Class: Oddb2xml::ZurroseExtractor

Inherits:
Extractor
  • Object
show all
Defined in:
lib/oddb2xml/extractor.rb

Instance Attribute Summary

Attributes inherited from Extractor

#xml

Instance Method Summary collapse

Constructor Details

#initialize(dat, extended = false, artikelstamm = false) ⇒ ZurroseExtractor



472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
# File 'lib/oddb2xml/extractor.rb', line 472

def initialize(dat, extended = false, artikelstamm = false)
  @@extended = extended
  @artikelstamm = artikelstamm
  FileUtils.makedirs(WorkDir)
  @@error_file ||= File.open(File.join(WorkDir, "duplicate_ean13_from_zur_rose.txt"), 'wb+:ISO-8859-14')
  @@items_without_ean13s ||= 0
  @@duplicated_ean13s ||= 0
  @@zur_rose_items ||= 0
  if dat
    if File.exists?(dat)
      @io = File.open(dat, 'rb:ISO-8859-14')
    else
      @io = StringIO.new(dat)
    end
    @io
  else
     nil
  end
end

Instance Method Details

#to_hashObject



491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
# File 'lib/oddb2xml/extractor.rb', line 491

def to_hash
  data = {}
  while line = @io.gets
    line = Oddb2xml.patch_some_utf8(line).chomp
    next if line =~ /(ad us\.* vet)|(\(vet\))/i
    if @@extended
      next unless line =~ /(\d{13})(\d{1})$/
    else
      next unless line =~ /(7680\d{9})(\d{1})$/
    end
    pharma_code = line[3..9]
    if $1.to_s == '0000000000000'
      @@items_without_ean13s += 1
      ean13 = '999999' + pharma_code.to_s # dummy ean13
    else
      ean13 = $1
    end
    next if @artikelstamm && /(0{13})(\d{1})$/.match(line)
    if data[ean13]
      @@error_file.puts "Duplicate ean13 #{ean13} in line \nact: #{line.chomp}\norg: #{data[ean13][:line]}"
      @@items_without_ean13s -= 1
      @@duplicated_ean13s += 1
      next
    end

    pexf = sprintf("%.2f", line[60,6].gsub(/(\d{2})$/, '.\1').to_f)
    ppub =  sprintf("%.2f", line[66,6].gsub(/(\d{2})$/, '.\1').to_f)
    next if  @artikelstamm && /^113/.match(line) && /^7680/.match(ean13)
    next if  @artikelstamm && /^113/.match(line) && ppub.eql?('0.0') && pexf.eql?('0.0')
    data[ean13] = {
      :data_origin   => 'zur_rose',
      :line   => line.chomp,
      :ean13 => ean13,
      :clag  => line[73],
      :vat   => line[96],
      :description => line[10..59].sub(/\s+$/, ''),
      :quantity => '',
      :pharmacode => pharma_code,
      :price => pexf,
      :pub_price => ppub,
      :type => :nonpharma,
      :cmut => line[2],
    }
    @@zur_rose_items += 1
  end if @io
  if defined?(@@extended) and @@extended
    @@error_file.puts get_error_msg
  end
  @@error_file.close
  @@error_file = nil
  data
end