MediaWiki  master
UploadFromUrl.php
Go to the documentation of this file.
1 <?php
31 class UploadFromUrl extends UploadBase {
32  protected $mUrl;
33 
34  protected $mTempPath, $mTmpHandle;
35 
36  protected static $allowedUrls = [];
37 
47  public static function isAllowed( $user ) {
48  if ( !$user->isAllowed( 'upload_by_url' ) ) {
49  return 'upload_by_url';
50  }
51 
52  return parent::isAllowed( $user );
53  }
54 
59  public static function isEnabled() {
60  global $wgAllowCopyUploads;
61 
62  return $wgAllowCopyUploads && parent::isEnabled();
63  }
64 
73  public static function isAllowedHost( $url ) {
74  global $wgCopyUploadsDomains;
75  if ( !count( $wgCopyUploadsDomains ) ) {
76  return true;
77  }
78  $parsedUrl = wfParseUrl( $url );
79  if ( !$parsedUrl ) {
80  return false;
81  }
82  $valid = false;
83  foreach ( $wgCopyUploadsDomains as $domain ) {
84  // See if the domain for the upload matches this whitelisted domain
85  $whitelistedDomainPieces = explode( '.', $domain );
86  $uploadDomainPieces = explode( '.', $parsedUrl['host'] );
87  if ( count( $whitelistedDomainPieces ) === count( $uploadDomainPieces ) ) {
88  $valid = true;
89  // See if all the pieces match or not (excluding wildcards)
90  foreach ( $whitelistedDomainPieces as $index => $piece ) {
91  if ( $piece !== '*' && $piece !== $uploadDomainPieces[$index] ) {
92  $valid = false;
93  }
94  }
95  if ( $valid ) {
96  // We found a match, so quit comparing against the list
97  break;
98  }
99  }
100  /* Non-wildcard test
101  if ( $parsedUrl['host'] === $domain ) {
102  $valid = true;
103  break;
104  }
105  */
106  }
107 
108  return $valid;
109  }
110 
117  public static function isAllowedUrl( $url ) {
118  if ( !isset( self::$allowedUrls[$url] ) ) {
119  $allowed = true;
120  Hooks::run( 'IsUploadAllowedFromUrl', [ $url, &$allowed ] );
121  self::$allowedUrls[$url] = $allowed;
122  }
123 
124  return self::$allowedUrls[$url];
125  }
126 
134  public function initialize( $name, $url ) {
135  $this->mUrl = $url;
136 
137  $tempPath = $this->makeTemporaryFile();
138  # File size and removeTempFile will be filled in later
139  $this->initializePathInfo( $name, $tempPath, 0, false );
140  }
141 
146  public function initializeFromRequest( &$request ) {
147  $desiredDestName = $request->getText( 'wpDestFile' );
148  if ( !$desiredDestName ) {
149  $desiredDestName = $request->getText( 'wpUploadFileURL' );
150  }
151  $this->initialize(
152  $desiredDestName,
153  trim( $request->getVal( 'wpUploadFileURL' ) )
154  );
155  }
156 
161  public static function isValidRequest( $request ) {
162  global $wgUser;
163 
164  $url = $request->getVal( 'wpUploadFileURL' );
165 
166  return !empty( $url )
167  && $wgUser->isAllowed( 'upload_by_url' );
168  }
169 
173  public function getSourceType() {
174  return 'url';
175  }
176 
184  public function fetchFile( $httpOptions = [] ) {
185  if ( !Http::isValidURI( $this->mUrl ) ) {
186  return Status::newFatal( 'http-invalid-url', $this->mUrl );
187  }
188 
189  if ( !self::isAllowedHost( $this->mUrl ) ) {
190  return Status::newFatal( 'upload-copy-upload-invalid-domain' );
191  }
192  if ( !self::isAllowedUrl( $this->mUrl ) ) {
193  return Status::newFatal( 'upload-copy-upload-invalid-url' );
194  }
195  return $this->reallyFetchFile( $httpOptions );
196  }
197 
203  protected function makeTemporaryFile() {
204  $tmpFile = TempFSFile::factory( 'URL', 'urlupload_', wfTempDir() );
205  $tmpFile->bind( $this );
206 
207  return $tmpFile->getPath();
208  }
209 
217  public function saveTempFileChunk( $req, $buffer ) {
218  wfDebugLog( 'fileupload', 'Received chunk of ' . strlen( $buffer ) . ' bytes' );
219  $nbytes = fwrite( $this->mTmpHandle, $buffer );
220 
221  if ( $nbytes == strlen( $buffer ) ) {
222  $this->mFileSize += $nbytes;
223  } else {
224  // Well... that's not good!
225  wfDebugLog(
226  'fileupload',
227  'Short write ' . $nbytes . '/' . strlen( $buffer ) .
228  ' bytes, aborting with ' . $this->mFileSize . ' uploaded so far'
229  );
230  fclose( $this->mTmpHandle );
231  $this->mTmpHandle = false;
232  }
233 
234  return $nbytes;
235  }
236 
244  protected function reallyFetchFile( $httpOptions = [] ) {
246  if ( $this->mTempPath === false ) {
247  return Status::newFatal( 'tmp-create-error' );
248  }
249 
250  // Note the temporary file should already be created by makeTemporaryFile()
251  $this->mTmpHandle = fopen( $this->mTempPath, 'wb' );
252  if ( !$this->mTmpHandle ) {
253  return Status::newFatal( 'tmp-create-error' );
254  }
255  wfDebugLog( 'fileupload', 'Temporary file created "' . $this->mTempPath . '"' );
256 
257  $this->mRemoveTempFile = true;
258  $this->mFileSize = 0;
259 
260  $options = $httpOptions + [ 'followRedirects' => true ];
261 
262  if ( $wgCopyUploadProxy !== false ) {
263  $options['proxy'] = $wgCopyUploadProxy;
264  }
265 
266  if ( $wgCopyUploadTimeout && !isset( $options['timeout'] ) ) {
267  $options['timeout'] = $wgCopyUploadTimeout;
268  }
269  wfDebugLog(
270  'fileupload',
271  'Starting download from "' . $this->mUrl . '" ' .
272  '<' . implode( ',', array_keys( array_filter( $options ) ) ) . '>'
273  );
274  $req = MWHttpRequest::factory( $this->mUrl, $options, __METHOD__ );
275  $req->setCallback( [ $this, 'saveTempFileChunk' ] );
276  $status = $req->execute();
277 
278  if ( $this->mTmpHandle ) {
279  // File got written ok...
280  fclose( $this->mTmpHandle );
281  $this->mTmpHandle = null;
282  } else {
283  // We encountered a write error during the download...
284  return Status::newFatal( 'tmp-write-error' );
285  }
286 
287  wfDebugLog( 'fileupload', $status );
288  if ( $status->isOK() ) {
289  wfDebugLog( 'fileupload', 'Download by URL completed successfully.' );
290  } else {
291  wfDebugLog(
292  'fileupload',
293  'Download by URL completed with HTTP status ' . $req->getStatus()
294  );
295  }
296 
297  return $status;
298  }
299 }
makeTemporaryFile()
Create a new temporary file in the URL subdirectory of wfTempDir().
static isValidRequest( $request)
static factory( $url, array $options=null, $caller=__METHOD__)
Generate a new request object Deprecated:
$wgCopyUploadsDomains
A list of domains copy uploads can come from.
static $allowedUrls
initializeFromRequest(&$request)
Entry point for SpecialUpload.
wfParseUrl( $url)
parse_url() work-alike, but non-broken.
initializePathInfo( $name, $tempPath, $fileSize, $removeTempFile=false)
Initialize the path information.
Definition: UploadBase.php:228
static isAllowedHost( $url)
Checks whether the URL is for an allowed host The domains in the whitelist can include wildcard chara...
$wgAllowCopyUploads
Allow for upload to be copied from an URL.
reallyFetchFile( $httpOptions=[])
Download the file, save it to the temporary file and update the file size and set $mRemoveTempFile to...
Implements uploading from a HTTP resource.
initialize( $name, $url)
Entry point for API upload.
null means default in associative array with keys and values unescaped Should be merged with default with a value of false meaning to suppress the attribute in associative array with keys and values unescaped noclasses just before the function returns a value If you return true
Definition: hooks.txt:1982
static newFatal( $message)
Factory function for fatal errors.
Definition: StatusValue.php:68
Status::newGood()` to allow deletion, and then `return false` from the hook function. Ensure you consume the 'ChangeTagAfterDelete' hook to carry out custom deletion actions. $tag:name of the tag $user:user initiating the action & $status:Status object. See above. 'ChangeTagsListActive':Allows you to nominate which of the tags your extension uses are in active use. & $tags:list of all active tags. Append to this array. 'ChangeTagsAfterUpdateTags':Called after tags have been updated with the ChangeTags::updateTags function. Params:$addedTags:tags effectively added in the update $removedTags:tags effectively removed in the update $prevTags:tags that were present prior to the update $rc_id:recentchanges table id $rev_id:revision table id $log_id:logging table id $params:tag params $rc:RecentChange being tagged when the tagging accompanies the action, or null $user:User who performed the tagging when the tagging is subsequent to the action, or null 'ChangeTagsAllowedAdd':Called when checking if a user can add tags to a change. & $allowedTags:List of all the tags the user is allowed to add. Any tags the user wants to add( $addTags) that are not in this array will cause it to fail. You may add or remove tags to this array as required. $addTags:List of tags user intends to add. $user:User who is adding the tags. 'ChangeUserGroups':Called before user groups are changed. $performer:The User who will perform the change $user:The User whose groups will be changed & $add:The groups that will be added & $remove:The groups that will be removed 'Collation::factory':Called if $wgCategoryCollation is an unknown collation. $collationName:Name of the collation in question & $collationObject:Null. Replace with a subclass of the Collation class that implements the collation given in $collationName. 'ConfirmEmailComplete':Called after a user 's email has been confirmed successfully. $user:user(object) whose email is being confirmed 'ContentAlterParserOutput':Modify parser output for a given content object. Called by Content::getParserOutput after parsing has finished. Can be used for changes that depend on the result of the parsing but have to be done before LinksUpdate is called(such as adding tracking categories based on the rendered HTML). $content:The Content to render $title:Title of the page, as context $parserOutput:ParserOutput to manipulate 'ContentGetParserOutput':Customize parser output for a given content object, called by AbstractContent::getParserOutput. May be used to override the normal model-specific rendering of page content. $content:The Content to render $title:Title of the page, as context $revId:The revision ID, as context $options:ParserOptions for rendering. To avoid confusing the parser cache, the output can only depend on parameters provided to this hook function, not on global state. $generateHtml:boolean, indicating whether full HTML should be generated. If false, generation of HTML may be skipped, but other information should still be present in the ParserOutput object. & $output:ParserOutput, to manipulate or replace 'ContentHandlerDefaultModelFor':Called when the default content model is determined for a given title. May be used to assign a different model for that title. $title:the Title in question & $model:the model name. Use with CONTENT_MODEL_XXX constants. 'ContentHandlerForModelID':Called when a ContentHandler is requested for a given content model name, but no entry for that model exists in $wgContentHandlers. Note:if your extension implements additional models via this hook, please use GetContentModels hook to make them known to core. $modeName:the requested content model name & $handler:set this to a ContentHandler object, if desired. 'ContentModelCanBeUsedOn':Called to determine whether that content model can be used on a given page. This is especially useful to prevent some content models to be used in some special location. $contentModel:ID of the content model in question $title:the Title in question. & $ok:Output parameter, 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. 'ContribsPager::getQueryInfo':Before the contributions query is about to run & $pager:Pager object for contributions & $queryInfo:The query for the contribs Pager 'ContribsPager::reallyDoQuery':Called before really executing the query for My Contributions & $data:an array of results of all contribs queries $pager:The ContribsPager object hooked into $offset:Index offset, inclusive $limit:Exact query limit $descending:Query direction, false for ascending, true for descending 'ContributionsLineEnding':Called before a contributions HTML line is finished $page:SpecialPage object for contributions & $ret:the HTML line $row:the DB row for this line & $classes:the classes to add to the surrounding< li > & $attribs:associative array of other HTML attributes for the< li > element. Currently only data attributes reserved to MediaWiki are allowed(see Sanitizer::isReservedDataAttribute). 'ContributionsToolLinks':Change tool links above Special:Contributions $id:User identifier $title:User page title & $tools:Array of tool links $specialPage:SpecialPage instance for context and services. Can be either SpecialContributions or DeletedContributionsPage. Extensions should type hint against a generic SpecialPage though. 'ConvertContent':Called by AbstractContent::convert when a conversion to another content model is requested. Handler functions that modify $result should generally return false to disable further attempts at conversion. $content:The Content object to be converted. $toModel:The ID of the content model to convert to. $lossy:boolean indicating whether lossy conversion is allowed. & $result:Output parameter, in case the handler function wants to provide a converted Content object. Note that $result->getContentModel() must return $toModel. 'ContentSecurityPolicyDefaultSource':Modify the allowed CSP load sources. This affects all directives except for the script directive. If you want to add a script source, see ContentSecurityPolicyScriptSource hook. & $defaultSrc:Array of Content-Security-Policy allowed sources $policyConfig:Current configuration for the Content-Security-Policy header $mode:ContentSecurityPolicy::REPORT_ONLY_MODE or ContentSecurityPolicy::FULL_MODE depending on type of header 'ContentSecurityPolicyDirectives':Modify the content security policy directives. Use this only if ContentSecurityPolicyDefaultSource and ContentSecurityPolicyScriptSource do not meet your needs. & $directives:Array of CSP directives $policyConfig:Current configuration for the CSP header $mode:ContentSecurityPolicy::REPORT_ONLY_MODE or ContentSecurityPolicy::FULL_MODE depending on type of header 'ContentSecurityPolicyScriptSource':Modify the allowed CSP script sources. Note that you also have to use ContentSecurityPolicyDefaultSource if you want non-script sources to be loaded from whatever you add. & $scriptSrc:Array of CSP directives $policyConfig:Current configuration for the CSP header $mode:ContentSecurityPolicy::REPORT_ONLY_MODE or ContentSecurityPolicy::FULL_MODE depending on type of header 'CustomEditor':When invoking the page editor Return true to allow the normal editor to be used, or false if implementing a custom editor, e.g. for a special namespace, etc. $article:Article being edited $user:User performing the edit 'DeletedContribsPager::reallyDoQuery':Called before really executing the query for Special:DeletedContributions Similar to ContribsPager::reallyDoQuery & $data:an array of results of all contribs queries $pager:The DeletedContribsPager object hooked into $offset:Index offset, inclusive $limit:Exact query limit $descending:Query direction, false for ascending, true for descending 'DeletedContributionsLineEnding':Called before a DeletedContributions HTML line is finished. Similar to ContributionsLineEnding $page:SpecialPage object for DeletedContributions & $ret:the HTML line $row:the DB row for this line & $classes:the classes to add to the surrounding< li > & $attribs:associative array of other HTML attributes for the< li > element. Currently only data attributes reserved to MediaWiki are allowed(see Sanitizer::isReservedDataAttribute). 'DeleteUnknownPreferences':Called by the cleanupPreferences.php maintenance script to build a WHERE clause with which to delete preferences that are not known about. This hook is used by extensions that have dynamically-named preferences that should not be deleted in the usual cleanup process. For example, the Gadgets extension creates preferences prefixed with 'gadget-', and so anything with that prefix is excluded from the deletion. &where:An array that will be passed as the $cond parameter to IDatabase::select() to determine what will be deleted from the user_properties table. $db:The IDatabase object, useful for accessing $db->buildLike() etc. 'DifferenceEngineAfterLoadNewText':called in DifferenceEngine::loadNewText() after the new revision 's content has been loaded into the class member variable $differenceEngine->mNewContent but before returning true from this function. $differenceEngine:DifferenceEngine object 'DifferenceEngineLoadTextAfterNewContentIsLoaded':called in DifferenceEngine::loadText() after the new revision 's content has been loaded into the class member variable $differenceEngine->mNewContent but before checking if the variable 's value is null. This hook can be used to inject content into said class member variable. $differenceEngine:DifferenceEngine object 'DifferenceEngineMarkPatrolledLink':Allows extensions to change the "mark as patrolled" link which is shown both on the diff header as well as on the bottom of a page, usually wrapped in a span element which has class="patrollink". $differenceEngine:DifferenceEngine object & $markAsPatrolledLink:The "mark as patrolled" link HTML(string) $rcid:Recent change ID(rc_id) for this change(int) 'DifferenceEngineMarkPatrolledRCID':Allows extensions to possibly change the rcid parameter. For example the rcid might be set to zero due to the user being the same as the performer of the change but an extension might still want to show it under certain conditions. & $rcid:rc_id(int) of the change or 0 $differenceEngine:DifferenceEngine object $change:RecentChange object $user:User object representing the current user 'DifferenceEngineNewHeader':Allows extensions to change the $newHeader variable, which contains information about the new revision, such as the revision 's author, whether the revision was marked as a minor edit or not, etc. $differenceEngine:DifferenceEngine object & $newHeader:The string containing the various #mw-diff-otitle[1-5] divs, which include things like revision author info, revision comment, RevisionDelete link and more $formattedRevisionTools:Array containing revision tools, some of which may have been injected with the DiffRevisionTools hook $nextlink:String containing the link to the next revision(if any) $status
Definition: hooks.txt:1263
static isEnabled()
Checks if the upload from URL feature is enabled.
wfTempDir()
Tries to get the system directory for temporary files.
static isAllowedUrl( $url)
Checks whether the URL is not allowed.
static factory( $prefix, $extension='', $tmpDirectory=null)
Make a new temporary file on the file system.
Definition: TempFSFile.php:55
UploadBase and subclasses are the backend of MediaWiki&#39;s file uploads.
Definition: UploadBase.php:40
int bool $wgCopyUploadTimeout
Different timeout for upload by url This could be useful since when fetching large files...
null means default in associative array with keys and values unescaped Should be merged with default with a value of false meaning to suppress the attribute in associative array with keys and values unescaped & $options
Definition: hooks.txt:1982
saveTempFileChunk( $req, $buffer)
Callback: save a chunk of the result of a HTTP request to the temporary file.
this hook is for auditing only or null if authentication failed before getting that far or null if we can t even determine that When $user is not null
Definition: hooks.txt:780
$buffer
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
static isValidURI( $uri)
Check that the given URI is a valid one.
Definition: Http.php:149
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
this hook is for auditing only $req
Definition: hooks.txt:979
wfDebugLog( $logGroup, $text, $dest='all', array $context=[])
Send a line to a supplementary debug log file, if configured, or main debug log if not...
static isAllowed( $user)
Checks if the user is allowed to use the upload-by-URL feature.
Allows to change the fields on the form that will be generated $name
Definition: hooks.txt:271
$wgCopyUploadProxy
Proxy to use for copy upload requests.
do that in ParserLimitReportFormat instead use this to modify the parameters of the image all existing parser cache entries will be invalid To avoid you ll need to handle that somehow(e.g. with the RejectParserCacheValue hook) because MediaWiki won 't do it for you. & $defaults also a ContextSource after deleting those rows but within the same transaction you ll probably need to make sure the header is varied on $request
Definition: hooks.txt:2633
return true to allow those checks to and false if checking is done & $user
Definition: hooks.txt:1473
static run( $event, array $args=[], $deprecatedVersion=null)
Call hook functions defined in Hooks::register and $wgHooks.
Definition: Hooks.php:200
fetchFile( $httpOptions=[])
Download the file.