SVG, FCP XML: bound XML entity expansion (#2129)

A few kilobytes of internal entities expanded to millions of nodes (each top-level
reference stays under roxmltree's per-reference cap), so a 1 MB file exhausted memory.
SVG keeps DTDs, which Illustrator's exports use for entities, but caps the expanded
document at 4 million nodes. Final Cut Pro XML and FCPXML, which FilmCraft parses
without a node limit, are refused when they declare an entity; real exports never do.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

crates/interchange/src/import.rs

@@ -143,6 +143,11 @@ pub fn import(
     if depth > MAX_DEPTH {
         return Err(Error::TooDeep(MAX_DEPTH));
     }
+    // FilmCraft's parser expands internal entities without a node limit (#2129); Final Cut Pro
+    // and FCPXML exports never declare any.
+    if matches!(format, TimelineFormat::Fcp7Xml | TimelineFormat::Fcpxml) && bytes.windows(8).any(|w| w == b"<!ENTITY") {
+        return Err(Error::XmlEntities);
+    }
     let fc_opts = filmcraft_interchange::ImportOptions {
         base_dir: opts.base_dir.clone(),
         edl_frame_rate: opts.edl_frame_rate.map(|f| {

crates/interchange/src/lib.rs

@@ -160,6 +160,8 @@ pub enum Error {
     NoComp(u64),
     #[error("the timeline document nests more than {0} levels deep")]
     TooDeep(usize),
+    #[error("the timeline document declares XML entities, which timeline files don't use and which can expand without bound")]
+    XmlEntities,
 }
 
 pub type Result<T> = std::result::Result<T, Error>;

crates/interchange/src/tests.rs

@@ -520,6 +520,21 @@ fn export_survives_a_precomp_layer_pointing_at_a_non_comp() {
     assert!(matches!(export(&p, footage, &ExportOptions::default(), &HashMap::new()), Err(Error::NoComp(_))));
 }
 
+#[test]
+fn xml_timelines_declaring_entities_are_refused() {
+    // #2129: the timeline parser expands internal entities without a node limit.
+    let dtd = "<!DOCTYPE x [<!ENTITY a \"aaaaaaaaaa\"><!ENTITY b \"&a;&a;&a;&a;&a;&a;&a;&a;&a;&a;\">]>";
+    let docs = [
+        (TimelineFormat::Fcp7Xml, format!("<?xml version=\"1.0\"?>{dtd}<xmeml version=\"5\">&b;</xmeml>")),
+        (TimelineFormat::Fcpxml, format!("<?xml version=\"1.0\"?>{dtd}<fcpxml version=\"1.9\">&b;</fcpxml>")),
+    ];
+    for (format, doc) in docs {
+        let mut p = Project::default();
+        let r = import(&mut p, doc.as_bytes(), Some(format), &ImportOptions::default(), &mut |_| None);
+        assert!(matches!(r, Err(Error::XmlEntities)), "{format:?}");
+    }
+}
+
 #[test]
 fn deeply_nested_documents_error_instead_of_overflowing_the_stack() {
     // A few hundred kilobytes nested 100 000 levels deep overflowed the parsers' stack.

crates/svg/src/lib.rs

@@ -1261,6 +1261,10 @@ fn xml_depth(text: &str) -> usize {
     max
 }
 
+/// Nodes a document may expand to: a few kilobytes of nested internal entities
+/// (`<!ENTITY b "&a;&a;…">`) otherwise expand to millions and exhaust memory.
+const MAX_XML_NODES: u32 = 4_000_000;
+
 /// Parse an SVG document.
 pub fn parse(bytes: &[u8]) -> Result<Doc, Error> {
     let text = String::from_utf8_lossy(bytes);
@@ -1268,7 +1272,8 @@ pub fn parse(bytes: &[u8]) -> Result<Doc, Error> {
     if xml_depth(text) > MAX_XML_DEPTH {
         return Err(Error::Xml(format!("elements nested more than {MAX_XML_DEPTH} levels deep")));
     }
-    let opts = roxmltree::ParsingOptions { allow_dtd: true, ..Default::default() };
+    // DTDs stay allowed (Illustrator's SVGs declare entities), but expansion is bounded (#2129).
+    let opts = roxmltree::ParsingOptions { allow_dtd: true, nodes_limit: MAX_XML_NODES, ..Default::default() };
     let xml = roxmltree::Document::parse_with_options(text, opts).map_err(|e| Error::Xml(e.to_string()))?;
     let root = xml.root_element();
     if root.tag_name().name() != "svg" {

crates/svg/src/tests.rs

@@ -220,6 +220,18 @@ fn transforms() {
     assert!((p.x - 6.0).abs() < 1e-9 && (p.y - 7.0).abs() < 1e-9);
 }
 
+#[test]
+fn nested_entities_are_refused_instead_of_expanding_without_bound() {
+    // #2129: each top-level `&b;` stays under roxmltree's per-reference cap but adds 12 500
+    // nodes, so 40 KB of references expands to 5 million.
+    let dtd = format!("<!ENTITY a \"{}\"><!ENTITY b \"{}\">", "<g/>".repeat(50), "&a;".repeat(250));
+    let bomb = format!("<?xml version=\"1.0\"?><!DOCTYPE svg [{dtd}]><svg xmlns=\"http://www.w3.org/2000/svg\">{}</svg>", "&b;".repeat(400));
+    assert!(matches!(crate::parse(bomb.as_bytes()), Err(crate::Error::Xml(_))));
+    // An ordinary entity, as Illustrator declares them, still parses.
+    let ok = "<?xml version=\"1.0\"?><!DOCTYPE svg [<!ENTITY w \"40\">]><svg xmlns=\"http://www.w3.org/2000/svg\"><rect width=\"&w;\" height=\"5\"/></svg>";
+    assert!(crate::parse(ok.as_bytes()).is_ok());
+}
+
 #[test]
 fn deep_nesting_errors_instead_of_overflowing_the_stack() {
     // 100 000 nested groups overflowed the XML parser's stack and aborted the process.