Release 0.6.0

Major Features And Improvements

Concat operation now supports special modes to concatenate large number of files in parallel when the downstream operation is passed to sort (reported by @ekg). This mode is ~30-fold faster / thread and is parallelizable.
Introducing the experimental aggregation subroutine. Modelled based on the Datashader pipeline, Tomahawk can now natively construct aggregated datasets of billions of points into datasets of pixels that can be plotted using raster operations efficiently.
Slicing operator for two files has been re-instated. Use -O for output type (two or ld) when invoking the view subroutine

Bug Fixes and Other Changes

Bug fixes
- Problem when importing variants within 1-base of each other (reported by @ekg)
- Memory requirement problems when parallel sort merging very large files (reported by @ekg)
- Removed default missingness filter (from 0.2 to 1.0) to support importing bcf with few samples
- Fixed problem with linked intervals during slicing. Now correctly selects targetted associations based on both mates (reported by @ekg)

Release 0.5.3

Minor Features And Improvements

Default build now uses SSE4.2 only
Will correctly build dylib on OSX
Added filtering for maxmimum cell counts when viewing two files (flags -1,-2,-3,-4)

Bug Fixes and Other Changes

Bug fixes
- Viewing a two file with -l or -u now correctly filters output when passed as the sole parameter

Release 0.5.2

Minor Features And Improvements

Now correctly filters two files

Release 0.5.1

Minor Features And Improvements

Meta index for faster queries of two files now builds correctly
Interval slicing for sorted two files now outputs filtering hits for the from positions only
Haplotype counts in binary two files are now stored as double instead of float

Release 0.5.0

Breaking Changes

All changes are breaking because of structural rechanges. No files generated prior to this release are supported.

Minor Features And Improvements

Build process now smoother
Shared C++ library is now built by default

Release 0.4.0

Breaking Changes

All changes are breaking because of structural rechanges. No files generated prior to this release are supported.

Major Features And Improvements

Added large number of new FLAG fields to two files
Reintroduced speed mode for calculations when contingency tables are not required

Release 0.3.3

Major Features And Improvements

None

Bug Fixes and Other Changes

Bug fixes
- Viewing sorted two files fixed for edge cases

Release 0.3.2

Breaking Changes

Using old sorted files does not break functionality but have to be resorted

Major Features And Improvements

Sorting two files not correctly produces two indices used for fast queries
Viewing sorted two files is significantly faster

Bug Fixes and Other Changes

Bug fixes
- Sort merge (tomahawk sort -M) now produces the correct index

Release 0.3.1

Major Features And Improvements

None

Bug Fixes and Other Changes

Bug fixes
- Sort merge (tomahawk sort -M) now produces the correct output
- view ABI command now correctly triggers -h/-H flag

Release 0.3.0

Breaking Changes

All changes are breaking. There is no backwards compatibility from this release!

Major Features And Improvements

All major classes are now implemented as STL-like container and are as decoupled as possible
External sort file handles now buffer a small amount of data to reduce random access lookups
two/twk entries are no longer forced to be accessed by unaligned memory addresses
Both two and twk are now self-indexing. All external indices are now invalid

Bug Fixes and Other Changes

Bug fixes
- Too many to list here
Examples updates:
- Added R script to demonstrate simple plotting using base
- Added several figures demonstrating some keys concepts of Tomahawk