maintenance/dumpTextPass.php

   1 <?php
   2 /**
   3  * Copyright (C) 2005 Brion Vibber <brion@pobox.com>
   4  * http://www.mediawiki.org/
   5  *
   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.
  10  *
  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.
  15  *
  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  * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  19  * http://www.gnu.org/copyleft/gpl.html
  20  *
  21  * @package MediaWiki
  22  * @subpackage SpecialPage
  23  */
  24
  25 $originalDir = getcwd();
  26
  27 $optionsWithArgs = array( 'server', 'pagelist', 'start', 'end' );
  28
  29 require_once( 'commandLine.inc' );
  30 require_once( 'SpecialExport.php' );
  31 require_once( 'maintenance/backup.inc' );
  32
  33 class TextPassDumper extends BackupDumper {
  34         var $prefetch = null;
  35
  36         function dump() {
  37                 # This shouldn't happen if on console... ;)
  38                 header( 'Content-type: text/html; charset=UTF-8' );
  39
  40                 # Notice messages will foul up your XML output even if they're
  41                 # relatively harmless.
  42 //              ini_set( 'display_errors', false );
  43
  44                 $this->startTime = wfTime();
  45
  46                 $this->db =& $this->backupDb();
  47                 $this->maxCount = $this->db->selectField( 'page', 'MAX(page_id)', '', 'BackupDumper::dump' );
  48                 $this->startTime = wfTime();
  49
  50                 $this->egress = new ExportProgressFilter( $this->sink, $this );
  51
  52                 $input = fopen( "php://stdin", "rt" );
  53                 $result = $this->readDump( $input );
  54
  55                 if( WikiError::isError( $result ) ) {
  56                         $this->progress( $result->getMessage() );
  57                 }
  58
  59                 $this->report( true );
  60         }
  61
  62         function processOption( $opt, $val, $param ) {
  63                 if( $opt == 'prefetch' ) {
  64                         require_once 'maintenance/backupPrefetch.inc';
  65                         switch( $val ) {
  66                         case "file":
  67                                 $filename = $param;
  68                                 break;
  69                         case "gzip":
  70                                 $filename = "compress.gzip://$param";
  71                                 break;
  72                         case "bzip2":
  73                                 $filename = "compress.bzip2://$param";
  74                                 break;
  75                         default:
  76                                 $filename = $val;
  77                         }
  78                         $this->prefetch = new BaseDump( $filename );
  79                 }
  80         }
  81
  82         function readDump( $input ) {
  83                 $this->buffer = "";
  84                 $this->openElement = false;
  85                 $this->atStart = true;
  86                 $this->state = "";
  87                 $this->lastName = "";
  88                 $this->thisPage = 0;
  89                 $this->thisRev = 0;
  90
  91                 $parser = xml_parser_create( "UTF-8" );
  92                 xml_parser_set_option( $parser, XML_OPTION_CASE_FOLDING, false );
  93
  94                 xml_set_element_handler( $parser, array( &$this, 'startElement' ), array( &$this, 'endElement' ) );
  95                 xml_set_character_data_handler( $parser, array( &$this, 'characterData' ) );
  96
  97                 $offset = 0; // for context extraction on error reporting
  98                 $bufferSize = 512 * 1024;
  99                 do {
 100                         $chunk = fread( $input, $bufferSize );
 101                         if( !xml_parse( $parser, $chunk, feof( $input ) ) ) {
 102                                 wfDebug( "TextDumpPass::readDump encountered XML parsing error\n" );
 103                                 return new WikiXmlError( $parser, 'XML import parse failure', $chunk, $offset );
 104                         }
 105                         $offset += strlen( $chunk );
 106                 } while( $chunk !== false && !feof( $input ) );
 107                 xml_parser_free( $parser );
 108         }
 109
 110         function getText( $id ) {
 111                 if( isset( $this->prefetch ) ) {
 112                         $text = $this->prefetch->prefetch( $this->thisPage, $this->thisRev );
 113                         if( !is_null( $text ) )
 114                                 return $text;
 115                 }
 116                 $id = intval( $id );
 117                 $row = $this->db->selectRow( 'text',
 118                         array( 'old_text', 'old_flags' ),
 119                         array( 'old_id' => $id ),
 120                         'TextPassDumper::getText' );
 121                 $text = Revision::getRevisionText( $row );
 122                 $stripped = str_replace( "\r", "", $text );
 123                 $normalized = UtfNormal::cleanUp( $stripped );
 124                 return $normalized;
 125         }
 126
 127         function startElement( $parser, $name, $attribs ) {
 128                 $this->clearOpenElement( null );
 129                 $this->lastName = $name;
 130
 131                 if( $name == 'revision' ) {
 132                         $this->state = $name;
 133                         $this->egress->writeOpenPage( null, $this->buffer );
 134                         $this->buffer = "";
 135                 } elseif( $name == 'page' ) {
 136                         $this->state = $name;
 137                         if( $this->atStart ) {
 138                                 $this->egress->writeOpenStream( $this->buffer );
 139                                 $this->buffer = "";
 140                                 $this->atStart = false;
 141                         }
 142                 }
 143
 144                 if( $name == "text" && isset( $attribs['id'] ) ) {
 145                         $text = $this->getText( $attribs['id'] );
 146                         $this->openElement = array( $name, array( 'xml:space' => 'preserve' ) );
 147                         if( strlen( $text ) > 0 ) {
 148                                 $this->characterData( $parser, $text );
 149                         }
 150                 } else {
 151                         $this->openElement = array( $name, $attribs );
 152                 }
 153         }
 154
 155         function endElement( $parser, $name ) {
 156                 if( $this->openElement ) {
 157                         $this->clearOpenElement( "" );
 158                 } else {
 159                         $this->buffer .= "</$name>";
 160                 }
 161
 162                 if( $name == 'revision' ) {
 163                         $this->egress->writeRevision( null, $this->buffer );
 164                         $this->buffer = "";
 165                         $this->thisRev = "";
 166                 } elseif( $name == 'page' ) {
 167                         $this->egress->writeClosePage( $this->buffer );
 168                         $this->buffer = "";
 169                         $this->thisPage = "";
 170                 } elseif( $name == 'mediawiki' ) {
 171                         $this->egress->writeCloseStream( $this->buffer );
 172                         $this->buffer = "";
 173                 }
 174         }
 175
 176         function characterData( $parser, $data ) {
 177                 $this->clearOpenElement( null );
 178                 if( $this->lastName == "id" ) {
 179                         if( $this->state == "revision" ) {
 180                                 $this->thisRev .= $data;
 181                         } elseif( $this->state == "page" ) {
 182                                 $this->thisPage .= $data;
 183                         }
 184                 }
 185                 $this->buffer .= htmlspecialchars( $data );
 186         }
 187
 188         function clearOpenElement( $style ) {
 189                 if( $this->openElement ) {
 190                         $this->buffer .= wfElement( $this->openElement[0], $this->openElement[1], $style );
 191                         $this->openElement = false;
 192                 }
 193         }
 194 }
 195
 196
 197 $dumper = new TextPassDumper( $argv );
 198 if( isset( $options['server'] ) ) {
 199         $dumper->server = $options['server'];
 200 }
 201
 202 if( true ) {
 203         $dumper->dump();
 204 } else {
 205         $dumper->progress( <<<END
 206 This script postprocesses XML dumps from dumpBackup.php to add
 207 page text which was stubbed out (using --stub).
 208
 209 XML input is accepted on stdin.
 210 XML output is sent to stdout; progress reports are sent to stderr.
 211
 212 Usage: php dumpTextPass.php [<options>]
 213 Options:
 214   --prefetch <file>  Use a prior dump file as a text source where possible.
 215               (Requires PHP 5.0+ and the XMLReader PECL extension)
 216   --quiet     Don't dump status reports to stderr.
 217   --report=n  Report position and speed after every n pages processed.
 218               (Default: 100)
 219   --server=h  Force reading from MySQL server h
 220 END
 221 );
 222 }
 223
 224 ?>