1
|
1 |
<?php
|
|
2 |
|
|
3 |
// A PHP diff engine for phpwiki. (Taken from phpwiki-1.3.3)
|
|
4 |
//
|
|
5 |
// Copyright (C) 2000, 2001 Geoffrey T. Dairiki <dairiki@dairiki.org>
|
|
6 |
// You may copy this code freely under the conditions of the GPL.
|
|
7 |
//
|
|
8 |
|
|
9 |
define('USE_ASSERTS', function_exists('assert'));
|
|
10 |
|
|
11 |
/**
|
|
12 |
* @todo document
|
|
13 |
* @package Enano
|
|
14 |
* @subpackage DifferenceEngine
|
|
15 |
*/
|
|
16 |
class _DiffOp {
|
|
17 |
var $type;
|
|
18 |
var $orig;
|
|
19 |
var $closing;
|
|
20 |
|
|
21 |
function reverse() {
|
|
22 |
trigger_error('pure virtual', E_USER_ERROR);
|
|
23 |
}
|
|
24 |
|
|
25 |
function norig() {
|
|
26 |
return $this->orig ? sizeof($this->orig) : 0;
|
|
27 |
}
|
|
28 |
|
|
29 |
function nclosing() {
|
|
30 |
return $this->closing ? sizeof($this->closing) : 0;
|
|
31 |
}
|
|
32 |
}
|
|
33 |
|
|
34 |
/**
|
|
35 |
* @todo document
|
|
36 |
* @package Enano
|
|
37 |
* @subpackage DifferenceEngine
|
|
38 |
*/
|
|
39 |
class _DiffOp_Copy extends _DiffOp {
|
|
40 |
var $type = 'copy';
|
|
41 |
|
|
42 |
function _DiffOp_Copy ($orig, $closing = false) {
|
|
43 |
if (!is_array($closing))
|
|
44 |
$closing = $orig;
|
|
45 |
$this->orig = $orig;
|
|
46 |
$this->closing = $closing;
|
|
47 |
}
|
|
48 |
|
|
49 |
function reverse() {
|
|
50 |
return new _DiffOp_Copy($this->closing, $this->orig);
|
|
51 |
}
|
|
52 |
}
|
|
53 |
|
|
54 |
/**
|
|
55 |
* @todo document
|
|
56 |
* @package Enano
|
|
57 |
* @subpackage DifferenceEngine
|
|
58 |
*/
|
|
59 |
class _DiffOp_Delete extends _DiffOp {
|
|
60 |
var $type = 'delete';
|
|
61 |
|
|
62 |
function _DiffOp_Delete ($lines) {
|
|
63 |
$this->orig = $lines;
|
|
64 |
$this->closing = false;
|
|
65 |
}
|
|
66 |
|
|
67 |
function reverse() {
|
|
68 |
return new _DiffOp_Add($this->orig);
|
|
69 |
}
|
|
70 |
}
|
|
71 |
|
|
72 |
/**
|
|
73 |
* @todo document
|
|
74 |
* @package Enano
|
|
75 |
* @subpackage DifferenceEngine
|
|
76 |
*/
|
|
77 |
class _DiffOp_Add extends _DiffOp {
|
|
78 |
var $type = 'add';
|
|
79 |
|
|
80 |
function _DiffOp_Add ($lines) {
|
|
81 |
$this->closing = $lines;
|
|
82 |
$this->orig = false;
|
|
83 |
}
|
|
84 |
|
|
85 |
function reverse() {
|
|
86 |
return new _DiffOp_Delete($this->closing);
|
|
87 |
}
|
|
88 |
}
|
|
89 |
|
|
90 |
/**
|
|
91 |
* @todo document
|
|
92 |
* @package Enano
|
|
93 |
* @subpackage DifferenceEngine
|
|
94 |
*/
|
|
95 |
class _DiffOp_Change extends _DiffOp {
|
|
96 |
var $type = 'change';
|
|
97 |
|
|
98 |
function _DiffOp_Change ($orig, $closing) {
|
|
99 |
$this->orig = $orig;
|
|
100 |
$this->closing = $closing;
|
|
101 |
}
|
|
102 |
|
|
103 |
function reverse() {
|
|
104 |
return new _DiffOp_Change($this->closing, $this->orig);
|
|
105 |
}
|
|
106 |
}
|
|
107 |
|
|
108 |
|
|
109 |
/**
|
|
110 |
* Class used internally by Diff to actually compute the diffs.
|
|
111 |
*
|
|
112 |
* The algorithm used here is mostly lifted from the perl module
|
|
113 |
* Algorithm::Diff (version 1.06) by Ned Konz, which is available at:
|
|
114 |
* http://www.perl.com/CPAN/authors/id/N/NE/NEDKONZ/Algorithm-Diff-1.06.zip
|
|
115 |
*
|
|
116 |
* More ideas are taken from:
|
|
117 |
* http://www.ics.uci.edu/~eppstein/161/960229.html
|
|
118 |
*
|
|
119 |
* Some ideas are (and a bit of code) are from from analyze.c, from GNU
|
|
120 |
* diffutils-2.7, which can be found at:
|
|
121 |
* ftp://gnudist.gnu.org/pub/gnu/diffutils/diffutils-2.7.tar.gz
|
|
122 |
*
|
|
123 |
* closingly, some ideas (subdivision by NCHUNKS > 2, and some optimizations)
|
|
124 |
* are my own.
|
|
125 |
*
|
|
126 |
* Line length limits for robustness added by Tim Starling, 2005-08-31
|
|
127 |
*
|
|
128 |
* @author Geoffrey T. Dairiki, Tim Starling
|
|
129 |
* @package Enano
|
|
130 |
* @subpackage DifferenceEngine
|
|
131 |
*/
|
|
132 |
define('MAX_XREF_LENGTH', 10000);
|
|
133 |
class _DiffEngine
|
|
134 |
{
|
|
135 |
function diff ($from_lines, $to_lines) {
|
|
136 |
$fname = '_DiffEngine::diff';
|
|
137 |
// wfProfileIn( $fname );
|
|
138 |
|
|
139 |
$n_from = sizeof($from_lines);
|
|
140 |
$n_to = sizeof($to_lines);
|
|
141 |
|
|
142 |
$this->xchanged = $this->ychanged = array();
|
|
143 |
$this->xv = $this->yv = array();
|
|
144 |
$this->xind = $this->yind = array();
|
|
145 |
unset($this->seq);
|
|
146 |
unset($this->in_seq);
|
|
147 |
unset($this->lcs);
|
|
148 |
|
|
149 |
// Skip leading common lines.
|
|
150 |
for ($skip = 0; $skip < $n_from && $skip < $n_to; $skip++) {
|
|
151 |
if ($from_lines[$skip] !== $to_lines[$skip])
|
|
152 |
break;
|
|
153 |
$this->xchanged[$skip] = $this->ychanged[$skip] = false;
|
|
154 |
}
|
|
155 |
// Skip trailing common lines.
|
|
156 |
$xi = $n_from; $yi = $n_to;
|
|
157 |
for ($endskip = 0; --$xi > $skip && --$yi > $skip; $endskip++) {
|
|
158 |
if ($from_lines[$xi] !== $to_lines[$yi])
|
|
159 |
break;
|
|
160 |
$this->xchanged[$xi] = $this->ychanged[$yi] = false;
|
|
161 |
}
|
|
162 |
|
|
163 |
// Ignore lines which do not exist in both files.
|
|
164 |
for ($xi = $skip; $xi < $n_from - $endskip; $xi++) {
|
|
165 |
$xhash[$this->_line_hash($from_lines[$xi])] = 1;
|
|
166 |
}
|
|
167 |
|
|
168 |
for ($yi = $skip; $yi < $n_to - $endskip; $yi++) {
|
|
169 |
$line = $to_lines[$yi];
|
|
170 |
if ( ($this->ychanged[$yi] = empty($xhash[$this->_line_hash($line)])) )
|
|
171 |
continue;
|
|
172 |
$yhash[$this->_line_hash($line)] = 1;
|
|
173 |
$this->yv[] = $line;
|
|
174 |
$this->yind[] = $yi;
|
|
175 |
}
|
|
176 |
for ($xi = $skip; $xi < $n_from - $endskip; $xi++) {
|
|
177 |
$line = $from_lines[$xi];
|
|
178 |
if ( ($this->xchanged[$xi] = empty($yhash[$this->_line_hash($line)])) )
|
|
179 |
continue;
|
|
180 |
$this->xv[] = $line;
|
|
181 |
$this->xind[] = $xi;
|
|
182 |
}
|
|
183 |
|
|
184 |
// Find the LCS.
|
|
185 |
$this->_compareseq(0, sizeof($this->xv), 0, sizeof($this->yv));
|
|
186 |
|
|
187 |
// Merge edits when possible
|
|
188 |
$this->_shift_boundaries($from_lines, $this->xchanged, $this->ychanged);
|
|
189 |
$this->_shift_boundaries($to_lines, $this->ychanged, $this->xchanged);
|
|
190 |
|
|
191 |
// Compute the edit operations.
|
|
192 |
$edits = array();
|
|
193 |
$xi = $yi = 0;
|
|
194 |
while ($xi < $n_from || $yi < $n_to) {
|
|
195 |
USE_ASSERTS && assert($yi < $n_to || $this->xchanged[$xi]);
|
|
196 |
USE_ASSERTS && assert($xi < $n_from || $this->ychanged[$yi]);
|
|
197 |
|
|
198 |
// Skip matching "snake".
|
|
199 |
$copy = array();
|
|
200 |
while ( $xi < $n_from && $yi < $n_to
|
|
201 |
&& !$this->xchanged[$xi] && !$this->ychanged[$yi]) {
|
|
202 |
$copy[] = $from_lines[$xi++];
|
|
203 |
++$yi;
|
|
204 |
}
|
|
205 |
if ($copy)
|
|
206 |
$edits[] = new _DiffOp_Copy($copy);
|
|
207 |
|
|
208 |
// Find deletes & adds.
|
|
209 |
$delete = array();
|
|
210 |
while ($xi < $n_from && $this->xchanged[$xi])
|
|
211 |
$delete[] = $from_lines[$xi++];
|
|
212 |
|
|
213 |
$add = array();
|
|
214 |
while ($yi < $n_to && $this->ychanged[$yi])
|
|
215 |
$add[] = $to_lines[$yi++];
|
|
216 |
|
|
217 |
if ($delete && $add)
|
|
218 |
$edits[] = new _DiffOp_Change($delete, $add);
|
|
219 |
elseif ($delete)
|
|
220 |
$edits[] = new _DiffOp_Delete($delete);
|
|
221 |
elseif ($add)
|
|
222 |
$edits[] = new _DiffOp_Add($add);
|
|
223 |
}
|
|
224 |
// wfProfileOut( $fname );
|
|
225 |
return $edits;
|
|
226 |
}
|
|
227 |
|
|
228 |
/**
|
|
229 |
* Returns the whole line if it's small enough, or the MD5 hash otherwise
|
|
230 |
*/
|
|
231 |
function _line_hash( $line ) {
|
|
232 |
if ( strlen( $line ) > MAX_XREF_LENGTH ) {
|
|
233 |
return md5( $line );
|
|
234 |
} else {
|
|
235 |
return $line;
|
|
236 |
}
|
|
237 |
}
|
|
238 |
|
|
239 |
|
|
240 |
/* Divide the Largest Common Subsequence (LCS) of the sequences
|
|
241 |
* [XOFF, XLIM) and [YOFF, YLIM) into NCHUNKS approximately equally
|
|
242 |
* sized segments.
|
|
243 |
*
|
|
244 |
* Returns (LCS, PTS). LCS is the length of the LCS. PTS is an
|
|
245 |
* array of NCHUNKS+1 (X, Y) indexes giving the diving points between
|
|
246 |
* sub sequences. The first sub-sequence is contained in [X0, X1),
|
|
247 |
* [Y0, Y1), the second in [X1, X2), [Y1, Y2) and so on. Note
|
|
248 |
* that (X0, Y0) == (XOFF, YOFF) and
|
|
249 |
* (X[NCHUNKS], Y[NCHUNKS]) == (XLIM, YLIM).
|
|
250 |
*
|
|
251 |
* This function assumes that the first lines of the specified portions
|
|
252 |
* of the two files do not match, and likewise that the last lines do not
|
|
253 |
* match. The caller must trim matching lines from the beginning and end
|
|
254 |
* of the portions it is going to specify.
|
|
255 |
*/
|
|
256 |
function _diag ($xoff, $xlim, $yoff, $ylim, $nchunks) {
|
|
257 |
$fname = '_DiffEngine::_diag';
|
|
258 |
// wfProfileIn( $fname );
|
|
259 |
$flip = false;
|
|
260 |
|
|
261 |
if ($xlim - $xoff > $ylim - $yoff) {
|
|
262 |
// Things seems faster (I'm not sure I understand why)
|
|
263 |
// when the shortest sequence in X.
|
|
264 |
$flip = true;
|
|
265 |
list ($xoff, $xlim, $yoff, $ylim)
|
|
266 |
= array( $yoff, $ylim, $xoff, $xlim);
|
|
267 |
}
|
|
268 |
|
|
269 |
if ($flip)
|
|
270 |
for ($i = $ylim - 1; $i >= $yoff; $i--)
|
|
271 |
$ymatches[$this->xv[$i]][] = $i;
|
|
272 |
else
|
|
273 |
for ($i = $ylim - 1; $i >= $yoff; $i--)
|
|
274 |
$ymatches[$this->yv[$i]][] = $i;
|
|
275 |
|
|
276 |
$this->lcs = 0;
|
|
277 |
$this->seq[0]= $yoff - 1;
|
|
278 |
$this->in_seq = array();
|
|
279 |
$ymids[0] = array();
|
|
280 |
|
|
281 |
$numer = $xlim - $xoff + $nchunks - 1;
|
|
282 |
$x = $xoff;
|
|
283 |
for ($chunk = 0; $chunk < $nchunks; $chunk++) {
|
|
284 |
// wfProfileIn( "$fname-chunk" );
|
|
285 |
if ($chunk > 0)
|
|
286 |
for ($i = 0; $i <= $this->lcs; $i++)
|
|
287 |
$ymids[$i][$chunk-1] = $this->seq[$i];
|
|
288 |
|
|
289 |
$x1 = $xoff + (int)(($numer + ($xlim-$xoff)*$chunk) / $nchunks);
|
|
290 |
for ( ; $x < $x1; $x++) {
|
|
291 |
$line = $flip ? $this->yv[$x] : $this->xv[$x];
|
|
292 |
if (empty($ymatches[$line]))
|
|
293 |
continue;
|
|
294 |
$matches = $ymatches[$line];
|
|
295 |
reset($matches);
|
|
296 |
while (list ($junk, $y) = each($matches))
|
|
297 |
if (empty($this->in_seq[$y])) {
|
|
298 |
$k = $this->_lcs_pos($y);
|
|
299 |
USE_ASSERTS && assert($k > 0);
|
|
300 |
$ymids[$k] = $ymids[$k-1];
|
|
301 |
break;
|
|
302 |
}
|
|
303 |
while (list ($junk, $y) = each($matches)) {
|
|
304 |
if ($y > $this->seq[$k-1]) {
|
|
305 |
USE_ASSERTS && assert($y < $this->seq[$k]);
|
|
306 |
// Optimization: this is a common case:
|
|
307 |
// next match is just replacing previous match.
|
|
308 |
$this->in_seq[$this->seq[$k]] = false;
|
|
309 |
$this->seq[$k] = $y;
|
|
310 |
$this->in_seq[$y] = 1;
|
|
311 |
} else if (empty($this->in_seq[$y])) {
|
|
312 |
$k = $this->_lcs_pos($y);
|
|
313 |
USE_ASSERTS && assert($k > 0);
|
|
314 |
$ymids[$k] = $ymids[$k-1];
|
|
315 |
}
|
|
316 |
}
|
|
317 |
}
|
|
318 |
// wfProfileOut( "$fname-chunk" );
|
|
319 |
}
|
|
320 |
|
|
321 |
$seps[] = $flip ? array($yoff, $xoff) : array($xoff, $yoff);
|
|
322 |
$ymid = $ymids[$this->lcs];
|
|
323 |
for ($n = 0; $n < $nchunks - 1; $n++) {
|
|
324 |
$x1 = $xoff + (int)(($numer + ($xlim - $xoff) * $n) / $nchunks);
|
|
325 |
$y1 = $ymid[$n] + 1;
|
|
326 |
$seps[] = $flip ? array($y1, $x1) : array($x1, $y1);
|
|
327 |
}
|
|
328 |
$seps[] = $flip ? array($ylim, $xlim) : array($xlim, $ylim);
|
|
329 |
|
|
330 |
// wfProfileOut( $fname );
|
|
331 |
return array($this->lcs, $seps);
|
|
332 |
}
|
|
333 |
|
|
334 |
function _lcs_pos ($ypos) {
|
|
335 |
$fname = '_DiffEngine::_lcs_pos';
|
|
336 |
// wfProfileIn( $fname );
|
|
337 |
|
|
338 |
$end = $this->lcs;
|
|
339 |
if ($end == 0 || $ypos > $this->seq[$end]) {
|
|
340 |
$this->seq[++$this->lcs] = $ypos;
|
|
341 |
$this->in_seq[$ypos] = 1;
|
|
342 |
// wfProfileOut( $fname );
|
|
343 |
return $this->lcs;
|
|
344 |
}
|
|
345 |
|
|
346 |
$beg = 1;
|
|
347 |
while ($beg < $end) {
|
|
348 |
$mid = (int)(($beg + $end) / 2);
|
|
349 |
if ( $ypos > $this->seq[$mid] )
|
|
350 |
$beg = $mid + 1;
|
|
351 |
else
|
|
352 |
$end = $mid;
|
|
353 |
}
|
|
354 |
|
|
355 |
USE_ASSERTS && assert($ypos != $this->seq[$end]);
|
|
356 |
|
|
357 |
$this->in_seq[$this->seq[$end]] = false;
|
|
358 |
$this->seq[$end] = $ypos;
|
|
359 |
$this->in_seq[$ypos] = 1;
|
|
360 |
// wfProfileOut( $fname );
|
|
361 |
return $end;
|
|
362 |
}
|
|
363 |
|
|
364 |
/* Find LCS of two sequences.
|
|
365 |
*
|
|
366 |
* The results are recorded in the vectors $this->{x,y}changed[], by
|
|
367 |
* storing a 1 in the element for each line that is an insertion
|
|
368 |
* or deletion (ie. is not in the LCS).
|
|
369 |
*
|
|
370 |
* The subsequence of file 0 is [XOFF, XLIM) and likewise for file 1.
|
|
371 |
*
|
|
372 |
* Note that XLIM, YLIM are exclusive bounds.
|
|
373 |
* All line numbers are origin-0 and discarded lines are not counted.
|
|
374 |
*/
|
|
375 |
function _compareseq ($xoff, $xlim, $yoff, $ylim) {
|
|
376 |
$fname = '_DiffEngine::_compareseq';
|
|
377 |
// wfProfileIn( $fname );
|
|
378 |
|
|
379 |
// Slide down the bottom initial diagonal.
|
|
380 |
while ($xoff < $xlim && $yoff < $ylim
|
|
381 |
&& $this->xv[$xoff] == $this->yv[$yoff]) {
|
|
382 |
++$xoff;
|
|
383 |
++$yoff;
|
|
384 |
}
|
|
385 |
|
|
386 |
// Slide up the top initial diagonal.
|
|
387 |
while ($xlim > $xoff && $ylim > $yoff
|
|
388 |
&& $this->xv[$xlim - 1] == $this->yv[$ylim - 1]) {
|
|
389 |
--$xlim;
|
|
390 |
--$ylim;
|
|
391 |
}
|
|
392 |
|
|
393 |
if ($xoff == $xlim || $yoff == $ylim)
|
|
394 |
$lcs = 0;
|
|
395 |
else {
|
|
396 |
// This is ad hoc but seems to work well.
|
|
397 |
//$nchunks = sqrt(min($xlim - $xoff, $ylim - $yoff) / 2.5);
|
|
398 |
//$nchunks = max(2,min(8,(int)$nchunks));
|
|
399 |
$nchunks = min(7, $xlim - $xoff, $ylim - $yoff) + 1;
|
|
400 |
list ($lcs, $seps)
|
|
401 |
= $this->_diag($xoff,$xlim,$yoff, $ylim,$nchunks);
|
|
402 |
}
|
|
403 |
|
|
404 |
if ($lcs == 0) {
|
|
405 |
// X and Y sequences have no common subsequence:
|
|
406 |
// mark all changed.
|
|
407 |
while ($yoff < $ylim)
|
|
408 |
$this->ychanged[$this->yind[$yoff++]] = 1;
|
|
409 |
while ($xoff < $xlim)
|
|
410 |
$this->xchanged[$this->xind[$xoff++]] = 1;
|
|
411 |
} else {
|
|
412 |
// Use the partitions to split this problem into subproblems.
|
|
413 |
reset($seps);
|
|
414 |
$pt1 = $seps[0];
|
|
415 |
while ($pt2 = next($seps)) {
|
|
416 |
$this->_compareseq ($pt1[0], $pt2[0], $pt1[1], $pt2[1]);
|
|
417 |
$pt1 = $pt2;
|
|
418 |
}
|
|
419 |
}
|
|
420 |
// wfProfileOut( $fname );
|
|
421 |
}
|
|
422 |
|
|
423 |
/* Adjust inserts/deletes of identical lines to join changes
|
|
424 |
* as much as possible.
|
|
425 |
*
|
|
426 |
* We do something when a run of changed lines include a
|
|
427 |
* line at one end and has an excluded, identical line at the other.
|
|
428 |
* We are free to choose which identical line is included.
|
|
429 |
* `compareseq' usually chooses the one at the beginning,
|
|
430 |
* but usually it is cleaner to consider the following identical line
|
|
431 |
* to be the "change".
|
|
432 |
*
|
|
433 |
* This is extracted verbatim from analyze.c (GNU diffutils-2.7).
|
|
434 |
*/
|
|
435 |
function _shift_boundaries ($lines, &$changed, $other_changed) {
|
|
436 |
$fname = '_DiffEngine::_shift_boundaries';
|
|
437 |
// wfProfileIn( $fname );
|
|
438 |
$i = 0;
|
|
439 |
$j = 0;
|
|
440 |
|
|
441 |
USE_ASSERTS && assert('sizeof($lines) == sizeof($changed)');
|
|
442 |
$len = sizeof($lines);
|
|
443 |
$other_len = sizeof($other_changed);
|
|
444 |
|
|
445 |
while (1) {
|
|
446 |
/*
|
|
447 |
* Scan forwards to find beginning of another run of changes.
|
|
448 |
* Also keep track of the corresponding point in the other file.
|
|
449 |
*
|
|
450 |
* Throughout this code, $i and $j are adjusted together so that
|
|
451 |
* the first $i elements of $changed and the first $j elements
|
|
452 |
* of $other_changed both contain the same number of zeros
|
|
453 |
* (unchanged lines).
|
|
454 |
* Furthermore, $j is always kept so that $j == $other_len or
|
|
455 |
* $other_changed[$j] == false.
|
|
456 |
*/
|
|
457 |
while ($j < $other_len && $other_changed[$j])
|
|
458 |
$j++;
|
|
459 |
|
|
460 |
while ($i < $len && ! $changed[$i]) {
|
|
461 |
USE_ASSERTS && assert('$j < $other_len && ! $other_changed[$j]');
|
|
462 |
$i++; $j++;
|
|
463 |
while ($j < $other_len && $other_changed[$j])
|
|
464 |
$j++;
|
|
465 |
}
|
|
466 |
|
|
467 |
if ($i == $len)
|
|
468 |
break;
|
|
469 |
|
|
470 |
$start = $i;
|
|
471 |
|
|
472 |
// Find the end of this run of changes.
|
|
473 |
while (++$i < $len && $changed[$i])
|
|
474 |
continue;
|
|
475 |
|
|
476 |
do {
|
|
477 |
/*
|
|
478 |
* Record the length of this run of changes, so that
|
|
479 |
* we can later determine whether the run has grown.
|
|
480 |
*/
|
|
481 |
$runlength = $i - $start;
|
|
482 |
|
|
483 |
/*
|
|
484 |
* Move the changed region back, so long as the
|
|
485 |
* previous unchanged line matches the last changed one.
|
|
486 |
* This merges with previous changed regions.
|
|
487 |
*/
|
|
488 |
while ($start > 0 && $lines[$start - 1] == $lines[$i - 1]) {
|
|
489 |
$changed[--$start] = 1;
|
|
490 |
$changed[--$i] = false;
|
|
491 |
while ($start > 0 && $changed[$start - 1])
|
|
492 |
$start--;
|
|
493 |
USE_ASSERTS && assert('$j > 0');
|
|
494 |
while ($other_changed[--$j])
|
|
495 |
continue;
|
|
496 |
USE_ASSERTS && assert('$j >= 0 && !$other_changed[$j]');
|
|
497 |
}
|
|
498 |
|
|
499 |
/*
|
|
500 |
* Set CORRESPONDING to the end of the changed run, at the last
|
|
501 |
* point where it corresponds to a changed run in the other file.
|
|
502 |
* CORRESPONDING == LEN means no such point has been found.
|
|
503 |
*/
|
|
504 |
$corresponding = $j < $other_len ? $i : $len;
|
|
505 |
|
|
506 |
/*
|
|
507 |
* Move the changed region forward, so long as the
|
|
508 |
* first changed line matches the following unchanged one.
|
|
509 |
* This merges with following changed regions.
|
|
510 |
* Do this second, so that if there are no merges,
|
|
511 |
* the changed region is moved forward as far as possible.
|
|
512 |
*/
|
|
513 |
while ($i < $len && $lines[$start] == $lines[$i]) {
|
|
514 |
$changed[$start++] = false;
|
|
515 |
$changed[$i++] = 1;
|
|
516 |
while ($i < $len && $changed[$i])
|
|
517 |
$i++;
|
|
518 |
|
|
519 |
USE_ASSERTS && assert('$j < $other_len && ! $other_changed[$j]');
|
|
520 |
$j++;
|
|
521 |
if ($j < $other_len && $other_changed[$j]) {
|
|
522 |
$corresponding = $i;
|
|
523 |
while ($j < $other_len && $other_changed[$j])
|
|
524 |
$j++;
|
|
525 |
}
|
|
526 |
}
|
|
527 |
} while ($runlength != $i - $start);
|
|
528 |
|
|
529 |
/*
|
|
530 |
* If possible, move the fully-merged run of changes
|
|
531 |
* back to a corresponding run in the other file.
|
|
532 |
*/
|
|
533 |
while ($corresponding < $i) {
|
|
534 |
$changed[--$start] = 1;
|
|
535 |
$changed[--$i] = 0;
|
|
536 |
USE_ASSERTS && assert('$j > 0');
|
|
537 |
while ($other_changed[--$j])
|
|
538 |
continue;
|
|
539 |
USE_ASSERTS && assert('$j >= 0 && !$other_changed[$j]');
|
|
540 |
}
|
|
541 |
}
|
|
542 |
// wfProfileOut( $fname );
|
|
543 |
}
|
|
544 |
}
|
|
545 |
|
|
546 |
/**
|
|
547 |
* Class representing a 'diff' between two sequences of strings.
|
|
548 |
* @todo document
|
|
549 |
* @package Enano
|
|
550 |
* @subpackage DifferenceEngine
|
|
551 |
*/
|
|
552 |
class Diff
|
|
553 |
{
|
|
554 |
var $edits;
|
|
555 |
|
|
556 |
/**
|
|
557 |
* Constructor.
|
|
558 |
* Computes diff between sequences of strings.
|
|
559 |
*
|
|
560 |
* @param $from_lines array An array of strings.
|
|
561 |
* (Typically these are lines from a file.)
|
|
562 |
* @param $to_lines array An array of strings.
|
|
563 |
*/
|
|
564 |
function Diff($from_lines, $to_lines) {
|
|
565 |
$eng = new _DiffEngine;
|
|
566 |
$this->edits = $eng->diff($from_lines, $to_lines);
|
|
567 |
//$this->_check($from_lines, $to_lines);
|
|
568 |
}
|
|
569 |
|
|
570 |
/**
|
|
571 |
* Compute reversed Diff.
|
|
572 |
*
|
|
573 |
* SYNOPSIS:
|
|
574 |
*
|
|
575 |
* $diff = new Diff($lines1, $lines2);
|
|
576 |
* $rev = $diff->reverse();
|
|
577 |
* @return object A Diff object representing the inverse of the
|
|
578 |
* original diff.
|
|
579 |
*/
|
|
580 |
function reverse () {
|
|
581 |
$rev = $this;
|
|
582 |
$rev->edits = array();
|
|
583 |
foreach ($this->edits as $edit) {
|
|
584 |
$rev->edits[] = $edit->reverse();
|
|
585 |
}
|
|
586 |
return $rev;
|
|
587 |
}
|
|
588 |
|
|
589 |
/**
|
|
590 |
* Check for empty diff.
|
|
591 |
*
|
|
592 |
* @return bool True iff two sequences were identical.
|
|
593 |
*/
|
|
594 |
function isEmpty () {
|
|
595 |
foreach ($this->edits as $edit) {
|
|
596 |
if ($edit->type != 'copy')
|
|
597 |
return false;
|
|
598 |
}
|
|
599 |
return true;
|
|
600 |
}
|
|
601 |
|
|
602 |
/**
|
|
603 |
* Compute the length of the Longest Common Subsequence (LCS).
|
|
604 |
*
|
|
605 |
* This is mostly for diagnostic purposes.
|
|
606 |
*
|
|
607 |
* @return int The length of the LCS.
|
|
608 |
*/
|
|
609 |
function lcs () {
|
|
610 |
$lcs = 0;
|
|
611 |
foreach ($this->edits as $edit) {
|
|
612 |
if ($edit->type == 'copy')
|
|
613 |
$lcs += sizeof($edit->orig);
|
|
614 |
}
|
|
615 |
return $lcs;
|
|
616 |
}
|
|
617 |
|
|
618 |
/**
|
|
619 |
* Get the original set of lines.
|
|
620 |
*
|
|
621 |
* This reconstructs the $from_lines parameter passed to the
|
|
622 |
* constructor.
|
|
623 |
*
|
|
624 |
* @return array The original sequence of strings.
|
|
625 |
*/
|
|
626 |
function orig() {
|
|
627 |
$lines = array();
|
|
628 |
|
|
629 |
foreach ($this->edits as $edit) {
|
|
630 |
if ($edit->orig)
|
|
631 |
array_splice($lines, sizeof($lines), 0, $edit->orig);
|
|
632 |
}
|
|
633 |
return $lines;
|
|
634 |
}
|
|
635 |
|
|
636 |
/**
|
|
637 |
* Get the closing set of lines.
|
|
638 |
*
|
|
639 |
* This reconstructs the $to_lines parameter passed to the
|
|
640 |
* constructor.
|
|
641 |
*
|
|
642 |
* @return array The sequence of strings.
|
|
643 |
*/
|
|
644 |
function closing() {
|
|
645 |
$lines = array();
|
|
646 |
|
|
647 |
foreach ($this->edits as $edit) {
|
|
648 |
if ($edit->closing)
|
|
649 |
array_splice($lines, sizeof($lines), 0, $edit->closing);
|
|
650 |
}
|
|
651 |
return $lines;
|
|
652 |
}
|
|
653 |
|
|
654 |
/**
|
|
655 |
* Check a Diff for validity.
|
|
656 |
*
|
|
657 |
* This is here only for debugging purposes.
|
|
658 |
*/
|
|
659 |
function _check ($from_lines, $to_lines) {
|
|
660 |
$fname = 'Diff::_check';
|
|
661 |
// wfProfileIn( $fname );
|
|
662 |
if (serialize($from_lines) != serialize($this->orig()))
|
|
663 |
trigger_error("Reconstructed original doesn't match", E_USER_ERROR);
|
|
664 |
if (serialize($to_lines) != serialize($this->closing()))
|
|
665 |
trigger_error("Reconstructed closing doesn't match", E_USER_ERROR);
|
|
666 |
|
|
667 |
$rev = $this->reverse();
|
|
668 |
if (serialize($to_lines) != serialize($rev->orig()))
|
|
669 |
trigger_error("Reversed original doesn't match", E_USER_ERROR);
|
|
670 |
if (serialize($from_lines) != serialize($rev->closing()))
|
|
671 |
trigger_error("Reversed closing doesn't match", E_USER_ERROR);
|
|
672 |
|
|
673 |
|
|
674 |
$prevtype = 'none';
|
|
675 |
foreach ($this->edits as $edit) {
|
|
676 |
if ( $prevtype == $edit->type )
|
|
677 |
trigger_error("Edit sequence is non-optimal", E_USER_ERROR);
|
|
678 |
$prevtype = $edit->type;
|
|
679 |
}
|
|
680 |
|
|
681 |
$lcs = $this->lcs();
|
|
682 |
trigger_error('Diff okay: LCS = '.$lcs, E_USER_NOTICE);
|
|
683 |
// wfProfileOut( $fname );
|
|
684 |
}
|
|
685 |
}
|
|
686 |
|
|
687 |
/**
|
|
688 |
* FIXME: bad name.
|
|
689 |
* @todo document
|
|
690 |
* @package Enano
|
|
691 |
* @subpackage DifferenceEngine
|
|
692 |
*/
|
|
693 |
class MappedDiff extends Diff
|
|
694 |
{
|
|
695 |
/**
|
|
696 |
* Constructor.
|
|
697 |
*
|
|
698 |
* Computes diff between sequences of strings.
|
|
699 |
*
|
|
700 |
* This can be used to compute things like
|
|
701 |
* case-insensitve diffs, or diffs which ignore
|
|
702 |
* changes in white-space.
|
|
703 |
*
|
|
704 |
* @param $from_lines array An array of strings.
|
|
705 |
* (Typically these are lines from a file.)
|
|
706 |
*
|
|
707 |
* @param $to_lines array An array of strings.
|
|
708 |
*
|
|
709 |
* @param $mapped_from_lines array This array should
|
|
710 |
* have the same size number of elements as $from_lines.
|
|
711 |
* The elements in $mapped_from_lines and
|
|
712 |
* $mapped_to_lines are what is actually compared
|
|
713 |
* when computing the diff.
|
|
714 |
*
|
|
715 |
* @param $mapped_to_lines array This array should
|
|
716 |
* have the same number of elements as $to_lines.
|
|
717 |
*/
|
|
718 |
function MappedDiff($from_lines, $to_lines,
|
|
719 |
$mapped_from_lines, $mapped_to_lines) {
|
|
720 |
$fname = 'MappedDiff::MappedDiff';
|
|
721 |
// wfProfileIn( $fname );
|
|
722 |
|
|
723 |
assert(sizeof($from_lines) == sizeof($mapped_from_lines));
|
|
724 |
assert(sizeof($to_lines) == sizeof($mapped_to_lines));
|
|
725 |
|
|
726 |
$this->Diff($mapped_from_lines, $mapped_to_lines);
|
|
727 |
|
|
728 |
$xi = $yi = 0;
|
|
729 |
for ($i = 0; $i < sizeof($this->edits); $i++) {
|
|
730 |
$orig = &$this->edits[$i]->orig;
|
|
731 |
if (is_array($orig)) {
|
|
732 |
$orig = array_slice($from_lines, $xi, sizeof($orig));
|
|
733 |
$xi += sizeof($orig);
|
|
734 |
}
|
|
735 |
|
|
736 |
$closing = &$this->edits[$i]->closing;
|
|
737 |
if (is_array($closing)) {
|
|
738 |
$closing = array_slice($to_lines, $yi, sizeof($closing));
|
|
739 |
$yi += sizeof($closing);
|
|
740 |
}
|
|
741 |
}
|
|
742 |
// wfProfileOut( $fname );
|
|
743 |
}
|
|
744 |
}
|
|
745 |
|
|
746 |
/**
|
|
747 |
* A class to format Diffs
|
|
748 |
*
|
|
749 |
* This class formats the diff in classic diff format.
|
|
750 |
* It is intended that this class be customized via inheritance,
|
|
751 |
* to obtain fancier outputs.
|
|
752 |
* @todo document
|
|
753 |
* @package Enano
|
|
754 |
* @subpackage DifferenceEngine
|
|
755 |
*/
|
|
756 |
class DiffFormatter
|
|
757 |
{
|
|
758 |
/**
|
|
759 |
* Number of leading context "lines" to preserve.
|
|
760 |
*
|
|
761 |
* This should be left at zero for this class, but subclasses
|
|
762 |
* may want to set this to other values.
|
|
763 |
*/
|
|
764 |
var $leading_context_lines = 0;
|
|
765 |
|
|
766 |
/**
|
|
767 |
* Number of trailing context "lines" to preserve.
|
|
768 |
*
|
|
769 |
* This should be left at zero for this class, but subclasses
|
|
770 |
* may want to set this to other values.
|
|
771 |
*/
|
|
772 |
var $trailing_context_lines = 0;
|
|
773 |
|
|
774 |
/**
|
|
775 |
* Format a diff.
|
|
776 |
*
|
|
777 |
* @param $diff object A Diff object.
|
|
778 |
* @return string The formatted output.
|
|
779 |
*/
|
|
780 |
function format($diff) {
|
|
781 |
$fname = 'DiffFormatter::format';
|
|
782 |
// wfProfileIn( $fname );
|
|
783 |
|
|
784 |
$xi = $yi = 1;
|
|
785 |
$block = false;
|
|
786 |
$context = array();
|
|
787 |
|
|
788 |
$nlead = $this->leading_context_lines;
|
|
789 |
$ntrail = $this->trailing_context_lines;
|
|
790 |
|
|
791 |
$this->_start_diff();
|
|
792 |
|
|
793 |
foreach ($diff->edits as $edit) {
|
|
794 |
if ($edit->type == 'copy') {
|
|
795 |
if (is_array($block)) {
|
|
796 |
if (sizeof($edit->orig) <= $nlead + $ntrail) {
|
|
797 |
$block[] = $edit;
|
|
798 |
}
|
|
799 |
else{
|
|
800 |
if ($ntrail) {
|
|
801 |
$context = array_slice($edit->orig, 0, $ntrail);
|
|
802 |
$block[] = new _DiffOp_Copy($context);
|
|
803 |
}
|
|
804 |
$this->_block($x0, $ntrail + $xi - $x0,
|
|
805 |
$y0, $ntrail + $yi - $y0,
|
|
806 |
$block);
|
|
807 |
$block = false;
|
|
808 |
}
|
|
809 |
}
|
|
810 |
$context = $edit->orig;
|
|
811 |
}
|
|
812 |
else {
|
|
813 |
if (! is_array($block)) {
|
|
814 |
$context = array_slice($context, sizeof($context) - $nlead);
|
|
815 |
$x0 = $xi - sizeof($context);
|
|
816 |
$y0 = $yi - sizeof($context);
|
|
817 |
$block = array();
|
|
818 |
if ($context)
|
|
819 |
$block[] = new _DiffOp_Copy($context);
|
|
820 |
}
|
|
821 |
$block[] = $edit;
|
|
822 |
}
|
|
823 |
|
|
824 |
if ($edit->orig)
|
|
825 |
$xi += sizeof($edit->orig);
|
|
826 |
if ($edit->closing)
|
|
827 |
$yi += sizeof($edit->closing);
|
|
828 |
}
|
|
829 |
|
|
830 |
if (is_array($block))
|
|
831 |
$this->_block($x0, $xi - $x0,
|
|
832 |
$y0, $yi - $y0,
|
|
833 |
$block);
|
|
834 |
|
|
835 |
$end = $this->_end_diff();
|
|
836 |
// wfProfileOut( $fname );
|
|
837 |
return $end;
|
|
838 |
}
|
|
839 |
|
|
840 |
function _block($xbeg, $xlen, $ybeg, $ylen, &$edits) {
|
|
841 |
$fname = 'DiffFormatter::_block';
|
|
842 |
// wfProfileIn( $fname );
|
|
843 |
$this->_start_block($this->_block_header($xbeg, $xlen, $ybeg, $ylen));
|
|
844 |
foreach ($edits as $edit) {
|
|
845 |
if ($edit->type == 'copy')
|
|
846 |
$this->_context($edit->orig);
|
|
847 |
elseif ($edit->type == 'add')
|
|
848 |
$this->_added($edit->closing);
|
|
849 |
elseif ($edit->type == 'delete')
|
|
850 |
$this->_deleted($edit->orig);
|
|
851 |
elseif ($edit->type == 'change')
|
|
852 |
$this->_changed($edit->orig, $edit->closing);
|
|
853 |
else
|
|
854 |
trigger_error('Unknown edit type', E_USER_ERROR);
|
|
855 |
}
|
|
856 |
$this->_end_block();
|
|
857 |
// wfProfileOut( $fname );
|
|
858 |
}
|
|
859 |
|
|
860 |
function _start_diff() {
|
|
861 |
ob_start();
|
|
862 |
}
|
|
863 |
|
|
864 |
function _end_diff() {
|
|
865 |
$val = ob_get_contents();
|
|
866 |
ob_end_clean();
|
|
867 |
return $val;
|
|
868 |
}
|
|
869 |
|
|
870 |
function _block_header($xbeg, $xlen, $ybeg, $ylen) {
|
|
871 |
if ($xlen > 1)
|
|
872 |
$xbeg .= "," . ($xbeg + $xlen - 1);
|
|
873 |
if ($ylen > 1)
|
|
874 |
$ybeg .= "," . ($ybeg + $ylen - 1);
|
|
875 |
|
|
876 |
return $xbeg . ($xlen ? ($ylen ? 'c' : 'd') : 'a') . $ybeg;
|
|
877 |
}
|
|
878 |
|
|
879 |
function _start_block($header) {
|
|
880 |
echo $header;
|
|
881 |
}
|
|
882 |
|
|
883 |
function _end_block() {
|
|
884 |
}
|
|
885 |
|
|
886 |
function _lines($lines, $prefix = ' ') {
|
|
887 |
foreach ($lines as $line)
|
|
888 |
echo "$prefix $line\n";
|
|
889 |
}
|
|
890 |
|
|
891 |
function _context($lines) {
|
|
892 |
$this->_lines($lines);
|
|
893 |
}
|
|
894 |
|
|
895 |
function _added($lines) {
|
|
896 |
$this->_lines($lines, '>');
|
|
897 |
}
|
|
898 |
function _deleted($lines) {
|
|
899 |
$this->_lines($lines, '<');
|
|
900 |
}
|
|
901 |
|
|
902 |
function _changed($orig, $closing) {
|
|
903 |
$this->_deleted($orig);
|
|
904 |
echo "---\n";
|
|
905 |
$this->_added($closing);
|
|
906 |
}
|
|
907 |
}
|
|
908 |
|
|
909 |
|
|
910 |
/**
|
|
911 |
* Additions by Axel Boldt follow, partly taken from diff.php, phpwiki-1.3.3
|
|
912 |
*
|
|
913 |
*/
|
|
914 |
|
|
915 |
define('NBSP', ' '); // iso-8859-x non-breaking space.
|
|
916 |
|
|
917 |
/**
|
|
918 |
* @todo document
|
|
919 |
* @package Enano
|
|
920 |
* @subpackage DifferenceEngine
|
|
921 |
*/
|
|
922 |
class _HWLDF_WordAccumulator {
|
|
923 |
function _HWLDF_WordAccumulator () {
|
|
924 |
$this->_lines = array();
|
|
925 |
$this->_line = '';
|
|
926 |
$this->_group = '';
|
|
927 |
$this->_tag = '';
|
|
928 |
}
|
|
929 |
|
|
930 |
function _flushGroup ($new_tag) {
|
|
931 |
if ($this->_group !== '') {
|
|
932 |
if ($this->_tag == 'mark')
|
|
933 |
$this->_line .= '<span class="diffchange">' .
|
|
934 |
htmlspecialchars ( $this->_group ) . '</span>';
|
|
935 |
else
|
|
936 |
$this->_line .= htmlspecialchars ( $this->_group );
|
|
937 |
}
|
|
938 |
$this->_group = '';
|
|
939 |
$this->_tag = $new_tag;
|
|
940 |
}
|
|
941 |
|
|
942 |
function _flushLine ($new_tag) {
|
|
943 |
$this->_flushGroup($new_tag);
|
|
944 |
if ($this->_line != '')
|
|
945 |
array_push ( $this->_lines, $this->_line );
|
|
946 |
else
|
|
947 |
# make empty lines visible by inserting an NBSP
|
|
948 |
array_push ( $this->_lines, NBSP );
|
|
949 |
$this->_line = '';
|
|
950 |
}
|
|
951 |
|
|
952 |
function addWords ($words, $tag = '') {
|
|
953 |
if ($tag != $this->_tag)
|
|
954 |
$this->_flushGroup($tag);
|
|
955 |
|
|
956 |
foreach ($words as $word) {
|
|
957 |
// new-line should only come as first char of word.
|
|
958 |
if ($word == '')
|
|
959 |
continue;
|
|
960 |
if ($word[0] == "\n") {
|
|
961 |
$this->_flushLine($tag);
|
|
962 |
$word = substr($word, 1);
|
|
963 |
}
|
|
964 |
assert(!strstr($word, "\n"));
|
|
965 |
$this->_group .= $word;
|
|
966 |
}
|
|
967 |
}
|
|
968 |
|
|
969 |
function getLines() {
|
|
970 |
$this->_flushLine('~done');
|
|
971 |
return $this->_lines;
|
|
972 |
}
|
|
973 |
}
|
|
974 |
|
|
975 |
/**
|
|
976 |
* @todo document
|
|
977 |
* @package Enano
|
|
978 |
* @subpackage DifferenceEngine
|
|
979 |
*/
|
|
980 |
define('MAX_LINE_LENGTH', 10000);
|
|
981 |
class WordLevelDiff extends MappedDiff
|
|
982 |
{
|
|
983 |
function WordLevelDiff ($orig_lines, $closing_lines) {
|
|
984 |
$fname = 'WordLevelDiff::WordLevelDiff';
|
|
985 |
// wfProfileIn( $fname );
|
|
986 |
|
|
987 |
list ($orig_words, $orig_stripped) = $this->_split($orig_lines);
|
|
988 |
list ($closing_words, $closing_stripped) = $this->_split($closing_lines);
|
|
989 |
|
|
990 |
$this->MappedDiff($orig_words, $closing_words,
|
|
991 |
$orig_stripped, $closing_stripped);
|
|
992 |
// wfProfileOut( $fname );
|
|
993 |
}
|
|
994 |
|
|
995 |
function _split($lines) {
|
|
996 |
$fname = 'WordLevelDiff::_split';
|
|
997 |
// wfProfileIn( $fname );
|
|
998 |
|
|
999 |
$words = array();
|
|
1000 |
$stripped = array();
|
|
1001 |
$first = true;
|
|
1002 |
foreach ( $lines as $line ) {
|
|
1003 |
# If the line is too long, just pretend the entire line is one big word
|
|
1004 |
# This prevents resource exhaustion problems
|
|
1005 |
if ( $first ) {
|
|
1006 |
$first = false;
|
|
1007 |
} else {
|
|
1008 |
$words[] = "\n";
|
|
1009 |
$stripped[] = "\n";
|
|
1010 |
}
|
|
1011 |
if ( strlen( $line ) > MAX_LINE_LENGTH ) {
|
|
1012 |
$words[] = $line;
|
|
1013 |
$stripped[] = $line;
|
|
1014 |
} else {
|
|
1015 |
if (preg_match_all('/ ( [^\S\n]+ | [0-9_A-Za-z\x80-\xff]+ | . ) (?: (?!< \n) [^\S\n])? /xs',
|
|
1016 |
$line, $m))
|
|
1017 |
{
|
|
1018 |
$words = array_merge( $words, $m[0] );
|
|
1019 |
$stripped = array_merge( $stripped, $m[1] );
|
|
1020 |
}
|
|
1021 |
}
|
|
1022 |
}
|
|
1023 |
// wfProfileOut( $fname );
|
|
1024 |
return array($words, $stripped);
|
|
1025 |
}
|
|
1026 |
|
|
1027 |
function orig () {
|
|
1028 |
$fname = 'WordLevelDiff::orig';
|
|
1029 |
// wfProfileIn( $fname );
|
|
1030 |
$orig = new _HWLDF_WordAccumulator;
|
|
1031 |
|
|
1032 |
foreach ($this->edits as $edit) {
|
|
1033 |
if ($edit->type == 'copy')
|
|
1034 |
$orig->addWords($edit->orig);
|
|
1035 |
elseif ($edit->orig)
|
|
1036 |
$orig->addWords($edit->orig, 'mark');
|
|
1037 |
}
|
|
1038 |
$lines = $orig->getLines();
|
|
1039 |
// wfProfileOut( $fname );
|
|
1040 |
return $lines;
|
|
1041 |
}
|
|
1042 |
|
|
1043 |
function closing () {
|
|
1044 |
$fname = 'WordLevelDiff::closing';
|
|
1045 |
// wfProfileIn( $fname );
|
|
1046 |
$closing = new _HWLDF_WordAccumulator;
|
|
1047 |
|
|
1048 |
foreach ($this->edits as $edit) {
|
|
1049 |
if ($edit->type == 'copy')
|
|
1050 |
$closing->addWords($edit->closing);
|
|
1051 |
elseif ($edit->closing)
|
|
1052 |
$closing->addWords($edit->closing, 'mark');
|
|
1053 |
}
|
|
1054 |
$lines = $closing->getLines();
|
|
1055 |
// wfProfileOut( $fname );
|
|
1056 |
return $lines;
|
|
1057 |
}
|
|
1058 |
}
|
|
1059 |
|
|
1060 |
/**
|
|
1061 |
* Wikipedia Table style diff formatter.
|
|
1062 |
* @todo document
|
|
1063 |
* @package Enano
|
|
1064 |
* @subpackage DifferenceEngine
|
|
1065 |
*/
|
|
1066 |
class TableDiffFormatter extends DiffFormatter
|
|
1067 |
{
|
|
1068 |
function TableDiffFormatter() {
|
|
1069 |
$this->leading_context_lines = 2;
|
|
1070 |
$this->trailing_context_lines = 2;
|
|
1071 |
}
|
|
1072 |
|
|
1073 |
function _block_header( $xbeg, $xlen, $ybeg, $ylen ) {
|
|
1074 |
$r = '<tr><td colspan="2" align="left"><strong><!--LINE '.$xbeg."--></strong></td>\n" .
|
|
1075 |
'<td colspan="2" align="left"><strong><!--LINE '.$ybeg."--></strong></td></tr>\n";
|
|
1076 |
return $r;
|
|
1077 |
}
|
|
1078 |
|
|
1079 |
function _start_block( $header ) {
|
|
1080 |
echo $header;
|
|
1081 |
}
|
|
1082 |
|
|
1083 |
function _end_block() {
|
|
1084 |
}
|
|
1085 |
|
|
1086 |
function _lines( $lines, $prefix=' ', $color='white' ) {
|
|
1087 |
}
|
|
1088 |
|
|
1089 |
# HTML-escape parameter before calling this
|
|
1090 |
function addedLine( $line ) {
|
|
1091 |
return "<td>+</td><td class='diff-addedline'>{$line}</td>";
|
|
1092 |
}
|
|
1093 |
|
|
1094 |
# HTML-escape parameter before calling this
|
|
1095 |
function deletedLine( $line ) {
|
|
1096 |
return "<td>-</td><td class='diff-deletedline'>{$line}</td>";
|
|
1097 |
}
|
|
1098 |
|
|
1099 |
# HTML-escape parameter before calling this
|
|
1100 |
function contextLine( $line ) {
|
|
1101 |
return "<td> </td><td class='diff-context'>{$line}</td>";
|
|
1102 |
}
|
|
1103 |
|
|
1104 |
function emptyLine() {
|
|
1105 |
return '<td colspan="2"> </td>';
|
|
1106 |
}
|
|
1107 |
|
|
1108 |
function _added( $lines ) {
|
|
1109 |
foreach ($lines as $line) {
|
|
1110 |
echo '<tr>' . $this->emptyLine() .
|
|
1111 |
$this->addedLine( htmlspecialchars ( $line ) ) . "</tr>\n";
|
|
1112 |
}
|
|
1113 |
}
|
|
1114 |
|
|
1115 |
function _deleted($lines) {
|
|
1116 |
foreach ($lines as $line) {
|
|
1117 |
echo '<tr>' . $this->deletedLine( htmlspecialchars ( $line ) ) .
|
|
1118 |
$this->emptyLine() . "</tr>\n";
|
|
1119 |
}
|
|
1120 |
}
|
|
1121 |
|
|
1122 |
function _context( $lines ) {
|
|
1123 |
foreach ($lines as $line) {
|
|
1124 |
echo '<tr>' .
|
|
1125 |
$this->contextLine( htmlspecialchars ( $line ) ) .
|
|
1126 |
$this->contextLine( htmlspecialchars ( $line ) ) . "</tr>\n";
|
|
1127 |
}
|
|
1128 |
}
|
|
1129 |
|
|
1130 |
function _changed( $orig, $closing ) {
|
|
1131 |
$fname = 'TableDiffFormatter::_changed';
|
|
1132 |
// wfProfileIn( $fname );
|
|
1133 |
|
|
1134 |
$diff = new WordLevelDiff( $orig, $closing );
|
|
1135 |
$del = $diff->orig();
|
|
1136 |
$add = $diff->closing();
|
|
1137 |
|
|
1138 |
# Notice that WordLevelDiff returns HTML-escaped output.
|
|
1139 |
# Hence, we will be calling addedLine/deletedLine without HTML-escaping.
|
|
1140 |
|
|
1141 |
while ( $line = array_shift( $del ) ) {
|
|
1142 |
$aline = array_shift( $add );
|
|
1143 |
echo '<tr>' . $this->deletedLine( $line ) .
|
|
1144 |
$this->addedLine( $aline ) . "</tr>\n";
|
|
1145 |
}
|
|
1146 |
foreach ($add as $line) { # If any leftovers
|
|
1147 |
echo '<tr>' . $this->emptyLine() .
|
|
1148 |
$this->addedLine( $line ) . "</tr>\n";
|
|
1149 |
}
|
|
1150 |
// wfProfileOut( $fname );
|
|
1151 |
}
|
|
1152 |
}
|
|
1153 |
|