v1.1.3

What's Changed

• WIP: switch to ubuntu latest for CI by @ctb in #101
• MRG: fix codecov upload by @ctb in #102
• MRG: rename test methods -> pytest; remove issue template by @ctb in #103
• MRG: switch screed tests to run monthly by @ctb in #105
• Docs: system_packages is deprecated in RTD by @luizirber in #107
• Bump actions/checkout from 3 to 4 by @dependabot in #104
• Close sequencefile if an error happens in open_reader by @luizirber in #106
• Bump actions/setup-python from 4 to 5 by @dependabot in #108

Full Changelog: v1.1.2...v1.1.3

v1.1.2

What's Changed

• Remove pkg_resources usage by @luizirber in #100

Full Changelog: v1.1.1...v1.1.2

v1.1.1

What's Changed

• [MRG] update release checklist after v1.1 by @ctb in #95
• [MRG] add patch to semver versioning in release notes by @ctb in #96
• [MRG] bump pip/tox cache version numbers by @ctb in #97
• [MRG] remove bz2file in favor of bz2 by @ctb in #99

Full Changelog: v1.1...v1.1.1

v1.1

What's Changed

• Add badges for Debian packages by @mr-c in #89
• [MRG] add dependabot by @ctb in #91
• Bump actions/checkout from 2 to 3 by @dependabot in #92
• Bump actions/setup-python from 2 to 4 by @dependabot in #93
• Bump actions/cache from 2 to 3 by @dependabot in #94
• [MRG] fix ResourceWarning from unclosed file by @ctb in #90

New Contributors

• @dependabot made their first contribution in #92

Full Changelog: v1.0.5...v1.1

Screed v1.0.5

This release REMOVES support for Python 3.6 and older (including Python 2).

• minor updates to Python code (#84)

Many substantial updates to screed's development infrastructure -

• use pyproject.toml and setup.cfg instead of setup.py (#83)
• switch to GitHub Actions instead of Travis (#86)
• use setuptools_scm instead of versioneer (#85)
• use tox (#87)

Screed 1.0

We are pleased to announce the release of screed 1.0. Screed is a biological sequence parsing and storage/retrieval library for DNA and protein sequences. It's designed to be lightweight and easy to use
from Python.

This version is the first with API compatibility guarantees, following the semantic versioning guidelines.
Most changes are internal or API clarifications, but there is a new shell command for screed functions and an unified function for writing FAST{A,Q} records.

Documentation is available at http://screed.readthedocs.org/en/v1.0

New items of note:

• New shell commands for common screed operations:
  • db for database creation (screed db <filename>)
  • dumping FAST{A,Q} records from a db (screed dump_fasta <db> <output> and screed dump_fastq <db> <output>). #55 @luizirber
• Remove \*_Writer classes and unify record writing in the write_fastx function. #53 @standage
• We now use pytest as a test runner, codecov for code coverage, and a simplified changelog format. #50 #49 #59 @luizirber @standage

Other bugs fixed/issues closed:

• Fix reverse complement problems for Python 2.7. #47 @ctb
• Fix operator comparison. #48 @luizirber
• Update tests & constrain behavior for screed Records. #54 @ctb
• Allow sqlite3 import to fail. #56 @ctb
• Cleanup user docs and code. #62 #57 @standage
• Simplify use of 'open' internally. #65 @ctb

Known Issues

These are all pre-existing

• Screed does not support gzip file streaming. This is an issue with Python 2.x and will likely not be fixed in future releases. This is being tracked in dib-lab/khmer#700

• Screed is overly tolerant of spaces in fast{a,q} which is against spec. This is being tracked in dib-lab/khmer#108

Contributors

@luizirber *@standage @ctb *@betatim

* Indicates new contributors

Release v0.9 - Python 3 support!

Release v0.9

We are pleased to announce the release of Screed v0.9. Screed is a database
engine capable of storing and retrieving short-read sequence data and is
designed to be fast and adaptable to different sequence file formats.

This version of Screed features Python 3 syntax with compatibility with Python 2. Additional changes have broken backwards compatibility in several small ways in preparation for our 1.0 release and adoption of strict semantic versioning from there on out.

It is also the first release since our move to the University of Davis, California and also under our new name, the Lab for Data Intensive Biology.

Documentation is available at http://screed.readthedocs.org/en/v0.9/

New items of note:

• Now a primarily Python 3 codebase with Python 2 compatibility. #41 @luizirber & @mr-c
• Tests now correctly run using temporary directories and the test data is now shipped allowing the tests to be run after installation. #30 @bocajnotnef #40 @mr-c
• The private method screed/screedRecord._screed_record_dict() has been renamed to screed.screedRecord.Record(). This is not a backwards compatible change. #35 @sguermond
• screed.open() now accepts - as a synonym for STDIN and is now an (optional) context manager. It no longer defaults to parsing out a separate description from the name. The description field will br removed altogether from the next release. This is not a backwards compatible change. #36 @anotherthomas #39 #41 @luizirber #43 @ctb
• The FASTQ parser was improved and it no longer hangs in the presence of empty lines. #38 @proteasome
• Screed records now slice correctly #41 @wrightmhw @luizirber

Other bugs fixed/issues closed:

• Release notes are now a part of the documentation. #33 @bocajnotnef
• A test was made more robust to prevent hangs. #37 @anotherthomas

Known Issues

These are all pre-existing

• Screed does not support gzip file streaming. This is an issue with Python 2.x and will likely not be fixed in future releases. This is being tracked in dib-lab/khmer#700
• Screed is overly tolerant of spaces in fast{a,q} which is against spec. This is being tracked in dib-lab/khmer#108

Contributors

@luizirber @mr-c @bocajnotnef @ctb *@proteasome *@anotherthomas *@sguermond

* Indicates new contributors

Release v0.8

Release v0.8

We are pleased to announce the release of Screed v0.8. Screed is a database
engine capable of storing and retrieving short-read sequence data and is
designed to be fast and adaptable to different sequence file formats.

This version of Screed contains developer documentation for contributing to the
Screed project and a code of conduct for interacting with other contributors
and project maintainers. Documentation is available at
http://screed.readthedocs.org/en/v0.8/

New items of note:

This release successfully installs and passes its unit tests on Ubuntu 14.04 and the latest release of Mac OS X 10 "Yosemite". It also passes the khmer acceptance tests as per the eelpond testing protocol.

This release of screed has renamed the 'accuracy' attribute of read records to 'quality;' this API change will need to be adopted by all users wanting to upgrade to this version. Unlike the khmer project, Screed is not currently under semantic versioning. It will be with the 1.0 release.

• Screed now has automatic compression detection via magic bit sniffing for gzip and bzip2 compressed files (from @mr-c in dib-lab/khmer#432)
• Screed now supports streaming of uncompressed FASTA and FASTQ formatted nucleotide sequence data. bzip2 compressed FASTA and FASTQ formatted nucleotide sequence data can also be streamed but not gzip compressed FASTA and FASTQ formatted nucleotide sequence data. (from @mr-c, see dib-lab/khmer#633)
• Screed now has a Changelog, developer documentation and a code of conduct (from @ctb, @mr-c, @bocajnotnef in dib-lab/khmer#625)
• Versions are now autogenerated using git tags via Versioneer (from
  @bocajnotnef in cadceb5)
• Documentation is now autogenerated using Doxygen (from @mr-c in d8ed05b)

Notable bugs fixed/issues closed:

• A khmer script was not accepting reads on the stdin dib-lab/khmer#633 by @mr-c
• screed returning the wrong version and breaking dev installs dib-lab/khmer#803 by @mr-c

Known Issues

These are all pre-existing

• Screed records cannot be sliced requiring un-Pythonic techniques to achieve the same behavior. This will be included in a future release. This is being tracked in dib-lab/khmer#768
• Screed self-tests do not use a temporary directory which causes tests run from package-based installs to fail. This is being tracked in dib-lab/khmer#748
• Screed does not support gzip file streaming. This is an issue with Python 2.x and will likely not be fixed in future releases. This is being tracked in dib-lab/khmer#700
• Screed is overly tolerant of spaces in fast{a,q} which is against spec. This is being tracked in dib-lab/khmer#108

Contributors

@bocajnotnef @mr-c @brtaylor92 @wrightmhw @kdmurray91 @luizirber @ctb