Releases: onecodex/needletail
Releases · onecodex/needletail
v0.7.1
v0.7.0
v0.6.3
- sdist (source) wheels are now pushed to PyPI
- Maturin Docker container is now used to generate manylinux-compliant wheels
v0.6.2
v0.6.2 contains backwards-compatible enhancements to the Python library (thanks @apcamargo!):
- Add new dunder methods (
__hash__,__eq__,__len__,__str__, and__repr__) toneedletail.Record needletail.Recordinstances can now be created directlyrecord.normalize()now takesiupacargument (default=False)parse_fastx_filecan now take apathlib.Path
v0.6.1
- Add
positionmethod to return current line number and byte offset of parser (thanks @olliecheng)
v0.6.0
v0.3.0
Added
- Improved error reporting (i.e., a parse failure now gives the record it failed on).
- Significant code cleanup and additional linting (
cargo clippy). - Significant additional test coverage, including via fuzzing.
- Significant improvements to library documentation.
Changed
- The
.kmersmethod has been simplified and a new.canonical_kmersmethod has been introduced with much of the original's functionality. - Added
parse_sequence_reader, which replacesfastx_streamandfastx_bytes. fastx_cliupdated and renamed toparse_sequence_path.SeqRecordis nowSequenceRecordand many of its methods are now in theSequencetrait (e.g., working on byte slices).- Automatic decompression now takes
Readinstead ofRead + Seekso we can handle e.g. gzip files piped in throughstdin. - See this link for additional details on updating code to
v0.3.0.
Removed
- Single-file zip handling (zip requires
Seek) 😞