biblicit 2.0.3 → 2.0.4
Sign up to get free protection for your applications and to get access to all the features.
- data/README.md +0 -2
- data/biblicit.gemspec +1 -1
- data/parscit/bin/citeExtract.pl +9 -161
- data/parscit/bin/sectExtract.pl +0 -14
- data/parscit/lib/ParsCit/Controller.pm +0 -59
- data/parscit/lib/ParsCit/PreProcess.pm +0 -4
- data/parscit/lib/ParsCit/Tr2crfpp.pm +1 -7
- metadata +4 -24
- data/parscit/bin/sectLabel/processOmniXML.pl +0 -1427
- data/parscit/bin/sectLabel/processOmniXML_new.pl +0 -1025
- data/parscit/bin/sectLabel/processOmniXMLv2.pl +0 -1529
- data/parscit/bin/sectLabel/processOmniXMLv3.pl +0 -964
- data/parscit/bin/sectLabel/simplifyOmniXML.pl +0 -382
- data/parscit/bin/xml2train.pl +0 -193
- data/parscit/lib/Omni/Config.pm +0 -93
- data/parscit/lib/Omni/Omnicell.pm +0 -263
- data/parscit/lib/Omni/Omnicol.pm +0 -292
- data/parscit/lib/Omni/Omnidd.pm +0 -328
- data/parscit/lib/Omni/Omnidoc.pm +0 -153
- data/parscit/lib/Omni/Omniframe.pm +0 -223
- data/parscit/lib/Omni/Omniline.pm +0 -423
- data/parscit/lib/Omni/Omnipage.pm +0 -282
- data/parscit/lib/Omni/Omnipara.pm +0 -232
- data/parscit/lib/Omni/Omnirun.pm +0 -303
- data/parscit/lib/Omni/Omnitable.pm +0 -336
- data/parscit/lib/Omni/Omniword.pm +0 -162
- data/parscit/lib/Omni/Traversal.pm +0 -313
- data/parscit/lib/SectLabel/AAMatching.pm +0 -1949
@@ -1,1025 +0,0 @@
|
|
1
|
-
#!/usr/bin/perl -wT
|
2
|
-
# Author: Luong Minh Thang <luongmin@comp.nus.edu.sg>, generated at Tue, 02 Jun 2009 01:30:42
|
3
|
-
|
4
|
-
# Modified from template by Min-Yen Kan <kanmy@comp.nus.edu.sg>
|
5
|
-
|
6
|
-
require 5.0;
|
7
|
-
use strict;
|
8
|
-
use Getopt::Long;
|
9
|
-
use HTML::Entities;
|
10
|
-
|
11
|
-
# I do not know a better solution to find a lib path in -T mode.
|
12
|
-
# So if you know a better solution, I'd be glad to hear.
|
13
|
-
# See this http://www.perlmonks.org/?node_id=585299 for why I used the below code
|
14
|
-
use FindBin;
|
15
|
-
FindBin::again(); # to get correct path in case 2 scripts in different directories use FindBin
|
16
|
-
my $path;
|
17
|
-
BEGIN {
|
18
|
-
if ($FindBin::Bin =~ /(.*)/) {
|
19
|
-
$path = $1;
|
20
|
-
}
|
21
|
-
}
|
22
|
-
use lib "$path/../../lib";
|
23
|
-
use SectLabel::PreProcess;
|
24
|
-
|
25
|
-
### USER customizable section
|
26
|
-
$0 =~ /([^\/]+)$/; my $progname = $1;
|
27
|
-
my $outputVersion = "1.0";
|
28
|
-
### END user customizable section
|
29
|
-
|
30
|
-
sub License {
|
31
|
-
print STDERR "# Copyright 2009 \251 by Luong Minh Thang\n";
|
32
|
-
}
|
33
|
-
|
34
|
-
### HELP Sub-procedure
|
35
|
-
sub Help {
|
36
|
-
print STDERR "Process Omnipage XML output (concatenated results fromm all pages of a PDF file), and extract text lines together with other XML infos\n";
|
37
|
-
print STDERR "usage: $progname -h\t[invokes help]\n";
|
38
|
-
print STDERR " $progname -in xmlFile -out outFile [-xmlFeature -decode -markup -para] [-tag tagFile -allowEmptyLine -log]\n";
|
39
|
-
print STDERR "Options:\n";
|
40
|
-
print STDERR "\t-q\tQuiet Mode (don't echo license)\n";
|
41
|
-
print STDERR "\t-xmlFeature: append XML feature together with text extracted\n";
|
42
|
-
print STDERR "\t-decode: decode HTML entities and then output, to avoid double entity encoding later\n";
|
43
|
-
print STDERR "\t-para: marking in the output each paragraph with # Para lineId numLines\n";
|
44
|
-
print STDERR "\t-markup: marking in the output detailed word-level info ### Page w h\\n## Para l t r b\\n# Line l t r b\\nword l t r b\n";
|
45
|
-
|
46
|
-
print STDERR "\t-tag tagFile: count XML tags/values for statistics purpose\n";
|
47
|
-
}
|
48
|
-
my $QUIET = 0;
|
49
|
-
my $HELP = 0;
|
50
|
-
my $outFile = undef;
|
51
|
-
my $inFile = undef;
|
52
|
-
|
53
|
-
my $isXmlFeature = 0;
|
54
|
-
my $isDecode = 0;
|
55
|
-
|
56
|
-
my $isMarkup = 0;
|
57
|
-
my $isParaDelimiter = 0;
|
58
|
-
|
59
|
-
my $tagFile = "";
|
60
|
-
my $isAllowEmpty = 0;
|
61
|
-
my $isDebug = 0;
|
62
|
-
$HELP = 1 unless GetOptions('in=s' => \$inFile,
|
63
|
-
'out=s' => \$outFile,
|
64
|
-
'decode' => \$isDecode,
|
65
|
-
'xmlFeature' => \$isXmlFeature,
|
66
|
-
|
67
|
-
'tag=s' => \$tagFile,
|
68
|
-
'allowEmptyLine' => \$isAllowEmpty,
|
69
|
-
'markup' => \$isMarkup,
|
70
|
-
|
71
|
-
'para' => \$isParaDelimiter,
|
72
|
-
'log' => \$isDebug,
|
73
|
-
'h' => \$HELP,
|
74
|
-
'q' => \$QUIET);
|
75
|
-
|
76
|
-
if ($HELP || !defined $inFile || !defined $outFile) {
|
77
|
-
Help();
|
78
|
-
exit(0);
|
79
|
-
}
|
80
|
-
|
81
|
-
if (!$QUIET) {
|
82
|
-
License();
|
83
|
-
}
|
84
|
-
|
85
|
-
### Untaint ###
|
86
|
-
$inFile = untaintPath($inFile);
|
87
|
-
$outFile = untaintPath($outFile);
|
88
|
-
$tagFile = untaintPath($tagFile);
|
89
|
-
$ENV{'PATH'} = '/bin:/usr/bin:/usr/local/bin';
|
90
|
-
### End untaint ###
|
91
|
-
|
92
|
-
### Mark page, para, line, word
|
93
|
-
my %gPageHash = ();
|
94
|
-
|
95
|
-
### Mark paragraph
|
96
|
-
my @gPara = ();
|
97
|
-
|
98
|
-
### XML features ###
|
99
|
-
# locFeature
|
100
|
-
my @gPosHash = (); my $gMinPos = 1000000; my $gMaxPos = 0;
|
101
|
-
my @gAlign = (); # alignFeature
|
102
|
-
my @gBold = (); # bold feature
|
103
|
-
my @gItalic = (); # italic feature
|
104
|
-
|
105
|
-
# font size feature
|
106
|
-
my %gFontSizeHash = (); my @gFontSize = ();
|
107
|
-
# font face feature
|
108
|
-
my %gFontFaceHash = (); my @gFontFace = ();
|
109
|
-
|
110
|
-
my @gPic = (); # pic feature
|
111
|
-
my @gTable = (); # table feature
|
112
|
-
my @gBullet = (); # bullet feature
|
113
|
-
|
114
|
-
# space feature
|
115
|
-
#my %gSpaceHash = (); my @gSpace = ();
|
116
|
-
### End XML features ###
|
117
|
-
|
118
|
-
my %tags = ();
|
119
|
-
|
120
|
-
if($isDebug){
|
121
|
-
print STDERR "\n# Processing file $inFile & output to $outFile\n";
|
122
|
-
}
|
123
|
-
|
124
|
-
my $markupOutput = "";
|
125
|
-
my $allText = processFile($inFile, $outFile, \%tags);
|
126
|
-
|
127
|
-
# Find header part
|
128
|
-
my @lines = split(/\n/, $allText);
|
129
|
-
my $numLines = scalar(@lines);
|
130
|
-
my ($headerLength, $bodyLength, $bodyStartId) =
|
131
|
-
SectLabel::PreProcess::findHeaderText(\@lines, 0, $numLines);
|
132
|
-
|
133
|
-
# Output
|
134
|
-
if($isMarkup){
|
135
|
-
open(OF, ">:utf8", "$outFile") || die"#Can't open file \"$outFile\"\n";
|
136
|
-
print OF "$markupOutput";
|
137
|
-
close OF;
|
138
|
-
} else {
|
139
|
-
output(\@lines, $outFile);
|
140
|
-
}
|
141
|
-
|
142
|
-
if($tagFile ne ""){
|
143
|
-
printTagInfo(\%tags, $tagFile);
|
144
|
-
}
|
145
|
-
|
146
|
-
sub processFile {
|
147
|
-
my ($inFile, $tags) = @_;
|
148
|
-
|
149
|
-
if (!(-e $inFile)) { die "# $progname crash\t\tFile \"$inFile\" doesn't exist"; }
|
150
|
-
open (IF, "<:utf8", $inFile) || die "# $progname crash\t\tCan't open \"$inFile\"";
|
151
|
-
|
152
|
-
my $isPara = 0;
|
153
|
-
my $isTable = 0;
|
154
|
-
my $isSpace = 0;
|
155
|
-
my $isPic = 0;
|
156
|
-
my $allText = "";
|
157
|
-
my $text = "";
|
158
|
-
|
159
|
-
my $lineId = 0;
|
160
|
-
my $isFirstTableCell = 0;
|
161
|
-
while (<IF>) { #each line contains a header
|
162
|
-
if (/^\#/) { next; } # skip comments
|
163
|
-
chomp;
|
164
|
-
s/\cM$//; # remove ^M character at the end of the file if any
|
165
|
-
my $line = $_;
|
166
|
-
|
167
|
-
if($tagFile ne ""){
|
168
|
-
processTagInfo($line, $tags);
|
169
|
-
}
|
170
|
-
|
171
|
-
# if ($line =~ /<\?xml version.+>/){ } ### Xml ###
|
172
|
-
# if ($line =~ /^<\/column>$/){ } ### Column ###
|
173
|
-
if ($isMarkup && $line =~ /<theoreticalPage (.*)\/>/ && $isMarkup){
|
174
|
-
$markupOutput .= "### Page $1\n";
|
175
|
-
}
|
176
|
-
|
177
|
-
### pic ###
|
178
|
-
if ($line =~ /^<dd (.*)>$/){
|
179
|
-
$isPic = 1;
|
180
|
-
if($isMarkup){
|
181
|
-
$markupOutput .= "### Figure $1\n";
|
182
|
-
}
|
183
|
-
}
|
184
|
-
elsif ($line =~ /^<\/dd>$/){
|
185
|
-
$isPic = 0;
|
186
|
-
}
|
187
|
-
|
188
|
-
### Table ###
|
189
|
-
elsif ($line =~ /^<table (.*)>$/){
|
190
|
-
$isTable = 1;
|
191
|
-
$isFirstTableCell = 1;
|
192
|
-
if($isMarkup){
|
193
|
-
$markupOutput .= "### Table $1\n";
|
194
|
-
}
|
195
|
-
}
|
196
|
-
elsif ($line =~ /^<\/table>$/){
|
197
|
-
$isTable = 0;
|
198
|
-
}
|
199
|
-
|
200
|
-
|
201
|
-
### Paragraph ###
|
202
|
-
# Note: table processing should have higher priority than paragraph, i.e. the priority does matter
|
203
|
-
elsif ($line =~ /^<para (.*)>$/){
|
204
|
-
$text .= $line."\n"; # we need the header
|
205
|
-
$isPara = 1;
|
206
|
-
|
207
|
-
if($isMarkup){
|
208
|
-
$markupOutput .= "## Para $1\n";
|
209
|
-
}
|
210
|
-
}
|
211
|
-
elsif ($line =~ /^<\/para>$/){
|
212
|
-
my ($paraText, $l, $t, $r, $b);
|
213
|
-
($paraText, $l, $t, $r, $b, $isSpace) = processPara($text, $isTable, $isPic, \$isFirstTableCell);
|
214
|
-
$allText .= $paraText;
|
215
|
-
|
216
|
-
my @tmpLines = split(/\n/, $paraText);
|
217
|
-
$lineId += scalar(@tmpLines);
|
218
|
-
$isPara = 0;
|
219
|
-
$text = "";
|
220
|
-
}
|
221
|
-
elsif($isPara){
|
222
|
-
$text .= $line."\n";
|
223
|
-
next;
|
224
|
-
}
|
225
|
-
}
|
226
|
-
close IF;
|
227
|
-
|
228
|
-
return $allText;
|
229
|
-
}
|
230
|
-
|
231
|
-
sub output {
|
232
|
-
my ($lines, $outFile) = @_;
|
233
|
-
|
234
|
-
open(OF, ">:utf8", "$outFile") || die"#Can't open file \"$outFile\"\n";
|
235
|
-
|
236
|
-
####### Final output ############
|
237
|
-
# xml feature label
|
238
|
-
my %gFontSizeLabels = ();
|
239
|
-
# my %gSpaceLabels = (); # yes, no
|
240
|
-
|
241
|
-
if($isXmlFeature){
|
242
|
-
getFontSizeLabels(\%gFontSizeHash, \%gFontSizeLabels);
|
243
|
-
# getSpaceLabels(\%gSpaceHash, \%gSpaceLabels);
|
244
|
-
}
|
245
|
-
|
246
|
-
my $id = -1;
|
247
|
-
my $output = "";
|
248
|
-
my $paraLineId = -1;
|
249
|
-
my $paraLineCount = 0;
|
250
|
-
foreach my $line (@{$lines}) {
|
251
|
-
$id++;
|
252
|
-
|
253
|
-
$line =~ s/\cM$//; # remove ^M character at the end of each line if any
|
254
|
-
|
255
|
-
if($line =~ /^\s*$/){ # # empty lines
|
256
|
-
if(!$isAllowEmpty){
|
257
|
-
next;
|
258
|
-
} else {
|
259
|
-
if($isDebug){
|
260
|
-
print STDERR "#! Line $id empty!\n";
|
261
|
-
}
|
262
|
-
}
|
263
|
-
}
|
264
|
-
|
265
|
-
if($gPara[$id] eq "yes"){
|
266
|
-
if($output ne ""){ ## mark para
|
267
|
-
if($isParaDelimiter){
|
268
|
-
print OF "# Para $paraLineId $paraLineCount\n$output";
|
269
|
-
$paraLineCount = 0;
|
270
|
-
} else {
|
271
|
-
if($isDecode){
|
272
|
-
$output = decode_entities($output);
|
273
|
-
}
|
274
|
-
print OF $output;
|
275
|
-
}
|
276
|
-
|
277
|
-
$output = "";
|
278
|
-
}
|
279
|
-
$paraLineId = $id;
|
280
|
-
}
|
281
|
-
|
282
|
-
$output .= $line;
|
283
|
-
$paraLineCount++;
|
284
|
-
|
285
|
-
## Output XML features ###
|
286
|
-
if($isXmlFeature){
|
287
|
-
# loc feature
|
288
|
-
my $locFeature;
|
289
|
-
if($gPosHash[$id] != -1){
|
290
|
-
$locFeature = "xmlLoc_".int(($gPosHash[$id] - $gMinPos)*8.0/($gMaxPos - $gMinPos + 1));
|
291
|
-
}
|
292
|
-
|
293
|
-
# align feature
|
294
|
-
my $alignFeature = "xmlAlign_".$gAlign[$id];
|
295
|
-
|
296
|
-
# fontSize feature
|
297
|
-
my $fontSizeFeature;
|
298
|
-
if($gFontSize[$id] == -1){
|
299
|
-
$fontSizeFeature = "xmlFontSize_none";
|
300
|
-
} else {
|
301
|
-
$fontSizeFeature = "xmlFontSize_".$gFontSizeLabels{$gFontSize[$id]};
|
302
|
-
}
|
303
|
-
|
304
|
-
my $boldFeature = "xmlBold_".$gBold[$id]; # bold feature
|
305
|
-
my $italicFeature = "xmlItalic_".$gItalic[$id]; # italic feature
|
306
|
-
my $picFeature = "xmlPic_".$gPic[$id]; # pic feature
|
307
|
-
my $tableFeature = "xmlTable_".$gTable[$id]; # table feature
|
308
|
-
my $bulletFeature = "xmlBullet_".$gBullet[$id]; # bullet feature
|
309
|
-
|
310
|
-
# space feature
|
311
|
-
# my $spaceFeature;
|
312
|
-
# if($gSpace[$id] eq "none"){
|
313
|
-
# $spaceFeature = "xmlSpace_none";
|
314
|
-
# } else {
|
315
|
-
# $spaceFeature = "xmlSpace_".$gSpaceLabels{$gSpace[$id]};
|
316
|
-
# }
|
317
|
-
|
318
|
-
## Differential features ##
|
319
|
-
my ($alignDiff, $fontSizeDiff, $fontFaceDiff, $fontSFDiff, $fontSFBIDiff, $fontSFBIADiff, $paraDiff) = getDifferentialFeatures($id);
|
320
|
-
|
321
|
-
$output .= " |XML| $locFeature $boldFeature $italicFeature $fontSizeFeature $picFeature $tableFeature $bulletFeature $fontSFBIADiff $paraDiff\n"; # $alignFeature $alignDiff $fontSizeDiff $fontFaceDiff $fontSFDiff $fontSFBIDiff
|
322
|
-
} else {
|
323
|
-
$output .= "\n";
|
324
|
-
}
|
325
|
-
}
|
326
|
-
|
327
|
-
if($output ne ""){ ## mark para
|
328
|
-
if($isParaDelimiter){
|
329
|
-
print OF "# Para $paraLineId $paraLineCount\n$output";
|
330
|
-
$paraLineCount = 0;
|
331
|
-
} else {
|
332
|
-
if($isDecode){
|
333
|
-
$output = decode_entities($output);
|
334
|
-
}
|
335
|
-
|
336
|
-
print OF $output;
|
337
|
-
}
|
338
|
-
$output = ""
|
339
|
-
}
|
340
|
-
close OF;
|
341
|
-
}
|
342
|
-
|
343
|
-
sub getDifferentialFeatures {
|
344
|
-
my ($id) = @_;
|
345
|
-
|
346
|
-
# alignChange feature
|
347
|
-
my $alignDiff = "bi_xmlA_";
|
348
|
-
if($id == 0){
|
349
|
-
$alignDiff .= $gAlign[$id];
|
350
|
-
} elsif($gAlign[$id] eq $gAlign[$id-1]){
|
351
|
-
$alignDiff .= "continue";
|
352
|
-
} else {
|
353
|
-
$alignDiff .= $gAlign[$id];
|
354
|
-
}
|
355
|
-
|
356
|
-
# fontFaceChange feature
|
357
|
-
my $fontFaceDiff = "bi_xmlF_";
|
358
|
-
if($id == 0){
|
359
|
-
$fontFaceDiff .= "new";
|
360
|
-
} elsif($gFontFace[$id] eq $gFontFace[$id-1]){
|
361
|
-
$fontFaceDiff .= "continue";
|
362
|
-
} else {
|
363
|
-
$fontFaceDiff .= "new";
|
364
|
-
}
|
365
|
-
|
366
|
-
# fontSizeChange feature
|
367
|
-
my $fontSizeDiff = "bi_xmlS_";
|
368
|
-
if($id == 0){
|
369
|
-
$fontSizeDiff .= "new";
|
370
|
-
} elsif($gFontSize[$id] == $gFontSize[$id-1]){
|
371
|
-
$fontSizeDiff .= "continue";
|
372
|
-
} else {
|
373
|
-
$fontSizeDiff .= "new";
|
374
|
-
}
|
375
|
-
|
376
|
-
# fontSFChange feature
|
377
|
-
my $fontSFDiff = "bi_xmlSF_";
|
378
|
-
if($id == 0){
|
379
|
-
$fontSFDiff .= "new";
|
380
|
-
} elsif($gFontSize[$id] == $gFontSize[$id-1] && $gFontFace[$id] eq $gFontFace[$id-1]){
|
381
|
-
$fontSFDiff .= "continue";
|
382
|
-
} else {
|
383
|
-
$fontSFDiff .= "new";
|
384
|
-
}
|
385
|
-
|
386
|
-
# fontSFBIChange feature
|
387
|
-
my $fontSFBIDiff = "bi_xmlSFBI_";
|
388
|
-
if($id == 0){
|
389
|
-
$fontSFBIDiff .= "new";
|
390
|
-
} elsif($gFontSize[$id] == $gFontSize[$id-1] && $gFontFace[$id] eq $gFontFace[$id-1] && $gBold[$id] eq $gBold[$id-1] && $gItalic[$id] eq $gItalic[$id-1]){
|
391
|
-
$fontSFBIDiff .= "continue";
|
392
|
-
} else {
|
393
|
-
$fontSFBIDiff .= "new";
|
394
|
-
}
|
395
|
-
|
396
|
-
# fontSFBIAChange feature
|
397
|
-
my $fontSFBIADiff = "bi_xmlSFBIA_";
|
398
|
-
if($id == 0){
|
399
|
-
$fontSFBIADiff .= "new";
|
400
|
-
} elsif($gFontSize[$id] == $gFontSize[$id-1] && $gFontFace[$id] eq $gFontFace[$id-1] && $gBold[$id] eq $gBold[$id-1] && $gItalic[$id] eq $gItalic[$id-1] && $gAlign[$id] eq $gAlign[$id-1]){
|
401
|
-
$fontSFBIADiff .= "continue";
|
402
|
-
} else {
|
403
|
-
$fontSFBIADiff .= "new";
|
404
|
-
}
|
405
|
-
|
406
|
-
# para change feature
|
407
|
-
my $paraDiff = "bi_xmlPara_";
|
408
|
-
if($id < $bodyStartId){ # header part, consider each line as a separate paragraph
|
409
|
-
$paraDiff .= "header";
|
410
|
-
} else {
|
411
|
-
if($gPara[$id] eq "yes"){
|
412
|
-
$paraDiff .= "new";
|
413
|
-
} else {
|
414
|
-
$paraDiff .= "continue";
|
415
|
-
}
|
416
|
-
}
|
417
|
-
|
418
|
-
return ($alignDiff, $fontSizeDiff, $fontFaceDiff, $fontSFDiff, $fontSFBIDiff, $fontSFBIADiff, $paraDiff);
|
419
|
-
}
|
420
|
-
|
421
|
-
sub getFontSizeLabels {
|
422
|
-
my ($gFontSizeHash, $gFontSizeLabels) = @_;
|
423
|
-
|
424
|
-
if($isDebug){ print STDERR "# Map fonts\n"; }
|
425
|
-
my @sortedFonts = sort { $gFontSizeHash->{$b} <=> $gFontSizeHash->{$a} } keys %{$gFontSizeHash}; # sort by values, obtain keys
|
426
|
-
|
427
|
-
my $commonSize = $sortedFonts[0];
|
428
|
-
@sortedFonts = sort { $a <=> $b } keys %{$gFontSizeHash}; # sort by keys, obtain keys
|
429
|
-
my $commonIndex = 0; # index of common font size
|
430
|
-
foreach(@sortedFonts){
|
431
|
-
if($commonSize == $_) { # found
|
432
|
-
last;
|
433
|
-
}
|
434
|
-
$commonIndex++;
|
435
|
-
}
|
436
|
-
|
437
|
-
# small fonts
|
438
|
-
for(my $i = 0; $i<$commonIndex; $i++){ # smallIndex $largeIndex
|
439
|
-
$gFontSizeLabels->{$sortedFonts[$i]} = "smaller";
|
440
|
-
|
441
|
-
if($isDebug){
|
442
|
-
print STDERR "$sortedFonts[$i] --> $gFontSizeLabels->{$sortedFonts[$i]}, freq = $gFontSizeHash->{$sortedFonts[$i]}\n";
|
443
|
-
}
|
444
|
-
}
|
445
|
-
|
446
|
-
# common fonts
|
447
|
-
$gFontSizeLabels->{$commonSize} = "common";
|
448
|
-
if($isDebug){
|
449
|
-
print STDERR "$sortedFonts[$commonIndex] --> $gFontSizeLabels->{$sortedFonts[$commonIndex]}, freq = $gFontSizeHash->{$sortedFonts[$commonIndex]}\n";
|
450
|
-
}
|
451
|
-
|
452
|
-
# large fonts
|
453
|
-
for(my $i = ($commonIndex+1); $i<scalar(@sortedFonts); $i++){ # ($largeIndex+1) (scalar(@sortedFonts)-1)
|
454
|
-
if((scalar(@sortedFonts)-$i) <= 3){
|
455
|
-
$gFontSizeLabels->{$sortedFonts[$i]} = "largest".($i+1-scalar(@sortedFonts));
|
456
|
-
} else {
|
457
|
-
$gFontSizeLabels->{$sortedFonts[$i]} = "larger";
|
458
|
-
}
|
459
|
-
|
460
|
-
if($isDebug){
|
461
|
-
print STDERR "$sortedFonts[$i] --> $gFontSizeLabels->{$sortedFonts[$i]}, freq = $gFontSizeHash->{$sortedFonts[$i]}\n";
|
462
|
-
}
|
463
|
-
}
|
464
|
-
}
|
465
|
-
|
466
|
-
sub getSpaceLabels {
|
467
|
-
my ($gSpaceHash, $gSpaceLabels) = @_;
|
468
|
-
|
469
|
-
if($isDebug){
|
470
|
-
print STDERR "\n# Map space\n";
|
471
|
-
}
|
472
|
-
my @sortedSpaces = sort { $gSpaceHash->{$b} <=> $gSpaceHash->{$a} } keys %{$gSpaceHash}; # sort by freqs, obtain space faces
|
473
|
-
|
474
|
-
my $commonSpace = $sortedSpaces[0];
|
475
|
-
my $commonFreq = $gSpaceHash->{$commonSpace};
|
476
|
-
# find similar common freq with larger spaces
|
477
|
-
for(my $i = 0; $i<scalar(@sortedSpaces); $i++){ # 0 ($smallIndex-1)
|
478
|
-
my $freq = $gSpaceHash->{$sortedSpaces[$i]};
|
479
|
-
if($freq/$commonFreq > 0.8){
|
480
|
-
if($sortedSpaces[$i] > $commonSpace){
|
481
|
-
$commonSpace = $sortedSpaces[$i];
|
482
|
-
}
|
483
|
-
} else {
|
484
|
-
last;
|
485
|
-
}
|
486
|
-
}
|
487
|
-
|
488
|
-
for(my $i = 0; $i<scalar(@sortedSpaces); $i++){ # 0 ($smallIndex-1)
|
489
|
-
if($sortedSpaces[$i] > $commonSpace){
|
490
|
-
$gSpaceLabels->{$sortedSpaces[$i]} = "yes";
|
491
|
-
} else {
|
492
|
-
$gSpaceLabels->{$sortedSpaces[$i]} = "no";
|
493
|
-
}
|
494
|
-
|
495
|
-
if($isDebug){
|
496
|
-
print STDERR "$sortedSpaces[$i] --> $gSpaceLabels->{$sortedSpaces[$i]}, freq = $gSpaceHash->{$sortedSpaces[$i]}\n";
|
497
|
-
}
|
498
|
-
}
|
499
|
-
}
|
500
|
-
|
501
|
-
sub getAttrValue {
|
502
|
-
my ($attrText, $attr) = @_;
|
503
|
-
|
504
|
-
my $value = "none";
|
505
|
-
if($attrText =~ /^.*$attr=\"(.+?)\".*$/){
|
506
|
-
$value = $1;
|
507
|
-
}
|
508
|
-
|
509
|
-
return $value;
|
510
|
-
}
|
511
|
-
|
512
|
-
sub checkFontAttr {
|
513
|
-
my ($attrText, $attr, $attrHash, $count) = @_;
|
514
|
-
|
515
|
-
if($attrText =~ /^.*$attr=\"(.+?)\".*$/){
|
516
|
-
my $attrValue = $1;
|
517
|
-
|
518
|
-
$attrHash->{$attrValue} = $attrHash->{$attrValue} ? ($attrHash->{$attrValue}+$count) : $count;
|
519
|
-
}
|
520
|
-
}
|
521
|
-
|
522
|
-
sub processPara {
|
523
|
-
my ($inputText, $isTable, $isPic, $isFirstTableCell) = @_;
|
524
|
-
|
525
|
-
my $isSpace = 0;
|
526
|
-
my $isSpecialSpace = 0;
|
527
|
-
my $isTab = 0;
|
528
|
-
my $isBullet = 0;
|
529
|
-
|
530
|
-
my $isForcedEOF = "none"; # 3 signals for end of L: forcedEOF=\"true\" in attribute of <ln> or || <nl orig=\"true\"\/> || end of </para> without encountering any of the above signal in the para plus $isSpace = 0
|
531
|
-
# xml feature
|
532
|
-
my $align = "none";
|
533
|
-
my ($l, $t, $r, $bottom);
|
534
|
-
my %fontSizeHash = ();
|
535
|
-
my %fontFaceHash = ();
|
536
|
-
my @boldArray = ();
|
537
|
-
my @italicArray = ();
|
538
|
-
my $space = "none";
|
539
|
-
|
540
|
-
my $lnAttr; my $isLn = 0; my $lnBold = "none"; my $lnItalic = "none";
|
541
|
-
my $runAttr; my $runText = ""; my $isRun = 0; my $runBold = "none"; my $runItalic = "none";
|
542
|
-
my $wdAttr; my $wdText = ""; my $isWd = 0;
|
543
|
-
|
544
|
-
my $wdIndex = 0; # word index in a line. When encountering </ln>, this parameter indicates the number of words in a line
|
545
|
-
my $lnBoldCount = 0;
|
546
|
-
my $lnItalicCount = 0;
|
547
|
-
|
548
|
-
my $allText = "";
|
549
|
-
my $text = ""; #invariant: when never enter a new line, $text will be copied into $allText, and $text is cleared
|
550
|
-
|
551
|
-
binmode(STDERR, ":utf8");
|
552
|
-
|
553
|
-
my $isFirstLinePara = 1;
|
554
|
-
my @lines = split(/\n/, $inputText);
|
555
|
-
for(my $i=0; $i<scalar(@lines); $i++){
|
556
|
-
my $line = $lines[$i];
|
557
|
-
|
558
|
-
## new para
|
559
|
-
if ($line =~ /^<para (.+?)>$/){
|
560
|
-
my $attr = $1;
|
561
|
-
$align = getAttrValue($attr, "alignment");
|
562
|
-
# $indent = getAttrValue($attr, "li");
|
563
|
-
$space = getAttrValue($attr, "spaceBefore");
|
564
|
-
}
|
565
|
-
|
566
|
-
## new ln
|
567
|
-
elsif ($line =~ /^<ln (.+)>$/){
|
568
|
-
$lnAttr = $1;
|
569
|
-
$isLn = 1;
|
570
|
-
|
571
|
-
if ($isMarkup){
|
572
|
-
$markupOutput .= "# Line $lnAttr\n";
|
573
|
-
}
|
574
|
-
|
575
|
-
if ($lnAttr =~ /^.*l=\"(\d+)\" t=\"(\d+)\" r=\"(\d+)\" b=\"(\d+)\".*$/){
|
576
|
-
($l, $t, $r, $bottom) = ($1, $2, $3, $4);
|
577
|
-
}
|
578
|
-
$isForcedEOF = getAttrValue($lnAttr, "forcedEOF");
|
579
|
-
|
580
|
-
if($isXmlFeature){ # Bold & Italic
|
581
|
-
$lnBold = getAttrValue($lnAttr, "bold");
|
582
|
-
$lnItalic = getAttrValue($lnAttr, "italic");
|
583
|
-
}
|
584
|
-
}
|
585
|
-
|
586
|
-
## new run
|
587
|
-
elsif ($line =~ /<run (.*)>$/){
|
588
|
-
$runAttr = $1;
|
589
|
-
|
590
|
-
$isSpace = 0;
|
591
|
-
$isTab = 0;
|
592
|
-
$isRun = 1;
|
593
|
-
|
594
|
-
if($line =~ /^<wd (.*?)>/){ # new wd, that consists of many runs
|
595
|
-
$isWd = 1;
|
596
|
-
$wdAttr = $1;
|
597
|
-
}
|
598
|
-
|
599
|
-
if($isXmlFeature){ # Bold & Italic
|
600
|
-
$runBold = getAttrValue($runAttr, "bold");
|
601
|
-
$runItalic = getAttrValue($runAttr, "italic");
|
602
|
-
}
|
603
|
-
}
|
604
|
-
|
605
|
-
## wd
|
606
|
-
elsif ($line =~ /^<wd (.+)?>(.+)<\/wd>$/){
|
607
|
-
$wdAttr = $1;
|
608
|
-
my $word = $2;
|
609
|
-
$isSpace = 0;
|
610
|
-
$isTab = 0;
|
611
|
-
|
612
|
-
if ($isMarkup){
|
613
|
-
$markupOutput .= "$word $wdAttr";
|
614
|
-
if($isRun && $runAttr =~ /(bold|italic)=\"true\"/){ # if both bold and italic, then just use one
|
615
|
-
$markupOutput .= " $1=\"true\"";
|
616
|
-
}
|
617
|
-
$markupOutput .= "\n";
|
618
|
-
}
|
619
|
-
|
620
|
-
if($isXmlFeature){ # FontSize & FontFace
|
621
|
-
checkFontAttr($wdAttr, "fontSize", \%fontSizeHash, 1);
|
622
|
-
checkFontAttr($wdAttr, "fontFace", \%fontFaceHash, 1);
|
623
|
-
}
|
624
|
-
|
625
|
-
if($isXmlFeature){ # Bold & Italic
|
626
|
-
my $wdBold = getAttrValue($wdAttr, "bold");
|
627
|
-
my $wdItalic = getAttrValue($wdAttr, "italic");
|
628
|
-
|
629
|
-
if($wdBold eq "true" || $runBold eq "true" || $lnBold eq "true"){
|
630
|
-
$boldArray[$wdIndex] = 1;
|
631
|
-
$lnBoldCount++;
|
632
|
-
}
|
633
|
-
|
634
|
-
if($wdItalic eq "true" || $runItalic eq "true" || $lnItalic eq "true"){
|
635
|
-
$italicArray[$wdIndex] = 1;
|
636
|
-
$lnItalicCount++;
|
637
|
-
}
|
638
|
-
} # if($isXmlFeature)
|
639
|
-
|
640
|
-
## add text
|
641
|
-
$text .= "$word";
|
642
|
-
|
643
|
-
if($isRun) {
|
644
|
-
$runText .= "$word ";
|
645
|
-
}
|
646
|
-
$wdIndex++;
|
647
|
-
}
|
648
|
-
|
649
|
-
## end wd
|
650
|
-
elsif ($line =~ /^<\/wd>$/){
|
651
|
-
$isWd = 0;
|
652
|
-
|
653
|
-
if($isMarkup){
|
654
|
-
$markupOutput .= "$wdText $wdAttr";
|
655
|
-
if($isRun && $runAttr =~ /(bold|italic)=\"true\"/){ # if both bold and italic, then just use one
|
656
|
-
$markupOutput .= " $1=\"true\"";
|
657
|
-
}
|
658
|
-
$markupOutput .= "\n";
|
659
|
-
|
660
|
-
$wdAttr = "";
|
661
|
-
}
|
662
|
-
}
|
663
|
-
|
664
|
-
## end run
|
665
|
-
elsif ($line =~ /^(.*)<\/run>$/){
|
666
|
-
my $word = $1;
|
667
|
-
|
668
|
-
## add text
|
669
|
-
if($word ne ""){
|
670
|
-
if($isXmlFeature){ # Bold & Italic
|
671
|
-
if($runBold eq "true" || $lnBold eq "true"){
|
672
|
-
$boldArray[$wdIndex] = 1;
|
673
|
-
$lnBoldCount++;
|
674
|
-
}
|
675
|
-
|
676
|
-
if($runItalic eq "true" || $lnItalic eq "true"){
|
677
|
-
$italicArray[$wdIndex] = 1;
|
678
|
-
$lnItalicCount++;
|
679
|
-
}
|
680
|
-
}
|
681
|
-
|
682
|
-
# appear in the final result
|
683
|
-
if($isLn){ $text .= "$word"; }
|
684
|
-
|
685
|
-
# for internal record
|
686
|
-
if($isRun){ $runText .= "$word "; }
|
687
|
-
if($isWd){ $wdText .= "$word"; }
|
688
|
-
|
689
|
-
$wdIndex++;
|
690
|
-
}
|
691
|
-
|
692
|
-
# xml feature
|
693
|
-
if($isXmlFeature && $runText ne "") { # not a space, tab or new-line run
|
694
|
-
my @words = split(/\s+/, $runText);
|
695
|
-
my $numWords = scalar(@words);
|
696
|
-
checkFontAttr($runAttr, "fontSize", \%fontSizeHash, $numWords);
|
697
|
-
checkFontAttr($runAttr, "fontFace", \%fontFaceHash, $numWords);
|
698
|
-
}
|
699
|
-
|
700
|
-
## reset run
|
701
|
-
if(!$isLn){ # <run> not enclosed within <ln>
|
702
|
-
$wdIndex = 0;
|
703
|
-
}
|
704
|
-
$runText = "";
|
705
|
-
$isRun = 0;
|
706
|
-
$isSpecialSpace = 0;
|
707
|
-
|
708
|
-
if($isXmlFeature){ # Bold & Italic
|
709
|
-
$runBold = "none";
|
710
|
-
$runItalic = "none";
|
711
|
-
|
712
|
-
if(!$isLn){ # <run> not enclosed within <ln>
|
713
|
-
$lnBoldCount = 0;
|
714
|
-
$lnItalicCount = 0;
|
715
|
-
}
|
716
|
-
}
|
717
|
-
}
|
718
|
-
|
719
|
-
## end ln
|
720
|
-
elsif ($line =~ /^<\/ln>$/){
|
721
|
-
if((!$isAllowEmpty && $text !~ /^\s*$/)
|
722
|
-
|| ($isAllowEmpty && $text ne "")){
|
723
|
-
if($isForcedEOF eq "true" || # there's a forced EOL?
|
724
|
-
!$isSpecialSpace # not an emply line with space character
|
725
|
-
){
|
726
|
-
$text .= "\n";
|
727
|
-
|
728
|
-
# update allText
|
729
|
-
$allText .= $text;
|
730
|
-
$text = "";
|
731
|
-
}
|
732
|
-
|
733
|
-
my $numWords = $wdIndex;
|
734
|
-
|
735
|
-
if(!$isTable){
|
736
|
-
if($isFirstLinePara){
|
737
|
-
push(@gPara, "yes");
|
738
|
-
$isFirstLinePara = 0;
|
739
|
-
} else {
|
740
|
-
push(@gPara, "no");
|
741
|
-
}
|
742
|
-
} else {
|
743
|
-
if($$isFirstTableCell){
|
744
|
-
push(@gPara, "yes");
|
745
|
-
$$isFirstTableCell = 0;
|
746
|
-
} else {
|
747
|
-
push(@gPara, "no");
|
748
|
-
}
|
749
|
-
}
|
750
|
-
|
751
|
-
if($isXmlFeature && $numWords >= 1){
|
752
|
-
# xml feature
|
753
|
-
# assumtion that: fontSize is either occur in <ln>, or within multiple <run> under <ln>, but not both
|
754
|
-
checkFontAttr($lnAttr, "fontSize", \%fontSizeHash, $numWords);
|
755
|
-
checkFontAttr($lnAttr, "fontFace", \%fontFaceHash, $numWords);
|
756
|
-
}
|
757
|
-
|
758
|
-
if($isXmlFeature && !$isSpecialSpace){
|
759
|
-
my $pos = ($t+$bottom)/2.0;
|
760
|
-
if($pos < $gMinPos){ $gMinPos = $pos; }
|
761
|
-
if($pos > $gMaxPos){ $gMaxPos = $pos; }
|
762
|
-
push(@gPosHash, $pos); # pos feature
|
763
|
-
push(@gAlign, $align); # alignment feature
|
764
|
-
|
765
|
-
if($isPic){
|
766
|
-
push(@gPic, "yes");
|
767
|
-
} else {
|
768
|
-
push(@gPic, "no");
|
769
|
-
}
|
770
|
-
if($isTable){
|
771
|
-
push(@gTable, "yes");
|
772
|
-
} else {
|
773
|
-
push(@gTable, "no");
|
774
|
-
}
|
775
|
-
|
776
|
-
if($isPic || $isTable){
|
777
|
-
### Not assign value ###
|
778
|
-
push(@gFontSize, -1); # bold feature
|
779
|
-
push(@gFontFace, "none"); # bold feature
|
780
|
-
push(@gBold, "no"); # bold feature
|
781
|
-
push(@gItalic, "no"); # italic feature
|
782
|
-
push(@gBullet, "no"); # bullet feature
|
783
|
-
} else {
|
784
|
-
updateXMLFontFeature(\%fontSizeHash, \%fontFaceHash);
|
785
|
-
%fontSizeHash = (); %fontFaceHash = ();
|
786
|
-
|
787
|
-
updateXMLFeatures($lnBoldCount, $lnItalicCount, $numWords, $isBullet, $space);
|
788
|
-
} # end if pic
|
789
|
-
} # end if($isXmlFeature && !$isSpecialSpace)
|
790
|
-
}
|
791
|
-
|
792
|
-
## reset ln
|
793
|
-
$isLn = 0;
|
794
|
-
$isForcedEOF = "none";
|
795
|
-
$isSpecialSpace = 0;
|
796
|
-
$wdIndex = 0;
|
797
|
-
|
798
|
-
if($isXmlFeature){ # Bold & Italic
|
799
|
-
$lnBold = "none";
|
800
|
-
$lnItalic = "none";
|
801
|
-
|
802
|
-
$lnBoldCount = 0;
|
803
|
-
$lnItalicCount = 0;
|
804
|
-
}
|
805
|
-
} # end else </ln>
|
806
|
-
|
807
|
-
## nl newline signal
|
808
|
-
elsif ($line =~ /^<nl orig=\"true\"\/>$/){
|
809
|
-
if($isLn){
|
810
|
-
$isSpace = 0;
|
811
|
-
} else {
|
812
|
-
if($isDebug){
|
813
|
-
print STDERR "#!!! Warning: found <nl orig=\"true\"\/> while not in tag <ln>: $line\n";
|
814
|
-
}
|
815
|
-
}
|
816
|
-
}
|
817
|
-
|
818
|
-
## space
|
819
|
-
elsif ($line =~ /^<space\/>$/){
|
820
|
-
my $startTag = "";
|
821
|
-
my $endTag = "";
|
822
|
-
if($i>0 && $lines[$i-1] =~ /^<(.+?)\b.*/){
|
823
|
-
$startTag = $1;
|
824
|
-
}
|
825
|
-
|
826
|
-
if($i < (scalar(@lines) -1) && $lines[$i+1] =~ /^<\/(.+)>/){
|
827
|
-
$endTag = $1;
|
828
|
-
}
|
829
|
-
|
830
|
-
if($startTag eq $endTag && $startTag ne ""){
|
831
|
-
# print STDERR "# Special space after \"$text\"\n";
|
832
|
-
$isSpecialSpace = 1;
|
833
|
-
}
|
834
|
-
|
835
|
-
## addText
|
836
|
-
$text .= " ";
|
837
|
-
$isSpace = 1;
|
838
|
-
}
|
839
|
-
|
840
|
-
## tab
|
841
|
-
elsif ($line =~ /^<tab .*\/>$/){
|
842
|
-
## add Text
|
843
|
-
$text .= "\t";
|
844
|
-
|
845
|
-
$isTab = 1;
|
846
|
-
}
|
847
|
-
|
848
|
-
## bullet
|
849
|
-
elsif ($line =~ /^<bullet .*>$/){
|
850
|
-
$isBullet = 1;
|
851
|
-
}
|
852
|
-
}
|
853
|
-
|
854
|
-
$allText .= $text;
|
855
|
-
return ($allText, $l, $t, $r, $bottom, $isSpace);
|
856
|
-
}
|
857
|
-
|
858
|
-
sub updateXMLFontFeature {
|
859
|
-
my ($fontSizeHash, $fontFaceHash) = @_;
|
860
|
-
|
861
|
-
# font size feature
|
862
|
-
if(scalar(keys %{$fontSizeHash}) == 0){
|
863
|
-
push(@gFontSize, -1);
|
864
|
-
} else {
|
865
|
-
my @sortedFonts = sort { $fontSizeHash->{$b} <=> $fontSizeHash->{$a} } keys %{$fontSizeHash};
|
866
|
-
|
867
|
-
my $fontSize = $sortedFonts[0];
|
868
|
-
push(@gFontSize, $fontSize);
|
869
|
-
|
870
|
-
$gFontSizeHash{$fontSize} = $gFontSizeHash{$fontSize} ? ($gFontSizeHash{$fontSize}+1) : 1;
|
871
|
-
}
|
872
|
-
|
873
|
-
# font face feature
|
874
|
-
if(scalar(keys %{$fontFaceHash}) == 0){
|
875
|
-
push(@gFontFace, "none");
|
876
|
-
} else {
|
877
|
-
my @sortedFonts = sort { $fontFaceHash->{$b} <=> $fontFaceHash->{$a} } keys %{$fontFaceHash};
|
878
|
-
my $fontFace = $sortedFonts[0];
|
879
|
-
push(@gFontFace, $fontFace);
|
880
|
-
|
881
|
-
$gFontFaceHash{$fontFace} = $gFontFaceHash{$fontFace} ? ($gFontFaceHash{$fontFace}+1) : 1;
|
882
|
-
}
|
883
|
-
}
|
884
|
-
|
885
|
-
sub updateXMLFeatures {
|
886
|
-
my ($lnBoldCount, $lnItalicCount, $numWords, $isBullet, $space) = @_;
|
887
|
-
# bold feature
|
888
|
-
my $boldFeature;
|
889
|
-
if ($lnBoldCount/$numWords >= 0.667){
|
890
|
-
$boldFeature = "yes";
|
891
|
-
} else {
|
892
|
-
$boldFeature = "no";
|
893
|
-
}
|
894
|
-
push(@gBold, $boldFeature);
|
895
|
-
|
896
|
-
# italic feature
|
897
|
-
my $italicFeature;
|
898
|
-
if ($lnItalicCount/$numWords >= 0.667){
|
899
|
-
$italicFeature = "yes";
|
900
|
-
} else {
|
901
|
-
$italicFeature = "no";
|
902
|
-
}
|
903
|
-
push(@gItalic, $italicFeature);
|
904
|
-
|
905
|
-
# bullet feature
|
906
|
-
if($isBullet){
|
907
|
-
push(@gBullet, "yes");
|
908
|
-
} else {
|
909
|
-
push(@gBullet, "no");
|
910
|
-
}
|
911
|
-
|
912
|
-
# space feature
|
913
|
-
# push(@gSpace, $space);
|
914
|
-
}
|
915
|
-
|
916
|
-
## Find the positions of header, body, and citation
|
917
|
-
sub getStructureInfo {
|
918
|
-
my ($lines, $numLines) = @_;
|
919
|
-
|
920
|
-
my ($bodyLength, $citationLength, $bodyEndId) =
|
921
|
-
SectLabel::PreProcess::findCitationText($lines, 0, $numLines);
|
922
|
-
|
923
|
-
my ($headerLength, $bodyStartId);
|
924
|
-
($headerLength, $bodyLength, $bodyStartId) =
|
925
|
-
SectLabel::PreProcess::findHeaderText($lines, 0, $bodyLength);
|
926
|
-
|
927
|
-
# sanity check
|
928
|
-
my $totalLength = $headerLength + $bodyLength + $citationLength;
|
929
|
-
if($numLines != $totalLength){
|
930
|
-
print STDOUT "Die in getStructureInfo(): different num lines $numLines != $totalLength\n"; # to display in Web
|
931
|
-
die "Die in getStructureInfo(): different num lines $numLines != $totalLength\n";
|
932
|
-
}
|
933
|
-
return ($headerLength, $bodyLength, $citationLength, $bodyStartId, $bodyEndId);
|
934
|
-
}
|
935
|
-
|
936
|
-
## Count XML tags/values for statistics purpose
|
937
|
-
sub processTagInfo {
|
938
|
-
my ($line, $tags) = @_;
|
939
|
-
|
940
|
-
my $tag;
|
941
|
-
my $attr;
|
942
|
-
if($line =~ /^<(.+?)\b(.*)/){
|
943
|
-
$tag = $1;
|
944
|
-
$attr = $2;
|
945
|
-
if(!$tags->{$tag}){
|
946
|
-
$tags->{$tag} = ();
|
947
|
-
}
|
948
|
-
if($attr =~ /^\s*(.+?)\s*\/?>/){
|
949
|
-
$attr = $1;
|
950
|
-
}
|
951
|
-
|
952
|
-
my @tokens = split(/\s+/, $attr);
|
953
|
-
foreach my $token (@tokens){
|
954
|
-
if($token =~ /^(.+)=(.+)$/){
|
955
|
-
my $attrName = $1;
|
956
|
-
my $value = $2;
|
957
|
-
if(!$tags->{$tag}->{$attrName}){
|
958
|
-
$tags->{$tag}->{$attrName} = ();
|
959
|
-
}
|
960
|
-
if(!$tags->{$tag}->{$attrName}->{$value}){
|
961
|
-
$tags->{$tag}->{$attrName}->{$value} = 0;
|
962
|
-
}
|
963
|
-
$tags->{$tag}->{$attrName}->{$value}++;
|
964
|
-
}
|
965
|
-
}
|
966
|
-
}
|
967
|
-
}
|
968
|
-
|
969
|
-
## Print tag info to file
|
970
|
-
sub printTagInfo {
|
971
|
-
my ($tags, $tagFile) = @_;
|
972
|
-
|
973
|
-
open(TAG, ">:utf8", "$tagFile") || die"#Can't open file \"$tagFile\"\n";
|
974
|
-
my @sortedTags = sort {$a cmp $b} keys %{$tags};
|
975
|
-
foreach(@sortedTags){
|
976
|
-
my @attrs = sort {$a cmp $b} keys %{$tags->{$_}};
|
977
|
-
print TAG "# Tag = $_\n";
|
978
|
-
foreach my $attr (@attrs) {
|
979
|
-
print TAG "$attr:";
|
980
|
-
my @values = sort {$a cmp $b} keys %{$tags->{$_}->{$attr}};
|
981
|
-
foreach my $value (@values){
|
982
|
-
print TAG " $value-$tags->{$_}->{$attr}->{$value}";
|
983
|
-
}
|
984
|
-
print TAG "\n";
|
985
|
-
}
|
986
|
-
}
|
987
|
-
close TAG;
|
988
|
-
}
|
989
|
-
|
990
|
-
sub untaintPath {
|
991
|
-
my ($path) = @_;
|
992
|
-
|
993
|
-
if ( $path =~ /^([-_\/\w\.]*)$/ ) {
|
994
|
-
$path = $1;
|
995
|
-
} else {
|
996
|
-
die "Bad path \"$path\"\n";
|
997
|
-
}
|
998
|
-
|
999
|
-
return $path;
|
1000
|
-
}
|
1001
|
-
|
1002
|
-
sub untaint {
|
1003
|
-
my ($s) = @_;
|
1004
|
-
if ($s =~ /^([\w \-\@\(\),\.\/]+)$/) {
|
1005
|
-
$s = $1; # $data now untainted
|
1006
|
-
} else {
|
1007
|
-
die "Bad data in $s"; # log this somewhere
|
1008
|
-
}
|
1009
|
-
return $s;
|
1010
|
-
}
|
1011
|
-
|
1012
|
-
sub execute {
|
1013
|
-
my ($cmd) = @_;
|
1014
|
-
if($isDebug){
|
1015
|
-
print STDERR "Executing: $cmd\n";
|
1016
|
-
}
|
1017
|
-
$cmd = untaint($cmd);
|
1018
|
-
system($cmd);
|
1019
|
-
}
|
1020
|
-
|
1021
|
-
sub newTmpFile {
|
1022
|
-
my $tmpFile = `date '+%Y%m%d-%H%M%S-$$'`;
|
1023
|
-
chomp($tmpFile);
|
1024
|
-
return $tmpFile;
|
1025
|
-
}
|