Skip to content

Latest commit

 

History

History
448 lines (248 loc) · 13 KB

CHANGELOG.md

File metadata and controls

448 lines (248 loc) · 13 KB
Raw

4.0.1 (2024-11-12)

4.0.0 (2024-11-12)

3.0.3 (2024-11-11)

3.0.2 (2024-11-11)

  • republish v3.0.1 since it got tagged on a deleted branch

3.0.1 (2024-11-11)

3.0.0 (2024-11-07)

2.0.4 (2024-08-09)

2.0.3 (2024-07-23)

Reverts

2.0.2 (2024-02-21)

  • Update typescript-eslint config and related fixes

2.0.1 (2024-2-20)

  • Update to buffer-crc32 1.0.0
  • Fix BAM header parsing of refNames containing a :

2.0.0 (2023-06-08)

Features

  • Add explicit buffer import
  • Remove cross-fetch and object.entries polyfills
  • Improve typescripting
  • Remove chunkSizeLimit and fetchSizeLimit

1.1.18 (2022-12-17)

  • Use es2015 for nodejs build

1.1.17 (2022-07-18)

  • Bump devDeps and generic-filehandle to 3.0.0

1.1.16 (2022-03-30)

  • Add src directory for better source maps

1.1.15 (2022-03-18)

  • Fix for htsget failing with message 'input must be buffer, number, or string, received object'
  • Speed improvement by caching chunks of features

1.1.14 (2022-03-14)

  • Fix seq function for corner case

1.1.13 (2022-02-25)

  • Optimize qual and sequence string record functions for less GC pressure

1.1.12 (2022-02-17)

  • Add blocksForRange method to BamFile class to help stats estimation in JBrowse 2

1.1.11 (2022-01-26)

  • Cache setup of index file parsing

1.1.10 (2022-01-18)

  • Make _refID and flags public fields
  • Small internal changes to the handling of opts

1.1.9 (2021-12-14)

  • Add ESM module export in package.json (smaller bundle size for consumers)
  • Cache BAI readFile result for compatibility with node.js native filehandles (which otherwise fail if re-reading the filehandle twice)

1.1.8 (2021-05-21)

  • Fix types for yieldThreadTime

1.1.7 (2021-05-21)

  • New param yieldThreadTime to constructor to yield while processing

1.1.6 (2021-02-20)

  • Add qualRaw function on records for getting raw qual score array instead of string

1.1.5 (2020-12-11)

  • Allow getHeaderText to accept cancellation options

1.1.4 (2020-12-11)

  • Add canMergeBlocks to CSI code (already existed in BAI)
  • Add suggestion from @jrobinso about reg2bins modification for memory saving (Thanks!)
  • Add getHeaderText() method for getting a text string of the header data

1.1.3 (2020-10-29)

  • Fix usage of feature.get('seq'), was using feature.getReadBases before this

1.1.2 (2020-10-02)

  • Fix signedness in BAM tags (#65)
  • Remove unused seq_reverse_complemented tag from _tags()

1.1.1 (2020-09-20)

  • Remove JBrowse specific results from tags

1.1.0 (2020-08-28)

  • Add support for the CG tag for long CIGAR strings

1.0.42 (2020-08-19)

  • Small bugfix for Htsget specifically

1.0.41 (2020-08-19)

  • Add htsget example
  • Support opts object to getHeader allowing things like auth headers to be passed right off the bat

1.0.40 (2020-07-30)

1.0.39 (2020-07-30)

  • Don't use origin master in the follow-tags postpublish command for cleaner version publishing

1.0.38 (2020-07-30)

  • Direct construction of qual/seq toString
  • Improve performance of the uniqueID calculation for pathological cases where there are tons of bins

1.0.37 (2020-06-06)

  • Typescript only release: export BamRecord types

1.0.36 (2020-03-05)

  • Adds a shortcut to stop parsing chunks after a record is detected to be outside the requested range while decoding

1.0.35 (2020-02-04)

  • Update scheme used to calculate unique fileOffset based IDs using @gmod/bgzf-filehandle updates

1.0.34 (2020-01-24)

  • Small fix for using id() instead of .get('id') for weird SAM records containing ID field

1.0.33 (2020-01-24)

  • Perform decoding of entire chunk up front to aid caching, reverts change in 1.0.29

1.0.32 (2019-11-16)

  • Add a speed improvement for long reads by pre-allocating sequence/quality scores array

1.0.31 (2019-11-07)

  • Fix example of the "ID" field failing to return the right data

1.0.30 (2019-11-07)

  • Add fix that was causing the parser to not return all tags from the _tags API

1.0.29 (2019-10-31)

  • Decoding of the BAM records at time of use instead of entire chunk decoded up front
  • Alternate chunk merging strategy inspired by igv.js code

1.0.28 (2019-10-29)

  • Add CSI index block merging
  • Change unique ID generator to be smaller numeric IDs

1.0.27 (2019-10-10)

  • Make feature IDs become generated based relative to the exact bgzip block

1.0.26 (2019-10-01)

  • Restore issue with getRecordsForRange not returning all features (#44)
  • Fix compatibility with electron (#43)
  • Fix usage of feature.get('seq')

1.0.25 (2019-09-29)

  • Fixed some typescript typings

1.0.24 (2019-09-27)

  • Added typescript typings

1.0.22 (2019-09-27)

  • Added typescript typings
  • Botched release, was removed from npm

1.0.22 (2019-09-03)

  • Fixed issue with features having different IDs across different chunks (#36)

1.0.21 (2019-08-06)

  • Add a fix for the small chunk unpacking re-seeking in the same bgzf block repeatedly (#35)

1.0.20 (2019-06-06)

  • Added a method for smaller chunk unpacking, by modifying the header parsing to return smaller chunks and the bgzf unzipping to respect chunk boundaries (#30)
  • Use fileOffset as bam feature ID which previously was crc32 of the BAM buffer which consequently speeds up processing and allows exact duplicate features

1.0.19 (2019-05-30)

  • Added lineCount and hasRefSeq functions to BamFile, each accepting a string seqName
  • Fixed aborting on index retrieval code

1.0.18 (2019-05-01)

  • Bump generic-filehandle to 1.0.9 to fix error with using native fetch (global fetch needed to be bound)
  • Bump abortable-promise-cache to 1.0.1 version to fix error with using native fetch and abort signals

1.0.17 (2019-04-28)

  • Fix wrong number of arguments being passed to the readRefSeqs file read() invocation resulting in bad range requests

1.0.16 (2019-04-28)

  • Added indexCov algorithm to retrieve approximate coverage of the BAM inferred from the size of the BAI linear index bins
  • Fixed abortSignal on read() calls
  • Updated API to allow bamUrl/baiUrl/csiUrl

1.0.15 (2019-04-04)

  • Added check for too large of chromosomes in the bai bins
  • Added aborting support (thanks @rbuels)
  • Refactored index file class

1.0.14 (2019-01-04)

  • Add hasRefSeq for CSI indexes

1.0.13 (2018-12-25)

  • Use ascii decoding for read names
  • Fix error with large BAM headers with many refseqs

1.0.12 (2018-11-25)

  • Faster viewAsPairs operation

1.0.11 (2018-11-23)

  • Fix for ie11

1.0.10 (2018-11-18)

  • Add a maxInsertSize parameter to getRecordsForRange

1.0.9 (2018-11-16)

  • Allow bases other than ACGT to be decoded
  • Make viewAsPairs only resolve pairs on given refSeq unless pairAcrossChr is enabled for query

1.0.8 (2018-10-31)

  • Add getPairOrientation for reads

1.0.7 (2018-10-19)

  • Re-release of 1.0.6 due to build machinery error

1.0.6 (2018-10-19)

  • Add bugfix for where bytes for an invalid request returns 0 resulting in pako unzip errors

1.0.5 (2018-10-16)

  • Add a bugfix for pairing reads related to adding duplicate records to results

1.0.4 (2018-10-13)

  • Support pairing reads
  • Fix pseudobin parsing containing feature count on certain BAM files

1.0.3 (2018-09-25)

  • Remove @gmod/tabix dependency

1.0.2 (2018-09-25)

  • Fix CSI indexing code

1.0.1 (2018-09-24)

  • Rename hasDataForReferenceSequence to hasRefSeq

1.0.0 (2018-09-24)

  • Initial implementation of BAM parsing code