fix(render): bound xref streams, JBIG2 referred-to counts, saved graphics states and /W2 runs (#4350)

* fix(render): bound xref streams, JBIG2 referred-to counts, saved graphics states and /W2 runs

Input bounds and crash fixes in the vendored hayro crates, ported from
the snapshot craft-libs carries (storytold/craft-libs#176, #239, #241):

- hayro-syntax: an xref stream subsection must be backed by its decoded
  data. All-zero /W widths and a count longer than the data are refused
  before any entry is inserted; a tiny stream with a large /Size or
  /Index count inserted that many entries.
- hayro-syntax: a subsection running past object number u32::MAX is
  refused; start + i overflowed.
- hayro-syntax: the xref stream type field is read big-endian at its
  full /W[0] width, and /W[0] over 8 bytes is refused like /W[1].
- hayro-jbig2: the long-form referred-to segment count is capped at
  u16::MAX before its retention bytes are read; the 29-bit count went
  straight to Vec::with_capacity (about 2 GiB) and aborted the process.
- hayro-interpret: at most 4096 saved graphics states holding at most
  64 MiB together. A q past either limit is counted but not copied, and
  its Q still removes the clips added since.
- hayro-interpret: a /W2 run whose first CID is past MAX_CID stops there,
  as /W does; a first CID of 4294967295 discarded the whole /W2 table.

Each patch is marked in the source, has a regression test in its crate,
and is listed in vendor/README.md.

* fix(render): a binary usecmap counts against the CMap nesting limit; /W2 checks the CID first

- hayro-cmap: a binary CMap's usecmap passed its own depth to the nested
  parse, so a self-referencing binary CMap recursed until the stack
  overflowed. It now counts like a text usecmap (ledger hayro-cmap 2).
- hayro-interpret: the /W2 first-CID check runs before the run's metrics
  are read, so an incomplete run past MAX_CID no longer discards the rest
  of the table (ledger 52).

* fix(render): refuse xref-stream object numbers past i32::MAX

ObjectIdentifier stores an i32; a number between i32::MAX and u32::MAX
was cast to a negative object number (ledger hayro-syntax 26).

vendor/hayro-cmap/src/bcmap/mod.rs

@@ -109,10 +109,13 @@ pub(crate) fn parse<'a>(
             SEGMENT_USECMAP => {
                 let base_data = get_cmap(CMapName::from_bytes(payload))?;
 
+                // PdfCraft patch: a binary `usecmap` counts against the nesting limit as a text
+                // one does; passing the same depth let a self-referencing CMap recurse until the
+                // stack overflowed.
                 base = Some(Box::new(parse::parse_inner(
                     base_data,
                     get_cmap.clone(),
-                    depth,
+                    depth + 1,
                 )?));
             }
             SEGMENT_WMODE => {

vendor/hayro-cmap/src/lib.rs

@@ -737,6 +737,17 @@ endcodespacerange
         CMap::parse(&data, |_| None).unwrap()
     }
 
+    /// PdfCraft patch test: a binary CMap whose `usecmap` names itself stops at the nesting limit.
+    #[cfg(feature = "embed-cmaps")]
+    #[test]
+    fn self_referencing_binary_usecmap_stops_at_the_nesting_limit() {
+        // Header (magic, version, file length 19), then one usecmap segment (length 9) naming "Self".
+        const SELF_REF: &[u8] = &[
+            b'b', b'c', b'm', b'a', b'p', 1, 0, 0, 0, 19, 0x09, 0, 0, 0, 9, b'S', b'e', b'l', b'f',
+        ];
+        assert!(CMap::parse(SELF_REF, |_| Some(SELF_REF)).is_none());
+    }
+
     #[test]
     fn metadata_parsing() {
         let data = br#"

vendor/hayro-interpret/src/context.rs

@@ -169,6 +169,30 @@ pub struct Context<'a> {
     pub(crate) xref: &'a XRef,
     pub(crate) ocg_state: OcgState,
     nesting_depth: u32,
+    /// PdfCraft patch: estimated bytes held by saved states, and for each `q` not saved because
+    /// a limit was reached, the clip count to return to at its `Q`.
+    saved_bytes: usize,
+    skipped_saves: Vec<usize>,
+}
+
+/// PdfCraft patch: most graphics states saved at once. Every `q` copies the current state, dash
+/// array and clip list included, so without a limit short operators multiply one large state.
+const MAX_SAVED_STATES: usize = 4096;
+/// PdfCraft patch: most bytes all saved states may hold together.
+const MAX_SAVED_STATE_BYTES: usize = 64 << 20;
+
+/// Roughly what a saved copy of `state` holds.
+fn state_bytes(state: &State<'_>) -> usize {
+    size_of::<State<'_>>()
+        .saturating_add(
+            state
+                .graphics_state
+                .stroke_props
+                .dash_array
+                .len()
+                .saturating_mul(size_of::<f32>()),
+        )
+        .saturating_add(state.clips.len().saturating_mul(size_of::<ClipType>()))
 }
 
 impl<'a> Context<'a> {
@@ -223,15 +247,33 @@ impl<'a> Context<'a> {
             interpreter_cache: cache.clone(),
             ocg_state,
             nesting_depth,
+            saved_bytes: 0,
+            skipped_saves: Vec::new(),
         }
     }
 
     pub(crate) fn save_state(&mut self) {
-        let Some(cur) = self.states.last().cloned() else {
+        let Some(cur) = self.states.last() else {
             warn!("attempted to save state without existing state");
             return;
         };
 
+        // PdfCraft patch: past either limit the `q` is recorded but nothing is copied. Its `Q`
+        // still removes the clips added since; other changes made in between are kept.
+        let bytes = state_bytes(cur);
+        if !self.skipped_saves.is_empty()
+            || self.states.len() >= MAX_SAVED_STATES
+            || self.saved_bytes.saturating_add(bytes) > MAX_SAVED_STATE_BYTES
+        {
+            if self.skipped_saves.is_empty() {
+                warn!("too many saved graphics states; further saves are not kept");
+            }
+            self.skipped_saves.push(cur.clips.len());
+            return;
+        }
+
+        let cur = cur.clone();
+        self.saved_bytes = self.saved_bytes.saturating_add(bytes);
         self.states.push(cur);
     }
 
@@ -308,6 +350,14 @@ impl<'a> Context<'a> {
     }
 
     pub(crate) fn restore_state(&mut self, device: &mut impl Device<'a>) {
+        // PdfCraft patch: a save that was not kept is the most recent one.
+        if let Some(target_clips) = self.skipped_saves.pop() {
+            while self.get().clips.len() > target_clips {
+                self.pop_clip_path(device);
+            }
+            return;
+        }
+
         let Some(target_clips) = self
             .states
             .get(self.states.len().saturating_sub(2))
@@ -324,6 +374,10 @@ impl<'a> Context<'a> {
         // The first state should never be popped.
         if self.states.len() > 1 {
             self.states.pop();
+            // PdfCraft patch: the state now on top is the copy that was charged when saved.
+            if let Some(top) = self.states.last() {
+                self.saved_bytes = self.saved_bytes.saturating_sub(state_bytes(top));
+            }
         }
 
         device.set_soft_mask(
@@ -395,7 +449,8 @@ impl<'a> Context<'a> {
     }
 
     pub(crate) fn num_states(&self) -> usize {
-        self.states.len()
+        // PdfCraft patch: saves not kept still count, so callers restore back past them.
+        self.states.len().saturating_add(self.skipped_saves.len())
     }
 
     pub(crate) fn nesting_depth(&self) -> u32 {
@@ -517,3 +572,97 @@ pub(crate) fn path_as_rect(path: &BezPath) -> Option<Rect> {
 
     Some(path.fast_bounding_box())
 }
+
+/// PdfCraft patch tests (ported from craft-libs): the graphics-state stack is bounded by count
+/// and by bytes, and every `q` still pairs with its `Q`.
+#[cfg(test)]
+mod saved_state_tests {
+    use super::*;
+    use crate::device::DummyDevice;
+    use hayro_syntax::Pdf;
+
+    fn pdf() -> Pdf {
+        let objects = [
+            "<< /Type /Catalog /Pages 2 0 R >>",
+            "<< /Type /Pages /Count 0 /Kids [] >>",
+        ];
+        let mut out = b"%PDF-1.7\n".to_vec();
+        let mut offsets = Vec::new();
+        for (i, body) in objects.iter().enumerate() {
+            offsets.push(out.len());
+            out.extend_from_slice(format!("{} 0 obj\n{body}\nendobj\n", i + 1).as_bytes());
+        }
+        let xref = out.len();
+        out.extend_from_slice(b"xref\n0 3\n0000000000 65535 f \n");
+        for offset in offsets {
+            out.extend_from_slice(format!("{offset:010} 00000 n \n").as_bytes());
+        }
+        out.extend_from_slice(
+            format!("trailer\n<< /Size 3 /Root 1 0 R >>\nstartxref\n{xref}\n%%EOF\n").as_bytes(),
+        );
+        Pdf::new(out).expect("synthetic pdf")
+    }
+
+    fn triangle() -> BezPath {
+        let mut path = BezPath::new();
+        path.move_to((10.0, 10.0));
+        path.line_to((90.0, 10.0));
+        path.line_to((50.0, 90.0));
+        path.close_path();
+        path
+    }
+
+    /// Saves `saves` more times, checks a clip added after the last save goes at its `Q`, and unwinds.
+    fn save_and_unwind(ctx: &mut Context<'_>, saves: usize) {
+        let before = ctx.num_states();
+        for _ in 0..saves {
+            ctx.save_state();
+        }
+        assert!(ctx.states.len() <= MAX_SAVED_STATES);
+        assert!(ctx.saved_bytes <= MAX_SAVED_STATE_BYTES);
+        assert_eq!(ctx.num_states(), before + saves, "every q still counts");
+
+        let mut device = DummyDevice;
+        let clips = ctx.get().clips.len();
+        ctx.push_clip_path(triangle(), FillRule::NonZero, &mut device);
+        assert_eq!(ctx.get().clips.len(), clips + 1);
+        ctx.restore_state(&mut device);
+        assert_eq!(ctx.get().clips.len(), clips);
+
+        while ctx.num_states() > 1 {
+            ctx.restore_state(&mut device);
+        }
+        assert_eq!(ctx.states.len(), 1);
+        assert!(ctx.skipped_saves.is_empty());
+        assert_eq!(ctx.saved_bytes, 0);
+    }
+
+    #[test]
+    fn repeated_saves_stay_bounded_and_balanced() {
+        let pdf = pdf();
+        let cache = InterpreterCache::new();
+        let bbox = Rect::new(0.0, 0.0, 100.0, 100.0);
+        let settings = InterpreterSettings::default();
+        let mut ctx = Context::new(Affine::IDENTITY, bbox, &cache, pdf.xref(), settings);
+
+        // A small state reaches the count limit.
+        save_and_unwind(&mut ctx, 20_000);
+
+        // A large one (a 64 Ki-entry dash array, 256 KiB a copy) reaches the byte limit first.
+        ctx.get_mut().graphics_state.stroke_props.dash_array = (0..1 << 16).map(|_| 2.0).collect();
+        ctx.save_state();
+        assert_eq!(ctx.states.len(), 2);
+        while ctx.skipped_saves.is_empty() {
+            ctx.save_state();
+        }
+        assert!(ctx.states.len() < MAX_SAVED_STATES);
+        save_and_unwind(&mut ctx, 1000);
+
+        // Once unwound, a kept save restores what changed after it.
+        let mut device = DummyDevice;
+        ctx.save_state();
+        ctx.pre_concat_affine(Affine::scale(2.0));
+        ctx.restore_state(&mut device);
+        assert_eq!(ctx.get().ctm, Affine::IDENTITY);
+    }
+}

vendor/hayro-interpret/src/font/cid.rs

@@ -618,6 +618,12 @@ fn read_widths2(arr: &Array<'_>) -> Option<HashMap<u32, [f32; 3]>> {
             let mut iter = range.iter::<f32>();
 
             while let Some(w) = iter.next() {
+                // PdfCraft patch: as in `read_widths`, checked before the run's metrics are
+                // read. Without it, a run starting at 4294967295 overflowed `first` and
+                // discarded the whole /W2 table.
+                if first > MAX_CID {
+                    break;
+                }
                 let v1 = iter.next()?;
                 let v2 = iter.next()?;
                 map.insert(first, [w, v1, v2]);
@@ -669,3 +675,24 @@ fn read_encoding(object: &Object<'_>, cmap_resolver: &CMapResolverFn) -> Option<
         _ => None,
     }
 }
+
+/// PdfCraft patch test (ported from craft-libs): a /W2 run whose first CID is past `MAX_CID`
+/// is skipped and the rest of the table kept.
+#[cfg(test)]
+mod tests {
+    use super::read_widths2;
+    use hayro_syntax::object::{Array, FromBytes};
+
+    #[test]
+    fn w2_run_past_the_largest_cid_keeps_the_rest_of_the_table() {
+        let arr =
+            Array::from_bytes(b"[4294967295 [1000 500] 10 [900 450 860] 20 30 800 400 870]")
+                .unwrap();
+        let map = read_widths2(&arr).unwrap();
+
+        assert_eq!(map.get(&4_294_967_295), None);
+        assert_eq!(map.get(&10), Some(&[900.0, 450.0, 860.0]));
+        assert_eq!(map.get(&25), Some(&[800.0, 400.0, 870.0]));
+        assert_eq!(map.len(), 12);
+    }
+}

vendor/hayro-jbig2/src/segment.rs

@@ -8,6 +8,12 @@ use alloc::vec::Vec;
 use crate::error::{ParseError, Result, SegmentError, bail, err};
 use crate::reader::Reader;
 
+/// PdfCraft patch: the most segments one segment may refer to. A long-form header can claim a
+/// 29-bit count, which `Vec::with_capacity` below would turn into a multi-gigabyte reservation
+/// before a single referred-to number is read; the allocation failure aborts the process. Real
+/// files refer to a handful of segments.
+pub(crate) const MAX_REFERRED_TO_SEGMENTS: u32 = u16::MAX as u32;
+
 /// "The segment type is a number between 0 and 63, inclusive. Not all values
 /// are allowed." (7.3)
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
@@ -183,6 +189,12 @@ pub(crate) fn parse_segment_header(reader: &mut Reader<'_>) -> Result<SegmentHea
         u32::from_be_bytes([count_and_retention & 0x1F, rest[0], rest[1], rest[2]])
     };
 
+    // PdfCraft patch: refuse an oversized count before reserving memory for it and before
+    // skipping the retention-flag bytes it claims.
+    if referred_to_count > MAX_REFERRED_TO_SEGMENTS {
+        bail!(SegmentError::InvalidReferredCount);
+    }
+
     // Skip retention flag bytes in long form.
     // "The first one-byte field following the initial four-byte field is formatted
     // as follows: Bit 0: Retain bit for this segment. Bit 1-7: Retain bits for
@@ -416,4 +428,34 @@ mod tests {
 
         assert_eq!(header.data_length, Some(32));
     }
+
+    /// PdfCraft patch test (ported from craft-libs): a long-form referred-to count over
+    /// `MAX_REFERRED_TO_SEGMENTS` is refused before its retention bytes are read.
+    #[test]
+    fn test_referred_to_count_is_capped() {
+        use crate::error::DecodeError;
+
+        let header = |count: u32| {
+            let mut data = vec![
+                0x00, 0x01, 0x00, 0x00, // Segment number = 65536
+                0x86, // Flags: type 6, page assoc 1 byte, deferred non-retain
+            ];
+            // Long form: bits 29-31 are 7, bits 0-28 the count.
+            data.extend_from_slice(&(0xE000_0000 | count).to_be_bytes());
+            parse_segment_header(&mut Reader::new(&data)).unwrap_err()
+        };
+
+        // The 29-bit maximum, 536,870,911, would reserve about 2 GiB.
+        for count in [0x1FFF_FFFF, MAX_REFERRED_TO_SEGMENTS + 1] {
+            assert!(matches!(
+                header(count),
+                DecodeError::Segment(SegmentError::InvalidReferredCount)
+            ));
+        }
+        // At the cap the header is read on and ends at its missing retention bytes.
+        assert!(matches!(
+            header(MAX_REFERRED_TO_SEGMENTS),
+            DecodeError::Parse(ParseError::UnexpectedEof)
+        ));
+    }
 }

vendor/hayro-syntax/src/xref.rs

@@ -965,17 +965,13 @@ fn populate_from_xref_stream<'a>(
 
     let [f1_len, f2_len, f3_len] = stream.dict().get::<[u8; 3]>(W)?;
 
-    if f2_len > size_of::<u64>() as u8 {
+    // PdfCraft patch: the type field is read at its full width, so it is bounded like the offset.
-        error!("xref offset length is larger than the allowed limit");
+    if f1_len > size_of::<u64>() as u8 || f2_len > size_of::<u64>() as u8 {
+        error!("xref field length is larger than the allowed limit");
 
         return None;
     }
 
-    // Do such files exist?
-    if f1_len != 1 {
-        warn!("first field in xref stream was longer than 1");
-    }
-
     let xref_data = stream.decoded().ok()?;
     // PdfCraft patch: (#2366) this section's entries are read first, for `prev_offset`.
     let mut own = XrefMap::default();
@@ -1042,15 +1038,38 @@ fn xref_stream_subsection<'a>(
     f3_len: u8,
     insert_map: &mut XrefMap,
 ) -> Option<()> {
+    // PdfCraft patch: every entry must be backed by stream bytes. With all widths zero an entry
+    // reads nothing, so a tiny stream could declare billions of entries; and a count past the
+    // decoded data is refused before any of it is inserted.
+    let width = f1_len as usize + f2_len as usize + f3_len as usize;
+    if width == 0 {
+        warn!("xref stream entries have no width");
+
+        return None;
+    }
+    let available = xref_reader.len().saturating_sub(xref_reader.offset());
+    if (num_elements as usize)
+        .checked_mul(width)
+        .is_none_or(|need| need > available)
+    {
+        warn!("xref stream subsection declares more entries than its data holds");
+
+        return None;
+    }
+
     for i in 0..num_elements {
+        // A zero-width type field defaults to 1 (ISO 32000-2 §7.5.8.2, `/W`).
         let f_type = if f1_len == 0 {
             1
         } else {
-            // We assume a length of 1.
+            // PdfCraft patch: the type field is read big-endian at its full `/W[0]` width.
-            xref_reader.read_bytes(1)?[0]
+            let bytes = xref_reader.read_bytes(f1_len as usize)?;
+            bytes.iter().fold(0u64, |acc, b| (acc << 8) | u64::from(*b))
         };
 
-        let obj_number = start + i;
+        // PdfCraft patch: object numbers are stored as `i32`; a subsection running past
+        // `i32::MAX` is refused instead of overflowing or turning negative.
+        let obj_number = i32::try_from(start.checked_add(i)?).ok()?;
 
         match f_type {
             // We don't care about free objects.
@@ -1073,7 +1092,7 @@ fn xref_stream_subsection<'a>(
                 };
 
                 insert_map.insert(
-                    ObjectIdentifier::new(obj_number as i32, gen_number as i32),
+                    ObjectIdentifier::new(obj_number, gen_number as i32),
                     EntryType::Normal(offset as usize),
                 );
             }
@@ -1091,7 +1110,7 @@ fn xref_stream_subsection<'a>(
                 };
 
                 insert_map.insert(
-                    ObjectIdentifier::new(obj_number as i32, gen_number),
+                    ObjectIdentifier::new(obj_number, gen_number),
                     EntryType::ObjStream(obj_stream_number, index),
                 );
             }
@@ -1248,6 +1267,72 @@ fn parse_metadata(info_dict: &Dict<'_>) -> Metadata {
 mod tests {
     use super::*;
 
+    // PdfCraft patch tests (ported from craft-libs): xref stream entries are bounded by the
+    // decoded data, and the type field is read at its full width.
+    #[test]
+    fn xref_stream_entries_need_backing_bytes() {
+        let mut map = XrefMap::default();
+        // All widths zero: an entry reads nothing, so none may be declared.
+        let mut empty = Reader::new(&[]);
+        assert!(xref_stream_subsection(&mut empty, 0, u32::MAX, 0, 0, 0, &mut map).is_none());
+        assert!(map.is_empty());
+        // Three entries of width 3 need nine bytes; eight are refused before any insert.
+        let data = [1, 0, 9, 1, 0, 9, 1, 0];
+        let mut short = Reader::new(&data);
+        assert!(xref_stream_subsection(&mut short, 0, 3, 1, 1, 1, &mut map).is_none());
+        assert!(map.is_empty());
+        let mut whole = Reader::new(&data[..6]);
+        assert!(xref_stream_subsection(&mut whole, 0, 2, 1, 1, 1, &mut map).is_some());
+        assert_eq!(map.len(), 2);
+        // A subsection running past object number i32::MAX is refused.
+        for start in [i32::MAX as u32, u32::MAX] {
+            let mut past_end = XrefMap::default();
+            let mut wrap = Reader::new(&data[..6]);
+            assert!(xref_stream_subsection(&mut wrap, start, 2, 1, 1, 1, &mut past_end).is_none());
+            assert!(past_end.keys().all(|id| id.obj_number >= 0));
+        }
+    }
+
+    #[test]
+    fn xref_stream_type_field_uses_the_full_w0_width() {
+        // /W [2 1 1]: two-byte type 0x0001 (normal), offset 7, gen 0.
+        let mut map = XrefMap::default();
+        let mut reader = Reader::new(&[0u8, 1, 7, 0]);
+        assert!(xref_stream_subsection(&mut reader, 5, 1, 2, 1, 1, &mut map).is_some());
+        assert!(matches!(
+            map.get(&ObjectIdentifier::new(5, 0)),
+            Some(EntryType::Normal(7))
+        ));
+    }
+
+    #[test]
+    fn xref_stream_loader_rejects_oversized_w0() {
+        // One normal entry (offset 7, gen 0) whose type field is `type_len` bytes wide, so only
+        // the width of /W[0] differs between the two streams.
+        let load = |type_len: usize| {
+            let mut entry = vec![0u8; type_len];
+            entry[type_len - 1] = 1;
+            entry.extend_from_slice(&[7, 0]);
+            let mut data = format!(
+                "1 0 obj\n<< /Type /XRef /Size 6 /Index [5 1] /W [{type_len} 1 1] /Length {} >>\nstream\n",
+                entry.len()
+            )
+            .into_bytes();
+            data.extend_from_slice(&entry);
+            data.extend_from_slice(b"\nendstream\nendobj\n");
+            let mut map = XrefMap::default();
+            let got = populate_from_xref_stream(
+                &data,
+                &mut Reader::new(&data),
+                &mut map,
+                &mut BTreeSet::new(),
+            );
+            got.is_some()
+        };
+        assert!(load(8));
+        assert!(!load(9));
+    }
+
     // Contributor-original, MIT OR Apache-2.0. Matches the bounded, uncompressed
     // synthetic regression used to reproduce the cold-publication race. A second
     // stream optionally checks that racing registration never aliases two ids.