ChangeLog

+2018-07-06  twu
+
+    * VERSION, config.site.rescomp.tst, index.html, path-solve.c, src, trunk:
+      Changed check on sense_endpoints to antisense_endpoints before trying to
+      remove an end segment
+
+2018-06-29  twu
+
+    * types.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to define Univcoordlist_pop
+
+    * stage3hrdef.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to add nmatches_amb as a field
+
+    * stage3hr.c, stage3hr.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to consider trim amount in the
+      final comparison among alignments
+
+    * stage3.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to take queryseq as an argument
+      in merge procedures
+
+    * stage3.c: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to copy all pairs when performing
+      a merge, to peelback to indels on the medial side when taking the
+      continuous solution, and to remove indels in insert_gapholders
+
+    * pair.c, pair.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to implement Pair_split_circular
+
+    * junction.c, junction.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to implement Junction_new_generic
+
+    * gmap.c: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to split circular alignments to
+      cross the origin
+
+    * chimera.c, chimera.h: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to disallow chimeras to circular
+      chromosomes
+
+    * path-solve.c: Merged revisions 215752 to 215897 from
+      branches/2018-06-15-path-solve-junctions to create a new junction at the
+      end, rather than use a precomputed junction
+
+    * path-solve.c: At ends, pushing NULL instead of the previously computed
+      junction
+
+    * pair.c, pair.h: Implemented checking procedures
+
+2018-06-15  twu
+
+    * stage3hr.c: Removed extraneous characters in debugging statement
+
+2018-06-14  twu
+
+    * stage3.c: Fixed handling of end exons in end trimming procedures.  Using
+      new interfaces to Pair_clip_bounded_list_5 and Pair_clip_bounded_list_3
+
+    * pair.c, pair.h: Implemented separate 5' and 3' versions of
+      Pair_clip_bounded_list
+
+    * chimera.c: Fixed debugging statements to use new interface to
+      Sequence_stdout
+
+2018-05-31  twu
+
+    * pair.c: In converting pairarray to substrings, now resetting exon
+      variables after an insertion or deletion
+
+    * iit-read.c: Added a warning in using IIT_read for a version 1 IIT
+
+    * iit_get.c, iit_store.c: Fixed some memory leaks
+
+    * getline.c, getline.h: Returning string_length with Getline_wlinefeed
+
+    * iit_store.c: Using Getline_wlinefeed instead of Getline_wlength
+
+2018-05-30  twu
+
+    * stage1hr.c: Fixed uninitialized variable
+
+    * substring.c: Fixed assertion
+
+    * stage3hr.c: Checking if there is enough space at ends of the chromosome
+      before resolving inner exons
+
+    * gmapindex.c: Assigning variable and not pointer when clearing empty space
+      at end of line
+
+    * atoiindex.c, cmetindex.c, indexdb-write.c, snpindex.c: Calling
+      Access_allocate_private properly for machines where mmap is not available
+      or disabled
+
+    * gregion.c: Improved procedure for finding unique gregions by sorting by
+      support instead of weight, by checking if query coordinates are
+      consistent, and handling cases where endpoints are equal
+
+    * Makefile.gsnaptoo.am: Not making iit_pileup
+
+2018-05-26  twu
+
+    * Makefile.gsnaptoo.am: Added getline.c and getline.h to library
+
+    * get-genome.c: Using variable line instead of Buffer
+
+    * config.site.rescomp.prd, config.site.rescomp.tst, gmap.c, index.html,
+      oligoindex_hr.c, src, stage2.c, stage3.c, trunk: Merged revisions 215481
+      through 215483 from branches/2018-05-25-restore-compute-ends to restore
+      GMAP behavior from 2018-03-20 and use Stage3middle_T,
+      Stage3_compute_middle, and Stage3_comput_ends for better alignment at ends
+
+    * datadir.c: Fixed extraneous parenthesis
+
+    * datadir.c: Using Getline
+
+    * Makefile.gsnaptoo.am, iit_get.c, iit_store.c: Using Getline
+
+2018-05-25  twu
+
+    * gmap.c: Ensuring that calls to strncpy are followed by setting the end to
+      be '\0'.  Using malloc instead of calloc in these situations
+
+    * Makefile.gsnaptoo.am, datadir.c, get-genome.c, getline.c, getline.h,
+      gmapindex.c: Using calls to Getline to prevent problems with buffer
+      overflow
+
+    * chrsubset.c, iit-read-univ.c, iit-read.c, iit_store.c, samread.c,
+      sequence.c, shortread.c, splicing-scan.c, stage2.c, stage3hr.c: Ensuring
+      that calls to strncpy are followed by setting the end to be '\0'.  Using
+      malloc instead of calloc in these situations
+
+    * substring.c: Fixed calculation of mandatory_trim_left and
+      mandatory_trim_right for alias to be positive, rather than negative
+
+2018-05-12  twu
+
+    * VERSION: Updated version number
+
+    * src: Made changes
+
+    * fa_coords.pl.in, gmap_build.pl.in, gmap_process.pl.in: Added options to
+      handle fastq files and to reverse complement all sequences
+
+    * stage1hr.c: No longer iterating through Segment_search
+
+    * stage1hr.c: Not calling Stage1_init or other Stage1 procedures when
+      querylength < index1part
+
+    * substring.c: Not doing aliasing and unaliasing on alt substrings. 
+      Computing mandatory_trim_querystart and mandatory_trim_queryend instead of
+      mandatory_trim_left and mandatory_trim_right
+
+    * trindex.c: Revised instructions
+
+    * iit-read-univ.c, iit_get.c, iit_pileup.c, iit_tally.c: Freeing div name
+      allocated by new versions of Parserange routines
+
+    * genome128_consec.c, genome128_hr.c: Added header files needed for SSSE3
+      computers
+
+2018-04-30  twu
+
+    * shortread.c: Formatting change
+
+    * parserange.c: Allow colons in accession names
+
+    * gmapindex.c: Allow colons in accession.  Implementing revcomp by using '-'
+      sign for contig length
+
+    * get-genome.c: Freeing chromosome string after call to Parserange
+
+2018-04-21  twu
+
+    * stage1hr.c: Merged revision 214805 from branches/2018-04-21-fix-anchors to
+      use Solve_segment_all instead of Concordance_filter_records for paired-end
+      reads
+
+    * stage3hr.c: Added variable remap_transcriptome_p
+
+    * stage1hr.c: Added variable remap_transcriptome_p
+
+    * stage3.c: Checking for a NULL stage3 in Stage3_split
+
+    * intersect.c: Fixed comments
+
+    * stage1.c: Using Oligospace_T to cast 0
+
+    * kmer-search.c: Fixed issues with transcript coordinates
+
+    * gmap.c, gsnap.c, inbuffer.c, inbuffer.h, sequence.c, sequence.h,
+      shortread.c, shortread.h: Adding a command-line option
+      --read-files-command
+
+    * extension-search.c, segment-search.c, stage1hr.c: Using (Oligospace_T) as
+      a cast for 0
+
+    * Makefile.gsnaptoo.am: Added a command-line option --read-files-command
+
+    * fopen.c, fopen.h: Added a command-line option --read-files-command
+
+2018-04-20  twu
+
+    * gsnap.c: Changed flag from --use-transcriptome to --use-transcriptome-only
+
+2018-03-26  twu
+
+    * samprint.c: Using new interface to Pair_print_sam
+
+    * block.c, config.site.rescomp.prd, gmap.c, index.html, indexdb.c,
+      oligoindex_hr.c, pair.c, src, stage1.c, stage2.c, stage3.c, translation.c,
+      trunk: Merged revisions 214439 through 214446 from
+      branches/2018-03-24-restore-gmap to restore speed and sensitivity of GMAP
+      by eliminating use of Stage3_compute_ends, restoring use of multiple
+      oligoindices, and fixing CDS phases for GFF3 output
+
+2018-03-24  twu
+
+    * indexdb.c: Removed option for expanding offsets
+
+    * gmap.c: Fixed memory issues with new calls to Stage3_merge_local and
+      Stage3_merge_chimera, and appending middlepieces to stage3list. Changed
+      default gff3_cds to be genomic.  Ignoring option for --expand-offsets
+
+    * stage3.c, stage3.h: Removed cigar_tokens and intronp as fields for
+      Stage3_T objects. Implemented Stage3_copy and using it in
+      Stage3_merge_local and Stage3_merge_chimera
+
+    * pair.c, pair.h: Removed cigar_tokens and intronp as paramters to
+      Pair_print_sam
+
+2018-03-23  twu
+
+    * substring.c: Added assertions about alts, to make sure we don't use
+      alignstart_trim or alignend_trim fields in those cases
+
+    * stage3hr.c: Checking for alts when getting chrpos_low or chrpos_high
+
+    * samprint.c: Formatting changes
+
+    * dynprog.h: Setting AMBIGUOUS score to be 3, to make cmet and atoi
+      alignments equivalent to standard
+
+    * cigar.c: For sam_hardclip_use_S option, returning hardclips to be 0 so
+      they don't affect the query sequence in SAM output
+
+    * stage3.c: Restored force of single gaps, to avoid problems with
+      add_dual_break later
+
+2018-03-21  twu
+
+    * iit_tally.c: Initial import
+
+    * VERSION: Updated version number
+
+    * dynprog.c, dynprog.h, dynprog_cdna.c, dynprog_cdna.h, dynprog_end.c,
+      dynprog_end.h, dynprog_genome.c, dynprog_genome.h, dynprog_simd.c,
+      dynprog_simd.h, dynprog_single.c, dynprog_single.h, gmap.c, gsnap.c,
+      pair.c, path-solve.c, splicetrie.c, splicetrie.h, src, stage3.c, stage3.h,
+      trunk: Merged revisions 214343 through 214360 from
+      branches/2018-03-20-cmet-gmap to make non-standard modes work in GMAP
+      alignments
+
+    * sam-exons.pl.in: Initial import
+
+2018-03-20  twu
+
+    * indexdb.c: Fixed preprocessor macro
+
+2018-03-19  twu
+
+    * VERSION: Updated version number
+
+    * Makefile.gsnaptoo.am, VERSION, atoi.c, atoi.h, atoiindex.c, cmet.c,
+      cmet.h, cmetindex.c, config.site.rescomp.tst, configure.ac,
+      epu16-bitpack64-access.c, epu16-bitpack64-access.h,
+      epu16-bitpack64-incr.c, epu16-bitpack64-incr.h, epu16-bitpack64-read.c,
+      epu16-bitpack64-read.h, epu16-bitpack64-readtwo.c,
+      epu16-bitpack64-readtwo.h, epu16-bitpack64-write.c,
+      epu16-bitpack64-write.h, extension-search.c, extension-search.h,
+      genomicpos.c, genomicpos.h, gsnap.c, index.html, indexdb.h, indexdbdef.h,
+      kmer-search.c, kmer-search.h, localdb-write.c, localdb.c, localdb.h,
+      localdbdef.h, oligo.c, path-solve.c, path-solve.h, reader.c, src,
+      stage1hr.c, trunk, types.h: Merged revisions 214117 through 214304 from
+      branches/2018-03-11-cmet-localdb to implement non-standard modes with
+      support for localdb
+
+    * substring.c: Removed an assertion that is not always valid
+
+    * stage3hr.c, stage3hr.h: Commented out code for Stage3end_substring_high,
+      which is not used anymore
+
+    * samprint.c: Fixed typo in using querylength5 instead of querylength3 for
+      mate
+
+    * samflags.h: Incremented values for number of filestream outputs, to handle
+      XS output files properly
+
+    * outbuffer.c: Always skipping output for OUTPUT_NONE filestream
+
+    * Makefile.gsnaptoo.am: Hiding program iit_pileup
+
+    * oligoindex_hr.c: Allowing debugging statements to work
+
+    * localdb.c: Allowing debugging statements to work
+
+    * kmer-search.c: Added debugging statements
+
+    * indexdb.c: Distinguishing between GSNAP pointer procedures, which need to
+      handle negative diagterms, and GMAP read procedures, where diagterms are
+      always non-negative
+
+    * gsnap.c: Added option for --sam-hardclip-use-S
+
+    * cigar.c, cigar.h: Calling low substring for all circular chromosome hits
+      to get correct chrpos in SAM output.  Added provisions for
+      SAM_hardclip_use_S_p
+
+2018-03-10  twu
+
+    * substring.c: Changed assertion to allow for '*'
+
+    * stage3.c: Added a clip of end5 against the chromosomal bound for
+      alignments on the minus strand
+
+    * intersect.c: Commenting out advance of positions past -diagterm
+
+    * indexdb.c: No longer checking for diagterm < 0, since diagterm <= 0 should
+      be true
+
+    * extension-search.c: Reverted advance of positions past -diagterm and added
+      assertions instead.  Changed algorithm to not add an elt when the
+      positions are invalid.
+
+    * shortread.c: Preventing invalid read when accession contains only one or
+      two characters
+
+    * kmer-search.c, merge-heap-diagonals.c, merge-heap-records.c,
+      merge-simd-diagonals.c, merge-simd-records.c: Added assertions to check
+      that positions >= -diagterm
+
+    * extension-search.c, intersect.c: Skipping or advancing positions so they
+      exceed -diagterm
+
+    * oligo.c: Fixed initialization code so that Clang compiler will accept it
+
+2018-03-09  twu
+
+    * localdb-write.c: Added a message to indicate when writing is done
+
+    * Makefile.gsnaptoo.am, iit_pileup.c: Added ability for the user to specify
+      a genomic range
+
+    * stage3hr.c, stage3hr.h: Fixed bug with circular read not being unaliased
+      because of a soft clip.  Removed macro for soft clips avoiding
+      circularization.  Removed circularalias values of +2 and -2, since
+      alignments now should stay within chromosomal bounds
+
+    * iit_pileup.c: Added code for using filestrings
+
+    * Makefile.gsnaptoo.am, iit-read.c, iit-read.h, iit_pileup.c: Restored
+      iit_pileup program.  Rewrote to take a FASTA file as input
+
+2018-03-05  twu
+
+    * stage3.c: Added back declaration of variables
+
+    * stage3.c: Removed unused variables and parameters
+
+    * samprint.c: Removed unused static variable
+
+    * path-solve.c: Using new interface to Stage3end_new_gmap
+
+    * stage3hr.c, stage3hr.h: Removed unused parameters to Stage3end_new_gmap. 
+      Removed unused code
+
+    * gsnap.c: Removed unused parameter to process_request
+
+    * extension-search.c: Removed unused code
+
+    * stage1hr.c: Fixed memory error when abort_pairing_p is true
+
+    * localdb.c: Defined variables needed for large genomes
+
+    * VERSION, block.c, block.h, distant-rna.c, distant-rna.h,
+      extension-search.c, gmap.c, gsnap.c, kmer-search.c, kmer-search.h,
+      ladder.c, ladder.h, localdb.c, localdb.h, oligo.c, oligo.h, pair.c,
+      pair.h, path-solve.c, path-solve.h, segment-search.c, segment-search.h,
+      splice.c, splice.h, src, stage1.c, stage1.h, stage1hr.c, stage1hr.h,
+      stage2.c, stage3.c, stage3hr.c, stage3hr.h, substring.c, substring.h,
+      terminal.c, terminal.h, trunk: Merged revisions 213978 through 214024 from
+      branches/2018-03-01-ptr-and-diagterm to add potential support for cmet,
+      atoi, and ttoc modes
+
+2018-03-03  twu
+
+    * cigar.c, cigar.h, gsnap.c: Obey behavior of --sam-use-0M flag
+
+    * Makefile.gsnaptoo.am, access.c, atoiindex.c, bitpack64-readtwo.c,
+      cmetindex.c, config.site.rescomp.prd, epu16-bitpack64-read.c,
+      epu16-bitpack64-readtwo.c, epu16-bitpack64-write.c, extension-search.c,
+      filesuffix.h, genome-write.c, genome128_hr.c, gmapindex.c, gsnap.c,
+      iit-write.c, iit_store.c, indel.c, indexdb-write.c, indexdb.c, indexdb.h,
+      indexdbdef.h, intersect.c, intersect.h, kmer-search.c, kmer-search.h,
+      localdb.c, merge-heap-diagonals.c, merge-heap-diagonals.h,
+      merge-heap-records.c, merge-heap-records.h, merge-simd-diagonals.c,
+      merge-simd-diagonals.h, oligoindex_hr.c, pair.c, path-solve.c,
+      segment-search.c, segment-search.h, smooth.c, snpindex.c, src, stage1hr.c,
+      stage1hr.h, stage2.c, stage3.c, stage3hr.c, substring.c, terminal.c,
+      trunk, types.h: Merged revisions 213921 to 213977 from
+      branches/2018-03-01-ptr-and-diagterm to add back support for large genomes
+
+2018-03-02  twu
+
+    * stage3hr.c: Improved debugging statements
+
+    * kmer-search.c: Removed extraneous calls to free lists
+
+    * ladder.c, ladder.h: Implemented Ladder_cutoff to allow for ladders, but
+      restrict computation to MAX_HITS
+
+    * concordance.c: Calling Ladder_cutoff instead of Ladder_maxscore
+
+    * cigar.c: Fixed computation of SAM chrpos.  Always calling
+      Stage3end_substring_low for a single SAM line
+
+    * concordance.c: Applying MAX_HITS to Concordance_pair_up_distant
+
+    * stage1hr.c: Fixed a typo: querypos_rc instead of querypos for minus
+      positions
+
+    * indexdb.c: Hiding Indexdb_ptr_with_diagterm from utility programs
+
+    * pairpool.c, pairpool.h, stage3.c: Using a better method for handling
+      chromosomal bounds in GMAP, by trimming pairs, rather than the pairarray
+
+    * Makefile.gsnaptoo.am, VERSION, config.site.rescomp.prd,
+      config.site.rescomp.tst, extension-search.c, extension-search.h,
+      indexdb.c, indexdb.h, intersect.c, intersect.h, kmer-search.c,
+      kmer-search.h, localdb.c, merge-heap-diagonals.c, merge-heap-diagonals.h,
+      merge-heap-records.c, merge-heap-records.h, merge-simd-diagonals.c,
+      merge-simd-diagonals.h, merge-simd-records.c, merge-simd-records.h,
+      segment-search.c, src, stage1hr.c, stage1hr.h, trunk: Merged revisions
+      213873 through 213921 from branches/2018-03-01-ptr-and-diagterm to use
+      pointers to positions and diagterms, rather than copying positions
+
+    * substring.c: In Substring_set_alt, handling amb_splice_pos properly as
+      being defined from left, rather than querystart
+
+    * Makefile.util.am: Added instructions for bitpack64-test
+
+2018-03-01  twu
+
+    * segment-search.c: Freeing lists provided to Merge_records
+
+    * path-solve.c: Using FREE_ALIGN for freeing results of merge procedures
+
+    * localdb.c: Localdb_read_with_bounds now always returns aligned memory,
+      which can happen if merging is performed.  Using new interface to merge
+      procedures
+
+    * kmer-search.c: Fixed a memory leak in exact procedure when the ends have
+      invalid oligos
+
+    * merge-heap-diagonals.c, merge-heap-records.c, merge-simd-diagonals.c,
+      merge-simd-records.c: Merge procedures no longer free their input lists or
+      streams.  Streams can be aligned or not, so the caller needs to free them
+
+    * pair.c, pairpool.c, pairpool.h, stage3.c: Prevent GMAP results from going
+      beyond chromosomal bounds, when making the final pairarray
+
+    * stage1hr.c: Using new interfaces to exact and approx algorithms
+
+    * stage3hr.c, stage3hr.h: Renamed Stage3end_list_free to Stage3end_gc
+
+    * segment-search.c: Fixing memory leak
+
+    * kmer-search.c, kmer-search.h: Limiting results for exact algorithm with
+      max_hits, and limiting approx algorithm with both max_hits and sizelimit
+
+    * concordance.c: Limiting results for newladder with MAX_HITS
+
+2018-02-28  twu
+
+    * uint8table.c, uint8table.h: Initial implementation
+
+    * Makefile.gsnaptoo.am, stage1hr.c: Using a Uint8table_T object to call
+      repeated occurrences of an oligo invalid
+
+    * substring.c: Allocating querylength+1 for mismatch_positions
+
+    * path-solve.c: Making a better call to Substring_new for the middle
+      diagonal, and handling the case where the result is NULL
+
+    * gsnap.c: Including header for concordance.h
+
+    * terminal.c: Using new interface to Substring_new
+
+    * path-solve.c: Removed all instances of middle_path
+
+    * extension-search.c, kmer-search.c: Using new interfaces to
+      Path_solve_from_diagonals and Substring_new
+
+    * segment-search.c: Using new interface to Path_solve_from_diagonals. 
+      Letting that procedure determine which left and right diagonals are in the
+      correct chrnum
+
+    * path-solve.c, path-solve.h: Path_solve_from_diagonals now takes a middle
+      diagonal, rather than a middle path.  Sets chrnum based on middle
+      diagonal, as determined by Substring_new.  Also considers only left and
+      right diagonals in that chrnum.
+
+    * stage3hr.c: Using new interface to Substring_new
+
+    * substring.c, substring.h: Substring_new now takes chrnum_fixed_p as a
+      parameter.  If true, then the procedure does not recompute chromosomal
+      bounds
+
+    * segment-search.c: Using left instead of lowpos to determine chrnum, to be
+      safe
+
+    * substring.c: Checking for substring being below given chroffset, and
+      recomputing chromosomal bounds
+
+    * segment-search.c: When adding segments to left and right, checking to make
+      sure they fall at least partially into the same chromosome as the anchor
+      segment
+
+    * pair.c: Fixed argument for setting querystart_pretrim and queryend_pretrim
+
+2018-02-27  twu
+
+    * substring.c: Fixed typo
+
+    * pair.c, substring.c, substring.h: Computing more accurate values for
+      querystart_pretrim and queryend_pretrim.  Removed assertion about
+      left_bound and right_bound in Substring_count_mismatches_region
+
+    * concordance.c, stage1hr.c, stage3hr.c, stage3hr.h, stage3hrdef.h: Removed
+      private5p and private3p fields from Stage3pair_T object. Always copying
+      hits when making a pair, needed because the concordance procedure can now
+      delete hits
+
+    * ladder.c: Assigning a value to nhits in all cases from
+      Ladder_hits_for_score
+
+    * concordance.c, stage1hr.c: Restored assignment of abort_pairing_p
+
+    * ladder.c, ladder.h, stage3hr.c, stage3hr.h: Removing duplicates before
+      returning hits at a given score.  Added a duplicates field to hold
+      duplicate hits, and a procedure for deleting them
+
+    * filestring.c: Now handling %p in format statement
+
+    * stage3hr.c: Changed one occurrence of Substring_ambiguous_p to
+      Substring_has_alts_p
+
+    * Makefile.gsnaptoo.am, cigar.c, concordance.c, concordance.h,
+      distant-rna.c, distant-rna.h, gsnap.c, kmer-search.c, kmer-search.h,
+      ladder.c, ladder.h, merge-heap-records.c, merge-heap-records.h,
+      merge-simd-records.c, merge-simd-records.h, merge-uint4.c, method.c,
+      method.h, pair.c, pair.h, path-solve.c, record.h, samprint.c,
+      segment-search.c, segment-search.h, splice.h, src, stage1hr.c, stage1hr.h,
+      stage3hr.c, stage3hr.h, stage3hrdef.h, substring.c, substring.h,
+      terminal.c, terminal.h, trunk, types.h, univdiag.c: Merged revisions
+      213654 to 213758 from branches/2018-02-22-limit-segment-search to add
+      concordance procedures, ladders, and filters for segment search, and to
+      revise the single-end and paired-end procedures
+
+2018-02-23  twu
+
+    * Makefile.gsnaptoo.am, VERSION, block.c, config.site.rescomp.tst,
+      extension-search.c, genomicpos.h, gmap.c, gmap_build.pl.in, gsnap.c,
+      indexdb.c, indexdb.h, intersect.c, intersect.h, intlist.c, intlist.h,
+      junction.c, kmer-search.c, localdb.c, localdb.h, merge-heap-diagonals.c,
+      merge-heap-diagonals.h, merge-heap-records.c, merge-heap-records.h,
+      merge-heap.c, merge-heap.h, merge-simd-diagonals.c,
+      merge-simd-diagonals.h, merge-simd-records.c, merge-simd-records.h,
+      merge-uint4.c, merge-uint4.h, merge.c, merge.h, oligo.c, pair.c, pair.h,
+      path-solve.c, segment-search.c, segment-search.h, splice.c, splicetrie.c,
+      splicetrie.h, src, stage1.c, stage1hr.c, stage1hr.h, stage2.c, stage2.h,
+      stage3.c, stage3hr.c, stage3hr.h, substring.c, substring.h, trunk,
+      types.h, uint8list.c, uint8list.h, uintlist.h, uniqscan.c, util: Merged
+      revisions 213593 through 213652 from branches/2018-02-21-large-genomes to
+      add support for large genomes
+
+2018-02-21  twu
+
+    * stage1hr.c: Checking for a set with zero elements before calling
+      Orderstat_int_pct
+
+    * segment-search.c: Excluding invalid oligos
+
+    * path-solve.c: When novelsplicing is false, creating just one hit
+
+    * oligo.c: Commented out unused procedures
+
+    * kmer-search.c: Handling invalid oligos correctly
+
+    * stage1hr.c: Making poly_A and poly_T oligos not valid.  Added a min
+      sizelimit
+
+    * stage1hr.c: Added comment
+
+    * cigar.c, samprint.c, stage3hr.c, stage3hr.h: Removed code designed for old
+      meaning of Substring_ambiguous_p, now distinct from Substring_has_alts_p
+
+    * indexdb.c: Commented out SIMD procedures for utility programs.  Added
+      support for AVX2
+
+    * extension-search.c, extension-search.h, indexdb.c, indexdb.h,
+      kmer-search.c, kmer-search.h, path-solve.c, path-solve.h,
+      segment-search.c, segment-search.h, splice.c, src, stage1hr.c, stage1hr.h,
+      stage3hr.c, stage3hr.h, substring.c, substring.h, trunk, types.h: Merged
+      revisions 213472 through 213572 from branches/2018-02-16-faster-one-kmer
+      to increase speed and accuracy
+
+    * VERSION, config.site.rescomp.prd, config.site.rescomp.tst, index.html:
+      Updated for latest version
+
+    * stage1hr.c: Added debugging statements
+
+2018-02-17  twu
+
+    * stage3.c: Replaced check with an assertion
+
+    * gmap.c: Filtering out Stage3_T objects with zero npairs
+
+    * junction.c, junction.h: Implemented Junction_typestring
+
+    * stage3.c: Making explicit check for npairs being 0 in Stage3_new
+
+    * pair.c: Handling extra exons with a specific type and transition rules
+
+    * gmap.c: Checking for Stage3_T object having zero pairs before pushing onto
+      list
+
+    * cigar.c: Checking hardclip when printing insertion token.  Calling either
+      Stage3end_substring_low or Stage3end_substring_high
+
+    * stage3hr.c: Added provisions for junctions_HtoL
+
+2018-02-13  twu
+
+    * localdb.c: Performing a check for possible negative coordinates
+
+    * cigar.c, cigar.h: Storing value of extended_cigar_p during setup
+
+    * gsnap.c: Added flags for --endtrim-length and --sam-extended-cigar
+
+    * stage3.c: In Stage3_recompute_coverage, added a check for npairs being 0
+
+    * shortread.c, shortread.h: Added support for end (3') trimming of each read
+
+    * kmer-search.c: Improved debugging statements
+
+    * gmap.c: Added checks for breakpoint being invalid, before running
+      Stage3_mergeable
+
+    * chimera.c: Added checks for npairs being zero
+
+    * gsnap.c: Removed the --expand-offsets flag
+
+    * access.c: Put a limit on the maximum number of attempts to kill an
+      unattached shared memory segment
+
+    * splice.c, substring.c: Checking for cases where the splice search
+      boundaries yield a negative coordinate.  Redefined middle coordinate to be
+      alignstart and alignend, and not the positions 1 bp distal to these
+      coordinates.
+
+    * stage1hr.c: Increased all sizelimits for diagonals from 100 to 3000
+
+    * stage3.c: Defined all parameters for allowed iterations to be 1
+
+    * stage2.c: Added a step to filter stage2 middles, starts, and ends for
+      uniqueness
+
+2018-02-12  twu
+
+    * substring.c: Improved debugging statement
+
+    * stage3hr.c: Added a comment
+
+    * splice.c: Improved debugging statements
+
+    * stage3hr.c: Now allowing Stage3pair_resolve_insides to resolve both ends
+      at the same time
+
+2018-02-11  twu
+
+    * VERSION: Updated version number
+
+    * stage3hr.c: Added debugging statements
+
+    * pair.c: Fixed counting routines to skip over bad pairs correctly
+
+    * substring.c: Removed unused code
+
+    * stage3hr.c: In Stage3pair_resolve_insides and resolve_inside_general
+      procedures, when hits are changed for the hitpair, revising the values of
+      alts_resolve_5 and alts_resolve_3.  Fixes a seg fault when
+      Stage3pair_eval_and_sort tries to resolve the hits a second time
+
+    * path-solve.c, samprint.c, stage3hr.c, stage3hr.h, substring.c,
+      substring.h: Changed variable names to distinguish between an ambiguous
+      splice length and alternative splice coords
+
+2018-02-10  twu
+
+    * config.site.rescomp.tst, pair.c, pair.h, segment-search.c,
+      segment-search.h, src, stage1hr.c, stage3.c, stage3hr.c, trunk: Merged
+      revisions 213287 through 213291 from branches/2018-02-10-fix-bugs to fix
+      bugs related to pairarrays that crossed chromosomal bounds, segments
+      subsuming others, and handling of dual breaks
+
+    * pair.c: Hiding function Pairarray_convert_to_substrings from GMAP
+
+    * cigar.c, cigar.h, distant-rna.c, extension-search.c, filestring.c,
+      filestring.h, gsnap.c, index.html, kmer-search.c, output.c, pair.c,
+      pair.h, path-solve.c, samprint.c, samprint.h, segment-search.c, splice.c,
+      src, stage1hr.c, stage3hr.c, stage3hr.h, substring.c, substring.h, trunk:
+      Merged revisions 213162 through 213277 from
+      branches/2018-02-07-improve-circular to standardize printing of circular
+      and translocation alignments
+
+    * stage3hr.c: Fixed printing of method label
+
+2018-02-08  twu
+
+    * extension-search.c: Making sure that queryoffset does not go outside
+      bounds of 0 to query_lastpos
+
+2018-02-07  twu
+
+    * substring.c: In Substring_new, giving initial values to trim_left and
+      trim_right
+
+    * substring.c: In Substring_new, calling trim_left_end and trim_right_end
+      from 0 and querylength, but trim_novel_spliceends from given querystart
+      and queryend.  If novel spliceends yields a short exon, then using the
+      non-spliced trimming results.
+
+    * stage3hr.c: Calling compute_circularpos on all Substring_T objects
+
+    * samprint.c: Added debugging statements
+
+    * substring.c: In Substring_new, initializing necessary values of
+      Substring_T object earlier than first possible abort and call to
+      Substring_free
+
+    * config.site.rescomp.tst, extension-search.c, gmap.c, index.html, oligo.c,
+      oligo.h, reader.c, reader.h, src, stage1hr.c, svncl.pl, trunk: Merged
+      revisions 213097 through 213130 from branches/2018-02-05-distant-dna to
+      allow for updating of querypos around N's and fix of fatal bug in GMAP in
+      merge_left_and_right_readthrough
+
+2018-02-06  twu
+
+    * substring.c: In trim_left_end and trim_right_end, computing trims relative
+      to querystart and queryend, rather than 0 and querylength.  In
+      embellish_genomic_sam, filling in beginning and end with dashes and stars
+      to avoid genomic_refdiff having a different length than querylength
+
+    * Makefile.gsnaptoo.am: When MAKE_LIB is not defined, not copying headers
+      either
+
+    * VERSION, config.site.rescomp.prd: Updated version number
+
+    * get-genome.c, parserange.c, parserange.h: Allowing Parserange_universal to
+      return a value for whole_chromosome_p
+
+    * substring.c: In Substring_new, always making querystart and queryend
+      correspond to trim_left and trim_right, to fix errors in CIGAR strings
+
+2018-02-05  twu
+
+    * access.c: Using a loop to create (and possibly deallocate) shared memory
+
+    * access.c: Upon startup, if a shared memory segment exists with no other
+      attached processes (possibly corrupted), then deallocating it and creating
+      a new one
+
+    * kmer-search.c: Simplified code in making sure that both ends are in the
+      same chromosome
+
+    * samprint.c: In SAM_compute chrpos, if hardclip_low yields a NULL
+      low_substring, then trying hardclip_high
+
+    * substring.c: Handling the case where both cdna and genome have 'N'
+
+    * stage3hr.c, stage3hr.h: Reversed revision 213080, which is giving CIGAR
+      errors.  Restored Stage3end_substring_high
+
+    * substring.c: In Substring_new, added a minimum of 8 bp in general test 2
+
+    * stage3hr.c: For Stage3end_substring_low and Stage3end_substring_high, if
+      the hardclip passes all substrings, return the last one processed
+
+    * stage3hr.c: In Stage3end_new_substrings, handling the case where
+      substring1 and substringN have different chrnums assigned by Substring_new
+
+    * stage3hr.c: In Stage3end_new_substitution, handling the case where
+      Substring_new has assigned a different chrnum than the one given
+
+    * substring.c: Made trim incremental when performed as a preliminary step
+      before finding novel splice ends
+
+    * kmer-search.c: Made a more rigorous check that both ends are on the same
+      chromosome
+
+    * kmer-search.c: Making sure the two ends are on the same chromosome
+
+    * substring.c: Handling the case where the alignment goes over the upper
+      bound of the next higher chromosome
+
+    * substring.c: Removed residue from an SVN merge conflict
+
+    * Makefile.gsnaptoo.am, distant-rna.c, gsnap.c, path-solve.c, src,
+      stage1hr.c, stage3hr.c, stage3hr.h, terminal.c, terminal.h, trunk,
+      types.h: Merged revisions 213033 through 213070 from
+      branches/2018-02-05-add-terminals to add terminal alignments
+
+    * substring.c: Not applying general test 1 if orig_nmismatches < 0
+
+    * substring.c: Making outofbounds adjustments to be increments to the
+      existing trim_left and trim_right
+
+    * cigar.c, distant-rna.c, gsnap.c, kmer-search.c, samprint.c, splice.c, src,
+      stage1hr.c, stage3hr.c, stage3hr.h, substring.c, substring.h, trunk:
+      Merged revisions 213027 through 213053 from
+      branches/2018-02-04-simplify-substring-new to simplify Substring_new
+      procedure
+
+    * config.site.rescomp.prd, config.site.rescomp.tst, index.html, svncl.pl:
+      Updated version number
+
+    * stage3.c: Added a List_reverse command
+
+    * types.h: Added comments
+
+    * stage1hr.c: Fixed wrong assignment of first_read_p
+
+    * stage3hr.c: Checking for the eventrim region having bounds that don't make
+      sense
+
+    * substring.c: Added an assertion
+
+    * stage3.c: Restored pass 7 to remove dual breaks at the ends
+
+    * stage1hr.c: Fixed a memory leak for queryrc
+
+    * filestring.c: Added a check for fp being NULL, which can occur with the
+      flags --omit-concordant-uniq or --omit-concordant-mult
+
+2018-02-02  twu
+
+    * VERSION: Updated version number
+
+    * configure.ac: Fixed command for --enable-lib
+
+    * block.c, block.h, extension-search.c, extension-search.h, filestring.c,
+      gmap.c, gsnap.c, index.html, indexdb.c, indexdb.h, intersect.c,
+      kmer-search.c, kmer-search.h, oligoindex_hr.c, pair.c, pair.h,
+      path-solve.c, segment-search.c, src, stage1.c, stage1.h, stage1hr.c,
+      stage1hr.h, stage2.c, stage3.c, stage3.h, stage3hr.c, substring.c,
+      transcriptome.c, trunk: Merged revisions 212952 through 212996 from
+      branches/2018-02-01-reverse-gmap-slowdown
+
+    * gmap.c: Fixed reference to array0 that should have been array1
+
+    * configure.ac: Disabled alloca by default
+
+2018-02-01  twu
+
+    * stage3.c: Restored computation of Stage2_compute_starts and
+      Stage2_compute_ends
+
+    * pair.c: Handling GFF3 output when chrstring or accession is NULL
+
+    * oligoindex_localdb.c, oligoindex_localdb.h: Not using
+      Oligoindex_localdb_tally because of speed
+
+    * inbuffer.c: Added debugging statements
+
+    * gmap.c: Restored function of --pairalign option
+
+2018-01-30  twu
+
+    * stage3hr.c: Made changes to debugging statements
+
+    * stage3.h: Updated interface for Stage3_setup
+
+    * stage3.c: Distinguishing between overall_end_distance_linear and
+      overall_end_distance_circular when making call to Stage2_compute_starts
+      and Stage2_compute_ends
+
+    * stage1hr.c: Making calls to remove_circular_alias and remove_duplicates
+      for single-end reads
+
+    * samprint.c: Fixed bug where alignment results to circular chromosome were
+      not being printed
+
+    * path-solve.c: Not allowing splice junctions for circular chromosomes
+
+    * kmer-search.c: Not allowing splice junctions for circular chromosomes
+
+    * gsnap.c: Using new interface to Stage3_setup
+
+    * gmap.c: Using genomelength instead of genome_totallength.  Initializing
+      circularp for usersegment
+
+    * gsnap.c, kmer-search.c, kmer-search.h, path-solve.c, path-solve.h: Not
+      allowing Junction_new_splice to be called when splicing is turned off
+
+    * trindex.c: Added some headers for open() procedure
+
+    * gmap.c, gsnap.c: Using interfaces to new setup procedures
+
+    * stage3.c, stage3.h: Limiting chrend for Stage2_compute_starts and
+      Stage2_compute_ends based on genome total length
+
+    * stage1hr.c, stage1hr.h: Added hook for distant DNA alignments
+
+    * samprint.c, samprint.h: Not printing XS field in SAM output when splicing
+      is turned off.
+
+    * pair.c, pair.h: Printing . in features field of GFF3 output when sense is
+      unknown. Not printing XS field in SAM output when splicing is turned off.
+
+2018-01-29  twu
+
+    * stage3.c: Before running Stage2_compute_starts and Stage2_compute_ends,
+      removed the check on chrend going past chrhigh.  The truncated coordinates
+      can cause chrend to be less than chrstart if the alignment straddles a
+      chromosomal bound
+
+    * stage3hr.c: Fixed a memory free error.  In Stage3end_new_substrings, when
+      Stage3end_T object fails due to circular alias, letting Stage3end_free and
+      not the caller free the junctions
+
+    * stage3.c: Accepting a single alignment regardless of final score, if the
+      original queryjump or genomejump is negative
+
+2018-01-27  twu
+
+    * substring.c: In Substring_new, computing trims first, then adjusting trims
+      for out-of-bound lengths, then adjusting query and genomic bounds
+
+2018-01-26  twu
+
+    * VERSION: Updated version number
+
+    * intlist.c, intlist.h, list.c, list.h, uintlist.c, uintlist.h: Added code
+      for non-inlined functions
+
+    * transcript.c, transcript.h: Made Transcript_num non-inline
+
+    * configure.ac: Added macro AC_C_INLINE
+
+    * get-genome.c: Added include for intlist.h
+
+2018-01-25  twu
+
+    * archive.html, index.html: Updated for latest version
+
+    * path-solve.c: Added a type conversion
+
+    * Makefile.gsnaptoo.am: Fixed typos in file names
+
+    * VERSION: Updated version number
+
+    * gsnap.c: Change in description for --action-if-cigar-error
+
+    * gmap.c: Implemented --action-if-cigar-error flag
+
+    * stage3.c: Putting an upper bound on the number of Boyer-Moore searches for
+      microexons, based on the number of 5' and 3' splice positions
+
+    * translation.c: Assigning aaphase_e for indels in CDS
+
+    * pair.c, pair.h: Implemented flag --action-if-cigar-error
+
+    * oligoindex_hr.h: Moved SIMD includes to header file, since definition of
+      Oligoindex_T object is now there
+
+    * oligoindex_hr.c: Moved SIMD includes to header file, since definition of
+      Oligoindex_T object is now there
+
+    * gsnap.c: Implemented flag --action-if-cigar-error
+
+    * gmap.c: Using new interface to Pair_setup
+
+    * Makefile.gsnaptoo.am: Turned off making of gmapl and gsnapl for all SIMD
+      types
+
+    * pair.c: Implemented patch by Nathan Weeks to remove extra token from
+      print_gff3_exons_forward.  For GFF3 code, changed types for genomic
+      coordinates from int to Chrpos_T.  Also initializing genomic coordinates
+      to 0 instead of -1
+
+    * oligoindex_hr.c: Added assertion
+
+    * stage3.c: Fixed computation of chrstart and chrend before final
+      Stage2_compute_start and Stage2_compute_ends for circular chromosomes
+
+    * Makefile.gsnaptoo.am, gmap.c, gsnap.c, localdb.c, localdb.h, merge.c,
+      oligoindex_hr.c, oligoindex_hr.h, oligoindex_localdb.c,
+      oligoindex_localdb.h, src, stage2.c, stage2.h, stage3.c, trunk: Merged
+      revisions 212704 through 212742 from branches/2018-01-23-gmap-localdb
+
+    * VERSION, config.site.rescomp.tst: Updated version
+
+    * configure.ac: Added the conditional MAKE_LIB and a flag to control it
+
+    * pair.c: Added an assertion about cds_phase being non-negative
+
+    * Makefile.gsnaptoo.am: Building library only if MAKE_LIB is true
+
+2018-01-23  twu
+
+    * gsnap.c: Allow --terminal-threshold flag for backward compatibility, but
+      ignore
+
+2018-01-22  twu
+
+    * gsnap.c: Allowing the --use-sarray flag and ignoring it
+
+    * stage1hr.c: Fixed bugs regarding the use of querylength3 instead of
+      querylength5, and for handling the case where all paired-end hits are NULL
+      and we need to run GMAP on the complete paths.
+
+2018-01-21  twu
+
+    * setup1.test.in: Turning off setup1.test
+
+2018-01-20  twu
+
+    * Makefile.gsnaptoo.am, access.c, atoiindex.c, bitpack64-access.h,
+      bitpack64-incr.c, bitpack64-incr.h, bitpack64-read.c, bitpack64-read.h,
+      bitpack64-readtwo.c, bitpack64-write.c, bitpack64-write.h, block.c,
+      cellpool.c, cmetindex.c, diagpool.c, distant-rna.c, distant-rna.h,
+      dynprog.c, dynprog_cdna.c, dynprog_end.c, dynprog_genome.c,
+      dynprog_single.c, epu16-bitpack64-access.c, epu16-bitpack64-access.h,
+      epu16-bitpack64-incr.c, epu16-bitpack64-incr.h, epu16-bitpack64-read.c,
+      epu16-bitpack64-read.h, epu16-bitpack64-readtwo.c,
+      epu16-bitpack64-readtwo.h, epu16-bitpack64-write.c,
+      epu16-bitpack64-write.h, extension-search.c, extension-search.h,
+      filesuffix.h, genome.c, genome128_consec.c, genome128_hr.c,
+      genome128_hr.h, genome_sites.c, genome_sites.h, gmap.c, gmap_build.pl.in,
+      gmapindex.c, gregion.c, gregion.h, gsnap.c, indel.c, indel.h,
+      indexdb-write.c, indexdb-write.h, indexdb.c, indexdb.h, indexdb_hr.c,
+      indexdbdef.h, intersect.c, intersect.h, intlist.c, intlist.h, junction.c,
+      kmer-search.c, kmer-search.h, list.c, list.h, littleendian.h,
+      localdb-write.c, localdb-write.h, localdb.c, localdb.h, localdbdef.h,
+      matchpool.c, maxent_hr.c, merge.c, merge.h, oligo.c, oligo.h,
+      oligoindex.c, output.c, pair.c, pair.h, pairpool.c, path-solve.c,
+      path-solve.h, reader.c, reader.h, resulthr.c, resulthr.h, samprint.c,
+      sarray-read.c, sarray-search.c, sarray-search.h, sedgesort.c, sedgesort.h,
+      segment-search.c, segment-search.h, smooth.c, snpindex.c, splice.c,
+      splice.h, splicestringpool.c, splicetrie.c, splicetrie.h, src, stage1.c,
+      stage1hr.c, stage1hr.h, stage2.c, stage3.c, stage3.h, stage3hr.c,
+      stage3hr.h, substring.c, substring.h, transcript.c, transcript.h, trunk,
+      types.h, uintlist.c, uintlist.h, uniqscan.c, univdiag.c, univdiag.h,
+      univdiagdef.h, util: Merged revisions 210996 to 212658 from
+      branches/2017-11-04-faster-transcriptome.  Put previous version into
+      tags/2018-01-19-version1-pre-transcriptome
+
+    * index.html: Revised for new version
+
+    * uniqscan.c: Using new interface to Stage1hr_setup
+
+    * transcript.c: Sorting transcripts before printing
+
+    * stage3hr.c, stage3hr.h: Stage3end_new_transcript aborting of number of
+      substrings and junctions don't match.  Stage3end_new_substrings returning
+      new junctions to caller
+
+    * stage1hr.h: Using transcriptome_end_accept
+
+    * stage1hr.c: Using transcriptome_end_accept
+
+    * sarray-search.c: Various improvements to algorithm
+
+2018-01-17  twu
+
+    * pair.c: Skipping gap characters in Pairarray_genomic_sequence
+
+    * gsnap.c: Fixed default value of max_middle_insertions
+
+2017-11-17  twu
+
+    * sarray-search.c: Turned off qsort in favor of sedgesort
+
 2017-11-16  twu
-    * VERSION, public-2017-09-05, src: Updated version number
-    * stage3.c: Merged revision 211227 from trunk to fix issue with trimming of
-      chimeras not being re-extended back to the breakpoint, and to fix issue
-      with CIGAR strings from shortgaps comps being treated differently from
-      indel comps.
-    * stage1hr.c: Appending greedy results to subs
-    * dynprog_end.c: Merged revision 211220 from trunk to handle the case where
-      rev_goffset is negative
-    * outbuffer.c: Merged revision 211219 from trunk to fix fatal bug when
-      trying to write SAM headers to OUTPUT_NONE split output, which is now NULL
+    * stopwatch.c: Added standalone code for testing
+    * stage3.c: Fixed issue with trimming of chimeras not being re-extended back
+      to the breakpoint.  Fixed issue with CIGAR strings from shortgap comps
+      being treated differently from indel comps.
+    * gsnap.c: Using new interface to Sarray_search.  Printing worker runtimes
+      to stderr.  Using --transcriptome-accept flag instead of old flags
+    * sarray-search.c, sarray-search.h: Removed genes_iit as a parameter
+    * indel.c: Added debugging statements
+
+    * dynprog_end.c: Handling the case where rev_goffset is negative
+
+    * outbuffer.c: Fixed fatal bug when trying to write SAM headers to
+      OUTPUT_NONE split output, which is now NULL
 2017-10-30  twu
-    * stage3.c: Merged revision 210882 from trunk to fix trim_end_indel
-      procedures to stop at a gap
+    * stage3.c: Fixed trim_end_indel procedures to stop at a gap
 2017-10-29  twu
-    * VERSION, public-2017-09-05, src: Updated version number
-    * substring.c: Merged revision 210861 from trunk to fix a memory leak when
-      embellish_genomic is called a second time on a Stage3end_T object
-    * stage3hr.c: Merged revision 210860 from trunk to set hardclips to 0 for
-      creating CIGAR for a GMAP alignment
-    * stage3.c: Merged revision 210859 from trunk to disable forcep in
-      traverse_single_gap
-    * pairpool.c: Merged revision 210857 from trunk to fix Pairpool_join_end5
-      and Pairpool_join_end3.  Not attaching the end if revision process still
-      results in a negative queryjump or genomejump.
-    * get-genome.c: Merged revision 210856 from trunk to look for map file in
-      genomesubdir as well as mapdir
-    * cigar.c, cigar.h, gsnap.c: Merged revision 210855 from trunk to call
-      Cigar_setup
-    * gmap.c: Merged revision 210647 from trunk to change option name from
-      --alt-initiation-codons to --alt-start-codons
-    * sarray-search.c: Merged part of revision 210755 from trunk.  Ensuring that
-      we are using the smallest indel.
+    * substring.c: Fixed a memory leak when embellish_genomic is called a second
+      time on a Stage3end_T object
+
+    * stage3hr.c: In Stage3end_new_gmap, in computing CIGAR, starting with
+      hardclips set to zero
+    * stage3.c: In traverse_single_gap, not allowing a force of the gappairs.
+      Implemented contiguous versions of peel_leftward and peel_rightward, but
+      not using yet.
+    * sarray-search.c: For transcriptome endpoints, restricting further the
+      cases where an indel should be ignored.  Added some potential code for
+      genome endpoints, but not implemented
+
+    * pairpool.c: In Pairpool_join_end5 and Pairpool_join_end3, if the final
+      revision process still results in a negative queryjump or genomejump, then
+      don't attach the end at all
+
+    * get-genome.c: If map file is not in mapdir, then look in genomesubdir
+
+    * cigar.c, cigar.h, gsnap.c: Calling Cigar_setup to initialize static
+      variables
+2017-10-20  twu
+    * trindex.c: Added error message if genes file not provided by user
+    * stage1hr.c, stage3hr.c, stage3hr.h: Doing remapping back to transcriptome
+      during pairing procedure
+    * get-genome.c: Added comment
+    * pair.c, pair.h, samprint.c: For GMAP-based alignments in GSNAP, printing
+      transcript info
+    * sarray-read.c, sarray-read.h: Changed relevant types from Univcoord_T to
+      Sarrayptr_T
+
+    * sarray-search.c, sarray-search.h: Changed relevant types from Univcoord_T
+      to Sarrayptr_T.  Choosing closest indel on left and right ends.  For
+      transcriptome, checking for overlapping or adjacent indels
 2017-10-18  twu
-    * sarray-search.c: Merged revision 210665 from trunk.  Fixed an issue where
-      querystart was -1, which can result when nmatches is 0
-2017-10-14  twu
-    * VERSION, src: Updated version number
-    * stage3hr.c: Merged revision 210536 from trunk.  Removed use of minimum of
-      npaths and maxpaths
-    * samprint.c: Merged revision 210535 from trunk.  Handling dinucleotides
-      gracefully if genomic sequence is NULL.
-    * gsnap.c: Merged revision 210534 from trunk.  Removed --maxsearch option.
-      Increased default expected_pairlength value.
+    * config.site.rescomp.prd, config.site.rescomp.tst, index.html: Updated for
+      latest version
+
+    * transcriptome.c, transcriptome.h: Implemented
+      Transcriptome_genomic_bounded_p
+
+    * transcript.c, transcript.h: Implemented Transcript_in_list_p.  Returning
+      min_insertlength from Transcript_intersect_p
+
+    * stage3hr.c: Fixed bug in retrieving parent hit in
+      Stage3end_remove_overlaps
+
+    * sarray-search.c: Fixed an issue where querystart was -1, which can result
+      when nmatches is 0
+
+2017-10-17  twu
+    * gmap.c: Changed option name from --alt-initiation-codons to
+      --alt-start-codons
+    * stage1hr.c: Fixed typo in variable name
+
+2017-10-13  twu
+    * stage3hr.c: Removed optimization to use the minimum of npaths and
+      maxpaths, since we need computations on all npaths to make a random
+      selection
+    * samprint.c: Handling dinucleotides gracefully if genomic sequence is NULL
+    * gsnap.c: Increased default expected_pairlength from 200 to 500.  Removed
+      --maxsearch option, since it can lead to poor answers.
 2017-10-12  twu
-    * outbuffer.c, samflags.h, samheader.c: Merged revision 210516 from trunk. 
-      Fixed problem introduced in 2017-05 which caused the --output-file option
-      to produce a NULL file pointer.
-    * chrom.c: Merged revision 210513 from trunk.  Dereferencing a character
-      pointer.
-    * pair.c: Merged revision 210506 from trunk.  Fixed problem with
-      initialization of endi in determining gff3 coordinates.  Avoiding reading
-      the pair at index of npairs.
-    * cellpool.c, cellpool.h, stage2.c: Merged revision 210504 from trunk. 
-      Adding non-overlapping paths, as well as high-scoring paths
-    * gmap.c, translation.c, translation.h: Merged revision 210493 from trunk. 
-      Adding option for only ATG as initiation codon.  Making this the default.
-    * stage3.c: Merged revision 210492 from trunk.  For trimming end exons,
-      using a percentage of the querylength instead of a fixed length
-    * stage3hr.c: Merged revision 210278 from trunk.  In
-      resolve_ambiguous_splice procedure, when ambiguity is resolved, setting
-      genomicstart or genomicend fields accordingly.
-2017-09-29  twu
-    * VERSION, public-2017-09-05, src: Updated version number
-    * stage3.c: Merged revision 210194 from trunk to move build_dual_breaks step
+    * outbuffer.c, samflags.h, samheader.c: Fixed problem introduced in 2017-05
+      which caused the --output-file option to produce a NULL file pointer
+    * chrom.c: Deferencing a character pointer
+    * pair.c: Fixed problem with initialization of endi in determining gff3
+      coordinates.  Avoiding reading the pair at index of npairs.
+    * cellpool.c, cellpool.h, stage2.c: Adding non-overlapping paths, as well as
+      high-scoring paths
+    * stage1hr.c: Using new interface to Sarray_search_transcriptome
+    * substring.c, substring.h: Added functions for Substring_chrpos_low and
+      Substring_chrpos_high
+    * stage3hr.h: Added interfaces for chrpos_low and chrpos_high
+    * stage3hr.c: In transferring Transcript_T objects, checking for duplicates
+
+    * sarray-search.c, sarray-search.h: Allowing for genomic bounds on
+      transcriptome alignment, for use in re-aligning genomic hit to find
+      transcriptome coordinates
+
+    * gmap.c, translation.c, translation.h: Adding option for only ATG as
+      initiation codon.  Making this the default.
+
+    * stage3.c: For trimming end exons, using a percentage of the querylength
+      instead of a fixed length
+
+2017-10-03  twu
+
+    * stage3hr.c: Removed old version of pair_up_concordant_transcriptome
+
+    * stage3hr.c: In resolve_ambiguous_splice procedures, when ambiguity is
+      resolved, setting genomicstart or genomicend fields accordingly
+
+    * stage3hr.h: Removed interface for Stage3pair_pairtype
+
+    * stage3hr.c: Fixed the order for constructing genomic sequence from
+      substrings
+
+    * substring.h: Not applying general test of goodness for substrings from
+      transcriptome-guided alignment
+
+    * substring.c: Not applying general test of goodness for substrings from
+      transcriptome-guided alignment
+    * sarray-search.c: Modified debugging statements
+    * samprint.c: Determining pairtype again, and allowing for the possibility
+      of concordant uniq
+
+    * trindex.c: Copying the genes IIT file to the transcriptome directory
+
+    * transcriptome.c: Allowing a transcriptome to be read without a genome, by
+      using divints instead of chrnums
+
+2017-10-02  twu
+
+    * transcript.c: Revised debugging statements
+
+    * gsnap.c, uniqscan.c: Using new interface to Stage1hr_setup
+
+    * stage3hr.c, stage3hr.h: Implemented a separate procedure for pairing up
+      transcriptome alignments.  Transcriptome hit types now take precedence. 
+      Implemented Stage3end_substrings_genomic_sequence.
+
+    * stage1hr.c, stage1hr.h: Using new interface to Stage3pair_new.  Re-mapping
+      to transcriptome from genomic suffix array alignment.
+
+    * junction.c, junction.h: Added function Junction_deletionpos
+
+2017-09-29  twu
+
+    * stage3.c: Moved location of build_dual_breaks step to get better behavior
 2017-09-27  twu
-    * stage3.c: Merged revision 210170 from trunk to stop over-aggressive use of
-      maxintronlen_end from wrong side of the sequence
-    * chimera.c: Merged revision 210169 from trunk to initialize a variable
+    * stage3.c: Fixed over-aggressive use of minintronlen_ends from wrong end of
+      sequence
+    * chimera.c: Initializing a variable
 2017-09-23  twu
-    * VERSION, public-2017-09-05, src, uinttableuint.c, uinttableuint.h: Merged
-      revision 210071 from trunk to add files
-    * gsnap.c, pair.c, pair.h, public-2017-09-05, src, uniqscan.c: Merged
-      revision 210062 from trunk to fix computation of cds bounds for GFF3
-      output
-    * translation.c: Merged revision 210057 from trunk to assign aaphase_g for
-      final genomic codon
-    * gmap.c: Merged revision 210059 from trunk to restore MAX_CHIMERA_ITER to
-      3, but not iterating multiple times for middle pieces.  Added option
-      --gff3-cds
-    * Makefile.gsnaptoo.am: Merged revision 210058 from trunk to add
-      uinttableuint to library
+    * uinttableuint.c, uinttableuint.h: Initial import
+    * stage1hr.c, stage3hr.c, stage3hr.h: For transcriptome-guided genomic
+      alignment, placing results into concordant_uniq instead of
+      paired_uniq_long, if a transcript matches both ends
+
+    * pair.c, pair.h: Fixed computation of cds bounds for GFF3 output
+    * gsnap.c, uniqscan.c: Using new interface to Pair_setup
+    * gmap.c: Restored MAX_CHIMERA_ITER to be 3, but not iterating multiple
+      times for middle pieces.  Added option --gff3-cds
+    * Makefile.gsnaptoo.am: Added uinttableuint.c to library
+
+    * translation.c: Assigning aaphase_g for final genomic codon
 2017-09-11  twu
-    * VERSION: Updated version number
-    * gmap.c, src: Merged revision 209800 from trunk to restore --intronlength
-      option
-    * pair.c: Merged revisions 209797 and 209798 from trunk to replace gff3
-      printing code for CDS with a call to the code for exons
-    * gmap.c, src: Merged revision 209788 from trunk to no longer iterate on
-      check_middle_local
+    * gmap.c: Restored --intronlength option
+    * pair.c: Fixed gff3 cds output so it ignores indels
+
+    * pair.c: Replaced gff3 printing code for CDS with a call to code for exons
+
+    * pair.c, pair.h: Using new interface to transcript print functions
+
+    * stage3hr.c, stage3hr.h, transcript.c, transcript.h: Replacing separate
+      trnums, trstarts, and trends fields with transcripts field
+
+    * stage1hr.c: No longer filtering initially by transcript concordance
+    * gsnap.c, samprint.c, samprint.h: Using new interface to transcripts field
+      for Stage3end_T and Stage3pair_T objects
+    * gmap.c: No longer iterating on check_middle_local
+
+    * Makefile.gsnaptoo.am: Added transcript.c and transcript.h to programs that
+      need it from pair.c
+
+2017-09-06  twu
+
+    * sarray-search.c, stage3hr.c: Commented out or fixed code for LARGE_GENOMES
+      to use Uint8list_T
+
+    * sarray-search.c, sarray-search.h: Commented out genome code for
+      LARGE_GENOMES
+
+    * trindex.c: Fixed code so it will compile.  Fixed memory leaks.
 2017-09-05  twu
-    * public-2017-09-05: Created public release
+    * Makefile.gsnaptoo.am, VERSION, bitpack64-readtwo.c,
+      config.site.rescomp.prd, config.site.rescomp.tst, genome128_consec.c,
+      genome128_consec.h, genome128_hr.c, genome128_hr.h, get-genome.c, gmap.c,
+      gmapindex.c, gsl.m4, gsnap.c, iit-read-univ.c, iit-read-univ.h,
+      iit-read.c, iit-read.h, indel.c, indel.h, index.html, intlist.c,
+      intlist.h, junction.c, junction.h, mapq.c, mapq.h, output.c, pair.c,
+      pair.h, pairpool.c, samprint.c, samprint.h, sarray-read.c, sarray-read.h,
+      sarray-search.c, sarray-search.h, sarray-write.c, sarray-write.h,
+      splicealt.c, src, stage1hr.c, stage1hr.h, stage3hr.c, stage3hr.h,
+      substring.c, substring.h, transcriptome.c, transcriptome.h, trindex.c,
+      trunk, uniqscan.c: Merged revisions 207858 to 209656 from
+      branches/2017-07-01-transcripts to allow for transcriptome-guided genomic
+      alignment
     * sarray-search.c: Fixed bug resulting from check of common diagonal over
       circular origin

Makefile.in

@@ -184,7 +184,7 @@ am__DIST_COMMON = $(srcdir)/Makefile.in $(top_srcdir)/config/compile \
 	$(top_srcdir)/config/config.sub \
 	$(top_srcdir)/config/install-sh $(top_srcdir)/config/ltmain.sh \
 	$(top_srcdir)/config/missing AUTHORS COPYING ChangeLog INSTALL \
-	NEWS README config/compile config/config.guess \
+	NEWS README TODO config/compile config/config.guess \
 	config/config.sub config/install-sh config/ltmain.sh \
 	config/missing
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
@@ -259,6 +259,7 @@ EXEEXT = @EXEEXT@
 FGREP = @FGREP@
 GMAPDB = @GMAPDB@
 GREP = @GREP@
+HAVE_INLINE = @HAVE_INLINE@
 INSTALL = @INSTALL@
 INSTALL_DATA = @INSTALL_DATA@
 INSTALL_PROGRAM = @INSTALL_PROGRAM@

TODO

+
+Add flag that allows for splitting afterwards.
+

VERSION

-2017-11-15
\ No newline at end of file
+2018-07-04
\ No newline at end of file

configure

 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.69 for gmap 2017-11-15.
+# Generated by GNU Autoconf 2.69 for gmap 2018-07-04.
 #
 # Report bugs to <Thomas Wu <twu@gene.com>>.
 #
@@ -590,8 +590,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='gmap'
 PACKAGE_TARNAME='gmap'
-PACKAGE_VERSION='2017-11-15'
-PACKAGE_STRING='gmap 2017-11-15'
+PACKAGE_VERSION='2018-07-04'
+PACKAGE_STRING='gmap 2018-07-04'
 PACKAGE_BUGREPORT='Thomas Wu <twu@gene.com>'
 PACKAGE_URL=''
@@ -660,6 +660,7 @@ SIMD_SSSE3_CFLAGS
 SIMD_SSE2_CFLAGS
 POPCNT_CFLAGS
 ALLOCA
+HAVE_INLINE
 PTHREAD_CFLAGS
 PTHREAD_LIBS
 PTHREAD_CC
@@ -689,6 +690,8 @@ SED
 LIBTOOL
 PERL
 BINDIR
+MAKE_LIB_FALSE
+MAKE_LIB_TRUE
 MAINTAINER_FALSE
 MAINTAINER_TRUE
 FULLDIST_FALSE
@@ -800,6 +803,7 @@ enable_silent_rules
 enable_maintainer_mode
 enable_fulldist
 enable_maintainer
+enable_lib
 enable_shared
 enable_static
 with_pic
@@ -1369,7 +1373,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures gmap 2017-11-15 to adapt to many kinds of systems.
+\`configure' configures gmap 2018-07-04 to adapt to many kinds of systems.
 Usage: $0 [OPTION]... [VAR=VALUE]...
@@ -1440,7 +1444,7 @@ fi
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of gmap 2017-11-15:";;
+     short | recursive ) echo "Configuration of gmap 2018-07-04:";;
    esac
   cat <<\_ACEOF
@@ -1460,6 +1464,9 @@ Optional Features:
                           sometimes confusing) to the casual installer
   --enable-fulldist       For use by program maintainer
   --enable-maintainer     For use by program maintainer
+  --enable-lib            Build and install a GMAP library (Warning: will put
+                          headers into include subdirectory)
+                          (default=disable-lib)
   --enable-shared[=PKGS]  build shared libraries [default=yes]
   --enable-static[=PKGS]  build static libraries [default=yes]
   --enable-fast-install[=PKGS]
@@ -1469,8 +1476,9 @@ Optional Features:
                           (default=yes). Note: Use only if you have a parallel
                           I/O file system, like Lustre
   --enable-pthreads       Enable pthreads (default=yes)
-  --enable-alloca         Enable alloca (default=yes) for stack-based memory
-                          allocation.
+  --enable-alloca         Enable alloca (default=no) for stack-based memory
+                          allocation. Warning: Can cause GMAP to crash on long
+                          sequences.
   --enable-mmap           Enable mmap (default=yes). Recommend that you do not
                           disable this unless absolutely necessary.
   --enable-simd           Enable simd commands in general if they compile and
@@ -1577,7 +1585,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-gmap configure 2017-11-15
+gmap configure 2018-07-04
 generated by GNU Autoconf 2.69
 Copyright (C) 2012 Free Software Foundation, Inc.
@@ -2183,7 +2191,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
-It was created by gmap $as_me 2017-11-15, which was
+It was created by gmap $as_me 2018-07-04, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
   $ $0 $@
@@ -2533,8 +2541,8 @@ ac_compiler_gnu=$ac_cv_c_compiler_gnu
 { $as_echo "$as_me:${as_lineno-$LINENO}: checking package version" >&5
 $as_echo_n "checking package version... " >&6; }
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: 2017-11-15" >&5
-$as_echo "2017-11-15" >&6; }
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: 2018-07-04" >&5
+$as_echo "2018-07-04" >&6; }
 ### Read defaults
@@ -4401,7 +4409,7 @@ fi
 # Define the identity of the package.
  PACKAGE='gmap'
- VERSION='2017-11-15'
+ VERSION='2018-07-04'
 cat >>confdefs.h <<_ACEOF
@@ -4679,6 +4687,22 @@ else
 fi
+ if test "x$enable_library" = xyes; then
+  MAKE_LIB_TRUE=
+  MAKE_LIB_FALSE='#'
+else
+  MAKE_LIB_TRUE='#'
+  MAKE_LIB_FALSE=
+fi
+
+# Check whether --enable-lib was given.
+if test "${enable_lib+set}" = set; then :
+  enableval=$enable_lib; enable_library="yes"
+else
+  enable_library="no"
+fi
+
+
 # For Perl scripts, because otherwise configure inserts
 # @bindir@ => "${exec_prefix}/bin";
 { $as_echo "$as_me:${as_lineno-$LINENO}: checking bindir" >&5
@@ -14502,6 +14526,55 @@ done
 # Checks for typedefs, structures, and compiler characteristics.
+{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for inline" >&5
+$as_echo_n "checking for inline... " >&6; }
+if ${ac_cv_c_inline+:} false; then :
+  $as_echo_n "(cached) " >&6
+else
+  ac_cv_c_inline=no
+for ac_kw in inline __inline__ __inline; do
+  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+/* end confdefs.h.  */
+#ifndef __cplusplus
+typedef int foo_t;
+static $ac_kw foo_t static_foo () {return 0; }
+$ac_kw foo_t foo () {return 0; }
+#endif
+
+_ACEOF
+if ac_fn_c_try_compile "$LINENO"; then :
+  ac_cv_c_inline=$ac_kw
+fi
+rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
+  test "$ac_cv_c_inline" != no && break
+done
+
+fi
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_c_inline" >&5
+$as_echo "$ac_cv_c_inline" >&6; }
+
+case $ac_cv_c_inline in
+  inline | yes) ;;
+  *)
+    case $ac_cv_c_inline in
+      no) ac_val=;;
+      *) ac_val=$ac_cv_c_inline;;
+    esac
+    cat >>confdefs.h <<_ACEOF
+#ifndef __cplusplus
+#define inline $ac_val
+#endif
+_ACEOF
+    ;;
+esac
+
+if test "$ac_cv_c_line" != no; then
+
+$as_echo "#define HAVE_INLINE 1" >>confdefs.h
+
+
+fi
+
  { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether byte ordering is bigendian" >&5
 $as_echo_n "checking whether byte ordering is bigendian... " >&6; }
 if ${ac_cv_c_bigendian+:} false; then :
@@ -15258,194 +15331,8 @@ $as_echo "disabled" >&6; }
      ;;
      x)
-     { $as_echo "$as_me:${as_lineno-$LINENO}: result: not specified so enabled by default" >&5
-$as_echo "not specified so enabled by default" >&6; }
-     # The Ultrix 4.2 mips builtin alloca declared by alloca.h only works
-# for constant arguments.  Useless!
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for working alloca.h" >&5
-$as_echo_n "checking for working alloca.h... " >&6; }
-if ${ac_cv_working_alloca_h+:} false; then :
-  $as_echo_n "(cached) " >&6
-else
-  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
-/* end confdefs.h.  */
-#include <alloca.h>
-int
-main ()
-{
-char *p = (char *) alloca (2 * sizeof (int));
-			  if (p) return 0;
-  ;
-  return 0;
-}
-_ACEOF
-if ac_fn_c_try_link "$LINENO"; then :
-  ac_cv_working_alloca_h=yes
-else
-  ac_cv_working_alloca_h=no
-fi
-rm -f core conftest.err conftest.$ac_objext \
-    conftest$ac_exeext conftest.$ac_ext
-fi
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_working_alloca_h" >&5
-$as_echo "$ac_cv_working_alloca_h" >&6; }
-if test $ac_cv_working_alloca_h = yes; then
-
-$as_echo "#define HAVE_ALLOCA_H 1" >>confdefs.h
-
-fi
-
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for alloca" >&5
-$as_echo_n "checking for alloca... " >&6; }
-if ${ac_cv_func_alloca_works+:} false; then :
-  $as_echo_n "(cached) " >&6
-else
-  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
-/* end confdefs.h.  */
-#ifdef __GNUC__
-# define alloca __builtin_alloca
-#else
-# ifdef _MSC_VER
-#  include <malloc.h>
-#  define alloca _alloca
-# else
-#  ifdef HAVE_ALLOCA_H
-#   include <alloca.h>
-#  else
-#   ifdef _AIX
- #pragma alloca
-#   else
-#    ifndef alloca /* predefined by HP cc +Olibcalls */
-void *alloca (size_t);
-#    endif
-#   endif
-#  endif
-# endif
-#endif
-
-int
-main ()
-{
-char *p = (char *) alloca (1);
-				    if (p) return 0;
-  ;
-  return 0;
-}
-_ACEOF
-if ac_fn_c_try_link "$LINENO"; then :
-  ac_cv_func_alloca_works=yes
-else
-  ac_cv_func_alloca_works=no
-fi
-rm -f core conftest.err conftest.$ac_objext \
-    conftest$ac_exeext conftest.$ac_ext
-fi
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_func_alloca_works" >&5
-$as_echo "$ac_cv_func_alloca_works" >&6; }
-
-if test $ac_cv_func_alloca_works = yes; then
-
-$as_echo "#define HAVE_ALLOCA 1" >>confdefs.h
-
-else
-  # The SVR3 libPW and SVR4 libucb both contain incompatible functions
-# that cause trouble.  Some versions do not even contain alloca or
-# contain a buggy version.  If you still want to use their alloca,
-# use ar to extract alloca.o from them instead of compiling alloca.c.
-
-ALLOCA=\${LIBOBJDIR}alloca.$ac_objext
-
-$as_echo "#define C_ALLOCA 1" >>confdefs.h
-
-
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether \`alloca.c' needs Cray hooks" >&5
-$as_echo_n "checking whether \`alloca.c' needs Cray hooks... " >&6; }
-if ${ac_cv_os_cray+:} false; then :
-  $as_echo_n "(cached) " >&6
-else
-  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
-/* end confdefs.h.  */
-#if defined CRAY && ! defined CRAY2
-webecray
-#else
-wenotbecray
-#endif
-
-_ACEOF
-if (eval "$ac_cpp conftest.$ac_ext") 2>&5 |
-  $EGREP "webecray" >/dev/null 2>&1; then :
-  ac_cv_os_cray=yes
-else
-  ac_cv_os_cray=no
-fi
-rm -f conftest*
-
-fi
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_os_cray" >&5
-$as_echo "$ac_cv_os_cray" >&6; }
-if test $ac_cv_os_cray = yes; then
-  for ac_func in _getb67 GETB67 getb67; do
-    as_ac_var=`$as_echo "ac_cv_func_$ac_func" | $as_tr_sh`
-ac_fn_c_check_func "$LINENO" "$ac_func" "$as_ac_var"
-if eval test \"x\$"$as_ac_var"\" = x"yes"; then :
-
-cat >>confdefs.h <<_ACEOF
-#define CRAY_STACKSEG_END $ac_func
-_ACEOF
-
-    break
-fi
-
-  done
-fi
-
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking stack direction for C alloca" >&5
-$as_echo_n "checking stack direction for C alloca... " >&6; }
-if ${ac_cv_c_stack_direction+:} false; then :
-  $as_echo_n "(cached) " >&6
-else
-  if test "$cross_compiling" = yes; then :
-  ac_cv_c_stack_direction=0
-else
-  cat confdefs.h - <<_ACEOF >conftest.$ac_ext
-/* end confdefs.h.  */
-$ac_includes_default
-int
-find_stack_direction (int *addr, int depth)
-{
-  int dir, dummy = 0;
-  if (! addr)
-    addr = &dummy;
-  *addr = addr < &dummy ? 1 : addr == &dummy ? 0 : -1;
-  dir = depth ? find_stack_direction (addr, depth - 1) : 0;
-  return dir + dummy;
-}
-
-int
-main (int argc, char **argv)
-{
-  return find_stack_direction (0, argc + !argv + 20) < 0;
-}
-_ACEOF
-if ac_fn_c_try_run "$LINENO"; then :
-  ac_cv_c_stack_direction=1
-else
-  ac_cv_c_stack_direction=-1
-fi
-rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext \
-  conftest.$ac_objext conftest.beam conftest.$ac_ext
-fi
-
-fi
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_c_stack_direction" >&5
-$as_echo "$ac_cv_c_stack_direction" >&6; }
-cat >>confdefs.h <<_ACEOF
-#define STACK_DIRECTION $ac_cv_c_stack_direction
-_ACEOF
-
-
-fi
-
+     { $as_echo "$as_me:${as_lineno-$LINENO}: result: not specified so disabled by default" >&5
+$as_echo "not specified so disabled by default" >&6; }
      ;;
 esac
@@ -19400,6 +19287,8 @@ ac_config_files="$ac_config_files util/gvf_iit.pl"
 ac_config_files="$ac_config_files util/vcf_iit.pl"
+ac_config_files="$ac_config_files util/sam-exons.pl"
+
 ac_config_files="$ac_config_files tests/Makefile"
 ac_config_files="$ac_config_files tests/align.test"
@@ -19556,6 +19445,10 @@ if test -z "${MAINTAINER_TRUE}" && test -z "${MAINTAINER_FALSE}"; then
   as_fn_error $? "conditional \"MAINTAINER\" was never defined.
 Usually this means the macro was only invoked conditionally." "$LINENO" 5
 fi
+if test -z "${MAKE_LIB_TRUE}" && test -z "${MAKE_LIB_FALSE}"; then
+  as_fn_error $? "conditional \"MAKE_LIB\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
 if test -z "${MAKE_AVX512_TRUE}" && test -z "${MAKE_AVX512_FALSE}"; then
   as_fn_error $? "conditional \"MAKE_AVX512\" was never defined.
@@ -19978,7 +19871,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by gmap $as_me 2017-11-15, which was
+This file was extended by gmap $as_me 2018-07-04, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
   CONFIG_FILES    = $CONFIG_FILES
@@ -20044,7 +19937,7 @@ _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
 ac_cs_version="\\
-gmap config.status 2017-11-15
+gmap config.status 2018-07-04
 configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
@@ -20483,6 +20376,7 @@ do
     "util/dbsnp_iit.pl") CONFIG_FILES="$CONFIG_FILES util/dbsnp_iit.pl" ;;
     "util/gvf_iit.pl") CONFIG_FILES="$CONFIG_FILES util/gvf_iit.pl" ;;
     "util/vcf_iit.pl") CONFIG_FILES="$CONFIG_FILES util/vcf_iit.pl" ;;
+    "util/sam-exons.pl") CONFIG_FILES="$CONFIG_FILES util/sam-exons.pl" ;;
     "tests/Makefile") CONFIG_FILES="$CONFIG_FILES tests/Makefile" ;;
     "tests/align.test") CONFIG_FILES="$CONFIG_FILES tests/align.test" ;;
     "tests/coords1.test") CONFIG_FILES="$CONFIG_FILES tests/coords1.test" ;;

configure.ac

@@ -131,6 +131,13 @@ AC_ARG_ENABLE([maintainer],
 	      [enable_maintainer="yes"],
               [enable_maintainer="no"])
 
+AM_CONDITIONAL(MAKE_LIB,test "x$enable_library" = xyes)
+AC_ARG_ENABLE([lib],
+	       AC_HELP_STRING([--enable-lib],
+                              [Build and install a GMAP library (Warning: will put headers into include subdirectory) (default=disable-lib)]),
+	      [enable_library="yes"],
+              [enable_library="no"])
+
 # For Perl scripts, because otherwise configure inserts 
 # @bindir@ => "${exec_prefix}/bin";
 AC_MSG_CHECKING(bindir)
@@ -223,6 +230,12 @@ AC_HEADER_DIRENT
 AC_CHECK_HEADERS([fcntl.h limits.h stddef.h stdlib.h string.h strings.h unistd.h sys/types.h])
 
 # Checks for typedefs, structures, and compiler characteristics.
+AC_C_INLINE
+if test "$ac_cv_c_line" != no; then
+  AC_DEFINE(HAVE_INLINE,1,[Define to 1 if compiler supports extern inline])
+  AC_SUBST(HAVE_INLINE)
+fi
+
 AC_C_BIGENDIAN
 AC_C_CONST
 AC_C_VOLATILE
@@ -247,7 +260,7 @@ AC_FUNC_FSEEKO
 AC_MSG_CHECKING(whether alloca is enabled)
 AC_ARG_ENABLE([alloca],
 	      AC_HELP_STRING([--enable-alloca],
-                             [Enable alloca (default=yes) for stack-based memory allocation.]),
+                             [Enable alloca (default=no) for stack-based memory allocation.  Warning: Can cause GMAP to crash on long sequences.]),
               [answer="$enableval"],
               [answer=""])
 case x"$answer" in
@@ -261,8 +274,7 @@ case x"$answer" in
      ;;
 
      x)
-     AC_MSG_RESULT([not specified so enabled by default])
-     AC_FUNC_ALLOCA
+     AC_MSG_RESULT([not specified so disabled by default])
      ;;
 esac	
 
@@ -586,6 +598,7 @@ AC_CONFIG_FILES([util/gff3_genes.pl])
 AC_CONFIG_FILES([util/dbsnp_iit.pl])
 AC_CONFIG_FILES([util/gvf_iit.pl])
 AC_CONFIG_FILES([util/vcf_iit.pl])
+AC_CONFIG_FILES([util/sam-exons.pl])
 AC_CONFIG_FILES([tests/Makefile])
 AC_CONFIG_FILES([tests/align.test],[chmod +x tests/align.test])
 AC_CONFIG_FILES([tests/coords1.test],[chmod +x tests/coords1.test])

debian/changelog

+gmap (2018-07-04-1) UNRELEASED; urgency=medium
+
+  * Cleanup d/rules
+  * New upstream version
+  * debhelper 11
+  * Point Vcs fields to salsa.debian.org
+  * Standards-Version: 4.1.5
+  * libgmap1-dev Depends: gmap, libgmap1 (instead of alternative)
+    Closes: #903793
+
+ -- Andreas Tille <tille@debian.org>  Tue, 17 Jul 2018 07:24:19 +0200
+
 gmap (2017-11-15-1) unstable; urgency=medium
 
   * New upstream version 2017-11-15

debian/compat

-10
+11

debian/control

@@ -6,19 +6,21 @@ Uploaders: Shaun Jackman <sjackman@debian.org>,
 Section: non-free/science
 XS-Autobuild: no
 Priority: optional
-Build-Depends: debhelper (>= 10),
+Build-Depends: debhelper (>= 11~),
                dh-exec,
                help2man,
                libbz2-dev,
                zlib1g-dev
-Standards-Version: 4.1.2
-Vcs-Browser: https://anonscm.debian.org/cgit/debian-med/gmap.git
-Vcs-Git: https://anonscm.debian.org/git/debian-med/gmap.git
+Standards-Version: 4.1.5
+Vcs-Browser: https://salsa.debian.org/med-team/gmap
+Vcs-Git: https://salsa.debian.org/med-team/gmap.git
 Homepage: http://research-pub.gene.com/gmap
 
 Package: gmap
 Architecture: any
-Depends: ${misc:Depends}, ${perl:Depends}, ${shlibs:Depends}
+Depends: ${misc:Depends},
+         ${perl:Depends},
+         ${shlibs:Depends}
 Description: spliced and SNP-tolerant alignment for mRNA and short reads
  This package contains the programs GMAP and GSNAP as well as
  utilities to manage genome databases in GMAP/GSNAP format.
@@ -34,7 +36,8 @@ Description: spliced and SNP-tolerant alignment for mRNA and short reads
 Package: libgmap1
 Architecture: any
 Section: non-free/libs
-Depends: ${misc:Depends}, ${shlibs:Depends}
+Depends: ${misc:Depends},
+         ${shlibs:Depends}
 Description: library for gmap tool
  This package contains library for gmap tool.
  GMAP (Genomic Mapping and Alignment Program) is a tool for aligning
@@ -49,8 +52,9 @@ Description: library for gmap tool
 Package: libgmap1-dev
 Architecture: any
 Section: non-free/libdevel
-Depends: gmap (= ${binary:Version}) | libgmap1 (= ${binary:Version}),
-         ${misc:Depends},
+Depends: gmap (= ${binary:Version}),
+         libgmap1 (= ${binary:Version}),
+         ${misc:Depends}
 Description: development files for gmap
  This package contains header files for gmap.
  GMAP (Genomic Mapping and Alignment Program) is a tool for aligning

debian/rules

@@ -2,8 +2,7 @@
 
 export DEB_BUILD_MAINT_OPTIONS = hardening=+all
 include /usr/share/dpkg/default.mk
-pkg 	:= gmap
-mandir  := $(CURDIR)/debian/$(pkg)/usr/share/man/man1
+mandir  := $(CURDIR)/debian/$(DEB_SOURCE)/usr/share/man/man1
 utildir := $(CURDIR)/util
 bindir  := $(CURDIR)/src
 
@@ -34,11 +33,11 @@ override_dh_auto_install:
 
 override_dh_install:
 	dh_install
-	mkdir -p debian/$(pkg)/usr/bin
+	mkdir -p debian/$(DEB_SOURCE)/usr/bin
 	for bin in `echo gmap gsnap gmap_build gmapl gsnapl; \
-		cd $(CURDIR)/debian/tmp/usr/lib/$(pkg)/; \
+		cd $(CURDIR)/debian/tmp/usr/lib/$(DEB_SOURCE)/; \
 		ls -1 gmap*.* gsnap*.* `; do \
-		ln -s /usr/lib/gmap/$$bin debian/$(pkg)/usr/bin/$$bin; \
+		ln -s /usr/lib/gmap/$$bin debian/$(DEB_SOURCE)/usr/bin/$$bin; \
 	done
 
 override_dh_auto_test:

src/access.c

-static char rcsid[] = "$Id: access.c 198277 2016-09-24 00:46:18Z twu $";
+static char rcsid[] = "$Id: access.c 213998 2018-03-03 04:59:34Z twu $";
 #ifdef HAVE_CONFIG_H
 #include <config.h>
 #endif
@@ -54,6 +54,9 @@ static char rcsid[] = "$Id: access.c 198277 2016-09-24 00:46:18Z twu $";
 #endif
 
 
+#define MAX_KILL_ATTEMPTS 5	/* Limits on attempts to kill unattached memory segments */
+
+
 #ifdef DEBUG
 #define debug(x) x
 #else
@@ -470,6 +473,45 @@ Access_deallocate (void *memory, int shmid, key_t key) {
 }
 
 
+/* Same as Access_deallocate, but without deleting semaphore */
+void
+Access_deallocate_wo_semaphore (void *memory, int shmid, key_t key) {
+  struct shmid_ds *buf = NULL;
+  short nattached;
+  int semid;
+
+  /* First, detach memory */
+  if (shmdt(memory) == -1) {
+#if 0      
+    /* Somehow, shmdt forks and prints the error message and continues with the rest of the code */
+    fprintf(stderr,"Error in Access_emergency_cleanup with shmdt on memory %p, shmid %d.  Error %d: %s\n",
+	    memory,shmid,errno,strerror(errno));
+#endif
+  }
+
+  /* Then, delete memory and semaphores, if applicable */
+  if ((semid = Semaphore_find(key)) != -1 && Semaphore_get_value(semid,SEMNO_KEEP) == SEMAPHORE_RESIDENT) {
+    fprintf(stderr,"Keeping memory for shmid %d resident, because it was pre-loaded.  To remove, run gsnap on this genome index with --unload-shared-memory\n",
+	    shmid);
+
+  } else if ((nattached = shmem_nattached(shmid)) > 0) {
+    fprintf(stderr,"Keeping memory for shmid %d resident, because it is being used by %d processes\n",
+	    shmid,nattached);
+
+  } else if (shmctl(shmid,IPC_RMID,buf) == -1) {
+    /* Somehow, shmctl forks and prints the error message and continues with the rest of the code */
+    /* fprintf(stderr,"Error in Access_deallocate with shmctl.  Error %d: %s\n",errno,strerror(errno)); */
+    /* Semaphore_delete(semid); */
+
+  } else {
+    fprintf(stderr,"Removed existing memory for shmid %d\n",shmid);
+    /* Semaphore_delete(semid); */
+  }
+
+  return;
+}
+
+
 
 #define FREAD_BATCH 100000000	/* 100 million elements at a time */
 
@@ -495,6 +537,17 @@ copy_memory_from_file (void *memory, char *filename, size_t filesize, size_t elt
       fread(p,sizeof(unsigned char),filesize/eltsize - i,fp);
     }
 
+  } else if (eltsize == 2) {
+    for (i = 0; i + FREAD_BATCH < filesize/eltsize; i += FREAD_BATCH) {
+      p = (void *) &(((UINT2 *) memory)[i]);
+      fread(p,sizeof(UINT2),FREAD_BATCH,fp);
+    }
+
+    if (i < filesize/eltsize) {
+      p = (void *) &(((UINT2 *) memory)[i]);
+      fread(p,sizeof(UINT2),filesize/eltsize - i,fp);
+    }
+
   } else if (eltsize == 4) {
     for (i = 0; i + FREAD_BATCH < filesize/eltsize; i += FREAD_BATCH) {
       p = (void *) &(((UINT4 *) memory)[i]);
@@ -545,6 +598,14 @@ copy_limited_from_file (void *memory, char *filename, size_t filesize, size_t el
       fread(p,sizeof(unsigned char),filesize/eltsize,fp);
     }
 
+  } else if (eltsize == 2) {
+    p = (void *) &(((UINT2 *) memory)[0]);
+    if (FREAD_BATCH < filesize/eltsize) {
+      fread(p,sizeof(UINT2),FREAD_BATCH,fp);
+    } else {
+      fread(p,sizeof(UINT2),filesize/eltsize,fp);
+    }
+
   } else if (eltsize == 4) {
     p = (void *) &(((UINT4 *) memory)[0]);
     if (FREAD_BATCH < filesize/eltsize) {
@@ -570,11 +631,15 @@ copy_limited_from_file (void *memory, char *filename, size_t filesize, size_t el
   return;
 }
 
+
 static void *
 shmem_attach (int *shmid, key_t *key, char *filename, size_t filesize, size_t eltsize) {
   void *memory = NULL;
   int semid = -1;
   ushort values[NSEMAPHORES];
+  int nattached;
+  bool donep;
+  int niter;
 
   values[SEMNO_LOCK] = -1;
   /* For some reason, these values are not being set, so using Semaphore_set_value below */
@@ -615,6 +680,9 @@ shmem_attach (int *shmid, key_t *key, char *filename, size_t filesize, size_t el
      others wait.  They will be woken up when the semaphore is
      removed. */
 
+  donep = false;
+  niter = 0;
+  while (donep == false) {
     if ((*shmid = shmget(*key,filesize,IPC_CREAT | IPC_EXCL |
 #ifdef HAVE_SHM_NORESERVE
 			 SHM_NORESERVE | 
@@ -628,28 +696,50 @@ shmem_attach (int *shmid, key_t *key, char *filename, size_t filesize, size_t el
 	shmem_memory = List_push(shmem_memory,memory);
 	shmem_ids = Intlist_push(shmem_ids,*shmid);
 	copy_limited_from_file(memory,filename,filesize,eltsize);
+	donep = true;
       } else {
 	semaphore_ids = Intlist_push(semaphore_ids,semid);
 	shmem_memory = List_push(shmem_memory,memory);
 	shmem_ids = Intlist_push(shmem_ids,*shmid);
 	copy_memory_from_file(memory,filename,filesize,eltsize);
 	fprintf(stderr,"Attached new memory for %s...",filename);
+	donep = true;
       }
 
     } else if ((*shmid = shmget(*key,0,0)) != -1) {
       /* Found existing shared memory */
       if ((memory = shmat(*shmid,NULL,0)) == (void *) -1) {
 	fprintf(stderr,"Error with shmat.  Error %d: %s\n",errno,strerror(errno));
-    } else {
+	
+      } else if ((nattached = shmem_nattached(*shmid)) > 1) {
+	/* Existing memory is indeed valid: attached (by a process other than this one) */
 	semaphore_ids = Intlist_push(semaphore_ids,semid);
 	shmem_memory = List_push(shmem_memory,memory);
 	shmem_ids = Intlist_push(shmem_ids,*shmid);
-      fprintf(stderr,"Attached existing memory for %s...",filename);
+	fprintf(stderr,"Attached existing memory (%d attached) for %s...",nattached,filename);
+	donep = true;
+
+      } else if (niter > MAX_KILL_ATTEMPTS) {
+	/* Existing memory is not attached by any other process, but unable to kill.  Could be corrupted */
+	fprintf(stderr,"Abandoned shared memory found, but could not be removed after %d attempts.  Using malloc instead of shared memory\n",niter);
+	memory = (void *) NULL;
+	donep = true;
+
+      } else {
+	/* Existing memory is not attached by any other process.  Could be corrupted. */
+	fprintf(stderr,"Clearing existing shared memory for %s...\n",filename);
+	Access_deallocate_wo_semaphore(memory,*shmid,*key);
+	
+	/* Now loop above to try again */
       }
     
     } else {
       fprintf(stderr,"Using malloc instead of shmget for file %s\n",filename);
       memory = (void *) NULL;
+      donep = true;
+    }
+
+    niter++;
   }
 
 #if 0
@@ -713,6 +803,8 @@ Access_allocate_private (Access_T *access, size_t *len, double *seconds, char *f
 
   if (eltsize == 1) {
     FREAD_CHARS(memory2,(*len)/eltsize,fp);
+  } else if (eltsize == 2) {
+    FREAD_USHORTS(memory2,(*len)/eltsize,fp);
   } else if (eltsize == 4) {
     FREAD_UINTS(memory2,(*len)/eltsize,fp);
   } else if (eltsize == 8) {
@@ -778,6 +870,8 @@ Access_allocate_shared (Access_T *access, int *shmid, key_t *key, int *fd, size_
 
   if (eltsize == 1) {
     FREAD_CHARS(memory2,(*len)/eltsize,fp);
+  } else if (eltsize == 2) {
+    FREAD_USHORTS(memory2,(*len)/eltsize,fp);
   } else if (eltsize == 4) {
     FREAD_UINTS(memory2,(*len)/eltsize,fp);
   } else if (eltsize == 8) {

src/atoi.c

-static char rcsid[] = "$Id: atoi.c 195989 2016-08-08 21:42:24Z twu $";
+static char rcsid[] = "$Id: atoi.c 214305 2018-03-19 23:40:43Z twu $";
 #ifdef HAVE_CONFIG_H
 #include <config.h>
 #endif
@@ -16435,3 +16435,18 @@ Atoi_reduce_ag (Oligospace_T oligo) {
   return reduced;
 }
+
+UINT2
+Atoi_reduce_tc_local (UINT2 oligo) {
+
+  /* Handles 16 bits, or k-mer up to 8 */
+  return convert_tc[oligo];
+}
+
+UINT2
+Atoi_reduce_ag_local (UINT2 oligo) {
+
+  /* Handles 16 bits, or k-mer up to 8 */
+  return convert_ag[oligo];
+}
+

src/atoi.h

-/* $Id: atoi.h 180699 2015-12-10 19:10:10Z twu $ */
+/* $Id: atoi.h 214305 2018-03-19 23:40:43Z twu $ */
 #ifndef ATOI_INCLUDED
 #define ATOI_INCLUDED
 
@@ -8,6 +8,10 @@ extern Oligospace_T
 Atoi_reduce_ag (Oligospace_T oligo);
 extern Oligospace_T
 Atoi_reduce_tc (Oligospace_T oligo);
+extern UINT2
+Atoi_reduce_tc_local (UINT2 oligo);
+extern UINT2
+Atoi_reduce_ag_local (UINT2 oligo);
 
 #endif
 

src/bitpack64-access.h

+/* $Id: bitpack64-access.h 212659 2018-01-20 00:58:14Z twu $ */
 #ifndef BITPACK64_ACCESS_INCLUDED
 #define BITPACK64_ACCESS_INCLUDED
 #include "types.h"

src/bitpack64-incr.c

-static char rcsid[] = "$Id: bitpack64-incr.c 199469 2016-10-23 03:49:07Z twu $";
+static char rcsid[] = "$Id: bitpack64-incr.c 212659 2018-01-20 00:58:14Z twu $";
 #ifdef HAVE_CONFIG_H
 #include <config.h>
 #endif
@@ -9537,7 +9537,7 @@ transfer_22_24 (UINT4 *out, const UINT4 *in) {
   value = ( CONVERT(*in) >>  18  )   % (1U << 22 ) ;
   in += WORD_INCR;
   value |= (CONVERT(*in) % (1U<< 8 ))<<( 22 - 8 );
-  *out |= (value << 8);	/* was 16 */
+  *out |= (value << 8);
 
   /* 12 */
   value = ( CONVERT(*in) >>  8  )   % (1U << 22 ) ;
@@ -9719,7 +9719,7 @@ transfer_26_28 (UINT4 *out, const UINT4 *in) {
   value |= (CONVERT(*in) % (1U<< 2 ))<<( 26 - 2 );
   *out |= (value << 16);
   out += WORD_INCR;
-  *out |= (value >> (28 - 12));	/* was (28 - 8) */
+  *out |= (value >> (28 - 12));
 
   /* 05 */
   value = ( CONVERT(*in) >>  2  )   % (1U << 26 ) ;

src/bitpack64-incr.h

+/* $Id: bitpack64-incr.h 212659 2018-01-20 00:58:14Z twu $ */
 #ifndef BITPACK64_INCR_INCLUDED
 #define BITPACK64_INCR_INCLUDED
 #include "types.h"