File Coverage

File:lib/CheckSpelling/UnknownWordSplitter.pm
Coverage:87.9%

linestmtbrancondsubtimecode
1#! -*-perl-*-
2
3# ~/bin/w
4# Search for potentially misspelled words
5# Output is:
6# misspellled
7# woord (WOORD, Woord, woord, woord's)
8package CheckSpelling::UnknownWordSplitter;
9
10
1
1
1
97609
2
use 5.022;
11
1
1
1
1
1
1
55
use feature 'unicode_strings';
12
1
1
1
1
2
0
7
use strict;
13
1
1
1
1
1
1
19
use warnings;
14
1
1
1
1
1
1
12
no warnings qw(experimental::vlb);
15
1
1
1
1
1
1
3
use utf8;
16
1
1
1
1
11
0
28
use Encode qw/decode_utf8 encode FB_DEFAULT/;
17
1
1
1
1
2
1
23
use File::Basename;
18
1
1
1
1
2
1
15
use Cwd 'abs_path';
19
1
1
1
1
1
1
15
use File::Temp qw/ tempfile tempdir /;
20
1
1
1
1
250
1
2904
use CheckSpelling::Util;
21our $VERSION='0.1.0';
22
23my ($longest_word, $shortest_word, $word_match, $forbidden_re, $patterns_re, $candidates_re, $disable_word_collating, $check_file_names);
24my ($ignore_pattern, $upper_pattern, $lower_pattern, $not_lower_pattern, $not_upper_or_lower_pattern, $punctuation_pattern);
25my ($shortest, $longest) = (255, 0);
26my @forbidden_re_list;
27my %forbidden_re_descriptions;
28my @candidates_re_list;
29my $hunspell_dictionary_path;
30my @hunspell_dictionaries;
31my %dictionary = ();
32my $base_dict;
33my %unique;
34my %unique_unrecognized;
35my ($last_file, $words, $unrecognized) = ('', 0, 0);
36
37my $disable_flags;
38
39sub test_re {
40
14
14
9
  my ($expression) = @_;
41
14
14
8
105
  return eval { qr /$expression/ };
42}
43
44sub quote_re {
45
14
14
7
  my ($expression) = @_;
46
14
50
15
  return $expression if $expression =~ /\?\{/;
47
14
31
  $expression =~ s/
48   \G
49   (
50      (?:[^\\]|\\[^Q])*
51   )
52   (?:
53      \\Q
54      (?:[^\\]|\\[^E])*
55      (?:\\E)?
56   )?
57/
58
28
50
46
   $1 . (defined($2) ? quotemeta($2) : '')
59/xge;
60
14
14
  return $expression;
61}
62
63sub file_to_lists {
64
3
3
4
  my ($re) = @_;
65
3
3
  my @patterns;
66  my %hints;
67
3
0
  my $fh;
68
3
50
32
  if (open($fh, '<:utf8', $re)) {
69
3
5
    local $/=undef;
70
3
15
    my $file=<$fh>;
71
3
7
    close $fh;
72
3
3
    my $line_number = 0;
73
3
1
    my $hint = '';
74
3
14
    for (split /\R/, $file) {
75
17
10
      ++$line_number;
76
17
12
      chomp;
77
17
100
18
      if (/^#(?:\s(.+)|)/) {
78
6
100
67
15
        $hint = $1 if ($hint eq '' && defined $1);
79
6
4
        next;
80      }
81
11
100
11
      $hint = '' unless $_ ne '';
82
11
9
      my $pattern = $_;
83
11
100
27
      next unless s/^(.+)/(?:$1)/;
84
7
10
      my $quoted = quote_re($1);
85
7
100
8
      unless (test_re $quoted) {
86
1
2
        my $error = $@;
87
1
46
        my $home = dirname(__FILE__);
88
1
31
        $error =~ s/$home.*?\.pm line \d+\./$re line $line_number (bad-regex)/;
89
1
9
        print STDERR $error;
90
1
1
        $_ = '(?:\$^ - skipped because bad-regex)';
91
1
1
        $hint = '';
92      }
93
7
100
12
      if (defined $hints{$_}) {
94
1
2
        my $pattern_length = length $pattern;
95
1
1
        my $wrapped = CheckSpelling::Util::wrap_in_backticks($pattern);
96
1
11
        print STDERR "$re:$line_number:1 ... $pattern_length, Warning - duplicate pattern: $wrapped (duplicate-pattern)\n";
97
1
1
        $_ = '(?:\$^ - skipped because duplicate-pattern on $line_number)';
98      } else {
99
6
6
        push @patterns, $_;
100
6
10
        $hints{$_} = $hint;
101      }
102
7
11
      $hint = '';
103    }
104  }
105
106  return {
107
3
12
    patterns => \@patterns,
108    hints => \%hints,
109  };
110}
111
112sub file_to_list {
113
2
2
1293
  my ($re) = @_;
114
2
4
  my $lists = file_to_lists($re);
115
116
2
2
2
7
  return @{$lists->{'patterns'}};
117}
118
119sub list_to_re {
120
2
2
2
  my (@list) = @_;
121
2
5
5
50
2
5
3
  @list = map { my $quoted = quote_re($_); test_re($quoted) ? $quoted : '' } @list;
122
2
5
3
5
  @list = grep { $_ ne '' } @list;
123
2
50
2
  return '$^' unless scalar @list;
124
2
6
  return join "|", (@list);
125}
126
127sub not_empty {
128
51
51
37
  my ($thing) = @_;
129
51
100
158
  return defined $thing && $thing ne ''
130}
131
132sub valid_word {
133  # shortest_word is an absolute
134
22
22
13
  our ($shortest, $longest, $shortest_word, $longest_word);
135
22
50
22
  $shortest = $shortest_word if $shortest_word;
136
22
100
100
18
  if ($longest_word) {
137    # longest_word is an absolute
138
20
18
    $longest = $longest_word;
139  } elsif (not_empty($longest)) {
140    # we allow for some sloppiness (a couple of stuck keys per word)
141    # it's possible that this should scale with word length
142
1
1
    $longest += 2;
143  }
144
22
15
  our ($upper_pattern, $lower_pattern, $punctuation_pattern);
145
22
66
100
18
124
  my $word_pattern = join '|', (grep { defined $_ && /./ } ($upper_pattern, $lower_pattern, $punctuation_pattern));
146
22
100
18
  $word_pattern = q<\\w|'> unless $word_pattern;
147
22
50
50
37
  if ((defined $shortest && not_empty($longest)) &&
148      ($shortest > $longest)) {
149
0
0
    $word_pattern = "(?:$word_pattern){3}";
150
0
0
    return qr/$word_pattern/;
151  }
152
22
100
26
  $shortest = 3 unless defined $shortest;
153
22
100
24
  $longest = '' unless defined $longest;
154
22
71
  $word_match = "(?:$word_pattern){$shortest,$longest}";
155
22
167
  return qr/\b$word_match\b/;
156}
157
158sub load_dictionary {
159
12
12
1683
  my ($dict) = @_;
160
12
6
  our ($word_match, $longest, $shortest, $longest_word, $shortest_word, %dictionary);
161
12
9
  $longest_word = CheckSpelling::Util::get_val_from_env('INPUT_LONGEST_WORD', undef);
162
12
12
  $shortest_word = CheckSpelling::Util::get_val_from_env('INPUT_SHORTEST_WORD', undef);
163
12
6
  our ($ignore_pattern, $upper_pattern, $lower_pattern, $not_lower_pattern, $not_upper_or_lower_pattern, $punctuation_pattern);
164
12
30
  $ignore_pattern = CheckSpelling::Util::get_file_from_env_utf8('INPUT_IGNORE_PATTERN', q<[^a-zA-Z']>);
165
12
41
  $upper_pattern = CheckSpelling::Util::get_file_from_env_utf8('INPUT_UPPER_PATTERN', '[A-Z]');
166
12
25
  $lower_pattern = CheckSpelling::Util::get_file_from_env_utf8('INPUT_LOWER_PATTERN', '[a-z]');
167
12
22
  $not_lower_pattern = CheckSpelling::Util::get_file_from_env_utf8('INPUT_NOT_LOWER_PATTERN', '[^a-z]');
168
12
21
  $not_upper_or_lower_pattern = CheckSpelling::Util::get_file_from_env_utf8('INPUT_NOT_UPPER_OR_LOWER_PATTERN', '[^A-Za-z]');
169
12
21
  $punctuation_pattern = CheckSpelling::Util::get_file_from_env_utf8('INPUT_PUNCTUATION_PATTERN', q<'>);
170
12
23
  %dictionary = ();
171
172
12
472
  open(DICT, '<:utf8', $dict);
173
12
46
  while (!eof(DICT)) {
174
31
31
    my $word = <DICT>;
175
31
28
    chomp $word;
176
31
100
64
    next unless $word =~ $word_match;
177
28
24
    my $l = length $word;
178
28
100
17
    $longest = -1 unless not_empty($longest);
179
28
100
24
    $longest = $l if $l > $longest;
180
28
100
23
    $shortest = $l if $l < $shortest;
181
28
65
    $dictionary{$word}=1;
182  }
183
12
19
  close DICT;
184
185
12
8
  $word_match = valid_word();
186}
187
188sub hunspell_dictionary {
189
3
3
3
  my ($dict) = @_;
190
3
4
  my $name = $dict;
191
3
6
  $name =~ s{/src/index/hunspell/index\.dic$}{};
192
3
9
  $name =~ s{.*/}{};
193
3
3
  my $aff = $dict;
194
3
3
  my $encoding;
195
3
7
  $aff =~ s/\.dic$/.aff/;
196
3
50
20
  if (open AFF, '<', $aff) {
197
3
19
    while (<AFF>) {
198
0
0
0
      next unless /^SET\s+(\S+)/;
199
0
0
0
      $encoding = $1 if ($1 !~ /utf-8/i);
200
0
0
      last;
201    }
202
3
6
    close AFF;
203  }
204  return {
205
3
291
    name => $name,
206    dict => $dict,
207    aff => $aff,
208    encoding => $encoding,
209    engine => Text::Hunspell->new($aff, $dict),
210  }
211}
212
213sub init {
214
9
9
9597
  my ($configuration) = @_;
215
9
9
  our ($word_match, %unique, $patterns_re, @forbidden_re_list, $forbidden_re, @candidates_re_list, $candidates_re);
216
9
20
  our $sandbox = CheckSpelling::Util::get_file_from_env('sandbox', '');
217
9
9
  our $hunspell_dictionary_path = CheckSpelling::Util::get_file_from_env('hunspell_dictionary_path', '');
218
9
18
  our $timeout = CheckSpelling::Util::get_val_from_env('splitter_timeout', 30);
219
9
5
  our %forbidden_re_descriptions;
220
9
100
12
  if ($hunspell_dictionary_path) {
221
3
32
    our @hunspell_dictionaries = ();
222
1
1
1
1
1
1
1
1
1
3
50
1
1
1
168
751
21
4
1
15
6
1
16
172
    if (eval 'use Text::Hunspell; 1') {
223
3
94
      my @hunspell_dictionaries_list = glob("$hunspell_dictionary_path/*.dic");
224
3
5
      for my $hunspell_dictionary_file (@hunspell_dictionaries_list) {
225
3
7
        push @hunspell_dictionaries, hunspell_dictionary($hunspell_dictionary_file);
226      }
227    } else {
228
0
0
      print STDERR "Could not load Text::Hunspell for dictionaries (hunspell-unavailable)\n";
229    }
230  }
231
9
15
  my (@patterns_re_list, %in_patterns_re_list);
232
9
50
41
  if (-e "$configuration/patterns.txt") {
233
0
0
    @patterns_re_list = file_to_list "$configuration/patterns.txt";
234
0
0
    $patterns_re = list_to_re @patterns_re_list;
235
0
0
0
0
    %in_patterns_re_list = map {$_ => 1} @patterns_re_list;
236  } else {
237
9
8
    $patterns_re = undef;
238  }
239
240
9
100
25
  if (-e "$configuration/forbidden.txt") {
241
1
2
    my $forbidden_re_info = file_to_lists "$configuration/forbidden.txt";
242
1
1
1
2
    @forbidden_re_list = @{$forbidden_re_info->{'patterns'}};
243
1
1
1
2
    %forbidden_re_descriptions = %{$forbidden_re_info->{'hints'}};
244
1
2
    $forbidden_re = list_to_re @forbidden_re_list;
245  } else {
246
8
5
    $forbidden_re = undef;
247  }
248
249
9
100
22
  if (-e "$configuration/candidates.txt") {
250
1
2
    @candidates_re_list = file_to_list "$configuration/candidates.txt";
251
1
2
2
50
33
2
2
6
    @candidates_re_list = map { my $quoted = quote_re($_); $in_patterns_re_list{$_} || !test_re($quoted) ? '' : $quoted } @candidates_re_list;
252
1
1
    $candidates_re = list_to_re @candidates_re_list;
253  } else {
254
8
6
    $candidates_re = undef;
255  }
256
257
9
11
  our $largest_file = CheckSpelling::Util::get_val_from_env('INPUT_LARGEST_FILE', 1024*1024);
258
259
9
24
  my $disable_flags = CheckSpelling::Util::get_file_from_env('INPUT_DISABLE_CHECKS', '');
260
9
10
  our $disable_word_collating = $disable_flags =~ /(?:^|,|\s)word-collating(?:,|\s|$)/;
261
9
5
  our $disable_minified_file = $disable_flags =~ /(?:^|,|\s)minified-file(?:,|\s|$)/;
262
9
9
  our $disable_single_line_file = $disable_flags =~ /(?:^|,|\s)single-line-file(?:,|\s|$)/;
263
264
9
4
  our $check_file_names = CheckSpelling::Util::get_file_from_env('check_file_names', '');
265
266
9
7
  our $use_magic_file = CheckSpelling::Util::get_val_from_env('INPUT_USE_MAGIC_FILE', '');
267
268
9
9
  $word_match = valid_word();
269
270
9
14
  our $base_dict = CheckSpelling::Util::get_file_from_env('dict', "$configuration/words");
271
9
100
26
  $base_dict = '/usr/share/dict/words' unless -e $base_dict;
272
9
8
  load_dictionary($base_dict);
273}
274
275sub split_line {
276
1164
1164
471
  our (%dictionary, $word_match, $disable_word_collating);
277
1164
447
  our ($ignore_pattern, $upper_pattern, $lower_pattern, $not_lower_pattern, $not_upper_or_lower_pattern, $punctuation_pattern);
278
1164
394
  our @hunspell_dictionaries;
279
1164
427
  our $shortest;
280
1164
474
  my $pattern = '.';
281  # $pattern = "(?:$upper_pattern){$shortest,}|$upper_pattern(?:$lower_pattern){2,}\n";
282
283  # https://www.fileformat.info/info/unicode/char/2019/
284
1164
455
  my $rsqm = "\xE2\x80\x99";
285
286
1164
599
  my ($words, $unrecognized) = (0, 0);
287
1164
757
  my ($line, $unique_ref, $unique_unrecognized_ref, $unrecognized_line_items_ref) = @_;
288
1164
5279
    $line =~ s/(?:$rsqm|&apos;|&#39;|\%27|&#8217;|&#x2019;|&rsquo;|\\u2019|\x{2019}|')+/'/g;
289
1164
2327
    $line =~ s/(?:$ignore_pattern)+/ /g;
290
1164
1437
    while ($line =~ s/($upper_pattern{2,})($upper_pattern$lower_pattern{2,})/ $1 $2 /g) {}
291
1164
3838
    while ($line =~ s/((?:$lower_pattern|$punctuation_pattern)+)($upper_pattern)/$1 $2/g) {}
292
1164
1339
    for my $token (split /\s+/, $line) {
293
5871
100
5030
      next unless $token =~ /$pattern/;
294
4708
3418
      $token =~ s/^(?:'|$rsqm)+//g;
295
4708
4485
      $token =~ s/(?:'|$rsqm)+s?$//g;
296
4708
2537
      my $raw_token = $token;
297
4708
2795
      $token =~ s/^[^Ii]?'+(.*)/$1/;
298
4708
2347
      $token =~ s/(.*?)'+$/$1/;
299
4708
100
6161
      next unless $token =~ $word_match;
300
4561
100
3658
      if (defined $dictionary{$token}) {
301
1032
449
        ++$words;
302
1032
520
        $unique_ref->{$token}=1;
303
1032
913
        next;
304      }
305
3529
100
2479
      if (@hunspell_dictionaries) {
306
3504
1547
        my $found = 0;
307
3504
1753
        for my $hunspell_dictionary (@hunspell_dictionaries) {
308          my $token_encoded = defined $hunspell_dictionary->{'encoding'} ?
309
3504
50
2310
            encode($hunspell_dictionary->{'encoding'}, $token) : $token;
310
3504
50
5439
          next unless ($hunspell_dictionary->{'engine'}->check($token_encoded));
311
0
0
          ++$words;
312
0
0
          $dictionary{$token} = 1;
313
0
0
          $unique_ref->{$token}=1;
314
0
0
          $found = 1;
315
0
0
          last;
316        }
317
3504
50
2286
        next if $found;
318      }
319
3529
2162
      my $key = lc $token;
320
3529
100
2895
      if (defined $dictionary{$key}) {
321
6
3
        ++$words;
322
6
4
        $unique_ref->{$key}=1;
323
6
7
        next;
324      }
325
3523
50
2260
      unless ($disable_word_collating) {
326
3523
1757
        $key =~ s/''+/'/g;
327
3523
1760
        $key =~ s/'[sd]$//;
328      }
329
3523
50
2383
      if (defined $dictionary{$key}) {
330
0
0
        ++$words;
331
0
0
        $unique_ref->{$key}=1;
332
0
0
        next;
333      }
334
3523
1561
      ++$unrecognized;
335
3523
1806
      $unique_unrecognized_ref->{$raw_token}=1;
336
3523
4108
      $unrecognized_line_items_ref->{$raw_token}=1;
337    }
338
1164
1448
    return ($words, $unrecognized);
339}
340
341sub skip_file {
342
15
15
47
  my ($temp_dir, $reason) = @_;
343
15
399
  open(SKIPPED, '>:utf8', "$temp_dir/skipped");
344
15
77
  print SKIPPED $reason;
345
15
312
  close SKIPPED;
346}
347
348sub split_file {
349
15
15
9544
  my ($file) = @_;
350  our (
351
15
12
    $unrecognized, $shortest, $largest_file, $words,
352    $word_match, %unique, %unique_unrecognized, $forbidden_re,
353    @forbidden_re_list, $patterns_re, %dictionary,
354    $candidates_re, @candidates_re_list, $check_file_names, $use_magic_file, $disable_minified_file,
355    $disable_single_line_file,
356    $sandbox,
357  );
358
15
11
  our %forbidden_re_descriptions;
359
15
7
  our ($ignore_pattern, $upper_pattern, $lower_pattern, $not_lower_pattern, $not_upper_or_lower_pattern, $punctuation_pattern);
360
361  # https://www.fileformat.info/info/unicode/char/2019/
362
15
9
  my $rsqm = "\xE2\x80\x99";
363
364
15
20
  my @candidates_re_hits = (0) x scalar @candidates_re_list;
365
15
10
  my @candidates_re_lines = (0) x scalar @candidates_re_list;
366
15
19
  my @forbidden_re_hits = (0) x scalar @forbidden_re_list;
367
15
11
  my @forbidden_re_lines = (0) x scalar @forbidden_re_list;
368
15
34
  my $temp_dir = tempdir(DIR=>$sandbox);
369
15
50
2020
  print STDERR "checking file: $file\n" if defined $ENV{'DEBUG'};
370
15
205
  open(NAME, '>', "$temp_dir/name");
371
15
31
    print NAME $file;
372
15
108
  close NAME;
373
15
42
  my $file_size = -s $file;
374
15
50
18
  if (defined $largest_file) {
375
15
50
17
    unless ($check_file_names eq $file) {
376
15
100
20
      if ($file_size > $largest_file) {
377
1
2
        skip_file($temp_dir, "size `$file_size` exceeds limit `$largest_file` (large-file)\n");
378
1
3
        return $temp_dir;
379      }
380    }
381  }
382
14
100
15
  if ($use_magic_file) {
383
8
50
10704
    if (open(my $file_fh, '-|',
384              '/usr/bin/file',
385              '-b',
386              '--mime',
387              '-e', 'cdf',
388              '-e', 'compress',
389              '-e', 'csv',
390              '-e', 'elf',
391              '-e', 'json',
392              '-e', 'tar',
393              $file)) {
394
8
26313
      my $file_kind = <$file_fh>;
395
8
3975
      close $file_fh;
396
8
100
119
      if ($file_kind =~ /^(.*?); charset=binary/) {
397
2
23
        skip_file($temp_dir, "it appears to be a binary file (`$1`) (binary-file)\n");
398
2
44
        return $temp_dir;
399      }
400    }
401  }
402
12
91
  open FILE, '<', $file;
403
12
14
  binmode FILE;
404
12
6
  my $head;
405
12
79
  read(FILE, $head, 4096);
406
12
763
  $head =~ s/(?:\r|\n)+$//;
407
12
59
  my $dos_new_lines = () = $head =~ /\r\n/gi;
408
12
34
  my $unix_new_lines = () = $head =~ /\n/gi;
409
12
110
  my $mac_new_lines = () = $head =~ /\r/gi;
410
12
53
  local $/;
411
12
100
100
100
100
100
84
  if ($unix_new_lines == 0 && $mac_new_lines == 0) {
412
3
7
    $/ = "\n";
413  } elsif ($dos_new_lines >= $unix_new_lines && $dos_new_lines >= $mac_new_lines) {
414
1
5
    $/ = "\r\n";
415  } elsif ($mac_new_lines > $unix_new_lines) {
416
2
8
    $/ = "\r";
417  } else {
418
6
7
    $/ = "\n";
419  }
420
12
21
  seek(FILE, 0, 0);
421
12
20
  ($words, $unrecognized) = (0, 0);
422
12
38
  %unique = ();
423
12
29
  %unique_unrecognized = ();
424
425  local $SIG{__WARN__} = sub {
426
0
0
0
    my $message = shift;
427
0
0
    $message =~ s/> line/> in $file - line/;
428
0
0
    chomp $message;
429
0
0
    print STDERR "$message\n";
430
12
115
  };
431
432
12
210
  open(WARNINGS, '>:utf8', "$temp_dir/warnings");
433
12
10
  our $timeout;
434
12
18
  eval {
435
12
0
0
82
0
    local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
436
12
35
    alarm $timeout;
437
438
12
9
    my $offset = 0;
439
12
73
    while (<FILE>) {
440
1167
100
952
      if ($. == 1) {
441
12
50
16
        unless ($disable_minified_file) {
442
12
100
100
56
          if ($file_size >= 512 && length($_) == $file_size) {
443
1
8
            skip_file($temp_dir, "file only has a single line (single-line-file)\n");
444
1
4
            last;
445          }
446        }
447      }
448
1166
1857
      $_ = decode_utf8($_, FB_DEFAULT);
449
1166
50
2336
      if (/[\x{D800}-\x{DFFF}]/) {
450
0
0
        skip_file($temp_dir, "file contains a UTF-16 surrogate -- UTF-16 surrogates are not supported (utf16-surrogate-file)\n");
451
0
0
        last;
452      }
453
1166
1223
      s/\R$//;
454
1166
100
968
      s/^\x{FEFF}// if $. == 1;
455
1166
100
1008
      next unless /./;
456
1164
674
      my $raw_line = $_;
457
458      # hook for custom line based text exclusions:
459
1164
100
684
      if (defined $patterns_re) {
460
2
6
11
7
        s/($patterns_re)/"="x length($1)/ge;
461      }
462
1164
670
      my $initial_line_state = $_;
463
1164
612
      my $previous_line_state = $_;
464
1164
530
      my $line_flagged;
465
1164
100
703
      if ($forbidden_re) {
466
9
5
58
11
        while (s/($forbidden_re)/"="x length($1)/e) {
467
5
3
          $line_flagged = 1;
468
5
9
          my ($begin, $end, $match) = ($-[0] + 1, $+[0] + 1, $1);
469
5
3
          my $found_trigger_re;
470
5
7
          for my $i (0 .. $#forbidden_re_list) {
471
7
5
            my $forbidden_re_singleton = $forbidden_re_list[$i];
472
7
3
            my $test_line = $previous_line_state;
473
7
4
100
54
6
            if ($test_line =~ s/($forbidden_re_singleton)/"="x length($1)/e) {
474
4
50
5
              next unless $test_line eq $_;
475
4
6
              my ($begin_test, $end_test, $match_test) = ($-[0] + 1, $+[0] + 1, $1);
476
4
50
4
              next unless $begin == $begin_test;
477
4
50
2
              next unless $end == $end_test;
478
4
50
3
              next unless $match eq $match_test;
479
4
3
              $found_trigger_re = $forbidden_re_singleton;
480
4
6
              my $hit = "$.:$begin:$end";
481
4
5
              $forbidden_re_hits[$i]++;
482
4
100
4
              $forbidden_re_lines[$i] = $hit unless $forbidden_re_lines[$i];
483
4
5
              last;
484            }
485          }
486
5
6
          my $wrapped = CheckSpelling::Util::wrap_in_backticks($match);
487
5
100
5
          if ($found_trigger_re) {
488
4
100
7
            my $description = $forbidden_re_descriptions{$found_trigger_re} || '';
489
4
8
            $found_trigger_re =~ s/^\(\?:(.*)\)$/$1/;
490
4
4
            my $quoted_trigger_re = CheckSpelling::Util::truncate_with_ellipsis(CheckSpelling::Util::wrap_in_backticks($found_trigger_re), 99);
491
4
100
4
            if ($description ne '') {
492
3
13
              print WARNINGS ":$.:$begin ... $end, Warning - $wrapped matches a line_forbidden.patterns rule: $description - $quoted_trigger_re (forbidden-pattern)\n";
493            } else {
494
1
6
              print WARNINGS ":$.:$begin ... $end, Warning - $wrapped matches a line_forbidden.patterns entry: $quoted_trigger_re (forbidden-pattern)\n";
495            }
496          } else {
497
1
4
            print WARNINGS ":$.:$begin ... $end, Warning - $wrapped matches a line_forbidden.patterns entry (forbidden-pattern)\n";
498          }
499
5
24
          $previous_line_state = $_;
500        }
501
9
9
        $_ = $initial_line_state;
502      }
503      # This is to make it easier to deal w/ rules:
504
1164
952
      s/^/ /;
505
1164
646
      my %unrecognized_line_items = ();
506
1164
738
      my ($new_words, $new_unrecognized) = split_line($_, \%unique, \%unique_unrecognized, \%unrecognized_line_items);
507
1164
556
      $words += $new_words;
508
1164
482
      $unrecognized += $new_unrecognized;
509
1164
704
      my $line_length = length($raw_line);
510
1164
1190
      for my $token (sort CheckSpelling::Util::case_biased keys %unrecognized_line_items) {
511
1029
493
        my $found_token = 0;
512
1029
461
        my $raw_token = $token;
513
1029
532
        $token =~ s/'/(?:'|\x{2019}|\&apos;|\&#39;)+/g;
514
1029
447
        my $before;
515
1029
100
50
1273
        if ($token =~ /^$upper_pattern$lower_pattern/) {
516
5
4
          $before = '(?<=.)';
517        } elsif ($token =~ /^$upper_pattern/) {
518
0
0
          $before = "(?<!$upper_pattern)";
519        } else {
520
1024
526
          $before = "(?<=$not_lower_pattern)";
521        }
522
1029
50
988
        my $after = ($token =~ /$upper_pattern$/) ? "(?=$not_upper_or_lower_pattern)|(?=$upper_pattern$lower_pattern)" : "(?=$not_lower_pattern)";
523
1029
1698
        while ($raw_line =~ /(?:\b|$before)($token)(?:\b|$after)/g) {
524
3520
1493
          $line_flagged = 1;
525
3520
1541
          $found_token = 1;
526
3520
4541
          my ($begin, $end, $match) = ($-[0] + 1, $+[0] + 1, $1);
527
3520
50
2952
          next unless $match =~ /./;
528
3520
1892
          my $wrapped = CheckSpelling::Util::wrap_in_backticks($match);
529
3520
11387
          print WARNINGS ":$.:$begin ... $end: $wrapped\n";
530        }
531
1029
100
1026
        unless ($found_token) {
532
3
50
33
26
          if ($raw_line !~ /$token.*$token/ && $raw_line =~ /($token)/) {
533
3
7
            my ($begin, $end, $match) = ($-[0] + 1, $+[0] + 1, $1);
534
3
3
            my $wrapped = CheckSpelling::Util::wrap_in_backticks($raw_token);
535
3
12
            print WARNINGS ":$.:$begin ... $end: $wrapped\n";
536          } else {
537
0
0
            my $offset = $line_length + 1;
538
0
0
            my $wrapped = CheckSpelling::Util::wrap_in_backticks($raw_token);
539
0
0
            print WARNINGS ":$.:1 ... $offset, Warning - Could not identify whole word $wrapped in line (token-is-substring)\n";
540          }
541        }
542      }
543
1164
100
100
1645
      if ($line_flagged && $candidates_re) {
544
1
1
        $_ = $previous_line_state = $initial_line_state;
545
1
1
19
2
        s/($candidates_re)/"="x length($1)/ge;
546
1
50
2
        if ($_ ne $initial_line_state) {
547
1
1
          $_ = $previous_line_state;
548
1
2
          for my $i (0 .. $#candidates_re_list) {
549
2
2
            my $candidate_re = $candidates_re_list[$i];
550
2
100
67
16
            next unless $candidate_re =~ /./ && $raw_line =~ /$candidate_re/;
551
1
1
50
6
2
            if (($_ =~ s/($candidate_re)/"="x length($1)/e)) {
552
1
2
              my ($begin, $end) = ($-[0] + 1, $+[0] + 1);
553
1
2
              my $hit = "$.:$begin:$end";
554
1
2
              $_ = $previous_line_state;
555
1
1
5
1
              my $replacements = ($_ =~ s/($candidate_re)/"="x length($1)/ge);
556
1
2
              $candidates_re_hits[$i] += $replacements;
557
1
50
2
              $candidates_re_lines[$i] = $hit unless $candidates_re_lines[$i];
558
1
2
              $_ = $previous_line_state;
559            }
560          }
561        }
562      }
563
1164
50
740
      unless ($disable_minified_file) {
564
1164
787
        s/={3,}//g;
565
1164
775
        $offset += length;
566
1164
1005
        my $ratio = int($offset / $.);
567
1164
536
        my $ratio_threshold = 1000;
568
1164
100
2666
        if ($ratio > $ratio_threshold) {
569
11
25
          skip_file($temp_dir, "average line width ($ratio) exceeds the threshold ($ratio_threshold) (minified-file)\n");
570        }
571      }
572    }
573
574
12
72
    alarm 0;
575  };
576
12
50
12
  if ($@) {
577
0
0
0
    die unless $@ eq "alarm\n";
578
0
0
    print WARNINGS ":$.:1 ... 1, Warning - Could not parse file within time limit (slow-file)\n";
579
0
0
    skip_file($temp_dir, "it could not be parsed file within time limit (slow-file)\n");
580  }
581
582
12
34
  close FILE;
583
12
99
  close WARNINGS;
584
585
12
100
75
33
  if ($unrecognized || @candidates_re_hits || @forbidden_re_hits) {
586
11
235
    open(STATS, '>:utf8', "$temp_dir/stats");
587
11
100
100
100
100
181
      print STATS "{words: $words, unrecognized: $unrecognized, unknown: ".(keys %unique_unrecognized).
588      ", unique: ".(keys %unique).
589      (@candidates_re_hits ? ", candidates: [".(join ',', @candidates_re_hits)."]" : "").
590      (@candidates_re_lines ? ", candidate_lines: [".(join ',', @candidates_re_lines)."]" : "").
591      (@forbidden_re_hits ? ", forbidden: [".(join ',', @forbidden_re_hits)."]" : "").
592      (@forbidden_re_lines ? ", forbidden_lines: [".(join ',', @forbidden_re_lines)."]" : "").
593      "}";
594
11
90
    close STATS;
595
11
119
    open(UNKNOWN, '>:utf8', "$temp_dir/unknown");
596
11
19
36
27
      print UNKNOWN map { "$_\n" } sort CheckSpelling::Util::case_biased keys %unique_unrecognized;
597
11
49
    close UNKNOWN;
598  }
599
600
12
128
  return $temp_dir;
601}
602
603sub main {
604
2
2
365
  my ($configuration, @ARGV) = @_;
605
2
1
  our %dictionary;
606
2
100
3
  unless (%dictionary) {
607
1
4
    init($configuration);
608  }
609
610  # read all input
611
2
1
  my @reports;
612
613
2
2
  for my $file (@ARGV) {
614
2
2
    my $temp_dir = split_file($file);
615
2
4
    push @reports, "$temp_dir\n";
616  }
617
2
8
  print join '', @reports;
618}
619
6201;