3 * Import one or more images from the local file system into the wiki without
4 * using the web-based interface.
6 * "Smart import" additions:
7 * - aim: preserve the essential metadata (user, description) when importing media
8 * files from an existing wiki.
10 * - interface with the source wiki, don't use bare files only (see --source-wiki-url).
11 * - fetch metadata from source wiki for each file to import.
12 * - commit the fetched metadata to the destination wiki while submitting.
14 * This program is free software; you can redistribute it and/or modify
15 * it under the terms of the GNU General Public License as published by
16 * the Free Software Foundation; either version 2 of the License, or
17 * (at your option) any later version.
19 * This program is distributed in the hope that it will be useful,
20 * but WITHOUT ANY WARRANTY; without even the implied warranty of
21 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 * GNU General Public License for more details.
24 * You should have received a copy of the GNU General Public License along
25 * with this program; if not, write to the Free Software Foundation, Inc.,
26 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
27 * http://www.gnu.org/copyleft/gpl.html
30 * @ingroup Maintenance
31 * @author Rob Church <robchur@gmail.com>
32 * @author Mij <mij@bitchx.it>
35 require_once __DIR__
. '/Maintenance.php';
37 use MediaWiki\MainConfigNames
;
38 use MediaWiki\Specials\SpecialUpload
;
39 use MediaWiki\StubObject\StubGlobalUser
;
40 use MediaWiki\Title\Title
;
41 use MediaWiki\User\User
;
43 class ImportImages
extends Maintenance
{
45 public function __construct() {
46 parent
::__construct();
48 $this->addDescription( 'Imports images and other media files into the wiki' );
49 $this->addArg( 'dir', 'Path to the directory containing images to be imported' );
51 $this->addOption( 'extensions',
52 'Comma-separated list of allowable extensions, defaults to $wgFileExtensions',
56 $this->addOption( 'overwrite',
57 'Overwrite existing images with the same name (default is to skip them)' );
58 $this->addOption( 'limit',
59 'Limit the number of images to process. Ignored or skipped images are not counted',
63 $this->addOption( 'from',
64 "Ignore all files until the one with the given name. Useful for resuming aborted "
65 . "imports. The name should be the file's canonical database form.",
69 $this->addOption( 'skip-dupes',
70 'Skip images that were already uploaded under a different name (check SHA1)' );
71 $this->addOption( 'search-recursively', 'Search recursively for files in subdirectories' );
72 $this->addOption( 'sleep',
73 'Sleep between files. Useful mostly for debugging',
77 $this->addOption( 'user',
78 "Set username of uploader, default 'Maintenance script'",
82 // This parameter can optionally have an argument. If none specified, getOption()
83 // returns 1 which is precisely what we need.
84 $this->addOption( 'check-userblock', 'Check if the user got blocked during import' );
85 $this->addOption( 'comment',
86 "Set file description, default 'Importing file'",
90 $this->addOption( 'comment-file',
91 'Set description to the content of this file',
95 $this->addOption( 'comment-ext',
96 'Causes the description for each file to be loaded from a file with the same name, but '
97 . 'the extension provided. If a global description is also given, it is appended.',
101 $this->addOption( 'summary',
102 'Upload summary, description will be used if not provided',
106 $this->addOption( 'license',
107 'Use an optional license template',
111 $this->addOption( 'timestamp',
112 'Override upload time/date, all MediaWiki timestamp formats are accepted',
116 $this->addOption( 'protect',
117 'Specify the protect value (autoconfirmed,sysop)',
121 $this->addOption( 'unprotect', 'Unprotects all uploaded images' );
122 $this->addOption( 'source-wiki-url',
123 'If specified, take User and Comment data for each imported file from this URL. '
124 . 'For example, --source-wiki-url="https://en.wikipedia.org/w/',
128 $this->addOption( 'dry', "Dry run, don't import anything" );
131 public function execute() {
132 $services = $this->getServiceContainer();
133 $permissionManager = $services->getPermissionManager();
145 $this->output( "Importing Files\n\n" );
147 $dir = $this->getArg( 0 );
150 if ( $this->hasOption( 'protect' ) && $this->hasOption( 'unprotect' ) ) {
151 $this->fatalError( "Cannot specify both protect and unprotect. Only 1 is allowed.\n" );
154 if ( $this->hasOption( 'protect' ) && trim( $this->getOption( 'protect' ) ) ) {
155 $this->fatalError( "You must specify a protection option.\n" );
158 # Prepare the list of allowed extensions
159 $extensions = $this->hasOption( 'extensions' )
160 ?
explode( ',', strtolower( $this->getOption( 'extensions' ) ) )
161 : $this->getConfig()->get( MainConfigNames
::FileExtensions
);
163 # Search the path provided for candidates for import
164 $files = $this->findFiles( $dir, $extensions, $this->hasOption( 'search-recursively' ) );
165 if ( !$files->valid() ) {
166 $this->output( "No suitable files could be found for import.\n" );
170 # Initialise the user for this operation
171 $user = $this->hasOption( 'user' )
172 ? User
::newFromName( $this->getOption( 'user' ) )
173 : User
::newSystemUser( User
::MAINTENANCE_SCRIPT_USER
, [ 'steal' => true ] );
174 if ( !$user instanceof User
) {
175 $user = User
::newSystemUser( User
::MAINTENANCE_SCRIPT_USER
, [ 'steal' => true ] );
177 '@phan-var User $user';
178 StubGlobalUser
::setUser( $user );
180 # Get block check. If a value is given, this specified how often the check is performed
181 $checkUserBlock = (int)$this->getOption( 'check-userblock' );
183 $from = $this->getOption( 'from' );
184 $sleep = (int)$this->getOption( 'sleep' );
185 $limit = (int)$this->getOption( 'limit' );
186 $timestamp = $this->getOption( 'timestamp', false );
188 # Get the upload comment. Provide a default one in case there's no comment given.
189 $commentFile = $this->getOption( 'comment-file' );
190 if ( $commentFile !== null ) {
191 $comment = file_get_contents( $commentFile );
192 if ( $comment === false ||
$comment === null ) {
193 $this->fatalError( "failed to read comment file: {$commentFile}\n" );
196 $comment = $this->getOption( 'comment', 'Importing file' );
198 $commentExt = $this->getOption( 'comment-ext' );
199 $summary = $this->getOption( 'summary', '' );
200 $license = $this->getOption( 'license', '' );
201 $sourceWikiUrl = $this->getOption( 'source-wiki-url' );
203 $tags = in_array( ChangeTags
::TAG_SERVER_SIDE_UPLOAD
, ChangeTags
::getSoftwareTags() )
204 ?
[ ChangeTags
::TAG_SERVER_SIDE_UPLOAD
]
207 # Batch "upload" operation
208 $restrictionStore = $services->getRestrictionStore();
209 foreach ( $files as $file ) {
211 if ( $sleep && ( $processed > 0 ) ) {
215 $base = UtfNormal\Validator
::cleanUp( wfBaseName( $file ) );
218 $title = Title
::makeTitleSafe( NS_FILE
, $base );
221 "{$base} could not be imported; a valid title cannot be produced\n"
227 if ( $from !== $title->getDBkey() ) {
228 $statistics['ignored']++
;
231 // Found the requested file, continue from here
235 if ( $checkUserBlock && ( ( $processed %
$checkUserBlock ) == 0 ) ) {
236 $user->clearInstanceCache( 'name' ); // reload from DB!
237 if ( $permissionManager->isBlockedFrom( $user, $title ) ) {
239 "{$user->getName()} is blocked from {$title->getPrefixedText()}! skipping.\n"
241 $statistics['skipped']++
;
247 $image = $services->getRepoGroup()->getLocalRepo()
249 if ( $image->exists() ) {
250 if ( $this->hasOption( 'overwrite' ) ) {
251 $this->output( "{$base} exists, overwriting..." );
252 $svar = 'overwritten';
254 $this->output( "{$base} exists, skipping\n" );
255 $statistics['skipped']++
;
259 if ( $this->hasOption( 'skip-dupes' ) ) {
260 $repo = $image->getRepo();
261 # XXX: we end up calculating this again when actually uploading. that sucks.
262 $sha1 = FSFile
::getSha1Base36FromPath( $file );
263 $dupes = $repo->findBySha1( $sha1 );
266 "{$base} already exists as {$dupes[0]->getName()}, skipping\n"
268 $statistics['skipped']++
;
273 $this->output( "Importing {$base}..." );
277 if ( $sourceWikiUrl ) {
278 /* find comment text directly from source wiki, through MW's API */
279 $real_comment = $this->getFileCommentFromSourceWiki( $sourceWikiUrl, $base );
280 $commentText = $real_comment !== false ?
$real_comment : $comment;
282 /* find user directly from source wiki, through MW's API */
283 $real_user = $this->getFileUserFromSourceWiki( $sourceWikiUrl, $base );
284 if ( $real_user !== false ) {
285 $realUser = User
::newFromName( $real_user );
286 if ( $realUser === false ) {
287 # user does not exist in target wiki
289 "failed: user '$real_user' does not exist in target wiki."
293 StubGlobalUser
::setUser( $realUser );
298 $commentText = false;
301 $f = $this->findAuxFile( $file, $commentExt );
303 $this->output( " No comment file with extension {$commentExt} found "
304 . "for {$file}, using default comment." );
306 $commentText = file_get_contents( $f );
307 if ( !$commentText ) {
309 " Failed to load comment file {$f}, using default comment."
315 if ( !$commentText ) {
316 $commentText = $comment;
321 if ( $this->hasOption( 'dry' ) ) {
323 " publishing {$file} by '{$user->getName()}', comment '$commentText'..."
326 $mwProps = new MWFileProps( $services->getMimeAnalyzer() );
327 $props = $mwProps->getPropsFromPath( $file, true );
329 $publishOptions = [];
330 $handler = MediaHandler
::getHandler( $props['mime'] );
332 $publishOptions['headers'] = $handler->getContentHeaders( $props['metadata'] );
334 $publishOptions['headers'] = [];
336 $archive = $image->publish( $file, $flags, $publishOptions );
337 if ( !$archive->isGood() ) {
338 $this->output( "failed. (" .
339 $archive->getMessage( false, false, 'en' )->text() .
341 $statistics['failed']++
;
346 $commentText = SpecialUpload
::getInitialPageText( $commentText, $license );
347 if ( !$this->hasOption( 'summary' ) ) {
348 $summary = $commentText;
351 if ( $this->hasOption( 'dry' ) ) {
352 $this->output( "done.\n" );
353 } elseif ( $image->recordUpload3(
354 // @phan-suppress-next-line PhanPossiblyUndeclaredVariable
359 // @phan-suppress-next-line PhanTypeMismatchArgumentNullable,PhanPossiblyUndeclaredVariable
364 $this->output( "done.\n" );
368 $protectLevel = $this->getOption( 'protect' );
369 $restrictionLevels = $this->getConfig()->get( MainConfigNames
::RestrictionLevels
);
371 if ( $protectLevel && in_array( $protectLevel, $restrictionLevels ) ) {
374 if ( $this->hasOption( 'unprotect' ) ) {
381 $this->output( "\nWaiting for replica DBs...\n" );
382 // Wait for replica DBs.
383 sleep( 2 ); # Why this sleep?
384 $this->waitForReplication();
386 $this->output( "\nSetting image restrictions ..." );
390 foreach ( $restrictionStore->listApplicableRestrictionTypes( $title ) as $type ) {
391 $restrictions[$type] = $protectLevel;
394 $page = $services->getWikiPageFactory()->newFromTitle( $title );
395 $status = $page->doUpdateRestrictions( $restrictions, [], $cascade, '', $user );
396 $this->output( ( $status->isOK() ?
'done' : 'failed' ) . "\n" );
399 $this->output( "failed. (at recordUpload stage)\n" );
403 $statistics[$svar]++
;
406 if ( $limit && $processed >= $limit ) {
411 # Print out some statistics
412 $this->output( "\n" );
413 foreach ( array_merge(
419 ) as $desc => $number ) {
421 $this->output( ucfirst( $desc ) . ": $number\n" );
427 * Search a directory for files with one of a set of extensions
429 * @param string $dir Path to directory to search
430 * @param array $exts Array of lowercase extensions to search for
431 * @param bool $recurse Search subdirectories recursively
432 * @return Generator<string> Generator that iterating filenames
434 private function findFiles( $dir, $exts, $recurse = false ) {
435 $dhl = is_dir( $dir ) ?
opendir( $dir ) : false;
440 while ( ( $file = readdir( $dhl ) ) !== false ) {
441 if ( is_file( $dir . '/' . $file ) ) {
442 $ext = pathinfo( $file, PATHINFO_EXTENSION
);
443 if ( in_array( strtolower( $ext ), $exts ) ) {
444 yield
$dir . '/' . $file;
446 } elseif ( $recurse && is_dir( $dir . '/' . $file ) && $file !== '..' && $file !== '.' ) {
447 yield from
$this->findFiles( $dir . '/' . $file, $exts, true );
453 * Find an auxiliary file with the given extension, matching
454 * the give base file path. $maxStrip determines how many extensions
455 * may be stripped from the original file name before appending the
456 * new extension. For example, with $maxStrip = 1 (the default),
457 * file files acme.foo.bar.txt and acme.foo.txt would be auxilliary
458 * files for acme.foo.bar and the extension ".txt". With $maxStrip = 2,
459 * acme.txt would also be acceptable.
461 * @param string $file Base path
462 * @param string $auxExtension The extension to be appended to the base path
463 * @param int $maxStrip The maximum number of extensions to strip from the base path (default: 1)
464 * @return string|false
466 private function findAuxFile( $file, $auxExtension, $maxStrip = 1 ) {
467 if ( !str_starts_with( $auxExtension, '.' ) ) {
468 $auxExtension = '.' . $auxExtension;
471 $d = dirname( $file );
472 $n = basename( $file );
474 while ( $maxStrip >= 0 ) {
475 $f = $d . '/' . $n . $auxExtension;
477 if ( file_exists( $f ) ) {
481 $idx = strrpos( $n, '.' );
486 $n = substr( $n, 0, $idx );
494 * @todo FIXME: Access the api in a better way and performing just one query
495 * (preferably batching files too).
497 * @param string $wiki_host
498 * @param string $file
500 * @return string|false
502 private function getFileCommentFromSourceWiki( $wiki_host, $file ) {
503 $url = $wiki_host . '/api.php?action=query&format=xml&titles=File:'
504 . rawurlencode( $file ) . '&prop=imageinfo&&iiprop=comment';
505 $body = $this->getServiceContainer()->getHttpRequestFactory()->get( $url, [], __METHOD__
);
506 if ( preg_match( '#<ii comment="([^"]*)" />#', $body, $matches ) == 0 ) {
510 return html_entity_decode( $matches[1] );
513 private function getFileUserFromSourceWiki( $wiki_host, $file ) {
514 $url = $wiki_host . '/api.php?action=query&format=xml&titles=File:'
515 . rawurlencode( $file ) . '&prop=imageinfo&&iiprop=user';
516 $body = $this->getServiceContainer()->getHttpRequestFactory()->get( $url, [], __METHOD__
);
517 if ( preg_match( '#<ii user="([^"]*)" />#', $body, $matches ) == 0 ) {
521 return html_entity_decode( $matches[1] );
526 $maintClass = ImportImages
::class;
527 require_once RUN_MAINTENANCE_IF_MAIN
;