fiveupgrade.inc

来自「php 开发的内容管理系统」· INC 代码 · 共 1,215 行 · 第 1/3 页

<?php

require_once( 'cleanupDupes.inc' );
require_once( 'userDupes.inc' );
require_once( 'updaters.inc' );

define( 'MW_UPGRADE_COPY',     false );
define( 'MW_UPGRADE_ENCODE',   true  );
define( 'MW_UPGRADE_NULL',     null  );
define( 'MW_UPGRADE_CALLBACK', null  ); // for self-documentation only

class FiveUpgrade {
	function FiveUpgrade() {
		global $wgDatabase;
		$this->conversionTables = $this->prepareWindows1252();

		$this->dbw =& $this->newConnection();
		$this->dbr =& $this->streamConnection();

		$this->cleanupSwaps = array();
		$this->emailAuth = false; # don't preauthenticate emails
		$this->maxLag    = 10; # if slaves are lagged more than 10 secs, wait
	}

	function doing( $step ) {
		return is_null( $this->step ) || $step == $this->step;
	}

	function upgrade( $step ) {
		$this->step = $step;

		$tables = array(
			'page',
			'links',
			'user',
			'image',
			'oldimage',
			'watchlist',
			'logging',
			'archive',
			'imagelinks',
			'categorylinks',
			'ipblocks',
			'recentchanges',
			'querycache' );
		foreach( $tables as $table ) {
			if( $this->doing( $table ) ) {
				$method = 'upgrade' . ucfirst( $table );
				$this->$method();
			}
		}

		if( $this->doing( 'cleanup' ) ) {
			$this->upgradeCleanup();
		}
	}


	/**
	 * Open a connection to the master server with the admin rights.
	 * @return Database
	 * @access private
	 */
	function &newConnection() {
		global $wgDBadminuser, $wgDBadminpassword;
		global $wgDBserver, $wgDBname;
		$db =& new Database( $wgDBserver, $wgDBadminuser, $wgDBadminpassword, $wgDBname );
		return $db;
	}

	/**
	 * Open a second connection to the master server, with buffering off.
	 * This will let us stream large datasets in and write in chunks on the
	 * other end.
	 * @return Database
	 * @access private
	 */
	function &streamConnection() {
		$timeout = 3600 * 24;
		$db =& $this->newConnection();
		$db->bufferResults( false );
		$db->query( "SET net_read_timeout=$timeout" );
		$db->query( "SET net_write_timeout=$timeout" );
		return $db;
	}

	/**
	 * Prepare a conversion array for converting Windows Code Page 1252 to
	 * UTF-8. This should provide proper conversion of text that was miscoded
	 * as Windows-1252 by naughty user-agents, and doesn't rely on an outside
	 * iconv library.
	 *
	 * @return array
	 * @access private
	 */
	function prepareWindows1252() {
		# Mappings from:
		# http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP1252.TXT
		static $cp1252 = array(
			0x80 => 0x20AC,	#EURO SIGN
			0x81 => UNICODE_REPLACEMENT,
			0x82 => 0x201A,	#SINGLE LOW-9 QUOTATION MARK
			0x83 => 0x0192,	#LATIN SMALL LETTER F WITH HOOK
			0x84 => 0x201E,	#DOUBLE LOW-9 QUOTATION MARK
			0x85 => 0x2026,	#HORIZONTAL ELLIPSIS
			0x86 => 0x2020,	#DAGGER
			0x87 => 0x2021,	#DOUBLE DAGGER
			0x88 => 0x02C6,	#MODIFIER LETTER CIRCUMFLEX ACCENT
			0x89 => 0x2030,	#PER MILLE SIGN
			0x8A => 0x0160,	#LATIN CAPITAL LETTER S WITH CARON
			0x8B => 0x2039,	#SINGLE LEFT-POINTING ANGLE QUOTATION MARK
			0x8C => 0x0152,	#LATIN CAPITAL LIGATURE OE
			0x8D => UNICODE_REPLACEMENT,
			0x8E => 0x017D,	#LATIN CAPITAL LETTER Z WITH CARON
			0x8F => UNICODE_REPLACEMENT,
			0x90 => UNICODE_REPLACEMENT,
			0x91 => 0x2018,	#LEFT SINGLE QUOTATION MARK
			0x92 => 0x2019,	#RIGHT SINGLE QUOTATION MARK
			0x93 => 0x201C,	#LEFT DOUBLE QUOTATION MARK
			0x94 => 0x201D,	#RIGHT DOUBLE QUOTATION MARK
			0x95 => 0x2022,	#BULLET
			0x96 => 0x2013,	#EN DASH
			0x97 => 0x2014,	#EM DASH
			0x98 => 0x02DC,	#SMALL TILDE
			0x99 => 0x2122,	#TRADE MARK SIGN
			0x9A => 0x0161,	#LATIN SMALL LETTER S WITH CARON
			0x9B => 0x203A,	#SINGLE RIGHT-POINTING ANGLE QUOTATION MARK
			0x9C => 0x0153,	#LATIN SMALL LIGATURE OE
			0x9D => UNICODE_REPLACEMENT,
			0x9E => 0x017E,	#LATIN SMALL LETTER Z WITH CARON
			0x9F => 0x0178,	#LATIN CAPITAL LETTER Y WITH DIAERESIS
			);
		$pairs = array();
		for( $i = 0; $i < 0x100; $i++ ) {
			$unicode = isset( $cp1252[$i] ) ? $cp1252[$i] : $i;
			$pairs[chr( $i )] = codepointToUtf8( $unicode );
		}
		return $pairs;
	}

	/**
	 * Convert from 8-bit Windows-1252 to UTF-8 if necessary.
	 * @param string $text
	 * @return string
	 * @access private
	 */
	function conv( $text ) {
		global $wgUseLatin1;
		return is_null( $text )
			? null
			: ( $wgUseLatin1
				? strtr( $text, $this->conversionTables )
				: $text );
	}

	/**
	 * Dump timestamp and message to output
	 * @param string $message
	 * @access private
	 */
	function log( $message ) {
		global $wgDBname;
		echo $wgDBname . ' ' . wfTimestamp( TS_DB ) . ': ' . $message . "\n";
		flush();
	}

	/**
	 * Initialize the chunked-insert system.
	 * Rows will be inserted in chunks of the given number, rather
	 * than in a giant INSERT...SELECT query, to keep the serialized
	 * MySQL database replication from getting hung up. This way other
	 * things can be going on during conversion without waiting for
	 * slaves to catch up as badly.
	 *
	 * @param int $chunksize Number of rows to insert at once
	 * @param int $final Total expected number of rows / id of last row,
	 *                   used for progress reports.
	 * @param string $table to insert on
	 * @param string $fname function name to report in SQL
	 * @access private
	 */
	function setChunkScale( $chunksize, $final, $table, $fname ) {
		$this->chunkSize  = $chunksize;
		$this->chunkFinal = $final;
		$this->chunkCount = 0;
		$this->chunkStartTime = wfTime();
		$this->chunkOptions = array( 'IGNORE' );
		$this->chunkTable = $table;
		$this->chunkFunction = $fname;
	}

	/**
	 * Chunked inserts: perform an insert if we've reached the chunk limit.
	 * Prints a progress report with estimated completion time.
	 * @param array &$chunk -- This will be emptied if an insert is done.
	 * @param int $key A key identifier to use in progress estimation in
	 *                 place of the number of rows inserted. Use this if
	 *                 you provided a max key number instead of a count
	 *                 as the final chunk number in setChunkScale()
	 * @access private
	 */
	function addChunk( &$chunk, $key = null ) {
		if( count( $chunk ) >= $this->chunkSize ) {
			$this->insertChunk( $chunk );

			$this->chunkCount += count( $chunk );
			$now = wfTime();
			$delta = $now - $this->chunkStartTime;
			$rate = $this->chunkCount / $delta;

			if( is_null( $key ) ) {
				$completed = $this->chunkCount;
			} else {
				$completed = $key;
			}
			$portion = $completed / $this->chunkFinal;

			$estimatedTotalTime = $delta / $portion;
			$eta = $this->chunkStartTime + $estimatedTotalTime;

			printf( "%s: %6.2f%% done on %s; ETA %s [%d/%d] %.2f/sec\n",
				wfTimestamp( TS_DB, intval( $now ) ),
				$portion * 100.0,
				$this->chunkTable,
				wfTimestamp( TS_DB, intval( $eta ) ),
				$completed,
				$this->chunkFinal,
				$rate );
			flush();

			$chunk = array();
		}
	}

	/**
	 * Chunked inserts: perform an insert unconditionally, at the end, and log.
	 * @param array &$chunk -- This will be emptied if an insert is done.
	 * @access private
	 */
	function lastChunk( &$chunk ) {
		$n = count( $chunk );
		if( $n > 0 ) {
			$this->insertChunk( $chunk );
		}
		$this->log( "100.00% done on $this->chunkTable (last chunk $n rows)." );
	}

	/**
	 * Chunked inserts: perform an insert.
	 * @param array &$chunk -- This will be emptied if an insert is done.
	 * @access private
	 */
	function insertChunk( &$chunk ) {
		// Give slaves a chance to catch up
		wfWaitForSlaves( $this->maxLag );
		$this->dbw->insert( $this->chunkTable, $chunk, $this->chunkFunction, $this->chunkOptions );
	}


	/**
	 * Copy and transcode a table to table_temp.
	 * @param string $name Base name of the source table
	 * @param string $tabledef CREATE TABLE definition, w/ $1 for the name
	 * @param array $fields set of destination fields to these constants:
	 *              MW_UPGRADE_COPY   - straight copy
	 *              MW_UPGRADE_ENCODE - for old Latin1 wikis, conv to UTF-8
	 *              MW_UPGRADE_NULL   - just put NULL
	 * @param callable $callback An optional callback to modify the data
	 *                           or perform other processing. Func should be
	 *                           ( object $row, array $copy ) and return $copy
	 * @access private
	 */
	function copyTable( $name, $tabledef, $fields, $callback = null ) {
		$fname = 'FiveUpgrade::copyTable';

		$name_temp = $name . '_temp';
		$this->log( "Migrating $name table to $name_temp..." );

		$table      = $this->dbw->tableName( $name );
		$table_temp = $this->dbw->tableName( $name_temp );

		// Create temporary table; we're going to copy everything in there,
		// then at the end rename the final tables into place.
		$def = str_replace( '$1', $table_temp, $tabledef );
		$this->dbw->query( $def, $fname );

		$numRecords = $this->dbw->selectField( $name, 'COUNT(*)', '', $fname );
		$this->setChunkScale( 100, $numRecords, $name_temp, $fname );

		// Pull all records from the second, streaming database connection.
		$sourceFields = array_keys( array_filter( $fields,
			create_function( '$x', 'return $x !== MW_UPGRADE_NULL;' ) ) );
		$result = $this->dbr->select( $name,
			$sourceFields,
			'',
			$fname );

		$add = array();
		while( $row = $this->dbr->fetchObject( $result ) ) {
			$copy = array();
			foreach( $fields as $field => $source ) {
				if( $source === MW_UPGRADE_COPY ) {
					$copy[$field] = $row->$field;
				} elseif( $source === MW_UPGRADE_ENCODE ) {
					$copy[$field] = $this->conv( $row->$field );
				} elseif( $source === MW_UPGRADE_NULL ) {
					$copy[$field] = null;
				} else {
					$this->log( "Unknown field copy type: $field => $source" );
				}
			}
			if( is_callable( $callback ) ) {
				$copy = call_user_func( $callback, $row, $copy );
			}
			$add[] = $copy;
			$this->addChunk( $add );
		}
		$this->lastChunk( $add );
		$this->dbr->freeResult( $result );

		$this->log( "Done converting $name." );
		$this->cleanupSwaps[] = $name;
	}

	function upgradePage() {
		$fname = "FiveUpgrade::upgradePage";
		$chunksize = 100;

		if( $this->dbw->tableExists( 'page' ) ) {
			$this->log( 'Page table already exists; aborting.' );
			die( -1 );
		}

		$this->log( "Checking cur table for unique title index and applying if necessary" );
		checkDupes( true );

		$this->log( "...converting from cur/old to page/revision/text DB structure." );

		extract( $this->dbw->tableNames( 'cur', 'old', 'page', 'revision', 'text' ) );

		$this->log( "Creating page and revision tables..." );
		$this->dbw->query("CREATE TABLE $page (
  			page_id int(8) unsigned NOT NULL auto_increment,
  			page_namespace int NOT NULL,
  			page_title varchar(255) binary NOT NULL,
  			page_restrictions tinyblob NOT NULL default '',
  			page_counter bigint(20) unsigned NOT NULL default '0',
  			page_is_redirect tinyint(1) unsigned NOT NULL default '0',
  			page_is_new tinyint(1) unsigned NOT NULL default '0',
  			page_random real unsigned NOT NULL,
  			page_touched char(14) binary NOT NULL default '',
  			page_latest int(8) unsigned NOT NULL,
  			page_len int(8) unsigned NOT NULL,

  			PRIMARY KEY page_id (page_id),
  			UNIQUE INDEX name_title (page_namespace,page_title),
  			INDEX (page_random),
  			INDEX (page_len)
			) TYPE=InnoDB", $fname );
		$this->dbw->query("CREATE TABLE $revision (
  			rev_id int(8) unsigned NOT NULL auto_increment,
  			rev_page int(8) unsigned NOT NULL,
  			rev_text_id int(8) unsigned NOT NULL,
  			rev_comment tinyblob NOT NULL default '',
  			rev_user int(5) unsigned NOT NULL default '0',
  			rev_user_text varchar(255) binary NOT NULL default '',
  			rev_timestamp char(14) binary NOT NULL default '',
  			rev_minor_edit tinyint(1) unsigned NOT NULL default '0',
			rev_deleted tinyint(1) unsigned NOT NULL default '0',

  			PRIMARY KEY rev_page_id (rev_page, rev_id),
  			UNIQUE INDEX rev_id (rev_id),
  			INDEX rev_timestamp (rev_timestamp),
  			INDEX page_timestamp (rev_page,rev_timestamp),
  			INDEX user_timestamp (rev_user,rev_timestamp),
  			INDEX usertext_timestamp (rev_user_text,rev_timestamp)
			) TYPE=InnoDB", $fname );

		$maxold = intval( $this->dbw->selectField( 'old', 'max(old_id)', '', $fname ) );
		$this->log( "Last old record is {$maxold}" );

		global $wgLegacySchemaConversion;
		if( $wgLegacySchemaConversion ) {
			// Create HistoryBlobCurStub entries.
			// Text will be pulled from the leftover 'cur' table at runtime.
			echo "......Moving metadata from cur; using blob references to text in cur table.\n";
			$cur_text = "concat('O:18:\"historyblobcurstub\":1:{s:6:\"mCurId\";i:',cur_id,';}')";
			$cur_flags = "'object'";
		} else {
			// Copy all cur text in immediately: this may take longer but avoids
			// having to keep an extra table around.
			echo "......Moving text from cur.\n";
			$cur_text = 'cur_text';
			$cur_flags = "''";
		}

		$maxcur = $this->dbw->selectField( 'cur', 'max(cur_id)', '', $fname );
		$this->log( "Last cur entry is $maxcur" );

		/**
		 * Copy placeholder records for each page's current version into old
		 * Don't do any conversion here; text records are converted at runtime
		 * based on the flags (and may be originally binary!) while the meta
		 * fields will be converted in the old -> rev and cur -> page steps.
		 */