new(cbc_decrypt(&key, &make_iv(info.hash, &info.salt, Some(&BK_HMAC_KEY), info.block_size), enc_key)?);
|
||||
// The key is the salt, zero-padded to the block size; HMAC pads short keys with zeros
|
||||
// anyway, so the padded key gives the same HMAC.
|
||||
let hmac_key = hmac_key.get(..hmac_key.len().min(info.hash.size())).unwrap_or(&[]);
|
||||
let expected = cbc_decrypt(&key, &make_iv(info.hash, &info.salt, Some(&BK_HMAC_VALUE), info.block_size), enc_value)?;
|
||||
let expected = expected.get(..info.hash.size()).ok_or_else(|| bad("the integrity check value is too short"))?;
|
||||
if !same(&info.hash.hmac(hmac_key, &package)?, expected) {
|
||||
return Err(bad("the document is damaged (its integrity check failed)"));
|
||||
}
|
||||
}
|
||||
// The package itself (§2.3.4.15): StreamSize, then 4096-byte segments, each with its own IV.
|
||||
let size = package
|
||||
@@ -147,92 +124,6 @@ pub fn decrypt(bytes: &[u8], password: Option<&str>) -> Result<Vec<u8>, DocxErro
|
||||
Ok(out)
|
||||
}
|
||||
|
||||
/// [`decrypt`] for Standard encryption (§2.3.4.5–§2.3.4.9): one AES key from the password,
|
||||
/// checked with the verifier, and the whole package AES-ECB encrypted after its 8-byte size.
|
||||
fn decrypt_standard(info: &[u8], package: &[u8], password: Option<&str>) -> Result<Vec<u8>, DocxError> {
|
||||
let s = parse_standard(info)?;
|
||||
let Some(password) = password else { return Err(DocxError::PasswordRequired) };
|
||||
let key = standard_key(s.salt, password, s.key_bytes);
|
||||
// The verifier (§2.3.4.9): a random value and its SHA-1, both encrypted with the key.
|
||||
let verifier = Zeroizing::new(ecb(&key, s.verifier, false)?);
|
||||
let hash = Zeroizing::new(ecb(&key, s.verifier_hash, false)?);
|
||||
let hash = hash.get(..SHA1_SIZE).ok_or_else(|| bad("the password verifier hash is too short"))?;
|
||||
if !same(&Hash::Sha1.hash(&[&verifier]), hash) {
|
||||
return Err(DocxError::WrongPassword);
|
||||
}
|
||||
// The package (§2.3.4.4): StreamSize, then the data; bytes past the last block are ignored.
|
||||
let size = package
|
||||
.get(..8)
|
||||
.and_then(|b| <[u8; 8]>::try_from(b).ok())
|
||||
.map(u64::from_le_bytes)
|
||||
.ok_or_else(|| bad("the encrypted package is truncated"))?;
|
||||
let data = package.get(8..).unwrap_or(&[]);
|
||||
let size = usize::try_from(size).ok().filter(|s| *s <= data.len()).ok_or_else(|| bad("the encrypted package is truncated"))?;
|
||||
let needed = size.div_ceil(BLOCK).saturating_mul(BLOCK);
|
||||
let data = data.get(..needed).ok_or_else(|| bad("the encrypted package is truncated"))?;
|
||||
let mut out = ecb(&key, data, false)?;
|
||||
out.truncate(size);
|
||||
Ok(out)
|
||||
}
|
||||
|
||||
/// What a Standard `EncryptionInfo` stream says (§2.3.4.5): the key size, and the verifier's salt
|
||||
/// and encrypted values.
|
||||
struct Standard<'a> {
|
||||
key_bytes: usize,
|
||||
salt: &'a [u8],
|
||||
verifier: &'a [u8],
|
||||
verifier_hash: &'a [u8],
|
||||
}
|
||||
|
||||
fn parse_standard(info: &[u8]) -> Result<Standard<'_>, DocxError> {
|
||||
let short = || bad("EncryptionInfo is truncated");
|
||||
let u32_at = |at: usize| -> Result<u32, DocxError> {
|
||||
let b = at.checked_add(4).and_then(|end| info.get(at..end)).ok_or_else(short)?;
|
||||
<[u8; 4]>::try_from(b).map(u32::from_le_bytes).map_err(|_| short())
|
||||
};
|
||||
let flags = u32_at(4)?;
|
||||
if flags & F_EXTERNAL != 0 {
|
||||
return Err(bad("this file uses extensible encryption, which WordCraft can't open"));
|
||||
}
|
||||
if flags & STANDARD_FLAGS != STANDARD_FLAGS {
|
||||
return Err(bad("this file's encryption isn't AES, which WordCraft can't open"));
|
||||
}
|
||||
// EncryptionHeader (§2.3.2): Flags, SizeExtra, AlgID, AlgIDHash, KeySize, ProviderType, two
|
||||
// reserved words, then the CSP name, which isn't needed.
|
||||
let header_size = usize::try_from(u32_at(8)?).map_err(|_| short())?;
|
||||
if header_size < 32 {
|
||||
return Err(bad("the encryption header is too short"));
|
||||
}
|
||||
let verifier_at = header_size.checked_add(12).filter(|&v| v <= info.len()).ok_or_else(short)?;
|
||||
// AlgID 0 and AlgIDHash 0 mean "set by the flags": AES-128 and SHA-1.
|
||||
let bits = match u32_at(20)? {
|
||||
0 | ALG_AES_128 => 128,
|
||||
ALG_AES_192 => 192,
|
||||
ALG_AES_256 => 256,
|
||||
id => return Err(bad(format!("encryption algorithm 0x{id:X} isn't supported"))),
|
||||
};
|
||||
let hash_id = u32_at(24)?;
|
||||
if !matches!(hash_id, 0 | ALG_SHA1) {
|
||||
return Err(bad(format!("hash algorithm 0x{hash_id:X} isn't supported")));
|
||||
}
|
||||
if u32_at(28)? != bits {
|
||||
return Err(bad("the key size doesn't match the encryption algorithm"));
|
||||
}
|
||||
// EncryptionVerifier (§2.3.3): SaltSize (16), Salt, EncryptedVerifier, VerifierHashSize (20
|
||||
// for SHA-1), EncryptedVerifierHash (32 bytes for AES).
|
||||
if u32_at(verifier_at)? != SALT as u32 {
|
||||
return Err(bad("the salt isn't 16 bytes"));
|
||||
}
|
||||
let field = |at: usize, len: usize| verifier_at.checked_add(at).and_then(|a| info.get(a..a.checked_add(len)?)).ok_or_else(short);
|
||||
let salt = field(4, SALT)?;
|
||||
let verifier = field(4 + SALT, BLOCK)?;
|
||||
if u32_at(verifier_at.saturating_add(4 + SALT + BLOCK))? != SHA1_SIZE as u32 {
|
||||
return Err(bad("the password verifier hash isn't SHA-1"));
|
||||
}
|
||||
let verifier_hash = field(8 + SALT + BLOCK, 2 * BLOCK)?;
|
||||
Ok(Standard { key_bytes: bits as usize / 8, salt, verifier, verifier_hash })
|
||||
}
|
||||
|
||||
/// Encrypt a zip package with `password` (agile encryption: AES-256-CBC, SHA-512, Word's spin
|
||||
/// count), as Word's "Encrypt with Password" does.
|
||||
pub fn encrypt(package: &[u8], password: &str) -> Result<Vec<u8>, DocxError> {
|
||||
@@ -321,47 +212,6 @@ pub fn encrypt_with_spin_count(package: &[u8], password: &str, spin_count: u32)
|
||||
Ok(comp.into_inner().into_inner())
|
||||
}
|
||||
|
||||
/// A package encrypted with the *Standard* method (§2.3.4.5–§2.3.4.9: AES-ECB, SHA-1, 50,000
|
||||
/// spins), as Word 2007 wrote it, for tests only: WordCraft saves with the agile method
|
||||
/// ([`encrypt`]). `key_bits` is 128, 192 or 256. The compound file holds just the two streams
|
||||
/// [`decrypt`] reads.
|
||||
#[doc(hidden)]
|
||||
pub fn encrypt_standard_for_tests(package: &[u8], password: &str, key_bits: u32) -> Result<Vec<u8>, DocxError> {
|
||||
check_password(password)?;
|
||||
let alg_id = match key_bits {
|
||||
128 => ALG_AES_128,
|
||||
192 => ALG_AES_192,
|
||||
256 => ALG_AES_256,
|
||||
_ => return Err(bad(format!("unsupported AES key size {key_bits}"))),
|
||||
};
|
||||
let salt = random(SALT)?;
|
||||
let verifier = random(SALT)?;
|
||||
let key = standard_key(&salt, password, key_bits as usize / 8);
|
||||
let mut header = Vec::new();
|
||||
for v in [STANDARD_FLAGS, 0, alg_id, ALG_SHA1, key_bits, PROV_RSA_AES, 0, 0] {
|
||||
header.extend_from_slice(&v.to_le_bytes());
|
||||
}
|
||||
// CSPName: an empty null-terminated UTF-16 string.
|
||||
header.extend_from_slice(&[0, 0]);
|
||||
let mut info = vec![3, 0, 2, 0];
|
||||
info.extend_from_slice(&STANDARD_FLAGS.to_le_bytes());
|
||||
info.extend_from_slice(&(header.len() as u32).to_le_bytes());
|
||||
info.extend_from_slice(&header);
|
||||
info.extend_from_slice(&(SALT as u32).to_le_bytes());
|
||||
info.extend_from_slice(&salt);
|
||||
info.extend_from_slice(&ecb(&key, &verifier, true)?);
|
||||
info.extend_from_slice(&(SHA1_SIZE as u32).to_le_bytes());
|
||||
info.extend_from_slice(&ecb(&key, &padded(&Hash::Sha1.hash(&[&verifier])), true)?);
|
||||
let mut stream = (package.len() as u64).to_le_bytes().to_vec();
|
||||
stream.extend_from_slice(&ecb(&key, &padded(package), true)?);
|
||||
let io = |e: std::io::Error| bad(format!("couldn't write the compound file ({e})"));
|
||||
let mut comp = cfb::CompoundFile::create_with_version(cfb::Version::V3, Cursor::new(Vec::new())).map_err(io)?;
|
||||
comp.create_stream("/EncryptionInfo").map_err(io)?.write_all(&info).map_err(io)?;
|
||||
comp.create_stream("/EncryptedPackage").map_err(io)?.write_all(&stream).map_err(io)?;
|
||||
comp.flush().map_err(io)?;
|
||||
Ok(comp.into_inner().into_inner())
|
||||
}
|
||||
|
||||
/// Refuse a password Word wouldn't accept: empty, or longer than [`MAX_PASSWORD_CHARS`].
|
||||
pub fn check_password(password: &str) -> Result<(), DocxError> {
|
||||
if password.is_empty() {
|
||||
@@ -646,45 +496,6 @@ fn derive_key(hash: Hash, hn: &[u8], block_key: &[u8], len: usize) -> Zeroizing<
|
||||
k
|
||||
}
|
||||
|
||||
/// Standard encryption's key (§2.3.4.7): the SHA-1 password hash after 50,000 iterations, hashed
|
||||
/// with block number 0; then X1 = SHA-1 of that hash XOR-ed into 64 bytes of 0x36, X2 the same
|
||||
/// with 0x5C, and the key is the first `len` bytes of X1 + X2.
|
||||
fn standard_key(salt: &[u8], password: &str, len: usize) -> Zeroizing<Vec<u8>> {
|
||||
let hn = iterate(Hash::Sha1, salt, password, STANDARD_SPIN_COUNT);
|
||||
let hfinal = Zeroizing::new(Hash::Sha1.hash(&[&hn, &0u32.to_le_bytes()]));
|
||||
let x = |fill: u8| {
|
||||
let mut buf = Zeroizing::new([fill; 64]);
|
||||
for (b, h) in buf.iter_mut().zip(hfinal.iter()) {
|
||||
*b ^= h;
|
||||
}
|
||||
Zeroizing::new(Hash::Sha1.hash(&[&buf[..]]))
|
||||
};
|
||||
let mut key = Zeroizing::new(x(0x36).to_vec());
|
||||
key.extend_from_slice(&x(0x5C));
|
||||
key.truncate(len);
|
||||
key
|
||||
}
|
||||
|
||||
/// AES-ECB without padding (Standard encryption, §2.3.4.7–§2.3.4.9): `data` must be whole blocks.
|
||||
fn ecb(key: &[u8], data: &[u8], encrypt: bool) -> Result<Vec<u8>, DocxError> {
|
||||
let aes = Aes::new(key)?;
|
||||
let (blocks, rest) = data.as_chunks::<BLOCK>();
|
||||
if !rest.is_empty() {
|
||||
return Err(bad("encrypted data isn't a whole number of blocks"));
|
||||
}
|
||||
let mut out = Vec::with_capacity(data.len());
|
||||
for b in blocks {
|
||||
let mut x = aes::Block::from(*b);
|
||||
if encrypt {
|
||||
aes.encrypt(&mut x);
|
||||
} else {
|
||||
aes.decrypt(&mut x);
|
||||
}
|
||||
out.extend_from_slice(&x);
|
||||
}
|
||||
Ok(out)
|
||||
}
|
||||
|
||||
/// An initialization vector (§2.3.4.12): H(salt + blockKey), or the salt itself, truncated to
|
||||
/// the block size or padded with 0x36.
|
||||
fn make_iv(hash: Hash, salt: &[u8], block_key: Option<&[u8]>, block_size: usize) -> Vec<u8> {
|
||||
@@ -799,7 +610,7 @@ struct Info {
|
||||
block_size: usize,
|
||||
key_bytes: usize,
|
||||
/// encryptedHmacKey, encryptedHmacValue.
|
||||
integrity: (Vec<u8>, Vec<u8>),
|
||||
integrity: Option<(Vec<u8>, Vec<u8>)>,
|
||||
password: PasswordEncryptor,
|
||||
}
|
||||
|
||||
@@ -831,6 +642,7 @@ fn parse_info(info: &[u8]) -> Result<Info, DocxError> {
|
||||
};
|
||||
match (major, minor) {
|
||||
(4, 4) => {}
|
||||
(2..=4, 2) => return Err(bad("this file uses Standard encryption (Office 2007), which WordCraft can't open yet")),
|
||||
(3 | 4, 3) => return Err(bad("this file uses extensible encryption, which WordCraft can't open")),
|
||||
_ => return Err(bad(format!("unknown encryption version {major}.{minor}"))),
|
||||
}
|
||||
@@ -840,12 +652,10 @@ fn parse_info(info: &[u8]) -> Result<Info, DocxError> {
|
||||
}
|
||||
let key_data = root.els().find(|e| e.local() == "keyData").ok_or_else(|| bad("EncryptionInfo has no keyData"))?;
|
||||
let kd = params(key_data)?;
|
||||
// The integrity check is mandatory (§2.3.4.14): a file is not trusted to opt out of it.
|
||||
let di = root
|
||||
.els()
|
||||
.find(|e| e.local() == "dataIntegrity")
|
||||
.ok_or_else(|| bad("EncryptionInfo has no dataIntegrity, so the document cannot be checked for tampering"))?;
|
||||
let integrity = (b64_attr(di, "encryptedHmacKey")?, b64_attr(di, "encryptedHmacValue")?);
|
||||
let integrity = match root.els().find(|e| e.local() == "dataIntegrity") {
|
||||
Some(di) => Some((b64_attr(di, "encryptedHmacKey")?, b64_attr(di, "encryptedHmacValue")?)),
|
||||
None => None,
|
||||
};
|
||||
// The password key encryptor is the `encryptedKey` with a spin count (a certificate key
|
||||
// encryptor has none).
|
||||
let encryptors = root.els().find(|e| e.local() == "keyEncryptors").ok_or_else(|| bad("EncryptionInfo has no keyEncryptors"))?;
|
||||
@@ -1086,22 +896,6 @@ mod tests {
|
||||
assert!(encrypt(&pkg, &"x".repeat(MAX_PASSWORD_CHARS + 1)).is_err());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_file_without_data_integrity_is_refused() {
|
||||
let enc = encrypt_with_spin_count(&zip_package(), "pw", 4).unwrap();
|
||||
let mut comp = cfb::CompoundFile::open(Cursor::new(enc)).unwrap();
|
||||
let mut info = Vec::new();
|
||||
comp.open_stream("/EncryptionInfo").unwrap().read_to_end(&mut info).unwrap();
|
||||
let xml = String::from_utf8(info[8..].to_vec()).unwrap();
|
||||
let mut stripped = info[..8].to_vec();
|
||||
stripped.extend_from_slice(xml.replacen("<dataIntegrity", "<nothing", 1).as_bytes());
|
||||
comp.create_stream("/EncryptionInfo").unwrap().write_all(&stripped).unwrap();
|
||||
comp.flush().unwrap();
|
||||
let enc = comp.into_inner().into_inner();
|
||||
assert!(matches!(decrypt(&enc, Some("pw")), Err(DocxError::Encryption(m)) if m.contains("dataIntegrity")));
|
||||
assert!(crate::read_with_password(&enc, Some("pw")).is_err());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn tampered_package_fails_the_integrity_check() {
|
||||
let pkg = zip_package();
|
||||
@@ -1161,8 +955,8 @@ mod tests {
|
||||
cases.push((with("ChainingModeCBC", "ChainingModeCFB"), package.clone(), true));
|
||||
cases.push((with("saltValue=\"", "saltValue=\"!!"), package.clone(), true));
|
||||
cases.push((with("encryptedKeyValue=\"", "encryptedKeyValue=\"AAAA"), package.clone(), true));
|
||||
// Data integrity is mandatory: a file without it is refused, never opened unchecked.
|
||||
cases.push((with("<dataIntegrity", "<nothing"), package.clone(), true));
|
||||
// Data integrity is optional; without it a good file still opens.
|
||||
cases.push((with("<dataIntegrity", "<nothing"), package.clone(), false));
|
||||
let mut huge = package.clone();
|
||||
huge[..8].copy_from_slice(&u64::MAX.to_le_bytes());
|
||||
cases.push((with("<dataIntegrity", "<nothing"), huge, true));
|
||||
@@ -1179,184 +973,4 @@ mod tests {
|
||||
assert!(!is_encrypted(&fake));
|
||||
assert!(decrypt(&fake, Some("pw")).is_err());
|
||||
}
|
||||
|
||||
/// Standard encryption's key derivation (§2.3.4.7) for the password "Password1234_" and the
|
||||
/// salt 00 01 … 0F. The expected keys were computed outside WordCraft (Python's hashlib)
|
||||
/// following the specification's steps, so the encryptor and decryptor can't share a mistake.
|
||||
#[test]
|
||||
fn standard_key_matches_an_independent_derivation() {
|
||||
let salt: Vec<u8> = (0..16).collect();
|
||||
let hex = |b: &[u8]| b.iter().map(|x| format!("{x:02x}")).collect::<String>();
|
||||
assert_eq!(hex(&standard_key(&salt, "Password1234_", 16)), "340fc93a8e2154db8304d055df81cebc");
|
||||
assert_eq!(hex(&standard_key(&salt, "Password1234_", 32)), "340fc93a8e2154db8304d055df81cebcef5e89fb8e5eb520efdca76f3330e032");
|
||||
assert_eq!(standard_key(&salt, "Password1234_", 24).len(), 24);
|
||||
// AES-ECB against FIPS-197 Appendix C.1 and C.3.
|
||||
let pt: Vec<u8> = (0..16u8).map(|i| i * 0x11).collect();
|
||||
let k128: Vec<u8> = (0..16).collect();
|
||||
let k256: Vec<u8> = (0..32).collect();
|
||||
assert_eq!(hex(&ecb(&k128, &pt, true).unwrap()), "69c4e0d86a7b0430d8cdb78070b4c55a");
|
||||
assert_eq!(hex(&ecb(&k256, &pt, true).unwrap()), "8ea2b7ca516745bfeafc49904b496089");
|
||||
assert_eq!(ecb(&k256, &ecb(&k256, &pt, true).unwrap(), false).unwrap(), pt);
|
||||
assert!(ecb(&k128, &pt[..15], false).is_err());
|
||||
}
|
||||
|
||||
/// Standard encryption (Word 2007's AES-ECB method) opens with the right password, for each
|
||||
/// AES key size, and says when the password is missing or wrong — as agile files do.
|
||||
#[test]
|
||||
fn standard_encryption_round_trips() {
|
||||
let pkg = zip_package();
|
||||
for bits in [128, 192, 256] {
|
||||
let enc = encrypt_standard_for_tests(&pkg, "pässwörd 🔑", bits).unwrap();
|
||||
assert!(is_encrypted(&enc));
|
||||
assert_eq!(decrypt(&enc, Some("pässwörd 🔑")), Ok(pkg.clone()), "AES-{bits}");
|
||||
assert_eq!(decrypt(&enc, None), Err(DocxError::PasswordRequired));
|
||||
assert_eq!(decrypt(&enc, Some("Pässwörd 🔑")), Err(DocxError::WrongPassword));
|
||||
assert_eq!(crate::read(&enc), Err(DocxError::PasswordRequired));
|
||||
}
|
||||
let enc = encrypt_standard_for_tests(&pkg, "pw", 128).unwrap();
|
||||
let doc = crate::read_with_password(&enc, Some("pw")).unwrap();
|
||||
assert!(doc.plain_text(wordcraft_doc::StoryRef::Body).contains("Top secret plans"));
|
||||
// Sizes on and off the block boundary, and empty.
|
||||
for n in [0usize, 1, 15, 16, 17, SEGMENT + 3] {
|
||||
let data: Vec<u8> = (0..n).map(|i| (i * 7) as u8).collect();
|
||||
assert_eq!(decrypt(&encrypt_standard_for_tests(&data, "x", 256).unwrap(), Some("x")).unwrap(), data);
|
||||
}
|
||||
assert!(encrypt_standard_for_tests(&pkg, "pw", 64).is_err());
|
||||
assert!(encrypt_standard_for_tests(&pkg, "", 128).is_err());
|
||||
}
|
||||
|
||||
/// A Standard-encrypted file assembled byte by byte from [MS-OFFCRYPTO] §2.3.4.5, §2.3.2, §2.3.3
|
||||
/// and §2.3.4.4 with literal field values, without the test encryptor, so a layout mistake
|
||||
/// shared by the encryptor and the reader can't pass. Only the key derivation and AES-ECB
|
||||
/// (pinned by `standard_key_matches_an_independent_derivation`) are reused.
|
||||
fn hand_built_standard(payload: &[u8], password: &str) -> Vec<u8> {
|
||||
let salt: [u8; 16] = [0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF];
|
||||
let verifier: [u8; 16] = *b"verifier-16bytes";
|
||||
let key = standard_key(&salt, password, 16);
|
||||
let mut hash = sha1::Sha1::digest(verifier).to_vec();
|
||||
hash.resize(32, 0); // the hash, zero-padded to two AES blocks
|
||||
let csp: Vec<u8> = "WordCraft test provider".encode_utf16().chain([0]).flat_map(u16::to_le_bytes).collect();
|
||||
|
||||
let mut header = Vec::new();
|
||||
header.extend_from_slice(&0x24u32.to_le_bytes()); // Flags: fCryptoAPI | fAES
|
||||
header.extend_from_slice(&0u32.to_le_bytes()); // SizeExtra
|
||||
header.extend_from_slice(&0x660Eu32.to_le_bytes()); // AlgID: AES-128
|
||||
header.extend_from_slice(&0x8004u32.to_le_bytes()); // AlgIDHash: SHA-1
|
||||
header.extend_from_slice(&128u32.to_le_bytes()); // KeySize, in bits
|
||||
header.extend_from_slice(&0x18u32.to_le_bytes()); // ProviderType: AES
|
||||
header.extend_from_slice(&0u32.to_le_bytes()); // Reserved1
|
||||
header.extend_from_slice(&0u32.to_le_bytes()); // Reserved2
|
||||
header.extend_from_slice(&csp); // CSPName: UTF-16LE, NUL-terminated
|
||||
|
||||
let mut info = Vec::new();
|
||||
info.extend_from_slice(&3u16.to_le_bytes()); // Version major
|
||||
info.extend_from_slice(&2u16.to_le_bytes()); // Version minor
|
||||
info.extend_from_slice(&0x24u32.to_le_bytes()); // Flags: fCryptoAPI | fAES
|
||||
info.extend_from_slice(&(header.len() as u32).to_le_bytes()); // HeaderSize
|
||||
info.extend_from_slice(&header);
|
||||
info.extend_from_slice(&16u32.to_le_bytes()); // SaltSize
|
||||
info.extend_from_slice(&salt); // Salt
|
||||
info.extend_from_slice(&ecb(&key, &verifier, true).unwrap()); // EncryptedVerifier
|
||||
info.extend_from_slice(&20u32.to_le_bytes()); // VerifierHashSize (SHA-1)
|
||||
info.extend_from_slice(&ecb(&key, &hash, true).unwrap()); // EncryptedVerifierHash
|
||||
|
||||
let mut padded = payload.to_vec();
|
||||
padded.resize(payload.len().next_multiple_of(16), 0);
|
||||
let mut package = (payload.len() as u64).to_le_bytes().to_vec(); // StreamSize
|
||||
package.extend_from_slice(&ecb(&key, &padded, true).unwrap());
|
||||
package.extend_from_slice(&[0xEE, 0xEE, 0xEE]); // trailing padding the spec allows
|
||||
cfb_with(&info, &package)
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn standard_package_built_by_hand_from_the_spec_opens() {
|
||||
let payload: Vec<u8> = (0..37u8).map(|i| i.wrapping_mul(5)).collect();
|
||||
let file = hand_built_standard(&payload, "pw");
|
||||
assert!(is_encrypted(&file));
|
||||
assert_eq!(decrypt(&file, Some("pw")), Ok(payload.clone()));
|
||||
assert_eq!(decrypt(&file, Some("PW")), Err(DocxError::WrongPassword));
|
||||
assert_eq!(decrypt(&file, None), Err(DocxError::PasswordRequired));
|
||||
// The test encryptor writes the same fixed fields: version and flags (bytes 0..8) and the
|
||||
// header's eight words (12..44); HeaderSize differs with the CSP name, the rest is random.
|
||||
let (hand, _) = streams(&file);
|
||||
let (made, _) = streams(&encrypt_standard_for_tests(&payload, "pw", 128).unwrap());
|
||||
assert_eq!(hand[..8], made[..8]);
|
||||
assert_eq!(hand[12..44], made[12..44]);
|
||||
}
|
||||
|
||||
/// The two streams of an encrypted package.
|
||||
fn streams(file: &[u8]) -> (Vec<u8>, Vec<u8>) {
|
||||
let mut comp = cfb::CompoundFile::open(Cursor::new(file)).unwrap();
|
||||
let mut info = Vec::new();
|
||||
comp.open_stream("/EncryptionInfo").unwrap().read_to_end(&mut info).unwrap();
|
||||
let mut package = Vec::new();
|
||||
comp.open_stream("/EncryptedPackage").unwrap().read_to_end(&mut package).unwrap();
|
||||
(info, package)
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn hostile_standard_encryption_info_errors_without_panicking() {
|
||||
let data = b"a small package".to_vec();
|
||||
let (info, package) = streams(&encrypt_standard_for_tests(&data, "pw", 128).unwrap());
|
||||
// Layout: version 0, flags 4, header size 8, header 12 (flags, sizeExtra, algId 20,
|
||||
// algIdHash 24, keySize 28, provider, 2 reserved, an empty CSP name), then the verifier at
|
||||
// 46: saltSize, salt 50, encrypted verifier 66, verifierHashSize 82, its hash 86..118.
|
||||
assert_eq!(info.len(), 118);
|
||||
let put = |at: usize, v: u32| {
|
||||
let mut i = info.clone();
|
||||
i[at..at + 4].copy_from_slice(&v.to_le_bytes());
|
||||
i
|
||||
};
|
||||
let mut cases: Vec<(Vec<u8>, Vec<u8>)> = Vec::new();
|
||||
for n in 0..info.len() {
|
||||
cases.push((info[..n].to_vec(), package.clone()));
|
||||
}
|
||||
for v in [0, 4, 31, 0x1000, u32::MAX] {
|
||||
cases.push((put(8, v), package.clone()));
|
||||
}
|
||||
// No AES, no CryptoAPI, an external (extensible) provider.
|
||||
for v in [F_CRYPTO_API, F_AES, STANDARD_FLAGS | F_EXTERNAL, 0] {
|
||||
cases.push((put(4, v), package.clone()));
|
||||
}
|
||||
// RC4, MD5, key sizes that don't fit the algorithm.
|
||||
cases.push((put(20, 0x6801), package.clone()));
|
||||
cases.push((put(24, 0x8003), package.clone()));
|
||||
for v in [0, 64, 192, 256, u32::MAX] {
|
||||
cases.push((put(28, v), package.clone()));
|
||||
}
|
||||
for v in [0, 15, 17, u32::MAX] {
|
||||
cases.push((put(46, v), package.clone()));
|
||||
}
|
||||
for v in [0, 16, 32, u32::MAX] {
|
||||
cases.push((put(82, v), package.clone()));
|
||||
}
|
||||
// The package: truncated, an absurd StreamSize, a StreamSize past the data, data that
|
||||
// isn't whole blocks.
|
||||
for n in [0, 4, 8, 9, package.len() - 1] {
|
||||
cases.push((info.clone(), package[..n].to_vec()));
|
||||
}
|
||||
for size in [u64::MAX, 1 << 40, package.len() as u64] {
|
||||
let mut p = package.clone();
|
||||
p[..8].copy_from_slice(&size.to_le_bytes());
|
||||
cases.push((info.clone(), p));
|
||||
}
|
||||
for (i, p) in cases {
|
||||
let file = cfb_with(&i, &p);
|
||||
assert!(decrypt(&file, Some("pw")).is_err());
|
||||
let _ = crate::read_with_password(&file, Some("pw"));
|
||||
assert!(crate::read(&file).is_err());
|
||||
}
|
||||
// Version 2.2 and 4.2 are Standard too; trailing bytes after the last block are ignored.
|
||||
for v in [2u8, 4] {
|
||||
let mut i = info.clone();
|
||||
i[0] = v;
|
||||
assert_eq!(decrypt(&cfb_with(&i, &package), Some("pw")).unwrap(), data);
|
||||
}
|
||||
let mut p = package.clone();
|
||||
p.extend_from_slice(&[1, 2, 3]);
|
||||
assert_eq!(decrypt(&cfb_with(&info, &p), Some("pw")).unwrap(), data);
|
||||
// AlgID and AlgIDHash 0 mean "from the flags": AES-128 and SHA-1.
|
||||
assert_eq!(decrypt(&cfb_with(&put(20, 0), &package), Some("pw")).unwrap(), data);
|
||||
assert_eq!(decrypt(&cfb_with(&put(24, 0), &package), Some("pw")).unwrap(), data);
|
||||
}
|
||||
}
|
||||
|
||||
+1
-99
@@ -437,47 +437,7 @@ fn xml_fragment<'a>(ct: &str, bytes: &'a [u8]) -> Option<&'a str> {
|
||||
Some(rest) => rest.split_once("?>").map(|(_, r)| r)?.trim_start(),
|
||||
None => s,
|
||||
};
|
||||
(s.starts_with('<') && !s.contains("<!DOCTYPE") && one_element(s)).then_some(s)
|
||||
}
|
||||
|
||||
/// Whether `s` is exactly one well-formed element, with only whitespace, comments and processing
|
||||
/// instructions around it. Kept bytes are inlined into `pkg:xmlData` only then; anything else could
|
||||
/// close the wrapper and add parts of its own, so it goes out as `pkg:binaryData` (#3742).
|
||||
fn one_element(s: &str) -> bool {
|
||||
let mut r = quick_xml::Reader::from_str(s);
|
||||
r.config_mut().check_comments = true;
|
||||
let (mut depth, mut roots) = (0usize, 0usize);
|
||||
loop {
|
||||
let attrs_ok = |e: &BytesStart| e.attributes().all(|a| a.is_ok());
|
||||
match r.read_event() {
|
||||
Ok(Event::Start(e)) => {
|
||||
if !attrs_ok(&e) || (depth == 0 && roots > 0) {
|
||||
return false;
|
||||
}
|
||||
roots += usize::from(depth == 0);
|
||||
depth += 1;
|
||||
}
|
||||
Ok(Event::Empty(e)) => {
|
||||
if !attrs_ok(&e) || (depth == 0 && roots > 0) {
|
||||
return false;
|
||||
}
|
||||
roots += usize::from(depth == 0);
|
||||
}
|
||||
Ok(Event::End(_)) => match depth.checked_sub(1) {
|
||||
Some(d) => depth = d,
|
||||
None => return false,
|
||||
},
|
||||
Ok(Event::Text(t)) if depth == 0 => {
|
||||
if !t.iter().all(u8::is_ascii_whitespace) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
Ok(Event::CData(_) | Event::GeneralRef(_)) if depth == 0 => return false,
|
||||
Ok(Event::Decl(_) | Event::DocType(_)) | Err(_) => return false,
|
||||
Ok(Event::Eof) => return depth == 0 && roots == 1,
|
||||
Ok(_) => {}
|
||||
}
|
||||
}
|
||||
(s.starts_with('<') && !s.contains("<!DOCTYPE")).then_some(s)
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
@@ -534,64 +494,6 @@ mod tests {
|
||||
assert!(matches!(unpack(package(xml::MAX_DEPTH + 1).as_bytes()), Err(DocxError::Limit(_))));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn kept_xml_that_would_break_out_of_its_part_is_written_in_base64() {
|
||||
// A part kept from a file is written back as it was read; inlined as is, markup that
|
||||
// closes the wrapper adds parts of its own (here a second core properties part, read
|
||||
// before the real one) (#3742).
|
||||
let evil = format!(
|
||||
"<a/></pkg:xmlData></pkg:part><pkg:part pkg:name=\"/docProps/core.xml\" pkg:contentType=\"application/vnd.openxmlformats-package.core-properties+xml\"><pkg:xmlData><cp:coreProperties xmlns:cp=\"{}\" xmlns:dc=\"http://purl.org/dc/elements/1.1/\"><dc:title>Injected</dc:title></cp:coreProperties></pkg:xmlData></pkg:part><pkg:part pkg:name=\"/x.xml\"><pkg:xmlData><b>",
|
||||
"http://schemas.openxmlformats.org/package/2006/metadata/core-properties"
|
||||
);
|
||||
let w_ns = "xmlns:w=\"http://schemas.openxmlformats.org/wordprocessingml/2006/main\"";
|
||||
let rel = |id: &str, kind: &str, target: &str| format!("<Relationship Id=\"{id}\" Type=\"{kind}\" Target=\"{target}\"/>");
|
||||
let rels = format!(
|
||||
"<Relationships xmlns=\"http://schemas.openxmlformats.org/package/2006/relationships\">{}{}</Relationships>",
|
||||
rel("rId1", crate::package::rt::OFFICE_DOC, "word/document.xml"),
|
||||
rel("rId2", crate::package::rt::CUSTOM_XML, "customXml/item1.xml")
|
||||
);
|
||||
let types = "<Types xmlns=\"http://schemas.openxmlformats.org/package/2006/content-types\"><Default Extension=\"xml\" ContentType=\"application/xml\"/></Types>";
|
||||
let z = crate::package::zip_entries(&[
|
||||
("[Content_Types].xml".into(), types.as_bytes().to_vec()),
|
||||
("_rels/.rels".into(), rels.into_bytes()),
|
||||
("word/document.xml".into(), format!("<w:document {w_ns}><w:body><w:p/></w:body></w:document>").into_bytes()),
|
||||
("customXml/item1.xml".into(), evil.clone().into_bytes()),
|
||||
])
|
||||
.unwrap();
|
||||
let mut doc = crate::read(&z).unwrap();
|
||||
assert!(doc.passthrough.values().any(|b| b.as_slice() == evil.as_bytes()), "the part is kept as read");
|
||||
doc.core.title = "Real".into();
|
||||
let flat = write(&doc).unwrap();
|
||||
let text = String::from_utf8_lossy(&flat);
|
||||
assert_eq!(text.matches("pkg:name=\"/docProps/core.xml\"").count(), 1, "{text}");
|
||||
let back = read(&flat).unwrap();
|
||||
assert_eq!(back.core.title, "Real");
|
||||
assert!(back.passthrough.values().any(|b| b.as_slice() == evil.as_bytes()), "the part survives, in base64");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn only_one_well_formed_element_is_inlined() {
|
||||
let ct = "application/xml";
|
||||
for ok in ["<a/>", "<?xml version=\"1.0\"?>\n<a x=\"1\"><b>t & <![CDATA[c]]></b></a>\n<!-- end -->", "<!-- c --><a/><?pi x?>"] {
|
||||
assert!(xml_fragment(ct, ok.as_bytes()).is_some(), "{ok}");
|
||||
}
|
||||
for bad in [
|
||||
"<a/><b/>",
|
||||
"<a></b>",
|
||||
"<a>",
|
||||
"<a/></x>",
|
||||
"<a/>text",
|
||||
"<a/>&",
|
||||
"<a x=1/>",
|
||||
"<a/><?xml version=\"1.0\"?>",
|
||||
"<a></a></pkg:xmlData>",
|
||||
"<a><!-- x -- y --></a>",
|
||||
"<a/><!-- x--->",
|
||||
] {
|
||||
assert!(xml_fragment(ct, bad.as_bytes()).is_none(), "{bad}");
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn repeated_and_nameless_parts_count_against_the_size_limits() {
|
||||
let (mut files, mut types, mut total) = (BTreeMap::new(), Vec::new(), 0u64);
|
||||
|
||||
@@ -4,8 +4,6 @@
|
||||
//! back. [`embedded_fonts`] hands their font files to whoever lays the document out, undoing the
|
||||
//! obfuscation Word applies to them (ECMA-376 Part 1, 17.8.1).
|
||||
|
||||
use std::collections::HashSet;
|
||||
|
||||
use wordcraft_doc::Document;
|
||||
|
||||
use crate::package::{EMBEDDED_PARTS, EmbeddedManifest, FONT_TABLE_PART};
|
||||
@@ -13,11 +11,6 @@ use crate::package::{EMBEDDED_PARTS, EmbeddedManifest, FONT_TABLE_PART};
|
||||
/// Most font files taken from one document.
|
||||
const MAX_FONTS: usize = 64;
|
||||
|
||||
/// Most bytes of font files copied out of one document: one part's limit, however many
|
||||
/// references the font table makes (each reference is a copy; 64 of them to one large part
|
||||
/// were 64 copies, #3692).
|
||||
const MAX_FONT_BYTES: u64 = crate::package::MAX_PART;
|
||||
|
||||
/// The font table's references to embedded font files, one per style.
|
||||
const EMBEDS: [&str; 4] = ["w:embedRegular", "w:embedBold", "w:embedItalic", "w:embedBoldItalic"];
|
||||
|
||||
@@ -35,12 +28,6 @@ pub struct EmbeddedFont {
|
||||
/// fonts: that is up to the font database. Empty for a document read from anything but a package
|
||||
/// with a font table.
|
||||
pub fn embedded_fonts(doc: &Document) -> Vec<EmbeddedFont> {
|
||||
fonts_within(doc, MAX_FONT_BYTES)
|
||||
}
|
||||
|
||||
/// [`embedded_fonts`], copying at most `budget` bytes of font files.
|
||||
fn fonts_within(doc: &Document, mut budget: u64) -> Vec<EmbeddedFont> {
|
||||
let budget_total = budget;
|
||||
let Some(table) = doc.passthrough.get(FONT_TABLE_PART).and_then(|b| crate::xml::parse(b).ok()) else {
|
||||
return Vec::new();
|
||||
};
|
||||
@@ -49,9 +36,6 @@ fn fonts_within(doc: &Document, mut budget: u64) -> Vec<EmbeddedFont> {
|
||||
return Vec::new();
|
||||
};
|
||||
let mut out = Vec::new();
|
||||
// A reference repeating one already taken (same file, key and name) gives the same font.
|
||||
let mut taken: HashSet<(&str, [u8; 16], &str)> = HashSet::new();
|
||||
let mut over_budget = 0usize;
|
||||
let embeds = table
|
||||
.children("w:font")
|
||||
.flat_map(|f| f.els().filter(|e| EMBEDS.contains(&e.name.as_str())).map(move |e| (f.attr("w:name").unwrap_or(""), e)));
|
||||
@@ -77,25 +61,12 @@ fn fonts_within(doc: &Document, mut budget: u64) -> Vec<EmbeddedFont> {
|
||||
},
|
||||
None => [0; 16],
|
||||
};
|
||||
if !taken.insert((rel.target.as_str(), key, name)) {
|
||||
continue;
|
||||
}
|
||||
// Checked before the copy is made.
|
||||
let Some(left) = budget.checked_sub(bytes.len() as u64) else {
|
||||
over_budget += 1;
|
||||
continue;
|
||||
};
|
||||
budget = left;
|
||||
let mut data = bytes.to_vec();
|
||||
for (i, b) in data.iter_mut().take(32).enumerate() {
|
||||
*b ^= key.get(i % 16).copied().unwrap_or(0);
|
||||
}
|
||||
out.push(EmbeddedFont { name: name.to_string(), data });
|
||||
}
|
||||
if over_budget > 0 {
|
||||
// Once for the document, not once per font.
|
||||
log::warn!("docx: embedded fonts past the {budget_total} byte budget; {over_budget} of them are not loaded");
|
||||
}
|
||||
out
|
||||
}
|
||||
|
||||
@@ -112,53 +83,3 @@ fn font_key(guid: &str) -> Option<[u8; 16]> {
|
||||
}
|
||||
Some(key)
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use std::sync::Arc;
|
||||
|
||||
use super::*;
|
||||
|
||||
/// A document whose font table makes `refs` references (`(font name, part)`) to font parts
|
||||
/// `parts` (`(name, size)`).
|
||||
fn doc_with(refs: &[(&str, &str)], parts: &[(&str, usize)]) -> Document {
|
||||
let mut table = String::from(
|
||||
r#"<w:fonts xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships">"#,
|
||||
);
|
||||
let mut manifest = format!("P\t{FONT_TABLE_PART}\tapplication/xml\n");
|
||||
for (i, (font, part)) in refs.iter().enumerate() {
|
||||
table.push_str(&format!(r#"<w:font w:name="{font}"><w:embedRegular r:id="rId{i}"/></w:font>"#));
|
||||
manifest.push_str(&format!("R\t{FONT_TABLE_PART}\trId{i}\tfont\t{part}\tI\n"));
|
||||
}
|
||||
table.push_str("</w:fonts>");
|
||||
let mut doc = Document::default();
|
||||
doc.passthrough.insert(FONT_TABLE_PART.into(), Arc::new(table.into_bytes()));
|
||||
doc.passthrough.insert(EMBEDDED_PARTS.into(), Arc::new(manifest.into_bytes()));
|
||||
for (part, size) in parts {
|
||||
doc.passthrough.insert((*part).into(), Arc::new(vec![7u8; *size]));
|
||||
}
|
||||
doc
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_font_referenced_again_is_copied_once() {
|
||||
// Each reference was its own copy: the 64 references allowed to one 256 MB part were
|
||||
// 16 GB of copies (#3692).
|
||||
let refs = vec![("Body", "word/fonts/a.odttf"); MAX_FONTS];
|
||||
let fonts = embedded_fonts(&doc_with(&refs, &[("word/fonts/a.odttf", 1000)]));
|
||||
assert_eq!(fonts.len(), 1);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn font_copies_stop_at_the_byte_budget() {
|
||||
let refs: Vec<(String, &str)> = (0..MAX_FONTS).map(|i| (format!("Font {i}"), "word/fonts/a.odttf")).collect();
|
||||
let refs: Vec<(&str, &str)> = refs.iter().map(|(n, p)| (n.as_str(), *p)).collect();
|
||||
let doc = doc_with(&refs, &[("word/fonts/a.odttf", 1000)]);
|
||||
let (fonts, warned) = crate::test_log::warnings(|| fonts_within(&doc, 3500));
|
||||
assert_eq!(fonts.len(), 3, "three copies fit");
|
||||
let budget: Vec<_> = warned.iter().filter(|w| w.contains("byte budget")).collect();
|
||||
assert_eq!(budget.len(), 1, "one warning for the document, not one per font: {warned:?}");
|
||||
assert!(budget.iter().all(|w| w.contains(&format!("{} of them", MAX_FONTS - 3))), "{budget:?}");
|
||||
assert_eq!(embedded_fonts(&doc).len(), MAX_FONTS, "well within the real budget");
|
||||
}
|
||||
}
|
||||
|
||||
@@ -16,8 +16,6 @@ mod fonts;
|
||||
mod package;
|
||||
mod read;
|
||||
mod smart_art;
|
||||
#[cfg(test)]
|
||||
mod test_log;
|
||||
mod units;
|
||||
mod wordml2003;
|
||||
mod wordml2003_write;
|
||||
@@ -27,8 +25,8 @@ mod xml;
|
||||
pub use bibliography::{MAX_SOURCES, Sources, read_sources};
|
||||
pub use chart_spec::{chart_items, chart_xml};
|
||||
pub use crypt::{
|
||||
DEFAULT_SPIN_COUNT, MAX_PASSWORD_CHARS, PasswordCheck, check_password, check_protection_password, decrypt, encrypt, encrypt_standard_for_tests,
|
||||
encrypt_with_spin_count, is_encrypted, protection_hash,
|
||||
DEFAULT_SPIN_COUNT, MAX_PASSWORD_CHARS, PasswordCheck, check_password, check_protection_password, decrypt, encrypt, encrypt_with_spin_count,
|
||||
is_encrypted, protection_hash,
|
||||
};
|
||||
pub use flat::{XmlKind, sniff_xml, write as write_flat};
|
||||
pub use fonts::{EmbeddedFont, embedded_fonts};
|
||||
|
||||
+12
-134
@@ -1,7 +1,6 @@
|
||||
//! OPC package access: bounded zip reading, relationships, part names.
|
||||
|
||||
use std::collections::hash_map::Entry;
|
||||
use std::collections::{BTreeMap, HashMap, HashSet};
|
||||
use std::collections::{BTreeMap, HashMap};
|
||||
use std::io::{Cursor, Read, Write};
|
||||
use std::sync::Arc;
|
||||
|
||||
@@ -192,9 +191,6 @@ pub fn rel_is(t: &str, full: &str) -> bool {
|
||||
/// picture, an embedded object) is handed over rather than copied.
|
||||
pub struct Package {
|
||||
files: BTreeMap<String, Arc<Vec<u8>>>,
|
||||
/// Lowercased part name → its stored spelling (the first in name order): part names are
|
||||
/// case-insensitive (ECMA-376 Part 2 §6.2.2.3), so every spelling is one part.
|
||||
lower: HashMap<String, String>,
|
||||
}
|
||||
|
||||
impl Package {
|
||||
@@ -227,60 +223,24 @@ impl Package {
|
||||
let room = max_part.min(max_total.saturating_sub(total));
|
||||
let mut buf = Vec::with_capacity(f.size().min(16 * 1024 * 1024) as usize);
|
||||
let read = (&mut f).take(room + 1).read_to_end(&mut buf);
|
||||
// Charged whether or not the read succeeds: an entry inflates in full before its CRC
|
||||
// is checked, so failed entries would otherwise inflate without limit (#3723).
|
||||
total = total.saturating_add(buf.len() as u64);
|
||||
if buf.len() as u64 > room {
|
||||
return Err(DocxError::Limit(format!("part {name} decompresses past the size limit")));
|
||||
}
|
||||
if let Err(e) = read {
|
||||
// A corrupt entry: keep going unless it's a part we can't do without.
|
||||
log::warn!("docx: zip entry {name} unreadable: {e}");
|
||||
continue;
|
||||
}
|
||||
if buf.len() as u64 > room {
|
||||
return Err(DocxError::Limit(format!("part {name} decompresses past the size limit")));
|
||||
}
|
||||
total = total.saturating_add(buf.len() as u64);
|
||||
files.insert(name, Arc::new(buf));
|
||||
}
|
||||
Ok(Package::new(files))
|
||||
Ok(Package { files })
|
||||
}
|
||||
|
||||
/// A package from parts already in memory (a Flat OPC file, a converted document), by part
|
||||
/// name without the leading `/`.
|
||||
pub fn from_parts(files: BTreeMap<String, Vec<u8>>) -> Package {
|
||||
Package::new(files.into_iter().map(|(name, bytes)| (name, Arc::new(bytes))).collect())
|
||||
}
|
||||
|
||||
fn new(mut files: BTreeMap<String, Arc<Vec<u8>>>) -> Package {
|
||||
let mut lower: HashMap<String, String> = HashMap::new();
|
||||
let mut collides = false;
|
||||
for name in files.keys() {
|
||||
match lower.entry(name.to_ascii_lowercase()) {
|
||||
Entry::Occupied(_) => collides = true,
|
||||
Entry::Vacant(v) => {
|
||||
v.insert(name.clone());
|
||||
}
|
||||
}
|
||||
}
|
||||
if !collides {
|
||||
return Package { files, lower };
|
||||
}
|
||||
// Entries whose names differ only in case are one part, so every spelling of the name
|
||||
// reaches the same stored entry (#3704). The spelling kept is the one a relationship
|
||||
// names, so the part the document points at is the one read; the first in byte order only
|
||||
// when no relationship names any of them.
|
||||
let named = rel_targets(&files);
|
||||
for name in files.keys().rev().filter(|n| named.contains(n.as_str())) {
|
||||
lower.insert(name.to_ascii_lowercase(), name.clone());
|
||||
}
|
||||
let before = files.len();
|
||||
files.retain(|name, _| lower.get(&name.to_ascii_lowercase()) == Some(name));
|
||||
let dropped = before.saturating_sub(files.len());
|
||||
log::warn!("docx: {dropped} zip entries are other spellings of parts already read; they are left out");
|
||||
Package { files, lower }
|
||||
}
|
||||
|
||||
/// The stored spelling of part `name` (leading `/` ignored; case-insensitive).
|
||||
pub fn canonical(&self, name: &str) -> Option<&str> {
|
||||
self.lower.get(&name.trim_start_matches('/').to_ascii_lowercase()).map(String::as_str)
|
||||
Package { files: files.into_iter().map(|(name, bytes)| (name, Arc::new(bytes))).collect() }
|
||||
}
|
||||
|
||||
/// Part bytes by name (leading `/` ignored; case-insensitive fallback).
|
||||
@@ -291,15 +251,11 @@ impl Package {
|
||||
/// [`Package::get`], as the shared bytes themselves: for a part the document keeps whole, which
|
||||
/// then costs no copy (a picture document's pictures used to be held twice while it opened).
|
||||
pub fn get_shared(&self, name: &str) -> Option<&Arc<Vec<u8>>> {
|
||||
self.canonical(name).and_then(|n| self.files.get(n))
|
||||
}
|
||||
|
||||
/// [`Package::xml`] for the main document part: malformed XML is an error, not a truncated read.
|
||||
pub fn xml_strict(&self, name: &str) -> Result<Option<El>, DocxError> {
|
||||
match self.get(name) {
|
||||
Some(b) => xml::parse_strict(b).map(Some),
|
||||
None => Ok(None),
|
||||
let n = name.trim_start_matches('/');
|
||||
if let Some(v) = self.files.get(n) {
|
||||
return Some(v);
|
||||
}
|
||||
self.files.iter().find(|(k, _)| k.eq_ignore_ascii_case(n)).map(|(_, v)| v)
|
||||
}
|
||||
|
||||
pub fn xml(&self, name: &str) -> Result<Option<El>, DocxError> {
|
||||
@@ -325,14 +281,7 @@ impl Package {
|
||||
for r in root.els().filter(|e| e.local() == "Relationship") {
|
||||
let (Some(id), Some(target)) = (r.attr("Id"), r.attr("Target")) else { continue };
|
||||
let external = r.attr("TargetMode").is_some_and(|m| m.eq_ignore_ascii_case("External"));
|
||||
// Internal targets name the part as stored, so a part reached under several spellings
|
||||
// is kept, cached and written once (#3704).
|
||||
let resolved = if external {
|
||||
target.to_string()
|
||||
} else {
|
||||
let path = resolve(dir, target);
|
||||
self.canonical(&path).map_or(path, str::to_string)
|
||||
};
|
||||
let resolved = if external { target.to_string() } else { resolve(dir, target) };
|
||||
rels.list.push(Rel { id: id.to_string(), kind: r.attr("Type").unwrap_or("").to_string(), target: resolved, external });
|
||||
}
|
||||
rels
|
||||
@@ -407,24 +356,6 @@ fn split_dir(part: &str) -> (&str, &str) {
|
||||
}
|
||||
}
|
||||
|
||||
/// Every internal part name the relationship parts in `files` target, spelled as they spell it.
|
||||
fn rel_targets(files: &BTreeMap<String, Arc<Vec<u8>>>) -> HashSet<String> {
|
||||
let mut named = HashSet::new();
|
||||
for (name, bytes) in files {
|
||||
// `dir/_rels/file.rels` relates `dir/file`, so its targets resolve against `dir`.
|
||||
let Some(at) = name.rfind("_rels/").filter(|_| name.to_ascii_lowercase().ends_with(".rels")) else { continue };
|
||||
let dir = name.get(..at).unwrap_or("").trim_end_matches('/');
|
||||
let Ok(root) = xml::parse(bytes) else { continue };
|
||||
for r in root.els().filter(|e| e.local() == "Relationship") {
|
||||
let external = r.attr("TargetMode").is_some_and(|m| m.eq_ignore_ascii_case("External"));
|
||||
if let (Some(target), false) = (r.attr("Target"), external) {
|
||||
named.insert(resolve(dir, target));
|
||||
}
|
||||
}
|
||||
}
|
||||
named
|
||||
}
|
||||
|
||||
/// Resolve a relative target against a source directory into a package path.
|
||||
pub fn resolve(dir: &str, target: &str) -> String {
|
||||
let target = target.split('#').next().unwrap_or(target);
|
||||
@@ -481,39 +412,6 @@ mod tests {
|
||||
zip_entries(&e).unwrap()
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn case_colliding_entries_are_one_part() {
|
||||
let z = zip_of(&[("word/media/Image.png", b"upper".to_vec()), ("word/media/image.png", b"lower".to_vec())]);
|
||||
let p = Package::open(&z).unwrap();
|
||||
for spelling in ["word/media/Image.png", "word/media/image.png", "/WORD/MEDIA/IMAGE.PNG"] {
|
||||
assert_eq!(p.canonical(spelling), Some("word/media/Image.png"), "{spelling}");
|
||||
assert_eq!(p.get(spelling), Some(&b"upper"[..]), "{spelling}");
|
||||
}
|
||||
assert_eq!(p.files.len(), 1);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn the_spelling_a_relationship_names_is_kept() {
|
||||
// The first spelling in byte order ("Image.png") was kept even when the document's
|
||||
// relationship named the other one, so the picture shown was not the one it pointed at.
|
||||
let rels = r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="image" Target="media/image.png"/></Relationships>"#;
|
||||
let z = zip_of(&[
|
||||
("word/_rels/document.xml.rels", rels.as_bytes().to_vec()),
|
||||
("word/media/Image.png", b"upper".to_vec()),
|
||||
("word/media/image.png", b"lower".to_vec()),
|
||||
("word/media/Other.png", b"first".to_vec()),
|
||||
("word/media/other.png", b"second".to_vec()),
|
||||
]);
|
||||
let (p, warned) = crate::test_log::warnings(|| Package::open(&z).unwrap());
|
||||
assert_eq!(p.canonical("word/media/IMAGE.png"), Some("word/media/image.png"));
|
||||
assert_eq!(p.get("word/media/Image.png"), Some(&b"lower"[..]));
|
||||
assert_eq!(p.get("word/media/OTHER.PNG"), Some(&b"first"[..]), "no relationship names it: first in byte order");
|
||||
assert_eq!(p.files.len(), 3);
|
||||
let folded: Vec<_> = warned.iter().filter(|w| w.contains("other spellings")).collect();
|
||||
assert_eq!(folded.len(), 1, "one warning for the package: {warned:?}");
|
||||
assert!(folded.iter().all(|w| w.contains("2 zip entries")), "{folded:?}");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn decompression_bomb_is_rejected() {
|
||||
let z = zip_of(&[("word/document.xml", vec![b' '; 2_000_000])]);
|
||||
@@ -523,26 +421,6 @@ mod tests {
|
||||
assert!(Package::open_limited(&z, 10_000_000, 10_000_000).is_ok());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn entries_that_fail_their_check_still_count_against_the_total() {
|
||||
// Each entry inflates in full before its CRC is checked; a failed one was dropped without
|
||||
// being charged, so any number of them inflated (#3723).
|
||||
let names: Vec<String> = (0..20).map(|i| format!("p{i}.bin")).collect();
|
||||
let entries: Vec<(&str, Vec<u8>)> = names.iter().map(|n| (n.as_str(), vec![0u8; 1_000_000])).collect();
|
||||
let mut z = zip_of(&entries);
|
||||
let mut patched = 0;
|
||||
for sig in [[0x50u8, 0x4B, 0x01, 0x02], [0x50, 0x4B, 0x03, 0x04]] {
|
||||
let crc_at = if sig[2] == 1 { 16 } else { 14 };
|
||||
let found: Vec<usize> = z.windows(4).enumerate().filter(|(_, w)| *w == sig).map(|(i, _)| i).collect();
|
||||
for pos in found {
|
||||
z[pos + crc_at] ^= 0xFF;
|
||||
patched += 1;
|
||||
}
|
||||
}
|
||||
assert_eq!(patched, 40);
|
||||
assert!(matches!(Package::open_limited(&z, 2_000_000, 4_000_000), Err(DocxError::Limit(_))));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn lying_declared_size_is_rejected_or_bounded() {
|
||||
let mut z = zip_of(&[("a.xml", b"<a/>".to_vec())]);
|
||||
|
||||
@@ -395,7 +395,7 @@ mod tests {
|
||||
// A hyperlink whose relationship id is written without the prefix: the reader resolves
|
||||
// it (the alias `El::attr` accepts), so the part it names is not carried over twice.
|
||||
let named = package(
|
||||
r#"<w:p><w:hyperlink id="rIdDoc1"><w:r><w:t>link</w:t></w:r></w:hyperlink></w:p>"#,
|
||||
r#"<w:p><w:r><w:hyperlink id="rIdDoc1"><w:r><w:t>link</w:t></w:r></w:hyperlink></w:p>"#,
|
||||
&[("rIdDoc1", "keep.bin")],
|
||||
&[("word/keep.bin", b"kept")],
|
||||
);
|
||||
|
||||
@@ -129,7 +129,7 @@ pub(crate) fn read_package(bytes: &[u8]) -> Result<Document, DocxError> {
|
||||
pub(crate) fn read_opened(pkg: &Package) -> Result<Document, DocxError> {
|
||||
let root_rels = pkg.rels("");
|
||||
let main = root_rels.by_type(rt::OFFICE_DOC).filter(|r| !r.external).map(|r| r.target.clone()).unwrap_or_else(|| "word/document.xml".to_string());
|
||||
let Some(root) = pkg.xml_strict(&main)? else {
|
||||
let Some(root) = pkg.xml(&main)? else {
|
||||
return Err(DocxError::MissingPart(main));
|
||||
};
|
||||
if root.name != "w:document" {
|
||||
@@ -513,9 +513,7 @@ fn toc_control_to_toc(body: &mut Blocks) {
|
||||
let _ = p.insert_object(len, InlineObject::FieldEnd, &props);
|
||||
}
|
||||
}
|
||||
// An empty paragraph opening the control takes the field. It may end a section, as above
|
||||
// entries in columns (#3889).
|
||||
if engine.get(s).and_then(|b| b.as_para()).is_some_and(|p| p.is_empty()) {
|
||||
if engine.get(s).and_then(|b| b.as_para()).is_some_and(|p| p.is_empty() && p.section.is_none()) {
|
||||
field_to_heading(&mut engine, s + 1);
|
||||
} else {
|
||||
toc_field_to_heading(&mut engine);
|
||||
@@ -903,24 +901,6 @@ impl Reader<'_> {
|
||||
{
|
||||
self.pc.alias.insert(id.to_string(), "Normal".into());
|
||||
}
|
||||
// Word in other languages names the TOC styles' ids in that language (`toc 1` is `Sumrio1`
|
||||
// in Portuguese): they take the engine's ids, which Update Table and the writer's TOC
|
||||
// recognise the entries by (#3889). Not when the file or an earlier style has that id.
|
||||
let mut taken: Vec<String> = root.children("w:style").take(10_000).filter_map(|s| s.attr("w:styleId")).map(str::to_string).collect();
|
||||
for s in root.children("w:style").take(10_000).filter(|s| s.attr("w:type").is_none_or(|t| t == "paragraph")) {
|
||||
let (Some(id), Some(name)) = (s.attr("w:styleId"), s.child_val("w:name")) else { continue };
|
||||
let canonical = match name.trim().to_ascii_lowercase().as_str() {
|
||||
"toc heading" => "TOCHeading".to_string(),
|
||||
n => match n.strip_prefix("toc ").and_then(|d| d.parse::<u8>().ok()).filter(|d| (1..=9).contains(d)) {
|
||||
Some(d) => format!("TOC{d}"),
|
||||
None => continue,
|
||||
},
|
||||
};
|
||||
if !taken.contains(&canonical) && !self.pc.alias.contains_key(id) {
|
||||
taken.push(canonical.clone());
|
||||
self.pc.alias.insert(id.to_string(), canonical);
|
||||
}
|
||||
}
|
||||
let mut sheet = StyleSheet::empty();
|
||||
sheet.default_para_present = root.child("w:docDefaults").and_then(|dd| dd.child("w:pPrDefault")).is_some();
|
||||
if let Some(dd) = root.child("w:docDefaults") {
|
||||
@@ -1212,7 +1192,6 @@ impl Reader<'_> {
|
||||
s.compat_mode = m.clamp(11, 99);
|
||||
}
|
||||
s.sum_paragraph_spacing = k.children("w:doNotUseHTMLParagraphAutoSpacing").any(on_off);
|
||||
s.do_not_expand_shift_return = k.children("w:doNotExpandShiftReturn").any(on_off);
|
||||
}
|
||||
"w:trackRevisions" => s.track_changes = on_off(k),
|
||||
"w:doNotTrackMoves" => s.track_moves = !on_off(k),
|
||||
@@ -1363,10 +1342,6 @@ impl Reader<'_> {
|
||||
let mut by_para: HashMap<String, u32> = HashMap::new();
|
||||
let mut next_id = 0u32;
|
||||
let mut entries: BTreeMap<u32, Comment> = BTreeMap::new();
|
||||
// Ids given before this part, looked up rather than scanned for every generated id: a
|
||||
// scan per comment made many non-numeric ids quadratic (#3692). Ids given here are in
|
||||
// `entries`.
|
||||
let given: HashSet<u32> = self.comment_map.values().copied().collect();
|
||||
for c in root.children("w:comment") {
|
||||
if self.doc.parts.len() >= MAX_PARTS {
|
||||
break;
|
||||
@@ -1378,7 +1353,7 @@ impl Reader<'_> {
|
||||
let id = match file_id.parse::<u32>() {
|
||||
Ok(v) if !entries.contains_key(&v) => v,
|
||||
_ => {
|
||||
while entries.contains_key(&next_id) || given.contains(&next_id) {
|
||||
while entries.contains_key(&next_id) || self.comment_map.values().any(|v| *v == next_id) {
|
||||
next_id += 1;
|
||||
}
|
||||
next_id
|
||||
|
||||
@@ -231,11 +231,22 @@ impl PropCtx {
|
||||
// `w:ascii`, `w:hAnsiTheme` over `w:hAnsi`).
|
||||
let named = |a: &str| k.attr(a).filter(|f| !f.is_empty()).map(str::to_string);
|
||||
let themed = |a: &str| k.attr(a).and_then(|t| self.theme_font(t));
|
||||
// `w:eastAsia` and `w:cs` name fonts for other scripts and stay in `font_ea` /
|
||||
// `font_cs`: `w:hint` only picks between them for ambiguous characters, never for
|
||||
// Latin letters. A Normal style with only `w:eastAsia`/`w:cs` keeps the
|
||||
// `docDefaults` font, and so does a hinted run, as in Word.
|
||||
c.font = themed("w:asciiTheme").or_else(|| named("w:ascii")).or_else(|| themed("w:hAnsiTheme")).or_else(|| named("w:hAnsi"));
|
||||
c.font =
|
||||
themed("w:asciiTheme").or_else(|| named("w:ascii")).or_else(|| themed("w:hAnsiTheme")).or_else(|| named("w:hAnsi")).or_else(
|
||||
|| {
|
||||
// `w:eastAsia` and `w:cs` name fonts for other scripts; they stand in for the
|
||||
// Latin font only when the run is hinted as that script. Otherwise the Latin
|
||||
// font is inherited (a Normal style with only `w:eastAsia`/`w:cs` keeps the
|
||||
// `docDefaults` font, as in Word).
|
||||
match k.attr("w:hint") {
|
||||
Some("eastAsia") => k.attr("w:eastAsia"),
|
||||
Some("cs") => k.attr("w:cs"),
|
||||
_ => None,
|
||||
}
|
||||
.filter(|f| !f.is_empty())
|
||||
.map(str::to_string)
|
||||
},
|
||||
);
|
||||
// The complex-script font (Persian, Arabic, Hebrew text). A `w:cstheme` font is
|
||||
// left to inherit: the theme's per-script fonts aren't read.
|
||||
c.font_cs = k.attr("w:cs").filter(|f| !f.is_empty() && f.len() < 256).map(str::to_string);
|
||||
|
||||
@@ -41,10 +41,9 @@ pub(crate) fn preset_kind(prst: &str) -> ShapeKind {
|
||||
"straightConnector1" => ShapeKind::StraightConnector,
|
||||
"bentConnector2" | "bentConnector3" | "bentConnector4" | "bentConnector5" => ShapeKind::ElbowConnector,
|
||||
"curvedConnector2" | "curvedConnector3" | "curvedConnector4" | "curvedConnector5" => ShapeKind::CurvedConnector,
|
||||
"rightArrow" | "leftRightArrow" | "upDownArrow" | "notchedRightArrow" | "stripedRightArrow" => ShapeKind::Arrow,
|
||||
"leftArrow" => ShapeKind::ArrowLeft,
|
||||
"upArrow" => ShapeKind::ArrowUp,
|
||||
"downArrow" => ShapeKind::ArrowDown,
|
||||
"rightArrow" | "leftArrow" | "upArrow" | "downArrow" | "leftRightArrow" | "upDownArrow" | "notchedRightArrow" | "stripedRightArrow" => {
|
||||
ShapeKind::Arrow
|
||||
}
|
||||
"star4" | "star5" | "star6" | "star7" | "star8" | "star10" | "star12" | "star16" | "star24" | "star32" => ShapeKind::Star,
|
||||
"heart" => ShapeKind::Heart,
|
||||
_ => ShapeKind::Rectangle,
|
||||
@@ -470,15 +469,6 @@ mod tests {
|
||||
assert_eq!(preset_kind("flowChartDecision"), ShapeKind::Diamond);
|
||||
}
|
||||
|
||||
/// #3919: Word's up, down and left block arrows were all read as the right-pointing arrow.
|
||||
#[test]
|
||||
fn block_arrows_keep_their_direction() {
|
||||
assert_eq!(preset_kind("rightArrow"), ShapeKind::Arrow);
|
||||
assert_eq!(preset_kind("leftArrow"), ShapeKind::ArrowLeft);
|
||||
assert_eq!(preset_kind("upArrow"), ShapeKind::ArrowUp);
|
||||
assert_eq!(preset_kind("downArrow"), ShapeKind::ArrowDown);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn vml_presets_map_like_drawingml_presets() {
|
||||
assert_eq!(vml_preset_kind(13), Some(ShapeKind::Arrow));
|
||||
@@ -487,7 +477,7 @@ mod tests {
|
||||
assert_eq!(vml_preset_kind(202), Some(ShapeKind::TextBox));
|
||||
// The nearest shape, as `preset_kind` gives it: left arrow, right triangle, four-point
|
||||
// star, rounded callout, elbow connector, a bevel (its box).
|
||||
assert_eq!(vml_preset_kind(66), Some(ShapeKind::ArrowLeft));
|
||||
assert_eq!(vml_preset_kind(66), Some(ShapeKind::Arrow));
|
||||
assert_eq!(vml_preset_kind(6), Some(ShapeKind::RightTriangle));
|
||||
assert_eq!(vml_preset_kind(187), Some(ShapeKind::Star));
|
||||
assert_eq!(vml_preset_kind(62), Some(ShapeKind::RoundedRectangle));
|
||||
|
||||
@@ -724,28 +724,22 @@ impl Reader<'_> {
|
||||
if sc.fields.iter().any(|f| f.phase == Phase::Instr) {
|
||||
return;
|
||||
}
|
||||
let ty = k.attr("w:type").map(|t| t.trim().to_ascii_lowercase());
|
||||
let c = match ty.as_deref() {
|
||||
let c = match k.attr("w:type") {
|
||||
Some("page") => "\u{000C}",
|
||||
Some("column") => "\u{000E}",
|
||||
_ => "\n",
|
||||
};
|
||||
self.emit_text(sc, pb, c, props);
|
||||
}
|
||||
"w:cr" if !sc.fields.iter().any(|f| f.phase == Phase::Instr) => {
|
||||
self.emit_text(sc, pb, "\n", props);
|
||||
"w:cr" => {
|
||||
if !sc.fields.iter().any(|f| f.phase == Phase::Instr) {
|
||||
self.emit_text(sc, pb, "\n", props);
|
||||
}
|
||||
}
|
||||
"w:cr" => {}
|
||||
"w:noBreakHyphen" => self.emit_text(sc, pb, "\u{2011}", props),
|
||||
"w:softHyphen" => self.emit_text(sc, pb, "\u{00AD}", props),
|
||||
"w:sym" => {
|
||||
let h_str = k.attr("w:char").map(|h| {
|
||||
let s = h.trim();
|
||||
let s = s.strip_prefix("0x").or_else(|| s.strip_prefix("0X")).unwrap_or(s);
|
||||
let s = s.strip_prefix("U+").or_else(|| s.strip_prefix("u+")).unwrap_or(s);
|
||||
s.strip_prefix("#x").or_else(|| s.strip_prefix("#X")).unwrap_or(s)
|
||||
});
|
||||
let ch = h_str.and_then(|h| u32::from_str_radix(h, 16).ok()).and_then(char::from_u32).filter(|c| *c as u32 >= 0x20);
|
||||
let ch = k.attr("w:char").and_then(|h| u32::from_str_radix(h.trim(), 16).ok()).and_then(char::from_u32).filter(|c| *c as u32 >= 0x20);
|
||||
if let Some(ch) = ch {
|
||||
let mut p = props.clone();
|
||||
// The symbol's own font, kept apart from the run's (ECMA-376 §17.3.3.30).
|
||||
@@ -789,7 +783,7 @@ impl Reader<'_> {
|
||||
}
|
||||
}
|
||||
"w:footnoteReference" | "w:endnoteReference" => {
|
||||
let foot = k.local() == "footnoteReference";
|
||||
let foot = k.name == "w:footnoteReference";
|
||||
let fid = k.attr("w:id").and_then(int).unwrap_or(i64::MIN);
|
||||
let map = if foot { &self.footnotes } else { &self.endnotes };
|
||||
if let Some(id) = map.get(&fid).copied() {
|
||||
@@ -800,7 +794,7 @@ impl Reader<'_> {
|
||||
}
|
||||
// The note's own number at the start of its text: a reference to the note being read.
|
||||
"w:footnoteRef" | "w:endnoteRef" => {
|
||||
let kind = if k.local() == "footnoteRef" { NoteKind::Footnote } else { NoteKind::Endnote };
|
||||
let kind = if k.name == "w:footnoteRef" { NoteKind::Footnote } else { NoteKind::Endnote };
|
||||
if let Some((nk, id)) = self.current_note
|
||||
&& nk == kind
|
||||
{
|
||||
@@ -1072,10 +1066,6 @@ impl Reader<'_> {
|
||||
|
||||
// ---- drawings ----
|
||||
|
||||
/// Read a `w:drawing` frame and dispatch its payload to the picture, shape or graphic reader.
|
||||
/// Frame dimensions are converted from EMUs to bounded point values; placement, alternative
|
||||
/// text and drawing ids are carried into the object. Returns `None` when the frame or payload
|
||||
/// cannot be read. Ink written by WordCraft keeps its tool's caps rather than an outline cap.
|
||||
fn read_drawing(&mut self, sc: &mut StoryCtx, d: &El, rels: &Rels) -> Option<InlineObject> {
|
||||
let (c, anchored) = match d.child("wp:inline") {
|
||||
Some(c) => (c, false),
|
||||
@@ -1152,16 +1142,13 @@ impl Reader<'_> {
|
||||
let wsp = direct_shape.or_else(|| gd.find("wps:wsp"))?;
|
||||
let mut obj = self.read_wsp(sc, wsp, rels, w, h, float);
|
||||
// A freeform: whether it's ink WordCraft wrote.
|
||||
if let InlineObject::Shape { freeform: Some(f), extra, .. } = &mut obj
|
||||
if let InlineObject::Shape { freeform: Some(f), .. } = &mut obj
|
||||
&& let Some(name) = c.child("wp:docPr").and_then(|p| p.attr("name"))
|
||||
&& let Some(tool) = super::freeform::ink_tool(name)
|
||||
{
|
||||
let f = Arc::make_mut(f);
|
||||
f.ink = Some(tool);
|
||||
f.order = super::freeform::ink_order(name);
|
||||
// Ink's round ends belong to its tool, not the optional shape outline. Keep
|
||||
// its canonical model unchanged when reading the round cap the writer adds.
|
||||
extra.cap = Default::default();
|
||||
}
|
||||
obj
|
||||
};
|
||||
@@ -1474,10 +1461,6 @@ impl Reader<'_> {
|
||||
}
|
||||
}
|
||||
|
||||
/// Build a shape from `wps:wsp`, using frame dimensions `w` and `h` in points and the caller's
|
||||
/// placement. Reads geometry, paint, effects and connector ends, and creates a text-box story
|
||||
/// when present within the nesting limit. Explicit outline caps are preserved; missing or
|
||||
/// unknown caps use flat ends, without resolving theme-inherited caps.
|
||||
fn read_wsp(&mut self, sc: &mut StoryCtx, wsp: &El, rels: &Rels, w: f32, h: f32, mut float: Float) -> InlineObject {
|
||||
let sppr = wsp.child("wps:spPr");
|
||||
float.set_spin(xfrm_spin(sppr));
|
||||
@@ -1550,12 +1533,7 @@ impl Reader<'_> {
|
||||
{
|
||||
part.body.auto_color = Some(rgb);
|
||||
}
|
||||
let mut extra = ShapeExtra {
|
||||
dash: super::shape_style::line_dash(ln),
|
||||
cap: wordcraft_doc::connector::LineCap::from_ooxml(ln.and_then(|e| e.attr("cap")).unwrap_or("flat")),
|
||||
outline_fill: paint.outline_fill,
|
||||
..ShapeExtra::default()
|
||||
};
|
||||
let mut extra = ShapeExtra { dash: super::shape_style::line_dash(ln), outline_fill: paint.outline_fill, ..ShapeExtra::default() };
|
||||
// A rounded rectangle's own corner (`adj`, ECMA-376 §20.1.10.56); other presets drawn as
|
||||
// rounded rectangles keep the default.
|
||||
if prst == "roundRect" {
|
||||
@@ -1750,15 +1728,6 @@ impl Reader<'_> {
|
||||
let stroke = on("stroked").then(|| attr("strokecolor").map_or(Some(Rgb::BLACK), vml_color)).flatten();
|
||||
let stroke_width = attr("strokeweight").and_then(|v| measure(v, 1.0)).unwrap_or(0.75).clamp(0.0, 100.0);
|
||||
let fill = vml_fill(shape, shapetype);
|
||||
let endcap = shape
|
||||
.child("v:stroke")
|
||||
.and_then(|s| s.attr("endcap"))
|
||||
.or_else(|| shapetype.and_then(|t| t.child("v:stroke")).and_then(|s| s.attr("endcap")));
|
||||
let cap = match endcap {
|
||||
Some("square") => wordcraft_doc::connector::LineCap::Square,
|
||||
Some("round") => wordcraft_doc::connector::LineCap::Round,
|
||||
_ => wordcraft_doc::connector::LineCap::Flat,
|
||||
};
|
||||
let kind = match shape.local() {
|
||||
"line" => Some(ShapeKind::Line),
|
||||
"rect" => Some(ShapeKind::Rectangle),
|
||||
@@ -1782,7 +1751,7 @@ impl Reader<'_> {
|
||||
let font = tp
|
||||
.attr("style")
|
||||
.and_then(|s| vml_style(s, "font-family"))
|
||||
.map(|f| css_string(f).chars().take(100).collect::<String>())
|
||||
.map(|f| f.trim_matches(|c: char| c == '"' || c == '\'' || c.is_whitespace()).chars().take(100).collect::<String>())
|
||||
.filter(|f| !f.is_empty());
|
||||
// The shapetype's WordArt shape (`o:spt` 136-175) bends the text like DrawingML's warp.
|
||||
let warp = vml_spt(shape, shapetype, type_id)
|
||||
@@ -1861,7 +1830,7 @@ impl Reader<'_> {
|
||||
story,
|
||||
freeform: None,
|
||||
effects: Default::default(),
|
||||
extra: ShapeExtra { cap, ..Default::default() },
|
||||
extra: Default::default(),
|
||||
});
|
||||
}
|
||||
// A freeform: its `path`, in its `coordsize` space; a `v:fill` with no opacity
|
||||
@@ -1890,11 +1859,11 @@ impl Reader<'_> {
|
||||
story: None,
|
||||
freeform: Some(Arc::new(f)),
|
||||
effects: Default::default(),
|
||||
extra: ShapeExtra { cap, ..Default::default() },
|
||||
extra: Default::default(),
|
||||
});
|
||||
}
|
||||
let kind = kind?;
|
||||
let mut extra = ShapeExtra { cap, ..Default::default() };
|
||||
let mut extra = ShapeExtra::default();
|
||||
if kind == ShapeKind::Line {
|
||||
// Arrowheads (`v:stroke` startarrow / endarrow): any but `none` is drawn.
|
||||
let arrow = |name: &str| shape.child("v:stroke").and_then(|s| s.attr(name)).is_some_and(|v| v != "none");
|
||||
@@ -2262,49 +2231,6 @@ fn vml_style<'a>(style: &'a str, key: &str) -> Option<&'a str> {
|
||||
style.rsplit(';').filter_map(|d| d.split_once(':')).find(|(k, _)| k.trim().eq_ignore_ascii_case(key)).map(|(_, v)| v.trim())
|
||||
}
|
||||
|
||||
/// A `v:textpath` style value with one pair of surrounding quotes removed and the writer's CSS
|
||||
/// escapes mapped back: `\` + up to six hex digits + one optional space is that character, and
|
||||
/// `\x` is `x` (#3994; `write::watermark::run` writes them). Only the surrounding pair goes, so a
|
||||
/// font name that starts or ends with an apostrophe keeps it. Word writes font names raw, and a
|
||||
/// lone trailing `\` none of CSS would accept stays literal.
|
||||
fn css_string(v: &str) -> String {
|
||||
let v = v.trim();
|
||||
let v = ['"', '\''].into_iter().find_map(|q| v.strip_prefix(q)?.strip_suffix(q)).unwrap_or(v).trim();
|
||||
let mut out = String::with_capacity(v.len());
|
||||
let mut chars = v.chars();
|
||||
while let Some(c) = chars.next() {
|
||||
if c != '\\' {
|
||||
out.push(c);
|
||||
continue;
|
||||
}
|
||||
let mut hex = String::new();
|
||||
while hex.len() < 6 {
|
||||
match chars.clone().next() {
|
||||
Some(h) if h.is_ascii_hexdigit() => {
|
||||
hex.push(h);
|
||||
chars.next();
|
||||
}
|
||||
_ => break,
|
||||
}
|
||||
}
|
||||
if hex.is_empty() {
|
||||
match chars.next() {
|
||||
Some(e) => out.push(e),
|
||||
None => out.push('\\'),
|
||||
}
|
||||
} else {
|
||||
if let Some(d) = u32::from_str_radix(&hex, 16).ok().and_then(char::from_u32) {
|
||||
out.push(d);
|
||||
}
|
||||
// One whitespace after the escape is its terminator, not part of the value.
|
||||
if chars.clone().next().is_some_and(char::is_whitespace) {
|
||||
chars.next();
|
||||
}
|
||||
}
|
||||
}
|
||||
out
|
||||
}
|
||||
|
||||
/// Placement of a VML shape (ECMA-376 Part 4, VML): `position:absolute` in its style makes it
|
||||
/// float, offset by `left`/`margin-left` and `top`/`margin-top` from the area named by
|
||||
/// `mso-position-horizontal-relative` / `mso-position-vertical-relative` (unless aligned with
|
||||
@@ -2406,7 +2332,7 @@ fn vml_watermark(shape: &El) -> Option<wordcraft_doc::Watermark> {
|
||||
let font = tp
|
||||
.attr("style")
|
||||
.and_then(|s| vml_style(s, "font-family"))
|
||||
.map(|f| css_string(f).chars().take(100).collect::<String>())
|
||||
.map(|f| f.trim_matches(|c: char| c == '"' || c == '\'' || c.is_whitespace()).chars().take(100).collect::<String>())
|
||||
.filter(|f| !f.is_empty())
|
||||
.unwrap_or(default.font);
|
||||
let color = shape.attr("fillcolor").and_then(vml_color).unwrap_or(default.color);
|
||||
|
||||
@@ -1,36 +0,0 @@
|
||||
//! Test helper: the warnings this crate logs, captured per test thread, so a test can check that
|
||||
//! a limit which leaves something out says so.
|
||||
|
||||
use std::cell::RefCell;
|
||||
|
||||
struct Capture;
|
||||
|
||||
thread_local! {
|
||||
static LINES: RefCell<Vec<String>> = const { RefCell::new(Vec::new()) };
|
||||
}
|
||||
|
||||
impl log::Log for Capture {
|
||||
fn enabled(&self, m: &log::Metadata) -> bool {
|
||||
m.level() <= log::Level::Warn
|
||||
}
|
||||
|
||||
fn log(&self, r: &log::Record) {
|
||||
if self.enabled(r.metadata()) {
|
||||
LINES.with(|l| l.borrow_mut().push(r.args().to_string()));
|
||||
}
|
||||
}
|
||||
|
||||
fn flush(&self) {}
|
||||
}
|
||||
|
||||
static CAPTURE: Capture = Capture;
|
||||
|
||||
/// `f`'s result and the warnings logged on this thread while it ran.
|
||||
pub(crate) fn warnings<T>(f: impl FnOnce() -> T) -> (T, Vec<String>) {
|
||||
// Only this logger is ever installed in a test binary; a second call keeps it.
|
||||
let _ = log::set_logger(&CAPTURE);
|
||||
log::set_max_level(log::LevelFilter::Warn);
|
||||
LINES.with(|l| l.borrow_mut().clear());
|
||||
let out = f();
|
||||
(out, LINES.with(|l| std::mem::take(&mut *l.borrow_mut())))
|
||||
}
|
||||
@@ -48,10 +48,6 @@ pub(crate) struct EmbedWriter<'d> {
|
||||
pub parts: Vec<(String, Vec<u8>, String, PartRels)>,
|
||||
/// Leave out the main document's kept glossary (it's written from the model instead).
|
||||
pub drop_glossary: bool,
|
||||
/// Kept parts written with these bytes instead of their own (see [`Self::scrub_glossary`]).
|
||||
scrubbed: HashMap<String, Vec<u8>>,
|
||||
/// A chart past the chart limits has been written (and the loss logged, once per save).
|
||||
chart_cut: bool,
|
||||
}
|
||||
|
||||
impl<'d> EmbedWriter<'d> {
|
||||
@@ -72,7 +68,7 @@ impl<'d> EmbedWriter<'d> {
|
||||
"[content_types].xml",
|
||||
]
|
||||
.iter()
|
||||
.map(ToString::to_string)
|
||||
.map(|s| s.to_string())
|
||||
.collect();
|
||||
taken.insert(VBA_PROJECT_PART.to_ascii_lowercase());
|
||||
// The bibliography writer uses these names when sources are present (#532).
|
||||
@@ -97,8 +93,6 @@ impl<'d> EmbedWriter<'d> {
|
||||
markup_bytes: MAX_MARKUP_BYTES,
|
||||
parts: Vec::new(),
|
||||
drop_glossary: false,
|
||||
scrubbed: HashMap::new(),
|
||||
chart_cut: false,
|
||||
}
|
||||
}
|
||||
|
||||
@@ -236,44 +230,6 @@ impl<'d> EmbedWriter<'d> {
|
||||
})
|
||||
}
|
||||
|
||||
/// Have the main document's kept glossary, and the parts it reaches, written without the
|
||||
/// personal information (`names`) and dates the privacy settings remove (#3692). False when a
|
||||
/// part's markup can't be rewritten: the glossary is then written from the model.
|
||||
pub fn scrub_glossary(&mut self, names: bool, dates: bool) -> bool {
|
||||
let Some(roots) = self.passthrough.get(crate::package::UNMODELLED_ROOTS) else { return true };
|
||||
let mut todo: Vec<String> = String::from_utf8_lossy(roots)
|
||||
.lines()
|
||||
.take(MAX_EMBEDDED_RELS * 2)
|
||||
.filter_map(|line| match line.split('\t').collect::<Vec<_>>().as_slice() {
|
||||
[source, kind, target] if *source == "D" && rel_is(kind, rt::GLOSSARY) => Some(target.to_string()),
|
||||
_ => None,
|
||||
})
|
||||
.collect();
|
||||
let mut seen: HashSet<String> = HashSet::new();
|
||||
while let Some(p) = todo.pop() {
|
||||
if seen.len() >= MAX_EMBEDDED_PARTS || !seen.insert(p.clone()) {
|
||||
continue;
|
||||
}
|
||||
let (Some(info), Some(bytes)) = (self.manifest.parts.get(&p), self.passthrough.get(&p)) else { continue };
|
||||
todo.extend(info.rels.iter().filter(|r| !r.external).map(|r| r.target.clone()));
|
||||
if !(info.content_type.ends_with("xml") || p.to_ascii_lowercase().ends_with(".xml")) {
|
||||
continue;
|
||||
}
|
||||
match scrub_private(bytes, names, dates) {
|
||||
Some(b) => {
|
||||
self.scrubbed.insert(p, b);
|
||||
}
|
||||
None => {
|
||||
log::warn!(
|
||||
"docx: the glossary's part {p} can't be cleaned of personal information; the glossary is written from its entries, without what WordCraft doesn't model"
|
||||
);
|
||||
return false;
|
||||
}
|
||||
}
|
||||
}
|
||||
true
|
||||
}
|
||||
|
||||
/// What copying the parts reached from `roots` costs against the copy budget.
|
||||
fn tree_cost(&self, roots: &[&str]) -> u64 {
|
||||
let mut seen: HashSet<&str> = HashSet::new();
|
||||
@@ -327,9 +283,8 @@ impl<'d> EmbedWriter<'d> {
|
||||
rels.add_with_id(&r.id, &r.kind, &relative(&new, &target), false);
|
||||
}
|
||||
let ct = if info.content_type.is_empty() { "application/octet-stream".to_string() } else { info.content_type };
|
||||
let bytes = self.scrubbed.get(&orig).cloned().unwrap_or_else(|| bytes.to_vec());
|
||||
written.push(self.parts.len());
|
||||
self.parts.push((new, bytes, ct, rels));
|
||||
self.parts.push((new, bytes.to_vec(), ct, rels));
|
||||
}
|
||||
Some(name)
|
||||
}
|
||||
@@ -453,12 +408,7 @@ impl super::Writer<'_> {
|
||||
/// related from the story part through `rels`.
|
||||
pub(super) fn chart_graphic(&mut self, spec: &wordcraft_doc::chart::ChartSpec, rels: &mut PartRels) -> String {
|
||||
let name = self.embeds.fresh("word/charts/chart", ".xml");
|
||||
let (xml, cut) = crate::chart_spec::chart_part(spec);
|
||||
if cut && !self.embeds.chart_cut {
|
||||
self.embeds.chart_cut = true;
|
||||
log::warn!("docx: a chart has more series or points than a chart part holds; the rest are left out of the file");
|
||||
}
|
||||
let bytes = xml.into_bytes();
|
||||
let bytes = crate::chart_spec::chart_xml(spec).into_bytes();
|
||||
self.embeds.parts.push((name.clone(), bytes, crate::chart_spec::CHART_CT.to_string(), PartRels::default()));
|
||||
let id = rels.add(crate::package::rt::CHART, &relative(STORY_PART, &name), false);
|
||||
let mut w = xml::W::default();
|
||||
@@ -690,80 +640,6 @@ fn patch_rel_ids(bytes: &[u8], ids: &HashMap<String, String>) -> Vec<u8> {
|
||||
out
|
||||
}
|
||||
|
||||
/// `bytes` (kept XML) without the personal information and dates the privacy settings remove, as
|
||||
/// the writer leaves them out of the parts it writes: annotation authors become "Author" and
|
||||
/// comment initials "A" (`names`); annotation dates and UTC dates go (`dates`); and with `names`,
|
||||
/// a people part's authors become "Author" and their presence information goes. Attributes are
|
||||
/// matched by namespace, so any prefix is caught. `None` when the markup can't be rewritten
|
||||
/// (malformed, or not UTF-8).
|
||||
fn scrub_private(bytes: &[u8], names: bool, dates: bool) -> Option<Vec<u8>> {
|
||||
use std::borrow::Cow;
|
||||
|
||||
use quick_xml::events::Event;
|
||||
use quick_xml::events::attributes::Attribute;
|
||||
use quick_xml::name::{QName, ResolveResult};
|
||||
use quick_xml::reader::NsReader;
|
||||
use quick_xml::writer::Writer;
|
||||
|
||||
const WML: &[u8] = b"http://schemas.openxmlformats.org/wordprocessingml/2006/main";
|
||||
const WML_STRICT: &[u8] = b"http://purl.oclc.org/ooxml/wordprocessingml/main";
|
||||
const W15: &[u8] = b"http://schemas.microsoft.com/office/word/2012/wordml";
|
||||
// UTF-16 (or other wide encodings) would pass through unread.
|
||||
if bytes.contains(&0) || bytes.starts_with(&[0xFF, 0xFE]) || bytes.starts_with(&[0xFE, 0xFF]) {
|
||||
return None;
|
||||
}
|
||||
let ns_is = |r: &ResolveResult, ns: &[u8]| matches!(r, ResolveResult::Bound(n) if n.as_ref() == ns);
|
||||
let wml = |r: &ResolveResult| ns_is(r, WML) || ns_is(r, WML_STRICT);
|
||||
let mut r = NsReader::from_reader(bytes);
|
||||
r.config_mut().check_end_names = false;
|
||||
let mut out = Writer::new(Vec::with_capacity(bytes.len()));
|
||||
let mut buf = Vec::new();
|
||||
let mut skip = Vec::new();
|
||||
loop {
|
||||
buf.clear();
|
||||
let (res, ev) = r.read_resolved_event_into(&mut buf).ok()?;
|
||||
let (in_wml, in_w15) = (wml(&res), ns_is(&res, W15));
|
||||
let ev = match ev {
|
||||
Event::Eof => break,
|
||||
Event::Start(ref e) | Event::Empty(ref e) if names && in_w15 && e.local_name().as_ref() == b"presenceInfo" => {
|
||||
if matches!(ev, Event::Start(_)) {
|
||||
let end = e.name().as_ref().to_vec();
|
||||
r.read_to_end_into(QName(&end), &mut skip).ok()?;
|
||||
skip.clear();
|
||||
}
|
||||
continue;
|
||||
}
|
||||
Event::Start(ref e) | Event::Empty(ref e) => {
|
||||
let mut el = e.to_owned();
|
||||
el.clear_attributes();
|
||||
for a in e.attributes().with_checks(false) {
|
||||
let a = a.ok()?;
|
||||
let (ares, local) = r.resolver_mut().resolve_attribute(a.key);
|
||||
// An unprefixed attribute of a WordprocessingML element is read as its own.
|
||||
let wml_attr = wml(&ares) || (matches!(ares, ResolveResult::Unbound) && in_wml);
|
||||
let value: Option<&[u8]> = match local.as_ref() {
|
||||
b"author" if names && (wml_attr || ns_is(&ares, W15)) => Some(super::ANONYMOUS_AUTHOR.as_bytes()),
|
||||
b"initials" if names && wml_attr => Some(super::ANONYMOUS_INITIALS.as_bytes()),
|
||||
b"date" if dates && wml_attr => None,
|
||||
b"dateUtc" if dates => None,
|
||||
_ => {
|
||||
el.push_attribute(a);
|
||||
continue;
|
||||
}
|
||||
};
|
||||
if let Some(v) = value {
|
||||
el.push_attribute(Attribute { key: a.key, value: Cow::Borrowed(v) });
|
||||
}
|
||||
}
|
||||
if matches!(ev, Event::Start(_)) { Event::Start(el) } else { Event::Empty(el) }
|
||||
}
|
||||
ev => ev,
|
||||
};
|
||||
out.write_event(ev).ok()?;
|
||||
}
|
||||
Some(out.into_inner())
|
||||
}
|
||||
|
||||
/// A VML `style` with the object's size, rotation and flips, and its offset when it floats
|
||||
/// (`position:absolute`).
|
||||
fn vml_style_with(style: &str, w: f32, h: f32, float: &Float) -> String {
|
||||
@@ -833,40 +709,6 @@ fn valid_part_name(name: &str) -> String {
|
||||
mod tests {
|
||||
use super::*;
|
||||
|
||||
#[test]
|
||||
fn charts_cut_to_the_limits_on_save_say_so_once() {
|
||||
// A model deserialised as is can hold more than a chart part keeps; the save left the rest
|
||||
// out without a word (review of #3872).
|
||||
use std::sync::Arc;
|
||||
use wordcraft_doc::chart::{ChartSeries, ChartSpec, ChartType, MAX_SERIES};
|
||||
use wordcraft_doc::graphic::{Graphic, GraphicKind};
|
||||
use wordcraft_doc::{CharProps, InlineObject, Pos};
|
||||
let chart = |series: usize| {
|
||||
let spec = ChartSpec {
|
||||
kind: ChartType::Column,
|
||||
categories: vec!["a".into(), "b".into()],
|
||||
series: (0..series).map(|i| ChartSeries { name: format!("S{i}"), values: vec![Some(1.0); 2] }).collect(),
|
||||
title: None,
|
||||
legend: None,
|
||||
data_labels: false,
|
||||
};
|
||||
let graphic = Graphic { kind: GraphicKind::Chart, w: 100.0, h: 100.0, chart: Some(Arc::new(spec)), ..Default::default() };
|
||||
InlineObject::Graphic { w: 100.0, h: 100.0, alt: String::new(), float: Default::default(), graphic: Arc::new(graphic) }
|
||||
};
|
||||
let save = |charts: Vec<InlineObject>| {
|
||||
let mut doc = wordcraft_doc::Document::new();
|
||||
for c in charts {
|
||||
doc.insert_object(&Pos::body(0, 0), c, &CharProps::default()).unwrap();
|
||||
}
|
||||
crate::test_log::warnings(|| crate::write(&doc).unwrap()).1
|
||||
};
|
||||
let cut = |w: &Vec<String>| w.iter().filter(|w| w.contains("chart")).count();
|
||||
let warned = save(vec![chart(MAX_SERIES + 1), chart(MAX_SERIES + 5), chart(2)]);
|
||||
assert_eq!(cut(&warned), 1, "one warning for the save, not one per chart: {warned:?}");
|
||||
let warned = save(vec![chart(3)]);
|
||||
assert_eq!(cut(&warned), 0, "nothing left out: {warned:?}");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn data_part_drawing_ids_are_rewritten_once() {
|
||||
let ids: HashMap<String, String> = [("rId9".to_string(), "rId2".to_string()), ("rId2".to_string(), "rId3".to_string())].into_iter().collect();
|
||||
|
||||
@@ -272,13 +272,7 @@ pub(crate) fn write_parts(doc: &Document, flavor: Flavor) -> Result<Vec<(String,
|
||||
|
||||
// The glossary: kept verbatim while its entries are as read and its parts were kept, else
|
||||
// written from the model (before the notes: references its entries hold are collected too).
|
||||
// When the privacy settings ask for changes, its kept parts are written without the names and
|
||||
// dates of its tracked changes and comments (#3692), or from the model if they can't be.
|
||||
let (names, dates) = (doc.settings.remove_personal_info, doc.settings.remove_date_time);
|
||||
if glossary::edited(doc)
|
||||
|| (!doc.glossary.is_empty() && !wr.embeds.kept_glossary())
|
||||
|| ((names || dates) && !wr.embeds.scrub_glossary(names, dates))
|
||||
{
|
||||
if glossary::edited(doc) || (!doc.glossary.is_empty() && !wr.embeds.kept_glossary()) {
|
||||
wr.embeds.drop_glossary = true;
|
||||
wr.glossary(&mut entries, &mut overrides, &mut rels);
|
||||
}
|
||||
@@ -726,15 +720,7 @@ fn styles_xml(doc: &Document) -> Vec<u8> {
|
||||
w.open("w:docDefaults", &[]);
|
||||
w.open("w:rPrDefault", &[]);
|
||||
w.open("w:rPr", &[]);
|
||||
if s.default_chr.size.is_none() {
|
||||
// The reader reads an omitted `w:sz` as 10 pt, but the resolver falls back to
|
||||
// DEFAULT_FONT_SIZE, so writing nothing would change the size on reopening.
|
||||
let mut chr = s.default_chr.clone();
|
||||
chr.size = Some(wordcraft_doc::props::DEFAULT_FONT_SIZE);
|
||||
props::rpr_inner(&mut w, &chr);
|
||||
} else {
|
||||
props::rpr_inner(&mut w, &s.default_chr);
|
||||
}
|
||||
props::rpr_inner(&mut w, &s.default_chr);
|
||||
w.close("w:rPr");
|
||||
w.close("w:rPrDefault");
|
||||
if s.default_para_present || s.default_para != wordcraft_doc::props::ParaProps::default() {
|
||||
@@ -1089,9 +1075,6 @@ fn settings_xml(doc: &Document, footnotes: bool, endnotes: bool, template_rel: O
|
||||
}
|
||||
w.open("w:compat", &[]);
|
||||
// CT_Compat: the older switches come before the `w:compatSetting`s.
|
||||
if doc.settings.do_not_expand_shift_return {
|
||||
w.empty("w:doNotExpandShiftReturn", &[]);
|
||||
}
|
||||
if doc.settings.sum_paragraph_spacing {
|
||||
w.empty("w:doNotUseHTMLParagraphAutoSpacing", &[]);
|
||||
}
|
||||
@@ -1144,9 +1127,9 @@ fn write_comment_ids(
|
||||
},
|
||||
};
|
||||
used.insert(durable.clone());
|
||||
// Only a known UTC time: `w:date` is local time, often with a `Z` it doesn't mean.
|
||||
// A local `w:date` with a zone designator is UTC already; otherwise only the file's own.
|
||||
// Remove date and time on save (#1062): no comment date here either.
|
||||
let utc = if doc.settings.remove_date_time { None } else { c.date_utc.clone() };
|
||||
let utc = if doc.settings.remove_date_time { None } else { c.date_utc.clone().or_else(|| c.date.ends_with('Z').then(|| c.date.clone())) };
|
||||
rows.push((pid.clone(), durable, utc));
|
||||
}
|
||||
let mc = "http://schemas.openxmlformats.org/markup-compatibility/2006";
|
||||
@@ -1237,7 +1220,7 @@ fn theme_xml(doc: &Document) -> Vec<u8> {
|
||||
/// The author the saved file gives a comment or tracked change when the document removes personal
|
||||
/// information on save (`w:removePersonalInformation`), as Word does.
|
||||
pub(crate) const ANONYMOUS_AUTHOR: &str = "Author";
|
||||
pub(crate) const ANONYMOUS_INITIALS: &str = "A";
|
||||
const ANONYMOUS_INITIALS: &str = "A";
|
||||
|
||||
/// The `w:author` written for an annotation by `author`: "Author" when the document removes
|
||||
/// personal information on save, "Unknown" when it has none.
|
||||
|
||||
@@ -1069,9 +1069,6 @@ impl Writer<'_> {
|
||||
ShapeKind::ElbowConnector => "bentConnector3",
|
||||
ShapeKind::CurvedConnector => "curvedConnector3",
|
||||
ShapeKind::Arrow => "rightArrow",
|
||||
ShapeKind::ArrowLeft => "leftArrow",
|
||||
ShapeKind::ArrowUp => "upArrow",
|
||||
ShapeKind::ArrowDown => "downArrow",
|
||||
ShapeKind::Star => "star5",
|
||||
ShapeKind::Heart => "heart",
|
||||
};
|
||||
@@ -1097,8 +1094,7 @@ impl Writer<'_> {
|
||||
match stroke {
|
||||
Some(c) => {
|
||||
let width = emu(stroke_width.clamp(0.0, 100.0));
|
||||
let cap = if ink { "rnd" } else { extra.cap.ooxml() };
|
||||
let ln: &[(&str, &str)] = if ink || !extra.cap.is_flat() { &[("w", width.as_str()), ("cap", cap)] } else { &[("w", width.as_str())] };
|
||||
let ln: &[(&str, &str)] = if ink { &[("w", width.as_str()), ("cap", "rnd")] } else { &[("w", width.as_str())] };
|
||||
w.open("a:ln", ln);
|
||||
// An outline gradient read from a file, while the outline keeps its colour (#1500).
|
||||
let grad = extra.outline_fill.as_ref().filter(|f| !ink && f.main_color() == *c);
|
||||
@@ -1123,9 +1119,7 @@ impl Writer<'_> {
|
||||
w.close("a:ln");
|
||||
}
|
||||
None => {
|
||||
let cap = extra.cap.ooxml();
|
||||
let ln: &[(&str, &str)] = if extra.cap.is_flat() { &[] } else { &[("cap", cap)] };
|
||||
w.open("a:ln", ln);
|
||||
w.open("a:ln", &[]);
|
||||
w.empty("a:noFill", &[]);
|
||||
line_dash(w, &extra.dash);
|
||||
w.close("a:ln");
|
||||
@@ -1723,11 +1717,12 @@ pub(crate) fn written_toc_span(bl: &Blocks) -> Option<(usize, usize)> {
|
||||
|
||||
/// The engine's TOC: a paragraph holding an empty `TOC` field, then its TOC-styled entries.
|
||||
/// Returns (heading, last entry) so the field can be written spanning the entries, as Word does.
|
||||
/// The heading may end a section: entries in columns follow a continuous break (#3889).
|
||||
fn toc_span(bl: &Blocks) -> Option<(usize, usize)> {
|
||||
let start = bl.iter().position(|b| {
|
||||
b.as_para()
|
||||
.is_some_and(|p| p.objects.iter().any(|o| matches!(o, InlineObject::Field { instr, result, .. } if is_toc(instr) && result.is_empty())))
|
||||
b.as_para().is_some_and(|p| {
|
||||
p.section.is_none()
|
||||
&& p.objects.iter().any(|o| matches!(o, InlineObject::Field { instr, result, .. } if is_toc(instr) && result.is_empty()))
|
||||
})
|
||||
})?;
|
||||
let entries = bl[start + 1..]
|
||||
.iter()
|
||||
|
||||
@@ -23,21 +23,8 @@ pub(super) fn text(wm: &Watermark) -> Option<String> {
|
||||
pub(super) fn run(wm: &Watermark, n: usize, page: &SectionProps) -> Option<String> {
|
||||
let text = text(wm)?;
|
||||
let [width, height] = wm.frame(page.page_w, page.page_h);
|
||||
// A font name can't break out of the textpath's CSS-like style or make it malformed (#3994):
|
||||
// the characters a quoted CSS string can't carry raw (`\22 ` is `"`, `\3b ` `;`, `\3a ` `:`,
|
||||
// `\5c ` `\`) are escaped and control characters are dropped (no font name carries one); the
|
||||
// reader (`read::story::css_string`) maps the escapes back. An apostrophe is legal inside a
|
||||
// double-quoted string, so it is kept.
|
||||
let mut font = String::with_capacity(wm.font.len());
|
||||
for c in wm.font.chars().filter(|c| !c.is_control()).take(100) {
|
||||
match c {
|
||||
'"' => font.push_str(r"\22 "),
|
||||
';' => font.push_str(r"\3b "),
|
||||
':' => font.push_str(r"\3a "),
|
||||
'\\' => font.push_str(r"\5c "),
|
||||
c => font.push(c),
|
||||
}
|
||||
}
|
||||
// A font name can't break out of the textpath's CSS-like style.
|
||||
let font: String = wm.font.chars().filter(|c| !matches!(c, '"' | '\'' | ';' | ':')).take(100).collect();
|
||||
let font = if font.trim().is_empty() { Watermark::default().font } else { font };
|
||||
let rotation = if wm.diagonal { "rotation:315;" } else { "" };
|
||||
let style = format!(
|
||||
|
||||
+4
-76
@@ -15,8 +15,6 @@ use crate::DocxError;
|
||||
pub const MAX_DEPTH: usize = 256;
|
||||
/// Maximum number of elements in one part.
|
||||
pub const MAX_ELEMENTS: usize = 8_000_000;
|
||||
/// Maximum number of attributes on one element (#3702).
|
||||
pub const MAX_ATTRS: usize = 1024;
|
||||
|
||||
/// Known namespaces and the canonical prefix we use for them (read and write).
|
||||
pub const NAMESPACES: &[(&str, &str)] = &[
|
||||
@@ -311,16 +309,6 @@ pub fn parse(bytes: &[u8]) -> Result<El, DocxError> {
|
||||
/// [`parse`] with `extra` namespaces (prefix, URI) mapped to canonical prefixes first, so a
|
||||
/// related vocabulary (WordprocessingML 2003) reads with the prefixes the reader knows.
|
||||
pub fn parse_ns(bytes: &[u8], extra: &[(&'static str, &str)]) -> Result<El, DocxError> {
|
||||
parse_with(bytes, extra, false)
|
||||
}
|
||||
|
||||
/// [`parse`] for a part whose loss would be silent data loss (the main document): a mismatched end
|
||||
/// tag or an element left open at the end of the data is an error, not a cue to stop reading.
|
||||
pub fn parse_strict(bytes: &[u8]) -> Result<El, DocxError> {
|
||||
parse_with(bytes, &[], true)
|
||||
}
|
||||
|
||||
fn parse_with(bytes: &[u8], extra: &[(&'static str, &str)], strict: bool) -> Result<El, DocxError> {
|
||||
// UTF-16 (with a BOM) is transcoded to UTF-8; a UTF-8 BOM is stripped.
|
||||
let transcoded = utf16_to_utf8(bytes);
|
||||
let bytes = match &transcoded {
|
||||
@@ -333,7 +321,7 @@ fn parse_with(bytes: &[u8], extra: &[(&'static str, &str)], strict: bool) -> Res
|
||||
c.trim_text_start = false;
|
||||
c.trim_text_end = false;
|
||||
c.expand_empty_elements = false;
|
||||
c.check_end_names = strict;
|
||||
c.check_end_names = false;
|
||||
}
|
||||
let mut stack: Vec<El> = Vec::new();
|
||||
let mut root: Option<El> = None;
|
||||
@@ -348,9 +336,6 @@ fn parse_with(bytes: &[u8], extra: &[(&'static str, &str)], strict: bool) -> Res
|
||||
};
|
||||
match ev {
|
||||
Event::Start(ref e) | Event::Empty(ref e) => {
|
||||
if strict && root.is_some() && stack.is_empty() {
|
||||
return Err(DocxError::Xml("more than one root element".into()));
|
||||
}
|
||||
count += 1;
|
||||
if count > MAX_ELEMENTS {
|
||||
return Err(DocxError::Limit("too many XML elements".into()));
|
||||
@@ -362,9 +347,6 @@ fn parse_with(bytes: &[u8], extra: &[(&'static str, &str)], strict: bool) -> Res
|
||||
};
|
||||
let mut el = El { name, attrs: Vec::new(), kids: Vec::new() };
|
||||
for a in e.attributes().with_checks(false) {
|
||||
if el.attrs.len() >= MAX_ATTRS {
|
||||
return Err(DocxError::Limit("too many attributes on one element".into()));
|
||||
}
|
||||
let Ok(a) = a else { continue };
|
||||
let key = a.key;
|
||||
let raw_key = key.as_ref();
|
||||
@@ -441,23 +423,18 @@ fn parse_with(bytes: &[u8], extra: &[(&'static str, &str)], strict: bool) -> Res
|
||||
}
|
||||
Err(_) => None,
|
||||
};
|
||||
match ch {
|
||||
Some(c) => push_text(top, &c),
|
||||
None if strict => return Err(DocxError::Xml("an entity is not defined".into())),
|
||||
None => {}
|
||||
if let Some(c) = ch {
|
||||
push_text(top, &c);
|
||||
}
|
||||
}
|
||||
}
|
||||
Event::Eof => break,
|
||||
_ => {}
|
||||
}
|
||||
if !strict && root.is_some() && stack.is_empty() {
|
||||
if root.is_some() && stack.is_empty() {
|
||||
break;
|
||||
}
|
||||
}
|
||||
if strict && !stack.is_empty() {
|
||||
return Err(DocxError::Xml("an element is not closed".into()));
|
||||
}
|
||||
// Unclosed elements at EOF: close them (lenient).
|
||||
while let Some(el) = stack.pop() {
|
||||
attach(&mut stack, &mut root, el);
|
||||
@@ -577,22 +554,6 @@ pub fn body_ns() -> Vec<(String, String)> {
|
||||
mod tests {
|
||||
use super::*;
|
||||
|
||||
#[test]
|
||||
fn strict_parsing_rejects_a_dropped_entity_and_trailing_markup() {
|
||||
let bad = [&b"<w:t>before &unknown; after</w:t>"[..], b"<a></a><b></b>", b"<a></a><b>"];
|
||||
for xml in bad {
|
||||
assert!(parse_strict(xml).is_err(), "{}", String::from_utf8_lossy(xml));
|
||||
}
|
||||
assert!(
|
||||
parse_strict(
|
||||
b"<a>x & y</a>
|
||||
"
|
||||
)
|
||||
.is_ok()
|
||||
);
|
||||
assert!(parse(b"<a></a><b></b>").is_ok(), "lenient parsing still stops at the root");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_relationship_attribute_is_the_namespaced_one_or_the_alias_of_one() {
|
||||
// The relationship namespace, and VML's `o:relid`.
|
||||
@@ -674,39 +635,6 @@ mod tests {
|
||||
assert!(parse(s.as_bytes()).is_err());
|
||||
}
|
||||
|
||||
/// One start tag carrying 200,000 attributes (distinct or duplicate names) is refused or
|
||||
/// truncated instead of allocated: a 59 KB `.docx` used to cost 611 MB (#3702, WC-06).
|
||||
#[test]
|
||||
fn too_many_attributes_on_one_element_is_capped() {
|
||||
// The cap the issue suggests; a fix may pick a lower one, never a higher one.
|
||||
const CAP: usize = 1024;
|
||||
let distinct: String = (0..200_000).map(|i| format!(" a{i}=\"\"")).collect();
|
||||
let duplicate = " a=\"\"".repeat(200_000);
|
||||
for (label, attrs) in [("distinct", distinct), ("duplicate", duplicate)] {
|
||||
let x = format!("<w:p xmlns:w=\"{}\"{attrs}/>", NAMESPACES[0].1);
|
||||
match parse(x.as_bytes()) {
|
||||
Err(DocxError::Limit(_)) => {}
|
||||
Ok(p) => assert!(p.attrs.len() <= CAP, "{label}: kept {} attributes", p.attrs.len()),
|
||||
Err(e) => panic!("{label}: expected a Limit error or a capped element, got {e}"),
|
||||
}
|
||||
}
|
||||
// Real-world attribute counts are untouched.
|
||||
let x = format!(
|
||||
r#"<w:p xmlns:w="{}" w:rsidR="00A1" w:rsidRDefault="00A1" w:rsidP="00A1" w:rsidRPr="00A1"><w:rPr><w:rFonts w:ascii="A" w:hAnsi="B" w:eastAsia="C" w:cs="D" w:asciiTheme="minorHAnsi" w:hAnsiTheme="minorHAnsi" w:eastAsiaTheme="minorEastAsia" w:cstheme="minorBidi" w:hint="eastAsia"/></w:rPr></w:p>"#,
|
||||
NAMESPACES[0].1
|
||||
);
|
||||
let p = parse(x.as_bytes()).unwrap();
|
||||
assert_eq!(p.attrs.len(), 4);
|
||||
let fonts = p.child("w:rPr").and_then(|r| r.child("w:rFonts")).unwrap();
|
||||
assert_eq!(fonts.attrs.len(), 9);
|
||||
assert_eq!(fonts.attr("w:cstheme"), Some("minorBidi"));
|
||||
// Far more than any real producer writes, still well under any sane cap.
|
||||
let many: String = (0..256).map(|i| format!(" a{i}=\"{i}\"")).collect();
|
||||
let p = parse(format!("<p{many}/>").as_bytes()).unwrap();
|
||||
assert_eq!(p.attrs.len(), 256);
|
||||
assert_eq!(p.attr("a255"), Some("255"));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn garbage_is_error_not_panic() {
|
||||
assert!(parse(b"").is_err());
|
||||
|
||||
@@ -1,69 +0,0 @@
|
||||
//! The document default font size (`w:docDefaults/w:rPrDefault/w:rPr/w:sz`) across a save and
|
||||
//! reopen. An unset default is resolved as `DEFAULT_FONT_SIZE`; the writer must say so, because
|
||||
//! the reader reads an omitted size as 10 pt.
|
||||
use std::io::{Read, Write};
|
||||
use wordcraft_doc::Document;
|
||||
use wordcraft_doc::props::{CharProps, DEFAULT_FONT_SIZE};
|
||||
fn resolved(d: &Document) -> f32 {
|
||||
d.styles.resolve_char(None, &CharProps::default()).size
|
||||
}
|
||||
fn styles_xml(bytes: &[u8]) -> String {
|
||||
let mut archive = zip::ZipArchive::new(std::io::Cursor::new(bytes)).unwrap();
|
||||
let mut xml = String::new();
|
||||
archive.by_name("word/styles.xml").unwrap().read_to_string(&mut xml).unwrap();
|
||||
xml
|
||||
}
|
||||
fn defaults(xml: &str) -> &str {
|
||||
xml.split("<w:docDefaults>").nth(1).unwrap().split("</w:docDefaults>").next().unwrap()
|
||||
}
|
||||
#[test]
|
||||
fn an_unset_default_size_keeps_its_effective_size_across_save_and_reopen() {
|
||||
let mut d = Document::new();
|
||||
d.styles.default_chr.size = None;
|
||||
assert_eq!(resolved(&d), DEFAULT_FONT_SIZE);
|
||||
let bytes = wordcraft_docx::write(&d).unwrap();
|
||||
assert!(defaults(&styles_xml(&bytes)).contains(r#"<w:sz w:val="22"/>"#));
|
||||
let reopened = wordcraft_docx::read(&bytes).unwrap();
|
||||
assert_eq!(resolved(&reopened), DEFAULT_FONT_SIZE);
|
||||
// Saving again changes nothing further.
|
||||
let again = wordcraft_docx::read(&wordcraft_docx::write(&reopened).unwrap()).unwrap();
|
||||
assert_eq!(resolved(&again), DEFAULT_FONT_SIZE);
|
||||
}
|
||||
#[test]
|
||||
fn an_explicit_default_size_is_written_as_it_is() {
|
||||
for size in [8.0, 10.0, 11.0, 12.5, 18.0] {
|
||||
let mut d = Document::new();
|
||||
d.styles.default_chr.size = Some(size);
|
||||
let bytes = wordcraft_docx::write(&d).unwrap();
|
||||
assert_eq!(resolved(&wordcraft_docx::read(&bytes).unwrap()), size);
|
||||
}
|
||||
}
|
||||
#[test]
|
||||
fn a_file_that_omits_the_default_size_is_still_read_as_ten_points() {
|
||||
// The reader is unchanged: only what WordCraft writes differs. A package without a size
|
||||
// anywhere keeps its established reading.
|
||||
let root = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>"#;
|
||||
let doc = r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body><w:p><w:r><w:t>x</w:t></w:r></w:p></w:body></w:document>"#;
|
||||
let styles = r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:styles xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:docDefaults><w:rPrDefault><w:rPr/></w:rPrDefault></w:docDefaults></w:styles>"#;
|
||||
let rels = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/></Relationships>"#;
|
||||
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
|
||||
for (name, body) in [("_rels/.rels", root), ("word/document.xml", doc), ("word/_rels/document.xml.rels", rels), ("word/styles.xml", styles)] {
|
||||
zw.start_file(name, zip::write::SimpleFileOptions::default()).unwrap();
|
||||
zw.write_all(body.as_bytes()).unwrap();
|
||||
}
|
||||
let d = wordcraft_docx::read(&zw.finish().unwrap().into_inner()).unwrap();
|
||||
assert_eq!(d.styles.default_chr.size, Some(10.0));
|
||||
}
|
||||
#[test]
|
||||
fn an_unset_default_size_does_not_take_a_styles_size() {
|
||||
// The written default is the resolver's own fallback, not what the Normal style resolves to,
|
||||
// so a style that sets a size keeps it and one that does not still falls back to the default.
|
||||
let mut d = Document::new();
|
||||
d.styles.default_chr.size = None;
|
||||
if let Some(normal) = d.styles.get_mut("Normal") {
|
||||
normal.chr.size = Some(14.0);
|
||||
}
|
||||
let reopened = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
|
||||
assert_eq!(reopened.styles.default_chr.size, Some(DEFAULT_FONT_SIZE));
|
||||
assert_eq!(resolved(&reopened), resolved(&d));
|
||||
}
|
||||
@@ -48,40 +48,6 @@ fn read_body(body: &str) -> Document {
|
||||
wordcraft_docx::read(&docx(body, &[], &[])).unwrap()
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn repeated_style_ids_use_last() {
|
||||
let styles = format!(
|
||||
r#"<w:styles {W_NS}>
|
||||
<w:style w:type="paragraph" w:styleId="Repeated"><w:name w:val="First"/>
|
||||
<w:pPr><w:spacing w:after="240"/></w:pPr><w:rPr><w:b/><w:smallCaps/><w:sz w:val="28"/></w:rPr>
|
||||
</w:style>
|
||||
<w:style w:type="paragraph" w:styleId="Repeated"><w:name w:val="Last"/>
|
||||
<w:pPr><w:spacing w:before="120"/></w:pPr><w:rPr><w:i/><w:caps/><w:sz w:val="20"/></w:rPr>
|
||||
</w:style></w:styles>"#
|
||||
);
|
||||
let bytes = docx(
|
||||
r#"<w:p><w:pPr><w:pStyle w:val="Repeated"/></w:pPr><w:r><w:t>Original style probe</w:t></w:r></w:p>"#,
|
||||
&[("styles", "styles", "styles.xml")],
|
||||
&[("word/styles.xml", &styles)],
|
||||
);
|
||||
let mut document = wordcraft_docx::read(&bytes).unwrap();
|
||||
for round in 0..=2 {
|
||||
let style = document.styles.get("Repeated").unwrap();
|
||||
assert_eq!(style.name, "Last");
|
||||
assert_eq!(style.chr.size, Some(10.0));
|
||||
assert_eq!(style.chr.italic, Some(true));
|
||||
assert_eq!(style.chr.bold, None);
|
||||
assert_eq!(style.chr.caps, Some(true));
|
||||
assert_eq!(style.chr.small_caps, None);
|
||||
assert_eq!(style.para.space_before, Some(6.0));
|
||||
assert_eq!(style.para.space_after, None);
|
||||
assert_eq!(document.styles.styles.iter().filter(|style| style.id == "Repeated").count(), 1);
|
||||
if round < 2 {
|
||||
document = wordcraft_docx::read(&wordcraft_docx::write(&document).unwrap()).unwrap();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn paragraph_default_presence_survives_repeated_saves() {
|
||||
// Omission leaves paragraph defaults application-defined (ISO 29500 pPrDefault).
|
||||
@@ -373,18 +339,6 @@ fn pcl6_symbol_roundtrip() {
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn sym_hex_prefixes_and_break_type_casing() {
|
||||
let body = r##"
|
||||
<w:p>
|
||||
<w:r><w:sym w:font="Wingdings" w:char="0xF0FC"/><w:sym w:font="Symbol" w:char="U+F061"/><w:sym w:font="Webdings" w:char="#xF061"/></w:r>
|
||||
<w:r><w:br w:type="Page"/><w:br w:type="Column"/><w:br w:type="textWrapping"/></w:r>
|
||||
</w:p>"##;
|
||||
let d = wordcraft_docx::read(&docx(body, &[], &[])).unwrap();
|
||||
let p = paras(&d);
|
||||
assert_eq!(p[0].plain_text(), "\u{F0FC}\u{F061}\u{F061}\u{C}\u{E}\n");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn alternate_content_choice_and_fallback() {
|
||||
let body = r#"
|
||||
@@ -729,32 +683,6 @@ fn a_style_defined_twice_is_its_last_definition() {
|
||||
assert_eq!(h.para.align, None, "nothing kept from the first definition");
|
||||
}
|
||||
|
||||
/// #3889: Word in another language writes the TOC styles under ids in that language (`toc 1` is
|
||||
/// `Sumrio1` in Portuguese). They take the engine's ids, which Update Table and the writer know
|
||||
/// the entries by, with their formatting and references; an id the file already uses stays.
|
||||
#[test]
|
||||
fn localized_toc_style_ids_take_the_engine_ids() {
|
||||
let styles = format!(
|
||||
r#"<w:styles {W_NS}>
|
||||
<w:style w:type="paragraph" w:default="1" w:styleId="Normal"><w:name w:val="Normal"/></w:style>
|
||||
<w:style w:type="paragraph" w:styleId="CabealhodoSumrio"><w:name w:val="TOC Heading"/><w:next w:val="Normal"/></w:style>
|
||||
<w:style w:type="paragraph" w:styleId="Sumrio1"><w:name w:val="toc 1"/><w:basedOn w:val="Normal"/><w:next w:val="Normal"/><w:pPr><w:spacing w:after="100"/></w:pPr></w:style>
|
||||
<w:style w:type="paragraph" w:styleId="Sumrio2"><w:name w:val="toc 2"/><w:basedOn w:val="Sumrio1"/><w:pPr><w:ind w:left="220"/></w:pPr></w:style>
|
||||
<w:style w:type="paragraph" w:styleId="TOC3"><w:name w:val="toc 3"/></w:style>
|
||||
<w:style w:type="paragraph" w:styleId="Sumrio3"><w:name w:val="toc 3"/></w:style>
|
||||
</w:styles>"#
|
||||
);
|
||||
let p = |st: &str| format!(r#"<w:p><w:pPr><w:pStyle w:val="{st}"/></w:pPr><w:r><w:t>{st}</w:t></w:r></w:p>"#);
|
||||
let body = ["CabealhodoSumrio", "Sumrio1", "Sumrio2", "Sumrio3"].map(p).concat();
|
||||
let d = wordcraft_docx::read(&docx(&body, &[("rId1", "styles", "styles.xml")], &[("word/styles.xml", &styles)])).unwrap();
|
||||
let ids: Vec<Option<&str>> = paras(&d).iter().map(|p| p.props.style.as_deref()).collect();
|
||||
assert_eq!(ids, [Some("TOCHeading"), Some("TOC1"), Some("TOC2"), Some("Sumrio3")]);
|
||||
let toc2 = d.styles.get("TOC2").unwrap();
|
||||
assert_eq!((toc2.name.as_str(), toc2.based_on.as_deref(), toc2.para.indent_left), ("TOC 2", Some("TOC1"), Some(11.0)));
|
||||
assert_eq!(d.styles.get("TOC1").unwrap().para.space_after, Some(5.0));
|
||||
assert!(d.styles.get("Sumrio1").is_none() && d.styles.get("TOC3").is_some() && d.styles.get("Sumrio3").is_some());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn vml_image_and_inline_drawing() {
|
||||
let png: &[u8] = b"\x89PNG\r\n\x1a\nnot really a png";
|
||||
@@ -1112,7 +1040,7 @@ fn documents_without_a_compatibility_mode_are_laid_out_as_word_2007() {
|
||||
#[test]
|
||||
fn style_rfonts_without_ascii_inherits_doc_defaults_font() {
|
||||
// Issue #93: Normal names only East Asian / complex-script fonts, so Latin text keeps the
|
||||
// docDefaults font. A run hinted as East Asian keeps its East Asian font in `font_ea` (#4031).
|
||||
// docDefaults font. A run hinted as East Asian still uses its East Asian font.
|
||||
let styles = format!(
|
||||
r#"<w:styles {W_NS}>
|
||||
<w:docDefaults><w:rPrDefault><w:rPr><w:rFonts w:ascii="Calibri" w:eastAsia="Calibri" w:hAnsi="Calibri" w:cs="Calibri"/></w:rPr></w:rPrDefault><w:pPrDefault/></w:docDefaults>
|
||||
@@ -1123,28 +1051,9 @@ fn style_rfonts_without_ascii_inherits_doc_defaults_font() {
|
||||
let d = wordcraft_docx::read(&docx(body, &[("rId1", "styles", "styles.xml")], &[("word/styles.xml", &styles)])).unwrap();
|
||||
assert_eq!(d.styles.get("Normal").unwrap().chr.font, None);
|
||||
let ps = paras(&d);
|
||||
let rc = |p: &Paragraph| d.styles.resolve_char(p.props.style.as_deref(), &p.runs[0].props);
|
||||
assert_eq!(rc(ps[0]).font, "Calibri");
|
||||
assert_eq!((rc(ps[1]).font.as_str(), rc(ps[1]).font_ea.as_deref()), ("Calibri", Some("SimSun")));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn hinted_east_asian_or_cs_font_does_not_replace_the_latin_font() {
|
||||
// Issue #4031: `w:hint` only picks the font for ambiguous characters, so a run with
|
||||
// `w:eastAsia` + `w:hint="eastAsia"` and no `w:ascii` keeps the style's Latin font for its
|
||||
// Latin text; the East Asian / complex-script fonts stay in their own fields.
|
||||
let styles = format!(
|
||||
r#"<w:styles {W_NS}>
|
||||
<w:style w:type="paragraph" w:styleId="Heading2"><w:name w:val="heading 2"/><w:rPr><w:rFonts w:ascii="Times New Roman" w:hAnsi="Times New Roman" w:cs="Times New Roman" w:eastAsiaTheme="majorEastAsia"/><w:b/></w:rPr></w:style>
|
||||
</w:styles>"#
|
||||
);
|
||||
let body = r#"<w:p><w:pPr><w:pStyle w:val="Heading2"/></w:pPr><w:r><w:rPr><w:rFonts w:eastAsia="宋体" w:hint="eastAsia"/></w:rPr><w:t>corn straw</w:t></w:r><w:r><w:rPr><w:rFonts w:cs="Arial" w:hint="cs"/></w:rPr><w:t>model fluids</w:t></w:r></w:p>"#;
|
||||
let d = wordcraft_docx::read(&docx(body, &[("rId1", "styles", "styles.xml")], &[("word/styles.xml", &styles)])).unwrap();
|
||||
let p = paras(&d)[0];
|
||||
let ea = d.styles.resolve_char(p.props.style.as_deref(), &p.runs[0].props);
|
||||
assert_eq!((ea.font.as_str(), ea.font_ea.as_deref()), ("Times New Roman", Some("宋体")));
|
||||
let cs = d.styles.resolve_char(p.props.style.as_deref(), &p.runs[1].props);
|
||||
assert_eq!((cs.font.as_str(), cs.font_cs.as_str()), ("Times New Roman", "Arial"));
|
||||
let font = |p: &Paragraph| d.styles.resolve_char(p.props.style.as_deref(), &p.runs[0].props).font;
|
||||
assert_eq!(font(ps[0]), "Calibri");
|
||||
assert_eq!(font(ps[1]), "SimSun");
|
||||
}
|
||||
|
||||
#[test]
|
||||
@@ -2478,7 +2387,7 @@ fn vml_preset_shapes_are_drawn() {
|
||||
assert_eq!(kind(15), Some((ShapeKind::Rectangle, false)));
|
||||
assert_eq!(kind(16), Some((ShapeKind::Diamond, false)));
|
||||
assert_eq!(kind(17), Some((ShapeKind::Ellipse, false)));
|
||||
assert_eq!(shape(18), Some((ShapeKind::ArrowLeft, 60.0, 20.0, Some(Rgb(0xFF, 0xA5, 0x00)), black, 0.75, true, false)));
|
||||
assert_eq!(shape(18), Some((ShapeKind::Arrow, 60.0, 20.0, Some(Rgb(0xFF, 0xA5, 0x00)), black, 0.75, true, false)));
|
||||
assert_eq!(kind(19), Some((ShapeKind::Rectangle, false)));
|
||||
assert_eq!(kind(20), Some((ShapeKind::Star, false)));
|
||||
assert_eq!(shape(21), None);
|
||||
@@ -2549,12 +2458,12 @@ fn vml_nearest_preset_shape_survives_saving() {
|
||||
let mut z = zip::ZipArchive::new(std::io::Cursor::new(saved.as_slice())).unwrap();
|
||||
let mut xml = String::new();
|
||||
std::io::Read::read_to_string(&mut z.by_name("word/document.xml").unwrap(), &mut xml).unwrap();
|
||||
assert!(xml.contains(r#"prst="leftArrow""#), "{xml}");
|
||||
assert!(xml.contains(r#"prst="rightArrow""#), "{xml}");
|
||||
let again = wordcraft_docx::read(&saved).unwrap();
|
||||
assert!(
|
||||
matches!(
|
||||
¶s(&again)[0].objects[..],
|
||||
[InlineObject::Shape { kind: ShapeKind::ArrowLeft, w: 60.0, h: 20.0, fill: Some(Fill::Solid { color: Rgb(0xFF, 0xA5, 0x00), .. }), .. }]
|
||||
[InlineObject::Shape { kind: ShapeKind::Arrow, w: 60.0, h: 20.0, fill: Some(Fill::Solid { color: Rgb(0xFF, 0xA5, 0x00), .. }), .. }]
|
||||
),
|
||||
"{:?}",
|
||||
paras(&again)[0].objects
|
||||
@@ -3618,22 +3527,6 @@ fn html_auto_spacing_off_is_read_and_kept() {
|
||||
assert_eq!(settings(""), (false, false));
|
||||
}
|
||||
|
||||
/// `w:doNotExpandShiftReturn` in the settings' `w:compat`: justified lines ending in a manual line
|
||||
/// break keep their spacing (#3924).
|
||||
#[test]
|
||||
fn do_not_expand_shift_return_is_read_and_kept() {
|
||||
let settings = |compat: &str| {
|
||||
let s = format!(r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:settings {W_NS}><w:compat>{compat}</w:compat></w:settings>"#);
|
||||
let rels = [("rId9", "http://schemas.openxmlformats.org/officeDocument/2006/relationships/settings", "settings.xml")];
|
||||
let d = wordcraft_docx::read(&docx("<w:p/>", &rels, &[("word/settings.xml", &s)])).unwrap();
|
||||
let again = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
|
||||
(d.settings.do_not_expand_shift_return, again.settings.do_not_expand_shift_return)
|
||||
};
|
||||
assert_eq!(settings("<w:doNotExpandShiftReturn/>"), (true, true));
|
||||
assert_eq!(settings(r#"<w:doNotExpandShiftReturn w:val="0"/>"#), (false, false));
|
||||
assert_eq!(settings(""), (false, false));
|
||||
}
|
||||
|
||||
/// A right triangle (`rtTriangle`) has its right angle at the bottom left: a cover page's white
|
||||
/// corner cut-outs drawn as isosceles triangles pointed the wrong way.
|
||||
#[test]
|
||||
@@ -3775,50 +3668,6 @@ fn shape_dashes_import_and_save() {
|
||||
}
|
||||
}
|
||||
|
||||
/// Preserve explicit caps, default missing/unknown caps, and retain caps on unpainted outlines.
|
||||
#[test]
|
||||
fn shape_caps_import_and_round_trip() {
|
||||
use wordcraft_doc::connector::LineCap;
|
||||
for (attr, expected) in
|
||||
[("", LineCap::Flat), ("flat", LineCap::Flat), ("sq", LineCap::Square), ("rnd", LineCap::Round), ("unknown", LineCap::Flat)]
|
||||
{
|
||||
for outline in [true, false] {
|
||||
let cap = if attr.is_empty() { String::new() } else { format!(r#" cap="{attr}""#) };
|
||||
let fill = if outline { r#"<a:solidFill><a:srgbClr val="C00000"/></a:solidFill>"# } else { "<a:noFill/>" };
|
||||
let sppr = format!(r#"<a:prstGeom prst="line"/><a:noFill/><a:ln w="12700"{cap}>{fill}</a:ln>"#);
|
||||
let body = anchored_wsp("Cap probe", 889000, 0, &sppr, "", "<wps:cNvSpPr/>");
|
||||
let d = read_body(&format!("<w:p>{body}</w:p>"));
|
||||
let saved = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
|
||||
for doc in [&d, &saved] {
|
||||
let InlineObject::Shape { extra, .. } = ¶s(doc)[0].objects[0] else { panic!("shape") };
|
||||
assert_eq!(extra.cap, expected, "{attr:?}, outline={outline}");
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// Preserve direct and shapetype-inherited VML end caps through a DrawingML save.
|
||||
#[test]
|
||||
fn vml_line_caps_import_and_round_trip() {
|
||||
use wordcraft_doc::connector::LineCap;
|
||||
for (attr, expected) in [("", LineCap::Flat), ("flat", LineCap::Flat), ("square", LineCap::Square), ("round", LineCap::Round)] {
|
||||
for inherited in [false, true] {
|
||||
let cap = if attr.is_empty() { String::new() } else { format!(r#" endcap="{attr}""#) };
|
||||
let stroke = format!("<v:stroke{cap}/>");
|
||||
let (base, own) = if inherited { (stroke.as_str(), "") } else { ("", stroke.as_str()) };
|
||||
let body = format!(
|
||||
r##"<w:p><w:r><w:pict><v:shapetype id="probe" o:spt="32">{base}</v:shapetype><v:shape type="#probe" style="width:40pt;height:1pt" strokecolor="#ff0000">{own}</v:shape></w:pict></w:r></w:p>"##
|
||||
);
|
||||
let d = read_body(&body);
|
||||
let saved = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
|
||||
for doc in [&d, &saved] {
|
||||
let InlineObject::Shape { extra, .. } = ¶s(doc)[0].objects[0] else { panic!("shape") };
|
||||
assert_eq!(extra.cap, expected, "{attr:?}, inherited={inherited}");
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn bookmark_end_name_and_id_trim() {
|
||||
let body = r#"<w:p>
|
||||
|
||||
@@ -234,131 +234,6 @@ fn edited_glossary_is_written_from_the_model() {
|
||||
assert_eq!(twice.get(&target), files.get(&target));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn an_unchanged_glossary_follows_the_privacy_settings() {
|
||||
// An untouched glossary was written back byte for byte, so the names and dates of its
|
||||
// tracked changes survived a save that removes personal information (#3692).
|
||||
let tracked = r#"<w:docPart><w:docPartPr><w:name w:val="Clause"/><w:category><w:name w:val="General"/><w:gallery w:val="docParts"/></w:category></w:docPartPr><w:docPartBody><w:p><w:ins w:id="1" w:author="Quill Ramsey" w:date="2026-03-04T05:06:07Z"><w:r><w:t>Added clause</w:t></w:r></w:ins></w:p></w:docPartBody></w:docPart>"#;
|
||||
let bytes = package(
|
||||
&[("rId7", GLOSSARY_REL, "glossary/document.xml")],
|
||||
&[
|
||||
("[Content_Types].xml", content_types(&[("/word/glossary/document.xml", GLOSSARY_CT)])),
|
||||
("word/glossary/document.xml", glossary_xml(tracked)),
|
||||
],
|
||||
);
|
||||
let doc = wordcraft_docx::read(&bytes).unwrap();
|
||||
let as_read = save(&doc);
|
||||
let kept = text(&as_read, &glossary_target(&as_read).expect("a glossary"));
|
||||
assert!(kept.contains("Quill Ramsey"), "the fixture's glossary is kept verbatim: {kept}");
|
||||
for (personal, dates) in [(true, false), (false, true), (true, true)] {
|
||||
let mut private = doc.clone();
|
||||
private.settings.remove_personal_info = personal;
|
||||
private.settings.remove_date_time = dates;
|
||||
let files = save(&private);
|
||||
let g = text(&files, &glossary_target(&files).expect("a glossary"));
|
||||
assert!(g.contains("Added clause"), "{g}");
|
||||
assert_eq!(g.contains("Quill Ramsey"), !personal, "{g}");
|
||||
assert_eq!(g.contains("2026-03-04T05:06:07Z"), !dates, "{g}");
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_private_save_keeps_the_glossary_entries_the_model_passed_over() {
|
||||
// The privacy settings had the glossary written from the model, which holds only the
|
||||
// entries the reader took: one without a name was lost, and with no other entry the whole
|
||||
// glossary went. The kept parts are now written without the names and dates instead.
|
||||
let nameless = r#"<w:docPart><w:docPartPr><w:category><w:name w:val="General"/><w:gallery w:val="docParts"/></w:category></w:docPartPr><w:docPartBody><w:p><w:ins w:id="1" w:author="Quill Ramsey" w:date="2026-03-04T05:06:07Z"><w:r><w:t>Nameless entry</w:t></w:r></w:ins><w:commentRangeStart w:id="0"/><w:r><w:t>noted</w:t></w:r><w:commentRangeEnd w:id="0"/><w:r><w:commentReference w:id="0"/></w:r></w:p></w:docPartBody></w:docPart>"#;
|
||||
// The comments part binds WordprocessingML to another prefix: names are matched by namespace.
|
||||
let comments = format!(
|
||||
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><x:comments xmlns:x="{W}"><x:comment x:id="0" x:author="Quill Ramsey" x:date="2026-03-04T05:06:07Z" x:initials="QR"><x:p><x:r><x:t>A note</x:t></x:r></x:p></x:comment></x:comments>"#
|
||||
);
|
||||
let bytes = package(
|
||||
&[("rId7", GLOSSARY_REL, "glossary/document.xml")],
|
||||
&[
|
||||
(
|
||||
"[Content_Types].xml",
|
||||
content_types(&[
|
||||
("/word/glossary/document.xml", GLOSSARY_CT),
|
||||
("/word/glossary/comments.xml", "application/vnd.openxmlformats-officedocument.wordprocessingml.comments+xml"),
|
||||
]),
|
||||
),
|
||||
("word/glossary/document.xml", glossary_xml(nameless)),
|
||||
("word/glossary/_rels/document.xml.rels", rels(&[("rId1", &format!("{REL}/comments"), "comments.xml")])),
|
||||
("word/glossary/comments.xml", comments),
|
||||
],
|
||||
);
|
||||
let doc = wordcraft_docx::read(&bytes).unwrap();
|
||||
assert!(doc.glossary.is_empty(), "the reader passes over an entry without a name");
|
||||
for (personal, dates) in [(true, false), (false, true), (true, true)] {
|
||||
let mut private = doc.clone();
|
||||
private.settings.remove_personal_info = personal;
|
||||
private.settings.remove_date_time = dates;
|
||||
let files = save(&private);
|
||||
let target = glossary_target(&files).expect("the glossary survives");
|
||||
let g = text(&files, &target);
|
||||
assert!(g.contains("Nameless entry"), "{g}");
|
||||
assert_eq!(g.contains("Quill Ramsey"), !personal, "{g}");
|
||||
assert_eq!(g.contains("2026-03-04T05:06:07Z"), !dates, "{g}");
|
||||
let c = files.iter().find(|(n, _)| n.starts_with("word/glossary/comments")).map(|(_, b)| String::from_utf8_lossy(b).into_owned());
|
||||
let c = c.expect("the glossary's comments survive");
|
||||
assert!(c.contains("A note"), "{c}");
|
||||
assert_eq!(c.contains("Quill Ramsey"), !personal, "{c}");
|
||||
assert_eq!(c.contains("x:initials=\"QR\""), !personal, "{c}");
|
||||
assert_eq!(c.contains("2026-03-04T05:06:07Z"), !dates, "{c}");
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_kept_glossary_that_cant_be_rewritten_privately_is_written_from_the_model() {
|
||||
// UTF-16 markup isn't rewritten in place: the model, which applies the settings, is written.
|
||||
let tracked = r#"<w:docPart><w:docPartPr><w:name w:val="Clause"/><w:category><w:name w:val="General"/><w:gallery w:val="docParts"/></w:category></w:docPartPr><w:docPartBody><w:p><w:ins w:id="1" w:author="Quill Ramsey" w:date="2026-03-04T05:06:07Z"><w:r><w:t>Added clause</w:t></w:r></w:ins></w:p></w:docPartBody></w:docPart>"#;
|
||||
let utf16: Vec<u8> = [0xFF, 0xFE].into_iter().chain(glossary_xml(tracked).encode_utf16().flat_map(u16::to_le_bytes)).collect();
|
||||
let bytes = zip(&[
|
||||
("_rels/.rels", rels(&[("rIdMain", &format!("{REL}/officeDocument"), "word/document.xml")]).into_bytes()),
|
||||
(
|
||||
"word/document.xml",
|
||||
format!(r#"<w:document xmlns:w="{W}"><w:body><w:p><w:r><w:t>Body</w:t></w:r></w:p></w:body></w:document>"#).into_bytes(),
|
||||
),
|
||||
("word/_rels/document.xml.rels", rels(&[("rId7", GLOSSARY_REL, "glossary/document.xml")]).into_bytes()),
|
||||
("[Content_Types].xml", content_types(&[("/word/glossary/document.xml", GLOSSARY_CT)]).into_bytes()),
|
||||
("word/glossary/document.xml", utf16),
|
||||
]);
|
||||
let mut doc = wordcraft_docx::read(&bytes).unwrap();
|
||||
assert_eq!(doc.glossary.len(), 1);
|
||||
doc.settings.remove_personal_info = true;
|
||||
let (files, warned) = warnings(|| save(&doc));
|
||||
let g = text(&files, &glossary_target(&files).expect("a glossary"));
|
||||
assert!(g.contains("Added clause") && !g.contains("Quill Ramsey"), "{g}");
|
||||
// What the model doesn't hold is lost: the save says so (review of #3872).
|
||||
assert_eq!(warned.iter().filter(|w| w.contains("written from its entries")).count(), 1, "{warned:?}");
|
||||
}
|
||||
|
||||
/// `f`'s result and the warnings logged on this thread while it ran.
|
||||
fn warnings<T>(f: impl FnOnce() -> T) -> (T, Vec<String>) {
|
||||
use std::cell::RefCell;
|
||||
struct Capture;
|
||||
thread_local! {
|
||||
static LINES: RefCell<Vec<String>> = const { RefCell::new(Vec::new()) };
|
||||
}
|
||||
impl log::Log for Capture {
|
||||
fn enabled(&self, m: &log::Metadata) -> bool {
|
||||
m.level() <= log::Level::Warn
|
||||
}
|
||||
fn log(&self, r: &log::Record) {
|
||||
if self.enabled(r.metadata()) {
|
||||
LINES.with(|l| l.borrow_mut().push(r.args().to_string()));
|
||||
}
|
||||
}
|
||||
fn flush(&self) {}
|
||||
}
|
||||
static CAPTURE: Capture = Capture;
|
||||
let _ = log::set_logger(&CAPTURE);
|
||||
log::set_max_level(log::LevelFilter::Warn);
|
||||
LINES.with(|l| l.borrow_mut().clear());
|
||||
let out = f();
|
||||
(out, LINES.with(|l| std::mem::take(&mut *l.borrow_mut())))
|
||||
}
|
||||
|
||||
/// The relationships part of a part in `word/`.
|
||||
fn glossary_rels(part: &str) -> String {
|
||||
let (dir, file) = part.rsplit_once('/').unwrap();
|
||||
|
||||
@@ -7,7 +7,6 @@ mod complex_script_inheritance;
|
||||
mod content_control_save_cost;
|
||||
mod content_controls;
|
||||
mod custom_xml;
|
||||
mod default_size;
|
||||
mod drawing_payload;
|
||||
mod east_asian_lang;
|
||||
mod east_asian_layout;
|
||||
|
||||
@@ -68,25 +68,6 @@ fn missing_or_wrong_main_part() {
|
||||
assert_eq!(d.plain_text(wordcraft_doc::StoryRef::Body), "ok");
|
||||
}
|
||||
|
||||
fn body_zip(body: &str) -> Vec<u8> {
|
||||
doc_zip(format!("<w:document {W_NS}><w:body>{body}</w:body></w:document>").as_bytes())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_malformed_main_part_is_an_error_not_a_truncated_document() {
|
||||
let cases = [
|
||||
("a less-than sign in a run", "<w:p><w:r><w:t>before a <- b after</w:t></w:r></w:p><w:p><w:r><w:t>last</w:t></w:r></w:p>"),
|
||||
("a paragraph left open", "<w:p><w:r><w:t>first</w:t></w:r></w:p><w:p><w:r><w:t>second</w:t></w:r><w:p><w:r><w:t>third</w:t></w:r></w:p>"),
|
||||
("a mismatched end tag", "<w:p><w:r><w:t>one</w:t></w:p></w:r><w:p><w:r><w:t>two</w:t></w:r></w:p>"),
|
||||
];
|
||||
for (what, body) in cases {
|
||||
assert!(matches!(read(&body_zip(body)), Err(DocxError::Xml(_))), "{what}");
|
||||
}
|
||||
let whole = "<w:p><w:r><w:t>a <- b</w:t></w:r></w:p><w:p><w:r><w:t>last</w:t></w:r></w:p>";
|
||||
let d = read(&body_zip(whole)).unwrap();
|
||||
assert_eq!(d.plain_text(wordcraft_doc::StoryRef::Body), "a <- b\nlast", "well-formed text still reads whole");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn deep_nesting_is_rejected() {
|
||||
let mut x = format!("<w:document {W_NS}><w:body>");
|
||||
@@ -150,21 +131,6 @@ fn many_fields_and_huge_attribute_values() {
|
||||
assert!(d.plain_text(wordcraft_doc::StoryRef::Body).contains("later"));
|
||||
}
|
||||
|
||||
/// A paragraph with 2,000,000 attributes deflates to ~15 KB. Opening it is refused with a
|
||||
/// limit error, or the extras are dropped and the text survives (#3702, WC-06).
|
||||
#[test]
|
||||
fn millions_of_attributes_on_one_paragraph_are_capped() {
|
||||
let attrs: String = (0..2_000_000).map(|i| format!(" a{i}=\"\"")).collect();
|
||||
let x = format!("<w:document {W_NS}><w:body><w:p{attrs}><w:r><w:t>kept</w:t></w:r></w:p></w:body></w:document>");
|
||||
let start = std::time::Instant::now();
|
||||
match read(&doc_zip(x.as_bytes())) {
|
||||
Err(DocxError::Limit(_)) => {}
|
||||
Ok(d) => assert!(d.plain_text(wordcraft_doc::StoryRef::Body).contains("kept")),
|
||||
Err(e) => panic!("expected a Limit error or a document, got {e}"),
|
||||
}
|
||||
assert!(start.elapsed() < std::time::Duration::from_secs(10), "took {:?}", start.elapsed());
|
||||
}
|
||||
|
||||
/// A small valid document.xml to mutate.
|
||||
fn base_xml() -> String {
|
||||
format!(
|
||||
@@ -273,49 +239,6 @@ proptest! {
|
||||
}
|
||||
}
|
||||
|
||||
/// The streams of a Standard-encrypted package (gap 9), and a compound file holding them.
|
||||
fn standard_streams() -> (Vec<u8>, Vec<u8>) {
|
||||
use std::io::Read;
|
||||
let file = wordcraft_docx::encrypt_standard_for_tests(&sample(), "pw", 128).unwrap();
|
||||
let mut comp = cfb::CompoundFile::open(std::io::Cursor::new(file)).unwrap();
|
||||
let mut info = Vec::new();
|
||||
comp.open_stream("/EncryptionInfo").unwrap().read_to_end(&mut info).unwrap();
|
||||
let mut package = Vec::new();
|
||||
comp.open_stream("/EncryptedPackage").unwrap().read_to_end(&mut package).unwrap();
|
||||
(info, package)
|
||||
}
|
||||
|
||||
fn compound(info: &[u8], package: &[u8]) -> Vec<u8> {
|
||||
let mut comp = cfb::CompoundFile::create(std::io::Cursor::new(Vec::new())).unwrap();
|
||||
comp.create_stream("/EncryptionInfo").unwrap().write_all(info).unwrap();
|
||||
comp.create_stream("/EncryptedPackage").unwrap().write_all(package).unwrap();
|
||||
comp.flush().unwrap();
|
||||
comp.into_inner().into_inner()
|
||||
}
|
||||
|
||||
proptest! {
|
||||
// Each case that parses runs the 50,000-spin key derivation, so fewer cases.
|
||||
#![proptest_config(ProptestConfig { cases: 48, ..ProptestConfig::default() })]
|
||||
|
||||
#[test]
|
||||
fn mutated_standard_encryption_never_panics(ops in proptest::collection::vec((0u8..3, any::<prop::sample::Index>(), any::<u8>(), any::<bool>()), 1..8)) {
|
||||
let (mut info, mut package) = standard_streams();
|
||||
for (k, idx, byte, in_package) in ops {
|
||||
let x = if in_package { &mut package } else { &mut info };
|
||||
let at = idx.index(x.len() + 1);
|
||||
match k {
|
||||
0 => { if at < x.len() { x[at] = byte; } }
|
||||
1 => { x.truncate(at); }
|
||||
_ => { x.splice(at..at, std::iter::repeat_n(byte, usize::from(byte % 9))); }
|
||||
}
|
||||
}
|
||||
let file = compound(&info, &package);
|
||||
let _ = wordcraft_docx::decrypt(&file, Some("pw"));
|
||||
let _ = wordcraft_docx::read_with_password(&file, Some("pw"));
|
||||
prop_assert!(read(&file).is_err());
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn long_header_reference_chain_is_cut_off() {
|
||||
const HDR_REL: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/header";
|
||||
@@ -342,38 +265,3 @@ fn long_header_reference_chain_is_cut_off() {
|
||||
assert!(!d.parts.is_empty());
|
||||
assert!(d.parts.len() <= 8, "followed {} chained headers", d.parts.len());
|
||||
}
|
||||
|
||||
/// #3642: a compound file whose DIFAT names one FAT sector thousands of times is refused by the
|
||||
/// encrypted-package sniff (reached for any `.docx` name) before `cfb` builds its FAT.
|
||||
#[test]
|
||||
fn a_repeating_difat_is_refused_before_cfb_opens_it() {
|
||||
let f = wordcraft_doc::cfb_guard::repeating_difat_sample();
|
||||
assert!(!wordcraft_docx::is_encrypted(&f));
|
||||
assert!(matches!(wordcraft_docx::decrypt(&f, None), Err(e) if e.to_string().contains("FAT sectors in a file of")));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn comments_with_non_numeric_ids_open_in_linear_time() {
|
||||
// Each generated id was checked against every id given so far, so a comments part with many
|
||||
// ids that aren't numbers took time quadratic in their count (#3692).
|
||||
const RELS_NS: &str = "http://schemas.openxmlformats.org/package/2006/relationships";
|
||||
const COMMENTS_REL: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/comments";
|
||||
let n = 49_000;
|
||||
let mut comments = format!("<w:comments {W_NS}>");
|
||||
for i in 0..n {
|
||||
comments.push_str(&format!(r#"<w:comment w:id="c{i}" w:author="A"><w:p/></w:comment>"#));
|
||||
}
|
||||
comments.push_str("</w:comments>");
|
||||
let doc = format!("<w:document {W_NS}><w:body><w:p><w:r><w:t>body</w:t></w:r></w:p></w:body></w:document>");
|
||||
let rels = format!(r#"<Relationships xmlns="{RELS_NS}"><Relationship Id="rId1" Type="{COMMENTS_REL}" Target="comments.xml"/></Relationships>"#);
|
||||
let z = zip(&[
|
||||
("_rels/.rels", ROOT_RELS.as_bytes()),
|
||||
("word/document.xml", doc.as_bytes()),
|
||||
("word/_rels/document.xml.rels", rels.as_bytes()),
|
||||
("word/comments.xml", comments.as_bytes()),
|
||||
]);
|
||||
let d = read(&z).unwrap();
|
||||
assert_eq!(d.comments.len(), n);
|
||||
let ids: std::collections::HashSet<_> = d.comments.keys().collect();
|
||||
assert_eq!(ids.len(), n, "every comment has its own id");
|
||||
}
|
||||
|
||||
@@ -249,21 +249,6 @@ fn comments_without_durable_ids_get_unique_ones() {
|
||||
assert!(dates.iter().all(|(_, utc)| utc.is_empty()), "{dates:?}");
|
||||
}
|
||||
|
||||
/// A local `w:date` with a `Z` (older files have no `dateUtc`) isn't saved as the UTC time either,
|
||||
/// so it doesn't shift by the user's offset after a round trip.
|
||||
#[test]
|
||||
fn a_local_date_with_a_z_is_not_saved_as_utc() {
|
||||
let mut doc = wordcraft_docx::read(&comments_package(false)).unwrap();
|
||||
for c in doc.comments.values_mut() {
|
||||
c.date.push('Z');
|
||||
}
|
||||
let files = save(&doc);
|
||||
let (_, dates) = written_ids(&files);
|
||||
assert!(dates.iter().all(|(_, utc)| utc.is_empty()), "{dates:?}");
|
||||
let back = wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap();
|
||||
assert!(back.comments.values().all(|c| c.date_utc.is_none() && c.date.ends_with('Z')));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn comment_references_use_the_documents_annotation_reference_style() {
|
||||
let files = save(&wordcraft_docx::read(&comments_package(true)).unwrap());
|
||||
@@ -357,49 +342,3 @@ fn extended_properties_are_kept_without_words_counts_and_personal_ones_follow_th
|
||||
assert!(!private.contains("Company") && !private.contains("Manager"), "{private}");
|
||||
assert!(private.contains("<Template>Normal.dotm</Template>"));
|
||||
}
|
||||
|
||||
/// `name` with its letters' case flipped by the bits of `i`.
|
||||
fn spelling(name: &str, i: u32) -> String {
|
||||
let mut bit = 0;
|
||||
name.chars()
|
||||
.map(|c| {
|
||||
if !c.is_ascii_alphabetic() {
|
||||
return c;
|
||||
}
|
||||
bit += 1;
|
||||
if i >> (bit - 1) & 1 == 1 { c.to_ascii_uppercase() } else { c.to_ascii_lowercase() }
|
||||
})
|
||||
.collect()
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn a_part_reached_under_many_spellings_is_kept_and_written_once() {
|
||||
// Part names are case-insensitive (ECMA-376 Part 2 §6.2.2.3): 64 relationships naming one picture
|
||||
// and one unmodelled part in 64 casings are two parts, not 128 copies on save (#3704).
|
||||
let ns = r#"xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:pic="http://schemas.openxmlformats.org/drawingml/2006/picture""#;
|
||||
let n = 64;
|
||||
let mut body = String::new();
|
||||
let mut doc_rels = Vec::new();
|
||||
for i in 0..n {
|
||||
body.push_str(&format!(
|
||||
r#"<w:p><w:r><w:drawing><wp:inline {ns}><wp:extent cx="9525" cy="9525"/><wp:docPr id="{}" name="p"/><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/picture"><pic:pic><pic:blipFill><a:blip r:embed="rIdP{i}"/></pic:blipFill></pic:pic></a:graphicData></a:graphic></wp:inline></w:drawing></w:r></w:p>"#,
|
||||
i + 1
|
||||
));
|
||||
doc_rels.push((format!("rIdP{i}"), format!("{REL}/image"), spelling("media/image1.png", i)));
|
||||
doc_rels.push((format!("rIdU{i}"), "urn:example:future-relationship".to_string(), spelling("unknownPart.xml", i)));
|
||||
}
|
||||
let doc_rels: Vec<(&str, &str, &str)> = doc_rels.iter().map(|(a, b, c)| (a.as_str(), b.as_str(), c.as_str())).collect();
|
||||
let picture = format!("\u{89}PNG\r\n\u{1a}\n{}", "0".repeat(64 * 1024));
|
||||
let unknown = format!(r#"<x:thing xmlns:x="urn:example:future">{}</x:thing>"#, "x".repeat(64 * 1024));
|
||||
let ct = CT.replace("<Default Extension=\"jpeg\"", "<Default Extension=\"png\" ContentType=\"image/png\"/><Default Extension=\"jpeg\"");
|
||||
let bytes =
|
||||
package(&body, &doc_rels, &[], &[("[Content_Types].xml", &ct), ("word/media/image1.png", &picture), ("word/unknownPart.xml", &unknown)]);
|
||||
let doc = wordcraft_docx::read(&bytes).unwrap();
|
||||
assert_eq!(doc.media.len(), 1, "{:?}", doc.media.keys());
|
||||
let files = save(&doc);
|
||||
let named = |s: &str| files.keys().filter(|k| k.to_ascii_lowercase().contains(s)).count();
|
||||
assert_eq!(named("unknownpart"), 1, "{:?}", files.keys());
|
||||
assert_eq!(files.keys().filter(|k| k.starts_with("word/media/")).count(), 1, "{:?}", files.keys());
|
||||
let written: usize = files.values().map(Vec::len).sum();
|
||||
assert!(written < 4 * 64 * 1024, "{written} bytes written");
|
||||
}
|
||||
|
||||
@@ -1545,33 +1545,6 @@ fn watermark_roundtrips() {
|
||||
}
|
||||
}
|
||||
|
||||
/// #3994: a watermark font name round-trips through the textpath's CSS-like style whatever it
|
||||
/// carries — apostrophes, semicolons, colons, quotes, backslashes. Before, those characters
|
||||
/// were deleted (the font was saved under a different name) and a trailing backslash escaped
|
||||
/// the style's closing quote.
|
||||
#[test]
|
||||
fn watermark_font_roundtrips_through_the_textpath_style() {
|
||||
for font in ["Georgia", "Bob's Serif", "A;Named:Font", "Quote\"Inside", "Ends with \\", "All \"of\" ;it: 'here\\", "'Serif'", "Trailing'"] {
|
||||
let mut d = doc_with(vec![Paragraph::with_text("Body", CharProps::default())]);
|
||||
d.settings.watermark = Some(Watermark { font: font.into(), ..Default::default() });
|
||||
assert_eq!(rt(&d).settings.watermark.map(|w| w.font), Some(font.into()), "{font:?}");
|
||||
}
|
||||
}
|
||||
|
||||
/// #3994: the written textpath style stays well-formed CSS whatever the font name: control
|
||||
/// characters are dropped (no font name carries one) and a backslash or quote is a CSS escape
|
||||
/// (`\5c `, `\22 `) instead of escaping the string's closing quote.
|
||||
#[test]
|
||||
fn watermark_font_name_writes_wellformed_css() {
|
||||
let mut d = doc_with(vec![Paragraph::with_text("Body", CharProps::default())]);
|
||||
d.settings.watermark = Some(Watermark { font: "Ends with \\ and \n\"quoted\"".into(), ..Default::default() });
|
||||
let bytes = wordcraft_docx::write(&d).unwrap();
|
||||
let hdr = part_xml(&bytes, "word/header1.xml");
|
||||
let style = hdr.split("v:textpath style=\"").nth(1).and_then(|s| s.split('"').next()).unwrap_or_default();
|
||||
assert!(!style.contains('\n'), "control characters are dropped: {style}");
|
||||
assert!(style.contains(r#"font-family:"Ends with \5c and \22 quoted\22 ";"#), "CSS escapes: {style}");
|
||||
}
|
||||
|
||||
/// #1290: a document without headers gains one default header holding only the watermark (an
|
||||
/// empty paragraph once read back); saving again neither adds another nor duplicates the shape.
|
||||
#[test]
|
||||
|
||||
@@ -8,14 +8,8 @@ description = "WordCraft engine: session, selection, undo, the command registry
|
||||
[lints]
|
||||
workspace = true
|
||||
|
||||
[features]
|
||||
# Front-end producer tests need to simulate a globally exhausted IME revision allocator. It only
|
||||
# ever makes receipt classification fail closed. Never enable this for a production build; the
|
||||
# UI's dev-dependencies enable it for its tests.
|
||||
test-support = []
|
||||
|
||||
[dependencies]
|
||||
wordcraft-geom = { workspace = true, features = ["image"] }
|
||||
wordcraft-geom = { workspace = true }
|
||||
wordcraft-doc = { workspace = true }
|
||||
wordcraft-fonts = { workspace = true }
|
||||
wordcraft-layout = { workspace = true }
|
||||
@@ -42,10 +36,6 @@ js-sys = "0.3"
|
||||
[target.'cfg(not(target_arch = "wasm32"))'.dependencies]
|
||||
getrandom = { workspace = true }
|
||||
|
||||
[target.'cfg(unix)'.dependencies]
|
||||
# O_NONBLOCK, so a path swapped for a named pipe can't block `open` (#3666).
|
||||
libc = { workspace = true }
|
||||
|
||||
[dev-dependencies]
|
||||
encoding_rs = { workspace = true }
|
||||
zip = { workspace = true }
|
||||
|
||||
@@ -346,9 +346,6 @@ View|Window|Switch Windows|view.switchWindows
|
||||
View|Macros|Macros|tools.macros
|
||||
View|Macros|Record Macro|tools.recordMacro
|
||||
View|SharePoint|Properties|file.properties
|
||||
Read Mode|View|Column Width|view.readColumnWidth
|
||||
Read Mode|View|Page Color|view.readPageColor
|
||||
Read Mode|View|Layout|view.readLayout
|
||||
Developer|Controls|Rich Text Content Control|developer.richText
|
||||
Developer|Controls|Plain Text Content Control|developer.plainText
|
||||
Developer|Controls|Picture Content Control|developer.picture
|
||||
@@ -594,17 +591,4 @@ mod tests {
|
||||
// The floor only ever rises.
|
||||
assert!(pct >= 90.4, "parity {pct}%");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn read_mode_view_menu_is_catalogued() {
|
||||
let e = entries();
|
||||
let read_mode: Vec<_> = e.iter().filter(|x| x.0 == "Read Mode").map(|x| (x.1, x.2, x.3)).collect();
|
||||
assert_eq!(
|
||||
read_mode,
|
||||
[("View", "Column Width", "view.readColumnWidth"), ("View", "Page Color", "view.readPageColor"), ("View", "Layout", "view.readLayout")]
|
||||
);
|
||||
let reg = crate::cmd::registry();
|
||||
assert!(reg.get("view.readColumnWidth").is_some(), "Column Width counts as live");
|
||||
assert!(reg.get("view.readPageColor").is_some(), "Page Color counts as live");
|
||||
}
|
||||
}
|
||||
|
||||
@@ -313,22 +313,6 @@ mod tests {
|
||||
}
|
||||
}
|
||||
|
||||
/// The per-case shape/geometry of one scaled-group reroute case.
|
||||
struct ScaledGroupSpec {
|
||||
w: f32,
|
||||
h: f32,
|
||||
kind: ShapeKind,
|
||||
rot: f32,
|
||||
flip_h: bool,
|
||||
flip_v: bool,
|
||||
}
|
||||
|
||||
impl ScaledGroupSpec {
|
||||
fn new(w: f32, h: f32, kind: ShapeKind, rot: f32, flip_h: bool, flip_v: bool) -> Self {
|
||||
Self { w, h, kind, rot, flip_h, flip_v }
|
||||
}
|
||||
}
|
||||
|
||||
fn visual_ends(group: &InlineObject) -> ((f32, f32), (f32, f32)) {
|
||||
let InlineObject::Group { w, h, float, .. } = group else { panic!("group") };
|
||||
let rects = group.group_rects(0.0, 0.0, *w, *h);
|
||||
@@ -343,61 +327,6 @@ mod tests {
|
||||
(turn((x, y)), turn((x + w, y + h)))
|
||||
}
|
||||
|
||||
/// One scaled-group case: rerouting keeps the visual free end and the resolved sites.
|
||||
fn assert_scaled_group_reroute(spec: ScaledGroupSpec, free: Option<ConnEnd>, free_start: bool) {
|
||||
let attached = Some(ConnEnd { id: 7, site: 3 });
|
||||
let (start, end) = if free_start { (free, attached) } else { (attached, free) };
|
||||
let group = InlineObject::Group {
|
||||
w: spec.w,
|
||||
h: spec.h,
|
||||
ch_w: 200.0,
|
||||
ch_h: 300.0,
|
||||
float: Float { rot: 25.0, flip_h: true, ..Default::default() },
|
||||
canvas: None,
|
||||
children: vec![
|
||||
GroupChild { x: 100.0, y: 200.0, obj: shape(ShapeKind::Rectangle, ShapeExtra { id: 7, ..Default::default() }, Float::default()) },
|
||||
GroupChild {
|
||||
x: 10.0,
|
||||
y: 30.0,
|
||||
obj: shape(
|
||||
spec.kind,
|
||||
ShapeExtra { start, end, ..Default::default() },
|
||||
Float { rot: spec.rot, flip_h: spec.flip_h, flip_v: spec.flip_v, ..Default::default() },
|
||||
),
|
||||
},
|
||||
],
|
||||
};
|
||||
// Rendering scales each child rectangle before applying its own spin.
|
||||
let before = visual_ends(&group);
|
||||
let expected = if free_start { before.0 } else { before.1 };
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
let pos = s.doc.start_of(StoryRef::Body);
|
||||
s.doc.para_mut(pos.story, &pos.path).unwrap().insert_object(pos.off, group, &Default::default()).unwrap();
|
||||
assert!(reroute(&mut s));
|
||||
let group = s.doc.para_at(&pos).unwrap().object_at(pos.off).unwrap();
|
||||
let after = visual_ends(group);
|
||||
let actual = if free_start { after.0 } else { after.1 };
|
||||
assert!(
|
||||
(actual.0 - expected.0).abs() < 0.001 && (actual.1 - expected.1).abs() < 0.001,
|
||||
"{}x{}, {:?}, {}, {}, {}, {}: {:?} != {:?}",
|
||||
spec.w,
|
||||
spec.h,
|
||||
spec.kind,
|
||||
spec.rot,
|
||||
spec.flip_h,
|
||||
spec.flip_v,
|
||||
free_start,
|
||||
actual,
|
||||
expected
|
||||
);
|
||||
let InlineObject::Group { children, .. } = group else { panic!("group") };
|
||||
let c = &children[1];
|
||||
let InlineObject::Shape { w, h, float, .. } = &c.obj else { panic!("connector") };
|
||||
let routed = ends(c.x, c.y, *w, *h, float.flip_h, float.flip_v);
|
||||
assert_eq!(if free_start { routed.1 } else { routed.0 }, (140.0, 210.0), "resolved sites stay in child coordinates");
|
||||
assert_eq!(float.rot, 0.0);
|
||||
assert!(!reroute(&mut s), "rerouting must settle");
|
||||
}
|
||||
#[test]
|
||||
fn rerouting_scaled_group_preserves_visual_free_and_unresolved_ends() {
|
||||
for (w, h) in [(200.0, 300.0), (400.0, 600.0), (400.0, 150.0)] {
|
||||
@@ -406,7 +335,53 @@ mod tests {
|
||||
for (flip_h, flip_v) in [(false, false), (true, false), (false, true), (true, true)] {
|
||||
for free in [None, Some(ConnEnd { id: 404, site: 0 }), Some(ConnEnd { id: 7, site: 99 })] {
|
||||
for free_start in [false, true] {
|
||||
assert_scaled_group_reroute(ScaledGroupSpec::new(w, h, kind, rot, flip_h, flip_v), free, free_start);
|
||||
let attached = Some(ConnEnd { id: 7, site: 3 });
|
||||
let (start, end) = if free_start { (free, attached) } else { (attached, free) };
|
||||
let group = InlineObject::Group {
|
||||
w,
|
||||
h,
|
||||
ch_w: 200.0,
|
||||
ch_h: 300.0,
|
||||
float: Float { rot: 25.0, flip_h: true, ..Default::default() },
|
||||
canvas: None,
|
||||
children: vec![
|
||||
GroupChild {
|
||||
x: 100.0,
|
||||
y: 200.0,
|
||||
obj: shape(ShapeKind::Rectangle, ShapeExtra { id: 7, ..Default::default() }, Float::default()),
|
||||
},
|
||||
GroupChild {
|
||||
x: 10.0,
|
||||
y: 30.0,
|
||||
obj: shape(
|
||||
kind,
|
||||
ShapeExtra { start, end, ..Default::default() },
|
||||
Float { rot, flip_h, flip_v, ..Default::default() },
|
||||
),
|
||||
},
|
||||
],
|
||||
};
|
||||
// Rendering scales each child rectangle before applying its own spin.
|
||||
let before = visual_ends(&group);
|
||||
let expected = if free_start { before.0 } else { before.1 };
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
let pos = s.doc.start_of(StoryRef::Body);
|
||||
s.doc.para_mut(pos.story, &pos.path).unwrap().insert_object(pos.off, group, &Default::default()).unwrap();
|
||||
assert!(reroute(&mut s));
|
||||
let group = s.doc.para_at(&pos).unwrap().object_at(pos.off).unwrap();
|
||||
let after = visual_ends(group);
|
||||
let actual = if free_start { after.0 } else { after.1 };
|
||||
assert!(
|
||||
(actual.0 - expected.0).abs() < 0.001 && (actual.1 - expected.1).abs() < 0.001,
|
||||
"{w}x{h}, {kind:?}, {rot}, {flip_h}, {flip_v}, {free_start}: {actual:?} != {expected:?}"
|
||||
);
|
||||
let InlineObject::Group { children, .. } = group else { panic!("group") };
|
||||
let c = &children[1];
|
||||
let InlineObject::Shape { w, h, float, .. } = &c.obj else { panic!("connector") };
|
||||
let routed = ends(c.x, c.y, *w, *h, float.flip_h, float.flip_v);
|
||||
assert_eq!(if free_start { routed.1 } else { routed.0 }, (140.0, 210.0), "resolved sites stay in child coordinates");
|
||||
assert_eq!(float.rot, 0.0);
|
||||
assert!(!reroute(&mut s), "rerouting must settle");
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -286,7 +286,7 @@ fn content_base(s: &Session, r: &ControlRange) -> CharProps {
|
||||
/// and stops showing it. Returns the formatting the text should get, when it changes.
|
||||
pub fn prepare_edit(s: &mut Session) -> Result<Option<CharProps>, CmdError> {
|
||||
if s.sel.is_collapsed() {
|
||||
let at = super::protect::typing_pos(s, &s.doc.typing_pos(&s.sel.focus));
|
||||
let at = s.doc.typing_pos(&s.sel.focus);
|
||||
if at != s.sel.focus {
|
||||
s.sel = Selection::caret(at);
|
||||
}
|
||||
@@ -864,25 +864,6 @@ mod editing_tests {
|
||||
assert_eq!(typed, None, "a placeholder style localized in id and name");
|
||||
}
|
||||
|
||||
/// Typed text with line breaks in it (`type_text` with paragraphs, a typed paste) is the
|
||||
/// Enter key pressed between the lines, so a one-line control, where Enter does nothing, stays
|
||||
/// one paragraph, and a multi-line one splits (#3973: the line breaks went in as one command
|
||||
/// and split a one-line control).
|
||||
#[test]
|
||||
fn typed_line_breaks_do_not_split_a_one_line_control() {
|
||||
for (multi_line, paragraphs) in [(false, 1), (true, 2)] {
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
run(&mut s, "text.insert", json!({"text": "Name: "}));
|
||||
run(&mut s, "developer.plainText", json!({"title": "Name", "tag": "name"}));
|
||||
run(&mut s, "developer.properties", json!({"control": "name", "multiLine": multi_line}));
|
||||
run(&mut s, "caret.right", json!({}));
|
||||
run(&mut s, "text.insert", json!({"text": "Ada\rLovelace", "raw": true}));
|
||||
assert_eq!(s.doc.body.len(), paragraphs, "multi-line control: {multi_line}");
|
||||
let r = s.doc.control_at(&s.sel.focus).unwrap();
|
||||
assert_eq!(r.control.tag, "name", "still inside the control");
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn typing_replaces_the_placeholder_and_emptying_brings_it_back() {
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
//! Design tab: themes, style sets, colours, fonts, paragraph spacing, watermark, page colour, borders.
|
||||
|
||||
use serde_json::{Value, json};
|
||||
use wordcraft_doc::props::{Border, BorderStyle, Borders, CharProps, LineSpacing, Rgb, TextColor};
|
||||
use wordcraft_doc::props::{Border, BorderStyle, Borders, LineSpacing, Rgb, TextColor};
|
||||
use wordcraft_doc::{Watermark, styles};
|
||||
|
||||
use super::sel_result;
|
||||
@@ -67,8 +67,6 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
.params(r#"{"name": "default|basic|lines|shaded|casual|centered|minimalist|title"}"#),
|
||||
CommandSpec::new("design.paragraphSpacing", "Paragraph Spacing", "Design › Document Formatting", para_spacing)
|
||||
.params(r#"{"value": "default|none|compact|tight|open|relaxed|double"}"#),
|
||||
CommandSpec::new("design.scaleFontSizes", "Scale Font Sizes", "Design › Document Formatting", scale_font_sizes)
|
||||
.params(r#"{"delta": pt (-1638..1638, not 0) | "percent": number (-99..1000, not 0; 10 = +10%)} (exactly one)"#),
|
||||
CommandSpec::new("design.watermark", "Watermark", "Design › Page Background", watermark)
|
||||
.params(r#"{"text"?: string, "remove"?: bool, "diagonal"?: bool, "color"?: "RRGGBB"}"#),
|
||||
CommandSpec::new("design.pageColor", "Page Color", "Design › Page Background", |s, v| {
|
||||
@@ -171,19 +169,11 @@ fn style_set(s: &mut Session, v: &Value) -> CmdResult {
|
||||
_ => {}
|
||||
}
|
||||
}
|
||||
// Overlay the set onto the document's sheet: every other style, the latent styles and the
|
||||
// document defaults stay as they are (no set defines its own defaults).
|
||||
for mut st in fresh.styles {
|
||||
match s.doc.styles.styles.iter_mut().find(|x| x.id == st.id) {
|
||||
Some(old) => {
|
||||
st.default = old.default;
|
||||
st.locked = old.locked;
|
||||
st.builtin = old.builtin;
|
||||
*old = st;
|
||||
}
|
||||
None => s.doc.styles.styles.push(st),
|
||||
}
|
||||
}
|
||||
// Keep custom styles and fonts.
|
||||
let customs: Vec<_> = s.doc.styles.styles.iter().filter(|x| !x.builtin).cloned().collect();
|
||||
fresh.styles.extend(customs);
|
||||
fresh.default_chr.font = s.doc.styles.default_chr.font.clone();
|
||||
s.doc.styles = fresh;
|
||||
let (h, b) = (s.doc.settings.major_font.clone(), s.doc.settings.minor_font.clone());
|
||||
set_fonts(s, &h, &b);
|
||||
set_accent(s, accent);
|
||||
@@ -206,45 +196,6 @@ fn para_spacing(s: &mut Session, v: &Value) -> CmdResult {
|
||||
sel_result(s)
|
||||
}
|
||||
|
||||
/// Grow or shrink every style's font size at once. Only explicit style sizes change (including
|
||||
/// table style parts); styles that inherit one (via `based_on` or the defaults) follow, and runs
|
||||
/// keep their own direct formatting.
|
||||
fn scale_font_sizes(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let (delta, percent) = (p::f32(v, "delta"), p::f32(v, "percent"));
|
||||
// A field that is present but not a number is an error, not "absent".
|
||||
if (v.get("delta").is_some() && delta.is_none()) || (v.get("percent").is_some() && percent.is_none()) {
|
||||
return Err(CmdError::Params("`delta` and `percent` must be numbers".into()));
|
||||
}
|
||||
if delta.is_some() == percent.is_some() {
|
||||
return Err(CmdError::Params("give exactly one of `delta` (points) or `percent`".into()));
|
||||
}
|
||||
if delta.is_some_and(|d| !(-1638.0..=1638.0).contains(&d) || d == 0.0) || percent.is_some_and(|x| !(-99.0..=1000.0).contains(&x) || x == 0.0) {
|
||||
return Err(CmdError::Params("`delta` must be non-zero within ±1638 pt and `percent` non-zero within -99..1000".into()));
|
||||
}
|
||||
let factor = 1.0 + percent.unwrap_or(0.0) / 100.0;
|
||||
let scale = |size: f32| ((size * factor + delta.unwrap_or(0.0)).clamp(1.0, 1638.0) * 2.0).round() / 2.0;
|
||||
let scale_chr = |c: &mut CharProps| {
|
||||
c.size = c.size.map(scale);
|
||||
c.size_cs = c.size_cs.map(scale);
|
||||
};
|
||||
let st = &mut s.doc.styles;
|
||||
// An unset document default resolves to the fallback size; make it explicit so inheriting styles follow.
|
||||
st.default_chr.size = Some(scale(st.default_chr.size.unwrap_or(wordcraft_doc::props::DEFAULT_FONT_SIZE)));
|
||||
st.default_chr.size_cs = st.default_chr.size_cs.map(scale);
|
||||
for style in &mut st.styles {
|
||||
scale_chr(&mut style.chr);
|
||||
if let Some(t) = style.table.as_mut() {
|
||||
for c in [&mut t.header_chr, &mut t.band_chr, &mut t.first_col_chr, &mut t.last_col_chr, &mut t.total_chr, &mut t.col_band_chr] {
|
||||
scale_chr(c);
|
||||
}
|
||||
for part in t.regions.values_mut() {
|
||||
scale_chr(&mut part.chr);
|
||||
}
|
||||
}
|
||||
}
|
||||
sel_result(s)
|
||||
}
|
||||
|
||||
fn watermark(s: &mut Session, v: &Value) -> CmdResult {
|
||||
if p::bool(v, "remove").unwrap_or(false) {
|
||||
s.doc.settings.watermark = None;
|
||||
|
||||
@@ -532,16 +532,6 @@ fn keep_lasso(s: &mut Session, sel: LassoSelection) {
|
||||
s.view.draw.lasso = Some(LassoSelection { rev: s.rev().wrapping_add(1), ..sel });
|
||||
}
|
||||
|
||||
/// After a command that changed nothing, which puts the revision back from `touched` to `rev`: a
|
||||
/// lasso selection the command kept for the revision it would have reached stays selected.
|
||||
pub(crate) fn rekey_lasso(s: &mut Session, touched: u64, rev: u64) {
|
||||
if let Some(l) = s.view.draw.lasso.as_mut()
|
||||
&& (l.rev == touched || l.rev == touched.wrapping_add(1))
|
||||
{
|
||||
l.rev = rev;
|
||||
}
|
||||
}
|
||||
|
||||
/// Whether (`x`, `y`) is inside the closed loop `poly` (even–odd rule).
|
||||
fn inside(poly: &[[f32; 2]], x: f32, y: f32) -> bool {
|
||||
let n = poly.len();
|
||||
@@ -1091,15 +1081,6 @@ mod tests {
|
||||
let objs = ink(&s);
|
||||
let blue = objs.iter().filter(|o| matches!(o, InlineObject::Shape { stroke: Some(c), .. } if c.hex() == "0070C0")).count();
|
||||
assert_eq!(blue, 1);
|
||||
// Recolouring to the colour it has changes nothing, and the selection stays; an edit
|
||||
// elsewhere ends it.
|
||||
s.run("draw.lassoColor", &json!({"color": "0070C0"})).unwrap();
|
||||
assert_eq!(s.undo_depth(), steps + 2);
|
||||
assert_eq!(super::lasso_selection(&s).map(|l| l.items.clone()), Some(vec![sel.clone()]));
|
||||
s.run("text.insert", &json!({"text": "x"})).unwrap();
|
||||
assert!(super::lasso_selection(&s).is_none());
|
||||
let r = s.run("draw.lasso", &json!({"page": 0, "points": [[80, 80], [260, 80], [260, 180], [80, 180]]})).unwrap();
|
||||
assert_eq!(r["count"], 1);
|
||||
// Deleted: only the selected stroke goes, and the selection with it.
|
||||
s.run("draw.lassoDelete", &json!({})).unwrap();
|
||||
assert_eq!(ink(&s).len(), 1);
|
||||
|
||||
@@ -142,7 +142,7 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
.params(r#"{"tab"?: "find|replace|goto" (default replace)}"#)
|
||||
.pure(),
|
||||
CommandSpec::new("edit.replaceAll", "Replace All", "Home › Editing › Replace", replace_all)
|
||||
.params(r#"{"text": string, "with": string (^p ^t ^& ^c …; \\1 with wildcards; with regex also $1 ${n} ${name} $0 $$ \\1 \\n (paragraph mark) \\t), …options of edit.find}"#),
|
||||
.params(r#"{"text": string, "with": string (^p ^t ^& ^c …; \\1 with wildcards), …options of edit.find}"#),
|
||||
CommandSpec::new("edit.goto", "Go To", "Home › Editing › Find", goto)
|
||||
.key("Mod+Alt+G / F5")
|
||||
.params(r#"{"page"?: n, "pageBy"?: int (pages forward, negative back, from the caret's page), "bookmark"?: string (selects its text), "paragraph"?: n}; none opens Find and Replace on its Go To tab"#)
|
||||
@@ -459,10 +459,7 @@ pub(crate) fn search_full(s: &Session, story: StoryRef, with_rep: bool) -> Resul
|
||||
.configure(meta::Config::new().nfa_size_limit(Some(1 << 20)))
|
||||
.build(&c.pattern)
|
||||
.map_err(|e| CmdError::Params(format!("bad pattern: {e}")))?;
|
||||
let pieces = with_rep.then(|| {
|
||||
let index = |name: &str| re.group_info().to_index(regex_automata::PatternID::ZERO, name);
|
||||
find_pattern::resolve_names(find_pattern::parse_replace(&f.replace, &opts), &index)
|
||||
});
|
||||
let pieces = with_rep.then(|| find_pattern::parse_replace(&f.replace, &opts));
|
||||
let paths = s.doc.para_paths(story);
|
||||
// Paragraphs read as one text: consecutive siblings when the pattern spans paragraphs.
|
||||
let mut chunks: Vec<std::ops::Range<usize>> = Vec::new();
|
||||
|
||||
@@ -10,7 +10,7 @@ use wordcraft_doc::para::OBJ;
|
||||
use wordcraft_doc::{InlineObject, Pos};
|
||||
|
||||
use super::{delete_selection, sel_result, split_para};
|
||||
use crate::{CmdError, CmdResult, CommandSpec, ImeEditOutcome, MathEdit, Selection, Session, p};
|
||||
use crate::{CmdError, CmdResult, CommandSpec, MathEdit, Selection, Session, p};
|
||||
|
||||
fn editing(s: &Session) -> Option<&'static str> {
|
||||
if s.math.is_none() { Some("not editing an equation") } else { None }
|
||||
@@ -20,9 +20,6 @@ fn near_equation(s: &Session) -> Option<&'static str> {
|
||||
if s.math.is_some() || equation_at_caret(s).is_some() { None } else { Some("no equation at the caret") }
|
||||
}
|
||||
|
||||
/// Builds the equation command table: insert, edit and exit, typing, structures, symbols, caret
|
||||
/// movement, deletion (including bounded surrounding deletion), enter, conversion, input format,
|
||||
/// normal text, display, justification, numbering, set and get, gallery and structure actions.
|
||||
pub fn specs() -> Vec<CommandSpec> {
|
||||
vec![
|
||||
CommandSpec::new("insert.equation", "Equation", "Insert › Symbols", insert)
|
||||
@@ -43,9 +40,6 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
.params(r#"{"dir": "left|right|up|down|home|end|next|prev"}"#),
|
||||
CommandSpec::new("equation.backspace", "Delete Back in Equation", "Equation", |s, _| delete(s, true)).when(editing),
|
||||
CommandSpec::new("equation.delete", "Delete in Equation", "Equation", |s, _| delete(s, false)).when(editing),
|
||||
CommandSpec::new("equation.deleteSurrounding", "Delete Around Caret in Equation", "Equation", delete_surrounding)
|
||||
.when(editing)
|
||||
.params(r#"{"before": n, "after": n}"#),
|
||||
CommandSpec::new("equation.enter", "Enter in Equation", "Equation", enter).when(editing),
|
||||
CommandSpec::new("equation.convert", "Convert", "Equation › Conversions", convert)
|
||||
.when(near_equation_or_all)
|
||||
@@ -106,73 +100,21 @@ fn equation_at_caret(s: &Session) -> Option<Pos> {
|
||||
None
|
||||
}
|
||||
|
||||
/// What one equation edit changed. `content_changed` gates an IME receipt; `state_changed`
|
||||
/// covers style, representation and lazy-normalization changes that must poison a mixed frame.
|
||||
#[derive(Clone, Copy, Debug, Default)]
|
||||
struct EquationEffect {
|
||||
content_changed: bool,
|
||||
state_changed: bool,
|
||||
}
|
||||
|
||||
impl EquationEffect {
|
||||
/// Merges another edit's effect flags, so a change on either side marks the merged effect.
|
||||
fn merge(&mut self, other: Self) {
|
||||
self.content_changed |= other.content_changed;
|
||||
self.state_changed |= other.state_changed;
|
||||
}
|
||||
}
|
||||
|
||||
/// Records the edit as a classified IME outcome; any content change also counts as a state
|
||||
/// change, and `transformed` marks automatic rewrites.
|
||||
fn apply_effect(s: &mut Session, effect: EquationEffect, transformed: bool) {
|
||||
s.set_ime_edit_outcome(ImeEditOutcome {
|
||||
classified: true,
|
||||
content_changed: effect.content_changed,
|
||||
state_changed: effect.state_changed || effect.content_changed,
|
||||
transformed,
|
||||
selection_changed: false,
|
||||
});
|
||||
}
|
||||
|
||||
/// Records a classified IME outcome with no content or state change, distinguishing a real
|
||||
/// no-op from a command that never classified its edit.
|
||||
fn classified_noop(s: &mut Session) {
|
||||
s.set_ime_edit_outcome(ImeEditOutcome { classified: true, ..ImeEditOutcome::default() });
|
||||
}
|
||||
|
||||
/// Run `f` on the equation at `at`; keeps its linear text in step and drops the source OMML
|
||||
/// (it no longer matches).
|
||||
///
|
||||
/// The logical before-state is captured *after* lazy `nodes` materialization: an imported
|
||||
/// equation that only gets normalized is a state change, never an IME content change.
|
||||
fn with_eq<R>(s: &mut Session, at: &Pos, f: impl FnOnce(&mut Math, &mut bool) -> R) -> Result<(R, EquationEffect), CmdError> {
|
||||
fn with_eq<R>(s: &mut Session, at: &Pos, f: impl FnOnce(&mut Math, &mut bool) -> R) -> Result<R, CmdError> {
|
||||
let para = s.doc.para_mut(at.story, &at.path)?;
|
||||
let Some(InlineObject::Equation { linear, display, math }) = para.object_at_mut(at.off) else {
|
||||
return Err(CmdError::Failed("no equation there".into()));
|
||||
};
|
||||
let normalized = math.nodes.is_empty() && !linear.is_empty() && !math.linear;
|
||||
if normalized {
|
||||
if math.nodes.is_empty() && !linear.is_empty() && !math.linear {
|
||||
math.nodes = parse_linear(linear);
|
||||
}
|
||||
let nodes_before = math.nodes.clone();
|
||||
let linear_mode_before = math.linear;
|
||||
let jc_before = math.jc;
|
||||
let display_before = *display;
|
||||
let cached_linear_before = linear.clone();
|
||||
let omml_before_nonempty = !math.omml.is_empty();
|
||||
let r = f(math, display);
|
||||
let new_linear = if math.linear { plain(&math.nodes) } else { to_linear(&math.nodes) };
|
||||
let content_changed = math.nodes != nodes_before || math.linear != linear_mode_before;
|
||||
let state_changed = normalized
|
||||
|| content_changed
|
||||
|| math.jc != jc_before
|
||||
|| *display != display_before
|
||||
|| omml_before_nonempty
|
||||
|| new_linear != cached_linear_before;
|
||||
*linear = new_linear;
|
||||
*linear = if math.linear { plain(&math.nodes) } else { to_linear(&math.nodes) };
|
||||
math.omml.clear();
|
||||
para.touch();
|
||||
Ok((r, EquationEffect { content_changed, state_changed }))
|
||||
Ok(r)
|
||||
}
|
||||
|
||||
/// All the text of an equation's runs (its linear form while shown linear).
|
||||
@@ -221,9 +163,6 @@ fn insert(s: &mut Session, v: &Value) -> CmdResult {
|
||||
insert_nodes_as_equation(s, nodes, v)
|
||||
}
|
||||
|
||||
/// Clears any editing session, deletes the selection, creates an equation object (display on an
|
||||
/// empty line unless the request overrides it), optionally opens editing at the first
|
||||
/// placeholder, and reports a content change.
|
||||
fn insert_nodes_as_equation(s: &mut Session, nodes: Arg, v: &Value) -> CmdResult {
|
||||
s.math = None;
|
||||
let props = s.typing_props();
|
||||
@@ -239,27 +178,14 @@ fn insert_nodes_as_equation(s: &mut Session, nodes: Arg, v: &Value) -> CmdResult
|
||||
let obj = InlineObject::Equation { linear, display, math: Math { nodes, ..Default::default() } };
|
||||
let end = s.doc.insert_object(&at, obj, &props)?;
|
||||
if p::bool(v, "edit").unwrap_or(true) {
|
||||
let Some(session) = crate::next_math_session() else {
|
||||
return Err(CmdError::Failed("no equation editing session available".into()));
|
||||
};
|
||||
s.sel = Selection::caret(at.clone());
|
||||
s.math = Some(MathEdit { at, pos: caret, session });
|
||||
s.math = Some(MathEdit { at, pos: caret });
|
||||
} else {
|
||||
s.sel = Selection::caret(end);
|
||||
}
|
||||
// Creating an equation inserts visible content, even when its tree is initially empty.
|
||||
s.set_ime_edit_outcome(ImeEditOutcome {
|
||||
classified: true,
|
||||
content_changed: true,
|
||||
state_changed: true,
|
||||
transformed: false,
|
||||
selection_changed: false,
|
||||
});
|
||||
result(s)
|
||||
}
|
||||
|
||||
/// Starts editing the equation at the caret, materializing lazy nodes and clamping the requested
|
||||
/// position, and returns the equation and caret.
|
||||
fn edit(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let at = equation_at_caret(s).ok_or_else(|| CmdError::Disabled("no equation at the caret".into()))?;
|
||||
let nodes = match s.doc.para_at(&at).and_then(|p| p.object_at(at.off)) {
|
||||
@@ -274,11 +200,8 @@ fn edit(s: &mut Session, v: &Value) -> CmdResult {
|
||||
};
|
||||
let pos =
|
||||
v.get("pos").and_then(|p| serde_json::from_value::<MathPos>(p.clone()).ok()).unwrap_or_else(|| MathPos::new(Vec::new(), me::units(&nodes)));
|
||||
let Some(session) = crate::next_math_session() else {
|
||||
return Err(CmdError::Failed("no equation editing session available".into()));
|
||||
};
|
||||
s.sel = Selection::caret(at.clone());
|
||||
s.math = Some(MathEdit { pos: me::clamp(&nodes, &pos), at, session });
|
||||
s.math = Some(MathEdit { pos: me::clamp(&nodes, &pos), at });
|
||||
result(s)
|
||||
}
|
||||
|
||||
@@ -295,30 +218,20 @@ fn current(s: &Session) -> Result<MathEdit, CmdError> {
|
||||
s.math.clone().ok_or_else(|| CmdError::Disabled("not editing an equation".into()))
|
||||
}
|
||||
|
||||
/// Caps typing at `MAX_EQUATION_TYPE_CHARS`; empty text is a classified no-op. Linear form
|
||||
/// inserts literal runs, while built-up form applies AutoCorrect and build-up before spaces and
|
||||
/// relational characters.
|
||||
fn type_text(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let text = p::req_str(v, "text")?.to_string();
|
||||
let m = current(s)?;
|
||||
if text.is_empty() {
|
||||
// No normalization, no linear-cache rewrite, no revision: an empty commit is a no-op.
|
||||
classified_noop(s);
|
||||
return result(s);
|
||||
}
|
||||
let latex = s.math_latex;
|
||||
let style = TypeStyle { nor: s.math_normal_text };
|
||||
let mut pos = m.pos.clone();
|
||||
let mut transformed = false;
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| {
|
||||
with_eq(s, &m.at, |math, _| {
|
||||
if math.linear {
|
||||
// Linear form is plain text: no build-up.
|
||||
let text: String = text.chars().take(crate::MAX_EQUATION_TYPE_CHARS).collect();
|
||||
pos = me::insert_text(&mut math.nodes, &pos, &text, TypeStyle::default());
|
||||
set_literal(&mut math.nodes);
|
||||
return;
|
||||
}
|
||||
for c in text.chars().take(crate::MAX_EQUATION_TYPE_CHARS) {
|
||||
for c in text.chars().take(10_000) {
|
||||
let rel = matches!(wordcraft_doc::math::math_class(c), wordcraft_doc::math::MClass::Rel);
|
||||
if !style.nor && (c == ' ' || c == '\t' || rel) {
|
||||
let mut changed = false;
|
||||
@@ -330,9 +243,6 @@ fn type_text(s: &mut Session, v: &Value) -> CmdResult {
|
||||
pos = p;
|
||||
changed = true;
|
||||
}
|
||||
if changed {
|
||||
transformed = true;
|
||||
}
|
||||
if c == ' ' && changed {
|
||||
continue;
|
||||
}
|
||||
@@ -343,9 +253,9 @@ fn type_text(s: &mut Session, v: &Value) -> CmdResult {
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, transformed);
|
||||
result(s)
|
||||
}
|
||||
|
||||
/// Runs of a linear-form equation are literal text.
|
||||
fn set_literal(nodes: &mut Arg) {
|
||||
for n in nodes.iter_mut() {
|
||||
@@ -355,8 +265,6 @@ fn set_literal(nodes: &mut Arg) {
|
||||
}
|
||||
}
|
||||
|
||||
/// Inserts a gallery, linear or LaTeX structure: a new equation when none is being edited,
|
||||
/// literal text in linear form, or nodes at the caret in built-up form.
|
||||
fn insert_structure(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let nodes = nodes_from(v).ok_or_else(|| CmdError::Params("give an id, linear or latex".into()))?;
|
||||
let Some(m) = s.math.clone() else {
|
||||
@@ -364,7 +272,7 @@ fn insert_structure(s: &mut Session, v: &Value) -> CmdResult {
|
||||
return insert_nodes_as_equation(s, nodes, &json!({}));
|
||||
};
|
||||
let mut pos = m.pos.clone();
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| {
|
||||
with_eq(s, &m.at, |math, _| {
|
||||
if math.linear {
|
||||
pos = me::insert_text(&mut math.nodes, &pos, &to_linear(&nodes), TypeStyle::default());
|
||||
set_literal(&mut math.nodes);
|
||||
@@ -375,12 +283,9 @@ fn insert_structure(s: &mut Session, v: &Value) -> CmdResult {
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
/// Rejects an empty character; outside editing creates an equation, otherwise inserts the
|
||||
/// character at the math caret, marked literal in linear form.
|
||||
fn insert_symbol(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let c = p::req_str(v, "char")?.to_string();
|
||||
if c.is_empty() {
|
||||
@@ -391,7 +296,7 @@ fn insert_symbol(s: &mut Session, v: &Value) -> CmdResult {
|
||||
};
|
||||
let mut pos = m.pos.clone();
|
||||
let style = TypeStyle { nor: s.math_normal_text };
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| {
|
||||
with_eq(s, &m.at, |math, _| {
|
||||
pos = me::insert_text(&mut math.nodes, &pos, &c, style);
|
||||
if math.linear {
|
||||
set_literal(&mut math.nodes);
|
||||
@@ -400,7 +305,6 @@ fn insert_symbol(s: &mut Session, v: &Value) -> CmdResult {
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
@@ -413,8 +317,6 @@ fn nodes_of(s: &Session, at: &Pos) -> Option<Arg> {
|
||||
}
|
||||
}
|
||||
|
||||
/// Validates the direction, commits an AutoCorrect or build-up triggered by leaving a slot as one
|
||||
/// undoable edit, moves or clamps the math caret, and exits past either end.
|
||||
fn move_caret(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let m = current(s)?;
|
||||
let dir = p::str(v, "dir").unwrap_or("right");
|
||||
@@ -422,49 +324,48 @@ fn move_caret(s: &mut Session, v: &Value) -> CmdResult {
|
||||
return Err(CmdError::Params(format!("unknown direction {dir}")));
|
||||
}
|
||||
// Leaving a slot finishes what was typed there (AutoCorrect names, build-up). That is worked
|
||||
// out on a local copy: plain navigation must leave the equation (and its source OMML)
|
||||
// untouched, and a real change is a document edit — undoable and marking the document dirty
|
||||
// (#446, #447). Any failure happens before the single commit, so no rollback is needed.
|
||||
// out on a copy: plain navigation must leave the equation (and its source OMML) untouched,
|
||||
// and a real change is a document edit — undoable and marking the document dirty (#446, #447).
|
||||
let shown_linear = matches!(
|
||||
s.doc.para_at(&m.at).and_then(|p| p.object_at(m.at.off)),
|
||||
Some(InlineObject::Equation { math, .. }) if math.linear
|
||||
);
|
||||
let mut nodes = nodes_of(s, &m.at).ok_or_else(|| CmdError::Failed("no equation there".into()))?;
|
||||
let mut pos = m.pos.clone();
|
||||
if !s.math_normal_text && !shown_linear {
|
||||
if !s.math_normal_text
|
||||
&& !shown_linear
|
||||
&& let Some(before) = nodes_of(s, &m.at)
|
||||
{
|
||||
let latex = s.math_latex;
|
||||
let before = nodes.clone();
|
||||
let mut candidate = nodes.clone();
|
||||
let mut candidate_pos = pos.clone();
|
||||
let mut nodes = before.clone();
|
||||
let mut pos = m.pos.clone();
|
||||
let mut changed = false;
|
||||
if let Some(p) = me::autocorrect(&mut candidate, &candidate_pos) {
|
||||
candidate_pos = p;
|
||||
if let Some(p) = me::autocorrect(&mut nodes, &pos) {
|
||||
pos = p;
|
||||
changed = true;
|
||||
}
|
||||
if let Some(p) = me::build_up(&mut candidate, &candidate_pos, latex) {
|
||||
candidate_pos = p;
|
||||
if let Some(p) = me::build_up(&mut nodes, &pos, latex) {
|
||||
pos = p;
|
||||
changed = true;
|
||||
}
|
||||
pos = candidate_pos;
|
||||
if changed && candidate != before {
|
||||
if changed && nodes != before {
|
||||
// `equation.move` is a pure command, so record the edit here.
|
||||
s.checkpoint("Build Up");
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| math.nodes = candidate.clone())?;
|
||||
apply_effect(s, effect, true);
|
||||
// `equation.move` is pure for ordinary navigation, so `Session::run` does not touch
|
||||
// the document for us. A build-up is a real edit: mark it dirty and refresh layout.
|
||||
with_eq(s, &m.at, |math, _| math.nodes = nodes)?;
|
||||
s.touch();
|
||||
nodes = candidate;
|
||||
if let Some(mm) = s.math.as_mut() {
|
||||
mm.pos = pos;
|
||||
}
|
||||
}
|
||||
}
|
||||
let m = current(s)?;
|
||||
let nodes = nodes_of(s, &m.at).ok_or_else(|| CmdError::Failed("no equation there".into()))?;
|
||||
let next = match dir {
|
||||
"left" => me::move_left(&nodes, &pos),
|
||||
"right" => me::move_right(&nodes, &pos),
|
||||
"up" => Some(me::move_vertical(&nodes, &pos, true).unwrap_or(pos.clone())),
|
||||
"down" => Some(me::move_vertical(&nodes, &pos, false).unwrap_or(pos.clone())),
|
||||
"home" => Some(me::home_end(&nodes, &pos, false)),
|
||||
"end" => Some(me::home_end(&nodes, &pos, true)),
|
||||
"next" | "prev" => Some(me::next_placeholder(&nodes, &pos, dir == "prev").unwrap_or(pos.clone())),
|
||||
"left" => me::move_left(&nodes, &m.pos),
|
||||
"right" => me::move_right(&nodes, &m.pos),
|
||||
"up" => Some(me::move_vertical(&nodes, &m.pos, true).unwrap_or(m.pos.clone())),
|
||||
"down" => Some(me::move_vertical(&nodes, &m.pos, false).unwrap_or(m.pos.clone())),
|
||||
"home" => Some(me::home_end(&nodes, &m.pos, false)),
|
||||
"end" => Some(me::home_end(&nodes, &m.pos, true)),
|
||||
"next" | "prev" => Some(me::next_placeholder(&nodes, &m.pos, dir == "prev").unwrap_or(m.pos.clone())),
|
||||
other => return Err(CmdError::Params(format!("unknown direction {other}"))),
|
||||
};
|
||||
match next {
|
||||
@@ -478,96 +379,34 @@ fn move_caret(s: &mut Session, v: &Value) -> CmdResult {
|
||||
None => exit(s, &json!({"before": dir == "left"})),
|
||||
}
|
||||
}
|
||||
/// With the caret in an empty equation, removes the whole equation as a content change;
|
||||
/// otherwise deletes backward or forward, updates the caret and records the effect.
|
||||
|
||||
fn delete(s: &mut Session, back: bool) -> CmdResult {
|
||||
let m = current(s)?;
|
||||
let nodes = nodes_of(s, &m.at).ok_or_else(|| CmdError::Failed("no equation there".into()))?;
|
||||
// An empty equation goes away entirely: removing the object is a content change.
|
||||
// An empty equation goes away entirely.
|
||||
if nodes.is_empty() {
|
||||
s.math = None;
|
||||
s.sel = Selection { anchor: m.at.clone(), focus: Pos { off: m.at.off + OBJ.len_utf8(), ..m.at.clone() } };
|
||||
let at = delete_selection(s)?;
|
||||
s.sel = Selection::caret(at);
|
||||
s.set_ime_edit_outcome(ImeEditOutcome {
|
||||
classified: true,
|
||||
content_changed: true,
|
||||
state_changed: true,
|
||||
transformed: false,
|
||||
selection_changed: false,
|
||||
});
|
||||
return sel_result(s);
|
||||
}
|
||||
let mut pos = m.pos.clone();
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| {
|
||||
with_eq(s, &m.at, |math, _| {
|
||||
pos = if back { me::delete_back(&mut math.nodes, &pos) } else { me::delete_forward(&mut math.nodes, &pos) };
|
||||
})?;
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
/// Delete a bounded number of equation units on each side of the caret in one transaction.
|
||||
///
|
||||
/// The requested steps are simulated on a local tree and caret. Any step that cannot make
|
||||
/// progress refuses the whole request before the document is touched; the commit (if any) is a
|
||||
/// single command and a single undo step.
|
||||
fn delete_surrounding(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let m = current(s)?;
|
||||
let before = p::u64(v, "before").and_then(|n| usize::try_from(n).ok()).unwrap_or(usize::MAX);
|
||||
let after = p::u64(v, "after").and_then(|n| usize::try_from(n).ok()).unwrap_or(usize::MAX);
|
||||
if before.saturating_add(after) == 0 {
|
||||
return Err(CmdError::Params("nothing to delete".into()));
|
||||
}
|
||||
if before > crate::MAX_EQUATION_DELETE_UNITS || after > crate::MAX_EQUATION_DELETE_UNITS {
|
||||
return Err(CmdError::Params(format!("at most {} units on each side", crate::MAX_EQUATION_DELETE_UNITS)));
|
||||
}
|
||||
let mut nodes = nodes_of(s, &m.at).ok_or_else(|| CmdError::Failed("no equation there".into()))?;
|
||||
let mut pos = m.pos.clone();
|
||||
for _ in 0..before {
|
||||
let prev_nodes = nodes.clone();
|
||||
let prev_pos = pos.clone();
|
||||
pos = me::delete_back(&mut nodes, &pos);
|
||||
if nodes == prev_nodes {
|
||||
// Walking into the next argument is not a deletion. A requested unit that only
|
||||
// moves the caret must refuse the whole request, or a later step can apply against
|
||||
// a tree the platform's surrounding snapshot did not describe.
|
||||
return Err(CmdError::Failed(if pos == prev_pos {
|
||||
"the equation cannot delete that many units".into()
|
||||
} else {
|
||||
"the equation delete made no progress".into()
|
||||
}));
|
||||
}
|
||||
}
|
||||
for _ in 0..after {
|
||||
let prev_nodes = nodes.clone();
|
||||
let prev_pos = pos.clone();
|
||||
pos = me::delete_forward(&mut nodes, &pos);
|
||||
if nodes == prev_nodes {
|
||||
return Err(CmdError::Failed(if pos == prev_pos {
|
||||
"the equation cannot delete that many units".into()
|
||||
} else {
|
||||
"the equation delete made no progress".into()
|
||||
}));
|
||||
}
|
||||
}
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| math.nodes = nodes)?;
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
Ok(json!({"before": before, "after": after, "changed": true}))
|
||||
}
|
||||
/// In built-up form runs AutoCorrect, build-up and equation numbering; when nothing changes,
|
||||
/// leaves the equation and splits a new paragraph.
|
||||
fn enter(s: &mut Session, _: &Value) -> CmdResult {
|
||||
let m = current(s)?;
|
||||
let latex = s.math_latex;
|
||||
let mut pos = m.pos.clone();
|
||||
let mut changed = false;
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| {
|
||||
with_eq(s, &m.at, |math, _| {
|
||||
if math.linear {
|
||||
return;
|
||||
}
|
||||
@@ -588,7 +427,6 @@ fn enter(s: &mut Session, _: &Value) -> CmdResult {
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, true);
|
||||
return result(s);
|
||||
}
|
||||
// Nothing to build: Enter ends the equation and starts a new paragraph.
|
||||
@@ -596,13 +434,6 @@ fn enter(s: &mut Session, _: &Value) -> CmdResult {
|
||||
let after = Pos { off: m.at.off + OBJ.len_utf8(), ..m.at };
|
||||
let new = split_para(s, &after)?;
|
||||
s.sel = Selection::caret(new);
|
||||
s.set_ime_edit_outcome(ImeEditOutcome {
|
||||
classified: true,
|
||||
content_changed: true,
|
||||
state_changed: true,
|
||||
transformed: false,
|
||||
selection_changed: false,
|
||||
});
|
||||
sel_result(s)
|
||||
}
|
||||
|
||||
@@ -621,14 +452,11 @@ fn convert_math(math: &mut Math, professional: bool, latex: bool) {
|
||||
}
|
||||
}
|
||||
|
||||
/// Converts one equation or all equations between built-up and linear form, honoring LaTeX mode,
|
||||
/// merging every effect and resetting the math caret.
|
||||
fn convert(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let professional = p::str(v, "to").map(|t| !t.eq_ignore_ascii_case("linear")).unwrap_or(true);
|
||||
let latex = s.math_latex;
|
||||
if p::bool(v, "all").unwrap_or(false) {
|
||||
let mut n = 0;
|
||||
let mut effect = EquationEffect::default();
|
||||
let stories: Vec<wordcraft_doc::StoryRef> =
|
||||
std::iter::once(wordcraft_doc::StoryRef::Body).chain(s.doc.parts.keys().map(|k| wordcraft_doc::StoryRef::Part(*k))).collect();
|
||||
for story in stories {
|
||||
@@ -640,8 +468,7 @@ fn convert(s: &mut Session, v: &Value) -> CmdResult {
|
||||
.unwrap_or_default();
|
||||
for off in offs {
|
||||
let at = Pos { story, path: path.clone(), off };
|
||||
let (_, one) = with_eq(s, &at, |math, _| convert_math(math, professional, latex))?;
|
||||
effect.merge(one);
|
||||
with_eq(s, &at, |math, _| convert_math(math, professional, latex))?;
|
||||
n += 1;
|
||||
}
|
||||
}
|
||||
@@ -650,32 +477,26 @@ fn convert(s: &mut Session, v: &Value) -> CmdResult {
|
||||
m.pos = MathPos::default();
|
||||
}
|
||||
s.clamp_selection();
|
||||
apply_effect(s, effect, false);
|
||||
return Ok(json!({"converted": n}));
|
||||
}
|
||||
let at = equation_at_caret(s).ok_or_else(|| CmdError::Disabled("no equation at the caret".into()))?;
|
||||
let mut end = 0;
|
||||
let (_, effect) = with_eq(s, &at, |math, _| {
|
||||
with_eq(s, &at, |math, _| {
|
||||
convert_math(math, professional, latex);
|
||||
end = me::units(&math.nodes);
|
||||
})?;
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = MathPos::new(Vec::new(), end);
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
/// Toggles or sets the equation's inline or display mode.
|
||||
fn set_display(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let at = equation_at_caret(s).ok_or_else(|| CmdError::Disabled("no equation at the caret".into()))?;
|
||||
let (_, effect) = with_eq(s, &at, |_, display| *display = p::bool(v, "value").unwrap_or(!*display))?;
|
||||
apply_effect(s, effect, false);
|
||||
with_eq(s, &at, |_, display| *display = p::bool(v, "value").unwrap_or(!*display))?;
|
||||
result(s)
|
||||
}
|
||||
|
||||
/// Maps `jc` to the math justification, with unknown values falling back to center-group, and
|
||||
/// forces display mode.
|
||||
fn justify(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let at = equation_at_caret(s).ok_or_else(|| CmdError::Disabled("no equation at the caret".into()))?;
|
||||
let jc = match p::str(v, "jc").unwrap_or("centerGroup") {
|
||||
@@ -684,11 +505,10 @@ fn justify(s: &mut Session, v: &Value) -> CmdResult {
|
||||
"center" => MathJc::Center,
|
||||
_ => MathJc::CenterGroup,
|
||||
};
|
||||
let (_, effect) = with_eq(s, &at, |math, display| {
|
||||
with_eq(s, &at, |math, display| {
|
||||
math.jc = jc;
|
||||
*display = true;
|
||||
})?;
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
@@ -696,7 +516,7 @@ fn justify(s: &mut Session, v: &Value) -> CmdResult {
|
||||
fn number(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let at = equation_at_caret(s).ok_or_else(|| CmdError::Disabled("no equation at the caret".into()))?;
|
||||
let mut pos_reset = false;
|
||||
let (_, effect) = with_eq(s, &at, |math, display| {
|
||||
with_eq(s, &at, |math, display| {
|
||||
let numbered = matches!(math.nodes.as_slice(), [MNode::EqArr { rows }] if rows.iter().any(|r| wordcraft_doc::math::has_number_mark(r)));
|
||||
let want = p::bool(v, "value").unwrap_or(!numbered);
|
||||
if want && !numbered {
|
||||
@@ -719,7 +539,6 @@ fn number(s: &mut Session, v: &Value) -> CmdResult {
|
||||
m.pos = MathPos::default();
|
||||
}
|
||||
s.clamp_selection();
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
@@ -743,20 +562,17 @@ fn without_number(row: Arg) -> Arg {
|
||||
out
|
||||
}
|
||||
|
||||
/// Replaces the equation's nodes from linear or LaTeX input, forces built-up form, and puts the
|
||||
/// caret at the new end.
|
||||
fn set(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let at = equation_at_caret(s).ok_or_else(|| CmdError::Disabled("no equation at the caret".into()))?;
|
||||
let nodes = nodes_from(v).ok_or_else(|| CmdError::Params("give linear or latex".into()))?;
|
||||
let end = me::units(&nodes);
|
||||
let (_, effect) = with_eq(s, &at, |math, _| {
|
||||
with_eq(s, &at, |math, _| {
|
||||
math.nodes = nodes;
|
||||
math.linear = false;
|
||||
})?;
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = MathPos::new(Vec::new(), end);
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
@@ -785,8 +601,6 @@ fn structure_actions(s: &mut Session, _: &Value) -> CmdResult {
|
||||
Ok(json!({"actions": acts}))
|
||||
}
|
||||
|
||||
/// Applies a named structure action at an explicit or inferred level, refusing an action that
|
||||
/// does not apply at the caret.
|
||||
fn structure(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let m = current(s)?;
|
||||
let action = p::req_str(v, "action")?.to_string();
|
||||
@@ -800,12 +614,11 @@ fn structure(s: &mut Session, v: &Value) -> CmdResult {
|
||||
.ok_or_else(|| CmdError::Params(format!("{action} doesn't apply here")))?,
|
||||
};
|
||||
let mut pos = None;
|
||||
let (_, effect) = with_eq(s, &m.at, |math, _| pos = me::apply_structure(&mut math.nodes, &m.pos, level, &action))?;
|
||||
with_eq(s, &m.at, |math, _| pos = me::apply_structure(&mut math.nodes, &m.pos, level, &action))?;
|
||||
let pos = pos.ok_or_else(|| CmdError::Params(format!("{action} doesn't apply here")))?;
|
||||
if let Some(m) = s.math.as_mut() {
|
||||
m.pos = pos;
|
||||
}
|
||||
apply_effect(s, effect, false);
|
||||
result(s)
|
||||
}
|
||||
|
||||
@@ -1045,24 +858,14 @@ mod tests {
|
||||
assert!(document_xml(&s).contains(">bx<"));
|
||||
}
|
||||
|
||||
/// A Left move that builds up `a/b` makes the document dirty, advances layout and IME revisions,
|
||||
/// reports a classified transformed change, and Undo restores the literal run; plain navigation
|
||||
/// leaves no undo step.
|
||||
#[test]
|
||||
fn navigation_build_up_is_an_undoable_edit() {
|
||||
// #447: building `a/b` up on a Left move left the document clean and Undo did nothing.
|
||||
let mut s = imported("<m:r><m:t>a/b</m:t></m:r>");
|
||||
assert!(!s.dirty);
|
||||
let layout_rev = s.rev();
|
||||
let ime_rev = revision(&mut s);
|
||||
s.run("equation.move", &json!({"dir": "left"})).unwrap();
|
||||
// Every command replaces the last outcome; sample the edit before the read-only query.
|
||||
let outcome = s.ime_edit_outcome();
|
||||
assert_eq!(s.run("equation.get", &json!({})).unwrap()["latex"], "\\frac{a}{b}");
|
||||
assert!(s.dirty, "build-up marks the document dirty");
|
||||
assert!(s.rev() > layout_rev, "build-up refreshes the layout revision");
|
||||
assert!(outcome.classified && outcome.content_changed && outcome.state_changed && outcome.transformed, "{outcome:?}");
|
||||
assert!(revision(&mut s) > ime_rev, "build-up advances the unavailable IME revision");
|
||||
s.run("edit.undo", &json!({})).unwrap();
|
||||
s.run("equation.edit", &json!({})).unwrap();
|
||||
assert_eq!(s.run("equation.get", &json!({})).unwrap()["latex"], "a/b", "Undo restores the plain run");
|
||||
@@ -1138,239 +941,4 @@ mod tests {
|
||||
assert_eq!(subs, ["i", "j"], "{nodes:#?}");
|
||||
assert!(nodes.iter().all(|n| matches!(n, MNode::Script { .. }) || matches!(n, MNode::Run(r) if r.text.trim().is_empty())), "{nodes:#?}");
|
||||
}
|
||||
|
||||
/// Returns the session's current IME state revision, allocating a new one when the caret state changed.
|
||||
fn revision(s: &mut Session) -> u64 {
|
||||
s.ime_state_revision().expect("an IME state revision")
|
||||
}
|
||||
|
||||
/// Clears the stored node tree of the equation currently being edited, leaving its linear text
|
||||
/// and lazy state in place.
|
||||
fn clear_nodes(s: &mut Session) {
|
||||
let at = s.math.as_ref().expect("editing").at.clone();
|
||||
if let Some(InlineObject::Equation { math, .. }) = s.doc.para_mut(at.story, &at.path).expect("paragraph").object_at_mut(at.off) {
|
||||
math.nodes.clear();
|
||||
}
|
||||
}
|
||||
|
||||
/// Backspace at the start of an equation is classified with no content or state change,
|
||||
/// allocates no revision and leaves the linear text intact.
|
||||
#[test]
|
||||
fn equation_backspace_at_start_is_a_classified_noop() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "x+y"})).unwrap();
|
||||
s.run("equation.move", &json!({"dir": "home"})).unwrap();
|
||||
let before = revision(&mut s);
|
||||
s.run("equation.backspace", &json!({})).unwrap();
|
||||
let outcome = s.ime_edit_outcome();
|
||||
assert!(outcome.classified);
|
||||
assert!(!outcome.content_changed && !outcome.state_changed, "{outcome:?}");
|
||||
assert_eq!(revision(&mut s), before, "a no-op allocates no revision");
|
||||
assert_eq!(linear(&s), "x+y");
|
||||
}
|
||||
|
||||
/// Materializing an imported equation's lazy nodes is a state change only: the outcome is
|
||||
/// classified, content is unchanged, and the revision advances.
|
||||
#[test]
|
||||
fn lazy_normalization_is_state_not_content() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "x+1"})).unwrap();
|
||||
clear_nodes(&mut s);
|
||||
// Editing parses the local tree but leaves the stored equation lazy.
|
||||
s.run("equation.edit", &json!({})).unwrap();
|
||||
s.run("equation.move", &json!({"dir": "home"})).unwrap();
|
||||
let before = revision(&mut s);
|
||||
s.run("equation.backspace", &json!({})).unwrap();
|
||||
let outcome = s.ime_edit_outcome();
|
||||
assert!(outcome.classified);
|
||||
assert!(!outcome.content_changed, "the normalized tree was already logical content");
|
||||
assert!(outcome.state_changed, "materializing the cached nodes is a state change");
|
||||
assert!(revision(&mut s) > before, "state changes advance the revision");
|
||||
}
|
||||
|
||||
/// Empty typing is a classified no-op in built-up and linear form, without a new revision.
|
||||
#[test]
|
||||
fn empty_equation_type_is_a_classified_noop_in_both_modes() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "x"})).unwrap();
|
||||
let before = revision(&mut s);
|
||||
s.run("equation.type", &json!({"text": ""})).unwrap();
|
||||
assert!(s.ime_edit_outcome().classified);
|
||||
assert!(!s.ime_edit_outcome().content_changed && !s.ime_edit_outcome().state_changed);
|
||||
assert_eq!(revision(&mut s), before, "built-up empty type is a no-op");
|
||||
s.run("equation.convert", &json!({"to": "linear"})).unwrap();
|
||||
let before = revision(&mut s);
|
||||
s.run("equation.type", &json!({"text": ""})).unwrap();
|
||||
assert!(!s.ime_edit_outcome().content_changed && !s.ime_edit_outcome().state_changed);
|
||||
assert_eq!(revision(&mut s), before, "linear empty type is a no-op");
|
||||
}
|
||||
|
||||
/// Typing `a/b ` or `\alpha ` reports both a content change and a transformed automatic rewrite.
|
||||
#[test]
|
||||
fn automatic_equation_rewrites_are_transformed() {
|
||||
for text in ["a/b ", "\\alpha "] {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({})).unwrap();
|
||||
s.run("equation.type", &json!({"text": text})).unwrap();
|
||||
let outcome = s.ime_edit_outcome();
|
||||
assert!(outcome.content_changed, "{text}: {outcome:?}");
|
||||
assert!(outcome.transformed, "{text}: {outcome:?}");
|
||||
}
|
||||
}
|
||||
|
||||
/// Justification and display changes advance the revision and report a state, not content,
|
||||
/// change.
|
||||
#[test]
|
||||
fn equation_style_changes_are_state_changes_without_content() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "x"})).unwrap();
|
||||
let before = revision(&mut s);
|
||||
s.run("equation.justify", &json!({"jc": "left"})).unwrap();
|
||||
let outcome = s.ime_edit_outcome();
|
||||
assert!(outcome.classified && outcome.state_changed && !outcome.content_changed, "{outcome:?}");
|
||||
assert!(revision(&mut s) > before);
|
||||
let before = revision(&mut s);
|
||||
s.run("equation.display", &json!({"value": false})).unwrap();
|
||||
assert!(s.ime_edit_outcome().state_changed && !s.ime_edit_outcome().content_changed);
|
||||
assert!(revision(&mut s) > before);
|
||||
}
|
||||
|
||||
/// Converting every equation reports one classified effect that merges their content and state
|
||||
/// changes.
|
||||
#[test]
|
||||
fn convert_all_merges_every_equations_effect() {
|
||||
let mut s = session();
|
||||
s.run("document.setText", &json!({"text": "\r"})).unwrap();
|
||||
s.run("insert.equation", &json!({"linear": "a/b"})).unwrap();
|
||||
s.run("equation.convert", &json!({"to": "linear"})).unwrap();
|
||||
s.run("equation.exit", &json!({})).unwrap();
|
||||
s.run("caret.docEnd", &json!({})).unwrap();
|
||||
s.run("insert.equation", &json!({"linear": "c/d"})).unwrap();
|
||||
s.run("equation.exit", &json!({})).unwrap();
|
||||
s.run("equation.convert", &json!({"to": "professional", "all": true})).unwrap();
|
||||
let outcome = s.ime_edit_outcome();
|
||||
assert!(outcome.classified && outcome.content_changed && outcome.state_changed, "{outcome:?}");
|
||||
}
|
||||
|
||||
/// A bounded deletion around the caret removes the requested units in one undo step and reports
|
||||
/// a content change.
|
||||
#[test]
|
||||
fn equation_delete_surrounding_is_one_transaction_and_one_undo_step() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "abcd"})).unwrap();
|
||||
let depth = s.undo_depth();
|
||||
s.run("equation.deleteSurrounding", &json!({"before": 2, "after": 0})).unwrap();
|
||||
assert_eq!(linear(&s), "ab");
|
||||
assert_eq!(s.undo_depth(), depth + 1, "one undo step, not one per unit");
|
||||
assert!(s.ime_edit_outcome().classified && s.ime_edit_outcome().content_changed);
|
||||
}
|
||||
|
||||
/// An impossible bounded deletion applies no prefix, adds no undo step, keeps the session and
|
||||
/// caret, and leaves no classified outcome.
|
||||
#[test]
|
||||
fn equation_delete_surrounding_refuses_partial_progress() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "ab"})).unwrap();
|
||||
let session_token = s.math.as_ref().unwrap().session;
|
||||
let at = s.math.as_ref().unwrap().at.clone();
|
||||
let depth = s.undo_depth();
|
||||
let err = s.run("equation.deleteSurrounding", &json!({"before": 3, "after": 0})).unwrap_err();
|
||||
assert!(err.to_string().contains("cannot delete"), "{err}");
|
||||
assert_eq!(linear(&s), "ab", "the whole request is refused, no prefix applied");
|
||||
assert_eq!(s.undo_depth(), depth, "no undo step for a refused transaction");
|
||||
let m = s.math.as_ref().expect("still editing");
|
||||
assert_eq!(m.session, session_token);
|
||||
assert_eq!(m.at, at);
|
||||
assert!(!s.ime_edit_outcome().classified, "a refusal has no outcome");
|
||||
assert_eq!(
|
||||
s.last_command.as_ref().map(|(id, _)| id.as_str()),
|
||||
Some("equation.deleteSurrounding"),
|
||||
"only the pre-dispatch Repeat record remains"
|
||||
);
|
||||
}
|
||||
|
||||
/// Zero or oversized counts and deletions from an empty equation are refused without mutation;
|
||||
/// exactly deleting all text leaves the empty equation object.
|
||||
#[test]
|
||||
fn equation_delete_surrounding_bounds_and_empty_equations() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({})).unwrap();
|
||||
assert!(s.run("equation.deleteSurrounding", &json!({"before": 0, "after": 0})).is_err(), "zero counts are refused");
|
||||
assert!(s.run("equation.deleteSurrounding", &json!({"before": 1, "after": 0})).is_err(), "an empty equation refuses");
|
||||
assert!(s.math.is_some(), "the empty equation object stays");
|
||||
s.run("equation.type", &json!({"text": "a"})).unwrap();
|
||||
let depth = s.undo_depth();
|
||||
assert!(s.run("equation.deleteSurrounding", &json!({"before": 65, "after": 0})).is_err());
|
||||
assert!(s.run("equation.deleteSurrounding", &json!({"before": 0, "after": 65})).is_err());
|
||||
assert_eq!(linear(&s), "a");
|
||||
assert_eq!(s.undo_depth(), depth);
|
||||
s.run("equation.deleteSurrounding", &json!({"before": 1, "after": 0})).unwrap();
|
||||
assert_eq!(linear(&s), "", "exactly deleting the last unit leaves the equation object");
|
||||
assert!(s.math.is_some());
|
||||
}
|
||||
|
||||
/// A forward deletion that would only step into another argument is refused whole, leaving the
|
||||
/// text and undo depth unchanged (the caret is not asserted here).
|
||||
#[test]
|
||||
fn equation_delete_surrounding_refuses_a_caret_only_step() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"latex": "\\frac{a}{b}"})).unwrap();
|
||||
// At the end of the numerator a forward delete only steps into the denominator: the
|
||||
// tree does not change, so a request that would report progress is refused whole.
|
||||
s.math.as_mut().expect("editing").pos = MathPos { path: vec![(0, 0)], off: 1 };
|
||||
let before = linear(&s);
|
||||
let depth = s.undo_depth();
|
||||
let err = s.run("equation.deleteSurrounding", &json!({"before": 0, "after": 1})).unwrap_err();
|
||||
assert!(err.to_string().contains("no progress"), "{err}");
|
||||
assert_eq!(linear(&s), before);
|
||||
assert_eq!(s.undo_depth(), depth);
|
||||
}
|
||||
|
||||
/// A deletion that really erases on one side cannot buy a caret-only step on the other:
|
||||
/// the mixed request is refused whole, so a following IME commit cannot land in the
|
||||
/// argument the platform never asked about.
|
||||
#[test]
|
||||
fn equation_delete_surrounding_refuses_a_mixed_caret_only_step() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"latex": "\\frac{a}{b}"})).unwrap();
|
||||
// At the end of the numerator. `before: 1` erases `a`, but `after: 1` would only step
|
||||
// into the denominator (the report's mixed-direction reproducer).
|
||||
s.math.as_mut().expect("editing").pos = MathPos { path: vec![(0, 0)], off: 1 };
|
||||
let before = linear(&s);
|
||||
let pos = s.math.as_ref().expect("editing").pos.clone();
|
||||
let depth = s.undo_depth();
|
||||
let err = s.run("equation.deleteSurrounding", &json!({"before": 1, "after": 1})).unwrap_err();
|
||||
assert!(err.to_string().contains("no progress"), "{err}");
|
||||
assert_eq!(linear(&s), before, "the successful-looking prefix is not applied");
|
||||
assert_eq!(s.undo_depth(), depth, "no undo step for a refused mixed request");
|
||||
assert_eq!(s.math.as_ref().expect("editing").pos, pos, "the caret stays where the request found it");
|
||||
}
|
||||
|
||||
/// Typing more than the cap inserts exactly `MAX_EQUATION_TYPE_CHARS` characters in both linear
|
||||
/// and built-up form.
|
||||
#[test]
|
||||
fn equation_type_caps_linear_and_built_up_input() {
|
||||
for linear_mode in [false, true] {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({})).unwrap();
|
||||
if linear_mode {
|
||||
s.run("equation.convert", &json!({"to": "linear"})).unwrap();
|
||||
}
|
||||
let text = "a".repeat(crate::MAX_EQUATION_TYPE_CHARS + 5);
|
||||
s.run("equation.type", &json!({"text": text})).unwrap();
|
||||
assert_eq!(linear(&s).chars().count(), crate::MAX_EQUATION_TYPE_CHARS, "linear={linear_mode}");
|
||||
}
|
||||
}
|
||||
|
||||
/// Deleting one character forward removes it and moves the math caret to the start of the
|
||||
/// remaining text.
|
||||
#[test]
|
||||
fn equation_delete_surrounding_moves_the_caret_for_a_one_step_delete() {
|
||||
let mut s = session();
|
||||
s.run("insert.equation", &json!({"linear": "abc"})).unwrap();
|
||||
s.run("equation.move", &json!({"dir": "home"})).unwrap();
|
||||
s.run("equation.deleteSurrounding", &json!({"after": 1, "before": 0})).unwrap();
|
||||
assert_eq!(linear(&s), "bc");
|
||||
assert_eq!(s.math.as_ref().unwrap().pos, MathPos::new(Vec::new(), 0));
|
||||
}
|
||||
}
|
||||
|
||||
@@ -79,10 +79,6 @@ fn rename(s: &mut Session, v: &Value) -> CmdResult {
|
||||
Ok(json!({"name": name, "path": s.path}))
|
||||
}
|
||||
|
||||
/// Builds the file and document command table: rename, new, open, save, save as and encryption,
|
||||
/// PDF/PNG/print output and print settings, close, properties and info, author and privacy
|
||||
/// settings, and document inspection and text access. Recovery, compatibility, sharing, autosave,
|
||||
/// `file.newFromTemplate` and `file.saveTemplate` are registered in [`super::tools::specs`].
|
||||
pub fn specs() -> Vec<CommandSpec> {
|
||||
vec![
|
||||
CommandSpec::new("file.rename", "Rename Document", "File › Info", rename).params(r#"{"name": string (document name; its extension is kept, so the name may be given with or without it)}"#).pure(),
|
||||
@@ -141,11 +137,8 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
sel_result(s)
|
||||
})
|
||||
.key("Mod+W / Mod+F4")
|
||||
.params(r#"{"force"?: bool (desktop app's control channel and MCP bridge only: without it a document with unsaved changes is an error and stays open; with it the changes are discarded. Headless sessions never ask; file.closeDocument closes a document there)}"#)
|
||||
.params(r#"{"force"?: bool (desktop app's control channel and MCP bridge only: without it a document with unsaved changes is an error and stays open; with it the changes are discarded. Headless sessions never ask)}"#)
|
||||
.pure(),
|
||||
CommandSpec::new("file.closeDocument", "Close Document", "Agents", close_document)
|
||||
.params(r#"{"id"?: n (an open document, see view.switchWindows; default the active one), "force"?: bool (discard unsaved changes; without it a document with unsaved changes is an error and stays open)} → {"closed": id, "active": id, "documents": [{"id", "title", "path", "dirty", "active"}]}; closing the last document leaves a new blank one"#)
|
||||
.pure(),
|
||||
CommandSpec::new("file.properties", "Properties", "File › Info", properties)
|
||||
.params(r#"{"title"?, "subject"?, "author"?, "keywords"?, "comments"?, "category"?, "custom"?: {name: string | null (removes)}}"#),
|
||||
CommandSpec::new("file.info", "Info", "File", info).pure(),
|
||||
@@ -195,10 +188,6 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
let text = p::req_str(v, "text")?;
|
||||
let d = Document::from_text(text);
|
||||
s.doc.body = d.body;
|
||||
// The body is new even though the document id stays: a live IME composition (or an
|
||||
// equation editing session at the same offset) belongs to the old content.
|
||||
s.note_body_replaced();
|
||||
s.math = None;
|
||||
s.sel = crate::Selection::caret(s.doc.start_of(StoryRef::Body));
|
||||
sel_result(s)
|
||||
})
|
||||
@@ -211,8 +200,7 @@ fn privacy(s: &Session) -> Value {
|
||||
json!({"removePersonalInfo": s.doc.settings.remove_personal_info, "removeDateTime": s.doc.settings.remove_date_time})
|
||||
}
|
||||
|
||||
/// Open a new document beside the open ones and make it active: a blank one in the saved
|
||||
/// proofing language, or the requested template.
|
||||
/// Create a blank document in the saved proofing language, or use the requested template.
|
||||
fn new(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let doc = match p::str(v, "template").unwrap_or("blank") {
|
||||
"sample" => crate::sample::sample_document(),
|
||||
@@ -221,13 +209,14 @@ fn new(s: &mut Session, v: &Value) -> CmdResult {
|
||||
"report" => crate::sample::report(),
|
||||
_ => normal_document(s),
|
||||
};
|
||||
s.add_document(doc);
|
||||
s.set_document(doc);
|
||||
s.path = None;
|
||||
sel_result(s)
|
||||
}
|
||||
|
||||
/// A blank document on the Normal template: the region's paper (#838), written in the editing
|
||||
/// language (the templates' text is English).
|
||||
pub(crate) fn normal_document(s: &Session) -> Document {
|
||||
fn normal_document(s: &Session) -> Document {
|
||||
let mut d = s.blank_document();
|
||||
if let Some(lang) = s.prefs.default_language.as_ref().or(s.editing_lang.as_ref()) {
|
||||
d.styles.default_chr.lang = Some(lang.clone());
|
||||
@@ -235,22 +224,12 @@ pub(crate) fn normal_document(s: &Session) -> Document {
|
||||
d
|
||||
}
|
||||
|
||||
/// Open a document, using the saved proofing default only for plain text and Markdown. A file
|
||||
/// that is already open becomes the active document instead (`"alreadyOpen": true`). The file
|
||||
/// replaces an untouched blank document; otherwise it opens beside the open documents.
|
||||
/// Open a document, using the saved proofing default only for plain text and Markdown.
|
||||
fn open(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let Some(path) = p::str(v, "path") else {
|
||||
s.push_ui_request(json!({"open": "openFile"}));
|
||||
return sel_result(s);
|
||||
};
|
||||
if p::str(v, "data").is_none()
|
||||
&& let Some(id) = s.find_open(std::path::Path::new(path))
|
||||
{
|
||||
s.activate(id)?;
|
||||
return Ok(
|
||||
json!({"path": s.path, "paragraphs": s.doc.paragraph_count(), "words": s.doc.word_count(), "encrypted": s.password.is_some(), "alreadyOpen": true, "id": id}),
|
||||
);
|
||||
}
|
||||
let password = p::str(v, "password");
|
||||
let (mut doc, encrypted) = if let Some(data) = p::str(v, "data") {
|
||||
let bytes = super::insert::base64_decode(data).ok_or_else(|| CmdError::Params("bad base64".into()))?;
|
||||
@@ -267,19 +246,13 @@ fn open(s: &mut Session, v: &Value) -> CmdResult {
|
||||
crate::io::open_path_in(std::path::Path::new(path), password, &s.region()).map_err(CmdError::Failed)?
|
||||
};
|
||||
// Plain text and Markdown carry no language: they are in the editing language.
|
||||
let ext = std::path::Path::new(path).extension().and_then(std::ffi::OsStr::to_str).unwrap_or("").to_ascii_lowercase();
|
||||
let ext = std::path::Path::new(path).extension().and_then(|e| e.to_str()).unwrap_or("").to_ascii_lowercase();
|
||||
if let Some(lang) = s.prefs.default_language.as_ref().or(s.editing_lang.as_ref())
|
||||
&& matches!(ext.as_str(), "txt" | "text" | "md" | "markdown")
|
||||
{
|
||||
doc.styles.default_chr.lang = Some(lang.clone());
|
||||
}
|
||||
match s.used_because() {
|
||||
None => s.set_document(doc),
|
||||
Some(why) => {
|
||||
log::info!("file.open: opened beside document {}, not in its place: {why}", s.document_id());
|
||||
s.add_opened_document(doc);
|
||||
}
|
||||
}
|
||||
s.set_document(doc);
|
||||
if let Some(z) = s.doc.settings.zoom {
|
||||
super::view::restore_zoom(s, z);
|
||||
}
|
||||
@@ -288,26 +261,7 @@ fn open(s: &mut Session, v: &Value) -> CmdResult {
|
||||
if encrypted {
|
||||
s.password = password.map(crate::io::Password::new);
|
||||
}
|
||||
Ok(
|
||||
json!({"path": path, "paragraphs": s.doc.paragraph_count(), "words": s.doc.word_count(), "encrypted": encrypted, "alreadyOpen": false, "id": s.document_id()}),
|
||||
)
|
||||
}
|
||||
|
||||
/// Close a document with no window to ask about it (agents, headless sessions).
|
||||
fn close_document(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let id = match v.get("id").filter(|x| !x.is_null()) {
|
||||
None => s.document_id(),
|
||||
Some(x) => x.as_u64().ok_or_else(|| CmdError::Params("`id` must be an open document's id".into()))?,
|
||||
};
|
||||
if !s.is_open(id) {
|
||||
return Err(CmdError::Params(format!("no open document has id {id}")));
|
||||
}
|
||||
let dirty = s.documents().iter().any(|d| d.id == id && d.dirty);
|
||||
if dirty && p::bool(v, "force") != Some(true) {
|
||||
return Err(CmdError::Failed("the document has unsaved changes: save it first, or pass \"force\": true to discard them".into()));
|
||||
}
|
||||
s.close_document(id)?;
|
||||
Ok(json!({"closed": id, "active": s.document_id(), "documents": s.documents()}))
|
||||
Ok(json!({"path": path, "paragraphs": s.doc.paragraph_count(), "words": s.doc.word_count(), "encrypted": encrypted}))
|
||||
}
|
||||
|
||||
/// File › Info › Protect Document › Encrypt with Password: the password Word documents are saved
|
||||
@@ -462,25 +416,15 @@ fn properties(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let Some(m) = custom.as_object() else {
|
||||
return Err(CmdError::Params("`custom` must be an object of name → string or null".into()));
|
||||
};
|
||||
// Removals first: a case-only rename sends the old spelling as null and the new one
|
||||
// with its value, and `set_custom_prop` matches names case-insensitively, so setting
|
||||
// first would keep the old spelling (object keys are sorted, either spelling may lead).
|
||||
for (name, val) in m {
|
||||
if val.is_null() {
|
||||
s.doc.remove_custom_prop(name);
|
||||
if name.trim().is_empty() || name.chars().count() > MAX_CUSTOM_NAME {
|
||||
return Err(CmdError::Params(format!("custom property names are 1–{MAX_CUSTOM_NAME} characters")));
|
||||
}
|
||||
}
|
||||
for (name, val) in m {
|
||||
match val {
|
||||
// Removals took the name as stored above: a file opened from an older Word may
|
||||
// hold one longer than today's limit, and it must still be deletable (#2982).
|
||||
Value::Null => {}
|
||||
Value::String(x) if x.len() <= MAX_CUSTOM_VALUE => {
|
||||
if name.trim().is_empty() || name.chars().count() > MAX_CUSTOM_NAME {
|
||||
return Err(CmdError::Params(format!("custom property names are 1–{MAX_CUSTOM_NAME} characters")));
|
||||
}
|
||||
s.doc.set_custom_prop(name, x);
|
||||
Value::Null => {
|
||||
s.doc.remove_custom_prop(name);
|
||||
}
|
||||
Value::String(x) if x.len() <= MAX_CUSTOM_VALUE => s.doc.set_custom_prop(name, x),
|
||||
_ => return Err(CmdError::Params(format!("custom property `{name}` must be a string (≤ {MAX_CUSTOM_VALUE} bytes) or null"))),
|
||||
}
|
||||
}
|
||||
@@ -492,12 +436,10 @@ fn properties(s: &mut Session, v: &Value) -> CmdResult {
|
||||
Ok(out)
|
||||
}
|
||||
|
||||
/// Longest custom property name (Word's limit). The Custom Properties dialog validates with it
|
||||
/// too (#2982).
|
||||
pub const MAX_CUSTOM_NAME: usize = 255;
|
||||
/// Largest custom property value we accept from a command. The Custom Properties dialog
|
||||
/// validates with it too (#2982).
|
||||
pub const MAX_CUSTOM_VALUE: usize = 1024 * 1024;
|
||||
/// Longest custom property name (Word's limit).
|
||||
const MAX_CUSTOM_NAME: usize = 255;
|
||||
/// Largest custom property value we accept from a command.
|
||||
const MAX_CUSTOM_VALUE: usize = 1024 * 1024;
|
||||
|
||||
fn info(s: &mut Session, _: &Value) -> CmdResult {
|
||||
let l = s.layout();
|
||||
|
||||
@@ -345,20 +345,15 @@ pub enum Piece {
|
||||
Lit(String),
|
||||
/// `^&`: the text found.
|
||||
Found,
|
||||
/// `\n` with wildcards, or `$n` / `${n}` / `\n` with a regex: group `n` (0 is the match).
|
||||
/// `\n` with wildcards: group `n`.
|
||||
Group(usize),
|
||||
/// `${name}` / `$name` with a regex: a named group, resolved to a [`Piece::Group`] once the
|
||||
/// pattern is built ([`resolve_names`]); one left unresolved expands to nothing.
|
||||
Named(String),
|
||||
/// `^c`: the clipboard.
|
||||
Clip,
|
||||
}
|
||||
|
||||
/// Read Replace with text: `^` codes (unless `literal`), `\1`–`\9` with wildcards, and with a
|
||||
/// regex also `$n`, `${n}`, `${name}`, `$name`, `$$`, `\n` (paragraph mark), `\t` and `\\`.
|
||||
/// Read Replace with text: `^` codes (unless `literal`), and `\1`–`\9` with wildcards.
|
||||
pub fn parse_replace(with: &str, o: &Opts) -> Vec<Piece> {
|
||||
// With a regex, `special: false` turns off only the `^` codes; group references still expand.
|
||||
if o.literal && !o.regex {
|
||||
if o.literal || o.regex {
|
||||
return vec![Piece::Lit(with.to_string())];
|
||||
}
|
||||
let chars: Vec<char> = with.chars().collect();
|
||||
@@ -372,22 +367,6 @@ pub fn parse_replace(with: &str, o: &Opts) -> Vec<Piece> {
|
||||
let mut i = 0;
|
||||
while let Some(&c) = chars.get(i) {
|
||||
i += 1;
|
||||
if o.regex && (c == '$' || c == '\\') {
|
||||
match regex_ref(&chars, i, c) {
|
||||
Some((Ref::Char(ch), used)) => {
|
||||
cur.push(ch);
|
||||
i += used;
|
||||
}
|
||||
Some((Ref::Group(pc), used)) => {
|
||||
flush(&mut cur, &mut out);
|
||||
out.extend(pc);
|
||||
i += used;
|
||||
}
|
||||
// Not a reference or an escape: the character as written.
|
||||
None => cur.push(c),
|
||||
}
|
||||
continue;
|
||||
}
|
||||
if c == '\\' && o.wildcards {
|
||||
match chars.get(i) {
|
||||
Some(d) if d.is_ascii_digit() && *d != '0' => {
|
||||
@@ -405,7 +384,6 @@ pub fn parse_replace(with: &str, o: &Opts) -> Vec<Piece> {
|
||||
}
|
||||
}
|
||||
if c == '^'
|
||||
&& !o.literal
|
||||
&& let Some(&n) = chars.get(i)
|
||||
{
|
||||
if n == '&' || n == 'c' {
|
||||
@@ -431,67 +409,6 @@ pub fn parse_replace(with: &str, o: &Opts) -> Vec<Piece> {
|
||||
out
|
||||
}
|
||||
|
||||
/// What a `$` or `\` starts in regex Replace with text.
|
||||
enum Ref {
|
||||
/// An escape: `$$`, `\n`, `\t`, `\\`.
|
||||
Char(char),
|
||||
/// A group reference; `None` for a number too large to be a group (expands to nothing).
|
||||
Group(Option<Piece>),
|
||||
}
|
||||
|
||||
/// A group named by its number or name, as the `regex` crate reads it.
|
||||
fn group_ref(name: &str) -> Option<Piece> {
|
||||
if name.bytes().all(|b| b.is_ascii_digit()) { name.parse().ok().map(Piece::Group) } else { Some(Piece::Named(name.to_string())) }
|
||||
}
|
||||
|
||||
/// Read a regex reference or escape after `c` (`$` or `\`) at `chars[i..]`: what it is and how
|
||||
/// many characters it uses after `c`, or `None` when `c` is a plain character.
|
||||
fn regex_ref(chars: &[char], i: usize, c: char) -> Option<(Ref, usize)> {
|
||||
let next = *chars.get(i)?;
|
||||
if c == '\\' {
|
||||
return match next {
|
||||
'n' => Some((Ref::Char(PARA_MARK), 1)),
|
||||
't' => Some((Ref::Char('\t'), 1)),
|
||||
// `\$` and `\^` keep a `$` or `^` as typed (no group, no `^` code).
|
||||
'\\' | '$' | '^' => Some((Ref::Char(next), 1)),
|
||||
'1'..='9' => Some((Ref::Group(next.to_digit(10).map(|d| Piece::Group(d as usize))), 1)),
|
||||
_ => None,
|
||||
};
|
||||
}
|
||||
if next == '$' {
|
||||
return Some((Ref::Char('$'), 1));
|
||||
}
|
||||
let rest = chars.get(i..).unwrap_or_default();
|
||||
if next == '{' {
|
||||
// `${…}`: everything up to the closing brace; unclosed or empty, the `$` is literal.
|
||||
let close = rest.iter().position(|&ch| ch == '}')?;
|
||||
let name: String = rest.get(1..close)?.iter().collect();
|
||||
if name.is_empty() {
|
||||
return None;
|
||||
}
|
||||
return Some((Ref::Group(group_ref(&name)), close + 1));
|
||||
}
|
||||
// `$name`: the longest run of letters, digits and `_`.
|
||||
let len = rest.iter().take_while(|ch| ch.is_ascii_alphanumeric() || **ch == '_').count();
|
||||
if len == 0 {
|
||||
return None;
|
||||
}
|
||||
let name: String = rest.get(..len)?.iter().collect();
|
||||
Some((Ref::Group(group_ref(&name)), len))
|
||||
}
|
||||
|
||||
/// Turn named group references into numbered ones with `index` (a name's group number);
|
||||
/// an unknown name is dropped, so it expands to nothing.
|
||||
pub fn resolve_names(pieces: Vec<Piece>, index: &dyn Fn(&str) -> Option<usize>) -> Vec<Piece> {
|
||||
pieces
|
||||
.into_iter()
|
||||
.filter_map(|p| match p {
|
||||
Piece::Named(n) => index(&n).map(Piece::Group),
|
||||
p => Some(p),
|
||||
})
|
||||
.collect()
|
||||
}
|
||||
|
||||
/// Searched text as it is inserted again: marks become paragraph breaks.
|
||||
fn as_inserted(s: &str) -> String {
|
||||
s.replace(SECT_MARK, "\r")
|
||||
@@ -506,7 +423,6 @@ pub fn expand(pieces: &[Piece], found: &str, group: &dyn Fn(usize) -> Option<Str
|
||||
Piece::Found => out.push_str(&as_inserted(found)),
|
||||
Piece::Group(n) => out.push_str(&as_inserted(&group(*n).unwrap_or_default())),
|
||||
Piece::Clip => out.push_str(&clip.replace("\r\n", "\r").replace('\n', "\r")),
|
||||
Piece::Named(_) => {}
|
||||
}
|
||||
}
|
||||
out
|
||||
|
||||
@@ -43,15 +43,6 @@ fn read_font_file(path: &str) -> Result<Vec<u8>, CmdError> {
|
||||
crate::io::read_file_capped(path, wordcraft_fonts::MAX_FONT_FILE_BYTES as u64, "font file").map_err(CmdError::Failed)
|
||||
}
|
||||
|
||||
/// Adding or removing a font writes to the user font folder: automation confined to a write root
|
||||
/// (#769) may do that only when the folder is inside it.
|
||||
fn check_font_store(dir: Option<&std::path::Path>) -> Result<(), CmdError> {
|
||||
match dir {
|
||||
Some(dir) => crate::file_access::check_write(&dir.to_string_lossy()).map_err(CmdError::Failed),
|
||||
None => Ok(()),
|
||||
}
|
||||
}
|
||||
|
||||
fn add(s: &mut Session, v: &Value) -> CmdResult {
|
||||
if p::str(v, "url").is_some() {
|
||||
return Err(CmdError::Failed(
|
||||
@@ -80,7 +71,6 @@ fn add(s: &mut Session, v: &Value) -> CmdResult {
|
||||
s.push_ui_request(json!({"open": "addFont"}));
|
||||
return Ok(json!({"pending": true}));
|
||||
};
|
||||
check_font_store(FontDb::global().user_font_dir().as_deref())?;
|
||||
let r = FontDb::global().install_font(&name, &bytes).map_err(|e| CmdError::Failed(e.to_string()))?;
|
||||
// Text already set in the family (shown in a fallback until now) takes the new font.
|
||||
s.relayout();
|
||||
@@ -110,33 +100,8 @@ fn list(_: &mut Session, v: &Value) -> CmdResult {
|
||||
|
||||
fn remove(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let family = p::str(v, "family").filter(|f| !f.trim().is_empty()).ok_or_else(|| CmdError::Params("family is required".into()))?;
|
||||
check_font_store(FontDb::global().user_font_dir().as_deref())?;
|
||||
let gone = FontDb::global().remove_user_font(family.trim()).map_err(|e| CmdError::Failed(e.to_string()))?;
|
||||
s.relayout();
|
||||
s.status = format!("Removed {}", family.trim());
|
||||
Ok(json!({"removed": gone.iter().map(font_json).collect::<Vec<_>>()}))
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::*;
|
||||
|
||||
#[test]
|
||||
fn confined_automation_changes_only_a_font_store_inside_its_write_root() {
|
||||
let root = std::env::temp_dir().join(format!("wordcraft-font-store-{}", std::process::id()));
|
||||
let outside = std::env::temp_dir().join(format!("wordcraft-font-store-out-{}", std::process::id()));
|
||||
std::fs::create_dir_all(&root).unwrap();
|
||||
std::fs::create_dir_all(&outside).unwrap();
|
||||
let write = crate::file_access::AutomationRoots::new(None, Some(&root)).unwrap();
|
||||
let read_only = crate::file_access::AutomationRoots::new(Some(&root), None).unwrap();
|
||||
crate::file_access::confine(write.as_ref(), || {
|
||||
assert!(check_font_store(None).is_ok(), "no folder: the font lasts for the session");
|
||||
assert!(check_font_store(Some(&root.join("fonts"))).is_ok());
|
||||
assert!(check_font_store(Some(&outside)).is_err());
|
||||
});
|
||||
crate::file_access::confine(read_only.as_ref(), || assert!(check_font_store(Some(&root)).is_err()));
|
||||
assert!(check_font_store(Some(&outside)).is_ok(), "unconfined: the person at the keyboard");
|
||||
let _ = std::fs::remove_dir_all(&root);
|
||||
let _ = std::fs::remove_dir_all(&outside);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -13,7 +13,6 @@ use crate::{CmdError, CmdResult, CommandSpec, Session, p};
|
||||
/// Word's font size list.
|
||||
pub const SIZES: [f32; 17] = [8.0, 9.0, 10.0, 10.5, 11.0, 12.0, 14.0, 16.0, 18.0, 20.0, 22.0, 24.0, 26.0, 28.0, 36.0, 48.0, 72.0];
|
||||
|
||||
/// Register character-formatting commands, their parameters and keyboard bindings.
|
||||
pub fn specs() -> Vec<CommandSpec> {
|
||||
vec![
|
||||
CommandSpec::new("format.bold", "Bold", "Home › Font", |s, v| {
|
||||
@@ -108,7 +107,7 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
.params(r#"{"color": "RRGGBB" | "auto"} (the underline's own colour; "auto" follows the text colour)"#),
|
||||
CommandSpec::new("format.highlight", "Text Highlight Color", "Home › Font", highlight)
|
||||
.params(
|
||||
r#"{"color"?: "yellow|brightGreen|turquoise|pink|blue|red|darkBlue|teal|green|violet|darkRed|darkYellow|gray50|gray25|black|none"} (explicit colour sets it; omitted toggles the last chosen colour, initially yellow)"#,
|
||||
r#"{"color": "yellow|brightGreen|turquoise|pink|blue|red|darkBlue|teal|green|violet|darkRed|darkYellow|gray50|gray25|black|none"}"#,
|
||||
)
|
||||
// ⌘⌥H hides other apps on macOS.
|
||||
.key(super::non_mac("Mod+Alt+H")),
|
||||
@@ -229,8 +228,7 @@ pub fn apply(s: &mut Session, f: &dyn Fn(&mut CharProps)) -> CmdResult {
|
||||
sel_result(s)
|
||||
}
|
||||
|
||||
/// Resolved target formatting, sampled up to a paragraph/column-row boundary after 10,000 runs.
|
||||
/// Empty targets use pending caret formatting or the selected text box paragraph mark.
|
||||
/// Resolved formatting of every character in the target (or the caret).
|
||||
fn resolved(s: &Session) -> Vec<ResolvedChar> {
|
||||
let mut out = Vec::new();
|
||||
if let Some(rows) = column_rows(s) {
|
||||
@@ -393,88 +391,18 @@ fn underline_color(s: &mut Session, v: &Value) -> CmdResult {
|
||||
apply(s, &|x| x.underline_color = rgb)
|
||||
}
|
||||
|
||||
/// Test the entire highlight target without the formatting-state query's run cap.
|
||||
/// Resolve one run at a time and stop at the first mismatch, without collecting them.
|
||||
fn target_has_highlight(s: &Session, h: Highlight) -> bool {
|
||||
let matches = |style: Option<&str>, props: &CharProps| s.doc.styles.resolve_char(style, props).highlight == h.rgb();
|
||||
let mut seen = false;
|
||||
if let Some(rows) = column_rows(s) {
|
||||
for (a, b) in rows {
|
||||
let Some(p) = s.doc.para_at(&a) else { continue };
|
||||
let mark = a == b || (a.off == 0 && b.off == p.len());
|
||||
if !span_matches(p, mark, a.off, b.off, &matches, &mut seen) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
if seen {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
if let Some((a, b)) = selected_text_box(s).or_else(|| target(s)) {
|
||||
for path in s.doc.paths_between(&a, &b) {
|
||||
let Some(p) = s.doc.para(a.story, &path) else { continue };
|
||||
let from = if path == a.path { a.off } else { 0 };
|
||||
let to = if path == b.path { b.off } else { p.len() };
|
||||
// Match format_range: fully selected paragraphs include their mark.
|
||||
let mark = to == p.len() && (from == 0 || path != a.path);
|
||||
if !span_matches(p, mark, from, to, &matches, &mut seen) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
}
|
||||
// No runs: retain the caret/pending and empty text-box paragraph-mark rules.
|
||||
seen || resolved(s).iter().all(|r| r.highlight == h.rgb())
|
||||
}
|
||||
|
||||
/// Whether the paragraph's mark (when `mark`) and the runs overlapping `from..to` all match
|
||||
/// `matches`; `seen` records that something was checked.
|
||||
fn span_matches(
|
||||
p: &wordcraft_doc::Paragraph,
|
||||
mark: bool,
|
||||
from: usize,
|
||||
to: usize,
|
||||
matches: &impl Fn(Option<&str>, &CharProps) -> bool,
|
||||
seen: &mut bool,
|
||||
) -> bool {
|
||||
let style = p.props.style.as_deref();
|
||||
if mark {
|
||||
*seen = true;
|
||||
if !matches(style, &p.mark) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
for (r, c) in p.run_ranges() {
|
||||
if r.end > from && r.start < to {
|
||||
*seen = true;
|
||||
if !matches(style, c) {
|
||||
return false;
|
||||
}
|
||||
}
|
||||
}
|
||||
true
|
||||
}
|
||||
|
||||
/// Explicit palette/automation colours set highlighting; the shortcut (no colour) toggles
|
||||
/// the last choice off only when the whole formatting target already has that colour.
|
||||
fn highlight(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let color = v.get("color").map(|_| p::req_str(v, "color")).transpose()?;
|
||||
let h = match color {
|
||||
None => s.last_highlight,
|
||||
Some(c) => Highlight::ALL
|
||||
.iter()
|
||||
.copied()
|
||||
.find(|h| {
|
||||
h.ooxml().eq_ignore_ascii_case(c)
|
||||
|| h.name().replace(['-', ' ', '%'], "").eq_ignore_ascii_case(&c.replace(['-', ' ', '%'], ""))
|
||||
|| format!("{h:?}").eq_ignore_ascii_case(c)
|
||||
})
|
||||
.ok_or_else(|| CmdError::Params(format!("unknown highlight colour `{c}`")))?,
|
||||
};
|
||||
let remove = color.is_none() && target_has_highlight(s, h);
|
||||
let value = if remove { Highlight::None } else { h };
|
||||
let result = apply(s, &|x| x.highlight = Some(value))?;
|
||||
s.last_highlight = h;
|
||||
Ok(result)
|
||||
let c = p::str(v, "color").unwrap_or("yellow");
|
||||
let h = Highlight::ALL
|
||||
.iter()
|
||||
.copied()
|
||||
.find(|h| {
|
||||
h.ooxml().eq_ignore_ascii_case(c)
|
||||
|| h.name().replace(['-', ' ', '%'], "").eq_ignore_ascii_case(&c.replace(['-', ' ', '%'], ""))
|
||||
|| format!("{h:?}").eq_ignore_ascii_case(c)
|
||||
})
|
||||
.ok_or_else(|| CmdError::Params(format!("unknown highlight colour `{c}`")))?;
|
||||
apply(s, &|x| x.highlight = Some(h))
|
||||
}
|
||||
|
||||
/// `format.changeCase`: `mode` (lower, upper, title, sentence, toggle), or Shift+F3's cycle
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
|
||||
use serde_json::{Value, json};
|
||||
use wordcraft_doc::para::{Float, InlineObject, ShapeKind, Wrap};
|
||||
use wordcraft_doc::props::{Align, Border, Borders, CharProps, Rgb, TabAlign, TabStop, TextColor};
|
||||
use wordcraft_doc::props::{Align, CharProps, Rgb, TabAlign, TabStop, TextColor};
|
||||
use wordcraft_doc::{Block, Paragraph, PartKind, Path, Pos, StoryRef, Table, para_block};
|
||||
|
||||
use super::{delete_selection, sel_result, split_para, type_text};
|
||||
@@ -227,8 +227,8 @@ pub(crate) fn picture(s: &mut Session, v: &Value) -> CmdResult {
|
||||
place_picture(s, bytes, p::f32(v, "width"), p::str(v, "alt").unwrap_or(""))
|
||||
}
|
||||
|
||||
/// The media extension of an encoded image by its magic bytes, as the picture probe accepted it
|
||||
/// (TIFF included when its compression is one the `image` crate decodes within limits).
|
||||
/// The media extension of an encoded image by its magic bytes, as `image_size` accepted it
|
||||
/// (TIFF included: the `image` crate decodes it).
|
||||
pub(crate) fn image_ext(bytes: &[u8]) -> &'static str {
|
||||
match bytes.get(..4) {
|
||||
Some([0x89, b'P', b'N', b'G']) => "png",
|
||||
@@ -249,15 +249,14 @@ fn place_picture(s: &mut Session, bytes: Vec<u8>, width: Option<f32>, alt: &str)
|
||||
if bytes.len() > 200 << 20 {
|
||||
return Err(CmdError::Failed("image is larger than 200 MB".into()));
|
||||
}
|
||||
// Natural size: pixels at 96 ppi, or a metafile's own frame (WMF, EMF: #802). A picture that
|
||||
// is never decoded (a TIFF with JPEG strips, #3645) isn't inserted.
|
||||
let (natural_w, natural_h, ext) = match wordcraft_geom::picture::probe(&bytes) {
|
||||
Ok(pic) => (pic.width as f32 * 0.75, pic.height as f32 * 0.75, image_ext(&bytes)),
|
||||
Err(wordcraft_geom::picture::Refusal::Unsupported(why)) => return Err(CmdError::Failed(why.into())),
|
||||
Err(_) => match wordcraft_render::metafile_size(&bytes) {
|
||||
Some(size) => size,
|
||||
None => return Err(CmdError::Failed("not a supported image (PNG, JPEG, GIF, WebP, BMP, WMF, EMF)".into())),
|
||||
},
|
||||
// Natural size: pixels at 96 ppi, or a metafile's own frame (WMF, EMF: #802).
|
||||
let (natural_w, natural_h, ext) = if let Some((pw, ph)) = wordcraft_render::image_size(&bytes) {
|
||||
let ext = image_ext(&bytes);
|
||||
(pw as f32 * 0.75, ph as f32 * 0.75, ext)
|
||||
} else if let Some(size) = wordcraft_render::metafile_size(&bytes) {
|
||||
size
|
||||
} else {
|
||||
return Err(CmdError::Failed("not a supported image (PNG, JPEG, GIF, WebP, BMP, WMF, EMF)".into()));
|
||||
};
|
||||
let key = s.doc.add_media(bytes, ext);
|
||||
// Fitted to the text width.
|
||||
@@ -586,42 +585,18 @@ fn hf_part(s: &mut Session, header: bool) -> u32 {
|
||||
id
|
||||
}
|
||||
|
||||
/// Three independently editable cells spanning the section's text width.
|
||||
fn three_column_hf(s: &Session, style: &str) -> wordcraft_doc::Blocks {
|
||||
let block = s.sel.focus.path.0.first().copied().unwrap_or(0) as usize;
|
||||
// The section `section_mut` would pick, read without making anything unique.
|
||||
let sections = s.doc.sections();
|
||||
let own = sections.iter().position(|(end, _)| block <= *end).filter(|i| i + 1 < sections.len()).and_then(|i| sections.get(i));
|
||||
let width = own.map_or_else(|| s.doc.last_section.text_width(), |(_, sec)| sec.text_width());
|
||||
let mut table = Table::new(1, 3, width);
|
||||
table.props.style = Some("TableNormal".into());
|
||||
table.props.width = Some(width);
|
||||
table.props.fixed = true;
|
||||
// Explicit nil borders also override borders inherited from the document's table style.
|
||||
table.props.borders = Some(Borders::all(Border::default()));
|
||||
for row in &mut table.rows {
|
||||
for (cell, align) in row.cells.iter_mut().zip([Align::Left, Align::Center, Align::Right]) {
|
||||
let mut para = Paragraph::new().styled(style);
|
||||
para.props.align = Some(align);
|
||||
cell.blocks = vec![para_block(para)];
|
||||
}
|
||||
}
|
||||
// Keep a trailing paragraph so the story can also be edited after the table.
|
||||
vec![std::sync::Arc::new(Block::Table(table)), para_block(Paragraph::new().styled(style))]
|
||||
}
|
||||
|
||||
fn header_footer(s: &mut Session, v: &Value, header: bool) -> CmdResult {
|
||||
let id = hf_part(s, header);
|
||||
let style = if header { "Header" } else { "Footer" };
|
||||
if let Some(text) = p::str(v, "text") {
|
||||
let blocks = text.split('\n').map(|l| para_block(Paragraph::with_text(l, CharProps::default()).styled(style))).collect();
|
||||
s.doc.set_story(StoryRef::Part(id), blocks)?;
|
||||
} else if p::str(v, "preset") == Some("blankThree") {
|
||||
let blocks = three_column_hf(s, style);
|
||||
s.doc.set_story(StoryRef::Part(id), blocks)?;
|
||||
} else if let Some(preset) = p::str(v, "preset") {
|
||||
let mut para = Paragraph::new().styled(style);
|
||||
match preset {
|
||||
"blankThree" => {
|
||||
para.insert_text(0, "[Type here]\t[Type here]\t[Type here]", &CharProps::default())?;
|
||||
}
|
||||
"title" => {
|
||||
para.insert_text(0, if s.doc.core.title.is_empty() { "[Document title]" } else { &s.doc.core.title }, &CharProps::default())?;
|
||||
para.props.align = Some(Align::Center);
|
||||
@@ -1075,36 +1050,6 @@ pub fn drop_cap_state(s: &Session) -> DropCapState {
|
||||
mod tests {
|
||||
use super::*;
|
||||
|
||||
/// A 4 × 4 RGB TIFF that says its strips are JPEG (they aren't): its header still reads.
|
||||
fn jpeg_strip_tiff() -> Vec<u8> {
|
||||
let mut t = Vec::new();
|
||||
image::RgbImage::new(4, 4).write_to(&mut std::io::Cursor::new(&mut t), image::ImageFormat::Tiff).unwrap();
|
||||
assert_eq!(&t[..4], b"II*\0");
|
||||
let ifd = u32::from_le_bytes(t[4..8].try_into().unwrap()) as usize;
|
||||
let n = usize::from(u16::from_le_bytes([t[ifd], t[ifd + 1]]));
|
||||
let e = (0..n).map(|i| ifd + 2 + 12 * i).find(|&e| t[e..e + 2] == 259u16.to_le_bytes()).unwrap();
|
||||
t[e + 8..e + 10].copy_from_slice(&7u16.to_le_bytes());
|
||||
t
|
||||
}
|
||||
|
||||
/// #3645: a TIFF with JPEG strips is never decoded, so it is neither inserted nor swapped in;
|
||||
/// before, its readable header was all insertion looked at.
|
||||
#[test]
|
||||
fn tiffs_with_jpeg_strips_are_refused() {
|
||||
let tiff = jpeg_strip_tiff();
|
||||
assert_eq!(wordcraft_render::image_size(&tiff), Some((4, 4)), "its header reads");
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
let e = s.run("insert.picture", &json!({"data": base64_encode(&tiff)})).unwrap_err().to_string();
|
||||
assert!(e.contains("TIFF"), "{e}");
|
||||
assert!(s.doc.media.is_empty());
|
||||
let mut plain = Vec::new();
|
||||
image::RgbImage::new(4, 4).write_to(&mut std::io::Cursor::new(&mut plain), image::ImageFormat::Tiff).unwrap();
|
||||
s.run("insert.picture", &json!({"data": base64_encode(&plain)})).unwrap();
|
||||
let e = s.run("picture.change", &json!({"data": base64_encode(&tiff)})).unwrap_err().to_string();
|
||||
assert!(e.contains("TIFF"), "{e}");
|
||||
assert_eq!(s.doc.media.len(), 1);
|
||||
}
|
||||
|
||||
/// Closing the header or footer (Esc, the ribbon) puts the caret back where it was in the
|
||||
/// body, not on page 1 (#3331).
|
||||
#[test]
|
||||
|
||||
@@ -522,32 +522,9 @@ fn insert_field(s: &mut Session, instr: &str, placeholder: &str) -> CmdResult {
|
||||
}
|
||||
|
||||
/// Parse CSV (RFC 4180-ish: quotes, doubled quotes, commas/semicolons/tabs).
|
||||
/// The field separator: whichever of `,` `;` and tab the first line (outside quotes) uses most.
|
||||
fn csv_separator(text: &str) -> char {
|
||||
let first = text.lines().next().unwrap_or("");
|
||||
let mut quoted = false;
|
||||
let mut unquoted = String::new();
|
||||
for c in first.chars() {
|
||||
if c == '"' {
|
||||
quoted = !quoted;
|
||||
} else if !quoted {
|
||||
unquoted.push(c);
|
||||
}
|
||||
}
|
||||
[',', ';', '\t'].into_iter().max_by_key(|c| unquoted.matches(*c).count()).unwrap_or(',')
|
||||
}
|
||||
|
||||
/// Ends the current row: kept when it has any non-empty field, otherwise dropped.
|
||||
fn push_csv_row(rows: &mut Vec<Vec<String>>, row: &mut Vec<String>) {
|
||||
if row.iter().any(|x| !x.is_empty()) {
|
||||
rows.push(std::mem::take(row));
|
||||
} else {
|
||||
row.clear();
|
||||
}
|
||||
}
|
||||
|
||||
pub fn parse_csv(text: &str) -> Vec<Vec<String>> {
|
||||
let sep = csv_separator(text);
|
||||
let first = text.lines().next().unwrap_or("");
|
||||
let sep = [',', ';', '\t'].into_iter().max_by_key(|c| first.matches(*c).count()).unwrap_or(',');
|
||||
let mut rows = Vec::new();
|
||||
let mut row = Vec::new();
|
||||
let mut cell = String::new();
|
||||
@@ -574,7 +551,11 @@ pub fn parse_csv(text: &str) -> Vec<Vec<String>> {
|
||||
chars.next();
|
||||
}
|
||||
row.push(std::mem::take(&mut cell));
|
||||
push_csv_row(&mut rows, &mut row);
|
||||
if row.iter().any(|x| !x.is_empty()) {
|
||||
rows.push(std::mem::take(&mut row));
|
||||
} else {
|
||||
row.clear();
|
||||
}
|
||||
} else {
|
||||
cell.push(c);
|
||||
}
|
||||
@@ -583,7 +564,9 @@ pub fn parse_csv(text: &str) -> Vec<Vec<String>> {
|
||||
}
|
||||
}
|
||||
row.push(cell);
|
||||
push_csv_row(&mut rows, &mut row);
|
||||
if row.iter().any(|x| !x.is_empty()) {
|
||||
rows.push(row);
|
||||
}
|
||||
rows
|
||||
}
|
||||
|
||||
@@ -793,13 +776,7 @@ fn merge_fields(doc: &Document) -> Vec<String> {
|
||||
}
|
||||
|
||||
fn mergefield_name(instr: &str) -> Option<String> {
|
||||
// The field type ignores case and ends at a space or a quote, so `MERGEFIELDX` is another field.
|
||||
let t = instr.trim();
|
||||
let (kind, rest) = t.split_at_checked(t.find(|c: char| c.is_whitespace() || c == '"').unwrap_or(t.len()))?;
|
||||
if !kind.eq_ignore_ascii_case("MERGEFIELD") {
|
||||
return None;
|
||||
}
|
||||
let rest = rest.trim();
|
||||
let rest = instr.trim().strip_prefix("MERGEFIELD")?.trim();
|
||||
let name = if let Some(r) = rest.strip_prefix('"') { r.split('"').next().unwrap_or("") } else { rest.split_whitespace().next().unwrap_or("") };
|
||||
Some(name.to_string())
|
||||
}
|
||||
@@ -854,15 +831,11 @@ enum Act {
|
||||
Skip,
|
||||
}
|
||||
|
||||
/// The merge field a rule tests: the name in its first nested field, if that is a MERGEFIELD.
|
||||
fn rule_field(t: &str) -> Option<String> {
|
||||
mergefield_name(t.split_once('{').and_then(|(_, r)| r.split('}').next())?)
|
||||
}
|
||||
|
||||
/// Comparison of a rule condition `{ MERGEFIELD X } op "value" …` (IF, SKIPIF, NEXTIF) for
|
||||
/// `row`: whether it holds, and the quoted texts after the value (IF's true / false text).
|
||||
fn condition<'a>(s: &Session, t: &'a str, row: usize) -> (bool, Vec<&'a str>) {
|
||||
let field = rule_field(t).unwrap_or_default();
|
||||
let inner = t.split_once('{').and_then(|(_, r)| r.split('}').next()).unwrap_or("");
|
||||
let field = mergefield_name(inner).unwrap_or_default();
|
||||
let rest = t.split_once('}').map(|(_, r)| r).unwrap_or("");
|
||||
let op: String = rest.trim_start().chars().take_while(|c| matches!(c, '<'..='>')).collect();
|
||||
let mut quoted = rest.split('"').skip(1).step_by(2);
|
||||
@@ -905,7 +878,7 @@ fn act(s: &Session, instr: &str, row: usize) -> Option<Act> {
|
||||
"SKIPIF" if condition(s, t, row).0 => Act::Skip,
|
||||
"NEXTIF" | "SKIPIF" => Act::Text(String::new()),
|
||||
// IF { MERGEFIELD X } = "v" "then" "else" (an IF on other fields isn't ours to touch).
|
||||
"IF" if rule_field(t).is_some() => {
|
||||
"IF" if t.contains("MERGEFIELD") => {
|
||||
let (holds, texts) = condition(s, t, row);
|
||||
Act::Text(texts.get(if holds { 0 } else { 1 }).copied().unwrap_or("").to_string())
|
||||
}
|
||||
@@ -922,7 +895,7 @@ fn placeholder(instr: &str) -> Option<String> {
|
||||
"ADDRESSBLOCK" => Some("«AddressBlock»".into()),
|
||||
"GREETINGLINE" => Some("«GreetingLine»".into()),
|
||||
"NEXT" | "NEXTIF" | "SKIPIF" | "MERGEREC" => Some(String::new()),
|
||||
"IF" if rule_field(instr).is_some() => Some(String::new()),
|
||||
"IF" if instr.contains("MERGEFIELD") => Some(String::new()),
|
||||
_ => None,
|
||||
}
|
||||
}
|
||||
@@ -1104,8 +1077,10 @@ fn finish(s: &mut Session, v: &Value) -> CmdResult {
|
||||
crate::io::save_path(std::path::Path::new(path), &out).map_err(CmdError::Failed)?;
|
||||
return Ok(json!({"records": count, "path": path}));
|
||||
}
|
||||
s.add_document(out);
|
||||
s.set_document(out);
|
||||
s.path = None;
|
||||
s.dirty = true;
|
||||
s.merge.preview = false;
|
||||
Ok(json!({"records": count}))
|
||||
}
|
||||
|
||||
@@ -1205,24 +1180,13 @@ fn envelope_size(v: &Value) -> Result<(f32, f32), CmdError> {
|
||||
.ok_or_else(|| CmdError::Params(format!("unknown envelope size `{name}`")))
|
||||
}
|
||||
|
||||
/// Longest label or envelope address text accepted.
|
||||
const MAX_LABEL_TEXT: usize = 2_000;
|
||||
|
||||
/// An optional label/envelope text parameter, rejected when longer than `MAX_LABEL_TEXT` characters.
|
||||
fn label_text(v: &Value, key: &str) -> Result<Option<String>, CmdError> {
|
||||
match p::str(v, key) {
|
||||
Some(t) if t.chars().count() > MAX_LABEL_TEXT => Err(CmdError::Params(format!("`{key}` is longer than {MAX_LABEL_TEXT} characters"))),
|
||||
t => Ok(t.map(str::to_string)),
|
||||
}
|
||||
}
|
||||
|
||||
/// Envelopes: a landscape page with the return address at the top left and the delivery address
|
||||
/// in the middle. As a new document, or (`add`) as a section of its own at the start of this one.
|
||||
fn envelopes(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let delivery = label_text(v, "delivery")?.unwrap_or_else(|| {
|
||||
let delivery = p::str(v, "delivery").map(str::to_string).unwrap_or_else(|| {
|
||||
if s.merge.rows.is_empty() { "Recipient Name\nStreet Address\nCity, ST 00000".into() } else { address_block(s, s.merge.record) }
|
||||
});
|
||||
let ret = label_text(v, "return")?.unwrap_or_default();
|
||||
let ret = p::str(v, "return").unwrap_or("").to_string();
|
||||
let (w, h) = envelope_size(v)?;
|
||||
let sect = SectionProps {
|
||||
page_w: w,
|
||||
@@ -1267,7 +1231,8 @@ fn envelopes(s: &mut Session, v: &Value) -> CmdResult {
|
||||
let mut d = Document::new();
|
||||
d.last_section = sect;
|
||||
d.body = blocks;
|
||||
s.add_document(d);
|
||||
s.set_document(d);
|
||||
s.path = None;
|
||||
Ok(json!({"width": w, "height": h}))
|
||||
}
|
||||
|
||||
@@ -1443,7 +1408,7 @@ fn labels(s: &mut Session, v: &Value) -> CmdResult {
|
||||
(at("row").min(rows), at("col").min(cols))
|
||||
});
|
||||
let from_rec = single.is_none() && p::bool(v, "fromRecipients").unwrap_or(!s.merge.rows.is_empty() && p::str(v, "text").is_none());
|
||||
let text = label_text(v, "text")?.unwrap_or_else(|| "Name\nAddress\nCity, ST 00000".into());
|
||||
let text = p::str(v, "text").unwrap_or("Name\nAddress\nCity, ST 00000").to_string();
|
||||
let (page_w, page_h) = if prod.paper == "A4" { (595.3, 841.9) } else { (612.0, 792.0) };
|
||||
let h_gap = (prod.h_pitch - prod.width).max(0.0);
|
||||
let v_gap = (prod.v_pitch - prod.height).max(0.0);
|
||||
@@ -1499,7 +1464,8 @@ fn labels(s: &mut Session, v: &Value) -> CmdResult {
|
||||
}
|
||||
}
|
||||
d.body = vec![std::sync::Arc::new(Block::Table(t)), para_block(Paragraph::new())];
|
||||
s.add_document(d);
|
||||
s.set_document(d);
|
||||
s.path = None;
|
||||
let _ = Pos::body(0, 0);
|
||||
Ok(json!({"labels": if single.is_some() { 1 } else { rows * cols }, "filled": made, "product": prod.id}))
|
||||
}
|
||||
@@ -1515,86 +1481,6 @@ mod tests {
|
||||
assert_eq!(parse_csv("a;b\n1;2")[1], vec!["1", "2"]);
|
||||
}
|
||||
|
||||
/// #4001: finishing a merge opens a new document, with a new undo history. It is not an
|
||||
/// edit of the open document, even when the merged document holds the same content.
|
||||
#[test]
|
||||
fn finishing_a_merge_opens_a_new_document_with_a_new_history() {
|
||||
let mut s = Session::new(Document::from_text("Hello"));
|
||||
s.run("text.insert", &json!({"text": "!"})).unwrap();
|
||||
assert_eq!(s.undo_depth(), 1);
|
||||
s.run("caret.docStart", &json!({})).unwrap();
|
||||
// One recipient and no merge fields: the merged document holds the same content.
|
||||
s.run("mailings.recipients", &json!({"rows": [{"Name": "Ada"}]})).unwrap();
|
||||
let id = s.document_id();
|
||||
s.run("mailings.finish", &json!({})).unwrap();
|
||||
assert_ne!(s.document_id(), id, "the document was replaced");
|
||||
assert_eq!(s.doc.plain_text(StoryRef::Body), "!Hello");
|
||||
assert_eq!(s.undo_depth(), 0, "the new document starts with an empty undo history");
|
||||
// Undo cannot bring the old document's text back.
|
||||
s.run("edit.undo", &json!({})).unwrap();
|
||||
assert_eq!(s.doc.plain_text(StoryRef::Body), "!Hello", "the old text does not come back");
|
||||
}
|
||||
|
||||
/// #4004: finishing a merge to a file saves the merged copy there and leaves the open
|
||||
/// document alone: it changes nothing, so it is not an edit.
|
||||
#[test]
|
||||
fn finishing_a_merge_to_a_file_is_not_an_edit() {
|
||||
let dir = std::env::temp_dir().join(format!("wordcraft-merge-to-file-{}", std::process::id()));
|
||||
std::fs::create_dir_all(&dir).unwrap();
|
||||
let out = dir.join("merged.docx");
|
||||
let mut s = Session::new(Document::from_text("Hello"));
|
||||
s.run("text.insert", &json!({"text": "!"})).unwrap();
|
||||
s.run("mailings.recipients", &json!({"rows": [{"Name": "Ada"}]})).unwrap();
|
||||
let (depth, dirty, text) = (s.undo_depth(), s.dirty, s.doc.plain_text(StoryRef::Body));
|
||||
s.run("mailings.finish", &json!({"path": out.to_string_lossy()})).unwrap();
|
||||
assert!(out.exists(), "the merged copy was written");
|
||||
assert_eq!(s.doc.plain_text(StoryRef::Body), text, "the open document is unchanged");
|
||||
assert_eq!(s.undo_depth(), depth, "no undo step");
|
||||
assert_eq!(s.dirty, dirty, "the unsaved-changes flag is unchanged");
|
||||
let _ = std::fs::remove_dir_all(&dir);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn csv_delimiter_ignores_quoted_text() {
|
||||
let r = parse_csv("\"Name, first, last\";City\nAnn;Paris\n");
|
||||
assert_eq!(r, vec![vec!["Name, first, last", "City"], vec!["Ann", "Paris"]]);
|
||||
}
|
||||
|
||||
/// Field types ignore case (ECMA-376 Part 1, 17.16.1), and `MERGEFIELDX` is another field.
|
||||
#[test]
|
||||
fn merge_field_types_ignore_case() {
|
||||
assert_eq!(mergefield_name(" mergefield City ").as_deref(), Some("City"));
|
||||
assert_eq!(mergefield_name(r#"MergeField "First Name""#).as_deref(), Some("First Name"));
|
||||
assert_eq!(mergefield_name(r#"MERGEFIELD"City""#).as_deref(), Some("City"));
|
||||
assert_eq!(mergefield_name("MERGEFIELDX City"), None);
|
||||
assert_eq!(placeholder("mergefield City").as_deref(), Some("«City»"));
|
||||
assert_eq!(placeholder(r#"if { mergefield City } = "x" "y" "z""#).as_deref(), Some(""));
|
||||
let mut s = Session::new(Document::new());
|
||||
s.run("mailings.recipients", &json!({"rows": [{"Name": "Alpha"}, {"Name": "Beta"}]})).unwrap();
|
||||
s.run("mailings.start", &json!({"kind": "directory"})).unwrap();
|
||||
s.run("insert.field", &json!({"instr": "mergefield Name", "result": ""})).unwrap();
|
||||
s.run("insert.field", &json!({"instr": r#"if { mergefield Name } = "Beta" " (b)" """#, "result": ""})).unwrap();
|
||||
s.run("mailings.finish", &json!({})).unwrap();
|
||||
assert_eq!(s.doc.plain_text(StoryRef::Body), "Alpha\nBeta (b)");
|
||||
}
|
||||
|
||||
/// An IF is a merge rule only when the field it tests is a MERGEFIELD: the word in a quoted
|
||||
/// value doesn't make `IF { PAGE } = "mergefield" …` one.
|
||||
#[test]
|
||||
fn an_if_on_another_field_is_not_a_merge_rule() {
|
||||
let other = r#"IF { PAGE } = "mergefield" "yes" "no""#;
|
||||
assert_eq!(placeholder(other), None);
|
||||
assert_eq!(placeholder(r#"IF { PAGE } = { MERGEFIELD City } "yes" "no""#), None);
|
||||
assert_eq!(placeholder(r#"IF { MERGEFIELD City } = "mergefield" "yes" "no""#).as_deref(), Some(""));
|
||||
let mut s = Session::new(Document::new());
|
||||
s.run("mailings.recipients", &json!({"rows": [{"Name": "Alpha"}]})).unwrap();
|
||||
s.run("mailings.start", &json!({"kind": "directory"})).unwrap();
|
||||
s.run("insert.field", &json!({"instr": "mergefield Name", "result": ""})).unwrap();
|
||||
s.run("insert.field", &json!({"instr": other, "result": " kept"})).unwrap();
|
||||
s.run("mailings.finish", &json!({})).unwrap();
|
||||
assert_eq!(s.doc.plain_text(StoryRef::Body), "Alpha kept");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn merge_end_to_end() {
|
||||
let mut s = Session::new(Document::new());
|
||||
@@ -1837,17 +1723,6 @@ mod tests {
|
||||
assert!(!s.doc.parts.contains_key(&hdr), "the template header is replaced by the letters' copies");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn label_and_envelope_text_is_capped() {
|
||||
let mut s = Session::new(Document::new());
|
||||
let long = "x".repeat(2001);
|
||||
assert!(matches!(s.run("mailings.labels", &json!({"text": long})), Err(CmdError::Params(_))));
|
||||
assert!(matches!(s.run("mailings.envelopes", &json!({"delivery": long})), Err(CmdError::Params(_))));
|
||||
assert!(matches!(s.run("mailings.envelopes", &json!({"return": long})), Err(CmdError::Params(_))));
|
||||
s.run("mailings.labels", &json!({"text": "x".repeat(2000), "rows": 40, "cols": 10})).unwrap();
|
||||
assert!(s.doc.plain_text(StoryRef::Body).chars().count() <= 400 * 2000 + 400 * 20);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn envelopes_and_labels() {
|
||||
let mut s = Session::new(Document::new());
|
||||
|
||||
+18
-107
@@ -150,43 +150,16 @@ pub fn selected_paths(s: &Session) -> Vec<wordcraft_doc::Path> {
|
||||
s.doc.para_paths_in(&a, &b)
|
||||
}
|
||||
|
||||
/// What one selection deletion actually did, for IME outcome classification.
|
||||
#[derive(Clone, Debug, Default)]
|
||||
pub(crate) struct SelectionDelete {
|
||||
pub at: Pos,
|
||||
pub content_changed: bool,
|
||||
pub state_changed: bool,
|
||||
}
|
||||
|
||||
/// Delete the selection (respecting track changes), reporting whether committed content changed.
|
||||
///
|
||||
/// Under Track Changes an all-already-deleted range is a real no-op: no revision record is
|
||||
/// created and the effect says so.
|
||||
pub(crate) fn delete_selection_effect(s: &mut Session) -> Result<SelectionDelete, CmdError> {
|
||||
let (a, b) = s.sel.ordered();
|
||||
if a == b {
|
||||
return Ok(SelectionDelete { at: a, content_changed: false, state_changed: false });
|
||||
}
|
||||
controls::check_delete(s, &a, &b)?;
|
||||
let effect = if s.doc.settings.track_changes {
|
||||
if !track_delete_would_change(s, &a, &b) {
|
||||
SelectionDelete { at: a.clone(), content_changed: false, state_changed: false }
|
||||
} else {
|
||||
let rid = new_revision(s, RevisionKind::Delete);
|
||||
let deleted = track_delete_as(s, &a, &b, rid)?;
|
||||
SelectionDelete { at: deleted.at, content_changed: deleted.changed, state_changed: deleted.changed }
|
||||
}
|
||||
} else {
|
||||
let at = s.doc.delete_range(&a, &b)?;
|
||||
SelectionDelete { at, content_changed: true, state_changed: true }
|
||||
};
|
||||
s.sel = crate::Selection::caret(effect.at.clone());
|
||||
Ok(effect)
|
||||
}
|
||||
|
||||
/// Delete the selection (respecting track changes); returns the collapsed position.
|
||||
pub fn delete_selection(s: &mut Session) -> Result<Pos, CmdError> {
|
||||
delete_selection_effect(s).map(|e| e.at)
|
||||
let (a, b) = s.sel.ordered();
|
||||
if a == b {
|
||||
return Ok(a);
|
||||
}
|
||||
controls::check_delete(s, &a, &b)?;
|
||||
let at = if s.doc.settings.track_changes { track_delete(s, &a, &b)? } else { s.doc.delete_range(&a, &b)? };
|
||||
s.sel = crate::Selection::caret(at.clone());
|
||||
Ok(at)
|
||||
}
|
||||
|
||||
/// A revision record for the current author.
|
||||
@@ -244,47 +217,15 @@ pub fn remove_empty_para(s: &mut Session, story: StoryRef, path: &wordcraft_doc:
|
||||
Ok(true)
|
||||
}
|
||||
|
||||
/// Whether a tracked deletion of `a..b` would change anything: any non-deleted run intersects
|
||||
/// the range, or a removable/markable paragraph mark lies inside it.
|
||||
pub(crate) fn track_delete_would_change(s: &Session, a: &Pos, b: &Pos) -> bool {
|
||||
let paths = s.doc.paths_between(a, b);
|
||||
for path in &paths {
|
||||
let Some(p) = s.doc.para(a.story, path) else { continue };
|
||||
let from = if *path == a.path { a.off } else { 0 };
|
||||
let to = if *path == b.path { b.off } else { p.len() };
|
||||
if p.run_ranges().any(|(r, c)| r.end > from && r.start < to && c.del.is_none()) {
|
||||
return true;
|
||||
}
|
||||
if *path != b.path && has_next_para(s, a.story, path) && p.mark.del.is_none() {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
false
|
||||
}
|
||||
|
||||
/// What one tracked deletion did.
|
||||
#[derive(Clone, Debug, Default, PartialEq, Eq)]
|
||||
pub(crate) struct TrackDeleteEffect {
|
||||
pub at: Pos,
|
||||
/// At least one run was removed/marked deleted, or a paragraph mark was removed/marked.
|
||||
pub changed: bool,
|
||||
}
|
||||
|
||||
/// Tracked deletion: own insertions are removed, other text is marked deleted.
|
||||
///
|
||||
/// Preflights first: an all-already-deleted range does nothing and creates no revision record.
|
||||
pub(crate) fn track_delete(s: &mut Session, a: &Pos, b: &Pos) -> Result<Pos, CmdError> {
|
||||
if !track_delete_would_change(s, a, b) {
|
||||
return Ok(a.clone());
|
||||
}
|
||||
let rid = new_revision(s, RevisionKind::Delete);
|
||||
track_delete_as(s, a, b, rid).map(|e| e.at)
|
||||
track_delete_as(s, a, b, rid)
|
||||
}
|
||||
|
||||
/// [`track_delete`] marking the deleted text with revision `rid`.
|
||||
pub(crate) fn track_delete_as(s: &mut Session, a: &Pos, b: &Pos, rid: u32) -> Result<TrackDeleteEffect, CmdError> {
|
||||
pub(crate) fn track_delete_as(s: &mut Session, a: &Pos, b: &Pos, rid: u32) -> Result<Pos, CmdError> {
|
||||
let author = s.author.clone();
|
||||
let mut changed = false;
|
||||
// Remove text this author inserted (it never existed for the reader); mark the rest.
|
||||
let paths = s.doc.paths_between(a, b);
|
||||
for path in paths.iter().rev() {
|
||||
@@ -307,7 +248,6 @@ pub(crate) fn track_delete_as(s: &mut Session, a: &Pos, b: &Pos, rid: u32) -> Re
|
||||
} else {
|
||||
para.format(x, y, &|c| c.del = Some(rid))?;
|
||||
}
|
||||
changed = true;
|
||||
}
|
||||
// The paragraph mark is inside the range unless this is its last paragraph. Like text,
|
||||
// a mark this author inserted is removed (the paragraphs join again), any other is
|
||||
@@ -324,27 +264,14 @@ pub(crate) fn track_delete_as(s: &mut Session, a: &Pos, b: &Pos, rid: u32) -> Re
|
||||
p.mark.del = Some(rid);
|
||||
p.touch();
|
||||
}
|
||||
changed = true;
|
||||
}
|
||||
}
|
||||
}
|
||||
Ok(TrackDeleteEffect { at: a.clone(), changed })
|
||||
}
|
||||
|
||||
/// What one text insertion/deletion commit changed, for IME outcome classification.
|
||||
#[derive(Clone, Copy, Debug, Default)]
|
||||
pub(crate) struct InsertEffect {
|
||||
pub content_changed: bool,
|
||||
pub state_changed: bool,
|
||||
Ok(a.clone())
|
||||
}
|
||||
|
||||
/// Insert typed text at the caret (replacing the selection), with pending formatting and track changes.
|
||||
pub fn type_text(s: &mut Session, text: &str) -> Result<(), CmdError> {
|
||||
insert_text(s, text, true).map(|_| ())
|
||||
}
|
||||
|
||||
/// [`type_text`] with the effect the insertion had.
|
||||
pub(crate) fn type_text_effect(s: &mut Session, text: &str) -> Result<InsertEffect, CmdError> {
|
||||
insert_text(s, text, true)
|
||||
}
|
||||
|
||||
@@ -352,38 +279,30 @@ pub(crate) fn type_text_effect(s: &mut Session, text: &str) -> Result<InsertEffe
|
||||
/// being transferred, not Enter key presses: an empty list item between them stays (Enter on
|
||||
/// one would end the list).
|
||||
pub fn paste_text(s: &mut Session, text: &str) -> Result<(), CmdError> {
|
||||
insert_text(s, text, false).map(|_| ())
|
||||
insert_text(s, text, false)
|
||||
}
|
||||
|
||||
/// Inserts typed or pasted text at the caret: resolves field ends and content-control placeholders,
|
||||
/// deletes the selection, applies tracked-change insertions, and splits paragraphs at every typed
|
||||
/// line break or pasted paragraph separator. A pasted separator keeps the list semantics of pasted
|
||||
/// text; a one-line control suppresses typed breaks only. It gives an otherwise undirected paragraph
|
||||
/// right-to-left direction when its first strong letter is RTL, and returns whether content or state
|
||||
/// changed.
|
||||
fn insert_text(s: &mut Session, text: &str, enter: bool) -> Result<InsertEffect, CmdError> {
|
||||
fn insert_text(s: &mut Session, text: &str, enter: bool) -> Result<(), CmdError> {
|
||||
let mut props = s.typing_props();
|
||||
if s.sel.is_collapsed() {
|
||||
s.sel = crate::Selection::caret(s.doc.past_field_ends(&s.sel.focus));
|
||||
}
|
||||
// Content controls: a placeholder is replaced, locked ones refuse.
|
||||
let prepared = controls::prepare_edit(s)?;
|
||||
let placeholder_changed = prepared.is_some();
|
||||
if let Some(p) = prepared {
|
||||
if let Some(p) = controls::prepare_edit(s)? {
|
||||
props = p;
|
||||
}
|
||||
// Pasted text is text: pending properties left by AutoCorrect don't give it a symbol's font (#2036).
|
||||
if !enter {
|
||||
props = props.without_symbol();
|
||||
}
|
||||
let deletion = delete_selection_effect(s)?;
|
||||
delete_selection(s)?;
|
||||
// Typed text is new: never deleted, and not a formatting change of the text around it.
|
||||
props.fmt_change = None;
|
||||
// Its direction comes from its own letters, as on screen, not from the run it's typed into:
|
||||
// Latin typed into reopened Hebrew must not be saved as right-to-left (#1543). The DOCX
|
||||
// writer marks the right-to-left pieces itself.
|
||||
props.rtl = None;
|
||||
if s.doc.settings.track_changes && !text.is_empty() {
|
||||
if s.doc.settings.track_changes {
|
||||
props.ins = Some(new_revision(s, RevisionKind::Insert));
|
||||
props.del = None;
|
||||
} else {
|
||||
@@ -392,20 +311,13 @@ fn insert_text(s: &mut Session, text: &str, enter: bool) -> Result<InsertEffect,
|
||||
}
|
||||
let mut at = s.sel.focus.clone();
|
||||
let mut first = true;
|
||||
let mut inserted = false;
|
||||
// Typed line breaks are the Enter key between the lines: in a one-line control it does
|
||||
// nothing, so the lines run together and the control stays one paragraph (pasted text keeps
|
||||
// its breaks, as before).
|
||||
let no_breaks = enter && controls::refuses_paragraph(s);
|
||||
for part in text.split(['\r', '\u{2029}']) {
|
||||
if !first && !no_breaks {
|
||||
if !first {
|
||||
at = split_para_as(s, &at, enter)?;
|
||||
inserted = true;
|
||||
}
|
||||
first = false;
|
||||
let part = part.replace('\u{000B}', "\n");
|
||||
if !part.is_empty() {
|
||||
inserted = true;
|
||||
let starts_para = direction_unset(s, &at);
|
||||
at = s.doc.insert_text(&at, &part, &props)?;
|
||||
// Typing Arabic, Hebrew… as the first letters of a paragraph whose direction nothing
|
||||
@@ -422,8 +334,7 @@ fn insert_text(s: &mut Session, text: &str, enter: bool) -> Result<InsertEffect,
|
||||
}
|
||||
s.sel = crate::Selection::caret(at);
|
||||
s.goal_x = None;
|
||||
let content_changed = deletion.content_changed || inserted;
|
||||
Ok(InsertEffect { content_changed, state_changed: deletion.state_changed || content_changed || placeholder_changed })
|
||||
Ok(())
|
||||
}
|
||||
|
||||
/// Is the paragraph at `at` still without letters (empty, or only spaces, digits and
|
||||
|
||||
@@ -58,7 +58,7 @@ pub fn cut_tracked(s: &mut Session) -> Result<bool, CmdError> {
|
||||
}
|
||||
s.doc.revisions.push(Revision::new(RevisionKind::Delete, s.author.clone(), now_iso()));
|
||||
let rid = (s.doc.revisions.len() - 1) as u32;
|
||||
let at = track_delete_as(s, &a, &b, rid)?.at;
|
||||
let at = track_delete_as(s, &a, &b, rid)?;
|
||||
s.sel = Selection::caret(at);
|
||||
// Only the author's own insertions were cut: they are gone, not moved.
|
||||
if in_use(s, rid) {
|
||||
|
||||
@@ -134,7 +134,7 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
// `blur` asserts on a subnormal width.
|
||||
img = img.blur(-sharpen / 25.0);
|
||||
}
|
||||
Ok(img)
|
||||
img
|
||||
})
|
||||
})
|
||||
.params(r#"{"brightness"?: -100..100, "contrast"?: -100..100, "sharpen"?: -100..100}"#)
|
||||
@@ -146,7 +146,7 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
let (_, o) = selected(s).ok_or_else(|| CmdError::Disabled("no picture".into()))?;
|
||||
let InlineObject::Image { media, .. } = o else { return Err(CmdError::Disabled("not a picture".into())) };
|
||||
let base = s.originals.get(&media).cloned().unwrap_or(media.clone());
|
||||
adjust_media(s, &base, move |img| Ok(recolor(img, &mode, sat)))
|
||||
adjust_media(s, &base, move |img| recolor(img, &mode, sat))
|
||||
})
|
||||
.params(r#"{"mode": "grayscale|sepia|washout|blackAndWhite|saturation|tint", "saturation"?: 0..400}"#)
|
||||
.when(has_picture),
|
||||
@@ -159,7 +159,7 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
for px in rgba.pixels_mut() {
|
||||
px.0 = [c.0, c.1, c.2, px.0[3]];
|
||||
}
|
||||
Ok(image::DynamicImage::ImageRgba8(rgba))
|
||||
image::DynamicImage::ImageRgba8(rgba)
|
||||
})
|
||||
})
|
||||
.params(r#"{"color": "RRGGBB"}"#)
|
||||
@@ -186,20 +186,20 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
for px in rgba.pixels_mut() {
|
||||
px.0[3] = (px.0[3] as f32 * (1.0 - t / 100.0)) as u8;
|
||||
}
|
||||
Ok(image::DynamicImage::ImageRgba8(rgba))
|
||||
image::DynamicImage::ImageRgba8(rgba)
|
||||
})
|
||||
})
|
||||
.params(r#"{"percent": 0..100}"#)
|
||||
.when(has_picture),
|
||||
CommandSpec::new("picture.removeBackground", "Remove Background", "Picture Format › Adjust", |s, v| {
|
||||
let tol = p::f32(v, "tolerance").unwrap_or(30.0).clamp(0.0, 255.0);
|
||||
adjust(s, move |img| Ok(remove_background(img, tol)))
|
||||
adjust(s, move |img| remove_background(img, tol))
|
||||
})
|
||||
.params(r#"{"tolerance"?: 0..255}"#)
|
||||
.when(has_picture),
|
||||
CommandSpec::new("picture.compress", "Compress Pictures", "Picture Format › Adjust", |s, v| {
|
||||
let max = p::u64(v, "maxPixels").unwrap_or(1600).clamp(32, 20000) as u32;
|
||||
adjust(s, move |img| Ok(if img.width().max(img.height()) > max { img.resize(max, max, image::imageops::FilterType::Lanczos3) } else { img }))
|
||||
adjust(s, move |img| if img.width().max(img.height()) > max { img.resize(max, max, image::imageops::FilterType::Lanczos3) } else { img })
|
||||
})
|
||||
.params(r#"{"maxPixels"?: n}"#)
|
||||
.when(has_picture),
|
||||
@@ -212,10 +212,7 @@ pub fn specs() -> Vec<CommandSpec> {
|
||||
} else {
|
||||
return Err(CmdError::Params("`path` or `data` required".into()));
|
||||
};
|
||||
wordcraft_geom::picture::probe(&bytes).map_err(|e| match e {
|
||||
wordcraft_geom::picture::Refusal::Unsupported(why) => CmdError::Failed(why.into()),
|
||||
_ => CmdError::Failed("not a supported image".into()),
|
||||
})?;
|
||||
wordcraft_render::image_size(&bytes).ok_or_else(|| CmdError::Failed("not a supported image".into()))?;
|
||||
let key = s.doc.add_media(bytes.clone(), super::insert::image_ext(&bytes));
|
||||
with_obj(s, |o| {
|
||||
if let InlineObject::Image { media, .. } = o {
|
||||
@@ -1185,7 +1182,7 @@ pub fn fit_size(w: f32, h: f32, keep: bool) -> (f32, f32) {
|
||||
}
|
||||
|
||||
/// Apply an image operation to the selected picture's bitmap (stored as a new PNG).
|
||||
fn adjust(s: &mut Session, f: impl Fn(image::DynamicImage) -> Result<image::DynamicImage, CmdError>) -> CmdResult {
|
||||
fn adjust(s: &mut Session, f: impl Fn(image::DynamicImage) -> image::DynamicImage) -> CmdResult {
|
||||
let (_, o) = selected(s).ok_or_else(|| CmdError::Disabled("no picture".into()))?;
|
||||
let InlineObject::Image { media, .. } = o else { return Err(CmdError::Disabled("not a picture".into())) };
|
||||
adjust_media(s, &media, f)
|
||||
@@ -1194,14 +1191,13 @@ fn adjust(s: &mut Session, f: impl Fn(image::DynamicImage) -> Result<image::Dyna
|
||||
/// Like [`adjust`], but decodes `base` (a media key, usually the pre-edit original) instead
|
||||
/// of the picture's current bitmap. The result still chains onto the originals map, so Reset
|
||||
/// Picture keeps working.
|
||||
fn adjust_media(s: &mut Session, base: &str, f: impl Fn(image::DynamicImage) -> Result<image::DynamicImage, CmdError>) -> CmdResult {
|
||||
fn adjust_media(s: &mut Session, base: &str, f: impl Fn(image::DynamicImage) -> image::DynamicImage) -> CmdResult {
|
||||
let bytes = s.doc.media.get(base).cloned().ok_or_else(|| CmdError::Failed("picture data missing".into()))?;
|
||||
// Within the budget pictures are drawn within, checked from the header first (#3644, #3647).
|
||||
let img = wordcraft_geom::picture::decode(&bytes, wordcraft_geom::MAX_DECODED_BYTES).map_err(|e| match e {
|
||||
wordcraft_geom::picture::Refusal::TooLarge { .. } => too_large(),
|
||||
e => CmdError::Failed(format!("can't decode the picture: {e}")),
|
||||
})?;
|
||||
let out = f(img)?;
|
||||
let img = image::load_from_memory(&bytes).map_err(|e| CmdError::Failed(format!("can't decode the picture: {e}")))?;
|
||||
if (img.width() as u64) * (img.height() as u64) > 80_000_000 {
|
||||
return Err(CmdError::Failed("picture is too large to edit".into()));
|
||||
}
|
||||
let out = f(img);
|
||||
let mut png = Vec::new();
|
||||
out.write_to(&mut std::io::Cursor::new(&mut png), image::ImageFormat::Png).map_err(|e| CmdError::Failed(e.to_string()))?;
|
||||
let key = s.doc.add_media(png, "png");
|
||||
@@ -1225,25 +1221,6 @@ fn adjust_media(s: &mut Session, base: &str, f: impl Fn(image::DynamicImage) ->
|
||||
})
|
||||
}
|
||||
|
||||
fn too_large() -> CmdError {
|
||||
CmdError::Failed("picture is too large to edit".into())
|
||||
}
|
||||
|
||||
/// A `w` × `h` bitmap of `fill` for an edit to draw on, refused when the picture it makes would be
|
||||
/// too large to draw (two RGBA buffers within the decode budget), before it is allocated (#3611).
|
||||
fn canvas(w: u32, h: u32, fill: [u8; 4]) -> Result<image::RgbaImage, CmdError> {
|
||||
if !wordcraft_geom::decode_fits(w, h, wordcraft_geom::MAX_DECODED_BYTES) {
|
||||
return Err(too_large());
|
||||
}
|
||||
Ok(image::RgbaImage::from_pixel(w, h, image::Rgba(fill)))
|
||||
}
|
||||
|
||||
/// `w` and `h` grown by `pad` on every side, None past `u32`.
|
||||
fn grown(w: u32, h: u32, pad: u32) -> Option<(u32, u32)> {
|
||||
let twice = pad.checked_mul(2)?;
|
||||
Some((w.checked_add(twice)?, h.checked_add(twice)?))
|
||||
}
|
||||
|
||||
/// What Artistic Effects, Picture Styles and Picture Border drew on a picture's bitmap: the
|
||||
/// bitmap they started from and the effect, style and border on it, so the next one draws again
|
||||
/// from that bitmap instead of over the last result. Picking a second style replaces the first,
|
||||
@@ -1290,12 +1267,12 @@ fn relook(s: &mut Session, change: impl FnOnce(&mut PictureLook)) -> CmdResult {
|
||||
None => img,
|
||||
};
|
||||
if let Some(st) = &l.style {
|
||||
img = picture_style(img, st)?;
|
||||
img = picture_style(img, st);
|
||||
}
|
||||
if let Some((c, w)) = l.border {
|
||||
img = frame(img, w, [c.0, c.1, c.2, 255])?;
|
||||
img = frame(img, w, [c.0, c.1, c.2, 255]);
|
||||
}
|
||||
Ok(img)
|
||||
img
|
||||
})?;
|
||||
if let Some((_, InlineObject::Image { media, .. })) = selected(s)
|
||||
&& let Some(drawn) = s.doc.media.get(&media).cloned()
|
||||
@@ -1379,49 +1356,14 @@ fn recolor(img: image::DynamicImage, mode: &str, sat: f32) -> image::DynamicImag
|
||||
/// Flood the border-connected region close to the corner colour with transparency.
|
||||
fn remove_background(img: image::DynamicImage, tol: f32) -> image::DynamicImage {
|
||||
let mut rgba = img.to_rgba8();
|
||||
clear_background(&mut rgba, tol);
|
||||
image::DynamicImage::ImageRgba8(rgba)
|
||||
}
|
||||
|
||||
/// [`remove_background`] on `rgba`, a row span at a time: besides the border's seeds, the stack
|
||||
/// holds one for each run of matching pixels beside a filled span, instead of four for every
|
||||
/// pixel filled (#3647). Returns the most seeds it held.
|
||||
fn clear_background(rgba: &mut image::RgbaImage, tol: f32) -> usize {
|
||||
let (w, h) = rgba.dimensions();
|
||||
let Some(bg) = rgba.get_pixel_checked(0, 0).map(|p| p.0) else { return 0 };
|
||||
let close = |p: [u8; 4]| (0..3).map(|i| (p[i] as f32 - bg[i] as f32).abs()).fold(0.0f32, f32::max) <= tol;
|
||||
let at = |x: u32, y: u32| y as usize * w as usize + x as usize;
|
||||
// Not cleared yet, and close to the corner colour.
|
||||
let open = |rgba: &image::RgbaImage, seen: &[bool], x: u32, y: u32| {
|
||||
!seen.get(at(x, y)).copied().unwrap_or(true) && rgba.get_pixel_checked(x, y).is_some_and(|p| close(p.0))
|
||||
};
|
||||
let mut seen = vec![false; (w as usize) * (h as usize)];
|
||||
let mut stack = border_pixels(w, h);
|
||||
let mut most = stack.len();
|
||||
while let Some((x, y)) = stack.pop() {
|
||||
if !open(rgba, &seen, x, y) {
|
||||
continue;
|
||||
}
|
||||
let (x0, x1) = run_span(x, y, w, |xi, yi| open(rgba, &seen, xi, yi));
|
||||
for xi in x0..=x1 {
|
||||
if let Some(s) = seen.get_mut(at(xi, y)) {
|
||||
*s = true;
|
||||
}
|
||||
if let Some(p) = rgba.get_pixel_mut_checked(xi, y) {
|
||||
p.0[3] = 0;
|
||||
}
|
||||
}
|
||||
for ny in [y.checked_sub(1), y.checked_add(1).filter(|&n| n < h)].into_iter().flatten() {
|
||||
push_open_runs(&mut stack, x0, x1, ny, |xi, yi| open(rgba, &seen, xi, yi));
|
||||
}
|
||||
most = most.max(stack.len());
|
||||
if w == 0 || h == 0 {
|
||||
return image::DynamicImage::ImageRgba8(rgba);
|
||||
}
|
||||
most
|
||||
}
|
||||
|
||||
/// Every pixel on the image's edge, the flood-fill seeds.
|
||||
fn border_pixels(w: u32, h: u32) -> Vec<(u32, u32)> {
|
||||
let mut stack = Vec::new();
|
||||
let bg = rgba.get_pixel(0, 0).0;
|
||||
let close = |p: [u8; 4]| (0..3).map(|i| (p[i] as f32 - bg[i] as f32).abs()).fold(0.0f32, f32::max) <= tol;
|
||||
let mut seen = vec![false; (w as usize) * (h as usize)];
|
||||
let mut stack: Vec<(u32, u32)> = Vec::new();
|
||||
for x in 0..w {
|
||||
stack.push((x, 0));
|
||||
stack.push((x, h - 1));
|
||||
@@ -1430,47 +1372,46 @@ fn border_pixels(w: u32, h: u32) -> Vec<(u32, u32)> {
|
||||
stack.push((0, y));
|
||||
stack.push((w - 1, y));
|
||||
}
|
||||
stack
|
||||
}
|
||||
|
||||
/// The horizontal run of open pixels around `(x, y)`, both ends included.
|
||||
fn run_span(x: u32, y: u32, w: u32, open: impl Fn(u32, u32) -> bool) -> (u32, u32) {
|
||||
let (mut x0, mut x1) = (x, x);
|
||||
while x0 > 0 && open(x0 - 1, y) {
|
||||
x0 -= 1;
|
||||
}
|
||||
while x1 + 1 < w && open(x1 + 1, y) {
|
||||
x1 += 1;
|
||||
}
|
||||
(x0, x1)
|
||||
}
|
||||
|
||||
/// Queues the first pixel of each open run in row `ny` between `x0` and `x1`.
|
||||
fn push_open_runs(stack: &mut Vec<(u32, u32)>, x0: u32, x1: u32, ny: u32, open: impl Fn(u32, u32) -> bool) {
|
||||
let mut xi = x0;
|
||||
while xi <= x1 {
|
||||
if open(xi, ny) {
|
||||
stack.push((xi, ny));
|
||||
while xi <= x1 && open(xi, ny) {
|
||||
xi += 1;
|
||||
}
|
||||
} else {
|
||||
xi += 1;
|
||||
while let Some((x, y)) = stack.pop() {
|
||||
let i = (y as usize) * (w as usize) + x as usize;
|
||||
if seen.get(i).copied().unwrap_or(true) {
|
||||
continue;
|
||||
}
|
||||
if let Some(sv) = seen.get_mut(i) {
|
||||
*sv = true;
|
||||
}
|
||||
let p = rgba.get_pixel(x, y).0;
|
||||
if !close(p) {
|
||||
continue;
|
||||
}
|
||||
rgba.put_pixel(x, y, image::Rgba([p[0], p[1], p[2], 0]));
|
||||
if x > 0 {
|
||||
stack.push((x - 1, y));
|
||||
}
|
||||
if x + 1 < w {
|
||||
stack.push((x + 1, y));
|
||||
}
|
||||
if y > 0 {
|
||||
stack.push((x, y - 1));
|
||||
}
|
||||
if y + 1 < h {
|
||||
stack.push((x, y + 1));
|
||||
}
|
||||
}
|
||||
image::DynamicImage::ImageRgba8(rgba)
|
||||
}
|
||||
|
||||
fn frame(img: image::DynamicImage, border: u32, color: [u8; 4]) -> Result<image::DynamicImage, CmdError> {
|
||||
let (w, h) = grown(img.width(), img.height(), border).ok_or_else(too_large)?;
|
||||
let mut out = canvas(w, h, color)?;
|
||||
fn frame(img: image::DynamicImage, border: u32, color: [u8; 4]) -> image::DynamicImage {
|
||||
let (w, h) = (img.width(), img.height());
|
||||
let mut out = image::RgbaImage::from_pixel(w + 2 * border, h + 2 * border, image::Rgba(color));
|
||||
image::imageops::overlay(&mut out, &img.to_rgba8(), border as i64, border as i64);
|
||||
Ok(image::DynamicImage::ImageRgba8(out))
|
||||
image::DynamicImage::ImageRgba8(out)
|
||||
}
|
||||
|
||||
fn picture_style(img: image::DynamicImage, style: &str) -> Result<image::DynamicImage, CmdError> {
|
||||
fn picture_style(img: image::DynamicImage, style: &str) -> image::DynamicImage {
|
||||
let m = (img.width().max(img.height()) / 40).max(4);
|
||||
Ok(match style {
|
||||
"thickFrame" => frame(frame(img, m * 2, [255, 255, 255, 255])?, m / 2, [40, 40, 40, 255])?,
|
||||
match style {
|
||||
"thickFrame" => frame(frame(img, m * 2, [255, 255, 255, 255]), m / 2, [40, 40, 40, 255]),
|
||||
"rounded" | "softEdge" => {
|
||||
let mut rgba = img.to_rgba8();
|
||||
let (w, h) = rgba.dimensions();
|
||||
@@ -1493,47 +1434,18 @@ fn picture_style(img: image::DynamicImage, style: &str) -> Result<image::Dynamic
|
||||
"shadow" => {
|
||||
let (w, h) = (img.width(), img.height());
|
||||
let off = m;
|
||||
let (pw, ph) = grown(w, h, off).ok_or_else(too_large)?;
|
||||
let mut out = canvas(pw, ph, [0, 0, 0, 0])?;
|
||||
// A rectangle of alpha 90 the picture's size, one and a half margins in, blurred. Its
|
||||
// blur is the product of its sides' blurs along each axis, so it is drawn from two
|
||||
// profiles, not by blurring every pixel with a kernel that grows with the picture (#3647).
|
||||
let (sigma, start) = (off as f32 / 2.0, off + off / 2);
|
||||
let (xs, ys) = (blurred_span(pw, start, w, sigma), blurred_span(ph, start, h, sigma));
|
||||
for (x, y, p) in out.enumerate_pixels_mut() {
|
||||
let k = xs.get(x as usize).copied().unwrap_or(0.0) * ys.get(y as usize).copied().unwrap_or(0.0);
|
||||
p.0[3] = (90.0 * k).round() as u8;
|
||||
}
|
||||
let mut out = image::RgbaImage::from_pixel(w + off * 2, h + off * 2, image::Rgba([0, 0, 0, 0]));
|
||||
let shadow = image::RgbaImage::from_pixel(w, h, image::Rgba([0, 0, 0, 90]));
|
||||
let shadow = image::DynamicImage::ImageRgba8(shadow);
|
||||
let mut padded = image::RgbaImage::from_pixel(w + off * 2, h + off * 2, image::Rgba([0, 0, 0, 0]));
|
||||
image::imageops::overlay(&mut padded, &shadow.to_rgba8(), (off + off / 2) as i64, (off + off / 2) as i64);
|
||||
let blurred = image::DynamicImage::ImageRgba8(padded).blur(off as f32 / 2.0).to_rgba8();
|
||||
image::imageops::overlay(&mut out, &blurred, 0, 0);
|
||||
image::imageops::overlay(&mut out, &img.to_rgba8(), 0, 0);
|
||||
image::DynamicImage::ImageRgba8(out)
|
||||
}
|
||||
_ => frame(img, m, [255, 255, 255, 255])?,
|
||||
})
|
||||
}
|
||||
|
||||
/// How much of a span `len` long from `start` covers each of `n` positions once blurred with the
|
||||
/// Gaussian `image::imageops::blur` uses for `sigma` (outside the `n` positions counts as clear):
|
||||
/// 0 to 1. Each position is one difference of the kernel's running sums.
|
||||
fn blurred_span(n: u32, start: u32, len: u32, sigma: f32) -> Vec<f32> {
|
||||
let kernel = gaussian_kernel(sigma);
|
||||
let mut sums = vec![0.0f32];
|
||||
for k in &kernel {
|
||||
sums.push(sums.last().copied().unwrap_or(0.0) + k);
|
||||
_ => frame(img, m, [255, 255, 255, 255]),
|
||||
}
|
||||
let (size, r) = (kernel.len() as i64, (kernel.len() / 2) as i64);
|
||||
let sum_to = |k: i64| sums.get(k.clamp(0, size) as usize).copied().unwrap_or(0.0);
|
||||
let (start, end) = (i64::from(start), i64::from(start) + i64::from(len));
|
||||
// Position i takes weight k from i + k - r, inside the span for k in [start - i + r, end - i + r).
|
||||
(0..i64::from(n)).map(|i| sum_to(end - i + r) - sum_to(start - i + r)).collect()
|
||||
}
|
||||
|
||||
/// The normalized kernel `image::imageops::blur` (0.25) uses for `sigma`: odd, about 6.7 σ wide.
|
||||
fn gaussian_kernel(sigma: f32) -> Vec<f32> {
|
||||
let size = ((((sigma - 0.8) / 0.3 + 1.0) * 2.0 + 1.0).max(3.0) as usize) | 1;
|
||||
let mean = (size / 2) as f32;
|
||||
let weights: Vec<f32> = (0..size).map(|x| (-0.5 * ((x as f32 - mean) / sigma).powi(2)).exp()).collect();
|
||||
let total: f32 = weights.iter().sum();
|
||||
weights.into_iter().map(|v| v / total).collect()
|
||||
}
|
||||
|
||||
/// The selected pictures, shapes, text boxes and groups: the one the selection holds, then
|
||||
@@ -2070,112 +1982,6 @@ mod tests {
|
||||
assert_eq!(s.run("arrange.selectionPane", &json!({})).unwrap().as_array().map(Vec::len), Some(1));
|
||||
}
|
||||
|
||||
/// A PNG whose header (valid CRC) declares `w` × `h` pixels, with one pixel of data.
|
||||
fn header_only_png(w: u32, h: u32) -> Vec<u8> {
|
||||
let mut b = png(8, 8);
|
||||
b[16..20].copy_from_slice(&w.to_be_bytes());
|
||||
b[20..24].copy_from_slice(&h.to_be_bytes());
|
||||
let mut crc = 0xFFFF_FFFFu32;
|
||||
for &byte in &b[12..29] {
|
||||
crc ^= u32::from(byte);
|
||||
for _ in 0..8 {
|
||||
crc = if crc & 1 == 1 { 0xEDB8_8320 ^ (crc >> 1) } else { crc >> 1 };
|
||||
}
|
||||
}
|
||||
b[29..33].copy_from_slice(&(!crc).to_be_bytes());
|
||||
b
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn picture_edits_refuse_an_over_budget_header_before_decoding() {
|
||||
// Inserting reads the size only; every edit then refuses it from the header (#3644, #3647).
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
s.run("insert.picture", &json!({"data": super::super::insert::base64_encode(&header_only_png(20_000, 20_000))})).unwrap();
|
||||
let media = s.doc.media.len();
|
||||
for (id, v) in [
|
||||
("picture.corrections", json!({"brightness": 10})),
|
||||
("picture.color", json!({"mode": "sepia"})),
|
||||
("picture.removeBackground", json!({})),
|
||||
("picture.style", json!({"style": "shadow"})),
|
||||
] {
|
||||
let e = s.run(id, &v).unwrap_err().to_string();
|
||||
assert!(e.contains("too large to edit"), "{id}: {e}");
|
||||
}
|
||||
assert_eq!(s.doc.media.len(), media, "nothing was added");
|
||||
}
|
||||
|
||||
/// The Shadow style before #3647: the rectangle blurred with `image`'s Gaussian.
|
||||
fn blurred_shadow(img: &image::DynamicImage) -> image::RgbaImage {
|
||||
let (w, h) = (img.width(), img.height());
|
||||
let off = (w.max(h) / 40).max(4);
|
||||
let mut out = image::RgbaImage::from_pixel(w + off * 2, h + off * 2, image::Rgba([0, 0, 0, 0]));
|
||||
let shadow = image::RgbaImage::from_pixel(w, h, image::Rgba([0, 0, 0, 90]));
|
||||
let mut padded = image::RgbaImage::from_pixel(w + off * 2, h + off * 2, image::Rgba([0, 0, 0, 0]));
|
||||
image::imageops::overlay(&mut padded, &shadow, (off + off / 2) as i64, (off + off / 2) as i64);
|
||||
let blurred = image::DynamicImage::ImageRgba8(padded).blur(off as f32 / 2.0).to_rgba8();
|
||||
image::imageops::overlay(&mut out, &blurred, 0, 0);
|
||||
image::imageops::overlay(&mut out, &img.to_rgba8(), 0, 0);
|
||||
out
|
||||
}
|
||||
|
||||
/// #3647: the shadow drawn from two profiles looks like the blur it replaces.
|
||||
#[test]
|
||||
fn the_shadow_style_matches_the_blur_it_replaces() {
|
||||
for (w, h) in [(80, 40), (400, 220), (37, 300)] {
|
||||
let img = image::DynamicImage::ImageRgba8(image::RgbaImage::from_fn(w, h, |x, y| {
|
||||
image::Rgba([(x * 7) as u8, (y * 5) as u8, 90, if (x + y) % 3 == 0 { 255 } else { 120 }])
|
||||
}));
|
||||
let new = picture_style(img.clone(), "shadow").unwrap().to_rgba8();
|
||||
let old = blurred_shadow(&img);
|
||||
assert_eq!(new.dimensions(), old.dimensions());
|
||||
let worst = new.pixels().zip(old.pixels()).flat_map(|(a, b)| a.0.into_iter().zip(b.0).map(|(a, b)| a.abs_diff(b))).max();
|
||||
assert!(worst.is_some_and(|d| d <= 2), "{w} x {h}: {worst:?}");
|
||||
}
|
||||
}
|
||||
|
||||
/// #3647: Remove Background fills a span at a time, so its stack stays near the border's seeds
|
||||
/// instead of growing with the picture, and it clears the same region.
|
||||
#[test]
|
||||
fn remove_background_keeps_a_small_stack() {
|
||||
let mut uniform = image::RgbaImage::from_pixel(2000, 2000, image::Rgba([250, 250, 250, 255]));
|
||||
let most = clear_background(&mut uniform, 30.0);
|
||||
assert!(most <= 4 * (2000 + 2000) + 8, "{most}");
|
||||
assert!(uniform.pixels().all(|p| p.0[3] == 0));
|
||||
// A square ring keeps what it encloses: only background reachable from the border goes.
|
||||
let mut ring = image::RgbaImage::from_fn(60, 60, |x, y| {
|
||||
let d = (x as i32 - 30).abs().max((y as i32 - 30).abs());
|
||||
if (10..14).contains(&d) { image::Rgba([200, 20, 20, 255]) } else { image::Rgba([255, 255, 255, 255]) }
|
||||
});
|
||||
clear_background(&mut ring, 30.0);
|
||||
assert_eq!(ring.get_pixel(0, 0).0, [255, 255, 255, 0], "outside: cleared, colour kept");
|
||||
assert_eq!(ring.get_pixel(5, 30).0[3], 0);
|
||||
assert_eq!(ring.get_pixel(30, 18).0[3], 255, "the ring");
|
||||
assert_eq!(ring.get_pixel(30, 30).0[3], 255, "inside the ring");
|
||||
assert_eq!(clear_background(&mut image::RgbaImage::new(0, 0), 30.0), 0);
|
||||
}
|
||||
|
||||
/// #3611: a style or border is drawn only when the picture it makes can still be drawn. On a
|
||||
/// 1 × 2,000,000 picture Thick Frame asked for about 1.8 TB and aborted the process.
|
||||
#[test]
|
||||
fn picture_styles_and_borders_check_the_size_they_draw() {
|
||||
let mut thin = Vec::new();
|
||||
image::GrayImage::new(1, 2_000_000).write_to(&mut std::io::Cursor::new(&mut thin), image::ImageFormat::Png).unwrap();
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
s.run("insert.picture", &json!({"data": super::super::insert::base64_encode(&thin)})).unwrap();
|
||||
let media = s.doc.media.len();
|
||||
for (id, v) in [
|
||||
("picture.style", json!({"style": "thickFrame"})),
|
||||
("picture.style", json!({"style": "shadow"})),
|
||||
("picture.style", json!({})),
|
||||
("picture.border", json!({"width": 200})),
|
||||
] {
|
||||
let e = s.run(id, &v).unwrap_err().to_string();
|
||||
assert!(e.contains("too large to edit"), "{id} {v}: {e}");
|
||||
}
|
||||
assert_eq!(s.doc.media.len(), media, "nothing was drawn");
|
||||
assert_eq!(grown(u32::MAX, 1, 1), None);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn picture_pipeline() {
|
||||
let mut s = Session::new(wordcraft_doc::Document::new());
|
||||
|
||||
@@ -263,8 +263,6 @@ fn folder_setting(key: &str, f: &str, local_only: bool) -> Result<Option<String>
|
||||
if !path.is_absolute() {
|
||||
return Err(CmdError::Params(format!("`{key}`: {f:?} isn't a full path")));
|
||||
}
|
||||
// Its files are read from now on, and on every start: only from inside the read root (#3737).
|
||||
crate::file_access::check_read(&path.to_string_lossy()).map_err(CmdError::Failed)?;
|
||||
if !path.is_dir() {
|
||||
return Err(CmdError::Failed(format!("no folder at {}", path.display())));
|
||||
}
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user