3 * Copyright (C) 2005 Brion Vibber <brion@pobox.com>
4 * http://www.mediawiki.org/
6 * This program is free software; you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License as published by
8 * the Free Software Foundation; either version 2 of the License, or
9 * (at your option) any later version.
11 * This program is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License along
17 * with this program; if not, write to the Free Software Foundation, Inc.,
18 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
19 * http://www.gnu.org/copyleft/gpl.html
22 * @ingroup Maintenance
25 $optionsWithArgs = array( 'report' );
27 require_once( dirname( __FILE__
) . '/commandLine.inc' );
30 * @ingroup Maintenance
31 * @todo port to Maintenance class
34 var $reportingInterval = 100;
35 var $reporting = true;
42 function __construct() {
43 $this->stderr
= fopen( "php://stderr", "wt" );
46 function reportPage( $page ) {
50 function handleRevision( $rev ) {
51 $title = $rev->getTitle();
53 $this->progress( "Got bogus revision with null title!" );
60 if ( !$this->dryRun
) {
61 call_user_func( $this->importCallback
, $rev );
65 function handleUpload( $revision ) {
66 if ( $this->uploads
) {
69 $this->progress( "upload: " . $revision->getFilename() );
71 if ( !$this->dryRun
) {
73 // call_user_func( $this->uploadCallback, $revision );
74 $dbw = wfGetDB( DB_MASTER
);
75 return $dbw->deadlockLoop( array( $revision, 'importUpload' ) );
80 function handleLogItem( $rev ) {
84 if ( !$this->dryRun
) {
85 call_user_func( $this->logItemCallback
, $rev );
89 function report( $final = false ) {
90 if ( $final xor ( $this->pageCount %
$this->reportingInterval
== 0 ) ) {
95 function showReport() {
96 if ( $this->reporting
) {
97 $delta = wfTime() - $this->startTime
;
99 $rate = sprintf( "%.2f", $this->pageCount
/ $delta );
100 $revrate = sprintf( "%.2f", $this->revCount
/ $delta );
105 # Logs dumps don't have page tallies
106 if ( $this->pageCount
)
107 $this->progress( "$this->pageCount ($rate pages/sec $revrate revs/sec)" );
109 $this->progress( "$this->revCount ($revrate revs/sec)" );
111 wfWaitForSlaves( 5 );
114 function progress( $string ) {
115 fwrite( $this->stderr
, $string . "\n" );
118 function importFromFile( $filename ) {
119 if ( preg_match( '/\.gz$/', $filename ) ) {
120 $filename = 'compress.zlib://' . $filename;
122 elseif ( preg_match( '/\.bz2$/', $filename ) ) {
123 $filename = 'compress.bzip2://' . $filename;
125 elseif ( preg_match( '/\.7z$/', $filename ) ) {
126 $filename = 'mediawiki.compress.7z://' . $filename;
129 $file = fopen( $filename, 'rt' );
130 return $this->importFromHandle( $file );
133 function importFromStdin() {
134 $file = fopen( 'php://stdin', 'rt' );
135 if( posix_isatty( $file ) ) {
139 return $this->importFromHandle( $file );
142 function importFromHandle( $handle ) {
143 $this->startTime
= wfTime();
145 $source = new ImportStreamSource( $handle );
146 $importer = new WikiImporter( $source );
148 $importer->setDebug( $this->debug
);
149 $importer->setPageCallback( array( &$this, 'reportPage' ) );
150 $this->importCallback
= $importer->setRevisionCallback(
151 array( &$this, 'handleRevision' ) );
152 $this->uploadCallback
= $importer->setUploadCallback(
153 array( &$this, 'handleUpload' ) );
154 $this->logItemCallback
= $importer->setLogItemCallback(
155 array( &$this, 'handleLogItem' ) );
157 if ( $this->dryRun
) {
158 $importer->setPageOutCallback( null );
161 return $importer->doImport();
164 function showHelp() {
165 $gz = in_array('compress.zlib', stream_get_wrappers()) ?
'ok' : '(disabled; requires PHP zlib module)';
166 $bz2 = in_array('compress.bzip2', stream_get_wrappers()) ?
'ok' : '(disabled; requires PHP bzip2 module)';
167 echo "This script reads pages from an XML file as produced from Special:Export\n";
168 echo "or dumpBackup.php, and saves them into the current wiki.\n";
170 echo "Note that for very large data sets, importDump.php may be slow; there are\n";
171 echo "alternate methods which can be much faster for full site restoration:\n";
172 echo "http://www.mediawiki.org/wiki/Manual:Importing_XML_dumps\n";
174 echo "Usage: php importDump.php [<options>] [<file>]\n";
175 echo "If no file is listed, input may be piped from stdin.\n";
178 echo " --quiet Don't dump status reports to stderr.\n";
179 echo " --report=n Report position and speed after every n pages processed.\n";
180 echo " --dry-run Parse dump without actually importing pages.\n";
181 echo " --debug Output extra verbose debug information\n";
182 echo " --uploads Process file upload data if included (experimental)\n";
184 echo "Compressed XML files may be read directly:\n";
187 echo " .7z (if 7za executable is in PATH)\n";
192 if ( wfReadOnly() ) {
193 wfDie( "Wiki is in read-only mode; you'll need to disable it for import to work.\n" );
196 $reader = new BackupReader();
197 if ( isset( $options['quiet'] ) ) {
198 $reader->reporting
= false;
200 if ( isset( $options['report'] ) ) {
201 $reader->reportingInterval
= intval( $options['report'] );
203 if ( isset( $options['dry-run'] ) ) {
204 $reader->dryRun
= true;
206 if ( isset( $options['debug'] ) ) {
207 $reader->debug
= true;
209 if ( isset( $options['uploads'] ) ) {
210 $reader->uploads
= true; // experimental!
213 if ( isset( $options['help'] ) ) {
216 } elseif ( isset( $args[0] ) ) {
217 $result = $reader->importFromFile( $args[0] );
219 $result = $reader->importFromStdin();
223 echo "You might want to run rebuildrecentchanges.php to regenerate\n";
224 echo "the recentchanges page.\n";