Use shared MTEXT escape decoding

Cargo.lock

@@ -1131,6 +1131,11 @@ dependencies = [
  "libc",
 ]
 
+[[package]]
+name = "craft-codepage"
+version = "0.1.0"
+source = "git+https://github.com/storytold/craft-libs?rev=6345459e1297ee85a7ca62b428a440691bd062c1#6345459e1297ee85a7ca62b428a440691bd062c1"
+
 [[package]]
 name = "craft-docpath"
 version = "0.1.0"
@@ -1148,9 +1153,11 @@ dependencies = [
 [[package]]
 name = "craft-dxf"
 version = "0.1.0"
-source = "git+https://github.com/storytold/craft-libs?rev=006e4df225fca57e95ead0fb1f91920900798a99#006e4df225fca57e95ead0fb1f91920900798a99"
+source = "git+https://github.com/storytold/craft-libs?rev=6345459e1297ee85a7ca62b428a440691bd062c1#6345459e1297ee85a7ca62b428a440691bd062c1"
 dependencies = [
+ "craft-codepage",
  "encoding_rs",
+ "kurbo",
  "log",
  "thiserror 2.0.21",
 ]

Cargo.toml

@@ -14,7 +14,7 @@ repository = "https://github.com/storytold/cadcraft"
 # internal
 cadcraft-geom = { path = "crates/geom" }
 # DXF tags: the shared reader and writer (moved from crates/dxf; craft-libs#791).
-craft-dxf = { git = "https://github.com/storytold/craft-libs", rev = "006e4df225fca57e95ead0fb1f91920900798a99" }
+craft-dxf = { git = "https://github.com/storytold/craft-libs", rev = "6345459e1297ee85a7ca62b428a440691bd062c1" }
 craft-dwg = { git = "https://github.com/storytold/craft-libs", rev = "7bb6c90076a0177d01cc4c482e7fd8dc9f31a8bf" }
 cadcraft-color = { path = "crates/color" }
 cadcraft-doc = { path = "crates/doc" }

crates/fonts/src/mtext.rs

@@ -225,11 +225,8 @@ const MAX_ATOMS: usize = 200_000;
 /// Scale of stacked text relative to the surrounding text.
 const STACK_SCALE: f64 = 0.7;
 
-/// Codes that take an argument terminated by `;` (besides `\S`, which has its own scan). Any
-/// other code stands alone, so a missing `;` must not swallow the text after it.
-pub(crate) fn takes_arg(code: char) -> bool {
-    matches!(code, 'f' | 'F' | 'H' | 'W' | 'Q' | 'T' | 'A' | 'C' | 'c' | 'p')
-}
+// Shared DXF escape handling and parameterized-command classification.
+pub(crate) use craft_dxf::mtext::{takes_arg, unescape_codes};
 
 /// The line-spacing factor to lay out with: AutoCAD's 0.25..=4 range, 1 for a missing or
 /// non-finite value.
@@ -237,77 +234,6 @@ pub(crate) fn sane_line_spacing(v: f64) -> f64 {
     if v.is_finite() && v > 0.0 { v.clamp(0.25, 4.0) } else { 1.0 }
 }
 
-/// Four hex digits at `chars[at..at + 4]`.
-fn hex4(chars: &[char], at: usize) -> Option<u32> {
-    let digits = chars.get(at..at.checked_add(4)?)?;
-    digits.iter().try_fold(0u32, |acc, c| Some(acc * 16 + c.to_digit(16)?))
-}
-
-/// Push a decoded character so that the tokenizers read it as text, never as markup.
-fn push_literal(out: &mut String, c: char) {
-    if matches!(c, '\\' | '{' | '}') {
-        out.push('\\');
-    }
-    out.push(c);
-}
-
-/// Decode `\U+hhhh` (one UTF-16 unit; a surrogate pair gives one character, a lone surrogate
-/// U+FFFD) and `\M+nhhhh` (double-byte East Asian characters) to the characters they stand
-/// for. Malformed escapes are left for the tokenizers, which drop the backslash and the code.
-/// An escaped backslash (`\\U+0041`) is not an escape.
-pub(crate) fn unescape_codes(s: &str) -> std::borrow::Cow<'_, str> {
-    if !s.contains("\\U+") && !s.contains("\\M+") {
-        return std::borrow::Cow::Borrowed(s);
-    }
-    let chars: Vec<char> = s.chars().collect();
-    let mut out = String::with_capacity(s.len());
-    let mut i = 0;
-    while let Some(&c) = chars.get(i) {
-        if c != '\\' {
-            out.push(c);
-            i += 1;
-            continue;
-        }
-        let (n, plus) = (chars.get(i + 1).copied(), chars.get(i + 2).copied());
-        if n == Some('U') && plus == Some('+') {
-            if let Some(unit) = hex4(&chars, i + 3) {
-                i += 7;
-                let low = |at: usize| {
-                    (chars.get(at) == Some(&'\\') && chars.get(at + 1) == Some(&'U') && chars.get(at + 2) == Some(&'+'))
-                        .then(|| hex4(&chars, at + 3))
-                        .flatten()
-                        .filter(|u| (0xDC00..0xE000).contains(u))
-                };
-                let ch = if (0xD800..0xDC00).contains(&unit) {
-                    match low(i) {
-                        Some(lo) => {
-                            i += 7;
-                            char::from_u32(0x10000 + ((unit - 0xD800) << 10) + (lo - 0xDC00))
-                        }
-                        None => None,
-                    }
-                } else {
-                    char::from_u32(unit)
-                };
-                push_literal(&mut out, ch.unwrap_or('\u{fffd}'));
-                continue;
-            }
-        } else if n == Some('M') && plus == Some('+') && chars.get(i + 3).is_some_and(char::is_ascii_digit) && hex4(&chars, i + 4).is_some() {
-            push_literal(&mut out, craft_dxf::mif_escape_at(&chars, i).unwrap_or('\u{fffd}'));
-            i += 8;
-            continue;
-        }
-        // Any other backslash pair stays as it is (and an escaped backslash stays one).
-        out.push('\\');
-        i += 1;
-        if let Some(&next) = chars.get(i) {
-            out.push(next);
-            i += 1;
-        }
-    }
-    std::borrow::Cow::Owned(out)
-}
-
 fn arg(chars: &[char], from: usize) -> (String, usize) {
     let mut j = from;
     let mut a = String::new();