source: documentation/trunk/packages/dokuwiki-2011-05-25a/inc/pageutils.php@ 25027

Last change on this file since 25027 was 25027, checked in by jmt12, 12 years ago

Adding the packages directory, and within it a configured version of dokuwiki all ready to run
File size: 15.8 KB
Line 
1<?php
2/**
3 * Utilities for handling pagenames
4 *
5 * @license GPL 2 (http://www.gnu.org/licenses/gpl.html)
6 * @author Andreas Gohr <[email protected]>
7 * @todo Combine similar functions like {wiki,media,meta}FN()
8 */
9
10/**
11 * Fetch the an ID from request
12 *
13 * Uses either standard $_REQUEST variable or extracts it from
14 * the full request URI when userewrite is set to 2
15 *
16 * For $param='id' $conf['start'] is returned if no id was found.
17 * If the second parameter is true (default) the ID is cleaned.
18 *
19 * @author Andreas Gohr <[email protected]>
20 */
21function getID($param='id',$clean=true){
22 global $conf;
23
24 $id = isset($_REQUEST[$param]) ? $_REQUEST[$param] : null;
25
26 //construct page id from request URI
27 if(empty($id) && $conf['userewrite'] == 2){
28 $request = $_SERVER['REQUEST_URI'];
29 $script = '';
30
31 //get the script URL
32 if($conf['basedir']){
33 $relpath = '';
34 if($param != 'id') {
35 $relpath = 'lib/exe/';
36 }
37 $script = $conf['basedir'].$relpath.basename($_SERVER['SCRIPT_FILENAME']);
38
39 }elseif($_SERVER['PATH_INFO']){
40 $request = $_SERVER['PATH_INFO'];
41 }elseif($_SERVER['SCRIPT_NAME']){
42 $script = $_SERVER['SCRIPT_NAME'];
43 }elseif($_SERVER['DOCUMENT_ROOT'] && $_SERVER['SCRIPT_FILENAME']){
44 $script = preg_replace ('/^'.preg_quote($_SERVER['DOCUMENT_ROOT'],'/').'/','',
45 $_SERVER['SCRIPT_FILENAME']);
46 $script = '/'.$script;
47 }
48
49 //clean script and request (fixes a windows problem)
50 $script = preg_replace('/\/\/+/','/',$script);
51 $request = preg_replace('/\/\/+/','/',$request);
52
53 //remove script URL and Querystring to gain the id
54 if(preg_match('/^'.preg_quote($script,'/').'(.*)/',$request, $match)){
55 $id = preg_replace ('/\?.*/','',$match[1]);
56 }
57 $id = urldecode($id);
58 //strip leading slashes
59 $id = preg_replace('!^/+!','',$id);
60 }
61
62 // Namespace autolinking from URL
63 if(substr($id,-1) == ':' || ($conf['useslash'] && substr($id,-1) == '/')){
64 if(page_exists($id.$conf['start'])){
65 // start page inside namespace
66 $id = $id.$conf['start'];
67 }elseif(page_exists($id.noNS(cleanID($id)))){
68 // page named like the NS inside the NS
69 $id = $id.noNS(cleanID($id));
70 }elseif(page_exists($id)){
71 // page like namespace exists
72 $id = substr($id,0,-1);
73 }else{
74 // fall back to default
75 $id = $id.$conf['start'];
76 }
77 send_redirect(wl($id,'',true));
78 }
79
80 if($clean) $id = cleanID($id);
81 if(empty($id) && $param=='id') $id = $conf['start'];
82
83 return $id;
84}
85
86/**
87 * Remove unwanted chars from ID
88 *
89 * Cleans a given ID to only use allowed characters. Accented characters are
90 * converted to unaccented ones
91 *
92 * @author Andreas Gohr <[email protected]>
93 * @param string $raw_id The pageid to clean
94 * @param boolean $ascii Force ASCII
95 * @param boolean $media Allow leading or trailing _ for media files
96 */
97function cleanID($raw_id,$ascii=false,$media=false){
98 global $conf;
99 static $sepcharpat = null;
100
101 global $cache_cleanid;
102 $cache = & $cache_cleanid;
103
104 // check if it's already in the memory cache
105 if (isset($cache[(string)$raw_id])) {
106 return $cache[(string)$raw_id];
107 }
108
109 $sepchar = $conf['sepchar'];
110 if($sepcharpat == null) // build string only once to save clock cycles
111 $sepcharpat = '#\\'.$sepchar.'+#';
112
113 $id = trim((string)$raw_id);
114 $id = utf8_strtolower($id);
115
116 //alternative namespace seperator
117 $id = strtr($id,';',':');
118 if($conf['useslash']){
119 $id = strtr($id,'/',':');
120 }else{
121 $id = strtr($id,'/',$sepchar);
122 }
123
124 if($conf['deaccent'] == 2 || $ascii) $id = utf8_romanize($id);
125 if($conf['deaccent'] || $ascii) $id = utf8_deaccent($id,-1);
126
127 //remove specials
128 $id = utf8_stripspecials($id,$sepchar,'\*');
129
130 if($ascii) $id = utf8_strip($id);
131
132 //clean up
133 $id = preg_replace($sepcharpat,$sepchar,$id);
134 $id = preg_replace('#:+#',':',$id);
135 $id = ($media ? trim($id,':.-') : trim($id,':._-'));
136 $id = preg_replace('#:[:\._\-]+#',':',$id);
137
138 $cache[(string)$raw_id] = $id;
139 return($id);
140}
141
142/**
143 * Return namespacepart of a wiki ID
144 *
145 * @author Andreas Gohr <[email protected]>
146 */
147function getNS($id){
148 $pos = strrpos((string)$id,':');
149 if($pos!==false){
150 return substr((string)$id,0,$pos);
151 }
152 return false;
153}
154
155/**
156 * Returns the ID without the namespace
157 *
158 * @author Andreas Gohr <[email protected]>
159 */
160function noNS($id) {
161 $pos = strrpos($id, ':');
162 if ($pos!==false) {
163 return substr($id, $pos+1);
164 } else {
165 return $id;
166 }
167}
168
169/**
170 * Returns the current namespace
171 *
172 * @author Nathan Fritz <[email protected]>
173 */
174function curNS($id) {
175 return noNS(getNS($id));
176}
177
178/**
179 * Returns the ID without the namespace or current namespace for 'start' pages
180 *
181 * @author Nathan Fritz <[email protected]>
182 */
183function noNSorNS($id) {
184 global $conf;
185
186 $p = noNS($id);
187 if ($p == $conf['start'] || $p == false) {
188 $p = curNS($id);
189 if ($p == false) {
190 return $conf['start'];
191 }
192 }
193 return $p;
194}
195
196/**
197 * Creates a XHTML valid linkid from a given headline title
198 *
199 * @param string $title The headline title
200 * @param array $check Existing IDs (title => number)
201 * @author Andreas Gohr <[email protected]>
202 */
203function sectionID($title,&$check) {
204 $title = str_replace(array(':','.'),'',cleanID($title));
205 $new = ltrim($title,'0123456789_-');
206 if(empty($new)){
207 $title = 'section'.preg_replace('/[^0-9]+/','',$title); //keep numbers from headline
208 }else{
209 $title = $new;
210 }
211
212 if(is_array($check)){
213 // make sure tiles are unique
214 if (!array_key_exists ($title,$check)) {
215 $check[$title] = 0;
216 } else {
217 $title .= ++ $check[$title];
218 }
219 }
220
221 return $title;
222}
223
224
225/**
226 * Wiki page existence check
227 *
228 * parameters as for wikiFN
229 *
230 * @author Chris Smith <[email protected]>
231 */
232function page_exists($id,$rev='',$clean=true) {
233 return @file_exists(wikiFN($id,$rev,$clean));
234}
235
236/**
237 * returns the full path to the datafile specified by ID and optional revision
238 *
239 * The filename is URL encoded to protect Unicode chars
240 *
241 * @param $raw_id string id of wikipage
242 * @param $rev string page revision, empty string for current
243 * @param $clean bool flag indicating that $raw_id should be cleaned. Only set to false
244 * when $id is guaranteed to have been cleaned already.
245 *
246 * @author Andreas Gohr <[email protected]>
247 */
248function wikiFN($raw_id,$rev='',$clean=true){
249 global $conf;
250
251 global $cache_wikifn;
252 $cache = & $cache_wikifn;
253
254 if (isset($cache[$raw_id]) && isset($cache[$raw_id][$rev])) {
255 return $cache[$raw_id][$rev];
256 }
257
258 $id = $raw_id;
259
260 if ($clean) $id = cleanID($id);
261 $id = str_replace(':','/',$id);
262 if(empty($rev)){
263 $fn = $conf['datadir'].'/'.utf8_encodeFN($id).'.txt';
264 }else{
265 $fn = $conf['olddir'].'/'.utf8_encodeFN($id).'.'.$rev.'.txt';
266 if($conf['compression']){
267 //test for extensions here, we want to read both compressions
268 if (@file_exists($fn . '.gz')){
269 $fn .= '.gz';
270 }else if(@file_exists($fn . '.bz2')){
271 $fn .= '.bz2';
272 }else{
273 //file doesnt exist yet, so we take the configured extension
274 $fn .= '.' . $conf['compression'];
275 }
276 }
277 }
278
279 if (!isset($cache[$raw_id])) { $cache[$raw_id] = array(); }
280 $cache[$raw_id][$rev] = $fn;
281 return $fn;
282}
283
284/**
285 * Returns the full path to the file for locking the page while editing.
286 *
287 * @author Ben Coburn <[email protected]>
288 */
289function wikiLockFN($id) {
290 global $conf;
291 return $conf['lockdir'].'/'.md5(cleanID($id)).'.lock';
292}
293
294
295/**
296 * returns the full path to the meta file specified by ID and extension
297 *
298 * The filename is URL encoded to protect Unicode chars
299 *
300 * @author Steven Danz <[email protected]>
301 */
302function metaFN($id,$ext){
303 global $conf;
304 $id = cleanID($id);
305 $id = str_replace(':','/',$id);
306 $fn = $conf['metadir'].'/'.utf8_encodeFN($id).$ext;
307 return $fn;
308}
309
310/**
311 * returns an array of full paths to all metafiles of a given ID
312 *
313 * @author Esther Brunner <[email protected]>
314 * @author Michael Hamann <[email protected]>
315 */
316function metaFiles($id){
317 $basename = metaFN($id, '');
318 $files = glob($basename.'.*', GLOB_MARK);
319 // filter files like foo.bar.meta when $id == 'foo'
320 return $files ? preg_grep('/^'.preg_quote($basename, '/').'\.[^.\/]*$/u', $files) : array();
321}
322
323/**
324 * returns the full path to the mediafile specified by ID
325 *
326 * The filename is URL encoded to protect Unicode chars
327 *
328 * @author Andreas Gohr <[email protected]>
329 */
330function mediaFN($id){
331 global $conf;
332 $id = cleanID($id);
333 $id = str_replace(':','/',$id);
334 $fn = $conf['mediadir'].'/'.utf8_encodeFN($id);
335 return $fn;
336}
337
338/**
339 * Returns the full filepath to a localized textfile if local
340 * version isn't found the english one is returned
341 *
342 * @author Andreas Gohr <[email protected]>
343 */
344function localeFN($id){
345 global $conf;
346 $file = DOKU_CONF.'/lang/'.$conf['lang'].'/'.$id.'.txt';
347 if(!@file_exists($file)){
348 $file = DOKU_INC.'inc/lang/'.$conf['lang'].'/'.$id.'.txt';
349 if(!@file_exists($file)){
350 //fall back to english
351 $file = DOKU_INC.'inc/lang/en/'.$id.'.txt';
352 }
353 }
354 return $file;
355}
356
357/**
358 * Resolve relative paths in IDs
359 *
360 * Do not call directly use resolve_mediaid or resolve_pageid
361 * instead
362 *
363 * Partyly based on a cleanPath function found at
364 * http://www.php.net/manual/en/function.realpath.php#57016
365 *
366 * @author <bart at mediawave dot nl>
367 */
368function resolve_id($ns,$id,$clean=true){
369 global $conf;
370
371 // some pre cleaning for useslash:
372 if($conf['useslash']) $id = str_replace('/',':',$id);
373
374 // if the id starts with a dot we need to handle the
375 // relative stuff
376 if($id{0} == '.'){
377 // normalize initial dots without a colon
378 $id = preg_replace('/^(\.+)(?=[^:\.])/','\1:',$id);
379 // prepend the current namespace
380 $id = $ns.':'.$id;
381
382 // cleanup relatives
383 $result = array();
384 $pathA = explode(':', $id);
385 if (!$pathA[0]) $result[] = '';
386 foreach ($pathA AS $key => $dir) {
387 if ($dir == '..') {
388 if (end($result) == '..') {
389 $result[] = '..';
390 } elseif (!array_pop($result)) {
391 $result[] = '..';
392 }
393 } elseif ($dir && $dir != '.') {
394 $result[] = $dir;
395 }
396 }
397 if (!end($pathA)) $result[] = '';
398 $id = implode(':', $result);
399 }elseif($ns !== false && strpos($id,':') === false){
400 //if link contains no namespace. add current namespace (if any)
401 $id = $ns.':'.$id;
402 }
403
404 if($clean) $id = cleanID($id);
405 return $id;
406}
407
408/**
409 * Returns a full media id
410 *
411 * @author Andreas Gohr <[email protected]>
412 */
413function resolve_mediaid($ns,&$page,&$exists){
414 $page = resolve_id($ns,$page);
415 $file = mediaFN($page);
416 $exists = @file_exists($file);
417}
418
419/**
420 * Returns a full page id
421 *
422 * @author Andreas Gohr <[email protected]>
423 */
424function resolve_pageid($ns,&$page,&$exists){
425 global $conf;
426 global $ID;
427 $exists = false;
428
429 //empty address should point to current page
430 if ($page === "") {
431 $page = $ID;
432 }
433
434 //keep hashlink if exists then clean both parts
435 if (strpos($page,'#')) {
436 list($page,$hash) = explode('#',$page,2);
437 } else {
438 $hash = '';
439 }
440 $hash = cleanID($hash);
441 $page = resolve_id($ns,$page,false); // resolve but don't clean, yet
442
443 // get filename (calls clean itself)
444 $file = wikiFN($page);
445
446 // if ends with colon or slash we have a namespace link
447 if(in_array(substr($page,-1), array(':', ';')) ||
448 ($conf['useslash'] && substr($page,-1) == '/')){
449 if(page_exists($page.$conf['start'])){
450 // start page inside namespace
451 $page = $page.$conf['start'];
452 $exists = true;
453 }elseif(page_exists($page.noNS(cleanID($page)))){
454 // page named like the NS inside the NS
455 $page = $page.noNS(cleanID($page));
456 $exists = true;
457 }elseif(page_exists($page)){
458 // page like namespace exists
459 $page = $page;
460 $exists = true;
461 }else{
462 // fall back to default
463 $page = $page.$conf['start'];
464 }
465 }else{
466 //check alternative plural/nonplural form
467 if(!@file_exists($file)){
468 if( $conf['autoplural'] ){
469 if(substr($page,-1) == 's'){
470 $try = substr($page,0,-1);
471 }else{
472 $try = $page.'s';
473 }
474 if(page_exists($try)){
475 $page = $try;
476 $exists = true;
477 }
478 }
479 }else{
480 $exists = true;
481 }
482 }
483
484 // now make sure we have a clean page
485 $page = cleanID($page);
486
487 //add hash if any
488 if(!empty($hash)) $page .= '#'.$hash;
489}
490
491/**
492 * Returns the name of a cachefile from given data
493 *
494 * The needed directory is created by this function!
495 *
496 * @author Andreas Gohr <[email protected]>
497 *
498 * @param string $data This data is used to create a unique md5 name
499 * @param string $ext This is appended to the filename if given
500 * @return string The filename of the cachefile
501 */
502function getCacheName($data,$ext=''){
503 global $conf;
504 $md5 = md5($data);
505 $file = $conf['cachedir'].'/'.$md5{0}.'/'.$md5.$ext;
506 io_makeFileDir($file);
507 return $file;
508}
509
510/**
511 * Checks a pageid against $conf['hidepages']
512 *
513 * @author Andreas Gohr <[email protected]>
514 */
515function isHiddenPage($id){
516 global $conf;
517 global $ACT;
518 if(empty($conf['hidepages'])) return false;
519 if($ACT == 'admin') return false;
520
521 if(preg_match('/'.$conf['hidepages'].'/ui',':'.$id)){
522 return true;
523 }
524 return false;
525}
526
527/**
528 * Reverse of isHiddenPage
529 *
530 * @author Andreas Gohr <[email protected]>
531 */
532function isVisiblePage($id){
533 return !isHiddenPage($id);
534}
535
536/**
537 * Format an id for output to a user
538 *
539 * Namespaces are denoted by a trailing “:*”. The root namespace is
540 * “*”. Output is escaped.
541 *
542 * @author Adrian Lang <[email protected]>
543 */
544
545function prettyprint_id($id) {
546 if (!$id || $id === ':') {
547 return '*';
548 }
549 if ((substr($id, -1, 1) === ':')) {
550 $id .= '*';
551 }
552 return hsc($id);
553}
554
555/**
556 * Encode a UTF-8 filename to use on any filesystem
557 *
558 * Uses the 'fnencode' option to determine encoding
559 *
560 * When the second parameter is true the string will
561 * be encoded only if non ASCII characters are detected -
562 * This makes it safe to run it multiple times on the
563 * same string (default is true)
564 *
565 * @author Andreas Gohr <[email protected]>
566 * @see urlencode
567 */
568function utf8_encodeFN($file,$safe=true){
569 global $conf;
570 if($conf['fnencode'] == 'utf-8') return $file;
571
572 if($safe && preg_match('#^[a-zA-Z0-9/_\-\.%]+$#',$file)){
573 return $file;
574 }
575
576 if($conf['fnencode'] == 'safe'){
577 return SafeFN::encode($file);
578 }
579
580 $file = urlencode($file);
581 $file = str_replace('%2F','/',$file);
582 return $file;
583}
584
585/**
586 * Decode a filename back to UTF-8
587 *
588 * Uses the 'fnencode' option to determine encoding
589 *
590 * @author Andreas Gohr <[email protected]>
591 * @see urldecode
592 */
593function utf8_decodeFN($file){
594 global $conf;
595 if($conf['fnencode'] == 'utf-8') return $file;
596
597 if($conf['fnencode'] == 'safe'){
598 return SafeFN::decode($file);
599 }
600
601 return urldecode($file);
602}
603
Note: See TracBrowser for help on using the repository browser.