summaryrefslogtreecommitdiff
path: root/lib/csv/parser.rb
blob: 7e943acf21465df98971527ca33dc48157bdacbb (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
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
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
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
575
576
577
578
579
580
581
582
583
584
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
620
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
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
# frozen_string_literal: true

require "strscan"

require_relative "delete_suffix"
require_relative "input_record_separator"
require_relative "match_p"
require_relative "row"
require_relative "table"

using CSV::DeleteSuffix if CSV.const_defined?(:DeleteSuffix)
using CSV::MatchP if CSV.const_defined?(:MatchP)

class CSV
  # Note: Don't use this class directly. This is an internal class.
  class Parser
    #
    # A CSV::Parser is m17n aware. The parser works in the Encoding of the IO
    # or String object being read from or written to. Your data is never transcoded
    # (unless you ask Ruby to transcode it for you) and will literally be parsed in
    # the Encoding it is in. Thus CSV will return Arrays or Rows of Strings in the
    # Encoding of your data. This is accomplished by transcoding the parser itself
    # into your Encoding.
    #

    # Raised when encoding is invalid.
    class InvalidEncoding < StandardError
    end

    #
    # CSV::Scanner receives a CSV output, scans it and return the content.
    # It also controls the life cycle of the object with its methods +keep_start+,
    # +keep_end+, +keep_back+, +keep_drop+.
    #
    # Uses StringScanner (the official strscan gem). Strscan provides lexical
    # scanning operations on a String. We inherit its object and take advantage
    # on the methods. For more information, please visit:
    # https://ruby-doc.org/stdlib-2.6.1/libdoc/strscan/rdoc/StringScanner.html
    #
    class Scanner < StringScanner
      alias_method :scan_all, :scan

      def initialize(*args)
        super
        @keeps = []
      end

      def each_line(row_separator)
        position = pos
        rest.each_line(row_separator) do |line|
          position += line.bytesize
          self.pos = position
          yield(line)
        end
      end

      def keep_start
        @keeps.push(pos)
      end

      def keep_end
        start = @keeps.pop
        string.byteslice(start, pos - start)
      end

      def keep_back
        self.pos = @keeps.pop
      end

      def keep_drop
        @keeps.pop
      end
    end

    #
    # CSV::InputsScanner receives IO inputs, encoding and the chunk_size.
    # It also controls the life cycle of the object with its methods +keep_start+,
    # +keep_end+, +keep_back+, +keep_drop+.
    #
    # CSV::InputsScanner.scan() tries to match with pattern at the current position.
    # If there's a match, the scanner advances the “scan pointer” and returns the matched string.
    # Otherwise, the scanner returns nil.
    #
    # CSV::InputsScanner.rest() returns the “rest” of the string (i.e. everything after the scan pointer).
    # If there is no more data (eos? = true), it returns "".
    #
    class InputsScanner
      def initialize(inputs, encoding, row_separator, chunk_size: 8192)
        @inputs = inputs.dup
        @encoding = encoding
        @row_separator = row_separator
        @chunk_size = chunk_size
        @last_scanner = @inputs.empty?
        @keeps = []
        read_chunk
      end

      def each_line(row_separator)
        buffer = nil
        input = @scanner.rest
        position = @scanner.pos
        offset = 0
        n_row_separator_chars = row_separator.size
        while true
          input.each_line(row_separator) do |line|
            @scanner.pos += line.bytesize
            if buffer
              if n_row_separator_chars == 2 and
                buffer.end_with?(row_separator[0]) and
                line.start_with?(row_separator[1])
                buffer << line[0]
                line = line[1..-1]
                position += buffer.bytesize + offset
                @scanner.pos = position
                offset = 0
                yield(buffer)
                buffer = nil
                next if line.empty?
              else
                buffer << line
                line = buffer
                buffer = nil
              end
            end
            if line.end_with?(row_separator)
              position += line.bytesize + offset
              @scanner.pos = position
              offset = 0
              yield(line)
            else
              buffer = line
            end
          end
          break unless read_chunk
          input = @scanner.rest
          position = @scanner.pos
          offset = -buffer.bytesize if buffer
        end
        yield(buffer) if buffer
      end

      def scan(pattern)
        value = @scanner.scan(pattern)
        return value if @last_scanner

        if value
          read_chunk if @scanner.eos?
          return value
        else
          nil
        end
      end

      def scan_all(pattern)
        value = @scanner.scan(pattern)
        return value if @last_scanner

        return nil if value.nil?
        while @scanner.eos? and read_chunk and (sub_value = @scanner.scan(pattern))
          value << sub_value
        end
        value
      end

      def eos?
        @scanner.eos?
      end

      def keep_start
        @keeps.push([@scanner.pos, nil])
      end

      def keep_end
        start, buffer = @keeps.pop
        keep = @scanner.string.byteslice(start, @scanner.pos - start)
        if buffer
          buffer << keep
          keep = buffer
        end
        keep
      end

      def keep_back
        start, buffer = @keeps.pop
        if buffer
          string = @scanner.string
          keep = string.byteslice(start, string.bytesize - start)
          if keep and not keep.empty?
            @inputs.unshift(StringIO.new(keep))
            @last_scanner = false
          end
          @scanner = StringScanner.new(buffer)
        else
          @scanner.pos = start
        end
        read_chunk if @scanner.eos?
      end

      def keep_drop
        @keeps.pop
      end

      def rest
        @scanner.rest
      end

      private
      def read_chunk
        return false if @last_scanner

        unless @keeps.empty?
          keep = @keeps.last
          keep_start = keep[0]
          string = @scanner.string
          keep_data = string.byteslice(keep_start, @scanner.pos - keep_start)
          if keep_data
            keep_buffer = keep[1]
            if keep_buffer
              keep_buffer << keep_data
            else
              keep[1] = keep_data.dup
            end
          end
          keep[0] = 0
        end

        input = @inputs.first
        case input
        when StringIO
          string = input.read
          raise InvalidEncoding unless string.valid_encoding?
          @scanner = StringScanner.new(string)
          @inputs.shift
          @last_scanner = @inputs.empty?
          true
        else
          chunk = input.gets(@row_separator, @chunk_size)
          if chunk
            raise InvalidEncoding unless chunk.valid_encoding?
            @scanner = StringScanner.new(chunk)
            if input.respond_to?(:eof?) and input.eof?
              @inputs.shift
              @last_scanner = @inputs.empty?
            end
            true
          else
            @scanner = StringScanner.new("".encode(@encoding))
            @inputs.shift
            @last_scanner = @inputs.empty?
            if @last_scanner
              false
            else
              read_chunk
            end
          end
        end
      end
    end

    def initialize(input, options)
      @input = input
      @options = options
      @samples = []

      prepare
    end

    def column_separator
      @column_separator
    end

    def row_separator
      @row_separator
    end

    def quote_character
      @quote_character
    end

    def field_size_limit
      @field_size_limit
    end

    def skip_lines
      @skip_lines
    end

    def unconverted_fields?
      @unconverted_fields
    end

    def headers
      @headers
    end

    def header_row?
      @use_headers and @headers.nil?
    end

    def return_headers?
      @return_headers
    end

    def skip_blanks?
      @skip_blanks
    end

    def liberal_parsing?
      @liberal_parsing
    end

    def lineno
      @lineno
    end

    def line
      last_line
    end

    def parse(&block)
      return to_enum(__method__) unless block_given?

      if @return_headers and @headers and @raw_headers
        headers = Row.new(@headers, @raw_headers, true)
        if @unconverted_fields
          headers = add_unconverted_fields(headers, [])
        end
        yield headers
      end

      begin
        @scanner ||= build_scanner
        if quote_character.nil?
          parse_no_quote(&block)
        elsif @need_robust_parsing
          parse_quotable_robust(&block)
        else
          parse_quotable_loose(&block)
        end
      rescue InvalidEncoding
        if @scanner
          ignore_broken_line
          lineno = @lineno
        else
          lineno = @lineno + 1
        end
        message = "Invalid byte sequence in #{@encoding}"
        raise MalformedCSVError.new(message, lineno)
      end
    end

    def use_headers?
      @use_headers
    end

    private
    # A set of tasks to prepare the file in order to parse it
    def prepare
      prepare_variable
      prepare_quote_character
      prepare_backslash
      prepare_skip_lines
      prepare_strip
      prepare_separators
      validate_strip_and_col_sep_options
      prepare_quoted
      prepare_unquoted
      prepare_line
      prepare_header
      prepare_parser
    end

    def prepare_variable
      @need_robust_parsing = false
      @encoding = @options[:encoding]
      liberal_parsing = @options[:liberal_parsing]
      if liberal_parsing
        @liberal_parsing = true
        if liberal_parsing.is_a?(Hash)
          @double_quote_outside_quote =
            liberal_parsing[:double_quote_outside_quote]
          @backslash_quote = liberal_parsing[:backslash_quote]
        else
          @double_quote_outside_quote = false
          @backslash_quote = false
        end
        @need_robust_parsing = true
      else
        @liberal_parsing = false
        @backslash_quote = false
      end
      @unconverted_fields = @options[:unconverted_fields]
      @field_size_limit = @options[:field_size_limit]
      @skip_blanks = @options[:skip_blanks]
      @fields_converter = @options[:fields_converter]
      @header_fields_converter = @options[:header_fields_converter]
    end

    def prepare_quote_character
      @quote_character = @options[:quote_character]
      if @quote_character.nil?
        @escaped_quote_character = nil
        @escaped_quote = nil
      else
        @quote_character = @quote_character.to_s.encode(@encoding)
        if @quote_character.length != 1
          message = ":quote_char has to be nil or a single character String"
          raise ArgumentError, message
        end
        @double_quote_character = @quote_character * 2
        @escaped_quote_character = Regexp.escape(@quote_character)
        @escaped_quote = Regexp.new(@escaped_quote_character)
      end
    end

    def prepare_backslash
      return unless @backslash_quote

      @backslash_character = "\\".encode(@encoding)

      @escaped_backslash_character = Regexp.escape(@backslash_character)
      @escaped_backslash = Regexp.new(@escaped_backslash_character)
      if @quote_character.nil?
        @backslash_quote_character = nil
      else
        @backslash_quote_character =
          @backslash_character + @escaped_quote_character
      end
    end

    def prepare_skip_lines
      skip_lines = @options[:skip_lines]
      case skip_lines
      when String
        @skip_lines = skip_lines.encode(@encoding)
      when Regexp, nil
        @skip_lines = skip_lines
      else
        unless skip_lines.respond_to?(:match)
          message =
            ":skip_lines has to respond to \#match: #{skip_lines.inspect}"
          raise ArgumentError, message
        end
        @skip_lines = skip_lines
      end
    end

    def prepare_strip
      @strip = @options[:strip]
      @escaped_strip = nil
      @strip_value = nil
      @rstrip_value = nil
      if @strip.is_a?(String)
        case @strip.length
        when 0
          raise ArgumentError, ":strip must not be an empty String"
        when 1
          # ok
        else
          raise ArgumentError, ":strip doesn't support 2 or more characters yet"
        end
        @strip = @strip.encode(@encoding)
        @escaped_strip = Regexp.escape(@strip)
        if @quote_character
          @strip_value = Regexp.new(@escaped_strip +
                                    "+".encode(@encoding))
          @rstrip_value = Regexp.new(@escaped_strip +
                                     "+\\z".encode(@encoding))
        end
        @need_robust_parsing = true
      elsif @strip
        strip_values = " \t\f\v"
        @escaped_strip = strip_values.encode(@encoding)
        if @quote_character
          @strip_value = Regexp.new("[#{strip_values}]+".encode(@encoding))
          @rstrip_value = Regexp.new("[#{strip_values}]+\\z".encode(@encoding))
        end
        @need_robust_parsing = true
      end
    end

    begin
      StringScanner.new("x").scan("x")
    rescue TypeError
      STRING_SCANNER_SCAN_ACCEPT_STRING = false
    else
      STRING_SCANNER_SCAN_ACCEPT_STRING = true
    end

    def prepare_separators
      column_separator = @options[:column_separator]
      @column_separator = column_separator.to_s.encode(@encoding)
      if @column_separator.size < 1
        message = ":col_sep must be 1 or more characters: "
        message += column_separator.inspect
        raise ArgumentError, message
      end
      @row_separator =
        resolve_row_separator(@options[:row_separator]).encode(@encoding)

      @escaped_column_separator = Regexp.escape(@column_separator)
      @escaped_first_column_separator = Regexp.escape(@column_separator[0])
      if @column_separator.size > 1
        @column_end = Regexp.new(@escaped_column_separator)
        @column_ends = @column_separator.each_char.collect do |char|
          Regexp.new(Regexp.escape(char))
        end
        @first_column_separators = Regexp.new(@escaped_first_column_separator +
                                              "+".encode(@encoding))
      else
        if STRING_SCANNER_SCAN_ACCEPT_STRING
          @column_end = @column_separator
        else
          @column_end = Regexp.new(@escaped_column_separator)
        end
        @column_ends = nil
        @first_column_separators = nil
      end

      escaped_row_separator = Regexp.escape(@row_separator)
      @row_end = Regexp.new(escaped_row_separator)
      if @row_separator.size > 1
        @row_ends = @row_separator.each_char.collect do |char|
          Regexp.new(Regexp.escape(char))
        end
      else
        @row_ends = nil
      end

      @cr = "\r".encode(@encoding)
      @lf = "\n".encode(@encoding)
      @line_end = Regexp.new("\r\n|\n|\r".encode(@encoding))
      @not_line_end = Regexp.new("[^\r\n]+".encode(@encoding))
    end

    # This method verifies that there are no (obvious) ambiguities with the
    # provided +col_sep+ and +strip+ parsing options. For example, if +col_sep+
    # and +strip+ were both equal to +\t+, then there would be no clear way to
    # parse the input.
    def validate_strip_and_col_sep_options
      return unless @strip

      if @strip.is_a?(String)
        if @column_separator.start_with?(@strip) || @column_separator.end_with?(@strip)
          raise ArgumentError,
                "The provided strip (#{@escaped_strip}) and " \
                "col_sep (#{@escaped_column_separator}) options are incompatible."
        end
      else
        if Regexp.new("\\A[#{@escaped_strip}]|[#{@escaped_strip}]\\z").match?(@column_separator)
          raise ArgumentError,
                "The provided strip (true) and " \
                "col_sep (#{@escaped_column_separator}) options are incompatible."
        end
      end
    end

    def prepare_quoted
      if @quote_character
        @quotes = Regexp.new(@escaped_quote_character +
                             "+".encode(@encoding))
        no_quoted_values = @escaped_quote_character.dup
        if @backslash_quote
          no_quoted_values << @escaped_backslash_character
        end
        @quoted_value = Regexp.new("[^".encode(@encoding) +
                                   no_quoted_values +
                                   "]+".encode(@encoding))
      end
      if @escaped_strip
        @split_column_separator = Regexp.new(@escaped_strip +
                                             "*".encode(@encoding) +
                                             @escaped_column_separator +
                                             @escaped_strip +
                                             "*".encode(@encoding))
      else
        if @column_separator == " ".encode(@encoding)
          @split_column_separator = Regexp.new(@escaped_column_separator)
        else
          @split_column_separator = @column_separator
        end
      end
    end

    def prepare_unquoted
      return if @quote_character.nil?

      no_unquoted_values = "\r\n".encode(@encoding)
      no_unquoted_values << @escaped_first_column_separator
      unless @liberal_parsing
        no_unquoted_values << @escaped_quote_character
      end
      @unquoted_value = Regexp.new("[^".encode(@encoding) +
                                   no_unquoted_values +
                                   "]+".encode(@encoding))
    end

    def resolve_row_separator(separator)
      if separator == :auto
        cr = "\r".encode(@encoding)
        lf = "\n".encode(@encoding)
        if @input.is_a?(StringIO)
          pos = @input.pos
          separator = detect_row_separator(@input.read, cr, lf)
          @input.seek(pos)
        elsif @input.respond_to?(:gets)
          if @input.is_a?(File)
            chunk_size = 32 * 1024
          else
            chunk_size = 1024
          end
          begin
            while separator == :auto
              #
              # if we run out of data, it's probably a single line
              # (ensure will set default value)
              #
              break unless sample = @input.gets(nil, chunk_size)

              # extend sample if we're unsure of the line ending
              if sample.end_with?(cr)
                sample << (@input.gets(nil, 1) || "")
              end

              @samples << sample

              separator = detect_row_separator(sample, cr, lf)
            end
          rescue IOError
            # do nothing:  ensure will set default
          end
        end
        separator = InputRecordSeparator.value if separator == :auto
      end
      separator.to_s.encode(@encoding)
    end

    def detect_row_separator(sample, cr, lf)
      lf_index = sample.index(lf)
      if lf_index
        cr_index = sample[0, lf_index].index(cr)
      else
        cr_index = sample.index(cr)
      end
      if cr_index and lf_index
        if cr_index + 1 == lf_index
          cr + lf
        elsif cr_index < lf_index
          cr
        else
          lf
        end
      elsif cr_index
        cr
      elsif lf_index
        lf
      else
        :auto
      end
    end

    def prepare_line
      @lineno = 0
      @last_line = nil
      @scanner = nil
    end

    def last_line
      if @scanner
        @last_line ||= @scanner.keep_end
      else
        @last_line
      end
    end

    def prepare_header
      @return_headers = @options[:return_headers]

      headers = @options[:headers]
      case headers
      when Array
        @raw_headers = headers
        @use_headers = true
      when String
        @raw_headers = parse_headers(headers)
        @use_headers = true
      when nil, false
        @raw_headers = nil
        @use_headers = false
      else
        @raw_headers = nil
        @use_headers = true
      end
      if @raw_headers
        @headers = adjust_headers(@raw_headers)
      else
        @headers = nil
      end
    end

    def parse_headers(row)
      CSV.parse_line(row,
                     col_sep:    @column_separator,
                     row_sep:    @row_separator,
                     quote_char: @quote_character)
    end

    def adjust_headers(headers)
      adjusted_headers = @header_fields_converter.convert(headers, nil, @lineno)
      adjusted_headers.each {|h| h.freeze if h.is_a? String}
      adjusted_headers
    end

    def prepare_parser
      @may_quoted = may_quoted?
    end

    def may_quoted?
      return false if @quote_character.nil?

      if @input.is_a?(StringIO)
        pos = @input.pos
        sample = @input.read
        @input.seek(pos)
      else
        return false if @samples.empty?
        sample = @samples.first
      end
      sample[0, 128].index(@quote_character)
    end

    SCANNER_TEST = (ENV["CSV_PARSER_SCANNER_TEST"] == "yes")
    if SCANNER_TEST
      class UnoptimizedStringIO
        def initialize(string)
          @io = StringIO.new(string, "rb:#{string.encoding}")
        end

        def gets(*args)
          @io.gets(*args)
        end

        def each_line(*args, &block)
          @io.each_line(*args, &block)
        end

        def eof?
          @io.eof?
        end
      end

      SCANNER_TEST_CHUNK_SIZE =
        Integer((ENV["CSV_PARSER_SCANNER_TEST_CHUNK_SIZE"] || "1"), 10)
      def build_scanner
        inputs = @samples.collect do |sample|
          UnoptimizedStringIO.new(sample)
        end
        if @input.is_a?(StringIO)
          inputs << UnoptimizedStringIO.new(@input.read)
        else
          inputs << @input
        end
        InputsScanner.new(inputs,
                          @encoding,
                          @row_separator,
                          chunk_size: SCANNER_TEST_CHUNK_SIZE)
      end
    else
      def build_scanner
        string = nil
        if @samples.empty? and @input.is_a?(StringIO)
          string = @input.read
        elsif @samples.size == 1 and
              @input != ARGF and
              @input.respond_to?(:eof?) and
              @input.eof?
          string = @samples[0]
        end
        if string
          unless string.valid_encoding?
            index = string.lines(@row_separator).index do |line|
              !line.valid_encoding?
            end
            if index
              message = "Invalid byte sequence in #{@encoding}"
              raise MalformedCSVError.new(message, @lineno + index + 1)
            end
          end
          Scanner.new(string)
        else
          inputs = @samples.collect do |sample|
            StringIO.new(sample)
          end
          inputs << @input
          InputsScanner.new(inputs, @encoding, @row_separator)
        end
      end
    end

    def skip_needless_lines
      return unless @skip_lines

      until @scanner.eos?
        @scanner.keep_start
        line = @scanner.scan_all(@not_line_end) || "".encode(@encoding)
        line << @row_separator if parse_row_end
        if skip_line?(line)
          @lineno += 1
          @scanner.keep_drop
        else
          @scanner.keep_back
          return
        end
      end
    end

    def skip_line?(line)
      line = line.delete_suffix(@row_separator)
      case @skip_lines
      when String
        line.include?(@skip_lines)
      when Regexp
        @skip_lines.match?(line)
      else
        @skip_lines.match(line)
      end
    end

    def parse_no_quote(&block)
      @scanner.each_line(@row_separator) do |line|
        next if @skip_lines and skip_line?(line)
        original_line = line
        line = line.delete_suffix(@row_separator)

        if line.empty?
          next if @skip_blanks
          row = []
        else
          line = strip_value(line)
          row = line.split(@split_column_separator, -1)
          n_columns = row.size
          i = 0
          while i < n_columns
            row[i] = nil if row[i].empty?
            i += 1
          end
        end
        @last_line = original_line
        emit_row(row, &block)
      end
    end

    def parse_quotable_loose(&block)
      @scanner.keep_start
      @scanner.each_line(@row_separator) do |line|
        if @skip_lines and skip_line?(line)
          @scanner.keep_drop
          @scanner.keep_start
          next
        end
        original_line = line
        line = line.delete_suffix(@row_separator)

        if line.empty?
          if @skip_blanks
            @scanner.keep_drop
            @scanner.keep_start
            next
          end
          row = []
        elsif line.include?(@cr) or line.include?(@lf)
          @scanner.keep_back
          @need_robust_parsing = true
          return parse_quotable_robust(&block)
        else
          row = line.split(@split_column_separator, -1)
          n_columns = row.size
          i = 0
          while i < n_columns
            column = row[i]
            if column.empty?
              row[i] = nil
            else
              n_quotes = column.count(@quote_character)
              if n_quotes.zero?
                # no quote
              elsif n_quotes == 2 and
                   column.start_with?(@quote_character) and
                   column.end_with?(@quote_character)
                row[i] = column[1..-2]
              else
                @scanner.keep_back
                @need_robust_parsing = true
                return parse_quotable_robust(&block)
              end
            end
            i += 1
          end
        end
        @scanner.keep_drop
        @scanner.keep_start
        @last_line = original_line
        emit_row(row, &block)
      end
      @scanner.keep_drop
    end

    def parse_quotable_robust(&block)
      row = []
      skip_needless_lines
      start_row
      while true
        @quoted_column_value = false
        @unquoted_column_value = false
        @scanner.scan_all(@strip_value) if @strip_value
        value = parse_column_value
        if value
          @scanner.scan_all(@strip_value) if @strip_value
          if @field_size_limit and value.size >= @field_size_limit
            ignore_broken_line
            raise MalformedCSVError.new("Field size exceeded", @lineno)
          end
        end
        if parse_column_end
          row << value
        elsif parse_row_end
          if row.empty? and value.nil?
            emit_row([], &block) unless @skip_blanks
          else
            row << value
            emit_row(row, &block)
            row = []
          end
          skip_needless_lines
          start_row
        elsif @scanner.eos?
          break if row.empty? and value.nil?
          row << value
          emit_row(row, &block)
          break
        else
          if @quoted_column_value
            ignore_broken_line
            message = "Any value after quoted field isn't allowed"
            raise MalformedCSVError.new(message, @lineno)
          elsif @unquoted_column_value and
                (new_line = @scanner.scan(@line_end))
            ignore_broken_line
            message = "Unquoted fields do not allow new line " +
                      "<#{new_line.inspect}>"
            raise MalformedCSVError.new(message, @lineno)
          elsif @scanner.rest.start_with?(@quote_character)
            ignore_broken_line
            message = "Illegal quoting"
            raise MalformedCSVError.new(message, @lineno)
          elsif (new_line = @scanner.scan(@line_end))
            ignore_broken_line
            message = "New line must be <#{@row_separator.inspect}> " +
                      "not <#{new_line.inspect}>"
            raise MalformedCSVError.new(message, @lineno)
          else
            ignore_broken_line
            raise MalformedCSVError.new("TODO: Meaningful message",
                                        @lineno)
          end
        end
      end
    end

    def parse_column_value
      if @liberal_parsing
        quoted_value = parse_quoted_column_value
        if quoted_value
          @scanner.scan_all(@strip_value) if @strip_value
          unquoted_value = parse_unquoted_column_value
          if unquoted_value
            if @double_quote_outside_quote
              unquoted_value = unquoted_value.gsub(@quote_character * 2,
                                                   @quote_character)
              if quoted_value.empty? # %Q{""...} case
                return @quote_character + unquoted_value
              end
            end
            @quote_character + quoted_value + @quote_character + unquoted_value
          else
            quoted_value
          end
        else
          parse_unquoted_column_value
        end
      elsif @may_quoted
        parse_quoted_column_value ||
          parse_unquoted_column_value
      else
        parse_unquoted_column_value ||
          parse_quoted_column_value
      end
    end

    def parse_unquoted_column_value
      value = @scanner.scan_all(@unquoted_value)
      return nil unless value

      @unquoted_column_value = true
      if @first_column_separators
        while true
          @scanner.keep_start
          is_column_end = @column_ends.all? do |column_end|
            @scanner.scan(column_end)
          end
          @scanner.keep_back
          break if is_column_end
          sub_separator = @scanner.scan_all(@first_column_separators)
          break if sub_separator.nil?
          value << sub_separator
          sub_value = @scanner.scan_all(@unquoted_value)
          break if sub_value.nil?
          value << sub_value
        end
      end
      value.gsub!(@backslash_quote_character, @quote_character) if @backslash_quote
      if @rstrip_value
        value.gsub!(@rstrip_value, "")
      end
      value
    end

    def parse_quoted_column_value
      quotes = @scanner.scan_all(@quotes)
      return nil unless quotes

      @quoted_column_value = true
      n_quotes = quotes.size
      if (n_quotes % 2).zero?
        quotes[0, (n_quotes - 2) / 2]
      else
        value = quotes[0, (n_quotes - 1) / 2]
        while true
          quoted_value = @scanner.scan_all(@quoted_value)
          value << quoted_value if quoted_value
          if @backslash_quote
            if @scanner.scan(@escaped_backslash)
              if @scanner.scan(@escaped_quote)
                value << @quote_character
              else
                value << @backslash_character
              end
              next
            end
          end

          quotes = @scanner.scan_all(@quotes)
          unless quotes
            ignore_broken_line
            message = "Unclosed quoted field"
            raise MalformedCSVError.new(message, @lineno)
          end
          n_quotes = quotes.size
          if n_quotes == 1
            break
          elsif (n_quotes % 2) == 1
            value << quotes[0, (n_quotes - 1) / 2]
            break
          else
            value << quotes[0, n_quotes / 2]
          end
        end
        value
      end
    end

    def parse_column_end
      return true if @scanner.scan(@column_end)
      return false unless @column_ends

      @scanner.keep_start
      if @column_ends.all? {|column_end| @scanner.scan(column_end)}
        @scanner.keep_drop
        true
      else
        @scanner.keep_back
        false
      end
    end

    def parse_row_end
      return true if @scanner.scan(@row_end)
      return false unless @row_ends
      @scanner.keep_start
      if @row_ends.all? {|row_end| @scanner.scan(row_end)}
        @scanner.keep_drop
        true
      else
        @scanner.keep_back
        false
      end
    end

    def strip_value(value)
      return value unless @strip
      return nil if value.nil?

      case @strip
      when String
        size = value.size
        while value.start_with?(@strip)
          size -= 1
          value = value[1, size]
        end
        while value.end_with?(@strip)
          size -= 1
          value = value[0, size]
        end
      else
        value.strip!
      end
      value
    end

    def ignore_broken_line
      @scanner.scan_all(@not_line_end)
      @scanner.scan_all(@line_end)
      @lineno += 1
    end

    def start_row
      if @last_line
        @last_line = nil
      else
        @scanner.keep_drop
      end
      @scanner.keep_start
    end

    def emit_row(row, &block)
      @lineno += 1

      raw_row = row
      if @use_headers
        if @headers.nil?
          @headers = adjust_headers(row)
          return unless @return_headers
          row = Row.new(@headers, row, true)
        else
          row = Row.new(@headers,
                        @fields_converter.convert(raw_row, @headers, @lineno))
        end
      else
        # convert fields, if needed...
        row = @fields_converter.convert(raw_row, nil, @lineno)
      end

      # inject unconverted fields and accessor, if requested...
      if @unconverted_fields and not row.respond_to?(:unconverted_fields)
        add_unconverted_fields(row, raw_row)
      end

      yield(row)
    end

    # This method injects an instance variable <tt>unconverted_fields</tt> into
    # +row+ and an accessor method for +row+ called unconverted_fields().  The
    # variable is set to the contents of +fields+.
    def add_unconverted_fields(row, fields)
      class << row
        attr_reader :unconverted_fields
      end
      row.instance_variable_set(:@unconverted_fields, fields)
      row
    end
  end
end