maintenance/importLogs.inc

   1 <?php
   2 # Copyright (C) 2004 Brion Vibber <brion@pobox.com>
   3 # http://www.mediawiki.org/
   4 #
   5 # This program is free software; you can redistribute it and/or modify
   6 # it under the terms of the GNU General Public License as published by
   7 # the Free Software Foundation; either version 2 of the License, or
   8 # (at your option) any later version.
   9 #
  10 # This program is distributed in the hope that it will be useful,
  11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13 # GNU General Public License for more details.
  14 #
  15 # You should have received a copy of the GNU General Public License along
  16 # with this program; if not, write to the Free Software Foundation, Inc.,
  17 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18 # http://www.gnu.org/copyleft/gpl.html
  19
  20 /**
  21  * Attempt to import existing log pages into the log tables.
  22  *
  23  * Not yet complete.
  24  *
  25  * @file
  26  * @todo document
  27  * @ingroup Maintenance
  28  */
  29
  30 /** */
  31 require_once( 'GlobalFunctions.php' );
  32 require_once( 'Database.php' );
  33 require_once( 'Article.php' );
  34 require_once( 'LogPage.php' );
  35
  36 /**
  37  * Log importer
  38  * @todo document
  39  * @ingroup Maintenance
  40  */
  41 class LogImporter {
  42         var $dummy = false;
  43
  44         function LogImporter( $type ) {
  45                 $this->type = $type;
  46                 $this->db = wfGetDB( DB_MASTER );
  47                 $this->actions = $this->setupActions();
  48         }
  49
  50         function setupActions() {
  51                 $actions = array();
  52                 foreach( LogPage::validActions( $this->type ) as $action ) {
  53                         $key = "{$this->type}/$action";
  54                         $actions[$key] = $this->makeLineRegexp( $this->type, $action );
  55                 }
  56                 return $actions;
  57         }
  58
  59         function makeLineRegexp( $type, $action ) {
  60                 $linkRegexp = '(?:\[\[)?([^|\]]+?)(?:\|[^\]]+?)?(?:\]\])?';
  61                 $linkRegexp2 = '\[\[([^|\]]+?)(?:\|[^\]]+?)?\]\]';
  62
  63                 $text = LogPage::actionText( $type, $action );
  64                 $text = preg_quote( $text, '/' );
  65                 $text = str_replace( '\$1', $linkRegexp, $text );
  66                 $text = '^(.*?) ' . $linkRegexp2 . ' ' . $text;
  67                 $text .= '(?: <em>\((.*)\)<\/em>)?';
  68                 $text = "/$text/";
  69                 return $text;
  70         }
  71
  72         function importText( $text ) {
  73                 if( $this->dummy ) {
  74                         print $text;
  75                         var_dump( $this->actions );
  76                 }
  77                 $lines = explode( '<li>', $text );
  78                 foreach( $lines as $line ) {
  79                         $matches = array();
  80                         if( preg_match( '!^(.*)</li>!', $line, $matches ) ) {
  81                                 $this->importLine( $matches[1] );
  82                         }
  83                 }
  84         }
  85
  86         function fixDate( $date ) {
  87                 # Yuck! Parsing multilingual date formats??!!!!???!!??!
  88                 # 01:55, 23 Aug 2004 - won't take in strtotimr
  89                 # "Aug 23 2004 01:55" - seems ok
  90                 # TODO: multilingual attempt to extract from the data in Language
  91                 $matches = array();
  92                 if( preg_match( '/^(\d+:\d+(?::\d+)?), (.*)$/', $date, $matches ) ) {
  93                         $date = $matches[2] . ' ' . $matches[1];
  94                 }
  95                 $n = strtotime( $date ) + date("Z");
  96                 # print gmdate( 'D, d M Y H:i:s T', $n ) . "\n";
  97                 $timestamp = wfTimestamp( TS_MW, $n );
  98                 return $timestamp;
  99         }
 100
 101         function importLine( $line ) {
 102                 foreach( $this->actions as $action => $regexp ) {
 103                         $matches = array();
 104                         if( preg_match( $regexp, $line, $matches ) ) {
 105                                 if( $this->dummy ) {
 106                                         #var_dump( $matches );
 107                                 }
 108                                 $date = $this->fixDate( $matches[1] );
 109                                 $user = Title::newFromText( $matches[2] );
 110                                 $target = Title::newFromText( $matches[3] );
 111                                 if( isset( $matches[4] ) ) {
 112                                         $comment = $matches[4];
 113                                 } else {
 114                                         $comment = '';
 115                                 }
 116
 117                                 $insert = array(
 118                                         'log_type' => $this->type,
 119                                         'log_action' => preg_replace( '!^.*/!', '', $action ),
 120                                         'log_timestamp' => $date,
 121                                         'log_user' => intval( User::idFromName( $user->getText() ) ),
 122                                         'log_namespace' => $target->getNamespace(),
 123                                         'log_title' => $target->getDBkey(),
 124                                         'log_comment' => wfUnescapeWikiText( $comment ),
 125                                 );
 126                                 if( $this->dummy ) {
 127                                         var_dump( $insert );
 128                                 } else {
 129                                         # FIXME: avoid duplicates!
 130                                         $this->db->insert( 'logging', $insert );
 131                                 }
 132                                 break;
 133                         }
 134                 }
 135         }
 136 }
 137
 138 function wfUnescapeWikiText( $text ) {
 139         $text = str_replace(
 140                 array( '&#91;', '&#124;', '&#39;', 'ISBN&#32;', '&#58;//' , "\n&#61;", '&#123;&#123;' ),
 141                 array( '[',             '|',      "'",     'ISBN '        , '://'         , "\n=", '{{' ),
 142                 $text );
 143         return $text;
 144 }