diff --git a/crates/optimize/src/links.rs b/crates/optimize/src/links.rs index 744881df7..082d56e7d 100644 --- a/crates/optimize/src/links.rs +++ b/crates/optimize/src/links.rs @@ -150,11 +150,14 @@ fn referenced_names(doc: &Document) -> HashSet> { while let Some(o) = stack.pop() { match o { Object::Dict(d) => { - if let Some(n) = d.get(b"Dest").and_then(dest_name) { + // Resolve first: /Dest and a GoTo's /D may be an indirect reference to the name + // string, and an unresolved ref has no name, so the destination it uses would + // look unreferenced and be removed. + if let Some(n) = d.get(b"Dest").and_then(|x| dest_name(doc.resolve(x).as_ref())) { out.insert(n); } if matches!(d.name(b"S"), Some(b"GoTo" | b"GoToR" | b"GoToE")) - && let Some(n) = d.get(b"D").and_then(dest_name) + && let Some(n) = d.get(b"D").and_then(|x| dest_name(doc.resolve(x).as_ref())) { out.insert(n); } diff --git a/crates/optimize/tests/optimize.rs b/crates/optimize/tests/optimize.rs index 6cf53d4fc..6d9a9bb44 100644 --- a/crates/optimize/tests/optimize.rs +++ b/crates/optimize/tests/optimize.rs @@ -1097,3 +1097,45 @@ fn a_hostile_soft_mask_leaves_its_image_unchanged() { assert_eq!(stream(&doc, photo), before); assert_eq!(stream(&doc, mask), hostile); } + +#[test] +fn indirectly_referenced_destination_is_kept() { + // A link's /Dest can be an indirect reference to the name string, not the string itself. Reduce + // File Size read the name only from a direct string, so it deleted a destination the link still + // used, silently breaking the link. + let mut doc = Document::new_empty(); + let p = page(&mut doc, &[], "BT ET"); + let s = |t: &str| Object::String(pdfcraft_cos::PdfString::literal(t.as_bytes().to_vec())); + // A link whose /Dest is `ref -> "target"`, and an unused destination for contrast. + let name_ref = Object::Ref(doc.add(s("target"))); + let mut link = Dict::new(); + link.set(b"Type".to_vec(), Object::name("Annot")); + link.set(b"Subtype".to_vec(), Object::name("Link")); + link.set(b"Rect".to_vec(), Object::Array(vec![0.into(), 0.into(), 10.into(), 10.into()])); + link.set(b"Dest".to_vec(), name_ref); + let link = Object::Ref(doc.add(Object::Dict(link))); + doc.update_dict(p, |d| d.set(b"Annots".to_vec(), Object::Array(vec![link]))).unwrap(); + let fit = Object::Array(vec![Object::Ref(p), Object::name("Fit")]); + let mut tree = Dict::new(); + tree.set(b"Names".to_vec(), Object::Array(vec![s("target"), fit.clone(), s("unused"), fit])); + let tree = doc.add(Object::Dict(tree)); + let mut names = Dict::new(); + names.set(b"Dests".to_vec(), Object::Ref(tree)); + let root = doc.root().unwrap(); + doc.update_dict(root, |c| c.set(b"Names".to_vec(), Object::Dict(names))).unwrap(); + let report = optimize(&mut doc, &Settings::default()).unwrap(); + // Only "unused" is removed; the indirectly-referenced "target" is kept. + assert_eq!(report.unreferenced_dests, 1); + let left = named_dest_keys(&doc); + assert!(left.contains(&b"target".to_vec()), "indirectly referenced destination was kept: {left:?}"); + assert!(!left.contains(&b"unused".to_vec())); +} + +/// The keys in the catalog's /Names /Dests name tree (one level, as the optimizer writes it). +fn named_dest_keys(doc: &Document) -> Vec> { + let cat = doc.get(doc.root().unwrap()).as_dict().cloned().unwrap(); + let names = doc.resolve(cat.get(b"Names").unwrap()).as_dict().cloned().unwrap(); + let tree = doc.resolve(names.get(b"Dests").unwrap()).as_dict().cloned().unwrap(); + let arr = doc.resolve(tree.get(b"Names").unwrap()).as_array().cloned().unwrap(); + arr.chunks(2).filter_map(|c| doc.resolve(&c[0]).as_string().map(|s| s.bytes.clone())).collect() +}