v0.5.1

• Spliced sequence retrieval #127 (Thanks @simonvh)

v0.5.0.1

• Removed BGZF warning (#123) since the test suite contains good coverage of this feature.

v0.5.0

• coordinates with start > end can be auto reverse complemented using --auto-strand argument (#120)
• --no-coords flag in faidx script suppresses "sequence:start-end" in output sequence headers (#121)
• Sequence.long_name is deprecated in favor of Sequence.fancy_name. Sequence.longname() is now removed from the codebase after previous DeprecationWarnings.

v0.4.9.2

• Fixed issue in key_function validation logic

v0.4.9.1

• Faster faidx -i nucleotide code

v0.4.9

• Support for resolving duplicate keys using duplicate_action argument to Fasta
• Support for BGZF (block gzip compressed) FASTA files (#77) (Thanks @terrycojones!)
• Support for numpy 32bit and 64bit integer comparisons (#114)
• Better file locking and buffer flushing (#110)

v0.4.8.5

• support for numpy integer types (#114)
• key_function now applies to indexing as well as sequence retrieval (#111)
• preliminary support for BGZF compressed files (#77) (Thanks @terrycojones)

v0.4.8.4

• Fasta(mutable = True) now performs synchronous writes (#110)

v0.4.8.3

• Sequence.longname is now Sequence.long_name to match FastaRecord.long_name (thanks @michaelbarton)
• Sequence.longname now raises a DeprecationWarning, and will be eliminated in a future release.

v0.4.8.2

Added faidx -e option.