1 | <?php |
---|
2 | /** |
---|
3 | * GeSHi - Generic Syntax Highlighter |
---|
4 | * |
---|
5 | * The GeSHi class for Generic Syntax Highlighting. Please refer to the |
---|
6 | * documentation at http://qbnz.com/highlighter/documentation.php for more |
---|
7 | * information about how to use this class. |
---|
8 | * |
---|
9 | * For changes, release notes, TODOs etc, see the relevant files in the docs/ |
---|
10 | * directory. |
---|
11 | * |
---|
12 | * This file is part of GeSHi. |
---|
13 | * |
---|
14 | * GeSHi is free software; you can redistribute it and/or modify |
---|
15 | * it under the terms of the GNU General Public License as published by |
---|
16 | * the Free Software Foundation; either version 2 of the License, or |
---|
17 | * (at your option) any later version. |
---|
18 | * |
---|
19 | * GeSHi is distributed in the hope that it will be useful, |
---|
20 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
21 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
22 | * GNU General Public License for more details. |
---|
23 | * |
---|
24 | * You should have received a copy of the GNU General Public License |
---|
25 | * along with GeSHi; if not, write to the Free Software |
---|
26 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
---|
27 | * |
---|
28 | * @package geshi |
---|
29 | * @subpackage core |
---|
30 | * @author Nigel McNie <nigel@geshi.org>, Benny Baumann <BenBE@omorphia.de> |
---|
31 | * @copyright (C) 2004 - 2007 Nigel McNie, (C) 2007 - 2008 Benny Baumann |
---|
32 | * @license http://gnu.org/copyleft/gpl.html GNU GPL |
---|
33 | * |
---|
34 | */ |
---|
35 | |
---|
36 | // |
---|
37 | // GeSHi Constants |
---|
38 | // You should use these constant names in your programs instead of |
---|
39 | // their values - you never know when a value may change in a future |
---|
40 | // version |
---|
41 | // |
---|
42 | |
---|
43 | /** The version of this GeSHi file */ |
---|
44 | define('GESHI_VERSION', '1.0.8.2'); |
---|
45 | |
---|
46 | // Define the root directory for the GeSHi code tree |
---|
47 | if (!defined('GESHI_ROOT')) { |
---|
48 | /** The root directory for GeSHi */ |
---|
49 | define('GESHI_ROOT', dirname(__FILE__) . DIRECTORY_SEPARATOR); |
---|
50 | } |
---|
51 | /** The language file directory for GeSHi |
---|
52 | @access private */ |
---|
53 | define('GESHI_LANG_ROOT', GESHI_ROOT . 'geshi' . DIRECTORY_SEPARATOR); |
---|
54 | |
---|
55 | // Define if GeSHi should be paranoid about security |
---|
56 | if (!defined('GESHI_SECURITY_PARANOID')) { |
---|
57 | /** Tells GeSHi to be paranoid about security settings */ |
---|
58 | define('GESHI_SECURITY_PARANOID', false); |
---|
59 | } |
---|
60 | |
---|
61 | // Line numbers - use with enable_line_numbers() |
---|
62 | /** Use no line numbers when building the result */ |
---|
63 | define('GESHI_NO_LINE_NUMBERS', 0); |
---|
64 | /** Use normal line numbers when building the result */ |
---|
65 | define('GESHI_NORMAL_LINE_NUMBERS', 1); |
---|
66 | /** Use fancy line numbers when building the result */ |
---|
67 | define('GESHI_FANCY_LINE_NUMBERS', 2); |
---|
68 | |
---|
69 | // Container HTML type |
---|
70 | /** Use nothing to surround the source */ |
---|
71 | define('GESHI_HEADER_NONE', 0); |
---|
72 | /** Use a "div" to surround the source */ |
---|
73 | define('GESHI_HEADER_DIV', 1); |
---|
74 | /** Use a "pre" to surround the source */ |
---|
75 | define('GESHI_HEADER_PRE', 2); |
---|
76 | /** Use a pre to wrap lines when line numbers are enabled or to wrap the whole code. */ |
---|
77 | define('GESHI_HEADER_PRE_VALID', 3); |
---|
78 | /** |
---|
79 | * Use a "table" to surround the source: |
---|
80 | * |
---|
81 | * <table> |
---|
82 | * <thead><tr><td colspan="2">$header</td></tr></thead> |
---|
83 | * <tbody><tr><td><pre>$linenumbers</pre></td><td><pre>$code></pre></td></tr></tbody> |
---|
84 | * <tfooter><tr><td colspan="2">$footer</td></tr></tfoot> |
---|
85 | * </table> |
---|
86 | * |
---|
87 | * this is essentially only a workaround for Firefox, see sf#1651996 or take a look at |
---|
88 | * https://bugzilla.mozilla.org/show_bug.cgi?id=365805 |
---|
89 | * @note when linenumbers are disabled this is essentially the same as GESHI_HEADER_PRE |
---|
90 | */ |
---|
91 | define('GESHI_HEADER_PRE_TABLE', 4); |
---|
92 | |
---|
93 | // Capatalisation constants |
---|
94 | /** Lowercase keywords found */ |
---|
95 | define('GESHI_CAPS_NO_CHANGE', 0); |
---|
96 | /** Uppercase keywords found */ |
---|
97 | define('GESHI_CAPS_UPPER', 1); |
---|
98 | /** Leave keywords found as the case that they are */ |
---|
99 | define('GESHI_CAPS_LOWER', 2); |
---|
100 | |
---|
101 | // Link style constants |
---|
102 | /** Links in the source in the :link state */ |
---|
103 | define('GESHI_LINK', 0); |
---|
104 | /** Links in the source in the :hover state */ |
---|
105 | define('GESHI_HOVER', 1); |
---|
106 | /** Links in the source in the :active state */ |
---|
107 | define('GESHI_ACTIVE', 2); |
---|
108 | /** Links in the source in the :visited state */ |
---|
109 | define('GESHI_VISITED', 3); |
---|
110 | |
---|
111 | // Important string starter/finisher |
---|
112 | // Note that if you change these, they should be as-is: i.e., don't |
---|
113 | // write them as if they had been run through htmlentities() |
---|
114 | /** The starter for important parts of the source */ |
---|
115 | define('GESHI_START_IMPORTANT', '<BEGIN GeSHi>'); |
---|
116 | /** The ender for important parts of the source */ |
---|
117 | define('GESHI_END_IMPORTANT', '<END GeSHi>'); |
---|
118 | |
---|
119 | /**#@+ |
---|
120 | * @access private |
---|
121 | */ |
---|
122 | // When strict mode applies for a language |
---|
123 | /** Strict mode never applies (this is the most common) */ |
---|
124 | define('GESHI_NEVER', 0); |
---|
125 | /** Strict mode *might* apply, and can be enabled or |
---|
126 | disabled by {@link GeSHi->enable_strict_mode()} */ |
---|
127 | define('GESHI_MAYBE', 1); |
---|
128 | /** Strict mode always applies */ |
---|
129 | define('GESHI_ALWAYS', 2); |
---|
130 | |
---|
131 | // Advanced regexp handling constants, used in language files |
---|
132 | /** The key of the regex array defining what to search for */ |
---|
133 | define('GESHI_SEARCH', 0); |
---|
134 | /** The key of the regex array defining what bracket group in a |
---|
135 | matched search to use as a replacement */ |
---|
136 | define('GESHI_REPLACE', 1); |
---|
137 | /** The key of the regex array defining any modifiers to the regular expression */ |
---|
138 | define('GESHI_MODIFIERS', 2); |
---|
139 | /** The key of the regex array defining what bracket group in a |
---|
140 | matched search to put before the replacement */ |
---|
141 | define('GESHI_BEFORE', 3); |
---|
142 | /** The key of the regex array defining what bracket group in a |
---|
143 | matched search to put after the replacement */ |
---|
144 | define('GESHI_AFTER', 4); |
---|
145 | /** The key of the regex array defining a custom keyword to use |
---|
146 | for this regexp's html tag class */ |
---|
147 | define('GESHI_CLASS', 5); |
---|
148 | |
---|
149 | /** Used in language files to mark comments */ |
---|
150 | define('GESHI_COMMENTS', 0); |
---|
151 | |
---|
152 | /** Used to work around missing PHP features **/ |
---|
153 | define('GESHI_PHP_PRE_433', !(version_compare(PHP_VERSION, '4.3.3') === 1)); |
---|
154 | |
---|
155 | /** make sure we can call stripos **/ |
---|
156 | if (!function_exists('stripos')) { |
---|
157 | // the offset param of preg_match is not supported below PHP 4.3.3 |
---|
158 | if (GESHI_PHP_PRE_433) { |
---|
159 | /** |
---|
160 | * @ignore |
---|
161 | */ |
---|
162 | function stripos($haystack, $needle, $offset = null) { |
---|
163 | if (!is_null($offset)) { |
---|
164 | $haystack = substr($haystack, $offset); |
---|
165 | } |
---|
166 | if (preg_match('/'. preg_quote($needle, '/') . '/', $haystack, $match, PREG_OFFSET_CAPTURE)) { |
---|
167 | return $match[0][1]; |
---|
168 | } |
---|
169 | return false; |
---|
170 | } |
---|
171 | } |
---|
172 | else { |
---|
173 | /** |
---|
174 | * @ignore |
---|
175 | */ |
---|
176 | function stripos($haystack, $needle, $offset = null) { |
---|
177 | if (preg_match('/'. preg_quote($needle, '/') . '/', $haystack, $match, PREG_OFFSET_CAPTURE, $offset)) { |
---|
178 | return $match[0][1]; |
---|
179 | } |
---|
180 | return false; |
---|
181 | } |
---|
182 | } |
---|
183 | } |
---|
184 | |
---|
185 | /** some old PHP / PCRE subpatterns only support up to xxx subpatterns in |
---|
186 | regular expressions. Set this to false if your PCRE lib is up to date |
---|
187 | @see GeSHi->optimize_regexp_list() |
---|
188 | **/ |
---|
189 | define('GESHI_MAX_PCRE_SUBPATTERNS', 500); |
---|
190 | /** it's also important not to generate too long regular expressions |
---|
191 | be generous here... but keep in mind, that when reaching this limit we |
---|
192 | still have to close open patterns. 12k should do just fine on a 16k limit. |
---|
193 | @see GeSHi->optimize_regexp_list() |
---|
194 | **/ |
---|
195 | define('GESHI_MAX_PCRE_LENGTH', 12288); |
---|
196 | |
---|
197 | //Number format specification |
---|
198 | /** Basic number format for integers */ |
---|
199 | define('GESHI_NUMBER_INT_BASIC', 1); //Default integers \d+ |
---|
200 | /** Enhanced number format for integers like seen in C */ |
---|
201 | define('GESHI_NUMBER_INT_CSTYLE', 2); //Default C-Style \d+[lL]? |
---|
202 | /** Number format to highlight binary numbers with a suffix "b" */ |
---|
203 | define('GESHI_NUMBER_BIN_SUFFIX', 16); //[01]+[bB] |
---|
204 | /** Number format to highlight binary numbers with a prefix % */ |
---|
205 | define('GESHI_NUMBER_BIN_PREFIX_PERCENT', 32); //%[01]+ |
---|
206 | /** Number format to highlight binary numbers with a prefix 0b (C) */ |
---|
207 | define('GESHI_NUMBER_BIN_PREFIX_0B', 64); //0b[01]+ |
---|
208 | /** Number format to highlight octal numbers with a leading zero */ |
---|
209 | define('GESHI_NUMBER_OCT_PREFIX', 256); //0[0-7]+ |
---|
210 | /** Number format to highlight octal numbers with a suffix of o */ |
---|
211 | define('GESHI_NUMBER_OCT_SUFFIX', 512); //[0-7]+[oO] |
---|
212 | /** Number format to highlight hex numbers with a prefix 0x */ |
---|
213 | define('GESHI_NUMBER_HEX_PREFIX', 4096); //0x[0-9a-fA-F]+ |
---|
214 | /** Number format to highlight hex numbers with a suffix of h */ |
---|
215 | define('GESHI_NUMBER_HEX_SUFFIX', 8192); //[0-9][0-9a-fA-F]*h |
---|
216 | /** Number format to highlight floating-point numbers without support for scientific notation */ |
---|
217 | define('GESHI_NUMBER_FLT_NONSCI', 65536); //\d+\.\d+ |
---|
218 | /** Number format to highlight floating-point numbers without support for scientific notation */ |
---|
219 | define('GESHI_NUMBER_FLT_NONSCI_F', 131072); //\d+(\.\d+)?f |
---|
220 | /** Number format to highlight floating-point numbers with support for scientific notation (E) and optional leading zero */ |
---|
221 | define('GESHI_NUMBER_FLT_SCI_SHORT', 262144); //\.\d+e\d+ |
---|
222 | /** Number format to highlight floating-point numbers with support for scientific notation (E) and required leading digit */ |
---|
223 | define('GESHI_NUMBER_FLT_SCI_ZERO', 524288); //\d+(\.\d+)?e\d+ |
---|
224 | //Custom formats are passed by RX array |
---|
225 | |
---|
226 | // Error detection - use these to analyse faults |
---|
227 | /** No sourcecode to highlight was specified |
---|
228 | * @deprecated |
---|
229 | */ |
---|
230 | define('GESHI_ERROR_NO_INPUT', 1); |
---|
231 | /** The language specified does not exist */ |
---|
232 | define('GESHI_ERROR_NO_SUCH_LANG', 2); |
---|
233 | /** GeSHi could not open a file for reading (generally a language file) */ |
---|
234 | define('GESHI_ERROR_FILE_NOT_READABLE', 3); |
---|
235 | /** The header type passed to {@link GeSHi->set_header_type()} was invalid */ |
---|
236 | define('GESHI_ERROR_INVALID_HEADER_TYPE', 4); |
---|
237 | /** The line number type passed to {@link GeSHi->enable_line_numbers()} was invalid */ |
---|
238 | define('GESHI_ERROR_INVALID_LINE_NUMBER_TYPE', 5); |
---|
239 | /**#@-*/ |
---|
240 | |
---|
241 | |
---|
242 | /** |
---|
243 | * The GeSHi Class. |
---|
244 | * |
---|
245 | * Please refer to the documentation for GeSHi 1.0.X that is available |
---|
246 | * at http://qbnz.com/highlighter/documentation.php for more information |
---|
247 | * about how to use this class. |
---|
248 | * |
---|
249 | * @package geshi |
---|
250 | * @author Nigel McNie <nigel@geshi.org>, Benny Baumann <BenBE@omorphia.de> |
---|
251 | * @copyright (C) 2004 - 2007 Nigel McNie, (C) 2007 - 2008 Benny Baumann |
---|
252 | */ |
---|
253 | class GeSHi { |
---|
254 | /**#@+ |
---|
255 | * @access private |
---|
256 | */ |
---|
257 | /** |
---|
258 | * The source code to highlight |
---|
259 | * @var string |
---|
260 | */ |
---|
261 | var $source = ''; |
---|
262 | |
---|
263 | /** |
---|
264 | * The language to use when highlighting |
---|
265 | * @var string |
---|
266 | */ |
---|
267 | var $language = ''; |
---|
268 | |
---|
269 | /** |
---|
270 | * The data for the language used |
---|
271 | * @var array |
---|
272 | */ |
---|
273 | var $language_data = array(); |
---|
274 | |
---|
275 | /** |
---|
276 | * The path to the language files |
---|
277 | * @var string |
---|
278 | */ |
---|
279 | var $language_path = GESHI_LANG_ROOT; |
---|
280 | |
---|
281 | /** |
---|
282 | * The error message associated with an error |
---|
283 | * @var string |
---|
284 | * @todo check err reporting works |
---|
285 | */ |
---|
286 | var $error = false; |
---|
287 | |
---|
288 | /** |
---|
289 | * Possible error messages |
---|
290 | * @var array |
---|
291 | */ |
---|
292 | var $error_messages = array( |
---|
293 | GESHI_ERROR_NO_SUCH_LANG => 'GeSHi could not find the language {LANGUAGE} (using path {PATH})', |
---|
294 | GESHI_ERROR_FILE_NOT_READABLE => 'The file specified for load_from_file was not readable', |
---|
295 | GESHI_ERROR_INVALID_HEADER_TYPE => 'The header type specified is invalid', |
---|
296 | GESHI_ERROR_INVALID_LINE_NUMBER_TYPE => 'The line number type specified is invalid' |
---|
297 | ); |
---|
298 | |
---|
299 | /** |
---|
300 | * Whether highlighting is strict or not |
---|
301 | * @var boolean |
---|
302 | */ |
---|
303 | var $strict_mode = false; |
---|
304 | |
---|
305 | /** |
---|
306 | * Whether to use CSS classes in output |
---|
307 | * @var boolean |
---|
308 | */ |
---|
309 | var $use_classes = false; |
---|
310 | |
---|
311 | /** |
---|
312 | * The type of header to use. Can be one of the following |
---|
313 | * values: |
---|
314 | * |
---|
315 | * - GESHI_HEADER_PRE: Source is outputted in a "pre" HTML element. |
---|
316 | * - GESHI_HEADER_DIV: Source is outputted in a "div" HTML element. |
---|
317 | * - GESHI_HEADER_NONE: No header is outputted. |
---|
318 | * |
---|
319 | * @var int |
---|
320 | */ |
---|
321 | var $header_type = GESHI_HEADER_PRE; |
---|
322 | |
---|
323 | /** |
---|
324 | * Array of permissions for which lexics should be highlighted |
---|
325 | * @var array |
---|
326 | */ |
---|
327 | var $lexic_permissions = array( |
---|
328 | 'KEYWORDS' => array(), |
---|
329 | 'COMMENTS' => array('MULTI' => true), |
---|
330 | 'REGEXPS' => array(), |
---|
331 | 'ESCAPE_CHAR' => true, |
---|
332 | 'BRACKETS' => true, |
---|
333 | 'SYMBOLS' => false, |
---|
334 | 'STRINGS' => true, |
---|
335 | 'NUMBERS' => true, |
---|
336 | 'METHODS' => true, |
---|
337 | 'SCRIPT' => true |
---|
338 | ); |
---|
339 | |
---|
340 | /** |
---|
341 | * The time it took to parse the code |
---|
342 | * @var double |
---|
343 | */ |
---|
344 | var $time = 0; |
---|
345 | |
---|
346 | /** |
---|
347 | * The content of the header block |
---|
348 | * @var string |
---|
349 | */ |
---|
350 | var $header_content = ''; |
---|
351 | |
---|
352 | /** |
---|
353 | * The content of the footer block |
---|
354 | * @var string |
---|
355 | */ |
---|
356 | var $footer_content = ''; |
---|
357 | |
---|
358 | /** |
---|
359 | * The style of the header block |
---|
360 | * @var string |
---|
361 | */ |
---|
362 | var $header_content_style = ''; |
---|
363 | |
---|
364 | /** |
---|
365 | * The style of the footer block |
---|
366 | * @var string |
---|
367 | */ |
---|
368 | var $footer_content_style = ''; |
---|
369 | |
---|
370 | /** |
---|
371 | * Tells if a block around the highlighted source should be forced |
---|
372 | * if not using line numbering |
---|
373 | * @var boolean |
---|
374 | */ |
---|
375 | var $force_code_block = false; |
---|
376 | |
---|
377 | /** |
---|
378 | * The styles for hyperlinks in the code |
---|
379 | * @var array |
---|
380 | */ |
---|
381 | var $link_styles = array(); |
---|
382 | |
---|
383 | /** |
---|
384 | * Whether important blocks should be recognised or not |
---|
385 | * @var boolean |
---|
386 | * @deprecated |
---|
387 | * @todo REMOVE THIS FUNCTIONALITY! |
---|
388 | */ |
---|
389 | var $enable_important_blocks = false; |
---|
390 | |
---|
391 | /** |
---|
392 | * Styles for important parts of the code |
---|
393 | * @var string |
---|
394 | * @deprecated |
---|
395 | * @todo As above - rethink the whole idea of important blocks as it is buggy and |
---|
396 | * will be hard to implement in 1.2 |
---|
397 | */ |
---|
398 | var $important_styles = 'font-weight: bold; color: red;'; // Styles for important parts of the code |
---|
399 | |
---|
400 | /** |
---|
401 | * Whether CSS IDs should be added to the code |
---|
402 | * @var boolean |
---|
403 | */ |
---|
404 | var $add_ids = false; |
---|
405 | |
---|
406 | /** |
---|
407 | * Lines that should be highlighted extra |
---|
408 | * @var array |
---|
409 | */ |
---|
410 | var $highlight_extra_lines = array(); |
---|
411 | |
---|
412 | /** |
---|
413 | * Styles of lines that should be highlighted extra |
---|
414 | * @var array |
---|
415 | */ |
---|
416 | var $highlight_extra_lines_styles = array(); |
---|
417 | |
---|
418 | /** |
---|
419 | * Styles of extra-highlighted lines |
---|
420 | * @var string |
---|
421 | */ |
---|
422 | var $highlight_extra_lines_style = 'background-color: #ffc;'; |
---|
423 | |
---|
424 | /** |
---|
425 | * The line ending |
---|
426 | * If null, nl2br() will be used on the result string. |
---|
427 | * Otherwise, all instances of \n will be replaced with $line_ending |
---|
428 | * @var string |
---|
429 | */ |
---|
430 | var $line_ending = null; |
---|
431 | |
---|
432 | /** |
---|
433 | * Number at which line numbers should start at |
---|
434 | * @var int |
---|
435 | */ |
---|
436 | var $line_numbers_start = 1; |
---|
437 | |
---|
438 | /** |
---|
439 | * The overall style for this code block |
---|
440 | * @var string |
---|
441 | */ |
---|
442 | var $overall_style = 'font-family:monospace;'; |
---|
443 | |
---|
444 | /** |
---|
445 | * The style for the actual code |
---|
446 | * @var string |
---|
447 | */ |
---|
448 | var $code_style = 'font: normal normal 1em/1.2em monospace; margin:0; padding:0; background:none; vertical-align:top;'; |
---|
449 | |
---|
450 | /** |
---|
451 | * The overall class for this code block |
---|
452 | * @var string |
---|
453 | */ |
---|
454 | var $overall_class = ''; |
---|
455 | |
---|
456 | /** |
---|
457 | * The overall ID for this code block |
---|
458 | * @var string |
---|
459 | */ |
---|
460 | var $overall_id = ''; |
---|
461 | |
---|
462 | /** |
---|
463 | * Line number styles |
---|
464 | * @var string |
---|
465 | */ |
---|
466 | var $line_style1 = 'font-weight: normal; vertical-align:top;'; |
---|
467 | |
---|
468 | /** |
---|
469 | * Line number styles for fancy lines |
---|
470 | * @var string |
---|
471 | */ |
---|
472 | var $line_style2 = 'font-weight: bold; vertical-align:top;'; |
---|
473 | |
---|
474 | /** |
---|
475 | * Style for line numbers when GESHI_HEADER_PRE_TABLE is chosen |
---|
476 | * @var string |
---|
477 | */ |
---|
478 | var $table_linenumber_style = 'width:1px;text-align:right;margin:0;padding:0 2px;vertical-align:top;'; |
---|
479 | |
---|
480 | /** |
---|
481 | * Flag for how line numbers are displayed |
---|
482 | * @var boolean |
---|
483 | */ |
---|
484 | var $line_numbers = GESHI_NO_LINE_NUMBERS; |
---|
485 | |
---|
486 | /** |
---|
487 | * Flag to decide if multi line spans are allowed. Set it to false to make sure |
---|
488 | * each tag is closed before and reopened after each linefeed. |
---|
489 | * @var boolean |
---|
490 | */ |
---|
491 | var $allow_multiline_span = true; |
---|
492 | |
---|
493 | /** |
---|
494 | * The "nth" value for fancy line highlighting |
---|
495 | * @var int |
---|
496 | */ |
---|
497 | var $line_nth_row = 0; |
---|
498 | |
---|
499 | /** |
---|
500 | * The size of tab stops |
---|
501 | * @var int |
---|
502 | */ |
---|
503 | var $tab_width = 8; |
---|
504 | |
---|
505 | /** |
---|
506 | * Should we use language-defined tab stop widths? |
---|
507 | * @var int |
---|
508 | */ |
---|
509 | var $use_language_tab_width = false; |
---|
510 | |
---|
511 | /** |
---|
512 | * Default target for keyword links |
---|
513 | * @var string |
---|
514 | */ |
---|
515 | var $link_target = ''; |
---|
516 | |
---|
517 | /** |
---|
518 | * The encoding to use for entity encoding |
---|
519 | * NOTE: Used with Escape Char Sequences to fix UTF-8 handling (cf. SF#2037598) |
---|
520 | * @var string |
---|
521 | */ |
---|
522 | var $encoding = 'utf-8'; |
---|
523 | |
---|
524 | /** |
---|
525 | * Should keywords be linked? |
---|
526 | * @var boolean |
---|
527 | */ |
---|
528 | var $keyword_links = true; |
---|
529 | |
---|
530 | /** |
---|
531 | * Currently loaded language file |
---|
532 | * @var string |
---|
533 | * @since 1.0.7.22 |
---|
534 | */ |
---|
535 | var $loaded_language = ''; |
---|
536 | |
---|
537 | /** |
---|
538 | * Wether the caches needed for parsing are built or not |
---|
539 | * |
---|
540 | * @var bool |
---|
541 | * @since 1.0.8 |
---|
542 | */ |
---|
543 | var $parse_cache_built = false; |
---|
544 | |
---|
545 | /** |
---|
546 | * Work around for Suhosin Patch with disabled /e modifier |
---|
547 | * |
---|
548 | * Note from suhosins author in config file: |
---|
549 | * <blockquote> |
---|
550 | * The /e modifier inside <code>preg_replace()</code> allows code execution. |
---|
551 | * Often it is the cause for remote code execution exploits. It is wise to |
---|
552 | * deactivate this feature and test where in the application it is used. |
---|
553 | * The developer using the /e modifier should be made aware that he should |
---|
554 | * use <code>preg_replace_callback()</code> instead |
---|
555 | * </blockquote> |
---|
556 | * |
---|
557 | * @var array |
---|
558 | * @since 1.0.8 |
---|
559 | */ |
---|
560 | var $_kw_replace_group = 0; |
---|
561 | var $_rx_key = 0; |
---|
562 | |
---|
563 | /** |
---|
564 | * some "callback parameters" for handle_multiline_regexps |
---|
565 | * |
---|
566 | * @since 1.0.8 |
---|
567 | * @access private |
---|
568 | * @var string |
---|
569 | */ |
---|
570 | var $_hmr_before = ''; |
---|
571 | var $_hmr_replace = ''; |
---|
572 | var $_hmr_after = ''; |
---|
573 | var $_hmr_key = 0; |
---|
574 | |
---|
575 | /**#@-*/ |
---|
576 | |
---|
577 | /** |
---|
578 | * Creates a new GeSHi object, with source and language |
---|
579 | * |
---|
580 | * @param string The source code to highlight |
---|
581 | * @param string The language to highlight the source with |
---|
582 | * @param string The path to the language file directory. <b>This |
---|
583 | * is deprecated!</b> I've backported the auto path |
---|
584 | * detection from the 1.1.X dev branch, so now it |
---|
585 | * should be automatically set correctly. If you have |
---|
586 | * renamed the language directory however, you will |
---|
587 | * still need to set the path using this parameter or |
---|
588 | * {@link GeSHi->set_language_path()} |
---|
589 | * @since 1.0.0 |
---|
590 | */ |
---|
591 | function GeSHi($source = '', $language = '', $path = '') { |
---|
592 | if (!empty($source)) { |
---|
593 | $this->set_source($source); |
---|
594 | } |
---|
595 | if (!empty($language)) { |
---|
596 | $this->set_language($language); |
---|
597 | } |
---|
598 | $this->set_language_path($path); |
---|
599 | } |
---|
600 | |
---|
601 | /** |
---|
602 | * Returns an error message associated with the last GeSHi operation, |
---|
603 | * or false if no error has occured |
---|
604 | * |
---|
605 | * @return string|false An error message if there has been an error, else false |
---|
606 | * @since 1.0.0 |
---|
607 | */ |
---|
608 | function error() { |
---|
609 | if ($this->error) { |
---|
610 | //Put some template variables for debugging here ... |
---|
611 | $debug_tpl_vars = array( |
---|
612 | '{LANGUAGE}' => $this->language, |
---|
613 | '{PATH}' => $this->language_path |
---|
614 | ); |
---|
615 | $msg = str_replace( |
---|
616 | array_keys($debug_tpl_vars), |
---|
617 | array_values($debug_tpl_vars), |
---|
618 | $this->error_messages[$this->error]); |
---|
619 | |
---|
620 | return "<br /><strong>GeSHi Error:</strong> $msg (code {$this->error})<br />"; |
---|
621 | } |
---|
622 | return false; |
---|
623 | } |
---|
624 | |
---|
625 | /** |
---|
626 | * Gets a human-readable language name (thanks to Simon Patterson |
---|
627 | * for the idea :)) |
---|
628 | * |
---|
629 | * @return string The name for the current language |
---|
630 | * @since 1.0.2 |
---|
631 | */ |
---|
632 | function get_language_name() { |
---|
633 | if (GESHI_ERROR_NO_SUCH_LANG == $this->error) { |
---|
634 | return $this->language_data['LANG_NAME'] . ' (Unknown Language)'; |
---|
635 | } |
---|
636 | return $this->language_data['LANG_NAME']; |
---|
637 | } |
---|
638 | |
---|
639 | /** |
---|
640 | * Sets the source code for this object |
---|
641 | * |
---|
642 | * @param string The source code to highlight |
---|
643 | * @since 1.0.0 |
---|
644 | */ |
---|
645 | function set_source($source) { |
---|
646 | $this->source = $source; |
---|
647 | $this->highlight_extra_lines = array(); |
---|
648 | } |
---|
649 | |
---|
650 | /** |
---|
651 | * Sets the language for this object |
---|
652 | * |
---|
653 | * @note since 1.0.8 this function won't reset language-settings by default anymore! |
---|
654 | * if you need this set $force_reset = true |
---|
655 | * |
---|
656 | * @param string The name of the language to use |
---|
657 | * @since 1.0.0 |
---|
658 | */ |
---|
659 | function set_language($language, $force_reset = false) { |
---|
660 | if ($force_reset) { |
---|
661 | $this->loaded_language = false; |
---|
662 | } |
---|
663 | |
---|
664 | //Clean up the language name to prevent malicious code injection |
---|
665 | $language = preg_replace('#[^a-zA-Z0-9\-_]#', '', $language); |
---|
666 | |
---|
667 | $language = strtolower($language); |
---|
668 | |
---|
669 | //Retreive the full filename |
---|
670 | $file_name = $this->language_path . $language . '.php'; |
---|
671 | if ($file_name == $this->loaded_language) { |
---|
672 | // this language is already loaded! |
---|
673 | return; |
---|
674 | } |
---|
675 | |
---|
676 | $this->language = $language; |
---|
677 | |
---|
678 | $this->error = false; |
---|
679 | $this->strict_mode = GESHI_NEVER; |
---|
680 | |
---|
681 | //Check if we can read the desired file |
---|
682 | if (!is_readable($file_name)) { |
---|
683 | $this->error = GESHI_ERROR_NO_SUCH_LANG; |
---|
684 | return; |
---|
685 | } |
---|
686 | |
---|
687 | // Load the language for parsing |
---|
688 | $this->load_language($file_name); |
---|
689 | } |
---|
690 | |
---|
691 | /** |
---|
692 | * Sets the path to the directory containing the language files. Note |
---|
693 | * that this path is relative to the directory of the script that included |
---|
694 | * geshi.php, NOT geshi.php itself. |
---|
695 | * |
---|
696 | * @param string The path to the language directory |
---|
697 | * @since 1.0.0 |
---|
698 | * @deprecated The path to the language files should now be automatically |
---|
699 | * detected, so this method should no longer be needed. The |
---|
700 | * 1.1.X branch handles manual setting of the path differently |
---|
701 | * so this method will disappear in 1.2.0. |
---|
702 | */ |
---|
703 | function set_language_path($path) { |
---|
704 | if(strpos($path,':')) { |
---|
705 | //Security Fix to prevent external directories using fopen wrappers. |
---|
706 | if(DIRECTORY_SEPARATOR == "\\") { |
---|
707 | if(!preg_match('#^[a-zA-Z]:#', $path) || false !== strpos($path, ':', 2)) { |
---|
708 | return; |
---|
709 | } |
---|
710 | } else { |
---|
711 | return; |
---|
712 | } |
---|
713 | } |
---|
714 | if(preg_match('#[^/a-zA-Z0-9_\.\-\\\s:]#', $path)) { |
---|
715 | //Security Fix to prevent external directories using fopen wrappers. |
---|
716 | return; |
---|
717 | } |
---|
718 | if(GESHI_SECURITY_PARANOID && false !== strpos($path, '/.')) { |
---|
719 | //Security Fix to prevent external directories using fopen wrappers. |
---|
720 | return; |
---|
721 | } |
---|
722 | if(GESHI_SECURITY_PARANOID && false !== strpos($path, '..')) { |
---|
723 | //Security Fix to prevent external directories using fopen wrappers. |
---|
724 | return; |
---|
725 | } |
---|
726 | if ($path) { |
---|
727 | $this->language_path = ('/' == $path[strlen($path) - 1]) ? $path : $path . '/'; |
---|
728 | $this->set_language($this->language); // otherwise set_language_path has no effect |
---|
729 | } |
---|
730 | } |
---|
731 | |
---|
732 | /** |
---|
733 | * Sets the type of header to be used. |
---|
734 | * |
---|
735 | * If GESHI_HEADER_DIV is used, the code is surrounded in a "div".This |
---|
736 | * means more source code but more control over tab width and line-wrapping. |
---|
737 | * GESHI_HEADER_PRE means that a "pre" is used - less source, but less |
---|
738 | * control. Default is GESHI_HEADER_PRE. |
---|
739 | * |
---|
740 | * From 1.0.7.2, you can use GESHI_HEADER_NONE to specify that no header code |
---|
741 | * should be outputted. |
---|
742 | * |
---|
743 | * @param int The type of header to be used |
---|
744 | * @since 1.0.0 |
---|
745 | */ |
---|
746 | function set_header_type($type) { |
---|
747 | //Check if we got a valid header type |
---|
748 | if (!in_array($type, array(GESHI_HEADER_NONE, GESHI_HEADER_DIV, |
---|
749 | GESHI_HEADER_PRE, GESHI_HEADER_PRE_VALID, GESHI_HEADER_PRE_TABLE))) { |
---|
750 | $this->error = GESHI_ERROR_INVALID_HEADER_TYPE; |
---|
751 | return; |
---|
752 | } |
---|
753 | |
---|
754 | //Set that new header type |
---|
755 | $this->header_type = $type; |
---|
756 | } |
---|
757 | |
---|
758 | /** |
---|
759 | * Sets the styles for the code that will be outputted |
---|
760 | * when this object is parsed. The style should be a |
---|
761 | * string of valid stylesheet declarations |
---|
762 | * |
---|
763 | * @param string The overall style for the outputted code block |
---|
764 | * @param boolean Whether to merge the styles with the current styles or not |
---|
765 | * @since 1.0.0 |
---|
766 | */ |
---|
767 | function set_overall_style($style, $preserve_defaults = false) { |
---|
768 | if (!$preserve_defaults) { |
---|
769 | $this->overall_style = $style; |
---|
770 | } else { |
---|
771 | $this->overall_style .= $style; |
---|
772 | } |
---|
773 | } |
---|
774 | |
---|
775 | /** |
---|
776 | * Sets the overall classname for this block of code. This |
---|
777 | * class can then be used in a stylesheet to style this object's |
---|
778 | * output |
---|
779 | * |
---|
780 | * @param string The class name to use for this block of code |
---|
781 | * @since 1.0.0 |
---|
782 | */ |
---|
783 | function set_overall_class($class) { |
---|
784 | $this->overall_class = $class; |
---|
785 | } |
---|
786 | |
---|
787 | /** |
---|
788 | * Sets the overall id for this block of code. This id can then |
---|
789 | * be used in a stylesheet to style this object's output |
---|
790 | * |
---|
791 | * @param string The ID to use for this block of code |
---|
792 | * @since 1.0.0 |
---|
793 | */ |
---|
794 | function set_overall_id($id) { |
---|
795 | $this->overall_id = $id; |
---|
796 | } |
---|
797 | |
---|
798 | /** |
---|
799 | * Sets whether CSS classes should be used to highlight the source. Default |
---|
800 | * is off, calling this method with no arguments will turn it on |
---|
801 | * |
---|
802 | * @param boolean Whether to turn classes on or not |
---|
803 | * @since 1.0.0 |
---|
804 | */ |
---|
805 | function enable_classes($flag = true) { |
---|
806 | $this->use_classes = ($flag) ? true : false; |
---|
807 | } |
---|
808 | |
---|
809 | /** |
---|
810 | * Sets the style for the actual code. This should be a string |
---|
811 | * containing valid stylesheet declarations. If $preserve_defaults is |
---|
812 | * true, then styles are merged with the default styles, with the |
---|
813 | * user defined styles having priority |
---|
814 | * |
---|
815 | * Note: Use this method to override any style changes you made to |
---|
816 | * the line numbers if you are using line numbers, else the line of |
---|
817 | * code will have the same style as the line number! Consult the |
---|
818 | * GeSHi documentation for more information about this. |
---|
819 | * |
---|
820 | * @param string The style to use for actual code |
---|
821 | * @param boolean Whether to merge the current styles with the new styles |
---|
822 | * @since 1.0.2 |
---|
823 | */ |
---|
824 | function set_code_style($style, $preserve_defaults = false) { |
---|
825 | if (!$preserve_defaults) { |
---|
826 | $this->code_style = $style; |
---|
827 | } else { |
---|
828 | $this->code_style .= $style; |
---|
829 | } |
---|
830 | } |
---|
831 | |
---|
832 | /** |
---|
833 | * Sets the styles for the line numbers. |
---|
834 | * |
---|
835 | * @param string The style for the line numbers that are "normal" |
---|
836 | * @param string|boolean If a string, this is the style of the line |
---|
837 | * numbers that are "fancy", otherwise if boolean then this |
---|
838 | * defines whether the normal styles should be merged with the |
---|
839 | * new normal styles or not |
---|
840 | * @param boolean If set, is the flag for whether to merge the "fancy" |
---|
841 | * styles with the current styles or not |
---|
842 | * @since 1.0.2 |
---|
843 | */ |
---|
844 | function set_line_style($style1, $style2 = '', $preserve_defaults = false) { |
---|
845 | //Check if we got 2 or three parameters |
---|
846 | if (is_bool($style2)) { |
---|
847 | $preserve_defaults = $style2; |
---|
848 | $style2 = ''; |
---|
849 | } |
---|
850 | |
---|
851 | //Actually set the new styles |
---|
852 | if (!$preserve_defaults) { |
---|
853 | $this->line_style1 = $style1; |
---|
854 | $this->line_style2 = $style2; |
---|
855 | } else { |
---|
856 | $this->line_style1 .= $style1; |
---|
857 | $this->line_style2 .= $style2; |
---|
858 | } |
---|
859 | } |
---|
860 | |
---|
861 | /** |
---|
862 | * Sets whether line numbers should be displayed. |
---|
863 | * |
---|
864 | * Valid values for the first parameter are: |
---|
865 | * |
---|
866 | * - GESHI_NO_LINE_NUMBERS: Line numbers will not be displayed |
---|
867 | * - GESHI_NORMAL_LINE_NUMBERS: Line numbers will be displayed |
---|
868 | * - GESHI_FANCY_LINE_NUMBERS: Fancy line numbers will be displayed |
---|
869 | * |
---|
870 | * For fancy line numbers, the second parameter is used to signal which lines |
---|
871 | * are to be fancy. For example, if the value of this parameter is 5 then every |
---|
872 | * 5th line will be fancy. |
---|
873 | * |
---|
874 | * @param int How line numbers should be displayed |
---|
875 | * @param int Defines which lines are fancy |
---|
876 | * @since 1.0.0 |
---|
877 | */ |
---|
878 | function enable_line_numbers($flag, $nth_row = 5) { |
---|
879 | if (GESHI_NO_LINE_NUMBERS != $flag && GESHI_NORMAL_LINE_NUMBERS != $flag |
---|
880 | && GESHI_FANCY_LINE_NUMBERS != $flag) { |
---|
881 | $this->error = GESHI_ERROR_INVALID_LINE_NUMBER_TYPE; |
---|
882 | } |
---|
883 | $this->line_numbers = $flag; |
---|
884 | $this->line_nth_row = $nth_row; |
---|
885 | } |
---|
886 | |
---|
887 | /** |
---|
888 | * Sets wether spans and other HTML markup generated by GeSHi can |
---|
889 | * span over multiple lines or not. Defaults to true to reduce overhead. |
---|
890 | * Set it to false if you want to manipulate the output or manually display |
---|
891 | * the code in an ordered list. |
---|
892 | * |
---|
893 | * @param boolean Wether multiline spans are allowed or not |
---|
894 | * @since 1.0.7.22 |
---|
895 | */ |
---|
896 | function enable_multiline_span($flag) { |
---|
897 | $this->allow_multiline_span = (bool) $flag; |
---|
898 | } |
---|
899 | |
---|
900 | /** |
---|
901 | * Get current setting for multiline spans, see GeSHi->enable_multiline_span(). |
---|
902 | * |
---|
903 | * @see enable_multiline_span |
---|
904 | * @return bool |
---|
905 | */ |
---|
906 | function get_multiline_span() { |
---|
907 | return $this->allow_multiline_span; |
---|
908 | } |
---|
909 | |
---|
910 | /** |
---|
911 | * Sets the style for a keyword group. If $preserve_defaults is |
---|
912 | * true, then styles are merged with the default styles, with the |
---|
913 | * user defined styles having priority |
---|
914 | * |
---|
915 | * @param int The key of the keyword group to change the styles of |
---|
916 | * @param string The style to make the keywords |
---|
917 | * @param boolean Whether to merge the new styles with the old or just |
---|
918 | * to overwrite them |
---|
919 | * @since 1.0.0 |
---|
920 | */ |
---|
921 | function set_keyword_group_style($key, $style, $preserve_defaults = false) { |
---|
922 | //Set the style for this keyword group |
---|
923 | if (!$preserve_defaults) { |
---|
924 | $this->language_data['STYLES']['KEYWORDS'][$key] = $style; |
---|
925 | } else { |
---|
926 | $this->language_data['STYLES']['KEYWORDS'][$key] .= $style; |
---|
927 | } |
---|
928 | |
---|
929 | //Update the lexic permissions |
---|
930 | if (!isset($this->lexic_permissions['KEYWORDS'][$key])) { |
---|
931 | $this->lexic_permissions['KEYWORDS'][$key] = true; |
---|
932 | } |
---|
933 | } |
---|
934 | |
---|
935 | /** |
---|
936 | * Turns highlighting on/off for a keyword group |
---|
937 | * |
---|
938 | * @param int The key of the keyword group to turn on or off |
---|
939 | * @param boolean Whether to turn highlighting for that group on or off |
---|
940 | * @since 1.0.0 |
---|
941 | */ |
---|
942 | function set_keyword_group_highlighting($key, $flag = true) { |
---|
943 | $this->lexic_permissions['KEYWORDS'][$key] = ($flag) ? true : false; |
---|
944 | } |
---|
945 | |
---|
946 | /** |
---|
947 | * Sets the styles for comment groups. If $preserve_defaults is |
---|
948 | * true, then styles are merged with the default styles, with the |
---|
949 | * user defined styles having priority |
---|
950 | * |
---|
951 | * @param int The key of the comment group to change the styles of |
---|
952 | * @param string The style to make the comments |
---|
953 | * @param boolean Whether to merge the new styles with the old or just |
---|
954 | * to overwrite them |
---|
955 | * @since 1.0.0 |
---|
956 | */ |
---|
957 | function set_comments_style($key, $style, $preserve_defaults = false) { |
---|
958 | if (!$preserve_defaults) { |
---|
959 | $this->language_data['STYLES']['COMMENTS'][$key] = $style; |
---|
960 | } else { |
---|
961 | $this->language_data['STYLES']['COMMENTS'][$key] .= $style; |
---|
962 | } |
---|
963 | } |
---|
964 | |
---|
965 | /** |
---|
966 | * Turns highlighting on/off for comment groups |
---|
967 | * |
---|
968 | * @param int The key of the comment group to turn on or off |
---|
969 | * @param boolean Whether to turn highlighting for that group on or off |
---|
970 | * @since 1.0.0 |
---|
971 | */ |
---|
972 | function set_comments_highlighting($key, $flag = true) { |
---|
973 | $this->lexic_permissions['COMMENTS'][$key] = ($flag) ? true : false; |
---|
974 | } |
---|
975 | |
---|
976 | /** |
---|
977 | * Sets the styles for escaped characters. If $preserve_defaults is |
---|
978 | * true, then styles are merged with the default styles, with the |
---|
979 | * user defined styles having priority |
---|
980 | * |
---|
981 | * @param string The style to make the escape characters |
---|
982 | * @param boolean Whether to merge the new styles with the old or just |
---|
983 | * to overwrite them |
---|
984 | * @since 1.0.0 |
---|
985 | */ |
---|
986 | function set_escape_characters_style($style, $preserve_defaults = false) { |
---|
987 | if (!$preserve_defaults) { |
---|
988 | $this->language_data['STYLES']['ESCAPE_CHAR'][0] = $style; |
---|
989 | } else { |
---|
990 | $this->language_data['STYLES']['ESCAPE_CHAR'][0] .= $style; |
---|
991 | } |
---|
992 | } |
---|
993 | |
---|
994 | /** |
---|
995 | * Turns highlighting on/off for escaped characters |
---|
996 | * |
---|
997 | * @param boolean Whether to turn highlighting for escape characters on or off |
---|
998 | * @since 1.0.0 |
---|
999 | */ |
---|
1000 | function set_escape_characters_highlighting($flag = true) { |
---|
1001 | $this->lexic_permissions['ESCAPE_CHAR'] = ($flag) ? true : false; |
---|
1002 | } |
---|
1003 | |
---|
1004 | /** |
---|
1005 | * Sets the styles for brackets. If $preserve_defaults is |
---|
1006 | * true, then styles are merged with the default styles, with the |
---|
1007 | * user defined styles having priority |
---|
1008 | * |
---|
1009 | * This method is DEPRECATED: use set_symbols_style instead. |
---|
1010 | * This method will be removed in 1.2.X |
---|
1011 | * |
---|
1012 | * @param string The style to make the brackets |
---|
1013 | * @param boolean Whether to merge the new styles with the old or just |
---|
1014 | * to overwrite them |
---|
1015 | * @since 1.0.0 |
---|
1016 | * @deprecated In favour of set_symbols_style |
---|
1017 | */ |
---|
1018 | function set_brackets_style($style, $preserve_defaults = false) { |
---|
1019 | if (!$preserve_defaults) { |
---|
1020 | $this->language_data['STYLES']['BRACKETS'][0] = $style; |
---|
1021 | } else { |
---|
1022 | $this->language_data['STYLES']['BRACKETS'][0] .= $style; |
---|
1023 | } |
---|
1024 | } |
---|
1025 | |
---|
1026 | /** |
---|
1027 | * Turns highlighting on/off for brackets |
---|
1028 | * |
---|
1029 | * This method is DEPRECATED: use set_symbols_highlighting instead. |
---|
1030 | * This method will be remove in 1.2.X |
---|
1031 | * |
---|
1032 | * @param boolean Whether to turn highlighting for brackets on or off |
---|
1033 | * @since 1.0.0 |
---|
1034 | * @deprecated In favour of set_symbols_highlighting |
---|
1035 | */ |
---|
1036 | function set_brackets_highlighting($flag) { |
---|
1037 | $this->lexic_permissions['BRACKETS'] = ($flag) ? true : false; |
---|
1038 | } |
---|
1039 | |
---|
1040 | /** |
---|
1041 | * Sets the styles for symbols. If $preserve_defaults is |
---|
1042 | * true, then styles are merged with the default styles, with the |
---|
1043 | * user defined styles having priority |
---|
1044 | * |
---|
1045 | * @param string The style to make the symbols |
---|
1046 | * @param boolean Whether to merge the new styles with the old or just |
---|
1047 | * to overwrite them |
---|
1048 | * @param int Tells the group of symbols for which style should be set. |
---|
1049 | * @since 1.0.1 |
---|
1050 | */ |
---|
1051 | function set_symbols_style($style, $preserve_defaults = false, $group = 0) { |
---|
1052 | // Update the style of symbols |
---|
1053 | if (!$preserve_defaults) { |
---|
1054 | $this->language_data['STYLES']['SYMBOLS'][$group] = $style; |
---|
1055 | } else { |
---|
1056 | $this->language_data['STYLES']['SYMBOLS'][$group] .= $style; |
---|
1057 | } |
---|
1058 | |
---|
1059 | // For backward compatibility |
---|
1060 | if (0 == $group) { |
---|
1061 | $this->set_brackets_style ($style, $preserve_defaults); |
---|
1062 | } |
---|
1063 | } |
---|
1064 | |
---|
1065 | /** |
---|
1066 | * Turns highlighting on/off for symbols |
---|
1067 | * |
---|
1068 | * @param boolean Whether to turn highlighting for symbols on or off |
---|
1069 | * @since 1.0.0 |
---|
1070 | */ |
---|
1071 | function set_symbols_highlighting($flag) { |
---|
1072 | // Update lexic permissions for this symbol group |
---|
1073 | $this->lexic_permissions['SYMBOLS'] = ($flag) ? true : false; |
---|
1074 | |
---|
1075 | // For backward compatibility |
---|
1076 | $this->set_brackets_highlighting ($flag); |
---|
1077 | } |
---|
1078 | |
---|
1079 | /** |
---|
1080 | * Sets the styles for strings. If $preserve_defaults is |
---|
1081 | * true, then styles are merged with the default styles, with the |
---|
1082 | * user defined styles having priority |
---|
1083 | * |
---|
1084 | * @param string The style to make the escape characters |
---|
1085 | * @param boolean Whether to merge the new styles with the old or just |
---|
1086 | * to overwrite them |
---|
1087 | * @since 1.0.0 |
---|
1088 | */ |
---|
1089 | function set_strings_style($style, $preserve_defaults = false) { |
---|
1090 | if (!$preserve_defaults) { |
---|
1091 | $this->language_data['STYLES']['STRINGS'][0] = $style; |
---|
1092 | } else { |
---|
1093 | $this->language_data['STYLES']['STRINGS'][0] .= $style; |
---|
1094 | } |
---|
1095 | } |
---|
1096 | |
---|
1097 | /** |
---|
1098 | * Turns highlighting on/off for strings |
---|
1099 | * |
---|
1100 | * @param boolean Whether to turn highlighting for strings on or off |
---|
1101 | * @since 1.0.0 |
---|
1102 | */ |
---|
1103 | function set_strings_highlighting($flag) { |
---|
1104 | $this->lexic_permissions['STRINGS'] = ($flag) ? true : false; |
---|
1105 | } |
---|
1106 | |
---|
1107 | /** |
---|
1108 | * Sets the styles for numbers. If $preserve_defaults is |
---|
1109 | * true, then styles are merged with the default styles, with the |
---|
1110 | * user defined styles having priority |
---|
1111 | * |
---|
1112 | * @param string The style to make the numbers |
---|
1113 | * @param boolean Whether to merge the new styles with the old or just |
---|
1114 | * to overwrite them |
---|
1115 | * @since 1.0.0 |
---|
1116 | */ |
---|
1117 | function set_numbers_style($style, $preserve_defaults = false) { |
---|
1118 | if (!$preserve_defaults) { |
---|
1119 | $this->language_data['STYLES']['NUMBERS'][0] = $style; |
---|
1120 | } else { |
---|
1121 | $this->language_data['STYLES']['NUMBERS'][0] .= $style; |
---|
1122 | } |
---|
1123 | } |
---|
1124 | |
---|
1125 | /** |
---|
1126 | * Turns highlighting on/off for numbers |
---|
1127 | * |
---|
1128 | * @param boolean Whether to turn highlighting for numbers on or off |
---|
1129 | * @since 1.0.0 |
---|
1130 | */ |
---|
1131 | function set_numbers_highlighting($flag) { |
---|
1132 | $this->lexic_permissions['NUMBERS'] = ($flag) ? true : false; |
---|
1133 | } |
---|
1134 | |
---|
1135 | /** |
---|
1136 | * Sets the styles for methods. $key is a number that references the |
---|
1137 | * appropriate "object splitter" - see the language file for the language |
---|
1138 | * you are highlighting to get this number. If $preserve_defaults is |
---|
1139 | * true, then styles are merged with the default styles, with the |
---|
1140 | * user defined styles having priority |
---|
1141 | * |
---|
1142 | * @param int The key of the object splitter to change the styles of |
---|
1143 | * @param string The style to make the methods |
---|
1144 | * @param boolean Whether to merge the new styles with the old or just |
---|
1145 | * to overwrite them |
---|
1146 | * @since 1.0.0 |
---|
1147 | */ |
---|
1148 | function set_methods_style($key, $style, $preserve_defaults = false) { |
---|
1149 | if (!$preserve_defaults) { |
---|
1150 | $this->language_data['STYLES']['METHODS'][$key] = $style; |
---|
1151 | } else { |
---|
1152 | $this->language_data['STYLES']['METHODS'][$key] .= $style; |
---|
1153 | } |
---|
1154 | } |
---|
1155 | |
---|
1156 | /** |
---|
1157 | * Turns highlighting on/off for methods |
---|
1158 | * |
---|
1159 | * @param boolean Whether to turn highlighting for methods on or off |
---|
1160 | * @since 1.0.0 |
---|
1161 | */ |
---|
1162 | function set_methods_highlighting($flag) { |
---|
1163 | $this->lexic_permissions['METHODS'] = ($flag) ? true : false; |
---|
1164 | } |
---|
1165 | |
---|
1166 | /** |
---|
1167 | * Sets the styles for regexps. If $preserve_defaults is |
---|
1168 | * true, then styles are merged with the default styles, with the |
---|
1169 | * user defined styles having priority |
---|
1170 | * |
---|
1171 | * @param string The style to make the regular expression matches |
---|
1172 | * @param boolean Whether to merge the new styles with the old or just |
---|
1173 | * to overwrite them |
---|
1174 | * @since 1.0.0 |
---|
1175 | */ |
---|
1176 | function set_regexps_style($key, $style, $preserve_defaults = false) { |
---|
1177 | if (!$preserve_defaults) { |
---|
1178 | $this->language_data['STYLES']['REGEXPS'][$key] = $style; |
---|
1179 | } else { |
---|
1180 | $this->language_data['STYLES']['REGEXPS'][$key] .= $style; |
---|
1181 | } |
---|
1182 | } |
---|
1183 | |
---|
1184 | /** |
---|
1185 | * Turns highlighting on/off for regexps |
---|
1186 | * |
---|
1187 | * @param int The key of the regular expression group to turn on or off |
---|
1188 | * @param boolean Whether to turn highlighting for the regular expression group on or off |
---|
1189 | * @since 1.0.0 |
---|
1190 | */ |
---|
1191 | function set_regexps_highlighting($key, $flag) { |
---|
1192 | $this->lexic_permissions['REGEXPS'][$key] = ($flag) ? true : false; |
---|
1193 | } |
---|
1194 | |
---|
1195 | /** |
---|
1196 | * Sets whether a set of keywords are checked for in a case sensitive manner |
---|
1197 | * |
---|
1198 | * @param int The key of the keyword group to change the case sensitivity of |
---|
1199 | * @param boolean Whether to check in a case sensitive manner or not |
---|
1200 | * @since 1.0.0 |
---|
1201 | */ |
---|
1202 | function set_case_sensitivity($key, $case) { |
---|
1203 | $this->language_data['CASE_SENSITIVE'][$key] = ($case) ? true : false; |
---|
1204 | } |
---|
1205 | |
---|
1206 | /** |
---|
1207 | * Sets the case that keywords should use when found. Use the constants: |
---|
1208 | * |
---|
1209 | * - GESHI_CAPS_NO_CHANGE: leave keywords as-is |
---|
1210 | * - GESHI_CAPS_UPPER: convert all keywords to uppercase where found |
---|
1211 | * - GESHI_CAPS_LOWER: convert all keywords to lowercase where found |
---|
1212 | * |
---|
1213 | * @param int A constant specifying what to do with matched keywords |
---|
1214 | * @since 1.0.1 |
---|
1215 | */ |
---|
1216 | function set_case_keywords($case) { |
---|
1217 | if (in_array($case, array( |
---|
1218 | GESHI_CAPS_NO_CHANGE, GESHI_CAPS_UPPER, GESHI_CAPS_LOWER))) { |
---|
1219 | $this->language_data['CASE_KEYWORDS'] = $case; |
---|
1220 | } |
---|
1221 | } |
---|
1222 | |
---|
1223 | /** |
---|
1224 | * Sets how many spaces a tab is substituted for |
---|
1225 | * |
---|
1226 | * Widths below zero are ignored |
---|
1227 | * |
---|
1228 | * @param int The tab width |
---|
1229 | * @since 1.0.0 |
---|
1230 | */ |
---|
1231 | function set_tab_width($width) { |
---|
1232 | $this->tab_width = intval($width); |
---|
1233 | |
---|
1234 | //Check if it fit's the constraints: |
---|
1235 | if ($this->tab_width < 1) { |
---|
1236 | //Return it to the default |
---|
1237 | $this->tab_width = 8; |
---|
1238 | } |
---|
1239 | } |
---|
1240 | |
---|
1241 | /** |
---|
1242 | * Sets whether or not to use tab-stop width specifed by language |
---|
1243 | * |
---|
1244 | * @param boolean Whether to use language-specific tab-stop widths |
---|
1245 | * @since 1.0.7.20 |
---|
1246 | */ |
---|
1247 | function set_use_language_tab_width($use) { |
---|
1248 | $this->use_language_tab_width = (bool) $use; |
---|
1249 | } |
---|
1250 | |
---|
1251 | /** |
---|
1252 | * Returns the tab width to use, based on the current language and user |
---|
1253 | * preference |
---|
1254 | * |
---|
1255 | * @return int Tab width |
---|
1256 | * @since 1.0.7.20 |
---|
1257 | */ |
---|
1258 | function get_real_tab_width() { |
---|
1259 | if (!$this->use_language_tab_width || |
---|
1260 | !isset($this->language_data['TAB_WIDTH'])) { |
---|
1261 | return $this->tab_width; |
---|
1262 | } else { |
---|
1263 | return $this->language_data['TAB_WIDTH']; |
---|
1264 | } |
---|
1265 | } |
---|
1266 | |
---|
1267 | /** |
---|
1268 | * Enables/disables strict highlighting. Default is off, calling this |
---|
1269 | * method without parameters will turn it on. See documentation |
---|
1270 | * for more details on strict mode and where to use it. |
---|
1271 | * |
---|
1272 | * @param boolean Whether to enable strict mode or not |
---|
1273 | * @since 1.0.0 |
---|
1274 | */ |
---|
1275 | function enable_strict_mode($mode = true) { |
---|
1276 | if (GESHI_MAYBE == $this->language_data['STRICT_MODE_APPLIES']) { |
---|
1277 | $this->strict_mode = ($mode) ? GESHI_ALWAYS : GESHI_NEVER; |
---|
1278 | } |
---|
1279 | } |
---|
1280 | |
---|
1281 | /** |
---|
1282 | * Disables all highlighting |
---|
1283 | * |
---|
1284 | * @since 1.0.0 |
---|
1285 | * @todo Rewrite with array traversal |
---|
1286 | * @deprecated In favour of enable_highlighting |
---|
1287 | */ |
---|
1288 | function disable_highlighting() { |
---|
1289 | $this->enable_highlighting(false); |
---|
1290 | } |
---|
1291 | |
---|
1292 | /** |
---|
1293 | * Enables all highlighting |
---|
1294 | * |
---|
1295 | * The optional flag parameter was added in version 1.0.7.21 and can be used |
---|
1296 | * to enable (true) or disable (false) all highlighting. |
---|
1297 | * |
---|
1298 | * @since 1.0.0 |
---|
1299 | * @param boolean A flag specifying whether to enable or disable all highlighting |
---|
1300 | * @todo Rewrite with array traversal |
---|
1301 | */ |
---|
1302 | function enable_highlighting($flag = true) { |
---|
1303 | $flag = $flag ? true : false; |
---|
1304 | foreach ($this->lexic_permissions as $key => $value) { |
---|
1305 | if (is_array($value)) { |
---|
1306 | foreach ($value as $k => $v) { |
---|
1307 | $this->lexic_permissions[$key][$k] = $flag; |
---|
1308 | } |
---|
1309 | } else { |
---|
1310 | $this->lexic_permissions[$key] = $flag; |
---|
1311 | } |
---|
1312 | } |
---|
1313 | |
---|
1314 | // Context blocks |
---|
1315 | $this->enable_important_blocks = $flag; |
---|
1316 | } |
---|
1317 | |
---|
1318 | /** |
---|
1319 | * Given a file extension, this method returns either a valid geshi language |
---|
1320 | * name, or the empty string if it couldn't be found |
---|
1321 | * |
---|
1322 | * @param string The extension to get a language name for |
---|
1323 | * @param array A lookup array to use instead of the default one |
---|
1324 | * @since 1.0.5 |
---|
1325 | * @todo Re-think about how this method works (maybe make it private and/or make it |
---|
1326 | * a extension->lang lookup?) |
---|
1327 | * @todo static? |
---|
1328 | */ |
---|
1329 | function get_language_name_from_extension( $extension, $lookup = array() ) { |
---|
1330 | if ( !is_array($lookup) || empty($lookup)) { |
---|
1331 | $lookup = array( |
---|
1332 | 'actionscript' => array('as'), |
---|
1333 | 'ada' => array('a', 'ada', 'adb', 'ads'), |
---|
1334 | 'apache' => array('conf'), |
---|
1335 | 'asm' => array('ash', 'asm', 'inc'), |
---|
1336 | 'asp' => array('asp'), |
---|
1337 | 'bash' => array('sh'), |
---|
1338 | 'bf' => array('bf'), |
---|
1339 | 'c' => array('c', 'h'), |
---|
1340 | 'c_mac' => array('c', 'h'), |
---|
1341 | 'caddcl' => array(), |
---|
1342 | 'cadlisp' => array(), |
---|
1343 | 'cdfg' => array('cdfg'), |
---|
1344 | 'cobol' => array('cbl'), |
---|
1345 | 'cpp' => array('cpp', 'hpp', 'C', 'H', 'CPP', 'HPP'), |
---|
1346 | 'csharp' => array('cs'), |
---|
1347 | 'css' => array('css'), |
---|
1348 | 'd' => array('d'), |
---|
1349 | 'delphi' => array('dpk', 'dpr', 'pp', 'pas'), |
---|
1350 | 'diff' => array('diff', 'patch'), |
---|
1351 | 'dos' => array('bat', 'cmd'), |
---|
1352 | 'gettext' => array('po', 'pot'), |
---|
1353 | 'gml' => array('gml'), |
---|
1354 | 'gnuplot' => array('plt'), |
---|
1355 | 'groovy' => array('groovy'), |
---|
1356 | 'haskell' => array('hs'), |
---|
1357 | 'html4strict' => array('html', 'htm'), |
---|
1358 | 'ini' => array('ini', 'desktop'), |
---|
1359 | 'java' => array('java'), |
---|
1360 | 'javascript' => array('js'), |
---|
1361 | 'klonec' => array('kl1'), |
---|
1362 | 'klonecpp' => array('klx'), |
---|
1363 | 'latex' => array('tex'), |
---|
1364 | 'lisp' => array('lisp'), |
---|
1365 | 'lua' => array('lua'), |
---|
1366 | 'matlab' => array('m'), |
---|
1367 | 'mpasm' => array(), |
---|
1368 | 'mysql' => array('sql'), |
---|
1369 | 'nsis' => array(), |
---|
1370 | 'objc' => array(), |
---|
1371 | 'oobas' => array(), |
---|
1372 | 'oracle8' => array(), |
---|
1373 | 'oracle10' => array(), |
---|
1374 | 'pascal' => array('pas'), |
---|
1375 | 'perl' => array('pl', 'pm'), |
---|
1376 | 'php' => array('php', 'php5', 'phtml', 'phps'), |
---|
1377 | 'povray' => array('pov'), |
---|
1378 | 'providex' => array('pvc', 'pvx'), |
---|
1379 | 'prolog' => array('pl'), |
---|
1380 | 'python' => array('py'), |
---|
1381 | 'qbasic' => array('bi'), |
---|
1382 | 'reg' => array('reg'), |
---|
1383 | 'ruby' => array('rb'), |
---|
1384 | 'sas' => array('sas'), |
---|
1385 | 'scala' => array('scala'), |
---|
1386 | 'scheme' => array('scm'), |
---|
1387 | 'scilab' => array('sci'), |
---|
1388 | 'smalltalk' => array('st'), |
---|
1389 | 'smarty' => array(), |
---|
1390 | 'tcl' => array('tcl'), |
---|
1391 | 'vb' => array('bas'), |
---|
1392 | 'vbnet' => array(), |
---|
1393 | 'visualfoxpro' => array(), |
---|
1394 | 'whitespace' => array('ws'), |
---|
1395 | 'xml' => array('xml', 'svg'), |
---|
1396 | 'z80' => array('z80', 'asm', 'inc') |
---|
1397 | ); |
---|
1398 | } |
---|
1399 | |
---|
1400 | foreach ($lookup as $lang => $extensions) { |
---|
1401 | if (in_array($extension, $extensions)) { |
---|
1402 | return $lang; |
---|
1403 | } |
---|
1404 | } |
---|
1405 | return ''; |
---|
1406 | } |
---|
1407 | |
---|
1408 | /** |
---|
1409 | * Given a file name, this method loads its contents in, and attempts |
---|
1410 | * to set the language automatically. An optional lookup table can be |
---|
1411 | * passed for looking up the language name. If not specified a default |
---|
1412 | * table is used |
---|
1413 | * |
---|
1414 | * The language table is in the form |
---|
1415 | * <pre>array( |
---|
1416 | * 'lang_name' => array('extension', 'extension', ...), |
---|
1417 | * 'lang_name' ... |
---|
1418 | * );</pre> |
---|
1419 | * |
---|
1420 | * @param string The filename to load the source from |
---|
1421 | * @param array A lookup array to use instead of the default one |
---|
1422 | * @todo Complete rethink of this and above method |
---|
1423 | * @since 1.0.5 |
---|
1424 | */ |
---|
1425 | function load_from_file($file_name, $lookup = array()) { |
---|
1426 | if (is_readable($file_name)) { |
---|
1427 | $this->set_source(file_get_contents($file_name)); |
---|
1428 | $this->set_language($this->get_language_name_from_extension(substr(strrchr($file_name, '.'), 1), $lookup)); |
---|
1429 | } else { |
---|
1430 | $this->error = GESHI_ERROR_FILE_NOT_READABLE; |
---|
1431 | } |
---|
1432 | } |
---|
1433 | |
---|
1434 | /** |
---|
1435 | * Adds a keyword to a keyword group for highlighting |
---|
1436 | * |
---|
1437 | * @param int The key of the keyword group to add the keyword to |
---|
1438 | * @param string The word to add to the keyword group |
---|
1439 | * @since 1.0.0 |
---|
1440 | */ |
---|
1441 | function add_keyword($key, $word) { |
---|
1442 | if (!in_array($word, $this->language_data['KEYWORDS'][$key])) { |
---|
1443 | $this->language_data['KEYWORDS'][$key][] = $word; |
---|
1444 | |
---|
1445 | //NEW in 1.0.8 don't recompile the whole optimized regexp, simply append it |
---|
1446 | if ($this->parse_cache_built) { |
---|
1447 | $subkey = count($this->language_data['CACHED_KEYWORD_LISTS'][$key]) - 1; |
---|
1448 | $this->language_data['CACHED_KEYWORD_LISTS'][$key][$subkey] .= '|' . preg_quote($word, '/'); |
---|
1449 | } |
---|
1450 | } |
---|
1451 | } |
---|
1452 | |
---|
1453 | /** |
---|
1454 | * Removes a keyword from a keyword group |
---|
1455 | * |
---|
1456 | * @param int The key of the keyword group to remove the keyword from |
---|
1457 | * @param string The word to remove from the keyword group |
---|
1458 | * @param bool Wether to automatically recompile the optimized regexp list or not. |
---|
1459 | * Note: if you set this to false and @see GeSHi->parse_code() was already called once, |
---|
1460 | * for the current language, you have to manually call @see GeSHi->optimize_keyword_group() |
---|
1461 | * or the removed keyword will stay in cache and still be highlighted! On the other hand |
---|
1462 | * it might be too expensive to recompile the regexp list for every removal if you want to |
---|
1463 | * remove a lot of keywords. |
---|
1464 | * @since 1.0.0 |
---|
1465 | */ |
---|
1466 | function remove_keyword($key, $word, $recompile = true) { |
---|
1467 | $key_to_remove = array_search($word, $this->language_data['KEYWORDS'][$key]); |
---|
1468 | if ($key_to_remove !== false) { |
---|
1469 | unset($this->language_data['KEYWORDS'][$key][$key_to_remove]); |
---|
1470 | |
---|
1471 | //NEW in 1.0.8, optionally recompile keyword group |
---|
1472 | if ($recompile && $this->parse_cache_built) { |
---|
1473 | $this->optimize_keyword_group($key); |
---|
1474 | } |
---|
1475 | } |
---|
1476 | } |
---|
1477 | |
---|
1478 | /** |
---|
1479 | * Creates a new keyword group |
---|
1480 | * |
---|
1481 | * @param int The key of the keyword group to create |
---|
1482 | * @param string The styles for the keyword group |
---|
1483 | * @param boolean Whether the keyword group is case sensitive ornot |
---|
1484 | * @param array The words to use for the keyword group |
---|
1485 | * @since 1.0.0 |
---|
1486 | */ |
---|
1487 | function add_keyword_group($key, $styles, $case_sensitive = true, $words = array()) { |
---|
1488 | $words = (array) $words; |
---|
1489 | if (empty($words)) { |
---|
1490 | // empty word lists mess up highlighting |
---|
1491 | return false; |
---|
1492 | } |
---|
1493 | |
---|
1494 | //Add the new keyword group internally |
---|
1495 | $this->language_data['KEYWORDS'][$key] = $words; |
---|
1496 | $this->lexic_permissions['KEYWORDS'][$key] = true; |
---|
1497 | $this->language_data['CASE_SENSITIVE'][$key] = $case_sensitive; |
---|
1498 | $this->language_data['STYLES']['KEYWORDS'][$key] = $styles; |
---|
1499 | |
---|
1500 | //NEW in 1.0.8, cache keyword regexp |
---|
1501 | if ($this->parse_cache_built) { |
---|
1502 | $this->optimize_keyword_group($key); |
---|
1503 | } |
---|
1504 | } |
---|
1505 | |
---|
1506 | /** |
---|
1507 | * Removes a keyword group |
---|
1508 | * |
---|
1509 | * @param int The key of the keyword group to remove |
---|
1510 | * @since 1.0.0 |
---|
1511 | */ |
---|
1512 | function remove_keyword_group ($key) { |
---|
1513 | //Remove the keyword group internally |
---|
1514 | unset($this->language_data['KEYWORDS'][$key]); |
---|
1515 | unset($this->lexic_permissions['KEYWORDS'][$key]); |
---|
1516 | unset($this->language_data['CASE_SENSITIVE'][$key]); |
---|
1517 | unset($this->language_data['STYLES']['KEYWORDS'][$key]); |
---|
1518 | |
---|
1519 | //NEW in 1.0.8 |
---|
1520 | unset($this->language_data['CACHED_KEYWORD_LISTS'][$key]); |
---|
1521 | } |
---|
1522 | |
---|
1523 | /** |
---|
1524 | * compile optimized regexp list for keyword group |
---|
1525 | * |
---|
1526 | * @param int The key of the keyword group to compile & optimize |
---|
1527 | * @since 1.0.8 |
---|
1528 | */ |
---|
1529 | function optimize_keyword_group($key) { |
---|
1530 | $this->language_data['CACHED_KEYWORD_LISTS'][$key] = |
---|
1531 | $this->optimize_regexp_list($this->language_data['KEYWORDS'][$key]); |
---|
1532 | } |
---|
1533 | |
---|
1534 | /** |
---|
1535 | * Sets the content of the header block |
---|
1536 | * |
---|
1537 | * @param string The content of the header block |
---|
1538 | * @since 1.0.2 |
---|
1539 | */ |
---|
1540 | function set_header_content($content) { |
---|
1541 | $this->header_content = $content; |
---|
1542 | } |
---|
1543 | |
---|
1544 | /** |
---|
1545 | * Sets the content of the footer block |
---|
1546 | * |
---|
1547 | * @param string The content of the footer block |
---|
1548 | * @since 1.0.2 |
---|
1549 | */ |
---|
1550 | function set_footer_content($content) { |
---|
1551 | $this->footer_content = $content; |
---|
1552 | } |
---|
1553 | |
---|
1554 | /** |
---|
1555 | * Sets the style for the header content |
---|
1556 | * |
---|
1557 | * @param string The style for the header content |
---|
1558 | * @since 1.0.2 |
---|
1559 | */ |
---|
1560 | function set_header_content_style($style) { |
---|
1561 | $this->header_content_style = $style; |
---|
1562 | } |
---|
1563 | |
---|
1564 | /** |
---|
1565 | * Sets the style for the footer content |
---|
1566 | * |
---|
1567 | * @param string The style for the footer content |
---|
1568 | * @since 1.0.2 |
---|
1569 | */ |
---|
1570 | function set_footer_content_style($style) { |
---|
1571 | $this->footer_content_style = $style; |
---|
1572 | } |
---|
1573 | |
---|
1574 | /** |
---|
1575 | * Sets whether to force a surrounding block around |
---|
1576 | * the highlighted code or not |
---|
1577 | * |
---|
1578 | * @param boolean Tells whether to enable or disable this feature |
---|
1579 | * @since 1.0.7.20 |
---|
1580 | */ |
---|
1581 | function enable_inner_code_block($flag) { |
---|
1582 | $this->force_code_block = (bool)$flag; |
---|
1583 | } |
---|
1584 | |
---|
1585 | /** |
---|
1586 | * Sets the base URL to be used for keywords |
---|
1587 | * |
---|
1588 | * @param int The key of the keyword group to set the URL for |
---|
1589 | * @param string The URL to set for the group. If {FNAME} is in |
---|
1590 | * the url somewhere, it is replaced by the keyword |
---|
1591 | * that the URL is being made for |
---|
1592 | * @since 1.0.2 |
---|
1593 | */ |
---|
1594 | function set_url_for_keyword_group($group, $url) { |
---|
1595 | $this->language_data['URLS'][$group] = $url; |
---|
1596 | } |
---|
1597 | |
---|
1598 | /** |
---|
1599 | * Sets styles for links in code |
---|
1600 | * |
---|
1601 | * @param int A constant that specifies what state the style is being |
---|
1602 | * set for - e.g. :hover or :visited |
---|
1603 | * @param string The styles to use for that state |
---|
1604 | * @since 1.0.2 |
---|
1605 | */ |
---|
1606 | function set_link_styles($type, $styles) { |
---|
1607 | $this->link_styles[$type] = $styles; |
---|
1608 | } |
---|
1609 | |
---|
1610 | /** |
---|
1611 | * Sets the target for links in code |
---|
1612 | * |
---|
1613 | * @param string The target for links in the code, e.g. _blank |
---|
1614 | * @since 1.0.3 |
---|
1615 | */ |
---|
1616 | function set_link_target($target) { |
---|
1617 | if (!$target) { |
---|
1618 | $this->link_target = ''; |
---|
1619 | } else { |
---|
1620 | $this->link_target = ' target="' . $target . '"'; |
---|
1621 | } |
---|
1622 | } |
---|
1623 | |
---|
1624 | /** |
---|
1625 | * Sets styles for important parts of the code |
---|
1626 | * |
---|
1627 | * @param string The styles to use on important parts of the code |
---|
1628 | * @since 1.0.2 |
---|
1629 | */ |
---|
1630 | function set_important_styles($styles) { |
---|
1631 | $this->important_styles = $styles; |
---|
1632 | } |
---|
1633 | |
---|
1634 | /** |
---|
1635 | * Sets whether context-important blocks are highlighted |
---|
1636 | * |
---|
1637 | * @param boolean Tells whether to enable or disable highlighting of important blocks |
---|
1638 | * @todo REMOVE THIS SHIZ FROM GESHI! |
---|
1639 | * @deprecated |
---|
1640 | * @since 1.0.2 |
---|
1641 | */ |
---|
1642 | function enable_important_blocks($flag) { |
---|
1643 | $this->enable_important_blocks = ( $flag ) ? true : false; |
---|
1644 | } |
---|
1645 | |
---|
1646 | /** |
---|
1647 | * Whether CSS IDs should be added to each line |
---|
1648 | * |
---|
1649 | * @param boolean If true, IDs will be added to each line. |
---|
1650 | * @since 1.0.2 |
---|
1651 | */ |
---|
1652 | function enable_ids($flag = true) { |
---|
1653 | $this->add_ids = ($flag) ? true : false; |
---|
1654 | } |
---|
1655 | |
---|
1656 | /** |
---|
1657 | * Specifies which lines to highlight extra |
---|
1658 | * |
---|
1659 | * The extra style parameter was added in 1.0.7.21. |
---|
1660 | * |
---|
1661 | * @param mixed An array of line numbers to highlight, or just a line |
---|
1662 | * number on its own. |
---|
1663 | * @param string A string specifying the style to use for this line. |
---|
1664 | * If null is specified, the default style is used. |
---|
1665 | * If false is specified, the line will be removed from |
---|
1666 | * special highlighting |
---|
1667 | * @since 1.0.2 |
---|
1668 | * @todo Some data replication here that could be cut down on |
---|
1669 | */ |
---|
1670 | function highlight_lines_extra($lines, $style = null) { |
---|
1671 | if (is_array($lines)) { |
---|
1672 | //Split up the job using single lines at a time |
---|
1673 | foreach ($lines as $line) { |
---|
1674 | $this->highlight_lines_extra($line, $style); |
---|
1675 | } |
---|
1676 | } else { |
---|
1677 | //Mark the line as being highlighted specially |
---|
1678 | $lines = intval($lines); |
---|
1679 | $this->highlight_extra_lines[$lines] = $lines; |
---|
1680 | |
---|
1681 | //Decide on which style to use |
---|
1682 | if ($style === null) { //Check if we should use default style |
---|
1683 | unset($this->highlight_extra_lines_styles[$lines]); |
---|
1684 | } else if ($style === false) { //Check if to remove this line |
---|
1685 | unset($this->highlight_extra_lines[$lines]); |
---|
1686 | unset($this->highlight_extra_lines_styles[$lines]); |
---|
1687 | } else { |
---|
1688 | $this->highlight_extra_lines_styles[$lines] = $style; |
---|
1689 | } |
---|
1690 | } |
---|
1691 | } |
---|
1692 | |
---|
1693 | /** |
---|
1694 | * Sets the style for extra-highlighted lines |
---|
1695 | * |
---|
1696 | * @param string The style for extra-highlighted lines |
---|
1697 | * @since 1.0.2 |
---|
1698 | */ |
---|
1699 | function set_highlight_lines_extra_style($styles) { |
---|
1700 | $this->highlight_extra_lines_style = $styles; |
---|
1701 | } |
---|
1702 | |
---|
1703 | /** |
---|
1704 | * Sets the line-ending |
---|
1705 | * |
---|
1706 | * @param string The new line-ending |
---|
1707 | * @since 1.0.2 |
---|
1708 | */ |
---|
1709 | function set_line_ending($line_ending) { |
---|
1710 | $this->line_ending = (string)$line_ending; |
---|
1711 | } |
---|
1712 | |
---|
1713 | /** |
---|
1714 | * Sets what number line numbers should start at. Should |
---|
1715 | * be a positive integer, and will be converted to one. |
---|
1716 | * |
---|
1717 | * <b>Warning:</b> Using this method will add the "start" |
---|
1718 | * attribute to the <ol> that is used for line numbering. |
---|
1719 | * This is <b>not</b> valid XHTML strict, so if that's what you |
---|
1720 | * care about then don't use this method. Firefox is getting |
---|
1721 | * support for the CSS method of doing this in 1.1 and Opera |
---|
1722 | * has support for the CSS method, but (of course) IE doesn't |
---|
1723 | * so it's not worth doing it the CSS way yet. |
---|
1724 | * |
---|
1725 | * @param int The number to start line numbers at |
---|
1726 | * @since 1.0.2 |
---|
1727 | */ |
---|
1728 | function start_line_numbers_at($number) { |
---|
1729 | $this->line_numbers_start = abs(intval($number)); |
---|
1730 | } |
---|
1731 | |
---|
1732 | /** |
---|
1733 | * Sets the encoding used for htmlspecialchars(), for international |
---|
1734 | * support. |
---|
1735 | * |
---|
1736 | * NOTE: This is not needed for now because htmlspecialchars() is not |
---|
1737 | * being used (it has a security hole in PHP4 that has not been patched). |
---|
1738 | * Maybe in a future version it may make a return for speed reasons, but |
---|
1739 | * I doubt it. |
---|
1740 | * |
---|
1741 | * @param string The encoding to use for the source |
---|
1742 | * @since 1.0.3 |
---|
1743 | */ |
---|
1744 | function set_encoding($encoding) { |
---|
1745 | if ($encoding) { |
---|
1746 | $this->encoding = strtolower($encoding); |
---|
1747 | } |
---|
1748 | } |
---|
1749 | |
---|
1750 | /** |
---|
1751 | * Turns linking of keywords on or off. |
---|
1752 | * |
---|
1753 | * @param boolean If true, links will be added to keywords |
---|
1754 | * @since 1.0.2 |
---|
1755 | */ |
---|
1756 | function enable_keyword_links($enable = true) { |
---|
1757 | $this->keyword_links = (bool) $enable; |
---|
1758 | } |
---|
1759 | |
---|
1760 | /** |
---|
1761 | * Setup caches needed for styling. This is automatically called in |
---|
1762 | * parse_code() and get_stylesheet() when appropriate. This function helps |
---|
1763 | * stylesheet generators as they rely on some style information being |
---|
1764 | * preprocessed |
---|
1765 | * |
---|
1766 | * @since 1.0.8 |
---|
1767 | * @access private |
---|
1768 | */ |
---|
1769 | function build_style_cache() { |
---|
1770 | //Build the style cache needed to highlight numbers appropriate |
---|
1771 | if($this->lexic_permissions['NUMBERS']) { |
---|
1772 | //First check what way highlighting information for numbers are given |
---|
1773 | if(!isset($this->language_data['NUMBERS'])) { |
---|
1774 | $this->language_data['NUMBERS'] = 0; |
---|
1775 | } |
---|
1776 | |
---|
1777 | if(is_array($this->language_data['NUMBERS'])) { |
---|
1778 | $this->language_data['NUMBERS_CACHE'] = $this->language_data['NUMBERS']; |
---|
1779 | } else { |
---|
1780 | $this->language_data['NUMBERS_CACHE'] = array(); |
---|
1781 | if(!$this->language_data['NUMBERS']) { |
---|
1782 | $this->language_data['NUMBERS'] = |
---|
1783 | GESHI_NUMBER_INT_BASIC | |
---|
1784 | GESHI_NUMBER_FLT_NONSCI; |
---|
1785 | } |
---|
1786 | |
---|
1787 | for($i = 0, $j = $this->language_data['NUMBERS']; $j > 0; ++$i, $j>>=1) { |
---|
1788 | //Rearrange style indices if required ... |
---|
1789 | if(isset($this->language_data['STYLES']['NUMBERS'][1<<$i])) { |
---|
1790 | $this->language_data['STYLES']['NUMBERS'][$i] = |
---|
1791 | $this->language_data['STYLES']['NUMBERS'][1<<$i]; |
---|
1792 | unset($this->language_data['STYLES']['NUMBERS'][1<<$i]); |
---|
1793 | } |
---|
1794 | |
---|
1795 | //Check if this bit is set for highlighting |
---|
1796 | if($j&1) { |
---|
1797 | //So this bit is set ... |
---|
1798 | //Check if it belongs to group 0 or the actual stylegroup |
---|
1799 | if(isset($this->language_data['STYLES']['NUMBERS'][$i])) { |
---|
1800 | $this->language_data['NUMBERS_CACHE'][$i] = 1 << $i; |
---|
1801 | } else { |
---|
1802 | if(!isset($this->language_data['NUMBERS_CACHE'][0])) { |
---|
1803 | $this->language_data['NUMBERS_CACHE'][0] = 0; |
---|
1804 | } |
---|
1805 | $this->language_data['NUMBERS_CACHE'][0] |= 1 << $i; |
---|
1806 | } |
---|
1807 | } |
---|
1808 | } |
---|
1809 | } |
---|
1810 | } |
---|
1811 | } |
---|
1812 | |
---|
1813 | /** |
---|
1814 | * Setup caches needed for parsing. This is automatically called in parse_code() when appropriate. |
---|
1815 | * This function makes stylesheet generators much faster as they do not need these caches. |
---|
1816 | * |
---|
1817 | * @since 1.0.8 |
---|
1818 | * @access private |
---|
1819 | */ |
---|
1820 | function build_parse_cache() { |
---|
1821 | // cache symbol regexp |
---|
1822 | //As this is a costy operation, we avoid doing it for multiple groups ... |
---|
1823 | //Instead we perform it for all symbols at once. |
---|
1824 | // |
---|
1825 | //For this to work, we need to reorganize the data arrays. |
---|
1826 | if ($this->lexic_permissions['SYMBOLS'] && !empty($this->language_data['SYMBOLS'])) { |
---|
1827 | $this->language_data['MULTIPLE_SYMBOL_GROUPS'] = count($this->language_data['STYLES']['SYMBOLS']) > 1; |
---|
1828 | |
---|
1829 | $this->language_data['SYMBOL_DATA'] = array(); |
---|
1830 | $symbol_preg_multi = array(); // multi char symbols |
---|
1831 | $symbol_preg_single = array(); // single char symbols |
---|
1832 | foreach ($this->language_data['SYMBOLS'] as $key => $symbols) { |
---|
1833 | if (is_array($symbols)) { |
---|
1834 | foreach ($symbols as $sym) { |
---|
1835 | $sym = $this->hsc($sym); |
---|
1836 | if (!isset($this->language_data['SYMBOL_DATA'][$sym])) { |
---|
1837 | $this->language_data['SYMBOL_DATA'][$sym] = $key; |
---|
1838 | if (isset($sym[1])) { // multiple chars |
---|
1839 | $symbol_preg_multi[] = preg_quote($sym, '/'); |
---|
1840 | } else { // single char |
---|
1841 | if ($sym == '-') { |
---|
1842 | // don't trigger range out of order error |
---|
1843 | $symbol_preg_single[] = '\-'; |
---|
1844 | } else { |
---|
1845 | $symbol_preg_single[] = preg_quote($sym, '/'); |
---|
1846 | } |
---|
1847 | } |
---|
1848 | } |
---|
1849 | } |
---|
1850 | } else { |
---|
1851 | $symbols = $this->hsc($symbols); |
---|
1852 | if (!isset($this->language_data['SYMBOL_DATA'][$symbols])) { |
---|
1853 | $this->language_data['SYMBOL_DATA'][$symbols] = 0; |
---|
1854 | if (isset($symbols[1])) { // multiple chars |
---|
1855 | $symbol_preg_multi[] = preg_quote($symbols, '/'); |
---|
1856 | } else if ($symbols == '-') { |
---|
1857 | // don't trigger range out of order error |
---|
1858 | $symbol_preg_single[] = '\-'; |
---|
1859 | } else { // single char |
---|
1860 | $symbol_preg_single[] = preg_quote($symbols, '/'); |
---|
1861 | } |
---|
1862 | } |
---|
1863 | } |
---|
1864 | } |
---|
1865 | |
---|
1866 | //Now we have an array with each possible symbol as the key and the style as the actual data. |
---|
1867 | //This way we can set the correct style just the moment we highlight ... |
---|
1868 | // |
---|
1869 | //Now we need to rewrite our array to get a search string that |
---|
1870 | $symbol_preg = array(); |
---|
1871 | if (!empty($symbol_preg_multi)) { |
---|
1872 | rsort($symbol_preg_multi); |
---|
1873 | $symbol_preg[] = implode('|', $symbol_preg_multi); |
---|
1874 | } |
---|
1875 | if (!empty($symbol_preg_single)) { |
---|
1876 | rsort($symbol_preg_single); |
---|
1877 | $symbol_preg[] = '[' . implode('', $symbol_preg_single) . ']'; |
---|
1878 | } |
---|
1879 | $this->language_data['SYMBOL_SEARCH'] = implode("|", $symbol_preg); |
---|
1880 | } |
---|
1881 | |
---|
1882 | // cache optimized regexp for keyword matching |
---|
1883 | // remove old cache |
---|
1884 | $this->language_data['CACHED_KEYWORD_LISTS'] = array(); |
---|
1885 | foreach (array_keys($this->language_data['KEYWORDS']) as $key) { |
---|
1886 | if (!isset($this->lexic_permissions['KEYWORDS'][$key]) || |
---|
1887 | $this->lexic_permissions['KEYWORDS'][$key]) { |
---|
1888 | $this->optimize_keyword_group($key); |
---|
1889 | } |
---|
1890 | } |
---|
1891 | |
---|
1892 | // brackets |
---|
1893 | if ($this->lexic_permissions['BRACKETS']) { |
---|
1894 | $this->language_data['CACHE_BRACKET_MATCH'] = array('[', ']', '(', ')', '{', '}'); |
---|
1895 | if (!$this->use_classes && isset($this->language_data['STYLES']['BRACKETS'][0])) { |
---|
1896 | $this->language_data['CACHE_BRACKET_REPLACE'] = array( |
---|
1897 | '<| style="' . $this->language_data['STYLES']['BRACKETS'][0] . '">[|>', |
---|
1898 | '<| style="' . $this->language_data['STYLES']['BRACKETS'][0] . '">]|>', |
---|
1899 | '<| style="' . $this->language_data['STYLES']['BRACKETS'][0] . '">(|>', |
---|
1900 | '<| style="' . $this->language_data['STYLES']['BRACKETS'][0] . '">)|>', |
---|
1901 | '<| style="' . $this->language_data['STYLES']['BRACKETS'][0] . '">{|>', |
---|
1902 | '<| style="' . $this->language_data['STYLES']['BRACKETS'][0] . '">}|>', |
---|
1903 | ); |
---|
1904 | } |
---|
1905 | else { |
---|
1906 | $this->language_data['CACHE_BRACKET_REPLACE'] = array( |
---|
1907 | '<| class="br0">[|>', |
---|
1908 | '<| class="br0">]|>', |
---|
1909 | '<| class="br0">(|>', |
---|
1910 | '<| class="br0">)|>', |
---|
1911 | '<| class="br0">{|>', |
---|
1912 | '<| class="br0">}|>', |
---|
1913 | ); |
---|
1914 | } |
---|
1915 | } |
---|
1916 | |
---|
1917 | //Build the parse cache needed to highlight numbers appropriate |
---|
1918 | if($this->lexic_permissions['NUMBERS']) { |
---|
1919 | //Check if the style rearrangements have been processed ... |
---|
1920 | //This also does some preprocessing to check which style groups are useable ... |
---|
1921 | if(!isset($this->language_data['NUMBERS_CACHE'])) { |
---|
1922 | $this->build_style_cache(); |
---|
1923 | } |
---|
1924 | |
---|
1925 | //Number format specification |
---|
1926 | //All this formats are matched case-insensitively! |
---|
1927 | static $numbers_format = array( |
---|
1928 | GESHI_NUMBER_INT_BASIC => |
---|
1929 | '(?<![0-9a-z_\.%])(?<![\d\.]e[+\-])([1-9]\d*?|0)(?![0-9a-z\.])', |
---|
1930 | GESHI_NUMBER_INT_CSTYLE => |
---|
1931 | '(?<![0-9a-z_\.%])(?<![\d\.]e[+\-])([1-9]\d*?|0)l(?![0-9a-z\.])', |
---|
1932 | GESHI_NUMBER_BIN_SUFFIX => |
---|
1933 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])[01]+?b(?![0-9a-z\.])', |
---|
1934 | GESHI_NUMBER_BIN_PREFIX_PERCENT => |
---|
1935 | '(?<![0-9a-z_\.%])(?<![\d\.]e[+\-])%[01]+?(?![0-9a-z\.])', |
---|
1936 | GESHI_NUMBER_BIN_PREFIX_0B => |
---|
1937 | '(?<![0-9a-z_\.%])(?<![\d\.]e[+\-])0b[01]+?(?![0-9a-z\.])', |
---|
1938 | GESHI_NUMBER_OCT_PREFIX => |
---|
1939 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])0[0-7]+?(?![0-9a-z\.])', |
---|
1940 | GESHI_NUMBER_OCT_SUFFIX => |
---|
1941 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])[0-7]+?o(?![0-9a-z\.])', |
---|
1942 | GESHI_NUMBER_HEX_PREFIX => |
---|
1943 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])0x[0-9a-f]+?(?![0-9a-z\.])', |
---|
1944 | GESHI_NUMBER_HEX_SUFFIX => |
---|
1945 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])\d[0-9a-f]*?h(?![0-9a-z\.])', |
---|
1946 | GESHI_NUMBER_FLT_NONSCI => |
---|
1947 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])\d+?\.\d+?(?![0-9a-z\.])', |
---|
1948 | GESHI_NUMBER_FLT_NONSCI_F => |
---|
1949 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])(?:\d+?(?:\.\d*?)?|\.\d+?)f(?![0-9a-z\.])', |
---|
1950 | GESHI_NUMBER_FLT_SCI_SHORT => |
---|
1951 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])\.\d+?(?:e[+\-]?\d+?)?(?![0-9a-z\.])', |
---|
1952 | GESHI_NUMBER_FLT_SCI_ZERO => |
---|
1953 | '(?<![0-9a-z_\.])(?<![\d\.]e[+\-])(?:\d+?(?:\.\d*?)?|\.\d+?)(?:e[+\-]?\d+?)?(?![0-9a-z\.])' |
---|
1954 | ); |
---|
1955 | |
---|
1956 | //At this step we have an associative array with flag groups for a |
---|
1957 | //specific style or an string denoting a regexp given its index. |
---|
1958 | $this->language_data['NUMBERS_RXCACHE'] = array(); |
---|
1959 | foreach($this->language_data['NUMBERS_CACHE'] as $key => $rxdata) { |
---|
1960 | if(is_string($rxdata)) { |
---|
1961 | $regexp = $rxdata; |
---|
1962 | } else { |
---|
1963 | //This is a bitfield of number flags to highlight: |
---|
1964 | //Build an array, implode them together and make this the actual RX |
---|
1965 | $rxuse = array(); |
---|
1966 | for($i = 1; $i <= $rxdata; $i<<=1) { |
---|
1967 | if($rxdata & $i) { |
---|
1968 | $rxuse[] = $numbers_format[$i]; |
---|
1969 | } |
---|
1970 | } |
---|
1971 | $regexp = implode("|", $rxuse); |
---|
1972 | } |
---|
1973 | |
---|
1974 | $this->language_data['NUMBERS_RXCACHE'][$key] = |
---|
1975 | "/(?<!<\|\/NUM!)(?<!\d\/>)($regexp)(?!\|>)/i"; |
---|
1976 | } |
---|
1977 | } |
---|
1978 | |
---|
1979 | $this->parse_cache_built = true; |
---|
1980 | } |
---|
1981 | |
---|
1982 | /** |
---|
1983 | * Returns the code in $this->source, highlighted and surrounded by the |
---|
1984 | * nessecary HTML. |
---|
1985 | * |
---|
1986 | * This should only be called ONCE, cos it's SLOW! If you want to highlight |
---|
1987 | * the same source multiple times, you're better off doing a whole lot of |
---|
1988 | * str_replaces to replace the <span>s |
---|
1989 | * |
---|
1990 | * @since 1.0.0 |
---|
1991 | */ |
---|
1992 | function parse_code () { |
---|
1993 | // Start the timer |
---|
1994 | $start_time = microtime(); |
---|
1995 | |
---|
1996 | // Firstly, if there is an error, we won't highlight |
---|
1997 | if ($this->error) { |
---|
1998 | //Escape the source for output |
---|
1999 | $result = $this->hsc($this->source); |
---|
2000 | |
---|
2001 | //This fix is related to SF#1923020, but has to be applied regardless of |
---|
2002 | //actually highlighting symbols. |
---|
2003 | $result = str_replace(array('<SEMI>', '<PIPE>'), array(';', '|'), $result); |
---|
2004 | |
---|
2005 | // Timing is irrelevant |
---|
2006 | $this->set_time($start_time, $start_time); |
---|
2007 | $this->finalise($result); |
---|
2008 | return $result; |
---|
2009 | } |
---|
2010 | |
---|
2011 | // make sure the parse cache is up2date |
---|
2012 | if (!$this->parse_cache_built) { |
---|
2013 | $this->build_parse_cache(); |
---|
2014 | } |
---|
2015 | |
---|
2016 | // Replace all newlines to a common form. |
---|
2017 | $code = str_replace("\r\n", "\n", $this->source); |
---|
2018 | $code = str_replace("\r", "\n", $code); |
---|
2019 | |
---|
2020 | // Add spaces for regular expression matching and line numbers |
---|
2021 | // $code = "\n" . $code . "\n"; |
---|
2022 | |
---|
2023 | // Initialise various stuff |
---|
2024 | $length = strlen($code); |
---|
2025 | $COMMENT_MATCHED = false; |
---|
2026 | $stuff_to_parse = ''; |
---|
2027 | $endresult = ''; |
---|
2028 | |
---|
2029 | // "Important" selections are handled like multiline comments |
---|
2030 | // @todo GET RID OF THIS SHIZ |
---|
2031 | if ($this->enable_important_blocks) { |
---|
2032 | $this->language_data['COMMENT_MULTI'][GESHI_START_IMPORTANT] = GESHI_END_IMPORTANT; |
---|
2033 | } |
---|
2034 | |
---|
2035 | if ($this->strict_mode) { |
---|
2036 | // Break the source into bits. Each bit will be a portion of the code |
---|
2037 | // within script delimiters - for example, HTML between < and > |
---|
2038 | $k = 0; |
---|
2039 | $parts = array(); |
---|
2040 | $matches = array(); |
---|
2041 | $next_match_pointer = null; |
---|
2042 | // we use a copy to unset delimiters on demand (when they are not found) |
---|
2043 | $delim_copy = $this->language_data['SCRIPT_DELIMITERS']; |
---|
2044 | $i = 0; |
---|
2045 | while ($i < $length) { |
---|
2046 | $next_match_pos = $length + 1; // never true |
---|
2047 | foreach ($delim_copy as $dk => $delimiters) { |
---|
2048 | if(is_array($delimiters)) { |
---|
2049 | foreach ($delimiters as $open => $close) { |
---|
2050 | // make sure the cache is setup properly |
---|
2051 | if (!isset($matches[$dk][$open])) { |
---|
2052 | $matches[$dk][$open] = array( |
---|
2053 | 'next_match' => -1, |
---|
2054 | 'dk' => $dk, |
---|
2055 | |
---|
2056 | 'open' => $open, // needed for grouping of adjacent code blocks (see below) |
---|
2057 | 'open_strlen' => strlen($open), |
---|
2058 | |
---|
2059 | 'close' => $close, |
---|
2060 | 'close_strlen' => strlen($close), |
---|
2061 | ); |
---|
2062 | } |
---|
2063 | // Get the next little bit for this opening string |
---|
2064 | if ($matches[$dk][$open]['next_match'] < $i) { |
---|
2065 | // only find the next pos if it was not already cached |
---|
2066 | $open_pos = strpos($code, $open, $i); |
---|
2067 | if ($open_pos === false) { |
---|
2068 | // no match for this delimiter ever |
---|
2069 | unset($delim_copy[$dk][$open]); |
---|
2070 | continue; |
---|
2071 | } |
---|
2072 | $matches[$dk][$open]['next_match'] = $open_pos; |
---|
2073 | } |
---|
2074 | if ($matches[$dk][$open]['next_match'] < $next_match_pos) { |
---|
2075 | //So we got a new match, update the close_pos |
---|
2076 | $matches[$dk][$open]['close_pos'] = |
---|
2077 | strpos($code, $close, $matches[$dk][$open]['next_match']+1); |
---|
2078 | |
---|
2079 | $next_match_pointer =& $matches[$dk][$open]; |
---|
2080 | $next_match_pos = $matches[$dk][$open]['next_match']; |
---|
2081 | } |
---|
2082 | } |
---|
2083 | } else { |
---|
2084 | //So we should match an RegExp as Strict Block ... |
---|
2085 | /** |
---|
2086 | * The value in $delimiters is expected to be an RegExp |
---|
2087 | * containing exactly 2 matching groups: |
---|
2088 | * - Group 1 is the opener |
---|
2089 | * - Group 2 is the closer |
---|
2090 | */ |
---|
2091 | if(!GESHI_PHP_PRE_433 && //Needs proper rewrite to work with PHP >=4.3.0; 4.3.3 is guaranteed to work. |
---|
2092 | preg_match($delimiters, $code, $matches_rx, PREG_OFFSET_CAPTURE, $i)) { |
---|
2093 | //We got a match ... |
---|
2094 | $matches[$dk] = array( |
---|
2095 | 'next_match' => $matches_rx[1][1], |
---|
2096 | 'dk' => $dk, |
---|
2097 | |
---|
2098 | 'close_strlen' => strlen($matches_rx[2][0]), |
---|
2099 | 'close_pos' => $matches_rx[2][1], |
---|
2100 | ); |
---|
2101 | } else { |
---|
2102 | // no match for this delimiter ever |
---|
2103 | unset($delim_copy[$dk]); |
---|
2104 | continue; |
---|
2105 | } |
---|
2106 | |
---|
2107 | if ($matches[$dk]['next_match'] <= $next_match_pos) { |
---|
2108 | $next_match_pointer =& $matches[$dk]; |
---|
2109 | $next_match_pos = $matches[$dk]['next_match']; |
---|
2110 | } |
---|
2111 | } |
---|
2112 | } |
---|
2113 | // non-highlightable text |
---|
2114 | $parts[$k] = array( |
---|
2115 | 1 => substr($code, $i, $next_match_pos - $i) |
---|
2116 | ); |
---|
2117 | ++$k; |
---|
2118 | |
---|
2119 | if ($next_match_pos > $length) { |
---|
2120 | // out of bounds means no next match was found |
---|
2121 | break; |
---|
2122 | } |
---|
2123 | |
---|
2124 | // highlightable code |
---|
2125 | $parts[$k][0] = $next_match_pointer['dk']; |
---|
2126 | |
---|
2127 | //Only combine for non-rx script blocks |
---|
2128 | if(is_array($delim_copy[$next_match_pointer['dk']])) { |
---|
2129 | // group adjacent script blocks, e.g. <foobar><asdf> should be one block, not three! |
---|
2130 | $i = $next_match_pos + $next_match_pointer['open_strlen']; |
---|
2131 | while (true) { |
---|
2132 | $close_pos = strpos($code, $next_match_pointer['close'], $i); |
---|
2133 | if ($close_pos == false) { |
---|
2134 | break; |
---|
2135 | } |
---|
2136 | $i = $close_pos + $next_match_pointer['close_strlen']; |
---|
2137 | if ($i == $length) { |
---|
2138 | break; |
---|
2139 | } |
---|
2140 | if ($code[$i] == $next_match_pointer['open'][0] && ($next_match_pointer['open_strlen'] == 1 || |
---|
2141 | substr($code, $i, $next_match_pointer['open_strlen']) == $next_match_pointer['open'])) { |
---|
2142 | // merge adjacent but make sure we don't merge things like <tag><!-- comment --> |
---|
2143 | foreach ($matches as $submatches) { |
---|
2144 | foreach ($submatches as $match) { |
---|
2145 | if ($match['next_match'] == $i) { |
---|
2146 | // a different block already matches here! |
---|
2147 | break 3; |
---|
2148 | } |
---|
2149 | } |
---|
2150 | } |
---|
2151 | } else { |
---|
2152 | break; |
---|
2153 | } |
---|
2154 | } |
---|
2155 | } else { |
---|
2156 | $close_pos = $next_match_pointer['close_pos'] + $next_match_pointer['close_strlen']; |
---|
2157 | $i = $close_pos; |
---|
2158 | } |
---|
2159 | |
---|
2160 | if ($close_pos === false) { |
---|
2161 | // no closing delimiter found! |
---|
2162 | $parts[$k][1] = substr($code, $next_match_pos); |
---|
2163 | ++$k; |
---|
2164 | break; |
---|
2165 | } else { |
---|
2166 | $parts[$k][1] = substr($code, $next_match_pos, $i - $next_match_pos); |
---|
2167 | ++$k; |
---|
2168 | } |
---|
2169 | } |
---|
2170 | unset($delim_copy, $next_match_pointer, $next_match_pos, $matches); |
---|
2171 | $num_parts = $k; |
---|
2172 | |
---|
2173 | if ($num_parts == 1 && $this->strict_mode == GESHI_MAYBE) { |
---|
2174 | // when we have only one part, we don't have anything to highlight at all. |
---|
2175 | // if we have a "maybe" strict language, this should be handled as highlightable code |
---|
2176 | $parts = array( |
---|
2177 | 0 => array( |
---|
2178 | 0 => '', |
---|
2179 | 1 => '' |
---|
2180 | ), |
---|
2181 | 1 => array( |
---|
2182 | 0 => null, |
---|
2183 | 1 => $parts[0][1] |
---|
2184 | ) |
---|
2185 | ); |
---|
2186 | $num_parts = 2; |
---|
2187 | } |
---|
2188 | |
---|
2189 | } else { |
---|
2190 | // Not strict mode - simply dump the source into |
---|
2191 | // the array at index 1 (the first highlightable block) |
---|
2192 | $parts = array( |
---|
2193 | 0 => array( |
---|
2194 | 0 => '', |
---|
2195 | 1 => '' |
---|
2196 | ), |
---|
2197 | 1 => array( |
---|
2198 | 0 => null, |
---|
2199 | 1 => $code |
---|
2200 | ) |
---|
2201 | ); |
---|
2202 | $num_parts = 2; |
---|
2203 | } |
---|
2204 | |
---|
2205 | //Unset variables we won't need any longer |
---|
2206 | unset($code); |
---|
2207 | |
---|
2208 | //Preload some repeatedly used values regarding hardquotes ... |
---|
2209 | $hq = isset($this->language_data['HARDQUOTE']) ? $this->language_data['HARDQUOTE'][0] : false; |
---|
2210 | $hq_strlen = strlen($hq); |
---|
2211 | |
---|
2212 | //Preload if line numbers are to be generated afterwards |
---|
2213 | //Added a check if line breaks should be forced even without line numbers, fixes SF#1727398 |
---|
2214 | $check_linenumbers = $this->line_numbers != GESHI_NO_LINE_NUMBERS || |
---|
2215 | !empty($this->highlight_extra_lines) || !$this->allow_multiline_span; |
---|
2216 | |
---|
2217 | //preload the escape char for faster checking ... |
---|
2218 | $escaped_escape_char = $this->hsc($this->language_data['ESCAPE_CHAR']); |
---|
2219 | |
---|
2220 | // this is used for single-line comments |
---|
2221 | $sc_disallowed_before = ""; |
---|
2222 | $sc_disallowed_after = ""; |
---|
2223 | |
---|
2224 | if (isset($this->language_data['PARSER_CONTROL'])) { |
---|
2225 | if (isset($this->language_data['PARSER_CONTROL']['COMMENTS'])) { |
---|
2226 | if (isset($this->language_data['PARSER_CONTROL']['COMMENTS']['DISALLOWED_BEFORE'])) { |
---|
2227 | $sc_disallowed_before = $this->language_data['PARSER_CONTROL']['COMMENTS']['DISALLOWED_BEFORE']; |
---|
2228 | } |
---|
2229 | if (isset($this->language_data['PARSER_CONTROL']['COMMENTS']['DISALLOWED_AFTER'])) { |
---|
2230 | $sc_disallowed_after = $this->language_data['PARSER_CONTROL']['COMMENTS']['DISALLOWED_AFTER']; |
---|
2231 | } |
---|
2232 | } |
---|
2233 | } |
---|
2234 | |
---|
2235 | //Fix for SF#1932083: Multichar Quotemarks unsupported |
---|
2236 | $is_string_starter = array(); |
---|
2237 | if ($this->lexic_permissions['STRINGS']) { |
---|
2238 | foreach ($this->language_data['QUOTEMARKS'] as $quotemark) { |
---|
2239 | if (!isset($is_string_starter[$quotemark[0]])) { |
---|
2240 | $is_string_starter[$quotemark[0]] = (string)$quotemark; |
---|
2241 | } else if (is_string($is_string_starter[$quotemark[0]])) { |
---|
2242 | $is_string_starter[$quotemark[0]] = array( |
---|
2243 | $is_string_starter[$quotemark[0]], |
---|
2244 | $quotemark); |
---|
2245 | } else { |
---|
2246 | $is_string_starter[$quotemark[0]][] = $quotemark; |
---|
2247 | } |
---|
2248 | } |
---|
2249 | } |
---|
2250 | |
---|
2251 | // Now we go through each part. We know that even-indexed parts are |
---|
2252 | // code that shouldn't be highlighted, and odd-indexed parts should |
---|
2253 | // be highlighted |
---|
2254 | for ($key = 0; $key < $num_parts; ++$key) { |
---|
2255 | $STRICTATTRS = ''; |
---|
2256 | |
---|
2257 | // If this block should be highlighted... |
---|
2258 | if (!($key & 1)) { |
---|
2259 | // Else not a block to highlight |
---|
2260 | $endresult .= $this->hsc($parts[$key][1]); |
---|
2261 | unset($parts[$key]); |
---|
2262 | continue; |
---|
2263 | } |
---|
2264 | |
---|
2265 | $result = ''; |
---|
2266 | $part = $parts[$key][1]; |
---|
2267 | |
---|
2268 | $highlight_part = true; |
---|
2269 | if ($this->strict_mode && !is_null($parts[$key][0])) { |
---|
2270 | // get the class key for this block of code |
---|
2271 | $script_key = $parts[$key][0]; |
---|
2272 | $highlight_part = $this->language_data['HIGHLIGHT_STRICT_BLOCK'][$script_key]; |
---|
2273 | if ($this->language_data['STYLES']['SCRIPT'][$script_key] != '' && |
---|
2274 | $this->lexic_permissions['SCRIPT']) { |
---|
2275 | // Add a span element around the source to |
---|
2276 | // highlight the overall source block |
---|
2277 | if (!$this->use_classes && |
---|
2278 | $this->language_data['STYLES']['SCRIPT'][$script_key] != '') { |
---|
2279 | $attributes = ' style="' . $this->language_data['STYLES']['SCRIPT'][$script_key] . '"'; |
---|
2280 | } else { |
---|
2281 | $attributes = ' class="sc' . $script_key . '"'; |
---|
2282 | } |
---|
2283 | $result .= "<span$attributes>"; |
---|
2284 | $STRICTATTRS = $attributes; |
---|
2285 | } |
---|
2286 | } |
---|
2287 | |
---|
2288 | if ($highlight_part) { |
---|
2289 | // Now, highlight the code in this block. This code |
---|
2290 | // is really the engine of GeSHi (along with the method |
---|
2291 | // parse_non_string_part). |
---|
2292 | |
---|
2293 | // cache comment regexps incrementally |
---|
2294 | $next_comment_regexp_key = ''; |
---|
2295 | $next_comment_regexp_pos = -1; |
---|
2296 | $next_comment_multi_pos = -1; |
---|
2297 | $next_comment_single_pos = -1; |
---|
2298 | $comment_regexp_cache_per_key = array(); |
---|
2299 | $comment_multi_cache_per_key = array(); |
---|
2300 | $comment_single_cache_per_key = array(); |
---|
2301 | $next_open_comment_multi = ''; |
---|
2302 | $next_comment_single_key = ''; |
---|
2303 | $escape_regexp_cache_per_key = array(); |
---|
2304 | $next_escape_regexp_key = ''; |
---|
2305 | $next_escape_regexp_pos = -1; |
---|
2306 | |
---|
2307 | $length = strlen($part); |
---|
2308 | for ($i = 0; $i < $length; ++$i) { |
---|
2309 | // Get the next char |
---|
2310 | $char = $part[$i]; |
---|
2311 | $char_len = 1; |
---|
2312 | |
---|
2313 | // update regexp comment cache if needed |
---|
2314 | if (isset($this->language_data['COMMENT_REGEXP']) && $next_comment_regexp_pos < $i) { |
---|
2315 | $next_comment_regexp_pos = $length; |
---|
2316 | foreach ($this->language_data['COMMENT_REGEXP'] as $comment_key => $regexp) { |
---|
2317 | $match_i = false; |
---|
2318 | if (isset($comment_regexp_cache_per_key[$comment_key]) && |
---|
2319 | ($comment_regexp_cache_per_key[$comment_key]['pos'] >= $i || |
---|
2320 | $comment_regexp_cache_per_key[$comment_key]['pos'] === false)) { |
---|
2321 | // we have already matched something |
---|
2322 | if ($comment_regexp_cache_per_key[$comment_key]['pos'] === false) { |
---|
2323 | // this comment is never matched |
---|
2324 | continue; |
---|
2325 | } |
---|
2326 | $match_i = $comment_regexp_cache_per_key[$comment_key]['pos']; |
---|
2327 | } else if ( |
---|
2328 | //This is to allow use of the offset parameter in preg_match and stay as compatible with older PHP versions as possible |
---|
2329 | (GESHI_PHP_PRE_433 && preg_match($regexp, substr($part, $i), $match, PREG_OFFSET_CAPTURE)) || |
---|
2330 | (!GESHI_PHP_PRE_433 && preg_match($regexp, $part, $match, PREG_OFFSET_CAPTURE, $i)) |
---|
2331 | ) { |
---|
2332 | $match_i = $match[0][1]; |
---|
2333 | if (GESHI_PHP_PRE_433) { |
---|
2334 | $match_i += $i; |
---|
2335 | } |
---|
2336 | |
---|
2337 | $comment_regexp_cache_per_key[$comment_key] = array( |
---|
2338 | 'key' => $comment_key, |
---|
2339 | 'length' => strlen($match[0][0]), |
---|
2340 | 'pos' => $match_i |
---|
2341 | ); |
---|
2342 | } else { |
---|
2343 | $comment_regexp_cache_per_key[$comment_key]['pos'] = false; |
---|
2344 | continue; |
---|
2345 | } |
---|
2346 | |
---|
2347 | if ($match_i !== false && $match_i < $next_comment_regexp_pos) { |
---|
2348 | $next_comment_regexp_pos = $match_i; |
---|
2349 | $next_comment_regexp_key = $comment_key; |
---|
2350 | if ($match_i === $i) { |
---|
2351 | break; |
---|
2352 | } |
---|
2353 | } |
---|
2354 | } |
---|
2355 | } |
---|
2356 | |
---|
2357 | $string_started = false; |
---|
2358 | |
---|
2359 | if (isset($is_string_starter[$char])) { |
---|
2360 | // Possibly the start of a new string ... |
---|
2361 | |
---|
2362 | //Check which starter it was ... |
---|
2363 | //Fix for SF#1932083: Multichar Quotemarks unsupported |
---|
2364 | if (is_array($is_string_starter[$char])) { |
---|
2365 | $char_new = ''; |
---|
2366 | foreach ($is_string_starter[$char] as $testchar) { |
---|
2367 | if ($testchar === substr($part, $i, strlen($testchar)) && |
---|
2368 | strlen($testchar) > strlen($char_new)) { |
---|
2369 | $char_new = $testchar; |
---|
2370 | $string_started = true; |
---|
2371 | } |
---|
2372 | } |
---|
2373 | if ($string_started) { |
---|
2374 | $char = $char_new; |
---|
2375 | } |
---|
2376 | } else { |
---|
2377 | $testchar = $is_string_starter[$char]; |
---|
2378 | if ($testchar === substr($part, $i, strlen($testchar))) { |
---|
2379 | $char = $testchar; |
---|
2380 | $string_started = true; |
---|
2381 | } |
---|
2382 | } |
---|
2383 | $char_len = strlen($char); |
---|
2384 | } |
---|
2385 | |
---|
2386 | if ($string_started && $i != $next_comment_regexp_pos) { |
---|
2387 | // Hand out the correct style information for this string |
---|
2388 | $string_key = array_search($char, $this->language_data['QUOTEMARKS']); |
---|
2389 | if (!isset($this->language_data['STYLES']['STRINGS'][$string_key]) || |
---|
2390 | !isset($this->language_data['STYLES']['ESCAPE_CHAR'][$string_key])) { |
---|
2391 | $string_key = 0; |
---|
2392 | } |
---|
2393 | |
---|
2394 | // parse the stuff before this |
---|
2395 | $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2396 | $stuff_to_parse = ''; |
---|
2397 | |
---|
2398 | if (!$this->use_classes) { |
---|
2399 | $string_attributes = ' style="' . $this->language_data['STYLES']['STRINGS'][$string_key] . '"'; |
---|
2400 | } else { |
---|
2401 | $string_attributes = ' class="st'.$string_key.'"'; |
---|
2402 | } |
---|
2403 | |
---|
2404 | // now handle the string |
---|
2405 | $string = "<span$string_attributes>" . GeSHi::hsc($char); |
---|
2406 | $start = $i + $char_len; |
---|
2407 | $string_open = true; |
---|
2408 | |
---|
2409 | if(empty($this->language_data['ESCAPE_REGEXP'])) { |
---|
2410 | $next_escape_regexp_pos = $length; |
---|
2411 | } |
---|
2412 | |
---|
2413 | do { |
---|
2414 | //Get the regular ending pos ... |
---|
2415 | $close_pos = strpos($part, $char, $start); |
---|
2416 | if(false === $close_pos) { |
---|
2417 | $close_pos = $length; |
---|
2418 | } |
---|
2419 | |
---|
2420 | if($this->lexic_permissions['ESCAPE_CHAR']) { |
---|
2421 | // update escape regexp cache if needed |
---|
2422 | if (isset($this->language_data['ESCAPE_REGEXP']) && $next_escape_regexp_pos < $start) { |
---|
2423 | $next_escape_regexp_pos = $length; |
---|
2424 | foreach ($this->language_data['ESCAPE_REGEXP'] as $escape_key => $regexp) { |
---|
2425 | $match_i = false; |
---|
2426 | if (isset($escape_regexp_cache_per_key[$escape_key]) && |
---|
2427 | ($escape_regexp_cache_per_key[$escape_key]['pos'] >= $start || |
---|
2428 | $escape_regexp_cache_per_key[$escape_key]['pos'] === false)) { |
---|
2429 | // we have already matched something |
---|
2430 | if ($escape_regexp_cache_per_key[$escape_key]['pos'] === false) { |
---|
2431 | // this comment is never matched |
---|
2432 | continue; |
---|
2433 | } |
---|
2434 | $match_i = $escape_regexp_cache_per_key[$escape_key]['pos']; |
---|
2435 | } else if ( |
---|
2436 | //This is to allow use of the offset parameter in preg_match and stay as compatible with older PHP versions as possible |
---|
2437 | (GESHI_PHP_PRE_433 && preg_match($regexp, substr($part, $start), $match, PREG_OFFSET_CAPTURE)) || |
---|
2438 | (!GESHI_PHP_PRE_433 && preg_match($regexp, $part, $match, PREG_OFFSET_CAPTURE, $start)) |
---|
2439 | ) { |
---|
2440 | $match_i = $match[0][1]; |
---|
2441 | if (GESHI_PHP_PRE_433) { |
---|
2442 | $match_i += $start; |
---|
2443 | } |
---|
2444 | |
---|
2445 | $escape_regexp_cache_per_key[$escape_key] = array( |
---|
2446 | 'key' => $escape_key, |
---|
2447 | 'length' => strlen($match[0][0]), |
---|
2448 | 'pos' => $match_i |
---|
2449 | ); |
---|
2450 | } else { |
---|
2451 | $escape_regexp_cache_per_key[$escape_key]['pos'] = false; |
---|
2452 | continue; |
---|
2453 | } |
---|
2454 | |
---|
2455 | if ($match_i !== false && $match_i < $next_escape_regexp_pos) { |
---|
2456 | $next_escape_regexp_pos = $match_i; |
---|
2457 | $next_escape_regexp_key = $escape_key; |
---|
2458 | if ($match_i === $start) { |
---|
2459 | break; |
---|
2460 | } |
---|
2461 | } |
---|
2462 | } |
---|
2463 | } |
---|
2464 | |
---|
2465 | //Find the next simple escape position |
---|
2466 | if('' != $this->language_data['ESCAPE_CHAR']) { |
---|
2467 | $simple_escape = strpos($part, $this->language_data['ESCAPE_CHAR'], $start); |
---|
2468 | if(false === $simple_escape) { |
---|
2469 | $simple_escape = $length; |
---|
2470 | } |
---|
2471 | } else { |
---|
2472 | $simple_escape = $length; |
---|
2473 | } |
---|
2474 | } else { |
---|
2475 | $next_escape_regexp_pos = $length; |
---|
2476 | $simple_escape = $length; |
---|
2477 | } |
---|
2478 | |
---|
2479 | if($simple_escape < $next_escape_regexp_pos && |
---|
2480 | $simple_escape < $length && |
---|
2481 | $simple_escape < $close_pos) { |
---|
2482 | //The nexxt escape sequence is a simple one ... |
---|
2483 | $es_pos = $simple_escape; |
---|
2484 | |
---|
2485 | //Add the stuff not in the string yet ... |
---|
2486 | $string .= $this->hsc(substr($part, $start, $es_pos - $start)); |
---|
2487 | |
---|
2488 | //Get the style for this escaped char ... |
---|
2489 | if (!$this->use_classes) { |
---|
2490 | $escape_char_attributes = ' style="' . $this->language_data['STYLES']['ESCAPE_CHAR'][0] . '"'; |
---|
2491 | } else { |
---|
2492 | $escape_char_attributes = ' class="es0"'; |
---|
2493 | } |
---|
2494 | |
---|
2495 | //Add the style for the escape char ... |
---|
2496 | $string .= "<span$escape_char_attributes>" . |
---|
2497 | GeSHi::hsc($this->language_data['ESCAPE_CHAR']); |
---|
2498 | |
---|
2499 | //Get the byte AFTER the ESCAPE_CHAR we just found |
---|
2500 | $es_char = $part[$es_pos + 1]; |
---|
2501 | if ($es_char == "\n") { |
---|
2502 | // don't put a newline around newlines |
---|
2503 | $string .= "</span>\n"; |
---|
2504 | $start = $es_pos + 2; |
---|
2505 | } else if (ord($es_char) >= 128) { |
---|
2506 | //This is an non-ASCII char (UTF8 or single byte) |
---|
2507 | //This code tries to work around SF#2037598 ... |
---|
2508 | if(function_exists('mb_substr')) { |
---|
2509 | $es_char_m = mb_substr(substr($part, $es_pos+1, 16), 0, 1, $this->encoding); |
---|
2510 | $string .= $es_char_m . '</span>'; |
---|
2511 | } else if (!GESHI_PHP_PRE_433 && 'utf-8' == $this->encoding) { |
---|
2512 | if(preg_match("/[\xC2-\xDF][\x80-\xBF]". |
---|
2513 | "|\xE0[\xA0-\xBF][\x80-\xBF]". |
---|
2514 | "|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}". |
---|
2515 | "|\xED[\x80-\x9F][\x80-\xBF]". |
---|
2516 | "|\xF0[\x90-\xBF][\x80-\xBF]{2}". |
---|
2517 | "|[\xF1-\xF3][\x80-\xBF]{3}". |
---|
2518 | "|\xF4[\x80-\x8F][\x80-\xBF]{2}/s", |
---|
2519 | $part, $es_char_m, null, $es_pos + 1)) { |
---|
2520 | $es_char_m = $es_char_m[0]; |
---|
2521 | } else { |
---|
2522 | $es_char_m = $es_char; |
---|
2523 | } |
---|
2524 | $string .= $this->hsc($es_char_m) . '</span>'; |
---|
2525 | } else { |
---|
2526 | $es_char_m = $this->hsc($es_char); |
---|
2527 | } |
---|
2528 | $start = $es_pos + strlen($es_char_m) + 1; |
---|
2529 | } else { |
---|
2530 | $string .= $this->hsc($es_char) . '</span>'; |
---|
2531 | $start = $es_pos + 2; |
---|
2532 | } |
---|
2533 | } else if ($next_escape_regexp_pos < $length && |
---|
2534 | $next_escape_regexp_pos < $close_pos) { |
---|
2535 | $es_pos = $next_escape_regexp_pos; |
---|
2536 | //Add the stuff not in the string yet ... |
---|
2537 | $string .= $this->hsc(substr($part, $start, $es_pos - $start)); |
---|
2538 | |
---|
2539 | //Get the key and length of this match ... |
---|
2540 | $escape = $escape_regexp_cache_per_key[$next_escape_regexp_key]; |
---|
2541 | $escape_str = substr($part, $es_pos, $escape['length']); |
---|
2542 | $escape_key = $escape['key']; |
---|
2543 | |
---|
2544 | //Get the style for this escaped char ... |
---|
2545 | if (!$this->use_classes) { |
---|
2546 | $escape_char_attributes = ' style="' . $this->language_data['STYLES']['ESCAPE_CHAR'][$escape_key] . '"'; |
---|
2547 | } else { |
---|
2548 | $escape_char_attributes = ' class="es' . $escape_key . '"'; |
---|
2549 | } |
---|
2550 | |
---|
2551 | //Add the style for the escape char ... |
---|
2552 | $string .= "<span$escape_char_attributes>" . |
---|
2553 | $this->hsc($escape_str) . '</span>'; |
---|
2554 | |
---|
2555 | $start = $es_pos + $escape['length']; |
---|
2556 | } else { |
---|
2557 | //Copy the remainder of the string ... |
---|
2558 | $string .= $this->hsc(substr($part, $start, $close_pos - $start + $char_len)) . '</span>'; |
---|
2559 | $start = $close_pos + $char_len; |
---|
2560 | $string_open = false; |
---|
2561 | } |
---|
2562 | } while($string_open); |
---|
2563 | |
---|
2564 | if ($check_linenumbers) { |
---|
2565 | // Are line numbers used? If, we should end the string before |
---|
2566 | // the newline and begin it again (so when <li>s are put in the source |
---|
2567 | // remains XHTML compliant) |
---|
2568 | // note to self: This opens up possibility of config files specifying |
---|
2569 | // that languages can/cannot have multiline strings??? |
---|
2570 | $string = str_replace("\n", "</span>\n<span$string_attributes>", $string); |
---|
2571 | } |
---|
2572 | |
---|
2573 | $result .= $string; |
---|
2574 | $string = ''; |
---|
2575 | $i = $start - 1; |
---|
2576 | continue; |
---|
2577 | } else if ($this->lexic_permissions['STRINGS'] && $hq && $hq[0] == $char && |
---|
2578 | substr($part, $i, $hq_strlen) == $hq) { |
---|
2579 | // The start of a hard quoted string |
---|
2580 | if (!$this->use_classes) { |
---|
2581 | $string_attributes = ' style="' . $this->language_data['STYLES']['STRINGS']['HARD'] . '"'; |
---|
2582 | $escape_char_attributes = ' style="' . $this->language_data['STYLES']['ESCAPE_CHAR']['HARD'] . '"'; |
---|
2583 | } else { |
---|
2584 | $string_attributes = ' class="st_h"'; |
---|
2585 | $escape_char_attributes = ' class="es_h"'; |
---|
2586 | } |
---|
2587 | // parse the stuff before this |
---|
2588 | $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2589 | $stuff_to_parse = ''; |
---|
2590 | |
---|
2591 | // now handle the string |
---|
2592 | $string = ''; |
---|
2593 | |
---|
2594 | // look for closing quote |
---|
2595 | $start = $i + $hq_strlen; |
---|
2596 | while ($close_pos = strpos($part, $this->language_data['HARDQUOTE'][1], $start)) { |
---|
2597 | $start = $close_pos + 1; |
---|
2598 | if ($this->lexic_permissions['ESCAPE_CHAR'] && $part[$close_pos - 1] == $this->language_data['ESCAPE_CHAR']) { |
---|
2599 | // make sure this quote is not escaped |
---|
2600 | foreach ($this->language_data['HARDESCAPE'] as $hardescape) { |
---|
2601 | if (substr($part, $close_pos - 1, strlen($hardescape)) == $hardescape) { |
---|
2602 | // check wether this quote is escaped or if it is something like '\\' |
---|
2603 | $escape_char_pos = $close_pos - 1; |
---|
2604 | while ($escape_char_pos > 0 |
---|
2605 | && $part[$escape_char_pos - 1] == $this->language_data['ESCAPE_CHAR']) { |
---|
2606 | --$escape_char_pos; |
---|
2607 | } |
---|
2608 | if (($close_pos - $escape_char_pos) & 1) { |
---|
2609 | // uneven number of escape chars => this quote is escaped |
---|
2610 | continue 2; |
---|
2611 | } |
---|
2612 | } |
---|
2613 | } |
---|
2614 | } |
---|
2615 | |
---|
2616 | // found closing quote |
---|
2617 | break; |
---|
2618 | } |
---|
2619 | |
---|
2620 | //Found the closing delimiter? |
---|
2621 | if (!$close_pos) { |
---|
2622 | // span till the end of this $part when no closing delimiter is found |
---|
2623 | $close_pos = $length; |
---|
2624 | } |
---|
2625 | |
---|
2626 | //Get the actual string |
---|
2627 | $string = substr($part, $i, $close_pos - $i + 1); |
---|
2628 | $i = $close_pos; |
---|
2629 | |
---|
2630 | // handle escape chars and encode html chars |
---|
2631 | // (special because when we have escape chars within our string they may not be escaped) |
---|
2632 | if ($this->lexic_permissions['ESCAPE_CHAR'] && $this->language_data['ESCAPE_CHAR']) { |
---|
2633 | $start = 0; |
---|
2634 | $new_string = ''; |
---|
2635 | while ($es_pos = strpos($string, $this->language_data['ESCAPE_CHAR'], $start)) { |
---|
2636 | // hmtl escape stuff before |
---|
2637 | $new_string .= $this->hsc(substr($string, $start, $es_pos - $start)); |
---|
2638 | // check if this is a hard escape |
---|
2639 | foreach ($this->language_data['HARDESCAPE'] as $hardescape) { |
---|
2640 | if (substr($string, $es_pos, strlen($hardescape)) == $hardescape) { |
---|
2641 | // indeed, this is a hardescape |
---|
2642 | $new_string .= "<span$escape_char_attributes>" . |
---|
2643 | $this->hsc($hardescape) . '</span>'; |
---|
2644 | $start = $es_pos + strlen($hardescape); |
---|
2645 | continue 2; |
---|
2646 | } |
---|
2647 | } |
---|
2648 | // not a hard escape, but a normal escape |
---|
2649 | // they come in pairs of two |
---|
2650 | $c = 0; |
---|
2651 | while (isset($string[$es_pos + $c]) && isset($string[$es_pos + $c + 1]) |
---|
2652 | && $string[$es_pos + $c] == $this->language_data['ESCAPE_CHAR'] |
---|
2653 | && $string[$es_pos + $c + 1] == $this->language_data['ESCAPE_CHAR']) { |
---|
2654 | $c += 2; |
---|
2655 | } |
---|
2656 | if ($c) { |
---|
2657 | $new_string .= "<span$escape_char_attributes>" . |
---|
2658 | str_repeat($escaped_escape_char, $c) . |
---|
2659 | '</span>'; |
---|
2660 | $start = $es_pos + $c; |
---|
2661 | } else { |
---|
2662 | // this is just a single lonely escape char... |
---|
2663 | $new_string .= $escaped_escape_char; |
---|
2664 | $start = $es_pos + 1; |
---|
2665 | } |
---|
2666 | } |
---|
2667 | $string = $new_string . $this->hsc(substr($string, $start)); |
---|
2668 | } else { |
---|
2669 | $string = $this->hsc($string); |
---|
2670 | } |
---|
2671 | |
---|
2672 | if ($check_linenumbers) { |
---|
2673 | // Are line numbers used? If, we should end the string before |
---|
2674 | // the newline and begin it again (so when <li>s are put in the source |
---|
2675 | // remains XHTML compliant) |
---|
2676 | // note to self: This opens up possibility of config files specifying |
---|
2677 | // that languages can/cannot have multiline strings??? |
---|
2678 | $string = str_replace("\n", "</span>\n<span$string_attributes>", $string); |
---|
2679 | } |
---|
2680 | |
---|
2681 | $result .= "<span$string_attributes>" . $string . '</span>'; |
---|
2682 | $string = ''; |
---|
2683 | continue; |
---|
2684 | } else { |
---|
2685 | //Have a look for regexp comments |
---|
2686 | if ($i == $next_comment_regexp_pos) { |
---|
2687 | $COMMENT_MATCHED = true; |
---|
2688 | $comment = $comment_regexp_cache_per_key[$next_comment_regexp_key]; |
---|
2689 | $test_str = $this->hsc(substr($part, $i, $comment['length'])); |
---|
2690 | |
---|
2691 | //@todo If remove important do remove here |
---|
2692 | if ($this->lexic_permissions['COMMENTS']['MULTI']) { |
---|
2693 | if (!$this->use_classes) { |
---|
2694 | $attributes = ' style="' . $this->language_data['STYLES']['COMMENTS'][$comment['key']] . '"'; |
---|
2695 | } else { |
---|
2696 | $attributes = ' class="co' . $comment['key'] . '"'; |
---|
2697 | } |
---|
2698 | |
---|
2699 | $test_str = "<span$attributes>" . $test_str . "</span>"; |
---|
2700 | |
---|
2701 | // Short-cut through all the multiline code |
---|
2702 | if ($check_linenumbers) { |
---|
2703 | // strreplace to put close span and open span around multiline newlines |
---|
2704 | $test_str = str_replace( |
---|
2705 | "\n", "</span>\n<span$attributes>", |
---|
2706 | str_replace("\n ", "\n ", $test_str) |
---|
2707 | ); |
---|
2708 | } |
---|
2709 | } |
---|
2710 | |
---|
2711 | $i += $comment['length'] - 1; |
---|
2712 | |
---|
2713 | // parse the rest |
---|
2714 | $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2715 | $stuff_to_parse = ''; |
---|
2716 | } |
---|
2717 | |
---|
2718 | // If we haven't matched a regexp comment, try multi-line comments |
---|
2719 | if (!$COMMENT_MATCHED) { |
---|
2720 | // Is this a multiline comment? |
---|
2721 | if (!empty($this->language_data['COMMENT_MULTI']) && $next_comment_multi_pos < $i) { |
---|
2722 | $next_comment_multi_pos = $length; |
---|
2723 | foreach ($this->language_data['COMMENT_MULTI'] as $open => $close) { |
---|
2724 | $match_i = false; |
---|
2725 | if (isset($comment_multi_cache_per_key[$open]) && |
---|
2726 | ($comment_multi_cache_per_key[$open] >= $i || |
---|
2727 | $comment_multi_cache_per_key[$open] === false)) { |
---|
2728 | // we have already matched something |
---|
2729 | if ($comment_multi_cache_per_key[$open] === false) { |
---|
2730 | // this comment is never matched |
---|
2731 | continue; |
---|
2732 | } |
---|
2733 | $match_i = $comment_multi_cache_per_key[$open]; |
---|
2734 | } else if (($match_i = stripos($part, $open, $i)) !== false) { |
---|
2735 | $comment_multi_cache_per_key[$open] = $match_i; |
---|
2736 | } else { |
---|
2737 | $comment_multi_cache_per_key[$open] = false; |
---|
2738 | continue; |
---|
2739 | } |
---|
2740 | if ($match_i !== false && $match_i < $next_comment_multi_pos) { |
---|
2741 | $next_comment_multi_pos = $match_i; |
---|
2742 | $next_open_comment_multi = $open; |
---|
2743 | if ($match_i === $i) { |
---|
2744 | break; |
---|
2745 | } |
---|
2746 | } |
---|
2747 | } |
---|
2748 | } |
---|
2749 | if ($i == $next_comment_multi_pos) { |
---|
2750 | $open = $next_open_comment_multi; |
---|
2751 | $close = $this->language_data['COMMENT_MULTI'][$open]; |
---|
2752 | $open_strlen = strlen($open); |
---|
2753 | $close_strlen = strlen($close); |
---|
2754 | $COMMENT_MATCHED = true; |
---|
2755 | $test_str_match = $open; |
---|
2756 | //@todo If remove important do remove here |
---|
2757 | if ($this->lexic_permissions['COMMENTS']['MULTI'] || |
---|
2758 | $open == GESHI_START_IMPORTANT) { |
---|
2759 | if ($open != GESHI_START_IMPORTANT) { |
---|
2760 | if (!$this->use_classes) { |
---|
2761 | $attributes = ' style="' . $this->language_data['STYLES']['COMMENTS']['MULTI'] . '"'; |
---|
2762 | } else { |
---|
2763 | $attributes = ' class="coMULTI"'; |
---|
2764 | } |
---|
2765 | $test_str = "<span$attributes>" . $this->hsc($open); |
---|
2766 | } else { |
---|
2767 | if (!$this->use_classes) { |
---|
2768 | $attributes = ' style="' . $this->important_styles . '"'; |
---|
2769 | } else { |
---|
2770 | $attributes = ' class="imp"'; |
---|
2771 | } |
---|
2772 | |
---|
2773 | // We don't include the start of the comment if it's an |
---|
2774 | // "important" part |
---|
2775 | $test_str = "<span$attributes>"; |
---|
2776 | } |
---|
2777 | } else { |
---|
2778 | $test_str = $this->hsc($open); |
---|
2779 | } |
---|
2780 | |
---|
2781 | $close_pos = strpos( $part, $close, $i + $open_strlen ); |
---|
2782 | |
---|
2783 | if ($close_pos === false) { |
---|
2784 | $close_pos = $length; |
---|
2785 | } |
---|
2786 | |
---|
2787 | // Short-cut through all the multiline code |
---|
2788 | $rest_of_comment = $this->hsc(substr($part, $i + $open_strlen, $close_pos - $i - $open_strlen + $close_strlen)); |
---|
2789 | if (($this->lexic_permissions['COMMENTS']['MULTI'] || |
---|
2790 | $test_str_match == GESHI_START_IMPORTANT) && |
---|
2791 | $check_linenumbers) { |
---|
2792 | |
---|
2793 | // strreplace to put close span and open span around multiline newlines |
---|
2794 | $test_str .= str_replace( |
---|
2795 | "\n", "</span>\n<span$attributes>", |
---|
2796 | str_replace("\n ", "\n ", $rest_of_comment) |
---|
2797 | ); |
---|
2798 | } else { |
---|
2799 | $test_str .= $rest_of_comment; |
---|
2800 | } |
---|
2801 | |
---|
2802 | if ($this->lexic_permissions['COMMENTS']['MULTI'] || |
---|
2803 | $test_str_match == GESHI_START_IMPORTANT) { |
---|
2804 | $test_str .= '</span>'; |
---|
2805 | } |
---|
2806 | |
---|
2807 | $i = $close_pos + $close_strlen - 1; |
---|
2808 | |
---|
2809 | // parse the rest |
---|
2810 | $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2811 | $stuff_to_parse = ''; |
---|
2812 | } |
---|
2813 | } |
---|
2814 | |
---|
2815 | // If we haven't matched a multiline comment, try single-line comments |
---|
2816 | if (!$COMMENT_MATCHED) { |
---|
2817 | // cache potential single line comment occurances |
---|
2818 | if (!empty($this->language_data['COMMENT_SINGLE']) && $next_comment_single_pos < $i) { |
---|
2819 | $next_comment_single_pos = $length; |
---|
2820 | foreach ($this->language_data['COMMENT_SINGLE'] as $comment_key => $comment_mark) { |
---|
2821 | $match_i = false; |
---|
2822 | if (isset($comment_single_cache_per_key[$comment_key]) && |
---|
2823 | ($comment_single_cache_per_key[$comment_key] >= $i || |
---|
2824 | $comment_single_cache_per_key[$comment_key] === false)) { |
---|
2825 | // we have already matched something |
---|
2826 | if ($comment_single_cache_per_key[$comment_key] === false) { |
---|
2827 | // this comment is never matched |
---|
2828 | continue; |
---|
2829 | } |
---|
2830 | $match_i = $comment_single_cache_per_key[$comment_key]; |
---|
2831 | } else if ( |
---|
2832 | // case sensitive comments |
---|
2833 | ($this->language_data['CASE_SENSITIVE'][GESHI_COMMENTS] && |
---|
2834 | ($match_i = stripos($part, $comment_mark, $i)) !== false) || |
---|
2835 | // non case sensitive |
---|
2836 | (!$this->language_data['CASE_SENSITIVE'][GESHI_COMMENTS] && |
---|
2837 | (($match_i = strpos($part, $comment_mark, $i)) !== false))) { |
---|
2838 | $comment_single_cache_per_key[$comment_key] = $match_i; |
---|
2839 | } else { |
---|
2840 | $comment_single_cache_per_key[$comment_key] = false; |
---|
2841 | continue; |
---|
2842 | } |
---|
2843 | if ($match_i !== false && $match_i < $next_comment_single_pos) { |
---|
2844 | $next_comment_single_pos = $match_i; |
---|
2845 | $next_comment_single_key = $comment_key; |
---|
2846 | if ($match_i === $i) { |
---|
2847 | break; |
---|
2848 | } |
---|
2849 | } |
---|
2850 | } |
---|
2851 | } |
---|
2852 | if ($next_comment_single_pos == $i) { |
---|
2853 | $comment_key = $next_comment_single_key; |
---|
2854 | $comment_mark = $this->language_data['COMMENT_SINGLE'][$comment_key]; |
---|
2855 | $com_len = strlen($comment_mark); |
---|
2856 | |
---|
2857 | // This check will find special variables like $# in bash |
---|
2858 | // or compiler directives of Delphi beginning {$ |
---|
2859 | if ((empty($sc_disallowed_before) || ($i == 0) || |
---|
2860 | (false === strpos($sc_disallowed_before, $part[$i-1]))) && |
---|
2861 | (empty($sc_disallowed_after) || ($length <= $i + $com_len) || |
---|
2862 | (false === strpos($sc_disallowed_after, $part[$i + $com_len])))) |
---|
2863 | { |
---|
2864 | // this is a valid comment |
---|
2865 | $COMMENT_MATCHED = true; |
---|
2866 | if ($this->lexic_permissions['COMMENTS'][$comment_key]) { |
---|
2867 | if (!$this->use_classes) { |
---|
2868 | $attributes = ' style="' . $this->language_data['STYLES']['COMMENTS'][$comment_key] . '"'; |
---|
2869 | } else { |
---|
2870 | $attributes = ' class="co' . $comment_key . '"'; |
---|
2871 | } |
---|
2872 | $test_str = "<span$attributes>" . $this->hsc($this->change_case($comment_mark)); |
---|
2873 | } else { |
---|
2874 | $test_str = $this->hsc($comment_mark); |
---|
2875 | } |
---|
2876 | |
---|
2877 | //Check if this comment is the last in the source |
---|
2878 | $close_pos = strpos($part, "\n", $i); |
---|
2879 | $oops = false; |
---|
2880 | if ($close_pos === false) { |
---|
2881 | $close_pos = $length; |
---|
2882 | $oops = true; |
---|
2883 | } |
---|
2884 | $test_str .= $this->hsc(substr($part, $i + $com_len, $close_pos - $i - $com_len)); |
---|
2885 | if ($this->lexic_permissions['COMMENTS'][$comment_key]) { |
---|
2886 | $test_str .= "</span>"; |
---|
2887 | } |
---|
2888 | |
---|
2889 | // Take into account that the comment might be the last in the source |
---|
2890 | if (!$oops) { |
---|
2891 | $test_str .= "\n"; |
---|
2892 | } |
---|
2893 | |
---|
2894 | $i = $close_pos; |
---|
2895 | |
---|
2896 | // parse the rest |
---|
2897 | $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2898 | $stuff_to_parse = ''; |
---|
2899 | } |
---|
2900 | } |
---|
2901 | } |
---|
2902 | } |
---|
2903 | |
---|
2904 | // Where are we adding this char? |
---|
2905 | if (!$COMMENT_MATCHED) { |
---|
2906 | $stuff_to_parse .= $char; |
---|
2907 | } else { |
---|
2908 | $result .= $test_str; |
---|
2909 | unset($test_str); |
---|
2910 | $COMMENT_MATCHED = false; |
---|
2911 | } |
---|
2912 | } |
---|
2913 | // Parse the last bit |
---|
2914 | $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2915 | $stuff_to_parse = ''; |
---|
2916 | } else { |
---|
2917 | $result .= $this->hsc($part); |
---|
2918 | } |
---|
2919 | // Close the <span> that surrounds the block |
---|
2920 | if ($STRICTATTRS != '') { |
---|
2921 | $result = str_replace("\n", "</span>\n<span$STRICTATTRS>", $result); |
---|
2922 | $result .= '</span>'; |
---|
2923 | } |
---|
2924 | |
---|
2925 | $endresult .= $result; |
---|
2926 | unset($part, $parts[$key], $result); |
---|
2927 | } |
---|
2928 | |
---|
2929 | //This fix is related to SF#1923020, but has to be applied regardless of |
---|
2930 | //actually highlighting symbols. |
---|
2931 | /** NOTE: memorypeak #3 */ |
---|
2932 | $endresult = str_replace(array('<SEMI>', '<PIPE>'), array(';', '|'), $endresult); |
---|
2933 | |
---|
2934 | // // Parse the last stuff (redundant?) |
---|
2935 | // $result .= $this->parse_non_string_part($stuff_to_parse); |
---|
2936 | |
---|
2937 | // Lop off the very first and last spaces |
---|
2938 | // $result = substr($result, 1, -1); |
---|
2939 | |
---|
2940 | // We're finished: stop timing |
---|
2941 | $this->set_time($start_time, microtime()); |
---|
2942 | |
---|
2943 | $this->finalise($endresult); |
---|
2944 | return $endresult; |
---|
2945 | } |
---|
2946 | |
---|
2947 | /** |
---|
2948 | * Swaps out spaces and tabs for HTML indentation. Not needed if |
---|
2949 | * the code is in a pre block... |
---|
2950 | * |
---|
2951 | * @param string The source to indent (reference!) |
---|
2952 | * @since 1.0.0 |
---|
2953 | * @access private |
---|
2954 | */ |
---|
2955 | function indent(&$result) { |
---|
2956 | /// Replace tabs with the correct number of spaces |
---|
2957 | if (false !== strpos($result, "\t")) { |
---|
2958 | $lines = explode("\n", $result); |
---|
2959 | $result = null;//Save memory while we process the lines individually |
---|
2960 | $tab_width = $this->get_real_tab_width(); |
---|
2961 | $tab_string = ' ' . str_repeat(' ', $tab_width); |
---|
2962 | |
---|
2963 | for ($key = 0, $n = count($lines); $key < $n; $key++) { |
---|
2964 | $line = $lines[$key]; |
---|
2965 | if (false === strpos($line, "\t")) { |
---|
2966 | continue; |
---|
2967 | } |
---|
2968 | |
---|
2969 | $pos = 0; |
---|
2970 | $length = strlen($line); |
---|
2971 | $lines[$key] = ''; // reduce memory |
---|
2972 | |
---|
2973 | $IN_TAG = false; |
---|
2974 | for ($i = 0; $i < $length; ++$i) { |
---|
2975 | $char = $line[$i]; |
---|
2976 | // Simple engine to work out whether we're in a tag. |
---|
2977 | // If we are we modify $pos. This is so we ignore HTML |
---|
2978 | // in the line and only workout the tab replacement |
---|
2979 | // via the actual content of the string |
---|
2980 | // This test could be improved to include strings in the |
---|
2981 | // html so that < or > would be allowed in user's styles |
---|
2982 | // (e.g. quotes: '<' '>'; or similar) |
---|
2983 | if ($IN_TAG) { |
---|
2984 | if ('>' == $char) { |
---|
2985 | $IN_TAG = false; |
---|
2986 | } |
---|
2987 | $lines[$key] .= $char; |
---|
2988 | } else if ('<' == $char) { |
---|
2989 | $IN_TAG = true; |
---|
2990 | $lines[$key] .= '<'; |
---|
2991 | } else if ('&' == $char) { |
---|
2992 | $substr = substr($line, $i + 3, 5); |
---|
2993 | $posi = strpos($substr, ';'); |
---|
2994 | if (false === $posi) { |
---|
2995 | ++$pos; |
---|
2996 | } else { |
---|
2997 | $pos -= $posi+2; |
---|
2998 | } |
---|
2999 | $lines[$key] .= $char; |
---|
3000 | } else if ("\t" == $char) { |
---|
3001 | $str = ''; |
---|
3002 | // OPTIMISE - move $strs out. Make an array: |
---|
3003 | // $tabs = array( |
---|
3004 | // 1 => ' ', |
---|
3005 | // 2 => ' ', |
---|
3006 | // 3 => ' ' etc etc |
---|
3007 | // to use instead of building a string every time |
---|
3008 | $tab_end_width = $tab_width - ($pos % $tab_width); //Moved out of the look as it doesn't change within the loop |
---|
3009 | if (($pos & 1) || 1 == $tab_end_width) { |
---|
3010 | $str .= substr($tab_string, 6, $tab_end_width); |
---|
3011 | } else { |
---|
3012 | $str .= substr($tab_string, 0, $tab_end_width+5); |
---|
3013 | } |
---|
3014 | $lines[$key] .= $str; |
---|
3015 | $pos += $tab_end_width; |
---|
3016 | |
---|
3017 | if (false === strpos($line, "\t", $i + 1)) { |
---|
3018 | $lines[$key] .= substr($line, $i + 1); |
---|
3019 | break; |
---|
3020 | } |
---|
3021 | } else if (0 == $pos && ' ' == $char) { |
---|
3022 | $lines[$key] .= ' '; |
---|
3023 | ++$pos; |
---|
3024 | } else { |
---|
3025 | $lines[$key] .= $char; |
---|
3026 | ++$pos; |
---|
3027 | } |
---|
3028 | } |
---|
3029 | } |
---|
3030 | $result = implode("\n", $lines); |
---|
3031 | unset($lines);//We don't need the lines separated beyond this --- free them! |
---|
3032 | } |
---|
3033 | // Other whitespace |
---|
3034 | // BenBE: Fix to reduce the number of replacements to be done |
---|
3035 | $result = preg_replace('/^ /m', ' ', $result); |
---|
3036 | $result = str_replace(' ', ' ', $result); |
---|
3037 | |
---|
3038 | if ($this->line_numbers == GESHI_NO_LINE_NUMBERS) { |
---|
3039 | if ($this->line_ending === null) { |
---|
3040 | $result = nl2br($result); |
---|
3041 | } else { |
---|
3042 | $result = str_replace("\n", $this->line_ending, $result); |
---|
3043 | } |
---|
3044 | } |
---|
3045 | } |
---|
3046 | |
---|
3047 | /** |
---|
3048 | * Changes the case of a keyword for those languages where a change is asked for |
---|
3049 | * |
---|
3050 | * @param string The keyword to change the case of |
---|
3051 | * @return string The keyword with its case changed |
---|
3052 | * @since 1.0.0 |
---|
3053 | * @access private |
---|
3054 | */ |
---|
3055 | function change_case($instr) { |
---|
3056 | switch ($this->language_data['CASE_KEYWORDS']) { |
---|
3057 | case GESHI_CAPS_UPPER: |
---|
3058 | return strtoupper($instr); |
---|
3059 | case GESHI_CAPS_LOWER: |
---|
3060 | return strtolower($instr); |
---|
3061 | default: |
---|
3062 | return $instr; |
---|
3063 | } |
---|
3064 | } |
---|
3065 | |
---|
3066 | /** |
---|
3067 | * Handles replacements of keywords to include markup and links if requested |
---|
3068 | * |
---|
3069 | * @param string The keyword to add the Markup to |
---|
3070 | * @return The HTML for the match found |
---|
3071 | * @since 1.0.8 |
---|
3072 | * @access private |
---|
3073 | * |
---|
3074 | * @todo Get rid of ender in keyword links |
---|
3075 | */ |
---|
3076 | function handle_keyword_replace($match) { |
---|
3077 | $k = $this->_kw_replace_group; |
---|
3078 | $keyword = $match[0]; |
---|
3079 | |
---|
3080 | $before = ''; |
---|
3081 | $after = ''; |
---|
3082 | |
---|
3083 | if ($this->keyword_links) { |
---|
3084 | // Keyword links have been ebabled |
---|
3085 | |
---|
3086 | if (isset($this->language_data['URLS'][$k]) && |
---|
3087 | $this->language_data['URLS'][$k] != '') { |
---|
3088 | // There is a base group for this keyword |
---|
3089 | |
---|
3090 | // Old system: strtolower |
---|
3091 | //$keyword = ( $this->language_data['CASE_SENSITIVE'][$group] ) ? $keyword : strtolower($keyword); |
---|
3092 | // New system: get keyword from language file to get correct case |
---|
3093 | if (!$this->language_data['CASE_SENSITIVE'][$k] && |
---|
3094 | strpos($this->language_data['URLS'][$k], '{FNAME}') !== false) { |
---|
3095 | foreach ($this->language_data['KEYWORDS'][$k] as $word) { |
---|
3096 | if (strcasecmp($word, $keyword) == 0) { |
---|
3097 | break; |
---|
3098 | } |
---|
3099 | } |
---|
3100 | } else { |
---|
3101 | $word = $keyword; |
---|
3102 | } |
---|
3103 | |
---|
3104 | $before = '<|UR1|"' . |
---|
3105 | str_replace( |
---|
3106 | array( |
---|
3107 | '{FNAME}', |
---|
3108 | '{FNAMEL}', |
---|
3109 | '{FNAMEU}', |
---|
3110 | '.'), |
---|
3111 | array( |
---|
3112 | str_replace('+', '%20', urlencode($this->hsc($word))), |
---|
3113 | str_replace('+', '%20', urlencode($this->hsc(strtolower($word)))), |
---|
3114 | str_replace('+', '%20', urlencode($this->hsc(strtoupper($word)))), |
---|
3115 | '<DOT>'), |
---|
3116 | $this->language_data['URLS'][$k] |
---|
3117 | ) . '">'; |
---|
3118 | $after = '</a>'; |
---|
3119 | } |
---|
3120 | } |
---|
3121 | |
---|
3122 | return $before . '<|/'. $k .'/>' . $this->change_case($keyword) . '|>' . $after; |
---|
3123 | } |
---|
3124 | |
---|
3125 | /** |
---|
3126 | * handles regular expressions highlighting-definitions with callback functions |
---|
3127 | * |
---|
3128 | * @note this is a callback, don't use it directly |
---|
3129 | * |
---|
3130 | * @param array the matches array |
---|
3131 | * @return The highlighted string |
---|
3132 | * @since 1.0.8 |
---|
3133 | * @access private |
---|
3134 | */ |
---|
3135 | function handle_regexps_callback($matches) { |
---|
3136 | // before: "' style=\"' . call_user_func(\"$func\", '\\1') . '\"\\1|>'", |
---|
3137 | return ' style="' . call_user_func($this->language_data['STYLES']['REGEXPS'][$this->_rx_key], $matches[1]) . '"'. $matches[1] . '|>'; |
---|
3138 | } |
---|
3139 | |
---|
3140 | /** |
---|
3141 | * handles newlines in REGEXPS matches. Set the _hmr_* vars before calling this |
---|
3142 | * |
---|
3143 | * @note this is a callback, don't use it directly |
---|
3144 | * |
---|
3145 | * @param array the matches array |
---|
3146 | * @return string |
---|
3147 | * @since 1.0.8 |
---|
3148 | * @access private |
---|
3149 | */ |
---|
3150 | function handle_multiline_regexps($matches) { |
---|
3151 | $before = $this->_hmr_before; |
---|
3152 | $after = $this->_hmr_after; |
---|
3153 | if ($this->_hmr_replace) { |
---|
3154 | $replace = $this->_hmr_replace; |
---|
3155 | $search = array(); |
---|
3156 | |
---|
3157 | foreach (array_keys($matches) as $k) { |
---|
3158 | $search[] = '\\' . $k; |
---|
3159 | } |
---|
3160 | |
---|
3161 | $before = str_replace($search, $matches, $before); |
---|
3162 | $after = str_replace($search, $matches, $after); |
---|
3163 | $replace = str_replace($search, $matches, $replace); |
---|
3164 | } else { |
---|
3165 | $replace = $matches[0]; |
---|
3166 | } |
---|
3167 | return $before |
---|
3168 | . '<|!REG3XP' . $this->_hmr_key .'!>' |
---|
3169 | . str_replace("\n", "|>\n<|!REG3XP" . $this->_hmr_key . '!>', $replace) |
---|
3170 | . '|>' |
---|
3171 | . $after; |
---|
3172 | } |
---|
3173 | |
---|
3174 | /** |
---|
3175 | * Takes a string that has no strings or comments in it, and highlights |
---|
3176 | * stuff like keywords, numbers and methods. |
---|
3177 | * |
---|
3178 | * @param string The string to parse for keyword, numbers etc. |
---|
3179 | * @since 1.0.0 |
---|
3180 | * @access private |
---|
3181 | * @todo BUGGY! Why? Why not build string and return? |
---|
3182 | */ |
---|
3183 | function parse_non_string_part($stuff_to_parse) { |
---|
3184 | $stuff_to_parse = ' ' . $this->hsc($stuff_to_parse); |
---|
3185 | |
---|
3186 | // Regular expressions |
---|
3187 | foreach ($this->language_data['REGEXPS'] as $key => $regexp) { |
---|
3188 | if ($this->lexic_permissions['REGEXPS'][$key]) { |
---|
3189 | if (is_array($regexp)) { |
---|
3190 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3191 | // produce valid HTML when we match multiple lines |
---|
3192 | $this->_hmr_replace = $regexp[GESHI_REPLACE]; |
---|
3193 | $this->_hmr_before = $regexp[GESHI_BEFORE]; |
---|
3194 | $this->_hmr_key = $key; |
---|
3195 | $this->_hmr_after = $regexp[GESHI_AFTER]; |
---|
3196 | $stuff_to_parse = preg_replace_callback( |
---|
3197 | "/" . $regexp[GESHI_SEARCH] . "/{$regexp[GESHI_MODIFIERS]}", |
---|
3198 | array($this, 'handle_multiline_regexps'), |
---|
3199 | $stuff_to_parse); |
---|
3200 | $this->_hmr_replace = false; |
---|
3201 | $this->_hmr_before = ''; |
---|
3202 | $this->_hmr_after = ''; |
---|
3203 | } else { |
---|
3204 | $stuff_to_parse = preg_replace( |
---|
3205 | '/' . $regexp[GESHI_SEARCH] . '/' . $regexp[GESHI_MODIFIERS], |
---|
3206 | $regexp[GESHI_BEFORE] . '<|!REG3XP'. $key .'!>' . $regexp[GESHI_REPLACE] . '|>' . $regexp[GESHI_AFTER], |
---|
3207 | $stuff_to_parse); |
---|
3208 | } |
---|
3209 | } else { |
---|
3210 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3211 | // produce valid HTML when we match multiple lines |
---|
3212 | $this->_hmr_key = $key; |
---|
3213 | $stuff_to_parse = preg_replace_callback( "/(" . $regexp . ")/", |
---|
3214 | array($this, 'handle_multiline_regexps'), $stuff_to_parse); |
---|
3215 | $this->_hmr_key = ''; |
---|
3216 | } else { |
---|
3217 | $stuff_to_parse = preg_replace( "/(" . $regexp . ")/", "<|!REG3XP$key!>\\1|>", $stuff_to_parse); |
---|
3218 | } |
---|
3219 | } |
---|
3220 | } |
---|
3221 | } |
---|
3222 | |
---|
3223 | // Highlight numbers. As of 1.0.8 we support diffent types of numbers |
---|
3224 | $numbers_found = false; |
---|
3225 | if ($this->lexic_permissions['NUMBERS'] && preg_match('#\d#', $stuff_to_parse )) { |
---|
3226 | $numbers_found = true; |
---|
3227 | |
---|
3228 | //For each of the formats ... |
---|
3229 | foreach($this->language_data['NUMBERS_RXCACHE'] as $id => $regexp) { |
---|
3230 | //Check if it should be highlighted ... |
---|
3231 | $stuff_to_parse = preg_replace($regexp, "<|/NUM!$id/>\\1|>", $stuff_to_parse); |
---|
3232 | } |
---|
3233 | } |
---|
3234 | |
---|
3235 | // Highlight keywords |
---|
3236 | $disallowed_before = "(?<![a-zA-Z0-9\$_\|\#;>|^&"; |
---|
3237 | $disallowed_after = "(?![a-zA-Z0-9_\|%\\-&;"; |
---|
3238 | if ($this->lexic_permissions['STRINGS']) { |
---|
3239 | $quotemarks = preg_quote(implode($this->language_data['QUOTEMARKS']), '/'); |
---|
3240 | $disallowed_before .= $quotemarks; |
---|
3241 | $disallowed_after .= $quotemarks; |
---|
3242 | } |
---|
3243 | $disallowed_before .= "])"; |
---|
3244 | $disallowed_after .= "])"; |
---|
3245 | |
---|
3246 | $parser_control_pergroup = false; |
---|
3247 | if (isset($this->language_data['PARSER_CONTROL'])) { |
---|
3248 | if (isset($this->language_data['PARSER_CONTROL']['KEYWORDS'])) { |
---|
3249 | $x = 0; // check wether per-keyword-group parser_control is enabled |
---|
3250 | if (isset($this->language_data['PARSER_CONTROL']['KEYWORDS']['DISALLOWED_BEFORE'])) { |
---|
3251 | $disallowed_before = $this->language_data['PARSER_CONTROL']['KEYWORDS']['DISALLOWED_BEFORE']; |
---|
3252 | ++$x; |
---|
3253 | } |
---|
3254 | if (isset($this->language_data['PARSER_CONTROL']['KEYWORDS']['DISALLOWED_AFTER'])) { |
---|
3255 | $disallowed_after = $this->language_data['PARSER_CONTROL']['KEYWORDS']['DISALLOWED_AFTER']; |
---|
3256 | ++$x; |
---|
3257 | } |
---|
3258 | $parser_control_pergroup = (count($this->language_data['PARSER_CONTROL']['KEYWORDS']) - $x) > 0; |
---|
3259 | } |
---|
3260 | } |
---|
3261 | |
---|
3262 | // if this is changed, don't forget to change it below |
---|
3263 | // if (!empty($disallowed_before)) { |
---|
3264 | // $disallowed_before = "(?<![$disallowed_before])"; |
---|
3265 | // } |
---|
3266 | // if (!empty($disallowed_after)) { |
---|
3267 | // $disallowed_after = "(?![$disallowed_after])"; |
---|
3268 | // } |
---|
3269 | |
---|
3270 | foreach (array_keys($this->language_data['KEYWORDS']) as $k) { |
---|
3271 | if (!isset($this->lexic_permissions['KEYWORDS'][$k]) || |
---|
3272 | $this->lexic_permissions['KEYWORDS'][$k]) { |
---|
3273 | |
---|
3274 | $case_sensitive = $this->language_data['CASE_SENSITIVE'][$k]; |
---|
3275 | $modifiers = $case_sensitive ? '' : 'i'; |
---|
3276 | |
---|
3277 | // NEW in 1.0.8 - per-keyword-group parser control |
---|
3278 | $disallowed_before_local = $disallowed_before; |
---|
3279 | $disallowed_after_local = $disallowed_after; |
---|
3280 | if ($parser_control_pergroup && isset($this->language_data['PARSER_CONTROL']['KEYWORDS'][$k])) { |
---|
3281 | if (isset($this->language_data['PARSER_CONTROL']['KEYWORDS'][$k]['DISALLOWED_BEFORE'])) { |
---|
3282 | $disallowed_before_local = |
---|
3283 | $this->language_data['PARSER_CONTROL']['KEYWORDS'][$k]['DISALLOWED_BEFORE']; |
---|
3284 | } |
---|
3285 | |
---|
3286 | if (isset($this->language_data['PARSER_CONTROL']['KEYWORDS'][$k]['DISALLOWED_AFTER'])) { |
---|
3287 | $disallowed_after_local = |
---|
3288 | $this->language_data['PARSER_CONTROL']['KEYWORDS'][$k]['DISALLOWED_AFTER']; |
---|
3289 | } |
---|
3290 | } |
---|
3291 | |
---|
3292 | $this->_kw_replace_group = $k; |
---|
3293 | |
---|
3294 | //NEW in 1.0.8, the cached regexp list |
---|
3295 | // since we don't want PHP / PCRE to crash due to too large patterns we split them into smaller chunks |
---|
3296 | for ($set = 0, $set_length = count($this->language_data['CACHED_KEYWORD_LISTS'][$k]); $set < $set_length; ++$set) { |
---|
3297 | $keywordset =& $this->language_data['CACHED_KEYWORD_LISTS'][$k][$set]; |
---|
3298 | // Might make a more unique string for putting the number in soon |
---|
3299 | // Basically, we don't put the styles in yet because then the styles themselves will |
---|
3300 | // get highlighted if the language has a CSS keyword in it (like CSS, for example ;)) |
---|
3301 | $stuff_to_parse = preg_replace_callback( |
---|
3302 | "/$disallowed_before_local({$keywordset})(?!\<DOT\>(?:htm|php))$disallowed_after_local/$modifiers", |
---|
3303 | array($this, 'handle_keyword_replace'), |
---|
3304 | $stuff_to_parse |
---|
3305 | ); |
---|
3306 | } |
---|
3307 | } |
---|
3308 | } |
---|
3309 | |
---|
3310 | // |
---|
3311 | // Now that's all done, replace /[number]/ with the correct styles |
---|
3312 | // |
---|
3313 | foreach (array_keys($this->language_data['KEYWORDS']) as $k) { |
---|
3314 | if (!$this->use_classes) { |
---|
3315 | $attributes = ' style="' . |
---|
3316 | (isset($this->language_data['STYLES']['KEYWORDS'][$k]) ? |
---|
3317 | $this->language_data['STYLES']['KEYWORDS'][$k] : "") . '"'; |
---|
3318 | } else { |
---|
3319 | $attributes = ' class="kw' . $k . '"'; |
---|
3320 | } |
---|
3321 | $stuff_to_parse = str_replace("<|/$k/>", "<|$attributes>", $stuff_to_parse); |
---|
3322 | } |
---|
3323 | |
---|
3324 | if ($numbers_found) { |
---|
3325 | // Put number styles in |
---|
3326 | foreach($this->language_data['NUMBERS_RXCACHE'] as $id => $regexp) { |
---|
3327 | //Commented out for now, as this needs some review ... |
---|
3328 | // if ($numbers_permissions & $id) { |
---|
3329 | //Get the appropriate style ... |
---|
3330 | //Checking for unset styles is done by the style cache builder ... |
---|
3331 | if (!$this->use_classes) { |
---|
3332 | $attributes = ' style="' . $this->language_data['STYLES']['NUMBERS'][$id] . '"'; |
---|
3333 | } else { |
---|
3334 | $attributes = ' class="nu'.$id.'"'; |
---|
3335 | } |
---|
3336 | |
---|
3337 | //Set in the correct styles ... |
---|
3338 | $stuff_to_parse = str_replace("/NUM!$id/", $attributes, $stuff_to_parse); |
---|
3339 | // } |
---|
3340 | } |
---|
3341 | } |
---|
3342 | |
---|
3343 | // Highlight methods and fields in objects |
---|
3344 | if ($this->lexic_permissions['METHODS'] && $this->language_data['OOLANG']) { |
---|
3345 | $oolang_spaces = "[\s]*"; |
---|
3346 | $oolang_before = ""; |
---|
3347 | $oolang_after = "[a-zA-Z][a-zA-Z0-9_]*"; |
---|
3348 | if (isset($this->language_data['PARSER_CONTROL'])) { |
---|
3349 | if (isset($this->language_data['PARSER_CONTROL']['OOLANG'])) { |
---|
3350 | if (isset($this->language_data['PARSER_CONTROL']['OOLANG']['MATCH_BEFORE'])) { |
---|
3351 | $oolang_before = $this->language_data['PARSER_CONTROL']['OOLANG']['MATCH_BEFORE']; |
---|
3352 | } |
---|
3353 | if (isset($this->language_data['PARSER_CONTROL']['OOLANG']['MATCH_AFTER'])) { |
---|
3354 | $oolang_after = $this->language_data['PARSER_CONTROL']['OOLANG']['MATCH_AFTER']; |
---|
3355 | } |
---|
3356 | if (isset($this->language_data['PARSER_CONTROL']['OOLANG']['MATCH_SPACES'])) { |
---|
3357 | $oolang_spaces = $this->language_data['PARSER_CONTROL']['OOLANG']['MATCH_SPACES']; |
---|
3358 | } |
---|
3359 | } |
---|
3360 | } |
---|
3361 | |
---|
3362 | foreach ($this->language_data['OBJECT_SPLITTERS'] as $key => $splitter) { |
---|
3363 | if (false !== strpos($stuff_to_parse, $splitter)) { |
---|
3364 | if (!$this->use_classes) { |
---|
3365 | $attributes = ' style="' . $this->language_data['STYLES']['METHODS'][$key] . '"'; |
---|
3366 | } else { |
---|
3367 | $attributes = ' class="me' . $key . '"'; |
---|
3368 | } |
---|
3369 | $stuff_to_parse = preg_replace("/($oolang_before)(" . preg_quote($this->language_data['OBJECT_SPLITTERS'][$key], '/') . ")($oolang_spaces)($oolang_after)/", "\\1\\2\\3<|$attributes>\\4|>", $stuff_to_parse); |
---|
3370 | } |
---|
3371 | } |
---|
3372 | } |
---|
3373 | |
---|
3374 | // |
---|
3375 | // Highlight brackets. Yes, I've tried adding a semi-colon to this list. |
---|
3376 | // You try it, and see what happens ;) |
---|
3377 | // TODO: Fix lexic permissions not converting entities if shouldn't |
---|
3378 | // be highlighting regardless |
---|
3379 | // |
---|
3380 | if ($this->lexic_permissions['BRACKETS']) { |
---|
3381 | $stuff_to_parse = str_replace( $this->language_data['CACHE_BRACKET_MATCH'], |
---|
3382 | $this->language_data['CACHE_BRACKET_REPLACE'], $stuff_to_parse ); |
---|
3383 | } |
---|
3384 | |
---|
3385 | |
---|
3386 | //FIX for symbol highlighting ... |
---|
3387 | if ($this->lexic_permissions['SYMBOLS'] && !empty($this->language_data['SYMBOLS'])) { |
---|
3388 | //Get all matches and throw away those witin a block that is already highlighted... (i.e. matched by a regexp) |
---|
3389 | $n_symbols = preg_match_all("/<\|(?:<DOT>|[^>])+>(?:(?!\|>).*?)\|>|<\/a>|(?:" . $this->language_data['SYMBOL_SEARCH'] . ")+/", $stuff_to_parse, $pot_symbols, PREG_OFFSET_CAPTURE | PREG_SET_ORDER); |
---|
3390 | $global_offset = 0; |
---|
3391 | for ($s_id = 0; $s_id < $n_symbols; ++$s_id) { |
---|
3392 | $symbol_match = $pot_symbols[$s_id][0][0]; |
---|
3393 | if (strpos($symbol_match, '<') !== false || strpos($symbol_match, '>') !== false) { |
---|
3394 | // already highlighted blocks _must_ include either < or > |
---|
3395 | // so if this conditional applies, we have to skip this match |
---|
3396 | // BenBE: UNLESS the block contains <SEMI> or <PIPE> |
---|
3397 | if(strpos($symbol_match, '<SEMI>') === false && |
---|
3398 | strpos($symbol_match, '<PIPE>') === false) { |
---|
3399 | continue; |
---|
3400 | } |
---|
3401 | } |
---|
3402 | |
---|
3403 | // if we reach this point, we have a valid match which needs to be highlighted |
---|
3404 | |
---|
3405 | $symbol_length = strlen($symbol_match); |
---|
3406 | $symbol_offset = $pot_symbols[$s_id][0][1]; |
---|
3407 | unset($pot_symbols[$s_id]); |
---|
3408 | $symbol_end = $symbol_length + $symbol_offset; |
---|
3409 | $symbol_hl = ""; |
---|
3410 | |
---|
3411 | // if we have multiple styles, we have to handle them properly |
---|
3412 | if ($this->language_data['MULTIPLE_SYMBOL_GROUPS']) { |
---|
3413 | $old_sym = -1; |
---|
3414 | // Split the current stuff to replace into its atomic symbols ... |
---|
3415 | preg_match_all("/" . $this->language_data['SYMBOL_SEARCH'] . "/", $symbol_match, $sym_match_syms, PREG_PATTERN_ORDER); |
---|
3416 | foreach ($sym_match_syms[0] as $sym_ms) { |
---|
3417 | //Check if consequtive symbols belong to the same group to save output ... |
---|
3418 | if (isset($this->language_data['SYMBOL_DATA'][$sym_ms]) |
---|
3419 | && ($this->language_data['SYMBOL_DATA'][$sym_ms] != $old_sym)) { |
---|
3420 | if (-1 != $old_sym) { |
---|
3421 | $symbol_hl .= "|>"; |
---|
3422 | } |
---|
3423 | $old_sym = $this->language_data['SYMBOL_DATA'][$sym_ms]; |
---|
3424 | if (!$this->use_classes) { |
---|
3425 | $symbol_hl .= '<| style="' . $this->language_data['STYLES']['SYMBOLS'][$old_sym] . '">'; |
---|
3426 | } else { |
---|
3427 | $symbol_hl .= '<| class="sy' . $old_sym . '">'; |
---|
3428 | } |
---|
3429 | } |
---|
3430 | $symbol_hl .= $sym_ms; |
---|
3431 | } |
---|
3432 | unset($sym_match_syms); |
---|
3433 | |
---|
3434 | //Close remaining tags and insert the replacement at the right position ... |
---|
3435 | //Take caution if symbol_hl is empty to avoid doubled closing spans. |
---|
3436 | if (-1 != $old_sym) { |
---|
3437 | $symbol_hl .= "|>"; |
---|
3438 | } |
---|
3439 | } else { |
---|
3440 | if (!$this->use_classes) { |
---|
3441 | $symbol_hl = '<| style="' . $this->language_data['STYLES']['SYMBOLS'][0] . '">'; |
---|
3442 | } else { |
---|
3443 | $symbol_hl = '<| class="sy0">'; |
---|
3444 | } |
---|
3445 | $symbol_hl .= $symbol_match . '|>'; |
---|
3446 | } |
---|
3447 | |
---|
3448 | $stuff_to_parse = substr_replace($stuff_to_parse, $symbol_hl, $symbol_offset + $global_offset, $symbol_length); |
---|
3449 | |
---|
3450 | // since we replace old text with something of different size, |
---|
3451 | // we'll have to keep track of the differences |
---|
3452 | $global_offset += strlen($symbol_hl) - $symbol_length; |
---|
3453 | } |
---|
3454 | } |
---|
3455 | //FIX for symbol highlighting ... |
---|
3456 | |
---|
3457 | // Add class/style for regexps |
---|
3458 | foreach (array_keys($this->language_data['REGEXPS']) as $key) { |
---|
3459 | if ($this->lexic_permissions['REGEXPS'][$key]) { |
---|
3460 | if (is_callable($this->language_data['STYLES']['REGEXPS'][$key])) { |
---|
3461 | $this->_rx_key = $key; |
---|
3462 | $stuff_to_parse = preg_replace_callback("/!REG3XP$key!(.*)\|>/U", |
---|
3463 | array($this, 'handle_regexps_callback'), |
---|
3464 | $stuff_to_parse); |
---|
3465 | } else { |
---|
3466 | if (!$this->use_classes) { |
---|
3467 | $attributes = ' style="' . $this->language_data['STYLES']['REGEXPS'][$key] . '"'; |
---|
3468 | } else { |
---|
3469 | if (is_array($this->language_data['REGEXPS'][$key]) && |
---|
3470 | array_key_exists(GESHI_CLASS, $this->language_data['REGEXPS'][$key])) { |
---|
3471 | $attributes = ' class="' . |
---|
3472 | $this->language_data['REGEXPS'][$key][GESHI_CLASS] . '"'; |
---|
3473 | } else { |
---|
3474 | $attributes = ' class="re' . $key . '"'; |
---|
3475 | } |
---|
3476 | } |
---|
3477 | $stuff_to_parse = str_replace("!REG3XP$key!", "$attributes", $stuff_to_parse); |
---|
3478 | } |
---|
3479 | } |
---|
3480 | } |
---|
3481 | |
---|
3482 | // Replace <DOT> with . for urls |
---|
3483 | $stuff_to_parse = str_replace('<DOT>', '.', $stuff_to_parse); |
---|
3484 | // Replace <|UR1| with <a href= for urls also |
---|
3485 | if (isset($this->link_styles[GESHI_LINK])) { |
---|
3486 | if ($this->use_classes) { |
---|
3487 | $stuff_to_parse = str_replace('<|UR1|', '<a' . $this->link_target . ' href=', $stuff_to_parse); |
---|
3488 | } else { |
---|
3489 | $stuff_to_parse = str_replace('<|UR1|', '<a' . $this->link_target . ' style="' . $this->link_styles[GESHI_LINK] . '" href=', $stuff_to_parse); |
---|
3490 | } |
---|
3491 | } else { |
---|
3492 | $stuff_to_parse = str_replace('<|UR1|', '<a' . $this->link_target . ' href=', $stuff_to_parse); |
---|
3493 | } |
---|
3494 | |
---|
3495 | // |
---|
3496 | // NOW we add the span thingy ;) |
---|
3497 | // |
---|
3498 | |
---|
3499 | $stuff_to_parse = str_replace('<|', '<span', $stuff_to_parse); |
---|
3500 | $stuff_to_parse = str_replace ( '|>', '</span>', $stuff_to_parse ); |
---|
3501 | return substr($stuff_to_parse, 1); |
---|
3502 | } |
---|
3503 | |
---|
3504 | /** |
---|
3505 | * Sets the time taken to parse the code |
---|
3506 | * |
---|
3507 | * @param microtime The time when parsing started |
---|
3508 | * @param microtime The time when parsing ended |
---|
3509 | * @since 1.0.2 |
---|
3510 | * @access private |
---|
3511 | */ |
---|
3512 | function set_time($start_time, $end_time) { |
---|
3513 | $start = explode(' ', $start_time); |
---|
3514 | $end = explode(' ', $end_time); |
---|
3515 | $this->time = $end[0] + $end[1] - $start[0] - $start[1]; |
---|
3516 | } |
---|
3517 | |
---|
3518 | /** |
---|
3519 | * Gets the time taken to parse the code |
---|
3520 | * |
---|
3521 | * @return double The time taken to parse the code |
---|
3522 | * @since 1.0.2 |
---|
3523 | */ |
---|
3524 | function get_time() { |
---|
3525 | return $this->time; |
---|
3526 | } |
---|
3527 | |
---|
3528 | /** |
---|
3529 | * Merges arrays recursively, overwriting values of the first array with values of later arrays |
---|
3530 | * |
---|
3531 | * @since 1.0.8 |
---|
3532 | * @access private |
---|
3533 | */ |
---|
3534 | function merge_arrays() { |
---|
3535 | $arrays = func_get_args(); |
---|
3536 | $narrays = count($arrays); |
---|
3537 | |
---|
3538 | // check arguments |
---|
3539 | // comment out if more performance is necessary (in this case the foreach loop will trigger a warning if the argument is not an array) |
---|
3540 | for ($i = 0; $i < $narrays; $i ++) { |
---|
3541 | if (!is_array($arrays[$i])) { |
---|
3542 | // also array_merge_recursive returns nothing in this case |
---|
3543 | trigger_error('Argument #' . ($i+1) . ' is not an array - trying to merge array with scalar! Returning false!', E_USER_WARNING); |
---|
3544 | return false; |
---|
3545 | } |
---|
3546 | } |
---|
3547 | |
---|
3548 | // the first array is in the output set in every case |
---|
3549 | $ret = $arrays[0]; |
---|
3550 | |
---|
3551 | // merege $ret with the remaining arrays |
---|
3552 | for ($i = 1; $i < $narrays; $i ++) { |
---|
3553 | foreach ($arrays[$i] as $key => $value) { |
---|
3554 | if (is_array($value) && isset($ret[$key])) { |
---|
3555 | // if $ret[$key] is not an array you try to merge an scalar value with an array - the result is not defined (incompatible arrays) |
---|
3556 | // in this case the call will trigger an E_USER_WARNING and the $ret[$key] will be false. |
---|
3557 | $ret[$key] = $this->merge_arrays($ret[$key], $value); |
---|
3558 | } else { |
---|
3559 | $ret[$key] = $value; |
---|
3560 | } |
---|
3561 | } |
---|
3562 | } |
---|
3563 | |
---|
3564 | return $ret; |
---|
3565 | } |
---|
3566 | |
---|
3567 | /** |
---|
3568 | * Gets language information and stores it for later use |
---|
3569 | * |
---|
3570 | * @param string The filename of the language file you want to load |
---|
3571 | * @since 1.0.0 |
---|
3572 | * @access private |
---|
3573 | * @todo Needs to load keys for lexic permissions for keywords, regexps etc |
---|
3574 | */ |
---|
3575 | function load_language($file_name) { |
---|
3576 | if ($file_name == $this->loaded_language) { |
---|
3577 | // this file is already loaded! |
---|
3578 | return; |
---|
3579 | } |
---|
3580 | |
---|
3581 | //Prepare some stuff before actually loading the language file |
---|
3582 | $this->loaded_language = $file_name; |
---|
3583 | $this->parse_cache_built = false; |
---|
3584 | $this->enable_highlighting(); |
---|
3585 | $language_data = array(); |
---|
3586 | |
---|
3587 | //Load the language file |
---|
3588 | require $file_name; |
---|
3589 | |
---|
3590 | // Perhaps some checking might be added here later to check that |
---|
3591 | // $language data is a valid thing but maybe not |
---|
3592 | $this->language_data = $language_data; |
---|
3593 | |
---|
3594 | // Set strict mode if should be set |
---|
3595 | $this->strict_mode = $this->language_data['STRICT_MODE_APPLIES']; |
---|
3596 | |
---|
3597 | // Set permissions for all lexics to true |
---|
3598 | // so they'll be highlighted by default |
---|
3599 | foreach (array_keys($this->language_data['KEYWORDS']) as $key) { |
---|
3600 | if (!empty($this->language_data['KEYWORDS'][$key])) { |
---|
3601 | $this->lexic_permissions['KEYWORDS'][$key] = true; |
---|
3602 | } else { |
---|
3603 | $this->lexic_permissions['KEYWORDS'][$key] = false; |
---|
3604 | } |
---|
3605 | } |
---|
3606 | |
---|
3607 | foreach (array_keys($this->language_data['COMMENT_SINGLE']) as $key) { |
---|
3608 | $this->lexic_permissions['COMMENTS'][$key] = true; |
---|
3609 | } |
---|
3610 | foreach (array_keys($this->language_data['REGEXPS']) as $key) { |
---|
3611 | $this->lexic_permissions['REGEXPS'][$key] = true; |
---|
3612 | } |
---|
3613 | |
---|
3614 | // for BenBE and future code reviews: |
---|
3615 | // we can use empty here since we only check for existance and emptiness of an array |
---|
3616 | // if it is not an array at all but rather false or null this will work as intended as well |
---|
3617 | // even if $this->language_data['PARSER_CONTROL'] is undefined this won't trigger a notice |
---|
3618 | if (!empty($this->language_data['PARSER_CONTROL']['ENABLE_FLAGS'])) { |
---|
3619 | foreach ($this->language_data['PARSER_CONTROL']['ENABLE_FLAGS'] as $flag => $value) { |
---|
3620 | // it's either true or false and maybe is true as well |
---|
3621 | $perm = $value !== GESHI_NEVER; |
---|
3622 | if ($flag == 'ALL') { |
---|
3623 | $this->enable_highlighting($perm); |
---|
3624 | continue; |
---|
3625 | } |
---|
3626 | if (!isset($this->lexic_permissions[$flag])) { |
---|
3627 | // unknown lexic permission |
---|
3628 | continue; |
---|
3629 | } |
---|
3630 | if (is_array($this->lexic_permissions[$flag])) { |
---|
3631 | foreach ($this->lexic_permissions[$flag] as $key => $val) { |
---|
3632 | $this->lexic_permissions[$flag][$key] = $perm; |
---|
3633 | } |
---|
3634 | } else { |
---|
3635 | $this->lexic_permissions[$flag] = $perm; |
---|
3636 | } |
---|
3637 | } |
---|
3638 | unset($this->language_data['PARSER_CONTROL']['ENABLE_FLAGS']); |
---|
3639 | } |
---|
3640 | |
---|
3641 | //NEW in 1.0.8: Allow styles to be loaded from a separate file to override defaults |
---|
3642 | $style_filename = substr($file_name, 0, -4) . '.style.php'; |
---|
3643 | if (is_readable($style_filename)) { |
---|
3644 | //Clear any style_data that could have been set before ... |
---|
3645 | if (isset($style_data)) { |
---|
3646 | unset($style_data); |
---|
3647 | } |
---|
3648 | |
---|
3649 | //Read the Style Information from the style file |
---|
3650 | include $style_filename; |
---|
3651 | |
---|
3652 | //Apply the new styles to our current language styles |
---|
3653 | if (isset($style_data) && is_array($style_data)) { |
---|
3654 | $this->language_data['STYLES'] = |
---|
3655 | $this->merge_arrays($this->language_data['STYLES'], $style_data); |
---|
3656 | } |
---|
3657 | } |
---|
3658 | } |
---|
3659 | |
---|
3660 | /** |
---|
3661 | * Takes the parsed code and various options, and creates the HTML |
---|
3662 | * surrounding it to make it look nice. |
---|
3663 | * |
---|
3664 | * @param string The code already parsed (reference!) |
---|
3665 | * @since 1.0.0 |
---|
3666 | * @access private |
---|
3667 | */ |
---|
3668 | function finalise(&$parsed_code) { |
---|
3669 | // Remove end parts of important declarations |
---|
3670 | // This is BUGGY!! My fault for bad code: fix coming in 1.2 |
---|
3671 | // @todo Remove this crap |
---|
3672 | if ($this->enable_important_blocks && |
---|
3673 | (strpos($parsed_code, $this->hsc(GESHI_START_IMPORTANT)) === false)) { |
---|
3674 | $parsed_code = str_replace($this->hsc(GESHI_END_IMPORTANT), '', $parsed_code); |
---|
3675 | } |
---|
3676 | |
---|
3677 | // Add HTML whitespace stuff if we're using the <div> header |
---|
3678 | if ($this->header_type != GESHI_HEADER_PRE && $this->header_type != GESHI_HEADER_PRE_VALID) { |
---|
3679 | $this->indent($parsed_code); |
---|
3680 | } |
---|
3681 | |
---|
3682 | // purge some unnecessary stuff |
---|
3683 | /** NOTE: memorypeak #1 */ |
---|
3684 | $parsed_code = preg_replace('#<span[^>]+>(\s*)</span>#', '\\1', $parsed_code); |
---|
3685 | |
---|
3686 | // If we are using IDs for line numbers, there needs to be an overall |
---|
3687 | // ID set to prevent collisions. |
---|
3688 | if ($this->add_ids && !$this->overall_id) { |
---|
3689 | $this->overall_id = 'geshi-' . substr(md5(microtime()), 0, 4); |
---|
3690 | } |
---|
3691 | |
---|
3692 | // Get code into lines |
---|
3693 | /** NOTE: memorypeak #2 */ |
---|
3694 | $code = explode("\n", $parsed_code); |
---|
3695 | $parsed_code = $this->header(); |
---|
3696 | |
---|
3697 | // If we're using line numbers, we insert <li>s and appropriate |
---|
3698 | // markup to style them (otherwise we don't need to do anything) |
---|
3699 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS && $this->header_type != GESHI_HEADER_PRE_TABLE) { |
---|
3700 | // If we're using the <pre> header, we shouldn't add newlines because |
---|
3701 | // the <pre> will line-break them (and the <li>s already do this for us) |
---|
3702 | $ls = ($this->header_type != GESHI_HEADER_PRE && $this->header_type != GESHI_HEADER_PRE_VALID) ? "\n" : ''; |
---|
3703 | |
---|
3704 | // Set vars to defaults for following loop |
---|
3705 | $i = 0; |
---|
3706 | |
---|
3707 | // Foreach line... |
---|
3708 | for ($i = 0, $n = count($code); $i < $n;) { |
---|
3709 | //Reset the attributes for a new line ... |
---|
3710 | $attrs = array(); |
---|
3711 | |
---|
3712 | // Make lines have at least one space in them if they're empty |
---|
3713 | // BenBE: Checking emptiness using trim instead of relying on blanks |
---|
3714 | if ('' == trim($code[$i])) { |
---|
3715 | $code[$i] = ' '; |
---|
3716 | } |
---|
3717 | |
---|
3718 | // If this is a "special line"... |
---|
3719 | if ($this->line_numbers == GESHI_FANCY_LINE_NUMBERS && |
---|
3720 | $i % $this->line_nth_row == ($this->line_nth_row - 1)) { |
---|
3721 | // Set the attributes to style the line |
---|
3722 | if ($this->use_classes) { |
---|
3723 | //$attr = ' class="li2"'; |
---|
3724 | $attrs['class'][] = 'li2'; |
---|
3725 | $def_attr = ' class="de2"'; |
---|
3726 | } else { |
---|
3727 | //$attr = ' style="' . $this->line_style2 . '"'; |
---|
3728 | $attrs['style'][] = $this->line_style2; |
---|
3729 | // This style "covers up" the special styles set for special lines |
---|
3730 | // so that styles applied to special lines don't apply to the actual |
---|
3731 | // code on that line |
---|
3732 | $def_attr = ' style="' . $this->code_style . '"'; |
---|
3733 | } |
---|
3734 | } else { |
---|
3735 | if ($this->use_classes) { |
---|
3736 | //$attr = ' class="li1"'; |
---|
3737 | $attrs['class'][] = 'li1'; |
---|
3738 | $def_attr = ' class="de1"'; |
---|
3739 | } else { |
---|
3740 | //$attr = ' style="' . $this->line_style1 . '"'; |
---|
3741 | $attrs['style'][] = $this->line_style1; |
---|
3742 | $def_attr = ' style="' . $this->code_style . '"'; |
---|
3743 | } |
---|
3744 | } |
---|
3745 | |
---|
3746 | //Check which type of tag to insert for this line |
---|
3747 | if ($this->header_type == GESHI_HEADER_PRE_VALID) { |
---|
3748 | $start = "<pre$def_attr>"; |
---|
3749 | $end = '</pre>'; |
---|
3750 | } else { |
---|
3751 | // Span or div? |
---|
3752 | $start = "<div$def_attr>"; |
---|
3753 | $end = '</div>'; |
---|
3754 | } |
---|
3755 | |
---|
3756 | ++$i; |
---|
3757 | |
---|
3758 | // Are we supposed to use ids? If so, add them |
---|
3759 | if ($this->add_ids) { |
---|
3760 | $attrs['id'][] = "$this->overall_id-$i"; |
---|
3761 | } |
---|
3762 | |
---|
3763 | //Is this some line with extra styles??? |
---|
3764 | if (in_array($i, $this->highlight_extra_lines)) { |
---|
3765 | if ($this->use_classes) { |
---|
3766 | if (isset($this->highlight_extra_lines_styles[$i])) { |
---|
3767 | $attrs['class'][] = "lx$i"; |
---|
3768 | } else { |
---|
3769 | $attrs['class'][] = "ln-xtra"; |
---|
3770 | } |
---|
3771 | } else { |
---|
3772 | array_push($attrs['style'], $this->get_line_style($i)); |
---|
3773 | } |
---|
3774 | } |
---|
3775 | |
---|
3776 | // Add in the line surrounded by appropriate list HTML |
---|
3777 | $attr_string = ''; |
---|
3778 | foreach ($attrs as $key => $attr) { |
---|
3779 | $attr_string .= ' ' . $key . '="' . implode(' ', $attr) . '"'; |
---|
3780 | } |
---|
3781 | |
---|
3782 | $parsed_code .= "<li$attr_string>$start{$code[$i-1]}$end</li>$ls"; |
---|
3783 | unset($code[$i - 1]); |
---|
3784 | } |
---|
3785 | } else { |
---|
3786 | $n = count($code); |
---|
3787 | if ($this->use_classes) { |
---|
3788 | $attributes = ' class="de1"'; |
---|
3789 | } else { |
---|
3790 | $attributes = ' style="'. $this->code_style .'"'; |
---|
3791 | } |
---|
3792 | if ($this->header_type == GESHI_HEADER_PRE_VALID) { |
---|
3793 | $parsed_code .= '<pre'. $attributes .'>'; |
---|
3794 | } elseif ($this->header_type == GESHI_HEADER_PRE_TABLE) { |
---|
3795 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3796 | if ($this->use_classes) { |
---|
3797 | $attrs = ' class="ln"'; |
---|
3798 | } else { |
---|
3799 | $attrs = ' style="'. $this->table_linenumber_style .'"'; |
---|
3800 | } |
---|
3801 | $parsed_code .= '<td'.$attrs.'><pre'.$attributes.'>'; |
---|
3802 | // get linenumbers |
---|
3803 | // we don't merge it with the for below, since it should be better for |
---|
3804 | // memory consumption this way |
---|
3805 | // @todo: but... actually it would still be somewhat nice to merge the two loops |
---|
3806 | // the mem peaks are at different positions |
---|
3807 | for ($i = 0; $i < $n; ++$i) { |
---|
3808 | $close = 0; |
---|
3809 | // fancy lines |
---|
3810 | if ($this->line_numbers == GESHI_FANCY_LINE_NUMBERS && |
---|
3811 | $i % $this->line_nth_row == ($this->line_nth_row - 1)) { |
---|
3812 | // Set the attributes to style the line |
---|
3813 | if ($this->use_classes) { |
---|
3814 | $parsed_code .= '<span class="xtra li2"><span class="de2">'; |
---|
3815 | } else { |
---|
3816 | // This style "covers up" the special styles set for special lines |
---|
3817 | // so that styles applied to special lines don't apply to the actual |
---|
3818 | // code on that line |
---|
3819 | $parsed_code .= '<span style="display:block;' . $this->line_style2 . '">' |
---|
3820 | .'<span style="' . $this->code_style .'">'; |
---|
3821 | } |
---|
3822 | $close += 2; |
---|
3823 | } |
---|
3824 | //Is this some line with extra styles??? |
---|
3825 | if (in_array($i + 1, $this->highlight_extra_lines)) { |
---|
3826 | if ($this->use_classes) { |
---|
3827 | if (isset($this->highlight_extra_lines_styles[$i])) { |
---|
3828 | $parsed_code .= "<span class=\"xtra lx$i\">"; |
---|
3829 | } else { |
---|
3830 | $parsed_code .= "<span class=\"xtra ln-xtra\">"; |
---|
3831 | } |
---|
3832 | } else { |
---|
3833 | $parsed_code .= "<span style=\"display:block;" . $this->get_line_style($i) . "\">"; |
---|
3834 | } |
---|
3835 | ++$close; |
---|
3836 | } |
---|
3837 | $parsed_code .= $this->line_numbers_start + $i; |
---|
3838 | if ($close) { |
---|
3839 | $parsed_code .= str_repeat('</span>', $close); |
---|
3840 | } else if ($i != $n) { |
---|
3841 | $parsed_code .= "\n"; |
---|
3842 | } |
---|
3843 | } |
---|
3844 | $parsed_code .= '</pre></td><td'.$attributes.'>'; |
---|
3845 | } |
---|
3846 | $parsed_code .= '<pre'. $attributes .'>'; |
---|
3847 | } |
---|
3848 | // No line numbers, but still need to handle highlighting lines extra. |
---|
3849 | // Have to use divs so the full width of the code is highlighted |
---|
3850 | $close = 0; |
---|
3851 | for ($i = 0; $i < $n; ++$i) { |
---|
3852 | // Make lines have at least one space in them if they're empty |
---|
3853 | // BenBE: Checking emptiness using trim instead of relying on blanks |
---|
3854 | if ('' == trim($code[$i])) { |
---|
3855 | $code[$i] = ' '; |
---|
3856 | } |
---|
3857 | // fancy lines |
---|
3858 | if ($this->line_numbers == GESHI_FANCY_LINE_NUMBERS && |
---|
3859 | $i % $this->line_nth_row == ($this->line_nth_row - 1)) { |
---|
3860 | // Set the attributes to style the line |
---|
3861 | if ($this->use_classes) { |
---|
3862 | $parsed_code .= '<span class="xtra li2"><span class="de2">'; |
---|
3863 | } else { |
---|
3864 | // This style "covers up" the special styles set for special lines |
---|
3865 | // so that styles applied to special lines don't apply to the actual |
---|
3866 | // code on that line |
---|
3867 | $parsed_code .= '<span style="display:block;' . $this->line_style2 . '">' |
---|
3868 | .'<span style="' . $this->code_style .'">'; |
---|
3869 | } |
---|
3870 | $close += 2; |
---|
3871 | } |
---|
3872 | //Is this some line with extra styles??? |
---|
3873 | if (in_array($i + 1, $this->highlight_extra_lines)) { |
---|
3874 | if ($this->use_classes) { |
---|
3875 | if (isset($this->highlight_extra_lines_styles[$i])) { |
---|
3876 | $parsed_code .= "<span class=\"xtra lx$i\">"; |
---|
3877 | } else { |
---|
3878 | $parsed_code .= "<span class=\"xtra ln-xtra\">"; |
---|
3879 | } |
---|
3880 | } else { |
---|
3881 | $parsed_code .= "<span style=\"display:block;" . $this->get_line_style($i) . "\">"; |
---|
3882 | } |
---|
3883 | ++$close; |
---|
3884 | } |
---|
3885 | |
---|
3886 | $parsed_code .= $code[$i]; |
---|
3887 | |
---|
3888 | if ($close) { |
---|
3889 | $parsed_code .= str_repeat('</span>', $close); |
---|
3890 | $close = 0; |
---|
3891 | } |
---|
3892 | elseif ($i + 1 < $n) { |
---|
3893 | $parsed_code .= "\n"; |
---|
3894 | } |
---|
3895 | unset($code[$i]); |
---|
3896 | } |
---|
3897 | |
---|
3898 | if ($this->header_type == GESHI_HEADER_PRE_VALID || $this->header_type == GESHI_HEADER_PRE_TABLE) { |
---|
3899 | $parsed_code .= '</pre>'; |
---|
3900 | } |
---|
3901 | if ($this->header_type == GESHI_HEADER_PRE_TABLE && $this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3902 | $parsed_code .= '</td>'; |
---|
3903 | } |
---|
3904 | } |
---|
3905 | |
---|
3906 | $parsed_code .= $this->footer(); |
---|
3907 | } |
---|
3908 | |
---|
3909 | /** |
---|
3910 | * Creates the header for the code block (with correct attributes) |
---|
3911 | * |
---|
3912 | * @return string The header for the code block |
---|
3913 | * @since 1.0.0 |
---|
3914 | * @access private |
---|
3915 | */ |
---|
3916 | function header() { |
---|
3917 | // Get attributes needed |
---|
3918 | /** |
---|
3919 | * @todo Document behaviour change - class is outputted regardless of whether |
---|
3920 | * we're using classes or not. Same with style |
---|
3921 | */ |
---|
3922 | $attributes = ' class="' . $this->language; |
---|
3923 | if ($this->overall_class != '') { |
---|
3924 | $attributes .= " ".$this->overall_class; |
---|
3925 | } |
---|
3926 | $attributes .= '"'; |
---|
3927 | |
---|
3928 | if ($this->overall_id != '') { |
---|
3929 | $attributes .= " id=\"{$this->overall_id}\""; |
---|
3930 | } |
---|
3931 | if ($this->overall_style != '') { |
---|
3932 | $attributes .= ' style="' . $this->overall_style . '"'; |
---|
3933 | } |
---|
3934 | |
---|
3935 | $ol_attributes = ''; |
---|
3936 | |
---|
3937 | if ($this->line_numbers_start != 1) { |
---|
3938 | $ol_attributes .= ' start="' . $this->line_numbers_start . '"'; |
---|
3939 | } |
---|
3940 | |
---|
3941 | // Get the header HTML |
---|
3942 | $header = $this->header_content; |
---|
3943 | if ($header) { |
---|
3944 | if ($this->header_type == GESHI_HEADER_PRE || $this->header_type == GESHI_HEADER_PRE_VALID) { |
---|
3945 | $header = str_replace("\n", '', $header); |
---|
3946 | } |
---|
3947 | $header = $this->replace_keywords($header); |
---|
3948 | |
---|
3949 | if ($this->use_classes) { |
---|
3950 | $attr = ' class="head"'; |
---|
3951 | } else { |
---|
3952 | $attr = " style=\"{$this->header_content_style}\""; |
---|
3953 | } |
---|
3954 | if ($this->header_type == GESHI_HEADER_PRE_TABLE && $this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3955 | $header = "<thead><tr><td colspan=\"2\" $attr>$header</td></tr></thead>"; |
---|
3956 | } else { |
---|
3957 | $header = "<div$attr>$header</div>"; |
---|
3958 | } |
---|
3959 | } |
---|
3960 | |
---|
3961 | if (GESHI_HEADER_NONE == $this->header_type) { |
---|
3962 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3963 | return "$header<ol$attributes$ol_attributes>"; |
---|
3964 | } |
---|
3965 | return $header . ($this->force_code_block ? '<div>' : ''); |
---|
3966 | } |
---|
3967 | |
---|
3968 | // Work out what to return and do it |
---|
3969 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
3970 | if ($this->header_type == GESHI_HEADER_PRE) { |
---|
3971 | return "<pre$attributes>$header<ol$ol_attributes>"; |
---|
3972 | } else if ($this->header_type == GESHI_HEADER_DIV || |
---|
3973 | $this->header_type == GESHI_HEADER_PRE_VALID) { |
---|
3974 | return "<div$attributes>$header<ol$ol_attributes>"; |
---|
3975 | } else if ($this->header_type == GESHI_HEADER_PRE_TABLE) { |
---|
3976 | return "<table$attributes>$header<tbody><tr class=\"li1\">"; |
---|
3977 | } |
---|
3978 | } else { |
---|
3979 | if ($this->header_type == GESHI_HEADER_PRE) { |
---|
3980 | return "<pre$attributes>$header" . |
---|
3981 | ($this->force_code_block ? '<div>' : ''); |
---|
3982 | } else { |
---|
3983 | return "<div$attributes>$header" . |
---|
3984 | ($this->force_code_block ? '<div>' : ''); |
---|
3985 | } |
---|
3986 | } |
---|
3987 | } |
---|
3988 | |
---|
3989 | /** |
---|
3990 | * Returns the footer for the code block. |
---|
3991 | * |
---|
3992 | * @return string The footer for the code block |
---|
3993 | * @since 1.0.0 |
---|
3994 | * @access private |
---|
3995 | */ |
---|
3996 | function footer() { |
---|
3997 | $footer = $this->footer_content; |
---|
3998 | if ($footer) { |
---|
3999 | if ($this->header_type == GESHI_HEADER_PRE) { |
---|
4000 | $footer = str_replace("\n", '', $footer);; |
---|
4001 | } |
---|
4002 | $footer = $this->replace_keywords($footer); |
---|
4003 | |
---|
4004 | if ($this->use_classes) { |
---|
4005 | $attr = ' class="foot"'; |
---|
4006 | } else { |
---|
4007 | $attr = " style=\"{$this->footer_content_style}\""; |
---|
4008 | } |
---|
4009 | if ($this->header_type == GESHI_HEADER_PRE_TABLE && $this->linenumbers != GESHI_NO_LINE_NUMBERS) { |
---|
4010 | $footer = "<tfoot><tr><td colspan=\"2\">$footer</td></tr></tfoot>"; |
---|
4011 | } else { |
---|
4012 | $footer = "<div$attr>$footer</div>"; |
---|
4013 | } |
---|
4014 | } |
---|
4015 | |
---|
4016 | if (GESHI_HEADER_NONE == $this->header_type) { |
---|
4017 | return ($this->line_numbers != GESHI_NO_LINE_NUMBERS) ? '</ol>' . $footer : $footer; |
---|
4018 | } |
---|
4019 | |
---|
4020 | if ($this->header_type == GESHI_HEADER_DIV || $this->header_type == GESHI_HEADER_PRE_VALID) { |
---|
4021 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
4022 | return "</ol>$footer</div>"; |
---|
4023 | } |
---|
4024 | return ($this->force_code_block ? '</div>' : '') . |
---|
4025 | "$footer</div>"; |
---|
4026 | } |
---|
4027 | elseif ($this->header_type == GESHI_HEADER_PRE_TABLE) { |
---|
4028 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
4029 | return "</tr></tbody>$footer</table>"; |
---|
4030 | } |
---|
4031 | return ($this->force_code_block ? '</div>' : '') . |
---|
4032 | "$footer</div>"; |
---|
4033 | } |
---|
4034 | else { |
---|
4035 | if ($this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
4036 | return "</ol>$footer</pre>"; |
---|
4037 | } |
---|
4038 | return ($this->force_code_block ? '</div>' : '') . |
---|
4039 | "$footer</pre>"; |
---|
4040 | } |
---|
4041 | } |
---|
4042 | |
---|
4043 | /** |
---|
4044 | * Replaces certain keywords in the header and footer with |
---|
4045 | * certain configuration values |
---|
4046 | * |
---|
4047 | * @param string The header or footer content to do replacement on |
---|
4048 | * @return string The header or footer with replaced keywords |
---|
4049 | * @since 1.0.2 |
---|
4050 | * @access private |
---|
4051 | */ |
---|
4052 | function replace_keywords($instr) { |
---|
4053 | $keywords = $replacements = array(); |
---|
4054 | |
---|
4055 | $keywords[] = '<TIME>'; |
---|
4056 | $keywords[] = '{TIME}'; |
---|
4057 | $replacements[] = $replacements[] = number_format($time = $this->get_time(), 3); |
---|
4058 | |
---|
4059 | $keywords[] = '<LANGUAGE>'; |
---|
4060 | $keywords[] = '{LANGUAGE}'; |
---|
4061 | $replacements[] = $replacements[] = $this->language_data['LANG_NAME']; |
---|
4062 | |
---|
4063 | $keywords[] = '<VERSION>'; |
---|
4064 | $keywords[] = '{VERSION}'; |
---|
4065 | $replacements[] = $replacements[] = GESHI_VERSION; |
---|
4066 | |
---|
4067 | $keywords[] = '<SPEED>'; |
---|
4068 | $keywords[] = '{SPEED}'; |
---|
4069 | if ($time <= 0) { |
---|
4070 | $speed = 'N/A'; |
---|
4071 | } else { |
---|
4072 | $speed = strlen($this->source) / $time; |
---|
4073 | if ($speed >= 1024) { |
---|
4074 | $speed = sprintf("%.2f KB/s", $speed / 1024.0); |
---|
4075 | } else { |
---|
4076 | $speed = sprintf("%.0f B/s", $speed); |
---|
4077 | } |
---|
4078 | } |
---|
4079 | $replacements[] = $replacements[] = $speed; |
---|
4080 | |
---|
4081 | return str_replace($keywords, $replacements, $instr); |
---|
4082 | } |
---|
4083 | |
---|
4084 | /** |
---|
4085 | * Secure replacement for PHP built-in function htmlspecialchars(). |
---|
4086 | * |
---|
4087 | * See ticket #427 (http://wush.net/trac/wikka/ticket/427) for the rationale |
---|
4088 | * for this replacement function. |
---|
4089 | * |
---|
4090 | * The INTERFACE for this function is almost the same as that for |
---|
4091 | * htmlspecialchars(), with the same default for quote style; however, there |
---|
4092 | * is no 'charset' parameter. The reason for this is as follows: |
---|
4093 | * |
---|
4094 | * The PHP docs say: |
---|
4095 | * "The third argument charset defines character set used in conversion." |
---|
4096 | * |
---|
4097 | * I suspect PHP's htmlspecialchars() is working at the byte-value level and |
---|
4098 | * thus _needs_ to know (or asssume) a character set because the special |
---|
4099 | * characters to be replaced could exist at different code points in |
---|
4100 | * different character sets. (If indeed htmlspecialchars() works at |
---|
4101 | * byte-value level that goes some way towards explaining why the |
---|
4102 | * vulnerability would exist in this function, too, and not only in |
---|
4103 | * htmlentities() which certainly is working at byte-value level.) |
---|
4104 | * |
---|
4105 | * This replacement function however works at character level and should |
---|
4106 | * therefore be "immune" to character set differences - so no charset |
---|
4107 | * parameter is needed or provided. If a third parameter is passed, it will |
---|
4108 | * be silently ignored. |
---|
4109 | * |
---|
4110 | * In the OUTPUT there is a minor difference in that we use ''' instead |
---|
4111 | * of PHP's ''' for a single quote: this provides compatibility with |
---|
4112 | * get_html_translation_table(HTML_SPECIALCHARS, ENT_QUOTES) |
---|
4113 | * (see comment by mikiwoz at yahoo dot co dot uk on |
---|
4114 | * http://php.net/htmlspecialchars); it also matches the entity definition |
---|
4115 | * for XML 1.0 |
---|
4116 | * (http://www.w3.org/TR/xhtml1/dtds.html#a_dtd_Special_characters). |
---|
4117 | * Like PHP we use a numeric character reference instead of ''' for the |
---|
4118 | * single quote. For the other special characters we use the named entity |
---|
4119 | * references, as PHP is doing. |
---|
4120 | * |
---|
4121 | * @author {@link http://wikkawiki.org/JavaWoman Marjolein Katsma} |
---|
4122 | * |
---|
4123 | * @license http://www.gnu.org/copyleft/lgpl.html |
---|
4124 | * GNU Lesser General Public License |
---|
4125 | * @copyright Copyright 2007, {@link http://wikkawiki.org/CreditsPage |
---|
4126 | * Wikka Development Team} |
---|
4127 | * |
---|
4128 | * @access private |
---|
4129 | * @param string $string string to be converted |
---|
4130 | * @param integer $quote_style |
---|
4131 | * - ENT_COMPAT: escapes &, <, > and double quote (default) |
---|
4132 | * - ENT_NOQUOTES: escapes only &, < and > |
---|
4133 | * - ENT_QUOTES: escapes &, <, >, double and single quotes |
---|
4134 | * @return string converted string |
---|
4135 | * @since 1.0.7.18 |
---|
4136 | */ |
---|
4137 | function hsc($string, $quote_style = ENT_COMPAT) { |
---|
4138 | // init |
---|
4139 | static $aTransSpecchar = array( |
---|
4140 | '&' => '&', |
---|
4141 | '"' => '"', |
---|
4142 | '<' => '<', |
---|
4143 | '>' => '>', |
---|
4144 | |
---|
4145 | //This fix is related to SF#1923020, but has to be applied |
---|
4146 | //regardless of actually highlighting symbols. |
---|
4147 | |
---|
4148 | //Circumvent a bug with symbol highlighting |
---|
4149 | //This is required as ; would produce undesirable side-effects if it |
---|
4150 | //was not to be processed as an entity. |
---|
4151 | ';' => '<SEMI>', // Force ; to be processed as entity |
---|
4152 | '|' => '<PIPE>' // Force | to be processed as entity |
---|
4153 | ); // ENT_COMPAT set |
---|
4154 | |
---|
4155 | switch ($quote_style) { |
---|
4156 | case ENT_NOQUOTES: // don't convert double quotes |
---|
4157 | unset($aTransSpecchar['"']); |
---|
4158 | break; |
---|
4159 | case ENT_QUOTES: // convert single quotes as well |
---|
4160 | $aTransSpecchar["'"] = '''; // (apos) htmlspecialchars() uses ''' |
---|
4161 | break; |
---|
4162 | } |
---|
4163 | |
---|
4164 | // return translated string |
---|
4165 | return strtr($string, $aTransSpecchar); |
---|
4166 | } |
---|
4167 | |
---|
4168 | /** |
---|
4169 | * Returns a stylesheet for the highlighted code. If $economy mode |
---|
4170 | * is true, we only return the stylesheet declarations that matter for |
---|
4171 | * this code block instead of the whole thing |
---|
4172 | * |
---|
4173 | * @param boolean Whether to use economy mode or not |
---|
4174 | * @return string A stylesheet built on the data for the current language |
---|
4175 | * @since 1.0.0 |
---|
4176 | */ |
---|
4177 | function get_stylesheet($economy_mode = true) { |
---|
4178 | // If there's an error, chances are that the language file |
---|
4179 | // won't have populated the language data file, so we can't |
---|
4180 | // risk getting a stylesheet... |
---|
4181 | if ($this->error) { |
---|
4182 | return ''; |
---|
4183 | } |
---|
4184 | |
---|
4185 | //Check if the style rearrangements have been processed ... |
---|
4186 | //This also does some preprocessing to check which style groups are useable ... |
---|
4187 | if(!isset($this->language_data['NUMBERS_CACHE'])) { |
---|
4188 | $this->build_style_cache(); |
---|
4189 | } |
---|
4190 | |
---|
4191 | // First, work out what the selector should be. If there's an ID, |
---|
4192 | // that should be used, the same for a class. Otherwise, a selector |
---|
4193 | // of '' means that these styles will be applied anywhere |
---|
4194 | if ($this->overall_id) { |
---|
4195 | $selector = '#' . $this->overall_id; |
---|
4196 | } else { |
---|
4197 | $selector = '.' . $this->language; |
---|
4198 | if ($this->overall_class) { |
---|
4199 | $selector .= '.' . $this->overall_class; |
---|
4200 | } |
---|
4201 | } |
---|
4202 | $selector .= ' '; |
---|
4203 | |
---|
4204 | // Header of the stylesheet |
---|
4205 | if (!$economy_mode) { |
---|
4206 | $stylesheet = "/**\n". |
---|
4207 | " * GeSHi Dynamically Generated Stylesheet\n". |
---|
4208 | " * --------------------------------------\n". |
---|
4209 | " * Dynamically generated stylesheet for {$this->language}\n". |
---|
4210 | " * CSS class: {$this->overall_class}, CSS id: {$this->overall_id}\n". |
---|
4211 | " * GeSHi (C) 2004 - 2007 Nigel McNie, 2007 - 2008 Benny Baumann\n" . |
---|
4212 | " * (http://qbnz.com/highlighter/ and http://geshi.org/)\n". |
---|
4213 | " * --------------------------------------\n". |
---|
4214 | " */\n"; |
---|
4215 | } else { |
---|
4216 | $stylesheet = "/**\n". |
---|
4217 | " * GeSHi (C) 2004 - 2007 Nigel McNie, 2007 - 2008 Benny Baumann\n" . |
---|
4218 | " * (http://qbnz.com/highlighter/ and http://geshi.org/)\n". |
---|
4219 | " */\n"; |
---|
4220 | } |
---|
4221 | |
---|
4222 | // Set the <ol> to have no effect at all if there are line numbers |
---|
4223 | // (<ol>s have margins that should be destroyed so all layout is |
---|
4224 | // controlled by the set_overall_style method, which works on the |
---|
4225 | // <pre> or <div> container). Additionally, set default styles for lines |
---|
4226 | if (!$economy_mode || $this->line_numbers != GESHI_NO_LINE_NUMBERS) { |
---|
4227 | //$stylesheet .= "$selector, {$selector}ol, {$selector}ol li {margin: 0;}\n"; |
---|
4228 | $stylesheet .= "$selector.de1, $selector.de2 {{$this->code_style}}\n"; |
---|
4229 | } |
---|
4230 | |
---|
4231 | // Add overall styles |
---|
4232 | // note: neglect economy_mode, empty styles are meaningless |
---|
4233 | if ($this->overall_style != '') { |
---|
4234 | $stylesheet .= "$selector {{$this->overall_style}}\n"; |
---|
4235 | } |
---|
4236 | |
---|
4237 | // Add styles for links |
---|
4238 | // note: economy mode does not make _any_ sense here |
---|
4239 | // either the style is empty and thus no selector is needed |
---|
4240 | // or the appropriate key is given. |
---|
4241 | foreach ($this->link_styles as $key => $style) { |
---|
4242 | if ($style != '') { |
---|
4243 | switch ($key) { |
---|
4244 | case GESHI_LINK: |
---|
4245 | $stylesheet .= "{$selector}a:link {{$style}}\n"; |
---|
4246 | break; |
---|
4247 | case GESHI_HOVER: |
---|
4248 | $stylesheet .= "{$selector}a:hover {{$style}}\n"; |
---|
4249 | break; |
---|
4250 | case GESHI_ACTIVE: |
---|
4251 | $stylesheet .= "{$selector}a:active {{$style}}\n"; |
---|
4252 | break; |
---|
4253 | case GESHI_VISITED: |
---|
4254 | $stylesheet .= "{$selector}a:visited {{$style}}\n"; |
---|
4255 | break; |
---|
4256 | } |
---|
4257 | } |
---|
4258 | } |
---|
4259 | |
---|
4260 | // Header and footer |
---|
4261 | // note: neglect economy_mode, empty styles are meaningless |
---|
4262 | if ($this->header_content_style != '') { |
---|
4263 | $stylesheet .= "$selector.head {{$this->header_content_style}}\n"; |
---|
4264 | } |
---|
4265 | if ($this->footer_content_style != '') { |
---|
4266 | $stylesheet .= "$selector.foot {{$this->footer_content_style}}\n"; |
---|
4267 | } |
---|
4268 | |
---|
4269 | // Styles for important stuff |
---|
4270 | // note: neglect economy_mode, empty styles are meaningless |
---|
4271 | if ($this->important_styles != '') { |
---|
4272 | $stylesheet .= "$selector.imp {{$this->important_styles}}\n"; |
---|
4273 | } |
---|
4274 | |
---|
4275 | // Simple line number styles |
---|
4276 | if ((!$economy_mode || $this->line_numbers != GESHI_NO_LINE_NUMBERS) && $this->line_style1 != '') { |
---|
4277 | $stylesheet .= "{$selector}li, {$selector}.li1 {{$this->line_style1}}\n"; |
---|
4278 | } |
---|
4279 | if ((!$economy_mode || $this->line_numbers != GESHI_NO_LINE_NUMBERS) && $this->table_linenumber_style != '') { |
---|
4280 | $stylesheet .= "{$selector}.ln {{$this->table_linenumber_style}}\n"; |
---|
4281 | } |
---|
4282 | // If there is a style set for fancy line numbers, echo it out |
---|
4283 | if ((!$economy_mode || $this->line_numbers == GESHI_FANCY_LINE_NUMBERS) && $this->line_style2 != '') { |
---|
4284 | $stylesheet .= "{$selector}.li2 {{$this->line_style2}}\n"; |
---|
4285 | } |
---|
4286 | |
---|
4287 | // note: empty styles are meaningless |
---|
4288 | foreach ($this->language_data['STYLES']['KEYWORDS'] as $group => $styles) { |
---|
4289 | if ($styles != '' && (!$economy_mode || |
---|
4290 | (isset($this->lexic_permissions['KEYWORDS'][$group]) && |
---|
4291 | $this->lexic_permissions['KEYWORDS'][$group]))) { |
---|
4292 | $stylesheet .= "$selector.kw$group {{$styles}}\n"; |
---|
4293 | } |
---|
4294 | } |
---|
4295 | foreach ($this->language_data['STYLES']['COMMENTS'] as $group => $styles) { |
---|
4296 | if ($styles != '' && (!$economy_mode || |
---|
4297 | (isset($this->lexic_permissions['COMMENTS'][$group]) && |
---|
4298 | $this->lexic_permissions['COMMENTS'][$group]) || |
---|
4299 | (!empty($this->language_data['COMMENT_REGEXP']) && |
---|
4300 | !empty($this->language_data['COMMENT_REGEXP'][$group])))) { |
---|
4301 | $stylesheet .= "$selector.co$group {{$styles}}\n"; |
---|
4302 | } |
---|
4303 | } |
---|
4304 | foreach ($this->language_data['STYLES']['ESCAPE_CHAR'] as $group => $styles) { |
---|
4305 | if ($styles != '' && (!$economy_mode || $this->lexic_permissions['ESCAPE_CHAR'])) { |
---|
4306 | // NEW: since 1.0.8 we have to handle hardescapes |
---|
4307 | if ($group === 'HARD') { |
---|
4308 | $group = '_h'; |
---|
4309 | } |
---|
4310 | $stylesheet .= "$selector.es$group {{$styles}}\n"; |
---|
4311 | } |
---|
4312 | } |
---|
4313 | foreach ($this->language_data['STYLES']['BRACKETS'] as $group => $styles) { |
---|
4314 | if ($styles != '' && (!$economy_mode || $this->lexic_permissions['BRACKETS'])) { |
---|
4315 | $stylesheet .= "$selector.br$group {{$styles}}\n"; |
---|
4316 | } |
---|
4317 | } |
---|
4318 | foreach ($this->language_data['STYLES']['SYMBOLS'] as $group => $styles) { |
---|
4319 | if ($styles != '' && (!$economy_mode || $this->lexic_permissions['SYMBOLS'])) { |
---|
4320 | $stylesheet .= "$selector.sy$group {{$styles}}\n"; |
---|
4321 | } |
---|
4322 | } |
---|
4323 | foreach ($this->language_data['STYLES']['STRINGS'] as $group => $styles) { |
---|
4324 | if ($styles != '' && (!$economy_mode || $this->lexic_permissions['STRINGS'])) { |
---|
4325 | // NEW: since 1.0.8 we have to handle hardquotes |
---|
4326 | if ($group === 'HARD') { |
---|
4327 | $group = '_h'; |
---|
4328 | } |
---|
4329 | $stylesheet .= "$selector.st$group {{$styles}}\n"; |
---|
4330 | } |
---|
4331 | } |
---|
4332 | foreach ($this->language_data['STYLES']['NUMBERS'] as $group => $styles) { |
---|
4333 | if ($styles != '' && (!$economy_mode || $this->lexic_permissions['NUMBERS'])) { |
---|
4334 | $stylesheet .= "$selector.nu$group {{$styles}}\n"; |
---|
4335 | } |
---|
4336 | } |
---|
4337 | foreach ($this->language_data['STYLES']['METHODS'] as $group => $styles) { |
---|
4338 | if ($styles != '' && (!$economy_mode || $this->lexic_permissions['METHODS'])) { |
---|
4339 | $stylesheet .= "$selector.me$group {{$styles}}\n"; |
---|
4340 | } |
---|
4341 | } |
---|
4342 | // note: neglect economy_mode, empty styles are meaningless |
---|
4343 | foreach ($this->language_data['STYLES']['SCRIPT'] as $group => $styles) { |
---|
4344 | if ($styles != '') { |
---|
4345 | $stylesheet .= "$selector.sc$group {{$styles}}\n"; |
---|
4346 | } |
---|
4347 | } |
---|
4348 | foreach ($this->language_data['STYLES']['REGEXPS'] as $group => $styles) { |
---|
4349 | if ($styles != '' && (!$economy_mode || |
---|
4350 | (isset($this->lexic_permissions['REGEXPS'][$group]) && |
---|
4351 | $this->lexic_permissions['REGEXPS'][$group]))) { |
---|
4352 | if (is_array($this->language_data['REGEXPS'][$group]) && |
---|
4353 | array_key_exists(GESHI_CLASS, $this->language_data['REGEXPS'][$group])) { |
---|
4354 | $stylesheet .= "$selector."; |
---|
4355 | $stylesheet .= $this->language_data['REGEXPS'][$group][GESHI_CLASS]; |
---|
4356 | $stylesheet .= " {{$styles}}\n"; |
---|
4357 | } else { |
---|
4358 | $stylesheet .= "$selector.re$group {{$styles}}\n"; |
---|
4359 | } |
---|
4360 | } |
---|
4361 | } |
---|
4362 | // Styles for lines being highlighted extra |
---|
4363 | if (!$economy_mode || (count($this->highlight_extra_lines)!=count($this->highlight_extra_lines_styles))) { |
---|
4364 | $stylesheet .= "{$selector}.ln-xtra, {$selector}li.ln-xtra, {$selector}div.ln-xtra {{$this->highlight_extra_lines_style}}\n"; |
---|
4365 | } |
---|
4366 | $stylesheet .= "{$selector}span.xtra { display:block; }\n"; |
---|
4367 | foreach ($this->highlight_extra_lines_styles as $lineid => $linestyle) { |
---|
4368 | $stylesheet .= "{$selector}.lx$lineid, {$selector}li.lx$lineid, {$selector}div.lx$lineid {{$linestyle}}\n"; |
---|
4369 | } |
---|
4370 | |
---|
4371 | return $stylesheet; |
---|
4372 | } |
---|
4373 | |
---|
4374 | /** |
---|
4375 | * Get's the style that is used for the specified line |
---|
4376 | * |
---|
4377 | * @param int The line number information is requested for |
---|
4378 | * @access private |
---|
4379 | * @since 1.0.7.21 |
---|
4380 | */ |
---|
4381 | function get_line_style($line) { |
---|
4382 | //$style = null; |
---|
4383 | $style = null; |
---|
4384 | if (isset($this->highlight_extra_lines_styles[$line])) { |
---|
4385 | $style = $this->highlight_extra_lines_styles[$line]; |
---|
4386 | } else { // if no "extra" style assigned |
---|
4387 | $style = $this->highlight_extra_lines_style; |
---|
4388 | } |
---|
4389 | |
---|
4390 | return $style; |
---|
4391 | } |
---|
4392 | |
---|
4393 | /** |
---|
4394 | * this functions creates an optimized regular expression list |
---|
4395 | * of an array of strings. |
---|
4396 | * |
---|
4397 | * Example: |
---|
4398 | * <code>$list = array('faa', 'foo', 'foobar'); |
---|
4399 | * => string 'f(aa|oo(bar)?)'</code> |
---|
4400 | * |
---|
4401 | * @param $list array of (unquoted) strings |
---|
4402 | * @param $regexp_delimiter your regular expression delimiter, @see preg_quote() |
---|
4403 | * @return string for regular expression |
---|
4404 | * @author Milian Wolff <mail@milianw.de> |
---|
4405 | * @since 1.0.8 |
---|
4406 | * @access private |
---|
4407 | */ |
---|
4408 | function optimize_regexp_list($list, $regexp_delimiter = '/') { |
---|
4409 | $regex_chars = array('.', '\\', '+', '*', '?', '[', '^', ']', '$', |
---|
4410 | '(', ')', '{', '}', '=', '!', '<', '>', '|', ':', $regexp_delimiter); |
---|
4411 | sort($list); |
---|
4412 | $regexp_list = array(''); |
---|
4413 | $num_subpatterns = 0; |
---|
4414 | $list_key = 0; |
---|
4415 | |
---|
4416 | // the tokens which we will use to generate the regexp list |
---|
4417 | $tokens = array(); |
---|
4418 | $prev_keys = array(); |
---|
4419 | // go through all entries of the list and generate the token list |
---|
4420 | $cur_len = 0; |
---|
4421 | for ($i = 0, $i_max = count($list); $i < $i_max; ++$i) { |
---|
4422 | if ($cur_len > GESHI_MAX_PCRE_LENGTH) { |
---|
4423 | // seems like the length of this pcre is growing exorbitantly |
---|
4424 | $regexp_list[++$list_key] = $this->_optimize_regexp_list_tokens_to_string($tokens); |
---|
4425 | $num_subpatterns = substr_count($regexp_list[$list_key], '(?:'); |
---|
4426 | $tokens = array(); |
---|
4427 | $cur_len = 0; |
---|
4428 | } |
---|
4429 | $level = 0; |
---|
4430 | $entry = preg_quote((string) $list[$i], $regexp_delimiter); |
---|
4431 | $pointer = &$tokens; |
---|
4432 | // properly assign the new entry to the correct position in the token array |
---|
4433 | // possibly generate smaller common denominator keys |
---|
4434 | while (true) { |
---|
4435 | // get the common denominator |
---|
4436 | if (isset($prev_keys[$level])) { |
---|
4437 | if ($prev_keys[$level] == $entry) { |
---|
4438 | // this is a duplicate entry, skip it |
---|
4439 | continue 2; |
---|
4440 | } |
---|
4441 | $char = 0; |
---|
4442 | while (isset($entry[$char]) && isset($prev_keys[$level][$char]) |
---|
4443 | && $entry[$char] == $prev_keys[$level][$char]) { |
---|
4444 | ++$char; |
---|
4445 | } |
---|
4446 | if ($char > 0) { |
---|
4447 | // this entry has at least some chars in common with the current key |
---|
4448 | if ($char == strlen($prev_keys[$level])) { |
---|
4449 | // current key is totally matched, i.e. this entry has just some bits appended |
---|
4450 | $pointer = &$pointer[$prev_keys[$level]]; |
---|
4451 | } else { |
---|
4452 | // only part of the keys match |
---|
4453 | $new_key_part1 = substr($prev_keys[$level], 0, $char); |
---|
4454 | $new_key_part2 = substr($prev_keys[$level], $char); |
---|
4455 | |
---|
4456 | if (in_array($new_key_part1[0], $regex_chars) |
---|
4457 | || in_array($new_key_part2[0], $regex_chars)) { |
---|
4458 | // this is bad, a regex char as first character |
---|
4459 | $pointer[$entry] = array('' => true); |
---|
4460 | array_splice($prev_keys, $level, count($prev_keys), $entry); |
---|
4461 | $cur_len += strlen($entry); |
---|
4462 | continue; |
---|
4463 | } else { |
---|
4464 | // relocate previous tokens |
---|
4465 | $pointer[$new_key_part1] = array($new_key_part2 => $pointer[$prev_keys[$level]]); |
---|
4466 | unset($pointer[$prev_keys[$level]]); |
---|
4467 | $pointer = &$pointer[$new_key_part1]; |
---|
4468 | // recreate key index |
---|
4469 | array_splice($prev_keys, $level, count($prev_keys), array($new_key_part1, $new_key_part2)); |
---|
4470 | $cur_len += strlen($new_key_part2); |
---|
4471 | } |
---|
4472 | } |
---|
4473 | ++$level; |
---|
4474 | $entry = substr($entry, $char); |
---|
4475 | continue; |
---|
4476 | } |
---|
4477 | // else: fall trough, i.e. no common denominator was found |
---|
4478 | } |
---|
4479 | if ($level == 0 && !empty($tokens)) { |
---|
4480 | // we can dump current tokens into the string and throw them away afterwards |
---|
4481 | $new_entry = $this->_optimize_regexp_list_tokens_to_string($tokens); |
---|
4482 | $new_subpatterns = substr_count($new_entry, '(?:'); |
---|
4483 | if (GESHI_MAX_PCRE_SUBPATTERNS && $num_subpatterns + $new_subpatterns > GESHI_MAX_PCRE_SUBPATTERNS) { |
---|
4484 | $regexp_list[++$list_key] = $new_entry; |
---|
4485 | $num_subpatterns = $new_subpatterns; |
---|
4486 | } else { |
---|
4487 | if (!empty($regexp_list[$list_key])) { |
---|
4488 | $new_entry = '|' . $new_entry; |
---|
4489 | } |
---|
4490 | $regexp_list[$list_key] .= $new_entry; |
---|
4491 | $num_subpatterns += $new_subpatterns; |
---|
4492 | } |
---|
4493 | $tokens = array(); |
---|
4494 | $cur_len = 0; |
---|
4495 | } |
---|
4496 | // no further common denominator found |
---|
4497 | $pointer[$entry] = array('' => true); |
---|
4498 | array_splice($prev_keys, $level, count($prev_keys), $entry); |
---|
4499 | |
---|
4500 | $cur_len += strlen($entry); |
---|
4501 | break; |
---|
4502 | } |
---|
4503 | unset($list[$i]); |
---|
4504 | } |
---|
4505 | // make sure the last tokens get converted as well |
---|
4506 | $new_entry = $this->_optimize_regexp_list_tokens_to_string($tokens); |
---|
4507 | if (GESHI_MAX_PCRE_SUBPATTERNS && $num_subpatterns + substr_count($new_entry, '(?:') > GESHI_MAX_PCRE_SUBPATTERNS) { |
---|
4508 | $regexp_list[++$list_key] = $new_entry; |
---|
4509 | } else { |
---|
4510 | if (!empty($regexp_list[$list_key])) { |
---|
4511 | $new_entry = '|' . $new_entry; |
---|
4512 | } |
---|
4513 | $regexp_list[$list_key] .= $new_entry; |
---|
4514 | } |
---|
4515 | return $regexp_list; |
---|
4516 | } |
---|
4517 | /** |
---|
4518 | * this function creates the appropriate regexp string of an token array |
---|
4519 | * you should not call this function directly, @see $this->optimize_regexp_list(). |
---|
4520 | * |
---|
4521 | * @param &$tokens array of tokens |
---|
4522 | * @param $recursed bool to know wether we recursed or not |
---|
4523 | * @return string |
---|
4524 | * @author Milian Wolff <mail@milianw.de> |
---|
4525 | * @since 1.0.8 |
---|
4526 | * @access private |
---|
4527 | */ |
---|
4528 | function _optimize_regexp_list_tokens_to_string(&$tokens, $recursed = false) { |
---|
4529 | $list = ''; |
---|
4530 | foreach ($tokens as $token => $sub_tokens) { |
---|
4531 | $list .= $token; |
---|
4532 | $close_entry = isset($sub_tokens['']); |
---|
4533 | unset($sub_tokens['']); |
---|
4534 | if (!empty($sub_tokens)) { |
---|
4535 | $list .= '(?:' . $this->_optimize_regexp_list_tokens_to_string($sub_tokens, true) . ')'; |
---|
4536 | if ($close_entry) { |
---|
4537 | // make sub_tokens optional |
---|
4538 | $list .= '?'; |
---|
4539 | } |
---|
4540 | } |
---|
4541 | $list .= '|'; |
---|
4542 | } |
---|
4543 | if (!$recursed) { |
---|
4544 | // do some optimizations |
---|
4545 | // common trailing strings |
---|
4546 | // BUGGY! |
---|
4547 | //$list = preg_replace_callback('#(?<=^|\:|\|)\w+?(\w+)(?:\|.+\1)+(?=\|)#', create_function( |
---|
4548 | // '$matches', 'return "(?:" . preg_replace("#" . preg_quote($matches[1], "#") . "(?=\||$)#", "", $matches[0]) . ")" . $matches[1];'), $list); |
---|
4549 | // (?:p)? => p? |
---|
4550 | $list = preg_replace('#\(\?\:(.)\)\?#', '\1?', $list); |
---|
4551 | // (?:a|b|c|d|...)? => [abcd...]? |
---|
4552 | // TODO: a|bb|c => [ac]|bb |
---|
4553 | static $callback_2; |
---|
4554 | if (!isset($callback_2)) { |
---|
4555 | $callback_2 = create_function('$matches', 'return "[" . str_replace("|", "", $matches[1]) . "]";'); |
---|
4556 | } |
---|
4557 | $list = preg_replace_callback('#\(\?\:((?:.\|)+.)\)#', $callback_2, $list); |
---|
4558 | } |
---|
4559 | // return $list without trailing pipe |
---|
4560 | return substr($list, 0, -1); |
---|
4561 | } |
---|
4562 | } // End Class GeSHi |
---|
4563 | |
---|
4564 | |
---|
4565 | if (!function_exists('geshi_highlight')) { |
---|
4566 | /** |
---|
4567 | * Easy way to highlight stuff. Behaves just like highlight_string |
---|
4568 | * |
---|
4569 | * @param string The code to highlight |
---|
4570 | * @param string The language to highlight the code in |
---|
4571 | * @param string The path to the language files. You can leave this blank if you need |
---|
4572 | * as from version 1.0.7 the path should be automatically detected |
---|
4573 | * @param boolean Whether to return the result or to echo |
---|
4574 | * @return string The code highlighted (if $return is true) |
---|
4575 | * @since 1.0.2 |
---|
4576 | */ |
---|
4577 | function geshi_highlight($string, $language, $path = null, $return = false) { |
---|
4578 | $geshi = new GeSHi($string, $language, $path); |
---|
4579 | $geshi->set_header_type(GESHI_HEADER_NONE); |
---|
4580 | |
---|
4581 | if ($return) { |
---|
4582 | return '<code>' . $geshi->parse_code() . '</code>'; |
---|
4583 | } |
---|
4584 | |
---|
4585 | echo '<code>' . $geshi->parse_code() . '</code>'; |
---|
4586 | |
---|
4587 | if ($geshi->error()) { |
---|
4588 | return false; |
---|
4589 | } |
---|
4590 | return true; |
---|
4591 | } |
---|
4592 | } |
---|
4593 | |
---|
4594 | ?> |
---|