MediaWiki  1.28.0
SearchExactMatchRescorer.php
Go to the documentation of this file.
1 <?php
42  public function rescore( $search, $namespaces, $srchres, $limit ) {
43  // Pick namespace (based on PrefixSearch::defaultSearchBackend)
44  $ns = in_array( NS_MAIN, $namespaces ) ? NS_MAIN : reset( $namespaces );
45  $t = Title::newFromText( $search, $ns );
46  if ( !$t || !$t->exists() ) {
47  // No exact match so just return the search results
48  return $srchres;
49  }
50  $string = $t->getPrefixedText();
51  $key = array_search( $string, $srchres );
52  if ( $key !== false ) {
53  // Exact match was in the results so just move it to the front
54  return $this->pullFront( $key, $srchres );
55  }
56  // Exact match not in the search results so check for some redirect handling cases
57  if ( $t->isRedirect() ) {
58  $target = $this->getRedirectTarget( $t );
59  $key = array_search( $target, $srchres );
60  if ( $key !== false ) {
61  // Exact match is a redirect to one of the returned matches so pull the
62  // returned match to the front. This might look odd but the alternative
63  // is to put the redirect in front and drop the match. The name of the
64  // found match is often more descriptive/better formed than the name of
65  // the redirect AND by definition they share a prefix. Hopefully this
66  // choice is less confusing and more helpful. But it might not be. But
67  // it is the choice we're going with for now.
68  return $this->pullFront( $key, $srchres );
69  }
70  $redirectTargetsToRedirect = $this->redirectTargetsToRedirect( $srchres );
71  if ( isset( $redirectTargetsToRedirect[$target] ) ) {
72  // The exact match and something in the results list are both redirects
73  // to the same thing! In this case we'll pull the returned match to the
74  // top following the same logic above. Again, it might not be a perfect
75  // choice but it'll do.
76  return $this->pullFront( $redirectTargetsToRedirect[$target], $srchres );
77  }
78  } else {
79  $redirectTargetsToRedirect = $this->redirectTargetsToRedirect( $srchres );
80  if ( isset( $redirectTargetsToRedirect[$string] ) ) {
81  // The exact match is the target of a redirect already in the results list so remove
82  // the redirect from the results list and push the exact match to the front
83  array_splice( $srchres, $redirectTargetsToRedirect[$string], 1 );
84  array_unshift( $srchres, $string );
85  return $srchres;
86  }
87  }
88 
89  // Exact match is totally unique from the other results so just add it to the front
90  array_unshift( $srchres, $string );
91  // And roll one off the end if the results are too long
92  if ( count( $srchres ) > $limit ) {
93  array_pop( $srchres );
94  }
95  return $srchres;
96  }
97 
103  private function redirectTargetsToRedirect( $titles ) {
104  $result = [];
105  foreach ( $titles as $key => $titleText ) {
106  $title = Title::newFromText( $titleText );
107  if ( !$title || !$title->isRedirect() ) {
108  continue;
109  }
110  $target = $this->getRedirectTarget( $title );
111  if ( !$target ) {
112  continue;
113  }
114  $result[$target] = $key;
115  }
116  return $result;
117  }
118 
125  private function pullFront( $key, $array ) {
126  $cut = array_splice( $array, $key, 1 );
127  array_unshift( $array, $cut[0] );
128  return $array;
129  }
130 
136  private function getRedirectTarget( $title ) {
138  if ( !$page->exists() ) {
139  return null;
140  }
141  $redir = $page->getRedirectTarget();
142  return $redir ? $redir->getPrefixedText() : null;
143  }
144 }
static factory(Title $title)
Create a WikiPage object of the appropriate class for the given title.
Definition: WikiPage.php:115
const NS_MAIN
Definition: Defines.php:56
pullFront($key, $array)
Returns an array where the element of $array at index $key becomes the first element.
static newFromText($text, $defaultNamespace=NS_MAIN)
Create a new Title from text, such as what one would find in a link.
Definition: Title.php:262
The index of the header message $result[1]=The index of the body text message $result[2 through n]=Parameters passed to body text message.Please note the header message cannot receive/use parameters. 'ImportHandleLogItemXMLTag':When parsing a XML tag in a log item.Return false to stop further processing of the tag $reader:XMLReader object $logInfo:Array of information 'ImportHandlePageXMLTag':When parsing a XML tag in a page.Return false to stop further processing of the tag $reader:XMLReader object &$pageInfo:Array of information 'ImportHandleRevisionXMLTag':When parsing a XML tag in a page revision.Return false to stop further processing of the tag $reader:XMLReader object $pageInfo:Array of page information $revisionInfo:Array of revision information 'ImportHandleToplevelXMLTag':When parsing a top level XML tag.Return false to stop further processing of the tag $reader:XMLReader object 'ImportHandleUploadXMLTag':When parsing a XML tag in a file upload.Return false to stop further processing of the tag $reader:XMLReader object $revisionInfo:Array of information 'ImportLogInterwikiLink':Hook to change the interwiki link used in log entries and edit summaries for transwiki imports.&$fullInterwikiPrefix:Interwiki prefix, may contain colons.&$pageTitle:String that contains page title. 'ImportSources':Called when reading from the $wgImportSources configuration variable.Can be used to lazy-load the import sources list.&$importSources:The value of $wgImportSources.Modify as necessary.See the comment in DefaultSettings.php for the detail of how to structure this array. 'InfoAction':When building information to display on the action=info page.$context:IContextSource object &$pageInfo:Array of information 'InitializeArticleMaybeRedirect':MediaWiki check to see if title is a redirect.&$title:Title object for the current page &$request:WebRequest &$ignoreRedirect:boolean to skip redirect check &$target:Title/string of redirect target &$article:Article object 'InternalParseBeforeLinks':during Parser's internalParse method before links but after nowiki/noinclude/includeonly/onlyinclude and other processings.&$parser:Parser object &$text:string containing partially parsed text &$stripState:Parser's internal StripState object 'InternalParseBeforeSanitize':during Parser's internalParse method just before the parser removes unwanted/dangerous HTML tags and after nowiki/noinclude/includeonly/onlyinclude and other processings.Ideal for syntax-extensions after template/parser function execution which respect nowiki and HTML-comments.&$parser:Parser object &$text:string containing partially parsed text &$stripState:Parser's internal StripState object 'InterwikiLoadPrefix':When resolving if a given prefix is an interwiki or not.Return true without providing an interwiki to continue interwiki search.$prefix:interwiki prefix we are looking for.&$iwData:output array describing the interwiki with keys iw_url, iw_local, iw_trans and optionally iw_api and iw_wikiid. 'InvalidateEmailComplete':Called after a user's email has been invalidated successfully.$user:user(object) whose email is being invalidated 'IRCLineURL':When constructing the URL to use in an IRC notification.Callee may modify $url and $query, URL will be constructed as $url.$query &$url:URL to index.php &$query:Query string $rc:RecentChange object that triggered url generation 'IsFileCacheable':Override the result of Article::isFileCacheable()(if true) &$article:article(object) being checked 'IsTrustedProxy':Override the result of IP::isTrustedProxy() &$ip:IP being check &$result:Change this value to override the result of IP::isTrustedProxy() 'IsUploadAllowedFromUrl':Override the result of UploadFromUrl::isAllowedUrl() $url:URL used to upload from &$allowed:Boolean indicating if uploading is allowed for given URL 'isValidEmailAddr':Override the result of Sanitizer::validateEmail(), for instance to return false if the domain name doesn't match your organization.$addr:The e-mail address entered by the user &$result:Set this and return false to override the internal checks 'isValidPassword':Override the result of User::isValidPassword() $password:The password entered by the user &$result:Set this and return false to override the internal checks $user:User the password is being validated for 'Language::getMessagesFileName':$code:The language code or the language we're looking for a messages file for &$file:The messages file path, you can override this to change the location. 'LanguageGetMagic':DEPRECATED!Use $magicWords in a file listed in $wgExtensionMessagesFiles instead.Use this to define synonyms of magic words depending of the language &$magicExtensions:associative array of magic words synonyms $lang:language code(string) 'LanguageGetNamespaces':Provide custom ordering for namespaces or remove namespaces.Do not use this hook to add namespaces.Use CanonicalNamespaces for that.&$namespaces:Array of namespaces indexed by their numbers 'LanguageGetSpecialPageAliases':DEPRECATED!Use $specialPageAliases in a file listed in $wgExtensionMessagesFiles instead.Use to define aliases of special pages names depending of the language &$specialPageAliases:associative array of magic words synonyms $lang:language code(string) 'LanguageGetTranslatedLanguageNames':Provide translated language names.&$names:array of language code=> language name $code:language of the preferred translations 'LanguageLinks':Manipulate a page's language links.This is called in various places to allow extensions to define the effective language links for a page.$title:The page's Title.&$links:Associative array mapping language codes to prefixed links of the form"language:title".&$linkFlags:Associative array mapping prefixed links to arrays of flags.Currently unused, but planned to provide support for marking individual language links in the UI, e.g.for featured articles. 'LanguageSelector':Hook to change the language selector available on a page.$out:The output page.$cssClassName:CSS class name of the language selector. 'LinkBegin':DEPRECATED!Use HtmlPageLinkRendererBegin instead.Used when generating internal and interwiki links in Linker::link(), before processing starts.Return false to skip default processing and return $ret.See documentation for Linker::link() for details on the expected meanings of parameters.$skin:the Skin object $target:the Title that the link is pointing to &$html:the contents that the< a > tag should have(raw HTML) $result
Definition: hooks.txt:1934
An utility class to rescore search results by looking for an exact match in the db and add the page f...
namespace and then decline to actually register it file or subcat img or subcat $title
Definition: hooks.txt:953
namespace and then decline to actually register it & $namespaces
Definition: hooks.txt:953
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such as
Definition: distributors.txt:9
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring php
Definition: injection.txt:35
linkcache txt The LinkCache class maintains a list of article titles and the information about whether or not the article exists in the database This is used to mark up links when displaying a page If the same link appears more than once on any page then it only has to be looked up once In most cases link lookups are done in batches with the LinkBatch class or the equivalent in so the link cache is mostly useful for short snippets of parsed and for links in the navigation areas of the skin The link cache was formerly used to track links used in a document for the purposes of updating the link tables This application is now deprecated To create a you can use the following $titles
Definition: linkcache.txt:17
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context as context the output can only depend on parameters provided to this hook not on global state indicating whether full HTML should be generated If generation of HTML may be but other information should still be present in the ParserOutput object to manipulate or replace but no entry for that model exists in $wgContentHandlers if desired whether it is OK to use $contentModel on $title Handler functions that modify $ok should generally return false to prevent further hooks from further modifying $ok inclusive $limit
Definition: hooks.txt:1046
rescore($search, $namespaces, $srchres, $limit)
Default search backend does proper prefix searching, but custom backends may sort based on other algo...
do that in ParserLimitReportFormat instead use this to modify the parameters of the image and a DIV can begin in one section and end in another Make sure your code can handle that case gracefully See the EditSectionClearerLink extension for an example zero but section is usually empty its values are the globals values before the output is cached $page
Definition: hooks.txt:2491
getRedirectTarget($title)
Get a redirect's destination from a title.