mirror of
https://github.com/storytold/deckcraft.git
synced 2026-10-26 03:27:03 -06:00
393 lines
16 KiB
Rust
393 lines
16 KiB
Rust
//! The native `.deckcraft` format: a zip holding `presentation.json` (the model, minus media
|
||
//! bytes), `media/<id>` files with the embedded media, and `mimetype`. Readers ignore unknown
|
||
//! fields, so newer files open in older builds with what those builds understand.
|
||
#![deny(clippy::unwrap_used, clippy::expect_used, clippy::panic, clippy::unimplemented, clippy::todo, clippy::unreachable)]
|
||
#![forbid(unsafe_code)]
|
||
|
||
use std::io::{Cursor, Read, Write};
|
||
use std::sync::Arc;
|
||
|
||
use deckcraft_model::text::{Paragraph, TextBody};
|
||
use deckcraft_model::{LayoutType, Presentation, defaults};
|
||
|
||
pub const MIMETYPE: &str = "application/vnd.storyteller.deckcraft+zip";
|
||
/// Files written before the SlideCraft → DeckCraft rename (`.slidecraft`) are still read.
|
||
pub const LEGACY_MIMETYPE: &str = concat!("application/vnd.storyteller.", "slide", "craft+zip");
|
||
pub const EXTENSION: &str = "deckcraft";
|
||
/// Format version written into the manifest.
|
||
pub const VERSION: u32 = 1;
|
||
/// Largest entry we will inflate (guards against zip bombs).
|
||
const MAX_ENTRY: u64 = 2 << 30;
|
||
|
||
#[derive(Debug, thiserror::Error)]
|
||
pub enum FormatError {
|
||
#[error("not a DeckCraft file: {0}")]
|
||
NotOurs(String),
|
||
#[error("damaged file: {0}")]
|
||
Damaged(String),
|
||
#[error("write failed: {0}")]
|
||
Write(String),
|
||
}
|
||
|
||
pub type Result<T> = std::result::Result<T, FormatError>;
|
||
|
||
#[derive(serde::Serialize, serde::Deserialize)]
|
||
struct Manifest {
|
||
format: String,
|
||
version: u32,
|
||
generator: String,
|
||
presentation: Presentation,
|
||
}
|
||
|
||
/// Serialize a presentation.
|
||
pub fn save(p: &Presentation) -> Result<Vec<u8>> {
|
||
let mut buf = Cursor::new(Vec::new());
|
||
{
|
||
let mut z = zip::ZipWriter::new(&mut buf);
|
||
let stored = zip::write::SimpleFileOptions::default().compression_method(zip::CompressionMethod::Stored);
|
||
let deflate = zip::write::SimpleFileOptions::default().compression_method(zip::CompressionMethod::Deflated);
|
||
let w = |e: zip::result::ZipError| FormatError::Write(e.to_string());
|
||
z.start_file("mimetype", stored).map_err(w)?;
|
||
z.write_all(MIMETYPE.as_bytes()).map_err(|e| FormatError::Write(e.to_string()))?;
|
||
let m = Manifest {
|
||
format: "deckcraft".into(),
|
||
version: VERSION,
|
||
generator: format!("DeckCraft {}", env!("CARGO_PKG_VERSION")),
|
||
presentation: p.clone(),
|
||
};
|
||
let json = serde_json::to_vec(&m).map_err(|e| FormatError::Write(e.to_string()))?;
|
||
z.start_file("presentation.json", deflate).map_err(w)?;
|
||
z.write_all(&json).map_err(|e| FormatError::Write(e.to_string()))?;
|
||
for item in &p.media {
|
||
if item.link.is_some() && item.data.is_empty() {
|
||
continue;
|
||
}
|
||
// Already-compressed media is stored, not deflated again.
|
||
let opts = if is_compressed(&item.content_type) { stored } else { deflate };
|
||
z.start_file(format!("media/{}", item.id.0), opts).map_err(w)?;
|
||
z.write_all(&item.data).map_err(|e| FormatError::Write(e.to_string()))?;
|
||
}
|
||
z.finish().map_err(w)?;
|
||
}
|
||
Ok(buf.into_inner())
|
||
}
|
||
|
||
fn is_compressed(ct: &str) -> bool {
|
||
ct.starts_with("image/jpeg")
|
||
|| ct.starts_with("image/png")
|
||
|| ct.starts_with("video/")
|
||
|| ct.starts_with("audio/mpeg")
|
||
|| ct.starts_with("audio/mp4")
|
||
|| ct.contains("zip")
|
||
}
|
||
|
||
/// Does this look like a `.deckcraft` file?
|
||
pub fn sniff(bytes: &[u8]) -> bool {
|
||
bytes.starts_with(b"PK") && [MIMETYPE, LEGACY_MIMETYPE].iter().any(|m| bytes.windows(m.len()).take(200).any(|w| w == m.as_bytes()))
|
||
}
|
||
|
||
fn read_entry(z: &mut zip::ZipArchive<Cursor<&[u8]>>, name: &str) -> Result<Vec<u8>> {
|
||
let f = z.by_name(name).map_err(|e| FormatError::Damaged(format!("{name}: {e}")))?;
|
||
if f.size() > MAX_ENTRY {
|
||
return Err(FormatError::Damaged(format!("{name} is too large")));
|
||
}
|
||
let mut out = Vec::with_capacity(f.size().min(64 << 20) as usize);
|
||
f.take(MAX_ENTRY).read_to_end(&mut out).map_err(|e| FormatError::Damaged(format!("{name}: {e}")))?;
|
||
Ok(out)
|
||
}
|
||
|
||
/// Read a presentation. Embedded media that is missing or damaged is left empty; use
|
||
/// [`load_checked`] to learn which.
|
||
pub fn load(bytes: &[u8]) -> Result<Presentation> {
|
||
load_checked(bytes).map(|(p, _)| p)
|
||
}
|
||
|
||
/// Read a presentation and list the embedded media that couldn't be read (missing, too large
|
||
/// or failing its CRC), one message each. Linked media saved without its bytes isn't damage.
|
||
pub fn load_checked(bytes: &[u8]) -> Result<(Presentation, Vec<String>)> {
|
||
let mut z = zip::ZipArchive::new(Cursor::new(bytes)).map_err(|e| FormatError::NotOurs(e.to_string()))?;
|
||
let json = read_entry(&mut z, "presentation.json").map_err(|_| FormatError::NotOurs("no presentation.json".into()))?;
|
||
let m: Manifest = serde_json::from_slice(&json).map_err(|e| FormatError::Damaged(e.to_string()))?;
|
||
if m.format != "deckcraft" {
|
||
return Err(FormatError::NotOurs(m.format));
|
||
}
|
||
let mut p = m.presentation;
|
||
let mut damaged = vec![];
|
||
for item in &mut p.media {
|
||
let name = format!("media/{}", item.id.0);
|
||
let linked_only = item.link.is_some() && z.index_for_name(&name).is_none();
|
||
match read_entry(&mut z, &name) {
|
||
Ok(data) => item.data = Arc::new(data),
|
||
Err(_) if linked_only => {}
|
||
Err(e) => damaged.push(e.to_string()),
|
||
}
|
||
}
|
||
repair(&mut p);
|
||
Ok((p, damaged))
|
||
}
|
||
|
||
/// Make a loaded presentation safe to edit: a master exists, every slide's layout exists,
|
||
/// ids are unique, tables are rectangular.
|
||
pub fn repair(p: &mut Presentation) {
|
||
if p.masters.is_empty() {
|
||
let fresh = defaults::blank_presentation(p.slide_size, Default::default(), false);
|
||
p.masters = fresh.masters;
|
||
}
|
||
p.fix_next_id();
|
||
let first_layout = p.masters.first().and_then(|m| m.layouts.first()).map(|l| l.id).unwrap_or_default();
|
||
let mut seen = std::collections::HashSet::new();
|
||
for i in 0..p.slides.len() {
|
||
let needs_layout = p.slides.get(i).is_some_and(|s| p.layout(s.layout).is_none());
|
||
let dup = p.slides.get(i).is_some_and(|s| !seen.insert(s.id));
|
||
if needs_layout || dup {
|
||
let new_id = if dup { Some(deckcraft_model::SlideId(p.alloc_id())) } else { None };
|
||
if let Some(s) = p.slides.get_mut(i) {
|
||
let s = Arc::make_mut(s);
|
||
if needs_layout {
|
||
s.layout = first_layout;
|
||
}
|
||
if let Some(id) = new_id {
|
||
s.id = id;
|
||
}
|
||
}
|
||
}
|
||
}
|
||
if !(p.slide_size.width.is_finite() && p.slide_size.width >= 1.0 && p.slide_size.height.is_finite() && p.slide_size.height >= 1.0) {
|
||
p.slide_size = defaults::WIDE;
|
||
}
|
||
p.slide_size.width = p.slide_size.width.min(4000.0);
|
||
p.slide_size.height = p.slide_size.height.min(4000.0);
|
||
}
|
||
|
||
/// Slides read from an outline: each title with its body paragraphs (level 0–8, text).
|
||
type OutlineSlides = Vec<(String, Vec<(u8, String)>)>;
|
||
|
||
/// Append a Title and Content slide per outline entry; the number made.
|
||
fn add_outline_slides(p: &mut Presentation, slides: OutlineSlides) -> usize {
|
||
let Some(lay) = defaults::layout_of_kind(p, LayoutType::TitleAndContent) else { return 0 };
|
||
let made = slides.len();
|
||
for (title, body) in slides {
|
||
let mut s = defaults::new_slide(p, lay);
|
||
if let Some(t) = s.shapes.first_mut() {
|
||
t.text = Some(TextBody::from_text(&title));
|
||
}
|
||
if let Some(b) = s.shapes.get_mut(1) {
|
||
for (level, line) in body {
|
||
let tb = b.text.get_or_insert_with(TextBody::default);
|
||
let mut para = Paragraph::new(&line);
|
||
para.level = level.min(8);
|
||
tb.paragraphs.push(para);
|
||
}
|
||
}
|
||
p.slides.push(Arc::new(s));
|
||
}
|
||
made
|
||
}
|
||
|
||
/// Outline text: one slide per unindented line; tab-indented lines become body bullets at their
|
||
/// level (Insert ▸ Slides from Outline, View ▸ Outline copy).
|
||
pub fn outline_to_slides(p: &mut Presentation, text: &str) -> usize {
|
||
let mut slides: OutlineSlides = Vec::new();
|
||
for raw in text.lines() {
|
||
if raw.trim().is_empty() {
|
||
continue;
|
||
}
|
||
let tabs = raw.chars().take_while(|c| *c == '\t').count();
|
||
let spaces = raw.chars().take_while(|c| *c == ' ').count() / 2;
|
||
let level = tabs.max(spaces);
|
||
let line = raw.trim().trim_start_matches(['-', '*', '•']).trim().to_string();
|
||
if level == 0 {
|
||
slides.push((line, Vec::new()));
|
||
} else if let Some((_, body)) = slides.last_mut() {
|
||
body.push(((level - 1).min(8) as u8, line));
|
||
}
|
||
}
|
||
add_outline_slides(p, slides)
|
||
}
|
||
|
||
/// An ATX heading's text (CommonMark: up to three spaces, one to six `#`, then a space or the end;
|
||
/// a closing run of `#` is dropped).
|
||
fn md_heading(line: &str) -> Option<&str> {
|
||
let t = line.strip_prefix(" ").or_else(|| line.strip_prefix(" ")).or_else(|| line.strip_prefix(' ')).unwrap_or(line);
|
||
let hashes = t.chars().take_while(|c| *c == '#').count();
|
||
let rest = t.get(hashes..)?;
|
||
if !(1..=6).contains(&hashes) || !(rest.is_empty() || rest.starts_with([' ', '\t'])) {
|
||
return None;
|
||
}
|
||
let rest = rest.trim();
|
||
let closed = rest.trim_end_matches('#');
|
||
Some(if closed.is_empty() || closed.ends_with([' ', '\t']) { closed.trim_end() } else { rest })
|
||
}
|
||
|
||
/// A list item's nesting level (two spaces or a tab each) and text: `-`, `*` or `+`, or digits
|
||
/// then `.` or `)`, followed by a space.
|
||
fn md_list_item(line: &str) -> Option<(u8, &str)> {
|
||
let indent: usize = line.chars().take_while(|c| matches!(c, ' ' | '\t')).map(|c| if c == '\t' { 2 } else { 1 }).sum();
|
||
let t = line.trim_start();
|
||
let rest = match t.strip_prefix(['-', '*', '+']) {
|
||
Some(r) => r,
|
||
None => {
|
||
let digits = t.chars().take_while(char::is_ascii_digit).count();
|
||
if digits == 0 || digits > 9 {
|
||
return None;
|
||
}
|
||
t.get(digits..)?.strip_prefix(['.', ')'])?
|
||
}
|
||
};
|
||
rest.starts_with([' ', '\t']).then(|| ((indent / 2).min(8) as u8, rest.trim()))
|
||
}
|
||
|
||
/// Markdown outline text (`.md`): each heading starts a slide with its text as the title; list
|
||
/// items under it become bullets nested by their indentation, other lines body paragraphs. Text
|
||
/// before the first heading is skipped, as indented lines before the first title are in the
|
||
/// plain format. A file without headings is read as a plain outline ([`outline_to_slides`]).
|
||
pub fn markdown_to_slides(p: &mut Presentation, text: &str) -> usize {
|
||
if !text.lines().any(|l| md_heading(l).is_some()) {
|
||
return outline_to_slides(p, text);
|
||
}
|
||
let mut slides: OutlineSlides = Vec::new();
|
||
for line in text.lines() {
|
||
if line.trim().is_empty() {
|
||
continue;
|
||
}
|
||
if let Some(title) = md_heading(line) {
|
||
slides.push((title.to_string(), Vec::new()));
|
||
} else if let Some((_, body)) = slides.last_mut() {
|
||
let (level, text) = md_list_item(line).unwrap_or((0, line.trim()));
|
||
body.push((level, text.to_string()));
|
||
}
|
||
}
|
||
add_outline_slides(p, slides)
|
||
}
|
||
|
||
/// The deck as outline text (titles unindented, body paragraphs tab-indented by level).
|
||
pub fn slides_to_outline(p: &Presentation) -> String {
|
||
let mut out = String::new();
|
||
for s in &p.slides {
|
||
out.push_str(&s.title());
|
||
out.push('\n');
|
||
for sh in &s.shapes {
|
||
if sh.ph_type().is_some_and(|k| k.is_title()) {
|
||
continue;
|
||
}
|
||
if let Some(t) = &sh.text {
|
||
for para in &t.paragraphs {
|
||
if para.is_empty() {
|
||
continue;
|
||
}
|
||
for _ in 0..=para.level {
|
||
out.push('\t');
|
||
}
|
||
out.push_str(¶.text().replace('\u{b}', " "));
|
||
out.push('\n');
|
||
}
|
||
}
|
||
}
|
||
}
|
||
out
|
||
}
|
||
|
||
#[cfg(test)]
|
||
mod tests {
|
||
use super::*;
|
||
|
||
/// The same package without the entries `drop` names.
|
||
fn without(bytes: &[u8], drop: &str) -> Vec<u8> {
|
||
let mut z = zip::ZipArchive::new(Cursor::new(bytes)).unwrap();
|
||
let mut out = zip::ZipWriter::new(Cursor::new(Vec::new()));
|
||
for i in 0..z.len() {
|
||
let f = z.by_index(i).unwrap();
|
||
if f.name() != drop {
|
||
out.raw_copy_file(f).unwrap();
|
||
}
|
||
}
|
||
out.finish().unwrap().into_inner()
|
||
}
|
||
|
||
#[test]
|
||
fn missing_or_corrupt_media_is_reported() {
|
||
let mut p = Presentation::default();
|
||
let id = p.add_media("a.png", "image/png", b"PICTURE-BYTES-0123456789".to_vec());
|
||
let good = save(&p).unwrap();
|
||
assert!(load_checked(&good).unwrap().1.is_empty());
|
||
// Missing member.
|
||
let (q, damaged) = load_checked(&without(&good, &format!("media/{}", id.0))).unwrap();
|
||
assert_eq!(damaged.len(), 1, "{damaged:?}");
|
||
assert!(q.media(id).is_some_and(|m| m.data.is_empty()));
|
||
// Bad CRC: PNGs are stored, so their bytes sit in the file as they are.
|
||
let mut bad = good.clone();
|
||
let at = bad.windows(13).position(|w| w == b"PICTURE-BYTES").unwrap();
|
||
bad[at] ^= 0xff;
|
||
let (_, damaged) = load_checked(&bad).unwrap();
|
||
assert_eq!(damaged.len(), 1, "{damaged:?}");
|
||
// Linked media saved without its bytes isn't damage.
|
||
let mut l = Presentation::default();
|
||
let lid = l.add_media("v.mp4", "video/mp4", vec![]);
|
||
if let Some(m) = l.media.iter_mut().find(|m| m.id == lid) {
|
||
m.link = Some("file:///movies/v.mp4".into());
|
||
}
|
||
assert!(load_checked(&save(&l).unwrap()).unwrap().1.is_empty());
|
||
}
|
||
|
||
#[test]
|
||
fn roundtrip_with_media() {
|
||
let mut p = Presentation::default();
|
||
let id = p.add_media("a.png", "image/png", vec![1, 2, 3, 4]);
|
||
let bytes = save(&p).unwrap();
|
||
assert!(sniff(&bytes));
|
||
let q = load(&bytes).unwrap();
|
||
assert_eq!(q.slides.len(), p.slides.len());
|
||
assert_eq!(q.media(id).map(|m| m.data.as_slice()), Some(&[1u8, 2, 3, 4][..]));
|
||
assert_eq!(serde_json::to_value(&q.slides).unwrap(), serde_json::to_value(&p.slides).unwrap());
|
||
}
|
||
|
||
#[test]
|
||
fn garbage_is_an_error_not_a_panic() {
|
||
assert!(load(b"").is_err());
|
||
assert!(load(b"PK\x03\x04garbage").is_err());
|
||
assert!(load(&[0u8; 1000]).is_err());
|
||
let mut b = save(&Presentation::default()).unwrap();
|
||
let n = b.len();
|
||
b.truncate(n / 2);
|
||
assert!(load(&b).is_err());
|
||
}
|
||
|
||
#[test]
|
||
fn repair_fixes_missing_layout_and_size() {
|
||
let mut p = Presentation::default();
|
||
Arc::make_mut(&mut p.slides[0]).layout = deckcraft_model::LayoutId(99999);
|
||
p.slide_size.width = f64::NAN;
|
||
repair(&mut p);
|
||
assert!(p.validate().is_empty());
|
||
assert_eq!(p.slide_size.width, 960.0);
|
||
}
|
||
|
||
#[test]
|
||
fn outline_roundtrip() {
|
||
let mut p = defaults::blank_presentation(defaults::WIDE, Default::default(), false);
|
||
let n = outline_to_slides(&mut p, "Intro\n\tPoint one\n\t\tSub point\nSecond\n\tMore");
|
||
assert_eq!(n, 2);
|
||
let o = slides_to_outline(&p);
|
||
assert!(o.contains("Intro\n\tPoint one\n\t\tSub point\nSecond"), "{o}");
|
||
}
|
||
|
||
#[test]
|
||
fn markdown_headings_and_list_items() {
|
||
assert_eq!(md_heading("# Intro"), Some("Intro"));
|
||
assert_eq!(md_heading(" ### Deep ### "), Some("Deep"));
|
||
assert_eq!(md_heading("## C# and F#"), Some("C# and F#"));
|
||
assert_eq!(md_heading("#"), Some(""));
|
||
assert_eq!(md_heading("## Ünïcödé"), Some("Ünïcödé"));
|
||
for not in ["#5 priority", "####### seven", " # code", "text", "", "é#"] {
|
||
assert_eq!(md_heading(not), None, "{not:?}");
|
||
}
|
||
assert_eq!(md_list_item("- a"), Some((0, "a")));
|
||
assert_eq!(md_list_item(" * b"), Some((2, "b")));
|
||
assert_eq!(md_list_item("\t+ c"), Some((1, "c")));
|
||
assert_eq!(md_list_item("12) d"), Some((0, "d")));
|
||
for not in ["-x", "1.x", "1", "1234567890. long", "• bullet", "é. x", ""] {
|
||
assert_eq!(md_list_item(not), None, "{not:?}");
|
||
}
|
||
}
|
||
}
|