CHANGELOG

MiLib 1.13 (15 Apr 2020)
========================

-- Fix quality trimming report json
-- Added CIGAR string for alignments
-- Implementation of barcodes generator in `SequencesUtils`
-- Upgrade to picocli 3


MiLib 1.12 (19 Nov 2019)
========================

-- `SequenceProvider.forceInitialize()` method added
-- Minor fix for `CachedSequenceProvider.size()`
-- Clustering: clustering strategy can now provide custom mutation guide for the candidate search 
   procedure
-- Clustering: tree search parameters now can be cluster dependent
-- Jacksion library upgraded to 2.9.10.1
-- Other dependencies upgrade


MiLib 1.11 (22 Jul 2019)
========================

-- Advanced high-quality island extraction algorithm
-- Jeckson library upgrade to 2.9.9


MiLib 1.10 (30 Jan 2019)
========================



MiLib 1.9 (23 Oct 2018)
========================

-- JDK 8
-- New `knownObject` record type for `PrimitivIO` (:warning: stream format changed, not backward 
   compatible)
-- Separate PrimitivI/O state objects
-- Better error messaging for fastq reader
-- SequenceTreeMap cleanup
-- `Clustering` class code cleanup
-- Serializers for some meta object types
-- Minor fixes and enhancements


MiLib 1.8.4 (14 Jun 2018)
========================

-- Major revision and fix for affine scoring alignment algorithms
-- Tools for (low) quality trimming
-- `setFloatingRightBound` and `setFloatingLeftBound` generalized on `AbstractKAlignerParameters`
-- Fix for `setReadId` for `SingleReadLazy`
-- Fixes IllegalState exception in Sorter with empty input


MiLib 1.8.3 (30 Mar 2018)
========================

-- New `greedy` search flag for SequenceTreeMap, with `greedy = false` SequenceTreeMap searches for 
   all possible alignments between target sequence and sequences stored in the Trie
-- Fixed bug in `SequenceTreeMap`, leading to missed matches for sequences with at least one 
   deletion and one insertion compared to the target one
-- Fixed bug in `MismatchOnlyPairedReadMerger`, leading to wrong `IllegalArgumentException`


MiLib 1.8.2 ( 7 Feb 2018)
========================

-- Upgrade to on-HDD `Sorter` (smarter memory / speed trade-offs)
-- Verbose exception for non-standard line breaks in fastq file
-- `shiftIndelsAtHomopolymers` for alignments
-- Reads/targets merging algorithms now supports differnet `IdentityType` (`Unweighted`, 
   `MinimalQualityWeighted`)
-- Added `SequenceReadUtil.setReadId`
-- Added `PipeDataInputReader`
-- `ArrayUtils` methods
-- Added `move` method in `Alignment` object
-- Many minor optimizations, enhancements


MiLib 1.8.1 (24 Jul 2017)
========================

-- Backward compatibility for JSON serialization of AffineGapAlignmentScoring


MiLib 1.8 ( 7 Jun 2017)
========================

-- `Randomizer` class, for memory-efficient randomization of huge streams of objects
-- `reverse()` method for Range
-- `AlignmentScoring` JSON serialization/deserialization fix
-- `size()` method added to `SequenceProvider` interface
-- Optional automatic inference of `size` value for `CachedSequenceProvider`
-- Added `com.milaboratory.core.Range#getAbsoluteRangeFor`
-- Added `com.milaboratory.core.sequence.provider.SequenceProviderUtils#subProvider`
-- `MultiNSequenceWithQuality` abstraction
-- Build host information added to output jar artefact


MiLib 1.7.2 ( 2 Mar 2017)
========================

-- Sorter for long streams of object, uses merge sort and temp file to store predsorted stream 
   chunks (`com.milaboratory.util.sorting.Sorter`).
-- Added `Range#hasSameDirection` method
-- Added `SmartProgressReporter#extractProgress` for `CountingOutputPort`
-- Minor corrections for `com.milaboratory.util.TempFileManager`


MiLib 1.7.1 (30 Jan 2017)
========================

-- Added two methods `com.milaboratory.core.alignment.Alignment#convertToSeq1Range` and 
   `com.milaboratory.core.alignment.Alignment#convertToSeq2Range`
-- Added `without` method in `Range` class
-- Added `getMinimalMatchScore`, `getMaximalMatchScore`, `getMinimalMismatchScore` and 
   `getMaximalMismatchScore` method to AlignmentScoring
-- Fix for noncommutativity of range `intersect` and `intersectsWith` methods


MiLib 1.7 (26 Dec 2016)
========================

-- Comprehansive set of methods to convert nucleotide mutations to amino-acid (e.g. 
   `nt2aaWithMapping`, `nt2aaDetailed`, etc...)
-- Encode method for mutations with separator
-- Mutations#extractMutationsForRange renamed to extractRelativeMutationsForRange
-- Addded Mutations#extractAbsoluteMutationsForRange
-- Added AlignmentIterator
-- minor: Added com.milaboratory.cli.ActionHelpProvider#printDefaultHelp
-- minor: Fix for toString of empty mutations


MiLib 1.6 (16 Dec 2016)
========================

-- ConvertPosition in Mutations and Alignments classes renamed to convertToSeq2Position 
   **[migrate]**
-- Introduced new method for reverse position conversion in alignments convertToSeq1Position (in 
   Mutations and Alignments classes)


MiLib 1.5.1 ( 8 Nov 2016)
========================

-- Additional sequence id extractor in RandomAccessFastaIndex
-- Redberry pipe -> 0.9.6


MiLib 1.5 ( 8 Sep 2016)
========================



MiLib 1.4 (20 Jun 2016)
========================

-- A bunch of classes migrated from MITools to MiLib
-- Classes for CLI abstraction moved from MiTools
-- Version of Jackson updated to 2.7.4
-- Minor fixes CLI
-- Correct offset values in PairedReadMergingResult
-- *(Alpha)* Mutations aggregator


MiLib 1.3 ( 1 Feb 2016)
========================

-- Major refactoring of wildcards
-- Reverse-complement for sequences with wildcards
-- Motif matching in sequences with wildcards
-- Bitap search in sequences with wildcards
-- Rename of `mismatchAndIndelMatcherFirst()` -> `substitutionAndIndelMatcherFirst()`


MiLib 1.2 (29 Dec 2015)
========================

-- KAligner2: new aligner for amplicon-like sequencing data processing (alignments with long gaps, 
   performance ~20-100us per query).
-- Common API for batch aligners.
-- Full set of banded aligners with affine gap
-- Nuc. Mutations -> AA. Mutations utility method
-- `removeMutationsInRanges` method for Mutations class
-- minor: Output jar file now contain version inforamtion accessible using 
   `VersionInfo.getVersionInfo(...)`
-- minor: FastqRecordsReader made public
-- minor: Access to RawFastaRecord


MiLib 1.1.2 ( 6 Oct 2015)
========================

-- Compatibility writer for Enums in PrimitivI


MiLib 1.1.1 (11 Sep 2015)
========================

-- exception in fastq files with `.` symbol in sequence fixed, `.` symbol in fastq files converts 
   to 'N'
-- verbose error message if fastq file parsing failed
-- added `MultiAlignmentHelper` for pretty output of several alignments with common sequence (like 
   V, D, J alignments in VDJ assembly)
-- `charFromCodeAt` renamed to `symbolAt` in base Sequence class
-- BLAST aligner now supports management of several concurrent BLAST instances (to align queries 
   from a common pipe)


MiLib 1.1 (21 Aug 2015)
========================

-- Substantial modification for basic Sequence abstraction
-- Wildcars are incorporated into normal alphabets
-- IncompleteSequences are removed
-- `FastaReader` isolated into sepparate entity (now it can read AminoAcid sequences) and special 
   wrapper is created (`FastaSequenceReaderWrapper`) to wire it back to SequenceReader abstraction
-- `BatchAligner` abstraction
-- `SimpleBatchAligner`
-- Extancive binding to Blast implemeting BatchAligner abstraction for interoperability with other 
   BatchAligners
-- many other small modifications to support new abstraction


MiLib 1.0.2 (12 Aug 2015)
========================

-- New automated release mechanism
-- More named presets for TreeSearchParameters (`two`/`three`/`four` x 
   smatches`/`indels`/`mismatchesOrIndels`)


MiLib v1.0.1 (16 Jun 2015)
========================

-- Added StringSerializer to PrimitivIO
-- Fixed annoying time format in SmartProgressReporter
-- Converter from BTOP to Mutations



MiLib v1.0 (23 Jan 2015)
========================

-- Initial release