mirror of
https://github.com/storytold/craft-libs.git
synced 2026-11-14 06:22:24 +00:00
craft-pdf: the PDF object layer from PdfCraft (`cos`), on filters and crypt * hayro-jbig2: bound a stream's total decode work and what a page retains; no i32 overflow Port of pdfcraft 85c10051 (patches 5-9), 82439525 (patch 10) and 10107d94 (patch 12). * hayro-jbig2: PdfCraft's JBIG2 follow-ups (refinement fixes, file header, work accounting, dictionary pixel cap) Ports pdfcraft 72e763ad, b7ec3669, e3bc4e85, 01d3479f, b9e068aa, 858ce960, 07f44038, e38b9d18, 8717a085, 58c30af3 and the hayro-jbig2 part of e2464e14 (with the hayro-jbig2 part of 0ed33990 they build on). * Add craft-stream-codecs: the codec modules of PdfCraft's stream filters LZW, ASCII85, ASCIIHex, RunLength, the TIFF and PNG predictors, bounded CCITT fax decoding (through hayro-ccitt) with Group 4 encoding (through fax), and the lossless JBIG2 encoder with its MQ coder, from pdfcraft-filters at PdfCraft revision 80e26e674d1aa33078e83c8139875047f3289eb6. The crate sits below craft-pdf and craft-tiff: it knows no container. The PDF filter chain, Flate and the bounded JBIG2 decoding entry are not part of it. * craft-pdf: the PDF filter chain from PdfCraft (`filters`) Adds the `filters` feature: the /Filter chain of pdfcraft-filters (PdfCraft revision 80e26e674d1aa33078e83c8139875047f3289eb6) as craft_pdf::filters, over the codecs in craft-stream-codecs. It adds Flate, strict and tolerant chain decoding, byte accounting and a shared work allowance, and encoding, with the chain's unit tests, its property tests and its fuzz target. The code is unchanged apart from the imports of the moved codecs; FilterError is craft-stream-codecs' CodecError under its old name. * craft-stream-codecs: bounded JBIG2 decoding (decode_jbig2_within) The JBIG2 decoding entry of pdfcraft-filters (PdfCraft revision 80e26e674d1aa33078e83c8139875047f3289eb6, src/jbig2/decode.rs) with its tests. It spends from the caller's work allowance through hayro-jbig2's Image::decode_within and charge_reading, which come with fix/sec-hayro-jbig2-followups (#886). * vendor/README.craft-reader.md: one #831 paragraph, the follow-ups after it * craft-pdf: filters re-exports decode_jbig2_within (and lib.rs module order) * craft-pdf: the object layer from PdfCraft (`cos`) Adds the `cos` feature: PdfCraft's pdfcraft-cos crate (revision 1d2af9907ba772b368f6b54811359805fa5428e9) as craft_pdf::cos, on crypt and filters, with its unit tests, integration tests and examples. Its test helpers come with it: pdfcraft-testkit as src/cos/testkit.rs (cfg(test)) and the qpdf runner of pdfcraft-oracle as tests/cos/oracle.rs. The code is unchanged apart from module paths and doc comments. The child test of file_identifier is named by its module path, since the integration tests are one binary. * craft-pdf: cos unit tests build for wasm The property tests in src/cos/writer.rs need proptest, a dev-dependency of the non-wasm targets only, and one test in src/cos/document/tests.rs times a load on a thread, which wasm32-unknown-unknown does not have. Both are compiled for non-wasm targets only; the other unit tests still build for wasm.
79 lines
3.2 KiB
Rust
79 lines
3.2 KiB
Rust
//! Stream data read from a file points into the file's bytes instead of being copied: the
|
|
//! document's object cache holds parsed dictionaries, not a second copy of every stream.
|
|
|
|
use std::sync::Arc;
|
|
|
|
use craft_pdf::cos::{Document, ObjRef, Object, SaveOptions, write_full, write_incremental};
|
|
|
|
const CONTENT: &[u8] = b"0 0 1 rg 10 10 30 20 re f";
|
|
|
|
fn fixture() -> Vec<u8> {
|
|
let objs = [
|
|
"<< /Type /Catalog /Pages 2 0 R >>".to_string(),
|
|
"<< /Type /Pages /Kids [3 0 R] /Count 1 /MediaBox [0 0 100 50] >>".to_string(),
|
|
"<< /Type /Page /Parent 2 0 R /Contents 4 0 R >>".to_string(),
|
|
format!("<< /Length {} >>\nstream\n{}\nendstream", CONTENT.len(), String::from_utf8_lossy(CONTENT)),
|
|
];
|
|
let mut out = b"%PDF-1.4\n".to_vec();
|
|
let mut offsets = Vec::new();
|
|
for (i, o) in objs.iter().enumerate() {
|
|
offsets.push(out.len());
|
|
out.extend_from_slice(format!("{} 0 obj\n{o}\nendobj\n", i + 1).as_bytes());
|
|
}
|
|
let xref = out.len();
|
|
out.extend_from_slice(format!("xref\n0 {}\n0000000000 65535 f \n", objs.len() + 1).as_bytes());
|
|
for o in offsets {
|
|
out.extend_from_slice(format!("{o:010} 00000 n \n").as_bytes());
|
|
}
|
|
out.extend_from_slice(format!("trailer\n<< /Size {} /Root 1 0 R >>\nstartxref\n{xref}\n%%EOF\n", objs.len() + 1).as_bytes());
|
|
out
|
|
}
|
|
|
|
const CONTENTS: ObjRef = ObjRef { num: 4, generation: 0 };
|
|
|
|
fn stream_bytes(doc: &Document, r: ObjRef) -> (Vec<u8>, *const u8) {
|
|
match &*doc.get(r) {
|
|
Object::Stream(s) => (s.raw.to_vec(), s.raw.as_ptr()),
|
|
other => panic!("{other:?}"),
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn cached_streams_are_the_files_own_bytes() {
|
|
let file = Arc::new(fixture());
|
|
let doc = Document::open(file.clone()).unwrap();
|
|
let (data, at) = stream_bytes(&doc, CONTENTS);
|
|
assert_eq!(data, CONTENT);
|
|
assert!(file.as_ptr_range().contains(&at), "not a copy");
|
|
// A full save reads and caches every object; the cache still holds no copy of stream data.
|
|
let full = write_full(&doc, &SaveOptions::default()).unwrap();
|
|
assert!(file.as_ptr_range().contains(&stream_bytes(&doc, CONTENTS).1));
|
|
// Saves are unchanged: the stream is written as it was read.
|
|
let back = Document::open(Arc::new(full)).unwrap();
|
|
assert_eq!(stream_bytes(&back, ObjRef::new(4, 0)).0, CONTENT);
|
|
}
|
|
|
|
#[test]
|
|
fn edits_within_a_document_keep_pointing_into_its_file() {
|
|
let file = Arc::new(fixture());
|
|
let mut doc = Document::open(file.clone()).unwrap();
|
|
doc.update_dict(CONTENTS, |d| d.set(b"Marked".to_vec(), Object::Bool(true))).unwrap();
|
|
assert!(file.as_ptr_range().contains(&stream_bytes(&doc, CONTENTS).1));
|
|
let updated = write_incremental(&doc, &SaveOptions::default()).unwrap();
|
|
let back = Document::open(Arc::new(updated)).unwrap();
|
|
assert_eq!(stream_bytes(&back, CONTENTS).0, CONTENT);
|
|
}
|
|
|
|
#[test]
|
|
fn streams_moved_to_another_document_get_their_own_bytes() {
|
|
let file = Arc::new(fixture());
|
|
let source = Document::open(file.clone()).unwrap();
|
|
let moved = (*source.get(CONTENTS)).clone();
|
|
let mut target = Document::new_empty();
|
|
let r = target.add(moved);
|
|
drop(source);
|
|
// The target keeps just the stream's bytes, not the whole source file.
|
|
assert_eq!(Arc::strong_count(&file), 1);
|
|
assert_eq!(stream_bytes(&target, r).0, CONTENT);
|
|
}
|