Revision 1441
Added by Luisehahne over 14 years ago
- recoded /modules/admin.php info_banner, now compare with modify in pages
- security fixes remove defined WB_PATH for backend templates
- fixed class.admin.php missing $TEXT declaration, add get_section_details
| search_modext.php | ||
|---|---|---|
| 5 | 5 |
* @package search |
| 6 | 6 |
* @author WebsiteBaker Project |
| 7 | 7 |
* @copyright 2004-2009, Ryan Djurovich |
| 8 |
* @copyright 2009-2011, Website Baker Org. e.V.
|
|
| 8 |
* @copyright 2009-2010, Website Baker Org. e.V.
|
|
| 9 | 9 |
* @link http://www.websitebaker2.org/ |
| 10 | 10 |
* @license http://www.gnu.org/licenses/gpl.html |
| 11 | 11 |
* @platform WebsiteBaker 2.8.x |
| 12 |
* @requirements PHP 5.2.2 and higher
|
|
| 12 |
* @requirements PHP 4.3.4 and higher
|
|
| 13 | 13 |
* @version $Id$ |
| 14 | 14 |
* @filesource $HeadURL$ |
| 15 | 15 |
* @lastmodified $Date$ |
| 16 | 16 |
* |
| 17 | 17 |
*/ |
| 18 | 18 |
|
| 19 |
// Must include code to stop this file being access directly |
|
| 20 |
if(defined('WB_PATH') == false) { die("Cannot access this file directly"); }
|
|
| 21 |
|
|
| 22 | 19 |
// make the url-string for highlighting |
| 23 |
function make_url_searchstring($search_match, $search_url_array) |
|
| 24 |
{
|
|
| 20 |
function make_url_searchstring($search_match, $search_url_array) {
|
|
| 25 | 21 |
$link = ""; |
| 26 |
if ($search_match != 'exact') |
|
| 27 |
{
|
|
| 22 |
if ($search_match != 'exact') {
|
|
| 28 | 23 |
$str = implode(" ", $search_url_array);
|
| 29 | 24 |
$link = "?searchresult=1&sstring=".urlencode($str); |
| 30 | 25 |
} else {
|
| ... | ... | |
| 35 | 30 |
} |
| 36 | 31 |
|
| 37 | 32 |
// make date and time for "last modified by... on ..."-string |
| 38 |
function get_page_modified($page_modified_when) |
|
| 39 |
{
|
|
| 33 |
function get_page_modified($page_modified_when) {
|
|
| 40 | 34 |
global $TEXT; |
| 41 |
if($page_modified_when > 0) |
|
| 42 |
{
|
|
| 35 |
if($page_modified_when > 0) {
|
|
| 43 | 36 |
$date = gmdate(DATE_FORMAT, $page_modified_when+TIMEZONE); |
| 44 | 37 |
$time = gmdate(TIME_FORMAT, $page_modified_when+TIMEZONE); |
| 45 | 38 |
} else {
|
| ... | ... | |
| 50 | 43 |
} |
| 51 | 44 |
|
| 52 | 45 |
// make username and displayname for "last modified by... on ..."-string |
| 53 |
function get_page_modified_by($page_modified_by, $users) |
|
| 54 |
{
|
|
| 46 |
function get_page_modified_by($page_modified_by, $users) {
|
|
| 55 | 47 |
global $TEXT; |
| 56 |
// check for existing user-id |
|
| 57 |
if(!isset($users[$page_modified_by])) |
|
| 58 |
{
|
|
| 59 |
$page_modified_by = 0; |
|
| 60 |
} |
|
| 61 |
|
|
| 62 |
$username = $users[$page_modified_by]['username']; |
|
| 63 |
$displayname = $users[$page_modified_by]['display_name']; |
|
| 48 |
if($page_modified_by>0) {
|
|
| 49 |
$username = $users[$page_modified_by]['username']; |
|
| 50 |
$displayname = $users[$page_modified_by]['display_name']; |
|
| 51 |
} else {
|
|
| 52 |
$username = ""; |
|
| 53 |
$displayname = $TEXT['UNKNOWN']; |
|
| 54 |
} |
|
| 64 | 55 |
return array($username, $displayname); |
| 65 | 56 |
} |
| 66 | 57 |
|
| 67 | 58 |
// checks if _all_ searchwords matches |
| 68 |
function is_all_matched($text, $search_words) |
|
| 69 |
{
|
|
| 59 |
function is_all_matched($text, $search_words) {
|
|
| 70 | 60 |
$all_matched = true; |
| 71 |
foreach ($search_words AS $word) |
|
| 72 |
{
|
|
| 73 |
if(!preg_match('/'.$word.'/ui', $text))
|
|
| 74 |
{
|
|
| 61 |
foreach ($search_words AS $word) {
|
|
| 62 |
if(!preg_match('/'.$word.'/i', $text)) {
|
|
| 75 | 63 |
$all_matched = false; |
| 76 | 64 |
break; |
| 77 | 65 |
} |
| ... | ... | |
| 80 | 68 |
} |
| 81 | 69 |
|
| 82 | 70 |
// checks if _any_ of the searchwords matches |
| 83 |
function is_any_matched($text, $search_words) |
|
| 84 |
{
|
|
| 71 |
function is_any_matched($text, $search_words) {
|
|
| 85 | 72 |
$any_matched = false; |
| 86 | 73 |
$word = '('.implode('|', $search_words).')';
|
| 87 |
if(preg_match('/'.$word.'/ui', $text))
|
|
| 88 |
{
|
|
| 74 |
if(preg_match('/'.$word.'/i', $text)) {
|
|
| 89 | 75 |
$any_matched = true; |
| 90 | 76 |
} |
| 91 | 77 |
return $any_matched; |
| 92 | 78 |
} |
| 93 | 79 |
|
| 94 | 80 |
// collects the matches from text in excerpt_array |
| 95 |
function get_excerpts($text, $search_words, $max_excerpt_num) |
|
| 96 |
{
|
|
| 97 |
$match_array = array(); |
|
| 98 |
$excerpt_array = array(); |
|
| 81 |
function get_excerpts($text, $search_words, $max_excerpt_num) {
|
|
| 82 |
$excerpt_array = FALSE; |
|
| 99 | 83 |
$word = '('.implode('|', $search_words).')';
|
| 100 |
|
|
| 101 |
//Filter droplets from the page data |
|
| 102 |
preg_match_all('~\[\[(.*?)\]\]~', $text, $matches);
|
|
| 103 |
foreach ($matches[1] as $match) |
|
| 104 |
{
|
|
| 105 |
$text = str_replace('[['.$match.']]', '', $text);
|
|
| 106 |
} |
|
| 107 |
|
|
| 108 |
// Build the regex-string |
|
| 109 |
if(strpos(strtoupper(PHP_OS), 'WIN')===0) // windows -> see below |
|
| 110 |
{
|
|
| 111 |
$str1=".!?;"; |
|
| 112 |
$str2=".!?;"; |
|
| 113 |
} else { // linux & Co.
|
|
| 114 |
// 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 |
|
| 115 |
$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"; |
|
| 116 |
// stop-sign: .!?; + DOUBLE EXCLAMATION MARK - INTERROBANG - EXCLAMATION QUESTION MARK - QUESTION EXCLAMATION MARK - DOUBLE QUESTION MARK - HALFWIDTH IDEOGRAPHIC FULL STOP - IDEOGRAPHIC FULL STOP - IDEOGRAPHIC COMMA |
|
| 117 |
$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"; |
|
| 118 |
} |
|
| 119 |
$regex='/(?:^|\b|['.$str1.'])([^'.$str1.']{0,200}?'.$word.'[^'.$str2.']{0,200}(?:['.$str2.']|\b|$))/uis';
|
|
| 120 |
if(version_compare(PHP_VERSION, '4.3.3', '>=') && |
|
| 121 |
strpos(strtoupper(PHP_OS), 'WIN')!==0 |
|
| 122 |
) { // this may crash windows server, so skip if on windows
|
|
| 123 |
// jump from match to match, get excerpt, stop if $max_excerpt_num is reached |
|
| 124 |
$last_end = 0; $offset = 0; |
|
| 125 |
while(preg_match('/'.$word.'/uis', $text, $match_array, PREG_OFFSET_CAPTURE, $last_end))
|
|
| 126 |
{
|
|
| 127 |
$offset = ($match_array[0][1]-206 < $last_end)?$last_end:$match_array[0][1]-206; |
|
| 128 |
if(preg_match($regex, $text, $matches, PREG_OFFSET_CAPTURE, $offset)) |
|
| 129 |
{
|
|
| 130 |
$last_end = $matches[1][1]+strlen($matches[1][0])-1; |
|
| 131 |
if(!preg_match('/\b[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\./', $matches[1][0])) // skip excerpts with email-addresses
|
|
| 132 |
{
|
|
| 133 |
$excerpt_array[] = trim($matches[1][0]); |
|
| 134 |
} |
|
| 135 |
if(count($excerpt_array)>=$max_excerpt_num) |
|
| 136 |
{
|
|
| 137 |
$excerpt_array = array_unique($excerpt_array); |
|
| 138 |
if(count($excerpt_array) >= $max_excerpt_num) { break; }
|
|
| 139 |
} |
|
| 140 |
} else { // problem: preg_match failed - can't find a start- or stop-sign
|
|
| 141 |
$last_end += 201; // jump forward and try again |
|
| 142 |
} |
|
| 84 |
// 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 |
|
| 85 |
$p_start=".!?;"."\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"; |
|
| 86 |
// stop-sign: .!?; + DOUBLE EXCLAMATION MARK - INTERROBANG - EXCLAMATION QUESTION MARK - QUESTION EXCLAMATION MARK - DOUBLE QUESTION MARK - HALFWIDTH IDEOGRAPHIC FULL STOP - IDEOGRAPHIC FULL STOP - IDEOGRAPHIC COMMA |
|
| 87 |
$p_stop=".!?;"."\xE2\x80\xBC"."\xE2\x80\xBD"."\xE2\x81\x89"."\xE2\x81\x88"."\xE2\x81\x87"."\xEF\xBD\xA1"."\xE3\x80\x82"."\xE3\x80\x81"; |
|
| 88 |
// jump from match to match, get excerpt, stop if $max_excerpt_num is reached |
|
| 89 |
$match_array = $matches = array(); |
|
| 90 |
$startpos = $wordpos = $endpos = 0; // although preg_match with u-switch handles unicode correctly, the ...pos-variables will count bytes (not chars) |
|
| 91 |
while(preg_match("/$word/i", $text, $match_array, PREG_OFFSET_CAPTURE, $startpos)) {
|
|
| 92 |
$wordpos = $match_array[0][1]; |
|
| 93 |
$startpos = ($wordpos-200 < $endpos)?$endpos:$wordpos-200; |
|
| 94 |
$endpos = $wordpos+200; |
|
| 95 |
// look for better start position |
|
| 96 |
if(preg_match_all("/[$p_start]/u", substr($text, $startpos, $wordpos-$startpos), $matches, PREG_OFFSET_CAPTURE))
|
|
| 97 |
$startpos += $matches[0][count($matches[0])-1][1]; // set startpos at last punctuation before word |
|
| 98 |
// look for better end position |
|
| 99 |
if(preg_match_all("/[$p_stop]/u", substr($text, $wordpos, $endpos-$wordpos), $matches, PREG_OFFSET_CAPTURE))
|
|
| 100 |
$endpos = $wordpos+$matches[0][0][1]; // set endpos at first punctuation after word |
|
| 101 |
$match = substr($text, $startpos+1, $endpos-$startpos); |
|
| 102 |
if(!preg_match('/\b[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\./', $match)) // skip excerpts with email-addresses
|
|
| 103 |
$excerpt_array[] = trim($match); |
|
| 104 |
if(count($excerpt_array)>=$max_excerpt_num) {
|
|
| 105 |
$excerpt_array = array_unique($excerpt_array); |
|
| 106 |
if(count($excerpt_array) >= $max_excerpt_num) |
|
| 107 |
break; |
|
| 143 | 108 |
} |
| 144 |
} else { // compatible, but may be very slow with large pages
|
|
| 145 |
if(preg_match_all($regex, $text, $match_array)) |
|
| 146 |
{
|
|
| 147 |
foreach($match_array[1] AS $string) |
|
| 148 |
{
|
|
| 149 |
if(!preg_match('/\b[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\./', $string)) // skip excerpts with email-addresses
|
|
| 150 |
{
|
|
| 151 |
$excerpt_array[] = trim($string); |
|
| 152 |
} |
|
| 153 |
|
|
| 154 |
} |
|
| 155 |
} |
|
| 109 |
// restart at last endpos |
|
| 110 |
$startpos = $endpos; |
|
| 156 | 111 |
} |
| 157 | 112 |
return $excerpt_array; |
| 158 | 113 |
} |
| 159 | 114 |
|
| 160 | 115 |
// makes excerpt_array a string ready to print out |
| 161 |
function prepare_excerpts($excerpt_array, $search_words, $max_excerpt_num) |
|
| 162 |
{
|
|
| 116 |
function prepare_excerpts($excerpt_array, $search_words, $max_excerpt_num) {
|
|
| 163 | 117 |
// excerpts: text before and after a single excerpt, html-tag for markup |
| 164 | 118 |
$EXCERPT_BEFORE = '... '; |
| 165 | 119 |
$EXCERPT_AFTER = ' ...<br />'; |
| ... | ... | |
| 168 | 122 |
// remove duplicate matches from $excerpt_array, if any. |
| 169 | 123 |
$excerpt_array = array_unique($excerpt_array); |
| 170 | 124 |
// use the first $max_excerpt_num excerpts only |
| 171 |
if(count($excerpt_array) > $max_excerpt_num) |
|
| 172 |
{
|
|
| 125 |
if(count($excerpt_array) > $max_excerpt_num) {
|
|
| 173 | 126 |
$excerpt_array = array_slice($excerpt_array, 0, $max_excerpt_num); |
| 174 | 127 |
} |
| 175 | 128 |
// prepare search-string |
| ... | ... | |
| 177 | 130 |
// we want markup on search-results page, |
| 178 | 131 |
// but we need some 'magic' to prevent <br />, <b>... from being highlighted |
| 179 | 132 |
$excerpt = ''; |
| 180 |
foreach($excerpt_array as $str) |
|
| 181 |
{
|
|
| 182 |
$excerpt .= '#,,#'.preg_replace("/($string)/iu","#,,,,#$1#,,,,,#",$str).'#,,,#';
|
|
| 133 |
foreach($excerpt_array as $str) {
|
|
| 134 |
$excerpt .= '#,,#'.preg_replace("/($string)/i","#,,,,#$1#,,,,,#",$str).'#,,,#';
|
|
| 183 | 135 |
} |
| 184 |
$excerpt = str_replace(array('&','<','>','"','\'',"\xC2\xA0"), array('&','<','>','"',''',' '), $excerpt);
|
|
| 136 |
$excerpt = str_replace(array('&','<','>','"','\'',"\xC2\xA0"), array('&','<','>','"',''',' '), $excerpt);
|
|
| 185 | 137 |
$excerpt = str_replace(array('#,,,,#','#,,,,,#'), array($EXCERPT_MARKUP_START,$EXCERPT_MARKUP_END), $excerpt);
|
| 186 | 138 |
$excerpt = str_replace(array('#,,#','#,,,#'), array($EXCERPT_BEFORE,$EXCERPT_AFTER), $excerpt);
|
| 187 | 139 |
// prepare to write out |
| 188 |
if(DEFAULT_CHARSET != 'utf-8') |
|
| 189 |
{
|
|
| 140 |
if(DEFAULT_CHARSET != 'utf-8') {
|
|
| 190 | 141 |
$excerpt = umlauts_to_entities($excerpt, 'UTF-8'); |
| 191 | 142 |
} |
| 192 | 143 |
return $excerpt; |
| 193 | 144 |
} |
| 194 | 145 |
|
| 195 | 146 |
// work out what the link-anchor should be |
| 196 |
function make_url_target($page_link_target, $text, $search_words) |
|
| 197 |
{
|
|
| 147 |
function make_url_target($page_link_target, $text, $search_words) {
|
|
| 198 | 148 |
// 1. e.g. $page_link_target=="&monthno=5&year=2007" - module-dependent target. Do nothing. |
| 199 | 149 |
// 2. $page_link_target=="#!wb_section_..." - the user wants the section-target, so do nothing. |
| 200 | 150 |
// 3. $page_link_target=="#wb_section_..." - try to find a better target, use the section-target as fallback. |
| 201 | 151 |
// 4. $page_link_target=="" - do nothing |
| 202 |
if(version_compare(PHP_VERSION, '4.3.3', ">=") && substr($page_link_target,0,12)=='#wb_section_') |
|
| 203 |
{
|
|
| 152 |
if(version_compare(PHP_VERSION, '4.3.3', ">=") && substr($page_link_target,0,12)=='#wb_section_') {
|
|
| 204 | 153 |
$word = '('.implode('|', $search_words).')';
|
| 205 |
preg_match('/'.$word.'/ui', $text, $match, PREG_OFFSET_CAPTURE);
|
|
| 206 |
if($match && is_array($match[0])) |
|
| 207 |
{
|
|
| 154 |
preg_match('/'.$word.'/i', $text, $match, PREG_OFFSET_CAPTURE);
|
|
| 155 |
if($match && is_array($match[0])) {
|
|
| 208 | 156 |
$x=$match[0][1]; // position of first match |
| 209 | 157 |
// is there an anchor nearby? |
| 210 |
if(preg_match_all('/<(?:[^>]+id|\s*a[^>]+name)\s*=\s*"(.*)"/iU', substr($text,0,$x), $match, PREG_OFFSET_CAPTURE))
|
|
| 211 |
{
|
|
| 212 |
$anchor=''; |
|
| 213 |
foreach($match[1] AS $array) |
|
| 214 |
{
|
|
| 215 |
if($array[1] > $x) |
|
| 216 |
{
|
|
| 217 |
break; |
|
| 218 |
} |
|
| 219 |
$anchor = $array[0]; |
|
| 220 |
} |
|
| 221 |
if($anchor != '') |
|
| 222 |
{
|
|
| 223 |
$page_link_target = '#'.$anchor; |
|
| 224 |
} |
|
| 158 |
if(preg_match_all('/<\s*(?:a[^>]+?name|[^>]+?id)\s*=\s*"([^"]+)"/i', substr($text,0,$x), $match)) {
|
|
| 159 |
$page_link_target = '#'.$match[1][count($match[1])-1]; |
|
| 225 | 160 |
} |
| 226 | 161 |
} |
| 227 |
} elseif(substr($page_link_target,0,13)=='#!wb_section_') {
|
|
| 162 |
} |
|
| 163 |
elseif(substr($page_link_target,0,13)=='#!wb_section_') {
|
|
| 228 | 164 |
$page_link_target = '#'.substr($page_link_target, 2); |
| 229 | 165 |
} |
| 230 | 166 |
|
| 231 | 167 |
// since wb 2.7.1 the section-anchor is configurable - SEC_ANCHOR holds the anchor name |
| 232 |
if(substr($page_link_target,0,12)=='#wb_section_') |
|
| 233 |
{
|
|
| 234 |
if(defined('SEC_ANCHOR') && SEC_ANCHOR!='')
|
|
| 235 |
{
|
|
| 168 |
if(substr($page_link_target,0,12)=='#wb_section_') {
|
|
| 169 |
if(defined('SEC_ANCHOR') && SEC_ANCHOR!='') {
|
|
| 236 | 170 |
$sec_id = substr($page_link_target, 12); |
| 237 | 171 |
$page_link_target = '#'.SEC_ANCHOR.$sec_id; |
| 238 | 172 |
} else { // section-anchors are disabled
|
| ... | ... | |
| 244 | 178 |
} |
| 245 | 179 |
|
| 246 | 180 |
// wrapper for compatibility with old print_excerpt() |
| 247 |
function print_excerpt($page_link, $page_link_target, $page_title, $page_description, $page_modified_when, $page_modified_by, $text, $max_excerpt_num, $func_vars, $pic_link="") |
|
| 248 |
{
|
|
| 181 |
function print_excerpt($page_link, $page_link_target, $page_title, $page_description, $page_modified_when, $page_modified_by, $text, $max_excerpt_num, $func_vars, $pic_link="") {
|
|
| 249 | 182 |
$mod_vars = array( |
| 250 | 183 |
'page_link' => $page_link, |
| 251 | 184 |
'page_link_target' => $page_link_target, |
| ... | ... | |
| 267 | 200 |
* list_files_dirs() - lists all files and dirs below a given directory |
| 268 | 201 |
* clear_filelist() - keeps only wanted or removes unwanted entries in file-list. |
| 269 | 202 |
*/ |
| 270 |
|
|
| 203 |
|
|
| 271 | 204 |
// prints the excerpts for one section |
| 272 |
function print_excerpt2($mod_vars, $func_vars) |
|
| 273 |
{
|
|
| 205 |
function print_excerpt2($mod_vars, $func_vars) {
|
|
| 274 | 206 |
extract($func_vars, EXTR_PREFIX_ALL, 'func'); |
| 275 | 207 |
extract($mod_vars, EXTR_PREFIX_ALL, 'mod'); |
| 276 | 208 |
global $TEXT; |
| 277 | 209 |
// check $mod_...vars |
| 278 |
if(!isset($mod_page_link)) { $mod_page_link = $func_page_link; }
|
|
| 279 |
if(!isset($mod_page_link_target)) { $mod_page_link_target = ''; }
|
|
| 280 |
if(!isset($mod_page_title)) { $mod_page_title = $func_page_title; }
|
|
| 281 |
if(!isset($mod_page_description)) { $mod_page_description = $func_page_description; }
|
|
| 282 |
if(!isset($mod_page_modified_when)) { $mod_page_modified_when = $func_page_modified_when; }
|
|
| 283 |
if(!isset($mod_page_modified_by)) { $mod_page_modified_by = $func_page_modified_by; }
|
|
| 284 |
if(!isset($mod_text)) { $mod_text = ''; }
|
|
| 285 |
if(!isset($mod_max_excerpt_num)) { $mod_max_excerpt_num = $func_default_max_excerpt; }
|
|
| 286 |
if(!isset($mod_pic_link)) { $mod_pic_link = ''; }
|
|
| 287 |
if(!isset($mod_no_highlight)) { $mod_no_highlight = false; }
|
|
| 288 |
if(!isset($func_enable_flush)) { $func_enable_flush = false; } // set this in db: wb_search.cfg_enable_flush [READ THE DOC BEFORE]
|
|
| 289 |
if(isset($mod_ext_charset)) |
|
| 290 |
{
|
|
| 291 |
$mod_ext_charset = strtolower($mod_ext_charset); |
|
| 292 |
} else {
|
|
| 293 |
$mod_ext_charset = ''; |
|
| 294 |
} |
|
| 210 |
if(!isset($mod_page_link)) $mod_page_link = $func_page_link; |
|
| 211 |
if(!isset($mod_page_link_target)) $mod_page_link_target = ""; |
|
| 212 |
if(!isset($mod_page_title)) $mod_page_title = $func_page_title; |
|
| 213 |
if(!isset($mod_page_description)) $mod_page_description = $func_page_description; |
|
| 214 |
if(!isset($mod_page_modified_when)) $mod_page_modified_when = $func_page_modified_when; |
|
| 215 |
if(!isset($mod_page_modified_by)) $mod_page_modified_by = $func_page_modified_by; |
|
| 216 |
if(!isset($mod_text)) $mod_text = ""; |
|
| 217 |
if(!isset($mod_max_excerpt_num)) $mod_max_excerpt_num = $func_default_max_excerpt; |
|
| 218 |
if(!isset($mod_pic_link)) $mod_pic_link = ""; |
|
| 219 |
if(!isset($mod_no_highlight)) $mod_no_highlight = false; |
|
| 220 |
if(!isset($func_enable_flush)) $func_enable_flush = false; // set this in db: wb_search.cfg_enable_flush [READ THE DOC BEFORE] |
|
| 221 |
if(isset($mod_ext_charset)) $mod_ext_charset = strtolower($mod_ext_charset); |
|
| 222 |
else $mod_ext_charset = ''; |
|
| 295 | 223 |
|
| 296 | 224 |
if($mod_text == "") // nothing to do |
| 297 | 225 |
{ return false; }
|
| ... | ... | |
| 299 | 227 |
if($mod_no_highlight) // no highlighting |
| 300 | 228 |
{ $mod_page_link_target = "&nohighlight=1".$mod_page_link_target; }
|
| 301 | 229 |
// clean the text: |
| 230 |
$mod_text = preg_replace('#<(br|dt|/dd|/?(?:h[1-6]|tr|table|p|li|ul|pre|code|div|hr))[^>]*>#i', '.', $mod_text);
|
|
| 302 | 231 |
$mod_text = preg_replace('#<(!--.*--|style.*</style|script.*</script)>#iU', ' ', $mod_text);
|
| 303 |
$mod_text = preg_replace('#<(br( /)?|dt|/dd|/?(h[1-6]|tr|table|p|li|ul|pre|code|div|hr))[^>]*>#i', '.', $mod_text);
|
|
| 304 |
// $mod_text = preg_replace('/(\v\s?|\s\s)+/', ' ', $mod_text);
|
|
| 305 |
$mod_text = preg_replace('/\s\./', '.', $mod_text);
|
|
| 232 |
$mod_text = preg_replace('#\[\[.*?\]\]#', '', $mod_text); //Filter droplets from the page data
|
|
| 233 |
// strip_tags() is called below |
|
| 306 | 234 |
if($mod_ext_charset!='') { // data from external database may have a different charset
|
| 307 | 235 |
require_once(WB_PATH.'/framework/functions-utf8.php'); |
| 308 | 236 |
switch($mod_ext_charset) {
|
| ... | ... | |
| 333 | 261 |
$mod_text = charset_to_utf8($mod_text, 'UTF-8'); |
| 334 | 262 |
} |
| 335 | 263 |
} else {
|
| 336 |
$mod_text = entities_to_umlauts($mod_text, 'UTF-8'); |
|
| 264 |
$mod_text = entities_to_umlauts($mod_text, 'UTF-8');
|
|
| 337 | 265 |
} |
| 338 | 266 |
$anchor_text = $mod_text; // make an copy containing html-tags |
| 339 | 267 |
$mod_text = strip_tags($mod_text); |
| 340 |
$mod_text = str_replace(array('>','<','&','"',''',''',' '), array('>','<','&','"','\'','\'',"\xC2\xA0"), $mod_text);
|
|
| 268 |
$mod_text = str_replace(array('>','<','&','"',''',''',' '), array('>','<','&','"','\'','\'',' '), $mod_text);
|
|
| 341 | 269 |
$mod_text = '.'.trim($mod_text).'.'; |
| 342 |
// Do a fast scan over $mod_text first. This will speedup things a lot.
|
|
| 270 |
// Do a fast scan over $mod_text first. This may speedup things a lot.
|
|
| 343 | 271 |
if($func_search_match == 'all') {
|
| 344 | 272 |
if(!is_all_matched($mod_text, $func_search_words)) |
| 345 | 273 |
return false; |
| ... | ... | |
| 349 | 277 |
} |
| 350 | 278 |
// search for an better anchor - this have to be done before strip_tags() (may fail if search-string contains <, &, amp, gt, lt, ...) |
| 351 | 279 |
$anchor = make_url_target($mod_page_link_target, $anchor_text, $func_search_words); |
| 352 |
|
|
| 353 | 280 |
// make the link from $mod_page_link, add anchor |
| 354 | 281 |
$link = ""; |
| 355 | 282 |
$link = page_link($mod_page_link); |
| ... | ... | |
| 366 | 293 |
} |
| 367 | 294 |
$excerpt = prepare_excerpts($excerpt_array, $func_search_words, $mod_max_excerpt_num); |
| 368 | 295 |
} |
| 369 |
|
|
| 370 | 296 |
// handle thumbs - to deactivate this look in the module's search.php: $show_thumb (or maybe in the module's settings-page) |
| 371 | 297 |
if($mod_pic_link != "") {
|
| 372 |
$excerpt = '<table class="excerpt_thumb" width="100%" cellspacing="0" cellpadding="0" border="0"><tbody><tr><td width="110" valign="top"><a href="'.$link.'"><img src="'.WB_URL.'/'.MEDIA_DIRECTORY.$mod_pic_link.'" alt="" /></a></td><td>'.$excerpt.'</td></tr></tbody></table>'; |
|
| 298 |
if(isset($mod_special) && $mod_special=='lightbox2_plus') |
|
| 299 |
$excerpt = '<table class="excerpt_thumb" width="100%" cellspacing="0" cellpadding="0" border="0"><tbody><tr><td width="110" valign="top">'.$mod_special_piclink.'<img src="'.WB_URL.'/'.MEDIA_DIRECTORY.$mod_pic_link.'" alt="" /></a></td><td>'.$excerpt.'</td></tr></tbody></table>'; |
|
| 300 |
else |
|
| 301 |
$excerpt = '<table class="excerpt_thumb" width="100%" cellspacing="0" cellpadding="0" border="0"><tbody><tr><td width="110" valign="top"><a href="'.$link.'"><img src="'.WB_URL.'/'.MEDIA_DIRECTORY.$mod_pic_link.'" alt="" /></a></td><td>'.$excerpt.'</td></tr></tbody></table>'; |
|
| 373 | 302 |
} |
| 374 | 303 |
|
| 375 | 304 |
// print-out the excerpt |
| ... | ... | |
| 377 | 306 |
$values = array(); |
| 378 | 307 |
list($date, $time) = get_page_modified($mod_page_modified_when); |
| 379 | 308 |
list($username, $displayname) = get_page_modified_by($mod_page_modified_by, $func_users); |
| 380 |
$vars = array('[LINK]', '[TITLE]', '[DESCRIPTION]', '[USERNAME]','[DISPLAY_NAME]','[DATE]','[TIME]','[TEXT_LAST_UPDATED_BY]','[TEXT_ON]','[EXCERPT]');
|
|
| 309 |
$vars = array('[LINK]', '[TITLE]','[PAGE_TITLE]' ,'[MENU_TITLE]' , '[DESCRIPTION]', '[USERNAME]','[DISPLAY_NAME]','[DATE]','[TIME]','[TEXT_LAST_UPDATED_BY]','[TEXT_ON]','[EXCERPT]');
|
|
| 381 | 310 |
$values = array( |
| 382 | 311 |
$link, |
| 383 | 312 |
$mod_page_title, |
| 313 |
$func_page_title, |
|
| 314 |
$func_page_menu_title, |
|
| 384 | 315 |
$mod_page_description, |
| 385 | 316 |
$username, |
| 386 | 317 |
$displayname, |
| ... | ... | |
| 423 | 354 |
} |
| 424 | 355 |
|
| 425 | 356 |
// keeps only wanted entries in array $files. $str have to be an eregi()-compatible regex |
| 426 |
/* |
|
| 427 |
function clear_filelist($files, $str, $keep=true) |
|
| 428 |
{
|
|
| 357 |
function clear_filelist($files, $str, $keep=true) {
|
|
| 429 | 358 |
// options: $keep = true : remove all non-matching entries |
| 430 | 359 |
// $keep = false : remove all matching entries |
| 431 | 360 |
$c_filelist = array(); |
| 432 | 361 |
if($str == '') |
| 433 | 362 |
return $files; |
| 434 |
foreach($files as $file) |
|
| 435 |
{
|
|
| 436 |
if($keep) |
|
| 437 |
{
|
|
| 438 |
if(eregi($str, $file)) |
|
| 439 |
{
|
|
| 363 |
foreach($files as $file) {
|
|
| 364 |
if($keep) {
|
|
| 365 |
if(eregi($str, $file)) {
|
|
| 440 | 366 |
$c_filelist[] = $file; |
| 441 | 367 |
} |
| 442 | 368 |
} else {
|
| 443 |
if(!eregi($str, $file)) |
|
| 444 |
{
|
|
| 369 |
if(!eregi($str, $file)) {
|
|
| 445 | 370 |
$c_filelist[] = $file; |
| 446 | 371 |
} |
| 447 | 372 |
} |
| 448 | 373 |
} |
| 449 | 374 |
return($c_filelist); |
| 450 | 375 |
} |
| 451 |
*/ |
|
| 452 |
?> |
|
| 376 |
|
|
| 377 |
function search_make_sql_part($words, $match, $columns) {
|
|
| 378 |
// $words are utf-8 encoded, will be converted to DEFAULT_CHARSET below |
|
| 379 |
if(empty($words) || empty($columns)) return('(1=1)');
|
|
| 380 |
global $database; |
|
| 381 |
|
|
| 382 |
// check if we can use SQL'S "LIKE" |
|
| 383 |
// work-around for WB'S missing-SET-NAMES-problem |
|
| 384 |
static $checked = FALSE; |
|
| 385 |
if($checked===FALSE) {
|
|
| 386 |
$checked = TRUE; |
|
| 387 |
$lowers = array('utf8'=>"\xc3\xa1", 'iso'=>"\xe1");
|
|
| 388 |
$uppers = array('utf8'=>"\xc3\x81", 'iso'=>"\xc1");
|
|
| 389 |
switch(DEFAULT_CHARSET) {
|
|
| 390 |
case 'utf-8': |
|
| 391 |
$lo = $lowers['utf8']; |
|
| 392 |
$up = $uppers['utf8']; |
|
| 393 |
break; |
|
| 394 |
case 'iso-8859-1': |
|
| 395 |
case 'iso-8859-2': |
|
| 396 |
case 'iso-8859-3': |
|
| 397 |
case 'iso-8859-4': |
|
| 398 |
case 'iso-8859-5': |
|
| 399 |
case 'iso-8859-7': |
|
| 400 |
case 'iso-8859-9': |
|
| 401 |
case 'iso-8859-10': |
|
| 402 |
$lo = $lowers['iso']; |
|
| 403 |
$up = $uppers['iso']; |
|
| 404 |
break; |
|
| 405 |
default: |
|
| 406 |
$checked = 'check failed'; // we can't handle arabic,hebrew,thai |
|
| 407 |
} |
|
| 408 |
if($checked===TRUE && $query = $database->query("SELECT UPPER('$lo')='$up'")) {
|
|
| 409 |
$res = $query->fetchRow(); |
|
| 410 |
if($res[0]==0) {
|
|
| 411 |
$checked = 'check failed'; |
|
| 412 |
} |
|
| 413 |
} else |
|
| 414 |
$checked = 'check failed'; |
|
| 415 |
} |
|
| 416 |
|
|
| 417 |
require_once(WB_PATH.'/framework/functions-utf8.php'); |
|
| 418 |
global $search_table_sql_local; |
|
| 419 |
$altnum = count($search_table_sql_local); |
|
| 420 |
|
|
| 421 |
if($match=='all') $op = 'AND'; |
|
| 422 |
else $op = ' OR'; // keep the leading space! |
|
| 423 |
|
|
| 424 |
// create sql-template |
|
| 425 |
$sql = ''; |
|
| 426 |
$i = 0; |
|
| 427 |
foreach($words as $w) {
|
|
| 428 |
if(empty($w)) continue; |
|
| 429 |
$w_alts = $e_alts = array(); |
|
| 430 |
if($altnum) {
|
|
| 431 |
for($x=0;$x<$altnum;$x++) |
|
| 432 |
$w_alts[$x] = strtr($w, $search_table_sql_local[$x]); |
|
| 433 |
} else {
|
|
| 434 |
$w_alts[0] = $w; |
|
| 435 |
} |
|
| 436 |
$w_alts = array_unique($w_alts); |
|
| 437 |
foreach($w_alts as $a) {
|
|
| 438 |
$tmp = htmlentities($a, ENT_COMPAT, 'UTF-8'); |
|
| 439 |
// if the missing-SET-NAMES-issue appears and $tmp contains non-ascii characters: exit and use the normal (slow) search-function instead |
|
| 440 |
if($checked!==TRUE && preg_match('/[\x80-\xFF]/', $tmp)) return('(1=1)'); // missing-SET-NAMES-issue
|
|
| 441 |
$e_alts[] = $tmp; |
|
| 442 |
} |
|
| 443 |
$sql .= ""; |
|
| 444 |
foreach($w_alts as $a) |
|
| 445 |
$sql .= "{{COL}} LIKE '%".addslashes(utf8_to_charset($a))."%' OR ";
|
|
| 446 |
if(isset($e_alts[$i]) && $e_alts[$i]!=$w) |
|
| 447 |
$sql .= " {{COL}} LIKE '%".addslashes($e_alts[$i])."%'";
|
|
| 448 |
else {
|
|
| 449 |
$sql = substr($sql, 0, strlen($sql)-4); |
|
| 450 |
$sql .= ''; |
|
| 451 |
} |
|
| 452 |
$sql .= " $op "; |
|
| 453 |
$i++; |
|
| 454 |
} |
|
| 455 |
$sql = substr($sql, 0, strlen($sql)-5); |
|
| 456 |
$sql_template = $sql; |
|
| 457 |
|
|
| 458 |
// create SQL-string from template |
|
| 459 |
$sql = '(';
|
|
| 460 |
foreach($columns as $c) {
|
|
| 461 |
$sql .= '(';
|
|
| 462 |
$sql .= str_replace('{{COL}}', $c, $sql_template);
|
|
| 463 |
$sql .= ") OR "; |
|
| 464 |
} |
|
| 465 |
$sql = substr($sql, 0, strlen($sql)-4); |
|
| 466 |
$sql .= ')'; |
|
| 467 |
|
|
| 468 |
return($sql); |
|
| 469 |
} |
|
Also available in: Unified diff