1 <?php
  2 
  3 namespace Icybee\Modules\Search;
  4 
  5 use ICanBoogie\I18n;
  6 
  7 use Brickrouge\Pager;
  8 
  9 if (!defined('PREG_CLASS_SEARCH_EXCLUDE'))
 10 {
 11     /**
 12      * Matches Unicode character classes to exclude from the search index.
 13      *
 14      * See: http://www.unicode.org/Public/UNIDATA/UCD.html#General_Category_Values
 15      *
 16      * The index only contains the following character classes:
 17      * Lu     Letter, Uppercase
 18      * Ll     Letter, Lowercase
 19      * Lt     Letter, Titlecase
 20      * Lo     Letter, Other
 21      * Nd     Number, Decimal Digit
 22      * No     Number, Other
 23      */
 24     define('PREG_CLASS_SEARCH_EXCLUDE',
 25     '\x{0}-\x{2f}\x{3a}-\x{40}\x{5b}-\x{60}\x{7b}-\x{bf}\x{d7}\x{f7}\x{2b0}-'.
 26     '\x{385}\x{387}\x{3f6}\x{482}-\x{489}\x{559}-\x{55f}\x{589}-\x{5c7}\x{5f3}-'.
 27     '\x{61f}\x{640}\x{64b}-\x{65e}\x{66a}-\x{66d}\x{670}\x{6d4}\x{6d6}-\x{6ed}'.
 28     '\x{6fd}\x{6fe}\x{700}-\x{70f}\x{711}\x{730}-\x{74a}\x{7a6}-\x{7b0}\x{901}-'.
 29     '\x{903}\x{93c}\x{93e}-\x{94d}\x{951}-\x{954}\x{962}-\x{965}\x{970}\x{981}-'.
 30     '\x{983}\x{9bc}\x{9be}-\x{9cd}\x{9d7}\x{9e2}\x{9e3}\x{9f2}-\x{a03}\x{a3c}-'.
 31     '\x{a4d}\x{a70}\x{a71}\x{a81}-\x{a83}\x{abc}\x{abe}-\x{acd}\x{ae2}\x{ae3}'.
 32     '\x{af1}-\x{b03}\x{b3c}\x{b3e}-\x{b57}\x{b70}\x{b82}\x{bbe}-\x{bd7}\x{bf0}-'.
 33     '\x{c03}\x{c3e}-\x{c56}\x{c82}\x{c83}\x{cbc}\x{cbe}-\x{cd6}\x{d02}\x{d03}'.
 34     '\x{d3e}-\x{d57}\x{d82}\x{d83}\x{dca}-\x{df4}\x{e31}\x{e34}-\x{e3f}\x{e46}-'.
 35     '\x{e4f}\x{e5a}\x{e5b}\x{eb1}\x{eb4}-\x{ebc}\x{ec6}-\x{ecd}\x{f01}-\x{f1f}'.
 36     '\x{f2a}-\x{f3f}\x{f71}-\x{f87}\x{f90}-\x{fd1}\x{102c}-\x{1039}\x{104a}-'.
 37     '\x{104f}\x{1056}-\x{1059}\x{10fb}\x{10fc}\x{135f}-\x{137c}\x{1390}-\x{1399}'.
 38     '\x{166d}\x{166e}\x{1680}\x{169b}\x{169c}\x{16eb}-\x{16f0}\x{1712}-\x{1714}'.
 39     '\x{1732}-\x{1736}\x{1752}\x{1753}\x{1772}\x{1773}\x{17b4}-\x{17db}\x{17dd}'.
 40     '\x{17f0}-\x{180e}\x{1843}\x{18a9}\x{1920}-\x{1945}\x{19b0}-\x{19c0}\x{19c8}'.
 41     '\x{19c9}\x{19de}-\x{19ff}\x{1a17}-\x{1a1f}\x{1d2c}-\x{1d61}\x{1d78}\x{1d9b}-'.
 42     '\x{1dc3}\x{1fbd}\x{1fbf}-\x{1fc1}\x{1fcd}-\x{1fcf}\x{1fdd}-\x{1fdf}\x{1fed}-'.
 43     '\x{1fef}\x{1ffd}-\x{2070}\x{2074}-\x{207e}\x{2080}-\x{2101}\x{2103}-\x{2106}'.
 44     '\x{2108}\x{2109}\x{2114}\x{2116}-\x{2118}\x{211e}-\x{2123}\x{2125}\x{2127}'.
 45     '\x{2129}\x{212e}\x{2132}\x{213a}\x{213b}\x{2140}-\x{2144}\x{214a}-\x{2b13}'.
 46     '\x{2ce5}-\x{2cff}\x{2d6f}\x{2e00}-\x{3005}\x{3007}-\x{303b}\x{303d}-\x{303f}'.
 47     '\x{3099}-\x{309e}\x{30a0}\x{30fb}\x{30fd}\x{30fe}\x{3190}-\x{319f}\x{31c0}-'.
 48     '\x{31cf}\x{3200}-\x{33ff}\x{4dc0}-\x{4dff}\x{a015}\x{a490}-\x{a716}\x{a802}'.
 49     '\x{a806}\x{a80b}\x{a823}-\x{a82b}\x{d800}-\x{f8ff}\x{fb1e}\x{fb29}\x{fd3e}'.
 50     '\x{fd3f}\x{fdfc}-\x{fe6b}\x{feff}-\x{ff0f}\x{ff1a}-\x{ff20}\x{ff3b}-\x{ff40}'.
 51     '\x{ff5b}-\x{ff65}\x{ff70}\x{ff9e}\x{ff9f}\x{ffe0}-\x{fffd}');
 52 
 53     /**
 54      * Matches all 'N' Unicode character classes (numbers)
 55      */
 56     define('PREG_CLASS_NUMBERS',
 57     '\x{30}-\x{39}\x{b2}\x{b3}\x{b9}\x{bc}-\x{be}\x{660}-\x{669}\x{6f0}-\x{6f9}'.
 58     '\x{966}-\x{96f}\x{9e6}-\x{9ef}\x{9f4}-\x{9f9}\x{a66}-\x{a6f}\x{ae6}-\x{aef}'.
 59     '\x{b66}-\x{b6f}\x{be7}-\x{bf2}\x{c66}-\x{c6f}\x{ce6}-\x{cef}\x{d66}-\x{d6f}'.
 60     '\x{e50}-\x{e59}\x{ed0}-\x{ed9}\x{f20}-\x{f33}\x{1040}-\x{1049}\x{1369}-'.
 61     '\x{137c}\x{16ee}-\x{16f0}\x{17e0}-\x{17e9}\x{17f0}-\x{17f9}\x{1810}-\x{1819}'.
 62     '\x{1946}-\x{194f}\x{2070}\x{2074}-\x{2079}\x{2080}-\x{2089}\x{2153}-\x{2183}'.
 63     '\x{2460}-\x{249b}\x{24ea}-\x{24ff}\x{2776}-\x{2793}\x{3007}\x{3021}-\x{3029}'.
 64     '\x{3038}-\x{303a}\x{3192}-\x{3195}\x{3220}-\x{3229}\x{3251}-\x{325f}\x{3280}-'.
 65     '\x{3289}\x{32b1}-\x{32bf}\x{ff10}-\x{ff19}');
 66 
 67     /**
 68      * Matches all 'P' Unicode character classes (punctuation)
 69      */
 70     define('PREG_CLASS_PUNCTUATION',
 71     '\x{21}-\x{23}\x{25}-\x{2a}\x{2c}-\x{2f}\x{3a}\x{3b}\x{3f}\x{40}\x{5b}-\x{5d}'.
 72     '\x{5f}\x{7b}\x{7d}\x{a1}\x{ab}\x{b7}\x{bb}\x{bf}\x{37e}\x{387}\x{55a}-\x{55f}'.
 73     '\x{589}\x{58a}\x{5be}\x{5c0}\x{5c3}\x{5f3}\x{5f4}\x{60c}\x{60d}\x{61b}\x{61f}'.
 74     '\x{66a}-\x{66d}\x{6d4}\x{700}-\x{70d}\x{964}\x{965}\x{970}\x{df4}\x{e4f}'.
 75     '\x{e5a}\x{e5b}\x{f04}-\x{f12}\x{f3a}-\x{f3d}\x{f85}\x{104a}-\x{104f}\x{10fb}'.
 76     '\x{1361}-\x{1368}\x{166d}\x{166e}\x{169b}\x{169c}\x{16eb}-\x{16ed}\x{1735}'.
 77     '\x{1736}\x{17d4}-\x{17d6}\x{17d8}-\x{17da}\x{1800}-\x{180a}\x{1944}\x{1945}'.
 78     '\x{2010}-\x{2027}\x{2030}-\x{2043}\x{2045}-\x{2051}\x{2053}\x{2054}\x{2057}'.
 79     '\x{207d}\x{207e}\x{208d}\x{208e}\x{2329}\x{232a}\x{23b4}-\x{23b6}\x{2768}-'.
 80     '\x{2775}\x{27e6}-\x{27eb}\x{2983}-\x{2998}\x{29d8}-\x{29db}\x{29fc}\x{29fd}'.
 81     '\x{3001}-\x{3003}\x{3008}-\x{3011}\x{3014}-\x{301f}\x{3030}\x{303d}\x{30a0}'.
 82     '\x{30fb}\x{fd3e}\x{fd3f}\x{fe30}-\x{fe52}\x{fe54}-\x{fe61}\x{fe63}\x{fe68}'.
 83     '\x{fe6a}\x{fe6b}\x{ff01}-\x{ff03}\x{ff05}-\x{ff0a}\x{ff0c}-\x{ff0f}\x{ff1a}'.
 84     '\x{ff1b}\x{ff1f}\x{ff20}\x{ff3b}-\x{ff3d}\x{ff3f}\x{ff5b}\x{ff5d}\x{ff5f}-'.
 85     '\x{ff65}');
 86 
 87     /**
 88      * Matches all CJK characters that are candidates for auto-splitting
 89      * (Chinese, Japanese, Korean).
 90      * Contains kana and BMP ideographs.
 91      */
 92     define('PREG_CLASS_CJK', '\x{3041}-\x{30ff}\x{31f0}-\x{31ff}\x{3400}-\x{4db5}'.
 93     '\x{4e00}-\x{9fbb}\x{f900}-\x{fad9}');
 94 }
 95 
 96 function query_google($search, $position, $limit)
 97 {
 98     $site = 'atalian.com';
 99     $query = 'http://ajax.googleapis.com/ajax/services/search/web?' . http_build_query
100     (
101         array
102         (
103             'q' => $search . ' site:' . $site,
104             'start' => $position * $limit,
105             'v' => '1.0'
106         )
107 
108         + array
109         (
110             'gl' => 'fr',
111             'hl' => 'fr',
112             'rsz' => 'large'
113         )
114     );
115 
116     $rc = file_get_contents($query);
117 
118     $response = json_decode($rc)->responseData;
119 
120     $entries = array();
121     $count = $response->cursor->estimatedResultCount;
122 
123     if ($response->results)
124     {
125         foreach ($response->results as $result)
126         {
127             $shortUrl = $result->unescapedUrl;
128             $shortUrl = substr($shortUrl, strpos($shortUrl, $site) + strlen($site));
129 
130             $entries[] = (object) array
131             (
132                 'url' => $shortUrl,
133                 'title' => $result->title,
134                 'body' => $result->content
135             );
136         }
137     }
138 
139     return array($entries, $count);
140 }
141 
142 function query_pages($search, $position, $limit)
143 {
144     global $core;
145 
146     $model = $core->models['pages'];
147 
148     $query_part = 'FROM {self_and_related} INNER JOIN {self}__contents content ON (nid = pageid AND contentid = "body")
149     WHERE is_online = 1 AND siteid = ? AND editor != "view" AND content LIKE ?';
150 
151     $query_args = array($core->site_id, '%' . $search . '%');
152 
153     $count = $model->query
154     (
155         'SELECT COUNT(nid) ' . $query_part, $query_args
156     )
157     ->fetchColumnAndClose();
158 
159     $entries = $model->query
160     (
161         'SELECT node.*, page.*, content.content AS body ' . $query_part . ' ORDER BY created_at DESC LIMIT ' . ($position * $limit) . ', ' . $limit, $query_args
162     )
163     ->fetchAll(\PDO::FETCH_CLASS, 'Icybee\Modules\Pages\Page', array($model));
164 
165     return array($entries, $count);
166 }
167 
168 function query_contents($constructor, $search, $position, $limit)
169 {
170     global $core;
171 
172     if ($constructor == 'contents')
173     {
174             $query_part = 'is_online = 1 AND (siteid = 0 OR siteid = ?)';
175         $query_args = array($core->site_id);
176     }
177     else
178     {
179         $query_part = 'is_online = 1 AND (siteid = 0 OR siteid = ?) AND constructor = ?';
180         $query_args = array($core->site_id, $constructor);
181     }
182 
183     $model = $core->models[$constructor];
184 
185     $words = explode(' ', $search);
186     $words = array_map(function($str) {
187 
188         return '%' . trim($str) . '%';
189 
190     }, $words);
191 
192     $properties = array('title', 'body');
193     $concat = str_repeat(' AND CONCAT_WS(" ", ' . implode(', ', $properties) . ') LIKE ?', count($words));
194 
195     $query_part .= ' AND (' . substr($concat, 4) . ')';
196     $query_args = array_merge($query_args, $words);
197 
198     $arq = $model->where($query_part, $query_args);
199 
200     $count = $arq->count;
201     $entries = $arq->limit($position * $limit, $limit)->order('date DESC')->all;
202 
203     return array($entries, $count);
204 }
205 
206 
207 
208 
209 
210 function make_set($constructor, $entries, $count, $search, $has_pager=false)
211 {
212     global $core;
213 
214     $flat_id = 'module.' . strtr($constructor, '.', '_');
215 
216     $rc  = '<div class="set">';
217 
218     if (empty($_GET['constructor']))
219     {
220         $title = ($constructor == 'google' ? 'Google' : $core->modules->descriptors[$constructor][Module::T_TITLE]);
221         $title = I18n\t(strtr($constructor, '.', '_'), array(), array('scope' => 'module_title', 'default' => $title));
222 
223         $rc .= '<h2>' . $title . '</h2>';
224     }
225 
226     $rc .= '<p class="count">';
227     $rc .= I18n\t('found', array(':count' => $count, '%search' => $search), array('scope' => array($flat_id, 'search')));
228     $rc .= '</p>';
229 
230     if ($entries)
231     {
232         $rc .= '<ol>';
233 
234         foreach ($entries as $entry)
235         {
236             $rc .= '<li>';
237             $rc .= '<h3><a href="' . $entry->url . '">' . \ICanBoogie\escape(\ICanBoogie\shorten($entry->title, 80)) . '</a></h3>';
238             $rc .= '<cite>' . \ICanBoogie\shorten($entry->url, 64) . '</cite>';
239 
240             $excerpt = search_excerpt($search, html_entity_decode($entry->body, ENT_COMPAT, 'utf-8'));
241 
242             $rc .= '<p class="excerpt">' . $excerpt . '</p>';
243             $rc .= '</li>';
244         }
245 
246         $rc .= '</ol>';
247 
248         if ($count > count($entries))
249         {
250             if ($has_pager)
251             {
252                 $rc .= new Pager
253                 (
254                     'div', array
255                     (
256                         Pager::T_COUNT => $count,
257                         Pager::T_LIMIT => $core->site->metas->get('search.limits.list', 10),
258                         Pager::T_POSITION => isset($_GET['page']) ? (int) $_GET['page'] : 0,
259                         Pager::T_WITH => 'q,constructor',
260 
261                         'class' => 'pagination'
262                     )
263                 );
264             }
265             else
266             {
267                 $more_url = '?' . http_build_query(array('q' => $search, 'constructor' => $constructor));
268 
269                 $rc .= '<p class="more"><a href="' . $more_url . '">';
270                 $rc .= I18n\t('more', array(':count' => $count, '%search' => $search), array('scope' => array($flat_id, 'search')));
271                 $rc .= '</a></p>';
272             }
273         }
274     }
275 
276     $rc .= '</div>';
277 
278     return $rc;
279 }
280 
281 
282 
283 
284 /**
285  * Returns snippets from a piece of text, with certain keywords highlighted.
286  * Used for formatting search results.
287  *
288  * @param $keys
289  *   A string containing a search query.
290  *
291  * @param $text
292  *   The text to extract fragments from.
293  *
294  * @return
295  *   A string containing HTML for the excerpt.
296  */
297 function search_excerpt($keys, $text) {
298     // We highlight around non-indexable or CJK characters.
299     $boundary = '(?:(?<=['. PREG_CLASS_SEARCH_EXCLUDE . PREG_CLASS_CJK .'])|(?=['. PREG_CLASS_SEARCH_EXCLUDE . PREG_CLASS_CJK .']))';
300 
301     // Extract positive keywords and phrases
302     preg_match_all('/ ("([^"]+)"|(?!OR)([^" ]+))/', ' '. $keys, $matches);
303     $keys = array_merge($matches[2], $matches[3]);
304 
305     // Prepare text
306     $text = ' '. strip_tags(str_replace(array('<', '>'), array(' <', '> '), $text)) .' ';
307     array_walk($keys, __NAMESPACE__ . '\_search_excerpt_replace');
308     $workkeys = $keys;
309 
310     // Extract a fragment per keyword for at most 4 keywords.
311     // First we collect ranges of text around each keyword, starting/ending
312     // at spaces.
313     // If the sum of all fragments is too short, we look for second occurrences.
314     $ranges = array();
315     $included = array();
316     $length = 0;
317     while ($length < 256 && count($workkeys)) {
318         foreach ($workkeys as $k => $key) {
319             if (strlen($key) == 0) {
320                 unset($workkeys[$k]);
321                 unset($keys[$k]);
322                 continue;
323             }
324             if ($length >= 256) {
325                 break;
326             }
327             // Remember occurrence of key so we can skip over it if more occurrences
328             // are desired.
329             if (!isset($included[$key])) {
330                 $included[$key] = 0;
331             }
332             // Locate a keyword (position $p), then locate a space in front (position
333             // $q) and behind it (position $s)
334             if (preg_match('/'. $boundary . $key . $boundary .'/iu', $text, $match, PREG_OFFSET_CAPTURE, $included[$key])) {
335                 $p = $match[0][1];
336                 if (($q = strpos($text, ' ', max(0, $p - 60))) !== FALSE) {
337                     $end = substr($text, $p, 80);
338                     if (($s = strrpos($end, ' ')) !== FALSE) {
339                         $ranges[$q] = $p + $s;
340                         $length += $p + $s - $q;
341                         $included[$key] = $p + 1;
342                     }
343                     else {
344                         unset($workkeys[$k]);
345                     }
346                 }
347                 else {
348                     unset($workkeys[$k]);
349                 }
350             }
351             else {
352                 unset($workkeys[$k]);
353             }
354         }
355     }
356 
357     // If we didn't find anything, return the beginning.
358     if (count($ranges) == 0) {
359         return truncate_utf8($text, 256) .' …';
360     }
361 
362     // Sort the text ranges by starting position.
363     ksort($ranges);
364 
365     // Now we collapse overlapping text ranges into one. The sorting makes it O(n).
366     $newranges = array();
367     foreach ($ranges as $from2 => $to2) {
368         if (!isset($from1)) {
369             $from1 = $from2;
370             $to1 = $to2;
371             continue;
372         }
373         if ($from2 <= $to1) {
374             $to1 = max($to1, $to2);
375         }
376         else {
377             $newranges[$from1] = $to1;
378             $from1 = $from2;
379             $to1 = $to2;
380         }
381     }
382     $newranges[$from1] = $to1;
383 
384     // Fetch text
385     $out = array();
386     foreach ($newranges as $from => $to) {
387         $out[] = substr($text, $from, $to - $from);
388     }
389     $text = (isset($newranges[0]) ? '' : '… ') . implode(' … ', $out) .' …';
390 
391     // Highlight keywords. Must be done at once to prevent conflicts ('strong' and '<strong>').
392     $text = preg_replace('/'. $boundary .'('. implode('|', $keys) .')'. $boundary .'/iu', '<strong>\0</strong>', $text);
393     return $text;
394 }
395 
396 /**
397  * @} End of "defgroup search".
398  */
399 
400 /**
401  * Helper function for array_walk in search_except.
402  */
403 function _search_excerpt_replace(&$text) {
404     $text = preg_quote($text, '/');
405 }
406 
407 function search_forms() {
408     $forms['search_theme_form']= array(
409         'callback' => 'search_box',
410         'callback arguments' => array('search_theme_form'),
411     );
412     $forms['search_block_form']= array(
413         'callback' => 'search_box',
414         'callback arguments' => array('search_block_form'),
415     );
416     return $forms;
417 }
418 
419 
420 
421 /**
422  * Truncate a UTF-8-encoded string safely to a number of characters.
423  *
424  * @param $string
425  *   The string to truncate.
426  * @param $len
427  *   An upper limit on the returned string length.
428  * @param $wordsafe
429  *   Flag to truncate at last space within the upper limit. Defaults to FALSE.
430  * @param $dots
431  *   Flag to add trailing dots. Defaults to FALSE.
432  * @return
433  *   The truncated string.
434  */
435 function truncate_utf8($string, $len, $wordsafe = FALSE, $dots = FALSE) {
436 
437     if (mb_strlen($string) <= $len) {
438         return $string;
439     }
440 
441     if ($dots) {
442         $len -= 4;
443     }
444 
445     if ($wordsafe) {
446         $string = mb_substr($string, 0, $len + 1); // leave one more character
447         if ($last_space = strrpos($string, ' ')) { // space exists AND is not on position 0
448             $string = substr($string, 0, $last_space);
449         }
450         else {
451             $string = mb_substr($string, 0, $len);
452         }
453     }
454     else {
455         $string = mb_substr($string, 0, $len);
456     }
457 
458     if ($dots) {
459         $string .= ' ...';
460     }
461 
462     return $string;
463 }

Autodoc API documentation generated by ApiGen 2.8.0