v0.8.1

What's Changed

• 75 fix-incorrect decoding of extended buffer on batch decoding in bq by @noamteyssier in #76
  • quick patch with bq decoding

Full Changelog: v0.8.0...v0.8.1

v0.8.0

What's Changed

• batch decoding for bq and vbq readers with zero-copy API by @noamteyssier in #74
  • Large performance gains for both BQ and VBQ (about 2x) making use of batch decoding.
  • Breaking sheader and xheader API with zero copy access to buffers instead of filling buffers
  • Addition of sseq and xseq functions to Record API which gives zero-copy access to pre-decoded sequence buffers

Full Changelog: v0.7.8...v0.8.0

v0.7.8

What's Changed

• 70 improve zstd decompression times by @noamteyssier in #71
• Simplify API for encoding blocks
• Zero copy vbq by @noamteyssier in #72

Full Changelog: v0.7.7...v0.7.8

v0.7.7

What's Changed

• 68 make contexts more flexible and efficient by @noamteyssier in #69

Full Changelog: v0.7.6...v0.7.7

v0.7.6

What's Changed

• 63 introduce a sequence context by @noamteyssier in #64
• 65 built in error transformations for generic errors by @noamteyssier in #66

Full Changelog: v0.7.5...v0.7.6

v0.7.5

What's Changed

• 49 support 4bit encodings by @noamteyssier in #50
• 48 move vbq index into file by @noamteyssier in #51
• 47 allow optional headers in vbq files by @noamteyssier in #52
• 53 convert vbq index cumulative records to u64 by @noamteyssier in #54
• 55 update documentation for new vbq formatting by @noamteyssier in #56
• 57 make per record flags optional by @noamteyssier in #58
• Binseq v2 by @noamteyssier in #60

Full Changelog: v0.6.5...v0.7.5

v0.6.5

What's Changed

• 1 support interleaved paired records by @noamteyssier in #2
• 4 use binseq encode and decode standards by @noamteyssier in #5
• 6 implement owned records and an iterator over owned records by @noamteyssier in #7
• 8 fix thread contention issue with processor by @noamteyssier in #9
• 10 build a mmap version of reader 1 by @noamteyssier in #11
• 12 expose direct writing functions by @noamteyssier in #13
• 15 pass thread id to all function in parallel by @noamteyssier in #16
• 17 refactor tid usage by @noamteyssier in #18
• 19 introduce invalid nucleotide policy in the writer by @noamteyssier in #20
• 23 allow users to reset mmap readers back to start by @noamteyssier in #24
• 25 consolidate mmap reader into single struct by @noamteyssier in #26
• 27 flatten code base by @noamteyssier in #28
• 29 setup thiserror system and remove anyhow by @noamteyssier in #30
• 31 setup a thread safe encoder by @noamteyssier in #32
• 33 improve documentation throughout repo by @noamteyssier in #34
• 37 refactor repository and include bq and vbq as separate submodules by @noamteyssier in #39
• Streaming read write by @drbh in #38
• 41 add num records as method on binseqreader enum by @noamteyssier in #42
• 43 process parallel range by @noamteyssier in #44
• 21 handle num threads equals zero by setting to max by @noamteyssier in #45

New Contributors

• @noamteyssier made their first contribution in #2
• @drbh made their first contribution in #38

Full Changelog: https://github.com/ArcInstitute/binseq/commits/v0.6.5