1Text::WordDiff(3pm) User Contributed Perl Documentation Text::WordDiff(3pm)
2
6 Text::WordDiff - Track changes between documents
7
9 use Text::WordDiff;
10 my $diff = word_diff 'file1.txt', 'file2.txt', { STYLE => 'HTML' };
12 my $diff = word_diff \$string1, \$string2, { STYLE => 'ANSIColor' };
13 my $diff = word_diff \*FH1, \*FH2; \%options;
14 my $diff = word_diff \&reader1, \&reader2;
15 my $diff = word_diff \@records1, \@records2;
16 # May also mix input types:
18 my $diff = word_diff \@records1, 'file_B.txt';
19
21 This module is a variation on the lovely Text::Diff module. Rather
22 than generating traditional line-oriented diffs, however, it generates
23 word-oriented diffs. This can be useful for tracking changes in
24 narrative documents or documents with very long lines. To diff source
25 code, one is still best off using Text::Diff. But if you want to see
26 how a short story changed from one version to the next, this module
27 will do the job very nicely.
28 What is a Word?
30 I'm glad you asked! Well, sort of. It's a really hard question to
31 answer. I consulted a number of sources, but really just did my best to
32 punt on the question by reformulating it as, "How do I split text up
33 into individual words?" The short answer is to split on word
34 boundaries. However, every word has two boundaries, one at the
35 beginning and one at the end. So splitting on "/\b/" didn't work so
36 well. What I really wanted to do was to split on the beginning of every
37 word. Fortunately, _Mastering Regular Expressions_ has a recipe for
38 that: "/(?<!\w)(?=\w)/". I've borrowed this regular expression for use
39 in Perls before 5.6.x, but go for the Unicode variant in 5.6.0 and
40 newer: "/(?<!\p{IsWord})(?=\p{IsWord})/". Adding some additional
41 controls for punctuation and control characters, this sentence, for
42 example, would be split up into the following tokens:
43 my @words = (
45 "Adding ",
46 "some ",
47 "additional ",
48 "controls",
49 "\n",
50 "for ",
51 "punctuation ",
52 "and ",
53 "control ",
54 "characters",
55 ", ",
56 "this ",
57 "sentence",
58 ", ",
59 "for ",
60 "example",
61 ", ",
62 "would ",
63 "be",
64 "\n",
65 "split ",
66 "up ",
67 "into ",
68 "the ",
69 "following ",
70 "tokens",
71 ":",
72 );
73 So it's not just comparing words, but word-like tokens and
75 control/punctuation tokens. This makes sense to me, at least, as the
76 diff is between these tokens, and thus leads to a nice word-and-space-
77 and-punctuation type diff. It's not unlike what a word processor might
78 do (although a lot of them are character-based, but that seemed a bit
79 extreme--feel free to dupe this module into Text::CharDiff!).
80 Now, I acknowledge that there are localization issues with this
82 approach. In particular, it will fail with Chinese, Japanese, and
83 Korean text, as these languages don't put non-word characters between
84 words. Ideally, Test::WordDiff would then split on every character
85 (since a single character often equals a word), but such is not the
86 case when the "utf8" flag is set on a string. For example, This simple
87 script:
88 use strict;
90 use utf8;
91 use Data::Dumper;
92 my $string = '뼈뼉뼘뼙뼛뼜뼝뽀뽁뽄뽈뽐뽑뽕뾔뾰뿅뿌뿍뿐뿔뿜뿟뿡쀼쁑쁘쁜쁠쁨쁩삐';
93 my @tokens = split /(?<!\p{IsWord})(?=\p{IsWord})/msx, $string;
94 print Dumper \@tokens;
95 Outputs:
97 $VAR1 = [
99 "\x{bf08}\x{bf09}\x{bf18}\x{bf19}\x{bf1b}\x{bf1c}\x{bf1d}\x{bf40}\x{bf41}\x{bf44}\x{bf48}\x{bf50}\x{bf51}\x{bf55}\x{bf94}\x{bfb0}\x{bfc5}\x{bfcc}\x{bfcd}\x{bfd0}\x{bfd4}\x{bfdc}\x{bfdf}\x{bfe1}\x{c03c}\x{c051}\x{c058}\x{c05c}\x{c060}\x{c068}\x{c069}\x{c090}"
100 ];
101 Not so useful. It seems to be less of a problem if the "use utf8;" line
103 is commented out, in which case we get:
104 $VAR1 = [
106 '뼈',
107 '뼉',
108 '뼘',
109 '뼙',
110 '뼛',
111 '뼜',
112 '뼝',
113 '뽀',
114 '뽁',
115 '뽄',
116 '뽈',
117 '뽐',
118 '뽑',
119 '뽕',
120 '뾔',
121 '뾰',
122 '뿅',
123 '뿌',
124 '뿍',
125 '뿐',
126 '뿔',
127 '뿜',
128 '뿟',
129 '뿡',
130 '?',
131 '?쁑',
132 '쁘',
133 '쁜',
134 '쁠',
135 '쁨',
136 '쁩',
137 '삐'
138 ];
139 Someone whose more familiar with non-space-using languages will have to
141 explain to me how I might be able to duplicate this pattern within the
142 scope of "use utf8;", seing as it may very well be important to have it
143 on in order to ensure proper character semantics.
144 However, if my word tokenization approach is just too naive, and you
146 decide that you need to take a different approach (maybe use
147 Lingua::ZH::Toke or similar module), you can still use this module;
148 you'll just have to tokenize your strings into words yourself, and pass
149 them to word_diff() as array references:
150 word_diff \@my_words1, \@my_words2;
152
154 word_diff() takes two arguments from which to draw input and an
155 optional hash reference of options to control its output. The first two
156 arguments contain the data to be diffed, and each may be in the form of
157 any of the following (that is, they can be in two different formats):
158 • String
160 A bare scalar will be assumed to be a file name. The file will be
162 opened and split up into words. word_diff() will also "stat" the
163 file to get the last modified time for use in the header, unless
164 the relevant option ("MTIME_A" or "MTIME_B") has been specified
165 explicitly.
166 • Scalar Reference
168 A scalar reference will be assumed to refer to a string. That
170 string will be split up into words.
171 • Array Reference
173 An array reference will be assumed to be a list of words.
175 • File Handle
177 A glob or IO::Handle-derived object will be read from and split up
179 into its constituent words.
180 The optional hash reference may contain the following options.
182 Additional options may be specified by the formattting class; see the
183 specific class for details.
184 • STYLE
186 "ANSIColor", "HTML" or an object or class name for a class
188 providing file_header(), hunk_header(), same_items(),
189 delete_items(), insert_items(), hunk_footer() and file_footer()
190 methods. Defaults to "ANSIColor" for nice display of diffs in an
191 ANSI Color-supporting terminal.
192 If the package indicated by the "STYLE" has no new() method,
194 word_diff() will load it automatically (lazy loading). It will then
195 instantiate an object of that class, passing in the options hash
196 reference with which the formatting class can initialize the
197 object.
198 Styles may be specified as class names ("STYLE => "My::Foo""), in
200 which case they will be instantiated by calling the new()
201 construcctor and passing in the options hash reference, or as
202 objects ("STYLE => My::Foo->new").
203 The simplest way to implement your own formatting style is to
205 create a new class that inherits from Text::WordDiff::Base, wherein
206 the new() method is already provided, and the file_header() returns
207 a Unified diff-style header. All of the other formatting methods
208 simply return empty strings, and are therefore ripe for overriding.
209 • FILENAME_A, MTIME_A, FILENAME_B, MTIME_B
211 The name of the file and the modification time "files" in epoch
213 seconds. Unless a defined value is specified for these options,
214 they will be filled in for each file when word_diff() is passed a
215 filename. If a filename is not passed in and "FILENAME_A" and
216 "FILENAME_B" are not defined, the header will not be printed by the
217 base formatting base class.
218 • OUTPUT
220 The method by which diff output should be, well, output. Examples
222 and their equivalent subroutines:
223 OUTPUT => \*FOOHANDLE, # like: sub { print FOOHANDLE shift() }
225 OUTPUT => \$output, # like: sub { $output .= shift }
226 OUTPUT => \@output, # like: sub { push @output, shift }
227 OUTPUT => sub { $output .= shift },
228 If "OUTPUT" is not defined, word_diff() will simply return the diff
230 as a string. If "OUTPUT" is a code reference, it will be called
231 once with the file header, once for each hunk body, and once for
232 each piece of content. If "OUTPUT" is an IO::Handle-derived object,
233 output will be sent to that handle.
234 • FILENAME_PREFIX_A, FILENAME_PREFIX_B
236 The string to print before the filename in the header. Defaults are
238 "---", "+++".
239 • DIFF_OPTS
241 A hash reference to be passed as the options to
243 "Algorithm::Diff->new". See Algorithm::Diff for details on
244 available options.
245
247 Text::WordDiff comes with two formatting classes:
248 Text::WordDiff::ANSIColor
250 This is the default formatting class. It emits a header and then
251 the diff content, with deleted text in bodfaced red and inserted
252 text in boldfaced green.
253 Text::WordDiff::HTML
255 Specify "STYLE => 'HTML'" to take advantage of this formatting
256 class. It outputs the diff content as XHTML, with deleted text in
257 "<del>" elements and inserted text in "<ins>" elements.
258 To implement your own formatting class, simply inherit from
260 Text::WordDiff::Base and override its methods as necssary. By default,
261 only the file_header() formatting method returns a value. All others
262 simply return empty strings, and are therefore ripe for overriding:
263 package My::WordDiff::Format;
265 use base 'Text::WordDiff::Base';
266 sub file_footer { return "End of diff\n"; }
268 The methods supplied by the base class are:
270 new()
272 Constructs and returns a new formatting object. It takes a single
273 hash reference as its argument, and uses it to construct the
274 object. The nice thing about this is that if you want to support
275 other options in your formatting class, you can just use them in
276 the formatting object constructed by the Text::WordDiff::Base class
277 and document that they can be passed as part of the options hash
278 refernce to word_diff().
279 file_header()
281 Called once for a single call to word_diff(), this method outputs
282 the header for the whole diff. This is the only formatting method
283 in the base class that returns anything other than an empty string.
284 It collects the filenames from filname_a() and filename_b() and, if
285 they're defined, uses the relevant prefixes and modification times
286 to return a unified diff-style header.
287 hunk_header()
289 This method is called for each diff hunk. It should output any
290 necessary header for the hunk.
291 same_items()
293 This method is called for items that have not changed between the
294 two sequnces being compared. The unchanged items will be passed as
295 a list to the method.
296 "delete_items"
298 This method is called for items in the first sequence that are not
299 present in the second sequcne. The deleted items will be passed as
300 a list to the method.
301 "insert_items"
303 This method is called for items in the second sequence that are not
304 present in the first sequcne. The inserted items will be passed as
305 a list to the method.
306 "hunk_footer"
308 This method is called at the end of a hunk. It should output any
309 necessary content to close out the hunk.
310 file_footer()
312 This method is called once when the whole diff has been procssed.
313 It should output any necessary content to close out the diff file.
314 "filename_a"
316 This accessor returns the value specified for the "FILENAME_A"
317 option to word_diff().
318 "filename_b"
320 This accessor returns the value specified for the "FILENAME_B"
321 option to word_diff().
322 "mtime_a"
324 This accessor returns the value specified for the "MTIME_A" option
325 to word_diff().
326 "mtime_b"
328 This accessor returns the value specified for the "MTIME_B" option
329 to word_diff().
330 "filename_prefix_a"
332 This accessor returns the value specified for the
333 "FILENAME_PREFIX_A" option to word_diff().
334 "filename_prefix_b"
336 This accessor returns the value specified for the
337 "FILENAME_PREFIX_B" option to word_diff().
338
340 Text::Diff
341 Inspired the interface and implementation of this module. Thanks
342 Barry!
343 Text::ParagraphDiff
345 A module that attempts to diff paragraphs and the words in them.
346 Algorithm::Diff
348 The module that makes this all possible.
349
351 This module is stored in an open GitHub repository
352 <http://github.com/theory/text-worddiff/>. Feel free to fork and
353 contribute!
354 Please file bug reports via GitHub Issues
356 <http://github.com/theory/text-worddiff/issues/> or by sending mail to
357 bug-Text-WordDiff@rt.cpan.org <mailto:bug-Text-WordDiff@rt.cpan.org>.
358
360 David E. Wheeler <david@justatheory.com>
361 Currently maintained by the developers of The Perl Shop <tps@cpan.org>.
363
365 Copyright (c) 2005-2011 David E. Wheeler. Some Rights Reserved.
366 This module is free software; you can redistribute it and/or modify it
368 under the same terms as Perl itself.
369perl v5.38.0 2023-07-21 Text::WordDiff(3pm)