xref: /plugin/siteexport/inc/functions.php (revision 4c9a39da61d6465fb1589b6d77b3c5161f21843f)
1<?php
2
3if(!defined('DOKU_PLUGIN')) die('meh');
4require_once(DOKU_PLUGIN.'siteexport/inc/settings.php');
5require_once(DOKU_PLUGIN.'siteexport/inc/debug.php');
6
7class siteexport_functions extends DokuWiki_Plugin
8{
9    public $debug = null;
10    public $settings = null;
11
12    public function siteexport_functions($init=true, $isAJAX=false)
13    {
14        if ( $init )
15        {
16            $this->debug = new siteexport_debug();
17            $this->debug->isAJAX = $isAJAX;
18
19            $this->settings = new settings_plugin_siteexport_settings($this);
20            $this->debug->message("Settings completed: zipFile", $this->settings->zipFile, 1);
21        }
22    }
23
24    public function getPluginName()
25    {
26        return 'siteexport';
27    }
28
29    public function downloadURL()
30    {
31        $params = array('cache' => 'nocache', 'siteexport' => $this->settings->pattern);
32
33        if ( $this->debug->debugLevel() < 5 ) {
34            // If debug, then debug!
35            $params['debug'] = $this->debug->debugLevel();
36        }
37
38        return ml($this->settings->origZipFile, $params, true, '&');
39    }
40
41    public function checkIfCacheFileExistsForFileWithPattern($file, $pattern)
42    {
43        if ( !@file_exists($file) )
44        {
45            // If the cache File does not exist, move the newly created one over ...
46            $this->debug->message("'{$file}' does not exist. Checking original ZipFile", null, 3 );
47            $newCacheFile = mediaFN($this->getSpecialExportFileName($this->settings->origZipFile, $pattern));
48
49            if ( !@file_exists($newCacheFile) )
50            {
51                $this->debug->message("The export must have gone wrong. The cached file does not exist.", array("pattern" => $pattern, "original File" => $this->settings->origZipFile, "expected cached file" => $newCacheFile), 3);
52            }
53
54            $status = io_rename($newCacheFile, $file);
55            $this->debug->message("had to move another original file over. Did it work? " . ($status ? 'Yes, it did.' : 'No, it did not.'), null, 2 );
56        } else {
57            $this->debug->message("The file does exist!", $file, 2 );
58        }
59    }
60
61
62    /**
63     * Returns an utf8 encoded Namespace for a Page and input Namespace
64     * @param $NS
65     * @param $PAGE
66     */
67    function getNamespaceFromID($NS, &$PAGE) {
68        global $conf;
69        // Check current page - if its an NS add the startpage
70        $clean = true;
71        resolve_pageid(getNS($NS), $NS, $clean);
72        if ( ! page_exists($NS) && array_pop(explode(':', $NS)) != strtolower($conf['start'] )) { // Compare to lowercase since clean lowers it.
73            $NS .= ':' . $conf['start'];
74            resolve_pageid(getNS($NS), $NS, $clean);
75        }
76
77        $PAGE = noNS($NS);
78        $NS = getNS($NS);
79
80        return utf8_encodeFN(str_replace(':', '/', $NS));
81    }
82
83    /**
84     * create a file name for the page
85     **/
86    public function getSiteName($ID, $overrideRewrite=false) {
87        global $conf;
88
89        if ( empty($ID) ) return false;
90
91        // Remove extensions
92        if ( $overrideRewrite ) {
93            $ID = preg_replace("#\.[^\.]+$#", '', $ID);
94        }
95
96        $url = $this->wl($this->cleanID($ID), null, true, null, null, $overrideRewrite); // this must be done with rewriting set to override
97        //$url = $this->wl($this->cleanID($ID), null, true); // this must be done with rewriting set to override
98        $uri = @parse_url($url);
99        if ( $uri['path'][0] == '/' ) {
100            $uri['path'] = substr($uri['path'], 1);
101        }
102
103        return $this->shortenName($uri['path'] . '.' . $this->settings->fileType);
104    }
105
106    /**
107     * get the Title for the page
108     **/
109    public function getSiteTitle($ID) {
110        if (useHeading('content') && $ID) {
111            $heading = p_get_first_heading($ID,true);
112            if ($heading) {
113                return $this->xmlEntities($heading);
114            }
115        }
116        return ucwords($this->xmlEntities(array_pop(explode(':', $ID))));
117    }
118
119    /**
120     * Encoding ()taken from DW - but without needing the renderer
121     **/
122    public function xmlEntities($string) {
123        return htmlspecialchars($string,ENT_QUOTES,'UTF-8');
124    }
125
126    /**
127     * Create name for the file inside the zip and the replacements
128     **/
129    function shortenName($NAME)
130    {
131        $NS = $this->settings->exportNamespace;
132        $NAME = preg_replace("%^" . DOKU_BASE . "%", "", $NAME);
133        $NAME = preg_replace("%^(_media/)?(" . $NS . "/)?%", "", $NAME);
134
135        $this->debug->message("Shortening file to '$NAME'", null, 1);
136        return $NAME;
137    }
138
139    /**
140     * Remove unwanted chars from ID
141     *
142     * Cleans a given ID to only use allowed characters. Accented characters are
143     * converted to unaccented ones
144     *
145     * @author Andreas Gohr <andi@splitbrain.org>
146     * @param  string  $raw_id    The pageid to clean
147     * @param  boolean $ascii     Force ASCII
148     * @param  boolean $media     Allow leading or trailing _ for media files
149     */
150    function cleanID($raw_id,$ascii=false,$media=false){
151        global $conf;
152        global $lang;
153        static $sepcharpat = null;
154
155        global $cache_cleanid;
156        $cache = & $cache_cleanid;
157
158        // check if it's already in the memory cache
159        if (isset($cache[(string)$raw_id])) {
160            return $cache[(string)$raw_id];
161        }
162
163        $sepchar = $conf['sepchar'];
164        if($sepcharpat == null) // build string only once to save clock cycles
165        $sepcharpat = '#\\'.$sepchar.'+#';
166
167        $id = trim((string)$raw_id);
168        //        $id = utf8_strtolower($id); // NO LowerCase for us!
169
170        //alternative namespace seperator
171        $id = strtr($id,';',':');
172        if($conf['useslash']){
173            $id = strtr($id,'/',':');
174        }else{
175            $id = strtr($id,'/',$sepchar);
176        }
177
178        if($conf['deaccent'] == 2 || $ascii) $id = utf8_romanize($id);
179        if($conf['deaccent'] || $ascii) $id = utf8_deaccent($id,-1);
180
181        // We want spaces to be preserved when they are in the link.
182        global $UTF8_SPECIAL_CHARS2;
183        $UTF8_SPECIAL_CHARS2_SAVE = (string)$UTF8_SPECIAL_CHARS2;
184        $UTF8_SPECIAL_CHARS2 = str_replace(' ', '', $UTF8_SPECIAL_CHARS2);
185
186        //remove specials
187        $id = utf8_stripspecials($id,$sepchar,'\*');
188        $UTF8_SPECIAL_CHARS2 = $UTF8_SPECIAL_CHARS2_SAVE;
189
190        if($ascii) $id = utf8_strip($id);
191
192        //clean up
193        $id = preg_replace($sepcharpat,$sepchar,$id);
194        $id = preg_replace('#:+#',':',$id);
195        $id = ($media ? trim($id,':.-') : trim($id,':._-'));
196        $id = preg_replace('#:[:\._\-]+#',':',$id);
197
198        $cache[(string)$raw_id] = $id;
199        return($id);
200    }
201
202
203    /**
204     * This builds a link to a wikipage - changed for internal use here
205     *
206     * It handles URL rewriting and adds additional parameter if
207     * given in $more
208     *
209     * @author Andreas Gohr <andi@splitbrain.org>
210     */
211
212    function wl($id='',$more='',$abs=false,$sep='&amp;', $IDexists=true, $overrideRewrite=false, $hadBase=false){
213        global $conf;
214
215        $this->debug->message("Starting to build WL-URL for '$id'", $more, 1);
216
217        if(is_array($more)){
218
219        	$intermediateMore = '';
220        	foreach( $more as $key => $value) {
221
222        		if ( strlen($intermediateMore) > 0 ) {
223	        		$intermediateMore .= $sep;
224        		}
225
226	        	if ( !is_array($value) ) {
227		        	$intermediateMore .= rawurlencode($key) . '=';
228		        	$intermediateMore .= rawurlencode($value);
229		        	continue;
230	        	}
231
232	        	foreach( $value as $val ) {
233	        		if ( strlen($intermediateMore) > 0 ) {
234		        		$intermediateMore .= $sep;
235	        		}
236
237		        	$intermediateMore .= rawurlencode($key) . '[]=';
238		        	$intermediateMore .= rawurlencode($val);
239	        	}
240        	}
241
242            $more = $intermediateMore;
243        }else{
244            $more = str_replace(',',$sep,$more);
245        }
246
247        $id = idfilter($id);
248
249        if($abs){
250            $xlink = DOKU_URL;
251            if ( !$IDexists && !$hadBase ) { // If the file does not exist, we have to remove the base. This link my be one to an parallel BASE.
252                $xlink = preg_replace('#' . DOKU_BASE . '$#', '', $xlink);
253            }
254        }else if ($IDexists || $hadBase) { // if the ID does exist, we may add the base.
255            $xlink = DOKU_BASE;
256        } else{
257            $xlink = "";
258        }
259
260        // $this->debug->message("internal WL function Before Replacing: '$xlink'", array(DOKU_REL, DOKU_URL, DOKU_BASE, $xlink), 2);
261        $xlink = preg_replace('#(?<!http:|https:)//+#','/', ($abs ? '' : '/') . "$xlink/"); // ensure slashes at beginning and ending, but strip doubles
262        $this->debug->message("'$xlink'", array(DOKU_REL, DOKU_URL, DOKU_BASE, $xlink), 2);
263
264        if ( $overrideRewrite ) {
265            $this->debug->message("Override enabled.", null, 1);
266            $id = strtr($id,':','/');
267
268            $xlink .= $id;
269            if($more) $xlink .= '?'.$more;
270        } else {
271            if($conf['userewrite'] == 2 ){
272                $xlink .= DOKU_SCRIPT.'/'.$id;
273                if($more) $xlink .= '?'.$more;
274            }elseif($conf['userewrite'] ){
275                $xlink .= $id;
276                if($more) $xlink .= '?'.$more;
277            }elseif($id){
278                $xlink .= DOKU_SCRIPT.'?id='.$id;
279                if($more) $xlink .= $sep.$more;
280            }else{
281                $xlink .= DOKU_SCRIPT;
282                if($more) $xlink .= '?'.$more;
283            }
284        }
285
286        $this->debug->message("internal WL function result: '$xlink'", null, 2);
287
288        return $xlink;
289    }
290
291    /**
292     * Create the export file name - this is the file where everything is being stored
293     * @param $FILE
294     * @param $PATTERN - additional pattern for re-using old files
295     */
296    public function getSpecialExportFileName($FILE, $PATTERN=null) {
297
298        if ( empty($FILE) )
299        {
300            $FILE = $this->settings->origZipFile;
301        }
302
303        if ( empty($PATTERN) && empty($this->settings->pattern) ){
304            $this->debug->message("Generating an internal md5 pattern. This will go wrong - and won't cache properly.", null, 3);
305            $PATTERN = md5(microtime(false));
306        }
307
308        // Set Pattern Global for other stuff
309        if ( empty($this->settings->pattern) ) {
310            $this->settings['pattern'] = $PATTERN;
311        } else {
312            $PATTERN = $this->settings->pattern;
313        }
314
315        $FA = explode('.', $FILE);
316        $EXT = array_pop($FA);
317        array_push($FA, 'auto');
318        array_push($FA, $PATTERN);
319        array_push($FA, $EXT);
320
321        $fileName = implode('.', $FA);
322        $this->debug->message("Export Filename for '$FILE' will be: '$fileName'", null, 2);
323        return $fileName;
324    }
325
326    public function getCacheFileNameForPattern($PATTERN = null)
327    {
328        if ( $PATTERN == null ) {
329            $PATTERN = $this->settings->pattern;
330        }
331
332        return getCacheName($this->getSpecialExportFileName($this->settings->origZipFile, $PATTERN), '.' . basename(mediaFN($this->settings->origZipFile)) );
333    }
334
335    function startRedirctProcess($counter) {
336        global $ID;
337
338        $URL = wl($ID);
339
340        $additionalParameters = $_REQUEST;
341        $additionalParameters['startcounter'] = $counter;
342        $additionalParameters['pattern'] = $this->settings->pattern;
343
344        unset($additionalParameters['id']);
345        unset($additionalParameters['u']);
346        unset($additionalParameters['p']);
347        unset($additionalParameters['r']);
348        unset($additionalParameters['http_credentials']);
349
350        $this->addAdditionalParametersToURL($URL, $additionalParameters);
351        $this->debug->message("Redirecting to '$URL'", null, 2);
352
353        send_redirect($URL);
354        exit(0); // Should not be reached, but anyways
355    }
356
357    /**
358     * Builds additional Parameters into the URL given
359     * @param $URL
360     * @param $newAdditionalParameters
361     */
362    function addAdditionalParametersToURL(&$URL, $newAdditionalParameters) {
363
364        // Add additionalParameters
365        if ( !empty($newAdditionalParameters) ) {
366            foreach($newAdditionalParameters as $key => $value ) {
367                if ( empty($key) || empty($value) ) { continue; }
368
369                $append = '';
370
371                if ( is_array($value) ) {
372                    foreach( array_values($value) as $aValue ) { // Array Handling
373                        $URL .= (strstr($URL, '?') ? '&' : '?') . $key . "[]=$aValue";
374                    }
375                } else {
376                    $append = "$key=$value";
377                    $URL .= empty($append) || strstr($URL, $append)  ? '' : (strstr($URL, '?') ? '&' : '?') . $append;
378                }
379
380            }
381        }
382    }
383
384    /**
385     * Cleans the wiki variables and returns a rebuild URL that has the new variables at hand
386     * @param $data
387     */
388    function prepare_POSTData($data)
389    {
390        $NS = !empty($data['ns']) ? $data['ns'] : $data['id'];
391
392        $this->removeWikiVariables($data);
393        $data['do'] = 'siteexport';
394        $additionalKeys = '';
395
396        ksort($data);
397
398        $this->debug->message("Prepared POST data:", $data, 1);
399
400        foreach( $data as $key => $value ) {
401
402            if ( !is_array($value) ) { continue; }
403            $this->debug->message("Found inner Array:", $value, 1);
404
405            asort($value);
406            foreach ( $value as $innerKey => $aValue )
407            {
408                if ( is_numeric($innerKey))
409                {
410                    $innerKey = '';
411                }
412
413                $additionalKeys .= "&$key" . "[$innerKey]=$aValue";
414            }
415
416            unset($data[$key]);
417        }
418
419        return wl($NS, $data, true, '&') . $additionalKeys;
420    }
421
422    /**
423     * Parses a String into a $_REQUEST Like variant. You have to tell if a decode of the values is needed
424     * @param $inputArray
425     * @param $decode
426     */
427    public function parseStringToRequestArray($inputArray, $decode=false)
428    {
429        global $plugin_controller;
430
431        $outputArray = $inputArray;
432        if ( !is_array($inputArray) )
433        {
434            $intermediate = str_replace("&amp;", "&", $inputArray);
435
436            $outputArray = array();
437            foreach( explode("&", $intermediate) as $param ) {
438                list($key, $value) = explode("=", $param, 2);
439
440                // This is needed if we do want to calculate $_REQUEST for a non HTTP-Request
441                if ( $decode)
442                {
443                    $value = urldecode($value);
444                }
445
446                if ( empty($key) ) { continue; } // Don't check on Value, because there may be only the key that should be preserved
447
448                if ( substr($key, -2) == '[]' ) {
449	                $key = substr($key, 0, -2);
450	                if ( !is_array($outputArray[$key]) ) {
451		                $outputArray[$key] = array();
452	                }
453
454                    array_push($outputArray[$key], $value); // Array Handling
455                } else {
456                    $outputArray[$key] = $value;
457                }
458            }
459        }
460
461        if ( !empty($outputArray['diPlu']) ) {
462
463            $allPlugins = array();
464            foreach($plugin_controller->getList(null,true) as $plugin ) {
465                // check for CSS or JS
466                if ( !file_exists(DOKU_PLUGIN."$plugin/script.js") && !file_exists(DOKU_PLUGIN."$p/style.css") ) { continue; }
467                $allPlugins[] = $plugin;
468            }
469
470            if ( count($outputArray['diPlu']) > (count($allPlugins) / 2) ) {
471                $outputArray['diInv'] = 1;
472                $outputArray['diPlu'] = array_diff($allPlugins, $outputArray['diPlu']);
473            }
474        }
475
476        return $outputArray;
477    }
478
479    /**
480     * Remove certain fields from the list.
481     * @param $removeArray
482     * @param $advanced
483     * @param $isString
484     */
485    function removeWikiVariables(&$removeArray, $advanced=false, $isString=false) {
486
487        $removeArray = $this->parseStringToRequestArray($removeArray);
488
489        // 2010-08-23 - If there is still the media set, retain the id for e.g. detail.php
490        if ( !isset($removeArray['media']) ) {
491            unset($removeArray['id']);
492        }
493
494        unset($removeArray['do']);
495        unset($removeArray['ns']);
496        unset($removeArray['call']);
497        unset($removeArray['sectok']);
498        unset($removeArray['rndval']);
499        unset($removeArray['tseed']);
500        unset($removeArray['http_credentials']);
501        unset($removeArray['u']);
502        unset($removeArray['p']);
503        unset($removeArray['r']);
504        unset($removeArray['base']);
505        unset($removeArray['siteexport']);
506        unset($removeArray['DokuWiki']);
507        unset($removeArray['cronOverwriteExisting']);
508
509        if ( $removeArray['renderer'] == 'xhtml' ) {
510            $removeArray['do'] = 'export_' . $removeArray['renderer'];
511            unset($removeArray['renderer']);
512        }
513
514        // Keep custom options
515        if ( is_array($removeArray['customoptionname']) && is_array($removeArray['customoptionvalue']) && count($removeArray['customoptionname']) == count($removeArray['customoptionvalue']) )
516        {
517            for( $index=0; $index<count($removeArray['customoptionname']); $index++)
518            {
519                $removeArray[$removeArray['customoptionname'][$index]] = $removeArray['customoptionvalue'][$index];
520            }
521	        unset($removeArray['customoptionname']);
522	        unset($removeArray['customoptionvalue']);
523        }
524
525        if ( $advanced ) {
526            if ( $removeArray['renderer'] != 'xhtml' && !empty($removeArray['renderer']) ) {
527                $removeArray['do'] = 'export_' . $removeArray['renderer'];
528            }
529
530            // 2010-08-25 - Need fakeMedia for some _detail cases with rewrite = 2
531            if ( isset($removeArray['fakeMedia'])  ) {
532                unset($removeArray['media']);
533                unset($removeArray['fakeMedia']);
534            }
535
536            /* remove internal params */
537            unset($removeArray['ens']);
538            unset($removeArray['renderer']);
539            unset($removeArray['site']);
540            unset($removeArray['namespace']);
541            unset($removeArray['exportbody']);
542            unset($removeArray['addParams']);
543            unset($removeArray['template']);
544            unset($removeArray['eclipseDocZip']);
545            unset($removeArray['useTocFile']);
546            unset($removeArray['JavaHelpDocZip']);
547            unset($removeArray['depth']);
548            unset($removeArray['depthType']);
549            unset($removeArray['startcounter']);
550            unset($removeArray['pattern']);
551            unset($removeArray['TOCMapWithoutTranslation']);
552			// unset($removeArray['disableCache']);
553			unset($removeArray['debug']);
554        }
555
556        if ( $isString && is_array($removeArray) ) {
557            $intermediate = $removeArray;
558            $removeArray = array();
559
560            foreach ( $intermediate as $key => $value ) {
561                if ( is_array($value) ) {
562                    foreach( array_values($value) as $aValue ) { // Array Handling
563                        $removeArray[] = $key . "[]=$aValue";
564                    }
565                } else {
566                    $value = trim($value);
567
568                    $removeArray[] = "$key" . ( ((empty($value) && intval($value) !== 0)) || $value == '' ? '' : "=$value" ); // If the Value is empty, the Key must be preserved
569                }
570            }
571
572            //$removeArray = implode( ($this->settings->fileType == 'pdf' ? "&" : "&amp;"), $removeArray);
573            $removeArray = implode( "&", $removeArray); // The &amp; made problems with the HTTPClient / Apache. It should not be a problem to have &
574        }
575    }
576
577    /**
578     * returns a hashed name for the parameters
579     * @param $parameters
580     */
581    public function cronJobNameForParameters($parameters)
582    {
583        return md5($parameters);
584    }
585
586    /**
587     * Takes an URL and transforms it into the path+query part
588     * Used several times, e.g. for genering the hash for the cache file
589     * @param $url
590     */
591    public function urlToPathAndParams($url)
592    {
593        $query = parse_url($url, PHP_URL_QUERY);
594        $path = preg_replace(":^".DOKU_REL.":", "", parse_url($url, PHP_URL_PATH));
595        return "{$path}?{$query}";
596    }
597
598    /**
599     * Transforms an $_REQUEST into a Hash that can be used for cron and cache file
600     * @param $request
601     */
602    public function requestParametersToCacheHash($request)
603    {
604        $params = $this->urlToPathAndParams($this->prepare_POSTData($request));
605        $this->debug->message("Calculated the following Cache Hash URL: ", $params, 2);
606        return $this->cronJobNameForParameters($params);
607    }
608
609    /**
610     * Check a replaceID against a baseID - and make the replaceID relative against it
611     * @param $replaceID - ID which will be made relative if needed
612     * @param $baseID - ID which is the reference to be made relative against
613     */
614    public function getRelativeURL($replaceID, $baseID, $separator=':')
615    {
616        $origReplaceID = $replaceID;
617
618        $replaceTmp = cleanID($replaceID);
619        $file = noNS($replaceTmp);
620
621        $replaceID = getNS($replaceTmp);
622        $baseID = getNS($baseID);
623
624        $replaceParts = explode($separator, $replaceID);
625        $baseParts = explode($separator, $baseID);
626        $exportNSParts = explode(':', cleanID($this->settings->exportNamespace));
627
628        $newBase = array();
629
630        foreach($exportNSParts as $exportNS)
631        {
632            if ( $replaceParts[0] == $exportNS && $baseParts[0] == $exportNS )
633            {
634                array_shift($replaceParts);
635                array_shift($baseParts);
636                array_shift($exportNSParts);
637            }
638            else {
639                // Nothing is matching anymore
640                break;
641            }
642        }
643
644        $i = count($exportNSParts);
645        $this->debug->message("Checking", array('current extra removing amount'=>$i,'replace'=>$replaceParts,'base'=>$baseParts,'exportNSParts'=>$exportNSParts),1);
646
647        // Now if there is just one item in the ens left and it matches the base, but not the replace, this miiiiiight be the case we want.
648        if ( count($exportNSParts) == 1 && $exportNSParts[0] == $baseParts[0] )
649        {
650            array_shift($replaceParts);
651            $newBase = implode($separator, $replaceParts);
652
653            if ( substr($newBase, -1) != $separator )
654            {
655                $newBase .= $separator;
656            }
657
658            $this->debug->message("new Base: ", $newBase, 1);
659
660            // Now check from the beginning ...
661            return $newBase . $file;
662        }
663
664        return $origReplaceID;
665    }
666}
667
668?>