Project

General

Profile

1
<?php
2

    
3
// $Id: search_modext.php 777 2008-03-27 20:02:43Z thorn $
4

    
5
/*
6

    
7
 Website Baker Project <http://www.websitebaker.org/>
8
 Copyright (C) 2004-2008, Ryan Djurovich
9

    
10
 Website Baker is free software; you can redistribute it and/or modify
11
 it under the terms of the GNU General Public License as published by
12
 the Free Software Foundation; either version 2 of the License, or
13
 (at your option) any later version.
14

    
15
 Website Baker is distributed in the hope that it will be useful,
16
 but WITHOUT ANY WARRANTY; without even the implied warranty of
17
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
18
 GNU General Public License for more details.
19

    
20
 You should have received a copy of the GNU General Public License
21
 along with Website Baker; if not, write to the Free Software
22
 Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
23

    
24
*/
25

    
26
// make the url-string for highlighting
27
function make_url_searchstring($search_match, $search_url_array) {
28
	$link = "";
29
	if ($search_match != 'exact') {
30
		$str = implode(" ", $search_url_array);
31
		$link = "?searchresult=1&amp;sstring=".urlencode($str);
32
	} else {
33
		$str = str_replace(' ', '_', $search_url_array[0]);
34
		$link = "?searchresult=2&amp;sstring=".urlencode($str);
35
	}
36
	return $link;
37
}
38

    
39
// make date and time for "last modified by... on ..."-string
40
function get_page_modified($page_modified_when) {
41
	global $TEXT;
42
	if($page_modified_when > 0) {
43
		$date = gmdate(DATE_FORMAT, $page_modified_when+TIMEZONE);
44
		$time = gmdate(TIME_FORMAT, $page_modified_when+TIMEZONE);
45
	} else {
46
		$date = $TEXT['UNKNOWN'].' '.$TEXT['DATE'];
47
		$time = $TEXT['UNKNOWN'].' '.$TEXT['TIME'];
48
	}
49
	return array($date, $time);
50
}
51

    
52
// make username and displayname for "last modified by... on ..."-string
53
function get_page_modified_by($page_modified_by, $users) {
54
	global $TEXT;
55
	if($page_modified_by > 0) {
56
			$username = $users[$page_modified_by]['username'];
57
			$displayname = $users[$page_modified_by]['display_name'];
58
		} else {
59
			$username = "";
60
			$displayname = $TEXT['UNKNOWN'];
61
		}
62
	return array($username, $displayname);
63
}
64

    
65
// checks if _all_ searchwords matches
66
function is_all_matched($text, $search_words) {
67
	$all_matched = true;
68
	foreach ($search_words AS $word) {
69
		if(!preg_match('/'.$word.'/i', $text)) {
70
			$all_matched = false;
71
			break;
72
		}
73
	}
74
	return $all_matched;
75
}
76

    
77
// checks if _any_ of the searchwords matches
78
function is_any_matched($text, $search_words) {
79
	$any_matched = false;
80
	$word = '('.implode('|', $search_words).')';
81
	if(preg_match('/'.$word.'/i', $text)) {
82
		$any_matched = true;
83
	}
84
	return $any_matched;
85
}
86

    
87
// collects the matches from text in excerpt_array
88
function get_excerpts($text, $search_words, $max_excerpt_num) {
89
	$match_array = array();
90
	$excerpt_array = array();
91
	$word = '('.implode('|', $search_words).')';
92
	// Build the regex-string
93
	// start-sign: .!?; + INVERTED EXCLAMATION MARK - INVERTED QUESTION MARK - DOUBLE EXCLAMATION MARK - INTERROBANG - EXCLAMATION QUESTION MARK - QUESTION EXCLAMATION MARK - DOUBLE QUESTION MARK - HALFWIDTH IDEOGRAPHIC FULL STOP - IDEOGRAPHIC FULL STOP - IDEOGRAPHIC COMMA
94
	$str1=".!?;"."\xC2\xA1"."\xC2\xBF"."\xE2\x80\xBC"."\xE2\x80\xBD"."\xE2\x81\x89"."\xE2\x81\x88"."\xE2\x81\x87"."\xEF\xBD\xA1"."\xE3\x80\x82"."\xE3\x80\x81";
95
	// stop-sign: .!?; + DOUBLE EXCLAMATION MARK - INTERROBANG - EXCLAMATION QUESTION MARK - QUESTION EXCLAMATION MARK - DOUBLE QUESTION MARK - HALFWIDTH IDEOGRAPHIC FULL STOP - IDEOGRAPHIC FULL STOP - IDEOGRAPHIC COMMA
96
	$str2=".!?;"."\xE2\x80\xBC"."\xE2\x80\xBD"."\xE2\x81\x89"."\xE2\x81\x88"."\xE2\x81\x87"."\xEF\xBD\xA1"."\xE3\x80\x82"."\xE3\x80\x81";
97
	$regex='/(?:^|\b|['.$str1.'])([^'.$str1.']{0,200}?'.$word.'[^'.$str2.']{0,200}(?:['.$str2.']|\b|$))/Sisu';
98
	if(version_compare(PHP_VERSION, '4.3.3', '>=')) {
99
		// jump from match to match, get excerpt, stop if $max_excerpt_num is reached
100
		$last_end = 0; $offset = 0;
101
		while(preg_match('/'.$word.'/Sisu', $text, $match_array, PREG_OFFSET_CAPTURE, $last_end)) {
102
			$offset = ($match_array[0][1]-206 < $last_end)?$last_end:$match_array[0][1]-206;
103
			if(preg_match($regex, $text, $matches, PREG_OFFSET_CAPTURE, $offset)) {
104
				$last_end = $matches[1][1]+strlen($matches[1][0])-1;
105
				if(!preg_match('/\b[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\./S', $matches[1][0])) // skip excerpts with email-addresses
106
					$excerpt_array[] = trim($matches[1][0]);
107
				if(count($excerpt_array)>=$max_excerpt_num) {
108
					$excerpt_array = array_unique($excerpt_array);
109
					if(count($excerpt_array) >= $max_excerpt_num)
110
						break;
111
				}
112
			} else { // problem - preg_match failed: can't find a start- or stop-sign
113
				$last_end += 201; // jump forward and try again
114
			}
115
		}
116
	} else { // compatile, but may be very slow with many large pages
117
		if(preg_match_all($regex, $text, $match_array)) {
118
			foreach($match_array[1] AS $string) {
119
				if(!preg_match('/\b[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\./S', $string)) // skip excerpts with email-addresses
120
					$excerpt_array[] = trim($string);
121
			}
122
		}
123
	}
124
	return $excerpt_array;
125
}
126

    
127
// makes excerpt_array a string ready to print out
128
function prepare_excerpts($excerpt_array, $search_words, $max_excerpt_num) {
129
	// excerpts: text before and after a single excerpt, html-tag for markup
130
	$EXCERPT_BEFORE =       '...&nbsp;';
131
	$EXCERPT_AFTER =        '&nbsp;...<br />';
132
	$EXCERPT_MARKUP_START = '<b>';
133
	$EXCERPT_MARKUP_END =   '</b>';
134
	// remove duplicate matches from $excerpt_array, if any.
135
	$excerpt_array = array_unique($excerpt_array);
136
	// use the first $max_excerpt_num excerpts only
137
	if(count($excerpt_array) > $max_excerpt_num) {
138
		$excerpt_array = array_slice($excerpt_array, 0, $max_excerpt_num);
139
	}
140
	// prepare search-string
141
	$string = "(".implode("|", $search_words).")";
142
	// we want markup on search-results page,
143
	// but we need some 'magic' to prevent <br />, <b>... from being highlighted
144
	$excerpt = '';
145
	foreach($excerpt_array as $str) {
146
		$excerpt .= '#,,#'.preg_replace("/($string)/iu","#,,,,#$1#,,,,,#",$str).'#,,,#';
147
	}
148
	$excerpt = str_replace(array('&','<','>','"','\'',"\xC2\xA0"), array('&amp;','&lt;','&gt;','&quot;','&#39;','&nbsp;'), $excerpt);
149
	$excerpt = str_replace(array('#,,,,#','#,,,,,#'), array($EXCERPT_MARKUP_START,$EXCERPT_MARKUP_END), $excerpt);
150
	$excerpt = str_replace(array('#,,#','#,,,#'), array($EXCERPT_BEFORE,$EXCERPT_AFTER), $excerpt);
151
	// prepare to write out
152
	if(DEFAULT_CHARSET != 'utf-8') {
153
		$excerpt = umlauts_to_entities($excerpt, 'UTF-8');
154
	}
155
	return $excerpt;
156
}
157

    
158
// work out what the link-anchor should be
159
function make_url_target($page_link_target, $text, $search_words) {
160
	// 1. e.g. $page_link_target=="&monthno=5&year=2007" - module-dependent target. Do nothing.
161
	// 2. $page_link_target=="#!wb_section_..." - the user wants the section-target, so do nothing.
162
	// 3. $page_link_target=="#wb_section_..." - try to find a better target, use the section-target as fallback.
163
	// 4. $page_link_target=="" - do nothing
164
	if(version_compare(PHP_VERSION, '4.3.3', ">=") && substr($page_link_target,0,12)=='#wb_section_') {
165
		$word = '('.implode('|', $search_words).')';
166
		preg_match('/'.$word.'/i', $text, $match, PREG_OFFSET_CAPTURE);
167
		if($match && is_array($match[0])) {
168
			$x=$match[0][1]; // position of first match
169
			// is there an anchor nearby?
170
			if(preg_match_all('/<(?:[^>]+id|\s*a[^>]+name)\s*=\s*"(.*)"/SiU', substr($text,0,$x), $match, PREG_OFFSET_CAPTURE)) {
171
				$anchor='';
172
				foreach($match[1] AS $array) {
173
					if($array[1] > $x) {
174
						break;
175
					}
176
					$anchor = $array[0];
177
				}
178
				if($anchor != '') {
179
					$page_link_target = '#'.$anchor;
180
				}
181
			}
182
		}
183
	}
184
	elseif(substr($page_link_target,0,13)=='#!wb_section_') {
185
		$page_link_target = '#'.substr($page_link_target, 2);
186
	}
187
	return $page_link_target;
188
}
189

    
190
// wrapper for compatibility with old print_excerpt()
191
function print_excerpt($page_link, $page_link_target, $page_title, $page_description, $page_modified_when, $page_modified_by, $text, $max_excerpt_num, $func_vars, $pic_link="") {
192
	$mod_vars = array(
193
		'page_link' => $page_link,
194
		'page_link_target' => $page_link_target,
195
		'page_title' => $page_title,
196
		'page_description' => $page_description,
197
		'page_modified_when' => $page_modified_when,
198
		'page_modified_by' => $page_modified_by,
199
		'text' => $text,
200
		'max_excerpt_num' => $max_excerpt_num,
201
		'pic_link' => $pic_link
202
	);
203
	print_excerpt2($mod_vars, $func_vars);
204
}
205

    
206
/* These functions can be used in module-supplied search_funcs
207
 * -----------------------------------------------------------
208
 * print_excerpt2() - the main-function to use in all search_funcs
209
 * print_excerpt() - wrapper for compatibility-reason. Use print_excerpt2() instead.
210
 * list_files_dirs() - lists all files and dirs below a given directory
211
 * clear_filelist() - keeps only wanted or removes unwanted entries in file-list.
212
 */
213
 
214
// prints the excerpts for one section
215
function print_excerpt2($mod_vars, $func_vars) {
216
	extract($func_vars, EXTR_PREFIX_ALL, 'func');
217
	extract($mod_vars, EXTR_PREFIX_ALL, 'mod');
218
	global $TEXT;
219
	// check $mod_...vars
220
	if(!isset($mod_page_link))          $mod_page_link = $func_page_link;
221
	if(!isset($mod_page_link_target))   $mod_page_link_target = "";
222
	if(!isset($mod_page_title))         $mod_page_title = $func_page_title;
223
	if(!isset($mod_page_description))   $mod_page_description = $func_page_description;
224
	if(!isset($mod_page_modified_when)) $mod_page_modified_when = $func_page_modified_when;
225
	if(!isset($mod_page_modified_by))   $mod_page_modified_by = $func_page_modified_by;
226
	if(!isset($mod_text))               $mod_text = "";
227
	if(!isset($mod_max_excerpt_num))    $mod_max_excerpt_num = $func_default_max_excerpt;
228
	if(!isset($mod_pic_link))           $mod_pic_link = "";
229
	if(!isset($mod_no_highlight))       $mod_no_highlight = false;
230
	if(!isset($func_enable_flush))      $func_enable_flush = false; // set this in db: wb_search.cfg_enable_flush [READ THE DOC BEFORE]
231
	if($mod_text == "") // nothing to do
232
		{ return false; }
233
	if($mod_no_highlight) // no highlighting
234
		{ $mod_page_link_target = "&amp;nohighlight=1".$mod_page_link_target; }
235
	// clean the text:
236
	$mod_text = str_replace(array("\x0D","\x0A"), ' ', $mod_text);
237
	$mod_text = preg_replace('#<(!--.*--|style.*</style|script.*</script)>#SiU', ' ', $mod_text);
238
	$mod_text = preg_replace('#<(br( /)?|dt|/dd|/?(h[1-6]|tr|table|p|li|ul|pre|code|div|hr))[^>]*>#Si', '.', $mod_text);
239
	$mod_text = entities_to_umlauts($mod_text, 'UTF-8');
240
	$anchor_text = $mod_text; // make an copy containing html-tags
241
	$mod_text = strip_tags($mod_text);
242
	$mod_text = str_replace(array('&gt;','&lt;','&amp;','&quot;','&#39;','&apos;','&nbsp;'), array('>','<','&','"','\'','\'',"\xC2\xA0"), $mod_text);
243
	// Do a fast scan over $mod_text first. This will speedup things a lot.
244
	if($func_search_match == 'all') {
245
		if(!is_all_matched($mod_text, $func_search_words))
246
			return false;
247
	}
248
	elseif(!is_any_matched($mod_text, $func_search_words)) {
249
		return false;
250
	}
251
	// search for an better anchor - this have to be done before strip_tags() (may fail if search-string contains <, &, amp, gt, lt, ...)
252
	$anchor =  make_url_target($mod_page_link_target, $anchor_text, $func_search_words);
253

    
254
	// make the link from $mod_page_link, add anchor
255
	$link = "";
256
	$link = page_link($mod_page_link);
257
	$link .= make_url_searchstring($func_search_match, $func_search_url_array);
258
	$link .= $anchor;
259

    
260
	// now get the excerpt
261
	$excerpt = "";
262
	$excerpt_array = array();
263
	if($mod_max_excerpt_num > 0) {
264
		if(!$excerpt_array = get_excerpts($mod_text, $func_search_words, $mod_max_excerpt_num)) {
265
			return false;
266
		}
267
		$excerpt = prepare_excerpts($excerpt_array, $func_search_words, $mod_max_excerpt_num);
268
	}
269

    
270
	// handle thumbs - to deactivate this look in the module's search.php: $show_thumb (or maybe in the module's settings-page)
271
	if($mod_pic_link != "") {
272
		$excerpt = '<table class="excerpt_thumb" width="100%" cellspacing="0" cellpadding="0" border="0"><tbody><tr><td width="110" valign="top"><a href="'.$link.'"><img src="'.WB_URL.'/'.MEDIA_DIRECTORY.$mod_pic_link.'" alt="" /></a></td><td>'.$excerpt.'</td></tr></tbody></table>';
273
	}
274

    
275
	// print-out the excerpt
276
	$vars = array();
277
	$values = array();
278
	list($date, $time) = get_page_modified($mod_page_modified_when);
279
	list($username, $displayname) = get_page_modified_by($mod_page_modified_by, $func_users);
280
	$vars = array('[LINK]', '[TITLE]', '[DESCRIPTION]', '[USERNAME]','[DISPLAY_NAME]','[DATE]','[TIME]','[TEXT_LAST_UPDATED_BY]','[TEXT_ON]','[EXCERPT]');
281
	$values = array(
282
		$link,
283
		$mod_page_title,
284
		$mod_page_description,
285
		$username,
286
		$displayname,
287
		$date,
288
		$time,
289
		$TEXT['LAST_UPDATED_BY'],
290
		$TEXT['ON'],
291
		$excerpt
292
	);
293
	echo str_replace($vars, $values, $func_results_loop_string);
294
	if($func_enable_flush) { // ATTN: this will bypass output-filters and may break template-layout or -filters
295
		ob_flush();flush();
296
	}
297
	return true;
298
}
299

    
300
// list all files and dirs in $dir (recursive), omits '.', '..', and hidden files/dirs
301
// returns an array of two arrays ($files[] and $dirs[]).
302
// usage: list($files,$dirs) = list_files_dirs($directory);
303
//        $depth: get subdirs (true/false)
304
function list_files_dirs($dir, $depth=true, $files=array(), $dirs=array()) {
305
	$dh=opendir($dir);
306
	while(($file = readdir($dh)) !== false) {
307
		if($file{0} == '.' || $file == '..') {
308
			continue;
309
		}
310
		if(is_dir($dir.'/'.$file)) {
311
			if($depth) {
312
				$dirs[] = $dir.'/'.$file;
313
				list($files, $dirs) = list_files_dirs($dir.'/'.$file, $depth, $files, $dirs);
314
			}
315
		} else {
316
			$files[] = $dir.'/'.$file;
317
		}
318
	}
319
	closedir($dh);
320
	natcasesort($files);
321
	natcasesort($dirs);
322
	return(array($files, $dirs));
323
}
324

    
325
// keeps only wanted entries in array $files. $str have to be an eregi()-compatible regex
326
function clear_filelist($files, $str, $keep=true) {
327
	// options: $keep = true  : remove all non-matching entries
328
	//          $keep = false : remove all matching entries
329
	$c_filelist = array();
330
	if($str == '')
331
		return $files;
332
	foreach($files as $file) {
333
		if($keep) {
334
			if(eregi($str, $file)) {
335
				$c_filelist[] = $file;
336
			}
337
		} else {
338
			if(!eregi($str, $file)) {
339
				$c_filelist[] = $file;
340
			}
341
		}
342
	}
343
	return($c_filelist);
344
}
345

    
346
?>
(4-4/4)