fastwarc

This project publishes no release notes. Between v1.0.3 and v1.0.4 there were 28 commits, 27 of them substantive:

• Move payload_buf initialisation out of loop
• Do not shrink GzipReader decompressed buffer
• Implement consume_n using seek
• Explicitly inline stream_position
• Make max header length configurable and increase default limit
• Fix typo
• Use memchr for : scan in headers
• Slightly optimise header parsing
• Add quirks mode parsing
• Make parsed headers copy-on-write
• Move headers to separate submodule
• Move helper function
• Fix mypy error
• Mark legacy shims as deprecated
• Mark alias names as deprecated
• Fix import error on newer Python versions
• Fix header parsing when EOH marker is split across buffer boundaries
• Explicitly inline more methods
• Avoid copying of raw header buffer
• Use existing read_line method instead of take()

…and 7 more.

This project publishes no release notes. Between v1.0.2 and v1.0.3 there were 17 commits, 16 of them substantive:

• Add performance profiling binary
• Add short circuits for stream_position() and seek() in RawReaderAdapter and GzipReader for much improved performance
• Simplify IntoWarcReader trait bound for BufReader<T> and add conversion for fs::File
• Increase buffer sizes
• Use cached stream position for member boundaries
• Print byte throughput in profile binary
• Move reader adapter infrastructure to separate module
• Add TrackingBufReader to replace BufReader at the lowest level
• Serve small relative forward seeks from fill_buf()
• Update IntoWarcReader impls for BufReader, File, and Cursor
• Add constructor to RawWriterAdapter and std::fs::File conversion trait impl
• Use TrackingBufReader in decompressing readers (except Brotli) and optimise Gzip buffer growing
• Use larger buffer sizes in Python wrapper
• Minor performance improvement in read_line
• Forward inner_seek to native inner_seek
• Fix incorrect test assumptions

This project publishes no release notes. Between v1.0.1 and v1.0.2 there were 46 commits, 45 of them substantive:

• Port test_record_writer test
• Add clone_warc_file test (ported test_record_writer test)
• Port create_new_warc_record
• Specify explicit packages
• Use newer Python version (maybe that fixes the twine upload issue?)
• Rename cargo-coverage.lcov to cargo-coverage.info
• Upgrade packaging on twine upload, fix wheel matching in build-documentation
• Install Apache Beam in Python 3.14
• Install apache_beam in docs requirements.txt as a workaround for 314t requiring us to build grpcio
• Publish crates from CI
• Uncomment crates publishing guard
• Update READMEs, add RustDoc link to Sphinx sidebar
• Unhide ArchiveIteratorImpl
• Replace explicit &[u8] arguments with AsRef<[u8]>
• Ignore additional file types
• Add coverage building helper script
• Add Python bindings tests
• Fix deadlock in pickle
• Implement PartialEq and eq for WarcRecords
• Dispatch close to runtime self

…and 25 more.