All notable changes to this project will be documented in this file. This project adheres to Semantic Versioning.
- Adding
bcf::Record::unpack()
for explicitely unpacking BCF records. - Fixed
bcf::synced::SyncedReader::record()
. bam::Record::cigar()
now returns a reference (in constant time) and needsbam::Record::unpack_cigar()
to be called first.- Allow to create Cigar string from
bio_types::Alignment
. - Provide a cached variant of obtaining cigar string.
- Lots of small usability improvements.
- Initial implementation of synced BCF reader interface.
- Several small helper methods for BAM readers.
- Not skipping
fileformat=
header any more. - BCF records are always unpacked when reading.
- Moved unpacking of BCF records into constructor to prevent race conditions.
- Fixed bug in retrieving BCF record id.
- Fixed bug in the filter iterator of BCF.
- more push functions for BCF.
- bcf::IndexedReader
- support for writing bcf FILTER field
- setting thread count in all readers and writers
- setting ID and alleles in bcf records
- support for tabix indexes
- convert CIGAR to and from strings
- Serde support for bam records.
- Various convenience improvements in the API.
- Raw Htslib bindings are now generated on the fly.
- Switched to Htslib 1.6.
- Fixed a potential dangling pointer to the header in bcf records.
- Various small API improvements.
- HeaderView of bam and bcf can now be cloned.
- An efficient ringbuffer for accessing BCF regions
- An efficient ringbuffer for accessing BAM regions
- Improved mutability annotation for readers.
- Ability to clone bam records.
- Ability to set only qname.
- Further improved CIGAR string API.
- Improved documentation.
- Adapt to changes in Rust 1.18 that caused compilation issues.
- Support seek and tell to handle virtual offsets.
- Renamed previous seek method into fetch (in line with pysam).
- Improved CIGAR API.
- Updated dependencies.
- A SAM writer.
- Improved CIGAR string API using a newtype wrapper.
- Improved pileup API.
- Support threaded writing for BAM files.
- Prelude module to easily import all relevant traits.
- fine-grained constructors for STDIN/STDOUT, paths and URLs
- better template handling with bam files
- improved genotype handling
- improved error handling
- improved missing value handling
- Finally converted the last unit error types to real error types (really now!).
- More error types.
- Error types now properly implement the Display and the Error trait.
- Mark all records as Send and Sync.
- Improved error messages.
- Check existence of BAM when instantiating Readers.
- Improved handling of memory allocation for BCF and BAM records.
- Fixed a memory leak occuring when creating a new BAM record (thanks to @andrelmartins).