M6: field detection reads operators one at a time; the PDF/A check reads XMP as preflight does

From the Resources review of 66415d737:
- page_shapes collected every parsed operator, about 80 bytes for each
  decoded byte of dense content, so a page within its 256 MiB allowance
  could take some 20 GB. It now reads them one at a time
  (pdfcraft_content::ops) and drops the pieces once joined.
- declares_pdfa decoded through a DecodeBudget, which counts every filter
  stage against the 8 MiB cap, so XMP stored through a chain of filters
  stopped declaring PDF/A while preflight's Standards panel, which caps
  each stage, still read it. It reads the metadata as preflight does.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AmsdeUCdTaD72jkDw2pJKY

crates/forms/src/appearances.rs

@@ -1,6 +1,6 @@
 use super::read::{acroform, fields};
 use super::values::merged_appearance;
-use super::{DecodeBudget, Dict, Document, FieldKind, ObjRef, Object, Widget, appearance, name_bytes};
+use super::{Dict, Document, FieldKind, ObjRef, Object, Widget, appearance, name_bytes};
 
 /// Whether the form asks viewers to draw its fields' appearances themselves (`/NeedAppearances`).
 pub fn need_appearances(doc: &Document) -> bool {
@@ -212,18 +212,21 @@ impl FieldAppearanceCache {
     }
 }
 
-/// The most of the XMP metadata stream decoded to find a PDF/A declaration, as preflight's
+/// The most each filter stage of the XMP metadata stream may decode for its PDF/A declaration to
-/// Standards panel reads it: past it, metadata declares nothing there either.
+/// be read, as preflight's Standards panel reads it: past it, metadata declares nothing there
+/// either.
 pub(crate) const MAX_XMP_BYTES: usize = 8 << 20;
 
-/// Whether the document's XMP metadata declares PDF/A conformance (`pdfaid:part`). Metadata that
+/// Whether the document's XMP metadata declares PDF/A conformance (`pdfaid:part`). Metadata a
-/// decodes past [`MAX_XMP_BYTES`] doesn't.
+/// filter stage of which decodes past [`MAX_XMP_BYTES`] doesn't.
 pub fn declares_pdfa(doc: &Document) -> bool {
     let Some(root) = doc.root() else { return false };
     let Some(m) = doc.get(root).as_dict().and_then(|c| c.get(b"Metadata").cloned()) else { return false };
     let m = doc.resolve(&m);
     match &*m {
-        Object::Stream(s) => DecodeBudget::new(MAX_XMP_BYTES, 0).decode(s).is_ok_and(|b| b.windows(11).any(|w| w == b"pdfaid:part")),
+        // Not a `DecodeBudget`, which counts every stage against one allowance: preflight bounds
+        // each stage, and the two must agree on metadata stored through a chain of filters.
+        Object::Stream(s) => s.decoded_within(MAX_XMP_BYTES).is_ok_and(|b| b.windows(11).any(|w| w == b"pdfaid:part")),
         _ => false,
     }
 }

crates/forms/src/detect.rs

@@ -156,7 +156,11 @@ pub(crate) fn shapes_within(doc: &Document, page: usize, budget: &mut DecodeBudg
     let mut subpaths: Vec<DetectionPath> = Vec::new();
     let mut path = DetectionPath::default();
     let mut cur: Option<[f64; 2]> = None;
-    for op in Pieces::join(&pieces).parse().ops {
+    let joined = Pieces::join(&pieces);
+    drop(pieces);
+    // An operator at a time: dense content (`n n n …`) parsed whole holds about 80 bytes for each
+    // byte it decoded.
+    for op in pdfcraft_content::ops(joined.data()) {
         match op.op.as_slice() {
             b"q" => stack.push(ctm),
             b"Q" => ctm = stack.pop().unwrap_or(ctm),

crates/forms/src/tests.rs

@@ -3645,8 +3645,9 @@ fn page_shapes_decode_a_repeated_content_stream_within_the_budget() {
     assert!(budget.exhausted());
 }
 
-/// The PDF/A declaration is read from at most [`crate::appearances::MAX_XMP_BYTES`] of metadata, as
+/// The PDF/A declaration is read from metadata each filter stage of which decodes within
-/// preflight reads it: a field drawn with characters its font lacks decodes no more than that.
+/// [`crate::appearances::MAX_XMP_BYTES`], as preflight reads it: a field drawn with characters its
+/// font lacks decodes no more than that a stage.
 #[test]
 fn a_pdfa_declaration_is_read_from_metadata_within_its_cap() {
     let declared = |xmp: String| declares_pdfa(&document(&["<< /Type /Catalog /Metadata 2 0 R >>".into(), raw_stream(&xmp)]));
@@ -3654,6 +3655,16 @@ fn a_pdfa_declaration_is_read_from_metadata_within_its_cap() {
     assert!(declared(pdfa(1 << 20)));
     assert!(!declared(pdfa(crate::appearances::MAX_XMP_BYTES)), "metadata past the cap declares nothing");
     assert!(!declared("<x:xmpmeta><dc:title>pdfaid</dc:title></x:xmpmeta>".into()), "metadata without pdfaid:part declares nothing");
+    // 3.5 MiB stored as hex, then Flate: its stages decode 7 and 3.5 MiB, each within the cap.
+    let xmp = pdfa(7 << 19);
+    let hex: Vec<u8> = xmp.bytes().flat_map(|b| [b"0123456789ABCDEF"[usize::from(b >> 4)], b"0123456789ABCDEF"[usize::from(b & 15)]]).collect();
+    let mut stream = pdfcraft_cos::Stream::flate(pdfcraft_cos::Dict::new(), &hex);
+    stream.dict.set(b"Filter".to_vec(), Object::Array(vec![Object::name("FlateDecode"), Object::name("ASCIIHexDecode")]));
+    let mut doc = Document::new_empty();
+    let metadata = doc.add(Object::Stream(stream));
+    let catalog = doc.add(Object::Dict([(b"Metadata".to_vec(), Object::Ref(metadata))].into_iter().collect()));
+    doc.trailer_mut().set(b"Root".to_vec(), Object::Ref(catalog));
+    assert!(declares_pdfa(&doc), "metadata through a chain of filters declares PDF/A, as preflight reads it");
 }
 
 /// A widget listed many times, by one field's `/Kids` or by a `/Kids` array several fields share,