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         var $input = "php://stdin";
  36
  37         function dump() {
  38                 # This shouldn't happen if on console... ;)
  39                 header( 'Content-type: text/html; charset=UTF-8' );
  40
  41                 # Notice messages will foul up your XML output even if they're
  42                 # relatively harmless.
  43 //              ini_set( 'display_errors', false );
  44
  45                 $this->startTime = wfTime();
  46
  47                 $this->db =& $this->backupDb();
  48                 $this->maxCount = $this->db->selectField( 'page', 'MAX(page_id)', '', 'BackupDumper::dump' );
  49                 $this->startTime = wfTime();
  50
  51                 $this->egress = new ExportProgressFilter( $this->sink, $this );
  52
  53                 $input = fopen( $this->input, "rt" );
  54                 $result = $this->readDump( $input );
  55
  56                 if( WikiError::isError( $result ) ) {
  57                         wfDie( $result->getMessage() );
  58                 }
  59
  60                 $this->report( true );
  61         }
  62
  63         function processOption( $opt, $val, $param ) {
  64                 $url = $this->processFileOpt( $val, $param );
  65
  66                 switch( $opt ) {
  67                 case 'prefetch':
  68                         require_once 'maintenance/backupPrefetch.inc';
  69                         $this->prefetch = new BaseDump( $url );
  70                         break;
  71                 case 'stub':
  72                         $this->input = $url;
  73                         break;
  74                 }
  75         }
  76
  77         function processFileOpt( $val, $param ) {
  78                 switch( $val ) {
  79                 case "file":
  80                         return $param;
  81                 case "gzip":
  82                         return "compress.zlib://$param";
  83                 case "bzip2":
  84                         return "compress.bzip2://$param";
  85                 default:
  86                         return $val;
  87                 }
  88         }
  89
  90         function readDump( $input ) {
  91                 $this->buffer = "";
  92                 $this->openElement = false;
  93                 $this->atStart = true;
  94                 $this->state = "";
  95                 $this->lastName = "";
  96                 $this->thisPage = 0;
  97                 $this->thisRev = 0;
  98
  99                 $parser = xml_parser_create( "UTF-8" );
 100                 xml_parser_set_option( $parser, XML_OPTION_CASE_FOLDING, false );
 101
 102                 xml_set_element_handler( $parser, array( &$this, 'startElement' ), array( &$this, 'endElement' ) );
 103                 xml_set_character_data_handler( $parser, array( &$this, 'characterData' ) );
 104
 105                 $offset = 0; // for context extraction on error reporting
 106                 $bufferSize = 512 * 1024;
 107                 do {
 108                         $chunk = fread( $input, $bufferSize );
 109                         if( !xml_parse( $parser, $chunk, feof( $input ) ) ) {
 110                                 wfDebug( "TextDumpPass::readDump encountered XML parsing error\n" );
 111                                 return new WikiXmlError( $parser, 'XML import parse failure', $chunk, $offset );
 112                         }
 113                         $offset += strlen( $chunk );
 114                 } while( $chunk !== false && !feof( $input ) );
 115                 xml_parser_free( $parser );
 116
 117                 return true;
 118         }
 119
 120         function getText( $id ) {
 121                 if( isset( $this->prefetch ) ) {
 122                         $text = $this->prefetch->prefetch( $this->thisPage, $this->thisRev );
 123                         if( !is_null( $text ) )
 124                                 return $text;
 125                 }
 126                 $id = intval( $id );
 127                 $row = $this->db->selectRow( 'text',
 128                         array( 'old_text', 'old_flags' ),
 129                         array( 'old_id' => $id ),
 130                         'TextPassDumper::getText' );
 131                 $text = Revision::getRevisionText( $row );
 132                 $stripped = str_replace( "\r", "", $text );
 133                 $normalized = UtfNormal::cleanUp( $stripped );
 134                 return $normalized;
 135         }
 136
 137         function startElement( $parser, $name, $attribs ) {
 138                 $this->clearOpenElement( null );
 139                 $this->lastName = $name;
 140
 141                 if( $name == 'revision' ) {
 142                         $this->state = $name;
 143                         $this->egress->writeOpenPage( null, $this->buffer );
 144                         $this->buffer = "";
 145                 } elseif( $name == 'page' ) {
 146                         $this->state = $name;
 147                         if( $this->atStart ) {
 148                                 $this->egress->writeOpenStream( $this->buffer );
 149                                 $this->buffer = "";
 150                                 $this->atStart = false;
 151                         }
 152                 }
 153
 154                 if( $name == "text" && isset( $attribs['id'] ) ) {
 155                         $text = $this->getText( $attribs['id'] );
 156                         $this->openElement = array( $name, array( 'xml:space' => 'preserve' ) );
 157                         if( strlen( $text ) > 0 ) {
 158                                 $this->characterData( $parser, $text );
 159                         }
 160                 } else {
 161                         $this->openElement = array( $name, $attribs );
 162                 }
 163         }
 164
 165         function endElement( $parser, $name ) {
 166                 if( $this->openElement ) {
 167                         $this->clearOpenElement( "" );
 168                 } else {
 169                         $this->buffer .= "</$name>";
 170                 }
 171
 172                 if( $name == 'revision' ) {
 173                         $this->egress->writeRevision( null, $this->buffer );
 174                         $this->buffer = "";
 175                         $this->thisRev = "";
 176                 } elseif( $name == 'page' ) {
 177                         $this->egress->writeClosePage( $this->buffer );
 178                         $this->buffer = "";
 179                         $this->thisPage = "";
 180                 } elseif( $name == 'mediawiki' ) {
 181                         $this->egress->writeCloseStream( $this->buffer );
 182                         $this->buffer = "";
 183                 }
 184         }
 185
 186         function characterData( $parser, $data ) {
 187                 $this->clearOpenElement( null );
 188                 if( $this->lastName == "id" ) {
 189                         if( $this->state == "revision" ) {
 190                                 $this->thisRev .= $data;
 191                         } elseif( $this->state == "page" ) {
 192                                 $this->thisPage .= $data;
 193                         }
 194                 }
 195                 $this->buffer .= htmlspecialchars( $data );
 196         }
 197
 198         function clearOpenElement( $style ) {
 199                 if( $this->openElement ) {
 200                         $this->buffer .= wfElement( $this->openElement[0], $this->openElement[1], $style );
 201                         $this->openElement = false;
 202                 }
 203         }
 204 }
 205
 206
 207 $dumper = new TextPassDumper( $argv );
 208 if( isset( $options['server'] ) ) {
 209         $dumper->server = $options['server'];
 210 }
 211
 212 if( true ) {
 213         $dumper->dump();
 214 } else {
 215         $dumper->progress( <<<END
 216 This script postprocesses XML dumps from dumpBackup.php to add
 217 page text which was stubbed out (using --stub).
 218
 219 XML input is accepted on stdin.
 220 XML output is sent to stdout; progress reports are sent to stderr.
 221
 222 Usage: php dumpTextPass.php [<options>]
 223 Options:
 224   --stub=<type>:<file> To load a compressed stub dump instead of stdin
 225   --prefetch=<type>:<file> Use a prior dump file as a text source, to save
 226               pressure on the database.
 227               (Requires PHP 5.0+ and the XMLReader PECL extension)
 228   --quiet     Don't dump status reports to stderr.
 229   --report=n  Report position and speed after every n pages processed.
 230               (Default: 100)
 231   --server=h  Force reading from MySQL server h
 232 END
 233 );
 234 }
 235
 236 ?>