mysociety /
theyworkforyou
| Total Complexity | 150 |
| Total Lines | 580 |
| Duplicated Lines | 0 % |
| Coverage | 0% |
| Changes | 1 | ||
| Bugs | 0 | Features | 0 |
Complex classes like SEARCHENGINE often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use SEARCHENGINE, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 36 | class SEARCHENGINE { |
||
| 37 | public $valid = false; |
||
| 38 | public $error; |
||
| 39 | |||
| 40 | public function __construct($query) { |
||
| 41 | if (!defined('XAPIANDB') || !XAPIANDB) |
||
|
1 ignored issue
–
show
|
|||
| 42 | return null; |
||
| 43 | |||
| 44 | global $xapiandb, $PAGE, $hansardmajors, $parties; |
||
| 45 | if (!$xapiandb) { |
||
| 46 | if (strstr(XAPIANDB, ":")) { |
||
| 47 | //ini_set('display_errors', 'On'); |
||
| 48 | list ($xapian_host, $xapian_port) = explode(":", XAPIANDB); |
||
| 49 | twfy_debug("SEARCH", "Using Xapian remote backend: " . $xapian_host . " port " . $xapian_port); |
||
| 50 | $xapiandb_remote = remote_open($xapian_host, intval($xapian_port)); |
||
| 51 | $xapiandb = new XapianDatabase($xapiandb_remote); |
||
|
1 ignored issue
–
show
|
|||
| 52 | } else { |
||
| 53 | $xapiandb = new XapianDatabase(XAPIANDB); |
||
| 54 | } |
||
| 55 | } |
||
| 56 | $this->query = $query; |
||
| 57 | if (!isset($this->stemmer)) $this->stemmer = new XapianStem('english'); |
||
|
1 ignored issue
–
show
|
|||
| 58 | if (!isset($this->enquire)) $this->enquire = new XapianEnquire($xapiandb); |
||
|
1 ignored issue
–
show
|
|||
| 59 | if (!isset($this->queryparser)) { |
||
| 60 | $this->queryparser = new XapianQueryParser(); |
||
|
1 ignored issue
–
show
|
|||
| 61 | $this->datevaluerange = new XapianDateValueRangeProcessor(1); |
||
|
1 ignored issue
–
show
|
|||
| 62 | $this->queryparser->set_stemmer($this->stemmer); |
||
| 63 | $this->queryparser->set_stemming_strategy(XapianQueryParser::STEM_SOME); |
||
| 64 | $this->queryparser->set_database($xapiandb); |
||
| 65 | $this->queryparser->set_default_op(Query_OP_AND); |
||
|
1 ignored issue
–
show
|
|||
| 66 | $this->queryparser->add_boolean_prefix('speaker', 'S'); |
||
| 67 | $this->queryparser->add_boolean_prefix('major', 'M'); |
||
| 68 | $this->queryparser->add_boolean_prefix('date', 'D'); |
||
| 69 | $this->queryparser->add_boolean_prefix('batch', 'B'); |
||
| 70 | $this->queryparser->add_boolean_prefix('segment', 'U'); |
||
| 71 | $this->queryparser->add_boolean_prefix('department', 'G'); |
||
| 72 | $this->queryparser->add_boolean_prefix('party', 'P'); |
||
| 73 | $this->queryparser->add_boolean_prefix('column', 'C'); |
||
| 74 | $this->queryparser->add_boolean_prefix('gid', 'Q'); |
||
| 75 | $this->queryparser->add_valuerangeprocessor($this->datevaluerange); |
||
| 76 | } |
||
| 77 | |||
| 78 | # Force words to lower case |
||
| 79 | $this->query = preg_replace('#(department|party):.+?\b#ie', 'strtolower("$0")', $this->query); |
||
| 80 | |||
| 81 | // Any characters other than this are treated as, basically, white space |
||
| 82 | // (apart from quotes and minuses, special case below) |
||
| 83 | // The colon is in here for prefixes speaker:10043 and so on. |
||
| 84 | $this->wordchars = "A-Za-z0-9,.'&:_\x80-\xbf\xc2-\xf4"; |
||
| 85 | $this->wordcharsnodigit = "A-Za-z0-9'&_\x80-\xbf\xc2-\xf4"; |
||
| 86 | |||
| 87 | // An array of normal words. |
||
| 88 | $this->words = array(); |
||
| 89 | // All quoted phrases, as an (array of (arrays of words in each phrase)). |
||
| 90 | $this->phrases = array(); |
||
| 91 | // Items prefixed with a colon (speaker:10024) as an (array of (name, value)) |
||
| 92 | $this->prefixed = array(); |
||
| 93 | |||
| 94 | // Split words up into individual words, and quoted phrases |
||
| 95 | preg_match_all('/(' . |
||
| 96 | '"|' . # match either a quote, or... |
||
| 97 | '(?:(?<![' .$this->wordchars. '])-)?' . # optionally a - (exclude) |
||
| 98 | # if at start of word (i.e. not preceded by a word character, in |
||
| 99 | # which case it is probably a hyphenated-word) |
||
| 100 | '['.$this->wordchars.']+' . # followed by a string of word-characters |
||
| 101 | ')/', $this->query, $all_words); |
||
| 102 | if ($all_words) { |
||
| 103 | $all_words = $all_words[0]; |
||
| 104 | } else { |
||
| 105 | $all_words = array(); |
||
| 106 | } |
||
| 107 | $in_quote = false; |
||
| 108 | $from = ''; $to = ''; |
||
| 109 | foreach ($all_words as $word) { |
||
| 110 | if ($word == '"') { |
||
| 111 | $in_quote = !$in_quote; |
||
| 112 | if ($in_quote) array_push($this->phrases, array()); |
||
| 113 | if (!$in_quote && !count($this->phrases[count($this->phrases) - 1])) { |
||
| 114 | array_pop($this->phrases); |
||
| 115 | } |
||
| 116 | continue; |
||
| 117 | } |
||
| 118 | if ($word == '') { |
||
| 119 | continue; |
||
| 120 | } |
||
| 121 | |||
| 122 | if (strpos($word, ':') !== false) { |
||
| 123 | $items = explode(":", strtolower($word)); |
||
| 124 | $type = $items[0]; |
||
| 125 | if (substr($type, 0, 1)=='-') $type = substr($type, 1); |
||
| 126 | $value = strtolower(join(":", array_slice($items,1))); |
||
| 127 | if ($type == 'section') { |
||
| 128 | $newv = $value; |
||
| 129 | if ($value == 'debates' || $value == 'debate') $newv = 1; |
||
| 130 | elseif ($value == 'whall' || $value == 'westminster' || $value == 'westminhall') $newv = 2; |
||
| 131 | elseif ($value == 'wrans' || $value == 'wran') $newv = 3; |
||
| 132 | elseif ($value == 'wms' || $value == 'statements' || $value == 'statement') $newv = 4; |
||
| 133 | elseif ($value == 'lordsdebates' || $value == 'lords') $newv = 101; |
||
| 134 | elseif ($value == 'ni' || $value == 'nidebates') $newv = 5; |
||
| 135 | elseif ($value == 'pbc' || $value == 'standing') $newv = 6; |
||
| 136 | elseif ($value == 'sp') $newv = 7; |
||
| 137 | elseif ($value == 'spwrans' || $value == 'spwran') $newv = 8; |
||
| 138 | elseif ($value == 'uk') $newv = array(1,2,3,4,6,101); |
||
| 139 | elseif ($value == 'scotland') $newv = array(7,8); |
||
| 140 | elseif ($value == 'future') $newv = 'F'; |
||
| 141 | if (is_array($newv)) { |
||
| 142 | $newv = 'major:' . join(' major:', $newv); |
||
| 143 | } else { |
||
| 144 | $newv = "major:$newv"; |
||
| 145 | } |
||
| 146 | $this->query = str_ireplace("$type:$value", $newv, $this->query); |
||
| 147 | } elseif ($type == 'groupby') { |
||
| 148 | $newv = $value; |
||
| 149 | if ($value == 'debates' || $value == 'debate') $newv = 'debate'; |
||
| 150 | if ($value == 'speech' || $value == 'speeches') $newv = 'speech'; |
||
| 151 | $this->query = str_ireplace("$type:$value", '', $this->query); |
||
| 152 | array_push($this->prefixed, array($type, $newv)); |
||
| 153 | } elseif ($type == 'from') { |
||
| 154 | $from = $value; |
||
| 155 | } elseif ($type == 'to') { |
||
| 156 | $to = $value; |
||
| 157 | } |
||
| 158 | } elseif (strpos($word, '-') !== false) { |
||
| 159 | } elseif ($in_quote) { |
||
| 160 | array_push($this->phrases[count($this->phrases) - 1], strtolower($word)); |
||
| 161 | } elseif (strpos($word, '..') !== false) { |
||
| 162 | } elseif ($word == 'OR' || $word == 'AND' || $word == 'XOR' || $word == 'NEAR') { |
||
| 163 | } else { |
||
| 164 | array_push($this->words, strtolower($word)); |
||
| 165 | } |
||
| 166 | } |
||
| 167 | if ($from && $to) { |
||
| 168 | $this->query = str_ireplace("from:$from", '', $this->query); |
||
| 169 | $this->query = str_ireplace("to:$to", '', $this->query); |
||
| 170 | $this->query .= " $from..$to"; |
||
| 171 | } elseif ($from) { |
||
| 172 | $this->query = str_ireplace("from:$from", '', $this->query); |
||
| 173 | $this->query .= " $from..".date('Ymd'); |
||
| 174 | } elseif ($to) { |
||
| 175 | $this->query = str_ireplace("to:$to", '', $this->query); |
||
| 176 | $this->query .= " 19990101..$to"; |
||
| 177 | } |
||
| 178 | |||
| 179 | # Merged people |
||
| 180 | $db = new ParlDB; |
||
| 181 | $merged = $db->query('SELECT * FROM gidredirect WHERE gid_from LIKE :gid_from', array(':gid_from' => "uk.org.publicwhip/person/%")); |
||
| 182 | for ($n=0; $n<$merged->rows(); $n++) { |
||
| 183 | $from_id = str_replace('uk.org.publicwhip/person/', '', $merged->field($n, 'gid_from')); |
||
| 184 | $to_id = str_replace('uk.org.publicwhip/person/', '', $merged->field($n, 'gid_to')); |
||
| 185 | $this->query = preg_replace("#speaker:($from_id|$to_id)#i", "(speaker:$from_id OR speaker:$to_id)", $this->query); |
||
| 186 | } |
||
| 187 | |||
| 188 | twfy_debug("SEARCH", "prefixed: " . var_export($this->prefixed, true)); |
||
| 189 | |||
| 190 | twfy_debug("SEARCH", "query -- ". $this->query); |
||
| 191 | $flags = XapianQueryParser::FLAG_BOOLEAN | XapianQueryParser::FLAG_LOVEHATE | |
||
| 192 | XapianQueryParser::FLAG_WILDCARD | XapianQueryParser::FLAG_SPELLING_CORRECTION; |
||
| 193 | $flags = $flags | XapianQueryParser::FLAG_PHRASE; |
||
| 194 | try { |
||
| 195 | $query = $this->queryparser->parse_query($this->query, $flags); |
||
| 196 | } catch (Exception $e) { |
||
| 197 | # Nothing we can really do with a bad query |
||
| 198 | $this->error = _htmlspecialchars($e->getMessage()); |
||
| 199 | |||
| 200 | return null; |
||
| 201 | } |
||
| 202 | |||
| 203 | $this->enquire->set_query($query); |
||
| 204 | |||
| 205 | # Now parse the parsed query back into a query string, yummy |
||
| 206 | |||
| 207 | $qd = $query->get_description(); |
||
| 208 | twfy_debug("SEARCH", "queryparser original description -- " . $qd); |
||
| 209 | $qd = substr($qd, 14, -1); # Strip Xapian::Query() |
||
| 210 | $qd = preg_replace('#:\(.*?\)#', '', $qd); # Don't need pos or weight |
||
| 211 | # Date range |
||
| 212 | $qd = preg_replace('#VALUE_RANGE 1 (\d+) (\d+)#e', 'preg_replace("#(\d{4})(\d\d)(\d\d)#", "\$3/\$2/\$1", $1) |
||
| 213 | . ".." . preg_replace("#(\d{4})(\d\d)(\d\d)#", "\$3/\$2/\$1", $2)', $qd); |
||
| 214 | # Replace phrases with the phrase in quotes |
||
| 215 | preg_match_all('#\(([^(]*? PHRASE [^(]*?)\)#', $qd, $m); |
||
| 216 | foreach ($m[1] as $phrase) { |
||
| 217 | $phrase_new = preg_replace('# PHRASE \d+#', '', $phrase); |
||
| 218 | #$this->phrases[] = preg_split('#\s+#', $phrase_new); |
||
| 219 | $qd = str_replace("($phrase)", '"'.$phrase_new.'"', $qd); |
||
| 220 | } |
||
| 221 | preg_match_all('#\(([^(]*? NEAR [^(]*?)\)#', $qd, $m); |
||
| 222 | foreach ($m[1] as $mm) { |
||
| 223 | $mmn = preg_replace('# NEAR \d+ #', ' NEAR ', $mm); |
||
| 224 | $qd = str_replace("($mm)", "($mmn)", $qd); |
||
| 225 | } |
||
| 226 | # Awesome regexes to get rid of superfluous matching brackets |
||
| 227 | $qd = preg_replace('/( \( ( (?: (?>[^ ()]+) | (?1) ) (?: [ ](?:AND|OR|XOR|FILTER|NEAR[ ]\d+|PHRASE[ ]\d+)[ ] (?: (?>[^ ()]+) | (?1) ) )* ) \) ) [ ] (FILTER|AND_NOT)/x', '$2 $3', $qd); |
||
| 228 | $qd = preg_replace('/(?:FILTER | 0 [ ] \* ) [ ] ( \( ( (?: (?>[^ ()]+) | (?1) ) (?: [ ](?:AND|OR|XOR)[ ] (?: (?>[^ ()]+) | (?1) ) )* ) \) )/x', '$2', $qd); |
||
| 229 | $qd = preg_replace('/(?:FILTER | 0 [ ] \* ) [ ] ( [^()] )/x', '$1', $qd); |
||
| 230 | $qd = str_replace('AND ', '', $qd); # AND is the default |
||
| 231 | $qd = preg_replace('/^ ( \( ( (?: (?>[^()]+) | (?1) )* ) \) ) $/x', '$2', $qd); |
||
| 232 | # Other prefixes |
||
| 233 | $qd = preg_replace('#\bU(\d+)\b#', 'segment:$1', $qd); |
||
| 234 | $qd = preg_replace('#\bC(\d+)\b#', 'column:$1', $qd); |
||
| 235 | $qd = preg_replace('#\bQ(.*?)\b#', 'gid:$1', $qd); |
||
| 236 | $qd = preg_replace('#\bP(.*?)\b#e', '"party:" . (isset($parties[ucfirst("$1")]) ? $parties[ucfirst("$1")] : "$1")', $qd); |
||
| 237 | $qd = preg_replace('#\bD(.*?)\b#', 'date:$1', $qd); |
||
| 238 | $qd = preg_replace('#\bG(.*?)\b#', 'department:$1', $qd); # XXX Lookup to show proper name of dept |
||
| 239 | if (strstr($qd, 'M1 OR M2 OR M3 OR M4 OR M6 OR M101')) { |
||
| 240 | $qd = str_replace('M1 OR M2 OR M3 OR M4 OR M6 OR M101', 'section:uk', $qd); |
||
| 241 | } elseif (strstr($qd, 'M7 OR M8')) { |
||
| 242 | $qd = str_replace('M7 OR M8', 'section:scotland', $qd); |
||
| 243 | } |
||
| 244 | $qd = preg_replace('#\bM(\d+)\b#e', '"in the \'" . (isset($hansardmajors[$1]["title"]) ? $hansardmajors[$1]["title"] . "\'" : "$1")', $qd); |
||
| 245 | $qd = preg_replace('#\bMF\b#', 'in Future Business', $qd); |
||
| 246 | |||
| 247 | # Replace stemmed things with their unstemmed terms from the query |
||
| 248 | $used = array(); |
||
| 249 | preg_match_all('#Z[^\s()]+#', $qd, $m); |
||
| 250 | foreach ($m[0] as $mm) { |
||
| 251 | $iter = $this->queryparser->unstem_begin($mm); |
||
| 252 | $end = $this->queryparser->unstem_end($mm); |
||
| 253 | while (!$iter->equals($end)) { |
||
| 254 | $tt = $iter->get_term(); |
||
| 255 | if (!in_array($tt, $used)) break; |
||
| 256 | $iter->next(); |
||
| 257 | } |
||
| 258 | $used[] = $tt; |
||
| 259 | $qd = preg_replace('#' . preg_quote($mm, '#') . '#', $tt, $qd, 1); |
||
| 260 | } |
||
| 261 | |||
| 262 | # Speakers |
||
| 263 | for ($n=0; $n<$merged->rows(); $n++) { |
||
| 264 | $from_id = str_replace('uk.org.publicwhip/person/', '', $merged->field($n, 'gid_from')); |
||
| 265 | $to_id = str_replace('uk.org.publicwhip/person/', '', $merged->field($n, 'gid_to')); |
||
| 266 | $qd = str_replace("(S$from_id OR S$to_id)", "S$to_id", $qd); |
||
| 267 | $qd = str_replace("S$from_id OR S$to_id", "S$to_id", $qd); |
||
| 268 | } |
||
| 269 | |||
| 270 | preg_match_all('#S(\d+)#', $qd, $m); |
||
| 271 | foreach ($m[1] as $mm) { |
||
| 272 | $member = new MEMBER(array('person_id' => $mm)); |
||
| 273 | $name = $member->full_name(); |
||
| 274 | $qd = str_replace("S$mm", "speaker:$name", $qd); |
||
| 275 | } |
||
| 276 | |||
| 277 | # Simplify display of excluded words |
||
| 278 | $qd = preg_replace('#AND_NOT ([a-z0-9"]+)#', '-$1', $qd); |
||
| 279 | preg_match_all('#AND_NOT \((.*?)\)#', $qd, $m); |
||
| 280 | foreach ($m[1] as $mm) { |
||
| 281 | $mmn = '-' . join(' -', explode(' OR ', $mm)); |
||
| 282 | $qd = str_replace("AND_NOT ($mm)", $mmn, $qd); |
||
| 283 | } |
||
| 284 | |||
| 285 | foreach ($this->prefixed as $items) { |
||
| 286 | if ($items[0] == 'groupby') { |
||
| 287 | if ($items[1] == 'debate') { |
||
| 288 | $qd .= ' grouped by debate'; |
||
| 289 | } elseif ($items[1] == 'speech') { |
||
| 290 | $qd .= ' showing all speeches'; |
||
| 291 | } else { |
||
| 292 | $PAGE->error_message("Unknown group by '$items[1]' ignored"); |
||
| 293 | } |
||
| 294 | } |
||
| 295 | } |
||
| 296 | |||
| 297 | $this->query_desc = trim($qd); |
||
| 298 | |||
| 299 | #print 'DEBUG: ' . $query->get_description(); |
||
| 300 | twfy_debug("SEARCH", "words: " . var_export($this->words, true)); |
||
| 301 | twfy_debug("SEARCH", "phrases: " . var_export($this->phrases, true)); |
||
| 302 | twfy_debug("SEARCH", "queryparser description -- " . $this->query_desc); |
||
| 303 | |||
| 304 | $this->valid = true; |
||
| 305 | } |
||
| 306 | |||
| 307 | public function query_description_internal($long) { |
||
| 308 | if (!defined('XAPIANDB') || !XAPIANDB) { |
||
|
1 ignored issue
–
show
|
|||
| 309 | return ''; |
||
| 310 | } |
||
| 311 | if (!$this->valid) { |
||
| 312 | return '[bad query]'; |
||
| 313 | } |
||
| 314 | |||
| 315 | return $this->query_desc; |
||
| 316 | } |
||
| 317 | |||
| 318 | // Return textual description of search |
||
| 319 | public function query_description_short() { |
||
| 320 | return $this->query_description_internal(false); |
||
| 321 | } |
||
| 322 | |||
| 323 | // Return textual description of search |
||
| 324 | public function query_description_long() { |
||
| 325 | return $this->query_description_internal(true); |
||
| 326 | } |
||
| 327 | |||
| 328 | // Return stem of a word |
||
| 329 | public function stem($word) { |
||
| 330 | return $this->stemmer->apply(strtolower($word)); |
||
| 331 | } |
||
| 332 | |||
| 333 | public function get_spelling_correction() { |
||
| 334 | if (!defined('XAPIANDB') || !XAPIANDB) |
||
|
1 ignored issue
–
show
|
|||
| 335 | return null; |
||
| 336 | |||
| 337 | $qd = $this->queryparser->get_corrected_query_string(); |
||
| 338 | return $qd; |
||
| 339 | } |
||
| 340 | |||
| 341 | // Perform partial query to get a count of number of matches |
||
| 342 | public function run_count($first_result, $results_per_page, $sort_order='relevance') { |
||
| 343 | if (!defined('XAPIANDB') || !XAPIANDB) |
||
|
1 ignored issue
–
show
|
|||
| 344 | return null; |
||
| 345 | |||
| 346 | $start = getmicrotime(); |
||
| 347 | |||
| 348 | switch ($sort_order) { |
||
| 349 | case 'date': |
||
| 350 | case 'newest': |
||
| 351 | $this->enquire->set_sort_by_value(0, true); |
||
| 352 | break; |
||
| 353 | case 'oldest': |
||
| 354 | $this->enquire->set_sort_by_value(0, false); |
||
| 355 | break; |
||
| 356 | case 'created': |
||
| 357 | $this->enquire->set_sort_by_value(2); |
||
| 358 | default: |
||
| 359 | //do nothing, default ordering is by relevance |
||
| 360 | break; |
||
| 361 | } |
||
| 362 | |||
| 363 | // Set collapsing and sorting |
||
| 364 | global $PAGE; |
||
| 365 | $collapsed = false; |
||
| 366 | if (preg_match('#(speaker|segment):\d+#', $this->query)) { |
||
| 367 | $collapsed = true; |
||
| 368 | } |
||
| 369 | foreach ($this->prefixed as $items) { |
||
| 370 | if ($items[0] == 'groupby') { |
||
| 371 | $collapsed = true; |
||
| 372 | if ($items[1] == 'speech') |
||
| 373 | ; // no collapse key |
||
| 374 | elseif ($items[1] == 'debate') |
||
| 375 | $this->enquire->set_collapse_key(3); |
||
| 376 | else |
||
| 377 | $PAGE->error_message("Unknown group by '$items[1]' ignored"); |
||
| 378 | } |
||
| 379 | } |
||
| 380 | |||
| 381 | // default to grouping by subdebate, i.e. by page |
||
| 382 | if (!$collapsed) |
||
| 383 | $this->enquire->set_collapse_key(3); |
||
| 384 | |||
| 385 | /* |
||
| 386 | XXX Helping to debug possible Xapian bug |
||
| 387 | foreach (array(0, 50, 100, 200, 300, 400, 460) as $fff) { |
||
| 388 | foreach (array(0, 100, 300, 500, 1000) as $cal) { |
||
| 389 | print "get_mset($fff, 20, $cal): "; |
||
| 390 | $m = $this->enquire->get_mset($fff, 20, $cal); |
||
| 391 | print $m->get_matches_estimated(). ' '; |
||
| 392 | print $m->get_matches_lower_bound() . ' '; |
||
| 393 | print $m->get_matches_upper_bound() . "\n"; |
||
| 394 | } |
||
| 395 | } |
||
| 396 | */ |
||
| 397 | |||
| 398 | #$matches = $this->enquire->get_mset(0, 500); |
||
| 399 | $this->matches = $this->enquire->get_mset($first_result, $results_per_page, 100); |
||
| 400 | // Take either: 1) the estimate which is sometimes too large or 2) the |
||
| 401 | // size which is sometimes too low (it is limited to the 500 in the line |
||
| 402 | // above). We get the exact mset we need later, according to which page |
||
| 403 | // we are on. |
||
| 404 | #if ($matches->size() < 500) { |
||
| 405 | #$count = $matches->size(); |
||
| 406 | #} else { |
||
| 407 | $count = $this->matches->get_matches_estimated(); |
||
| 408 | # print "DEBUG bounds: "; |
||
| 409 | # print $this->matches->get_matches_lower_bound(); |
||
| 410 | # print ' - '; |
||
| 411 | # print $this->matches->get_matches_upper_bound(); |
||
| 412 | #} |
||
| 413 | |||
| 414 | $duration = getmicrotime() - $start; |
||
| 415 | twfy_debug ("SEARCH", "Search count took $duration seconds."); |
||
| 416 | |||
| 417 | return $count; |
||
| 418 | } |
||
| 419 | |||
| 420 | // Perform the full search... |
||
| 421 | public function run_search($first_result, $results_per_page, $sort_order='relevance') { |
||
| 422 | $start = getmicrotime(); |
||
| 423 | |||
| 424 | #$matches = $this->enquire->get_mset($first_result, $results_per_page); |
||
| 425 | $matches = $this->matches; |
||
| 426 | $this->gids = array(); |
||
| 427 | $this->created = array(); |
||
| 428 | $this->collapsed = array(); |
||
| 429 | $this->relevances = array(); |
||
| 430 | $iter = $matches->begin(); |
||
| 431 | $end = $matches->end(); |
||
| 432 | while (!$iter->equals($end)) { |
||
| 433 | $relevancy = $iter->get_percent(); |
||
| 434 | $weight = $iter->get_weight(); |
||
| 435 | $collapsed = $iter->get_collapse_count(); |
||
| 436 | $doc = $iter->get_document(); |
||
| 437 | $gid = $doc->get_data(); |
||
| 438 | if ($sort_order == 'created') { |
||
| 439 | array_push($this->created, join('', unpack('N', $doc->get_value(2)))); # XXX Needs fixing |
||
| 440 | } |
||
| 441 | twfy_debug("SEARCH", "gid: $gid relevancy: $relevancy% weight: $weight"); |
||
| 442 | array_push($this->gids, "uk.org.publicwhip/".$gid); |
||
| 443 | array_push($this->collapsed, $collapsed); |
||
| 444 | array_push($this->relevances, $relevancy); |
||
| 445 | $iter->next(); |
||
| 446 | } |
||
| 447 | $duration = getmicrotime() - $start; |
||
| 448 | twfy_debug ("SEARCH", "Run search took $duration seconds."); |
||
| 449 | } |
||
| 450 | // ... use these to get the results |
||
| 451 | public function get_gids() { |
||
| 453 | } |
||
| 454 | public function get_relevances() { |
||
| 455 | return $this->relevances; |
||
| 456 | } |
||
| 457 | public function get_createds() { |
||
| 458 | return $this->created; |
||
| 459 | } |
||
| 460 | |||
| 461 | // Puts HTML highlighting round all the matching words in the text |
||
| 462 | public function highlight($body) { |
||
| 463 | if (!defined('XAPIANDB') || !XAPIANDB) |
||
|
1 ignored issue
–
show
|
|||
| 464 | return $body; |
||
| 465 | |||
| 466 | $stemmed_words = array_map(array($this, 'stem'), $this->words); |
||
| 467 | if (is_array($body)) { |
||
| 468 | foreach ($body as $k => $b) { |
||
| 469 | $body[$k] = $this->highlight_internal($b, $stemmed_words); |
||
| 470 | } |
||
| 471 | |||
| 472 | return $body; |
||
| 473 | } else { |
||
| 474 | return $this->highlight_internal($body, $stemmed_words); |
||
| 475 | } |
||
| 476 | } |
||
| 477 | |||
| 478 | private $specialchars = array('<', '>', '"', '&'); |
||
| 479 | private $specialchars_upper = array('<', '>', '"', '&'); |
||
| 480 | |||
| 481 | public function highlight_internal($body, $stemmed_words) { |
||
| 546 | } |
||
| 547 | |||
| 548 | // Find the position of the first of the search words/phrases in $body. |
||
| 549 | public function position_of_first_word($body) { |
||
| 616 | } |
||
| 617 | } |
||
| 618 | |||
| 619 | global $SEARCHENGINE; |
||
| 620 | $SEARCHENGINE = null; |
||
| 621 |
Loading history...
The PSR-1: Basic Coding Standard recommends that a file should either introduce new symbols, that is classes, functions, constants or similar, or have side effects. Side effects are anything that executes logic, like for example printing output, changing ini settings or writing to a file.
The idea behind this recommendation is that merely auto-loading a class should not change the state of an application. It also promotes a cleaner style of programming and makes your code less prone to errors, because the logic is not spread out all over the place.
To learn more about the PSR-1, please see the PHP-FIG site on the PSR-1.