summaryrefslogtreecommitdiff
path: root/includes/search/SearchEngine.php
diff options
context:
space:
mode:
Diffstat (limited to 'includes/search/SearchEngine.php')
-rw-r--r--includes/search/SearchEngine.php83
1 files changed, 45 insertions, 38 deletions
diff --git a/includes/search/SearchEngine.php b/includes/search/SearchEngine.php
index 27a321ac..6b3e62b1 100644
--- a/includes/search/SearchEngine.php
+++ b/includes/search/SearchEngine.php
@@ -45,7 +45,7 @@ class SearchEngine {
*/
protected $db;
- function __construct($db = null) {
+ function __construct( $db = null ) {
if ( $db ) {
$this->db = $db;
} else {
@@ -58,7 +58,7 @@ class SearchEngine {
* If title searches are not supported or disabled, return null.
* STUB
*
- * @param $term String: raw search term
+ * @param string $term raw search term
* @return SearchResultSet
*/
function searchText( $term ) {
@@ -70,7 +70,7 @@ class SearchEngine {
* If title searches are not supported or disabled, return null.
* STUB
*
- * @param $term String: raw search term
+ * @param string $term raw search term
* @return SearchResultSet
*/
function searchTitle( $term ) {
@@ -118,7 +118,7 @@ class SearchEngine {
* on text to be used for searching or updating search index.
* Default implementation does nothing (simply returns $string).
*
- * @param $string string: String to process
+ * @param string $string String to process
* @return string
*/
public function normalizeText( $string ) {
@@ -163,7 +163,7 @@ class SearchEngine {
/**
* Really find the title match.
- * @return null|\Title
+ * @return null|Title
*/
private static function getNearMatchInternal( $searchterm ) {
global $wgContLang, $wgEnableSearchContributorsByIP;
@@ -183,10 +183,15 @@ class SearchEngine {
# Exact match? No need to look further.
$title = Title::newFromText( $term );
- if ( is_null( $title ) ){
+ if ( is_null( $title ) ) {
return null;
}
+ # Try files if searching in the Media: namespace
+ if ( $title->getNamespace() == NS_MEDIA ) {
+ $title = Title::makeTitle( NS_FILE, $title->getText() );
+ }
+
if ( $title->isSpecialPage() || $title->isExternal() || $title->exists() ) {
return $title;
}
@@ -197,22 +202,23 @@ class SearchEngine {
return $title;
}
+ if ( !wfRunHooks( 'SearchAfterNoDirectMatch', array( $term, &$title ) ) ) {
+ return $title;
+ }
+
# Now try all lower case (i.e. first letter capitalized)
- #
$title = Title::newFromText( $wgContLang->lc( $term ) );
if ( $title && $title->exists() ) {
return $title;
}
# Now try capitalized string
- #
$title = Title::newFromText( $wgContLang->ucwords( $term ) );
if ( $title && $title->exists() ) {
return $title;
}
# Now try all upper case
- #
$title = Title::newFromText( $wgContLang->uc( $term ) );
if ( $title && $title->exists() ) {
return $title;
@@ -233,7 +239,6 @@ class SearchEngine {
$title = Title::newFromText( $searchterm );
-
# Entering an IP address goes to the contributions page
if ( $wgEnableSearchContributorsByIP ) {
if ( ( $title->getNamespace() == NS_USER && User::isIP( $title->getText() ) )
@@ -242,7 +247,6 @@ class SearchEngine {
}
}
-
# Entering a user goes to the user page whether it's there or not
if ( $title->getNamespace() == NS_USER ) {
return $title;
@@ -669,7 +673,6 @@ class SearchResultTooMany {
# # Some search engines may bail out if too many matches are found
}
-
/**
* @todo FIXME: This class is horribly factored. It would probably be better to
* have a useful base class to which you pass some standard information, then
@@ -791,21 +794,26 @@ class SearchResult {
*/
protected function initText() {
if ( !isset( $this->mText ) ) {
- if ( $this->mRevision != null )
- $this->mText = $this->mRevision->getText();
- else // TODO: can we fetch raw wikitext for commons images?
+ if ( $this->mRevision != null ) {
+ //TODO: if we could plug in some code that knows about special content models *and* about
+ // special features of the search engine, the search could benefit.
+ $content = $this->mRevision->getContent();
+ $this->mText = $content ? $content->getTextForSearchIndex() : '';
+ } else { // TODO: can we fetch raw wikitext for commons images?
$this->mText = '';
-
+ }
}
}
/**
- * @param $terms Array: terms to highlight
+ * @param array $terms terms to highlight
* @return String: highlighted text snippet, null (and not '') if not supported
*/
function getTextSnippet( $terms ) {
global $wgUser, $wgAdvancedSearchHighlighting;
$this->initText();
+
+ // TODO: make highliter take a content object. Make ContentHandler a factory for SearchHighliter.
list( $contextlines, $contextchars ) = SearchEngine::userHighlightPrefs( $wgUser );
$h = new SearchHighlighter();
if ( $wgAdvancedSearchHighlighting )
@@ -815,7 +823,7 @@ class SearchResult {
}
/**
- * @param $terms Array: terms to highlight
+ * @param array $terms terms to highlight
* @return String: highlighted title, '' if not supported
*/
function getTitleSnippet( $terms ) {
@@ -823,7 +831,7 @@ class SearchResult {
}
/**
- * @param $terms Array: terms to highlight
+ * @param array $terms terms to highlight
* @return String: highlighted redirect name (redirect to this page), '' if none or not supported
*/
function getRedirectSnippet( $terms ) {
@@ -934,7 +942,7 @@ class SearchHighlighter {
* Default implementation of wikitext highlighting
*
* @param $text String
- * @param $terms Array: terms to highlight (unescaped)
+ * @param array $terms terms to highlight (unescaped)
* @param $contextlines Integer
* @param $contextchars Integer
* @return String
@@ -962,7 +970,7 @@ class SearchHighlighter {
}
$spat .= '/';
$textExt = array(); // text extracts
- $otherExt = array(); // other extracts
+ $otherExt = array(); // other extracts
wfProfileIn( "$fname-split" );
$start = 0;
$textLen = strlen( $text );
@@ -1131,8 +1139,8 @@ class SearchHighlighter {
// add more lines
$add = $index + 1;
while ( $len < $targetchars - 20
- && array_key_exists( $add, $all )
- && !array_key_exists( $add, $snippets ) ) {
+ && array_key_exists( $add, $all )
+ && !array_key_exists( $add, $snippets ) ) {
$offsets[$add] = 0;
$tt = "\n" . $this->extract( $all[$add], 0, $targetchars - $len, $offsets[$add] );
$extended[$add] = $tt;
@@ -1142,7 +1150,7 @@ class SearchHighlighter {
}
}
- // $snippets = array_map('htmlspecialchars', $extended);
+ // $snippets = array_map( 'htmlspecialchars', $extended );
$snippets = $extended;
$last = - 1;
$extract = '';
@@ -1177,7 +1185,7 @@ class SearchHighlighter {
/**
* Split text into lines and add it to extracts array
*
- * @param $extracts Array: index -> $line
+ * @param array $extracts index -> $line
* @param $count Integer
* @param $text String
*/
@@ -1232,7 +1240,7 @@ class SearchHighlighter {
$posEnd = $end;
}
- if ( $end > $start ) {
+ if ( $end > $start ) {
return substr( $text, $start, $end - $start );
} else {
return '';
@@ -1272,12 +1280,12 @@ class SearchHighlighter {
/**
* Search extracts for a pattern, and return snippets
*
- * @param $pattern String: regexp for matching lines
- * @param $extracts Array: extracts to search
+ * @param string $pattern regexp for matching lines
+ * @param array $extracts extracts to search
* @param $linesleft Integer: number of extracts to make
* @param $contextchars Integer: length of snippet
- * @param $out Array: map for highlighted snippets
- * @param $offsets Array: map of starting points of snippets
+ * @param array $out map for highlighted snippets
+ * @param array $offsets map of starting points of snippets
* @protected
*/
function process( $pattern, $extracts, &$linesleft, &$contextchars, &$out, &$offsets ) {
@@ -1321,12 +1329,12 @@ class SearchHighlighter {
$fname = __METHOD__;
wfProfileIn( $fname );
- // $text = preg_replace("/'{2,5}/", "", $text);
- // $text = preg_replace("/\[[a-z]+:\/\/[^ ]+ ([^]]+)\]/", "\\2", $text);
- // $text = preg_replace("/\[\[([^]|]+)\]\]/", "\\1", $text);
- // $text = preg_replace("/\[\[([^]]+\|)?([^|]]+)\]\]/", "\\2", $text);
- // $text = preg_replace("/\\{\\|(.*?)\\|\\}/", "", $text);
- // $text = preg_replace("/\\[\\[[A-Za-z_-]+:([^|]+?)\\]\\]/", "", $text);
+ // $text = preg_replace( "/'{2,5}/", "", $text );
+ // $text = preg_replace( "/\[[a-z]+:\/\/[^ ]+ ([^]]+)\]/", "\\2", $text );
+ // $text = preg_replace( "/\[\[([^]|]+)\]\]/", "\\1", $text );
+ // $text = preg_replace( "/\[\[([^]]+\|)?([^|]]+)\]\]/", "\\2", $text );
+ // $text = preg_replace( "/\\{\\|(.*?)\\|\\}/", "", $text );
+ // $text = preg_replace( "/\\[\\[[A-Za-z_-]+:([^|]+?)\\]\\]/", "", $text );
$text = preg_replace( "/\\{\\{([^|]+?)\\}\\}/", "", $text );
$text = preg_replace( "/\\{\\{([^|]+\\|)(.*?)\\}\\}/", "\\2", $text );
$text = preg_replace( "/\\[\\[([^|]+?)\\]\\]/", "\\1", $text );
@@ -1408,8 +1416,7 @@ class SearchHighlighter {
$line = htmlspecialchars( $pre . $found . $post );
$pat2 = '/(' . $terms . ")/i";
- $line = preg_replace( $pat2,
- "<span class='searchmatch'>\\1</span>", $line );
+ $line = preg_replace( $pat2, "<span class='searchmatch'>\\1</span>", $line );
$extract .= "${line}\n";
}
generated by cgit v1.2.2 (git 2.25.0) at 2024-04-23 11:54:38 +0000