Module: Bio::FlatFileIndex::Indexer

Defined in:
lib/bio/io/flatfile/indexer.rb

Defined Under Namespace

Modules: Parser Classes: NameSpace, NameSpaces

Constant Summary collapse

DEFAULT_SORT =

default sort program

'/usr/bin/sort'
DEFAULT_ENV =

default env program (run a program in a modified environment)

'/usr/bin/env'
DEFAULT_ENV_ARGS =

default arguments for env program

[ 'LC_ALL=C' ]

Class Method Summary collapse

Class Method Details

.addindex_bdb(db, flag, need_update, parser, options) ⇒ Object

def



476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
# File 'lib/bio/io/flatfile/indexer.rb', line 476

def self.addindex_bdb(db, flag, need_update, parser, options)
  DEBUG.print "reading files...\n"

  pn = db.primary
  pn.file.close
  pn.file.flag = flag

  db.secondary.each_files do |x|
    x.file.close
    x.file.flag = flag
    x.file.open
    x.file.close
  end

  need_update.each do |fileid|
    filename = db.fileids[fileid].filename
    parser.open_flatfile(fileid, filename)
    parser.each do |pos, len|
      p = parser.parse_primary
      #pn.file.add_exclusive(p, [ fileid, pos, len ])
      pn.file.add_overwrite(p, [ fileid, pos, len ])
      #DEBUG.print "#{p} #{fileid} #{pos} #{len}\n"
      parser.parse_secondary do |sn, sp|
        db.secondary[sn].file.add_nr(sp, p)
        #DEBUG.print "#{sp} #{p}\n"
      end
    end
    parser.close_flatfile
  end
  true
end

.addindex_flat(db, mode, need_update, parser, options) ⇒ Object

def



525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
# File 'lib/bio/io/flatfile/indexer.rb', line 525

def self.addindex_flat(db, mode, need_update, parser, options)
  require 'tempfile'
  prog = options['sort_program']
  env = options['env_program']
  env_args = options['env_program_arguments']

  return false if need_update.to_a.size == 0

  DEBUG.print "prepare temporary files...\n"
  tempbase = "bioflat#{rand(10000)}-"
  pfile = Tempfile.open(tempbase + 'primary-')
  DEBUG.print "open temporary file #{pfile.path.inspect}\n"
  sfiles = {}
  parser.secondary.names.each do |x|
    sfiles[x] =  Tempfile.open(tempbase + 'secondary-')
    DEBUG.print "open temporary file #{sfiles[x].path.inspect}\n"
  end

  DEBUG.print "reading files...\n"
  need_update.each do |fileid|
    filename = db.fileids[fileid].filename
    parser.open_flatfile(fileid, filename)
    parser.each do |pos, len|
      p = parser.parse_primary
      pfile << "#{p}\t#{fileid}\t#{pos}\t#{len}\n"
      #DEBUG.print "#{p} #{fileid} #{pos} #{len}\n"
      parser.parse_secondary do |sn, sp|
        sfiles[sn] << "#{sp}\t#{p}\n"
        #DEBUG.print "#{sp} #{p}\n"
      end
    end
    parser.close_flatfile
    fileid += 1
  end

  sort_proc = chose_sort_proc(prog, mode, env, env_args)
  pfile.close(false)
  DEBUG.print "sorting primary (#{parser.primary.name})...\n"
  db.primary.file.import_tsv_files(true, mode, sort_proc, pfile.path)
  pfile.close(true)

  parser.secondary.names.each do |x|
    DEBUG.print "sorting secondary (#{x})...\n"
    sfiles[x].close(false)
    db.secondary[x].file.import_tsv_files(false, mode, sort_proc,
                                          sfiles[x].path)
    sfiles[x].close(true)
  end
  true
end

.chose_sort_proc(prog, mode = :new, env = nil, env_args = nil) ⇒ Object



585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
# File 'lib/bio/io/flatfile/indexer.rb', line 585

def self.chose_sort_proc(prog, mode = :new,
                         env = nil, env_args = nil)
  case prog
  when /^builtin$/i, /^hs$/i, /^lm$/i
    DEBUG.print "sort: internal sort routine\n"
    sort_proc = Flat_1::FlatMappingFile::internal_sort_proc
  when nil, ''
    if FileTest.executable?(DEFAULT_SORT)
      return chose_sort_proc(DEFAULT_SORT, mode, env, env_args)
    else
      DEBUG.print "sort: internal sort routine\n"
      sort_proc = Flat_1::FlatMappingFile::internal_sort_proc
    end
  else
    env_args ||= DEFAULT_ENV_ARGS
    if env == '' or env == false then # inhibit to use env program
      prefixes = [ prog ]
    elsif env then # uses given env program
      prefixes = [ env ] + env_args + [ prog ]
    else # env == nil; uses default env program if possible
      if FileTest.executable?(DEFAULT_ENV)
        prefixes = [ DEFAULT_ENV ] + env_args + [ prog ]
      else
        prefixes = [ prog ]
      end
    end
    DEBUG.print "sort: #{prefixes.join(' ')}\n"
    if mode == :new then
      sort_proc = Flat_1::FlatMappingFile::external_sort_proc(prefixes)
    else
      sort_proc = Flat_1::FlatMappingFile::external_merge_sort_proc(prefixes)
    end
  end
  sort_proc
end

.makeindexBDB(name, parser, options, *files) ⇒ Object

module Parser



451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
# File 'lib/bio/io/flatfile/indexer.rb', line 451

def self.makeindexBDB(name, parser, options, *files)
  # options are not used in this method
  unless defined?(BDB)
    raise RuntimeError, "Berkeley DB support not found"
  end
  DEBUG.print "makeing BDB DataBank...\n"
  db = DataBank.new(name, MAGIC_BDB)
  db.format = parser.format
  db.fileids.add(*files)
  db.fileids.recalc

  db.primary = parser.primary.name
  db.secondary = parser.secondary.names

  DEBUG.print "writing config.dat, config, fileids ...\n"
  db.write('wb', BDBdefault::flag_write)

  DEBUG.print "reading files...\n"

  addindex_bdb(db, BDBdefault::flag_write, (0...(files.size)),
               parser, options)
  db.close
  true
end

.makeindexFlat(name, parser, options, *files) ⇒ Object

def



508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
# File 'lib/bio/io/flatfile/indexer.rb', line 508

def self.makeindexFlat(name, parser, options, *files)
  DEBUG.print "makeing flat/1 DataBank using temporary files...\n"

  db = DataBank.new(name, nil)
  db.format = parser.format
  db.fileids.add(*files)
  db.primary = parser.primary.name
  db.secondary = parser.secondary.names
  db.fileids.recalc
  DEBUG.print "writing DabaBank...\n"
  db.write('wb')

  addindex_flat(db, :new, (0...(files.size)), parser, options)
  db.close
  true
end

.update_index(name, parser, options, *files) ⇒ Object



621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
# File 'lib/bio/io/flatfile/indexer.rb', line 621

def self.update_index(name, parser, options, *files)
  db = DataBank.open(name)

  if parser then
    raise 'file format mismatch' if db.format != parser.format
  else

    begin
      dbclass_orig =
        Bio::FlatFile.autodetect_file(db.fileids[0].filename)
    rescue TypeError, Errno::ENOENT
    end
    begin
      dbclass_new =
        Bio::FlatFile.autodetect_file(files[0])
    rescue TypeError, Errno::ENOENT
    end

    case db.format
    when 'swiss', 'embl'
      parser = Parser.new(db.format)
      if dbclass_new and dbclass_new != parser.dbclass
        raise 'file format mismatch'
      end
    when 'genbank'
      dbclass = dbclass_orig or dbclass_new
      if dbclass == Bio::GenBank or dbclass == Bio::GenPept
        parser = Parser.new(dbclass_orig)
      elsif !dbclass then
        raise 'cannnot determine format. please specify manually.'
      else
        raise 'file format mismatch'
      end
      if dbclass_new and dbclass_new != parser.dbclass
        raise 'file format mismatch'
      end
    else
      raise 'unsupported format'
    end
  end

  parser.set_primary_namespace(db.primary.name)
  parser.add_secondary_namespaces(*db.secondary.names)

  if options['renew'] then
    newfiles = db.fileids.filenames.find_all do |x|
      FileTest.exist?(x)
    end
    newfiles.concat(files)
    newfiles2 = newfiles.sort
    newfiles2.uniq!
    newfiles3 = []
    newfiles.each do |x|
      newfiles3 << x if newfiles2.delete(x)
    end
    t = db.index_type
    db.close
    case t
    when MAGIC_BDB
      Indexer::makeindexBDB(name, parser, options, *newfiles3)
    when MAGIC_FLAT
      Indexer::makeindexFlat(name, parser, options, *newfiles3)
    else
      raise 'Unsupported index type'
    end
    return true
  end

  need_update = []
  newfiles = files.dup
  db.fileids.cache_all
  db.fileids.each_with_index do |f, i|
    need_update << i unless f.check
    newfiles.delete(f.filename)
  end

  b = db.fileids.size
  begin
    db.fileids.recalc
  rescue Errno::ENOENT => evar
    DEBUG.print "Error: #{evar}\n"
    DEBUG.print "assumed --renew option\n"
    db.close
    options = options.dup
    options['renew'] = true
    update_index(name, parser, options, *files)
    return true
  end
  # add new files
  db.fileids.add(*newfiles)
  db.fileids.recalc

  need_update.concat((b...(b + newfiles.size)).to_a)

  DEBUG.print "writing DabaBank...\n"
  db.write('wb', BDBdefault::flag_append)

  case db.index_type
  when MAGIC_BDB
    addindex_bdb(db, BDBdefault::flag_append,
                 need_update, parser, options)
  when MAGIC_FLAT
    addindex_flat(db, :add, need_update, parser, options)
  else
    raise 'Unsupported index type'
  end

  db.close
  true
end