X-Git-Url: https://git.heureux-cyclage.org/?a=blobdiff_plain;f=includes%2FDifferenceEngine.php;h=e6218b6d9c1600a334dd0e26326e177e12085300;hb=5fa40fe0fbfb379022430c24fcf66392e229d9a5;hp=198c9b3931c8903c6555880dd80fbc5218a26a22;hpb=332a0cf2684024869536a2af1ef83990efcb7b3c;p=lhc%2Fweb%2Fwiklou.git diff --git a/includes/DifferenceEngine.php b/includes/DifferenceEngine.php index 198c9b3931..e6218b6d9c 100644 --- a/includes/DifferenceEngine.php +++ b/includes/DifferenceEngine.php @@ -1,47 +1,53 @@ mTitle = $titleObj; + wfDebug("DifferenceEngine old '$old' new '$new' rcid '$rcid'\n"); + + if ( 'prev' === $new ) { # Show diff between revision $old and the previous one. # Get previous one from DB. # $this->mNewid = intval($old); - $dbr =& wfGetDB( DB_SLAVE ); - $this->mOldid = $dbr->selectField( 'old', 'old_id', - "old_title='" . $wgTitle->getDBkey() . "'" . - ' AND old_namespace=' . $wgTitle->getNamespace() . - " AND old_id<{$this->mNewid} ORDER BY old_id DESC" ); - } elseif ( 'next' == $new ) { + $this->mOldid = $this->mTitle->getPreviousRevisionID( $this->mNewid ); + } elseif ( 'next' === $new ) { # Show diff between revision $old and the previous one. # Get previous one from DB. # $this->mOldid = intval($old); - $dbr =& wfGetDB( DB_SLAVE ); - $this->mNewid = $dbr->selectField( 'old', 'old_id', - "old_title='" . $wgTitle->getDBkey() . "'" . - ' AND old_namespace=' . $wgTitle->getNamespace() . - " AND old_id>{$this->mOldid} ORDER BY old_id " ); + $this->mNewid = $this->mTitle->getNextRevisionID( $this->mOldid ); if ( false === $this->mNewid ) { # if no result, NewId points to the newest old revision. The only newer # revision is cur, which is "0". @@ -49,39 +55,72 @@ class DifferenceEngine { } } else { - $this->mOldid = intval($old); $this->mNewid = intval($new); } $this->mRcidMarkPatrolled = intval($rcid); # force it to be an integer } - function showDiffPage() - { - global $wgUser, $wgTitle, $wgOut, $wgContLang, $wgOnlySysopsCanPatrol, $wgUseRCPatrol; + function showDiffPage( $diffOnly = false ) { + global $wgUser, $wgOut, $wgUseExternalEditor, $wgUseRCPatrol; $fname = 'DifferenceEngine::showDiffPage'; wfProfileIn( $fname ); + # If external diffs are enabled both globally and for the user, + # we'll use the application/x-external-editor interface to call + # an external diff tool like kompare, kdiff3, etc. + if($wgUseExternalEditor && $wgUser->getOption('externaldiff')) { + global $wgInputEncoding,$wgServer,$wgScript,$wgLang; + $wgOut->disable(); + header ( "Content-type: application/x-external-editor; charset=".$wgInputEncoding ); + $url1=$this->mTitle->getFullURL("action=raw&oldid=".$this->mOldid); + $url2=$this->mTitle->getFullURL("action=raw&oldid=".$this->mNewid); + $special=$wgLang->getNsText(NS_SPECIAL); + $control=<<setArticleFlag( false ); + if ( ! $this->loadRevisionData() ) { + $t = $this->mTitle->getPrefixedText() . " (Diff: {$this->mOldid}, {$this->mNewid})"; + $mtext = wfMsg( 'missingarticle', "$t" ); + $wgOut->setPagetitle( wfMsg( 'errorpagetitle' ) ); + $wgOut->addWikitext( $mtext ); + wfProfileOut( $fname ); + return; + } + + wfRunHooks( 'DiffViewHeader', array( $this, $this->mOldRev, $this->mNewRev ) ); + + if ( $this->mNewRev->isCurrent() ) { + $wgOut->setArticleFlag( true ); + } + # mOldid is false if the difference engine is called with a "vague" query for # a diff between a version V and its previous version V' AND the version V # is the first version of that article. In that case, V' does not exist. if ( $this->mOldid === false ) { $this->showFirstRevision(); + $this->renderNewRevision(); // should we respect $diffOnly here or not? wfProfileOut( $fname ); return; } - $t = $wgTitle->getPrefixedText() . " (Diff: {$this->mOldid}, " . - "{$this->mNewid})"; - $mtext = wfMsg( 'missingarticle', $t ); - - $wgOut->setArticleFlag( false ); - if ( ! $this->loadText() ) { - $wgOut->setPagetitle( wfMsg( 'errorpagetitle' ) ); - $wgOut->addHTML( $mtext ); - wfProfileOut( $fname ); - return; - } $wgOut->suppressQuickbar(); $oldTitle = $this->mOldPage->getPrefixedText(); @@ -92,7 +131,7 @@ class DifferenceEngine { $wgOut->setPageTitle( $oldTitle . ', ' . $newTitle ); } $wgOut->setSubtitle( wfMsg( 'difference' ) ); - $wgOut->setRobotpolicy( 'noindex,follow' ); + $wgOut->setRobotpolicy( 'noindex,nofollow' ); if ( !( $this->mOldPage->userCanRead() && $this->mNewPage->userCanRead() ) ) { $wgOut->loginToUse(); @@ -102,85 +141,117 @@ class DifferenceEngine { } $sk = $wgUser->getSkin(); - $talk = $wgContLang->getNsText( NS_TALK ); - $contribs = wfMsg( 'contribslink' ); - - $this->mOldComment = $sk->formatComment($this->mOldComment); - $this->mNewComment = $sk->formatComment($this->mNewComment); - - $oldUserLink = $sk->makeLinkObj( Title::makeTitleSafe( NS_USER, $this->mOldUser ), $this->mOldUser ); - $newUserLink = $sk->makeLinkObj( Title::makeTitleSafe( NS_USER, $this->mNewUser ), $this->mNewUser ); - $oldUTLink = $sk->makeLinkObj( Title::makeTitleSafe( NS_USER_TALK, $this->mOldUser ), $talk ); - $newUTLink = $sk->makeLinkObj( Title::makeTitleSafe( NS_USER_TALK, $this->mNewUser ), $talk ); - $oldContribs = $sk->makeKnownLinkObj( Title::makeTitle( NS_SPECIAL, 'Contributions' ), $contribs, - 'target=' . urlencode($this->mOldUser) ); - $newContribs = $sk->makeKnownLinkObj( Title::makeTitle( NS_SPECIAL, 'Contributions' ), $contribs, - 'target=' . urlencode($this->mNewUser) ); - if ( !$this->mNewid && $wgUser->isSysop() ) { - $rollback = '   [' . $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'rollbacklink' ), - 'action=rollback&from=' . urlencode($this->mNewUser) ) . ']'; + + if ( $this->mNewRev->isCurrent() && $wgUser->isAllowed('rollback') ) { + $rollback = '   ' . $sk->generateRollback( $this->mNewRev ); } else { $rollback = ''; } - if ( $wgUseRCPatrol && $this->mRcidMarkPatrolled != 0 && $wgUser->getID() != 0 && - ( $wgUser->isSysop() || !$wgOnlySysopsCanPatrol ) ) - { - $patrol = ' [' . $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'markaspatrolleddiff' ), - "action=markpatrolled&rcid={$this->mRcidMarkPatrolled}" ) . ']'; + if( $wgUseRCPatrol && $this->mRcidMarkPatrolled != 0 && $wgUser->isAllowed( 'patrol' ) ) { + $patrol = ' [' . $sk->makeKnownLinkObj( $this->mTitle, wfMsg( 'markaspatrolleddiff' ), "action=markpatrolled&rcid={$this->mRcidMarkPatrolled}" ) . ']'; } else { $patrol = ''; } - $prevlink = $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'previousdiff' ), 'diff=prev&oldid='.$this->mOldid ); - if ( $this->mNewid == 0 ) { - $nextlink = ''; + $prevlink = $sk->makeKnownLinkObj( $this->mTitle, wfMsgHtml( 'previousdiff' ), + 'diff=prev&oldid='.$this->mOldid, '', '', 'id="differences-prevlink"' ); + if ( $this->mNewRev->isCurrent() ) { + $nextlink = ' '; } else { - $nextlink = $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'nextdiff' ), 'diff=next&oldid='.$this->mNewid ); + $nextlink = $sk->makeKnownLinkObj( $this->mTitle, wfMsgHtml( 'nextdiff' ), + 'diff=next&oldid='.$this->mNewid, '', '', 'id="differences-nextlink"' ); + } + + $oldminor = ''; + $newminor = ''; + + if ($this->mOldRev->mMinorEdit == 1) { + $oldminor = wfElement( 'span', array( 'class' => 'minor' ), + wfMsg( 'minoreditletter') ) . ' '; } - $oldHeader = "{$this->mOldtitle}
$oldUserLink " . - "($oldUTLink | $oldContribs)
" . $this->mOldComment . - '
' . $prevlink; - $newHeader = "{$this->mNewtitle}
$newUserLink " . - "($newUTLink | $newContribs) $rollback
" . $this->mNewComment . - '
' . $nextlink . $patrol; + if ($this->mNewRev->mMinorEdit == 1) { + $newminor = wfElement( 'span', array( 'class' => 'minor' ), + wfMsg( 'minoreditletter') ) . ' '; + } + + $oldHeader = "{$this->mOldtitle}
" . + $sk->revUserTools( $this->mOldRev ) . "
" . + $oldminor . $sk->revComment( $this->mOldRev, !$diffOnly ) . "
" . + $prevlink; + $newHeader = "{$this->mNewtitle}
" . + $sk->revUserTools( $this->mNewRev ) . " $rollback
" . + $newminor . $sk->revComment( $this->mNewRev, !$diffOnly ) . "
" . + $nextlink . $patrol; + + $this->showDiff( $oldHeader, $newHeader ); + + if ( !$diffOnly ) + $this->renderNewRevision(); + + wfProfileOut( $fname ); + } + + /** + * Show the new revision of the page. + */ + function renderNewRevision() { + global $wgOut; + $fname = 'DifferenceEngine::renderNewRevision'; + wfProfileIn( $fname ); - DifferenceEngine::showDiff( $this->mOldtext, $this->mNewtext, - $oldHeader, $newHeader ); $wgOut->addHTML( "

{$this->mPagetitle}

\n" ); - $wgOut->addWikiText( $this->mNewtext ); + #add deleted rev tag if needed + if ( !$this->mNewRev->userCan(Revision::DELETED_TEXT) ) { + $wgOut->addWikiText( wfMsg( 'rev-deleted-text-permission' ) ); + } + + if( !$this->mNewRev->isCurrent() ) { + $oldEditSectionSetting = $wgOut->parserOptions()->setEditSection( false ); + } + + $this->loadNewText(); + if( is_object( $this->mNewRev ) ) { + $wgOut->setRevisionId( $this->mNewRev->getId() ); + } + + $wgOut->addWikiTextTidy( $this->mNewtext ); + + if( !$this->mNewRev->isCurrent() ) { + $wgOut->parserOptions()->setEditSection( $oldEditSectionSetting ); + } wfProfileOut( $fname ); } - # Show the first revision of an article. Uses normal diff headers in contrast to normal - # "old revision" display style. - # - function showFirstRevision() - { - global $wgOut, $wgTitle, $wgUser, $wgLang; + /** + * Show the first revision of an article. Uses normal diff headers in + * contrast to normal "old revision" display style. + */ + function showFirstRevision() { + global $wgOut, $wgUser; $fname = 'DifferenceEngine::showFirstRevision'; wfProfileIn( $fname ); - - $this->mOldid = $this->mNewid; # hack to make loadText() work. - # Get article text from the DB # - if ( ! $this->loadText() ) { - $t = $wgTitle->getPrefixedText() . " (Diff: {$this->mOldid}, " . + if ( ! $this->loadNewText() ) { + $t = $this->mTitle->getPrefixedText() . " (Diff: {$this->mOldid}, " . "{$this->mNewid})"; - $mtext = wfMsg( 'missingarticle', $t ); + $mtext = wfMsg( 'missingarticle', "$t" ); $wgOut->setPagetitle( wfMsg( 'errorpagetitle' ) ); - $wgOut->addHTML( $mtext ); + $wgOut->addWikitext( $mtext ); wfProfileOut( $fname ); return; } + if ( $this->mNewRev->isCurrent() ) { + $wgOut->setArticleFlag( true ); + } # Check if user is allowed to look at this page. If not, bail out. # - if ( !( $this->mOldPage->userCanRead() ) ) { + if ( !( $this->mTitle->userCanRead() ) ) { $wgOut->loginToUse(); $wgOut->output(); wfProfileOut( $fname ); @@ -191,140 +262,387 @@ class DifferenceEngine { # $sk = $wgUser->getSkin(); - $uTLink = $sk->makeLinkObj( Title::makeTitleSafe( NS_USER_TALK, $this->mOldUser ), $wgLang->getNsText( NS_TALK ) ); - $userLink = $sk->makeLinkObj( Title::makeTitleSafe( NS_USER, $this->mOldUser ), $this->mOldUser ); - $contribs = $sk->makeKnownLinkObj( Title::makeTitle( NS_SPECIAL, 'Contributions' ), wfMsg( 'contribslink' ), - 'target=' . urlencode($this->mOldUser) ); - $nextlink = $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'nextdiff' ), 'diff=next&oldid='.$this->mNewid ); - $header = "
{$this->mOldtitle}
$userLink " . - "($uTLink | $contribs)
" . $this->mOldComment . - '
' . $nextlink. "
\n"; + $nextlink = $sk->makeKnownLinkObj( $this->mTitle, wfMsgHtml( 'nextdiff' ), 'diff=next&oldid='.$this->mNewid, '', '', 'id="differences-nextlink"' ); + $header = "
{$this->mOldtitle}
" . + $sk->revUserTools( $this->mNewRev ) . "
" . + $sk->revComment( $this->mNewRev ) . "
" . + $nextlink . "
\n"; $wgOut->addHTML( $header ); $wgOut->setSubtitle( wfMsg( 'difference' ) ); - $wgOut->setRobotpolicy( 'noindex,follow' ); + $wgOut->setRobotpolicy( 'noindex,nofollow' ); + wfProfileOut( $fname ); + } - # Show current revision - # - $wgOut->addHTML( "

{$this->mPagetitle}

\n" ); - $wgOut->addWikiText( $this->mNewtext ); + /** + * Get the diff text, send it to $wgOut + * Returns false if the diff could not be generated, otherwise returns true + */ + function showDiff( $otitle, $ntitle ) { + global $wgOut; + $diff = $this->getDiff( $otitle, $ntitle ); + if ( $diff === false ) { + $wgOut->addWikitext( wfMsg( 'missingarticle', "(fixme, bug)" ) ); + return false; + } else { + $wgOut->addHTML( $diff ); + return true; + } + } + + /** + * Get diff table, including header + * Note that the interface has changed, it's no longer static. + * Returns false on error + */ + function getDiff( $otitle, $ntitle ) { + $body = $this->getDiffBody(); + if ( $body === false ) { + return false; + } else { + $multi = $this->getMultiNotice(); + return $this->addHeader( $body, $otitle, $ntitle, $multi ); + } + } + + /** + * Get the diff table body, without header + * Results are cached + * Returns false on error + */ + function getDiffBody() { + global $wgMemc; + $fname = 'DifferenceEngine::getDiffBody'; + wfProfileIn( $fname ); + + // Cacheable? + $key = false; + if ( $this->mOldid && $this->mNewid ) { + // Try cache + $key = wfMemcKey( 'diff', 'oldid', $this->mOldid, 'newid', $this->mNewid ); + $difftext = $wgMemc->get( $key ); + if ( $difftext ) { + wfIncrStats( 'diff_cache_hit' ); + $difftext = $this->localiseLineNumbers( $difftext ); + $difftext .= "\n\n"; + wfProfileOut( $fname ); + return $difftext; + } + } + + #loadtext is permission safe, this just clears out the diff + if ( !$this->loadText() ) { + wfProfileOut( $fname ); + return false; + } else if ( $this->mOldRev && !$this->mOldRev->userCan(Revision::DELETED_TEXT) ) { + return ''; + } else if ( $this->mNewRev && !$this->mNewRev->userCan(Revision::DELETED_TEXT) ) { + return ''; + } + $difftext = $this->generateDiffBody( $this->mOldtext, $this->mNewtext ); + + // Save to cache for 7 days + if ( $key !== false && $difftext !== false ) { + wfIncrStats( 'diff_cache_miss' ); + $wgMemc->set( $key, $difftext, 7*86400 ); + } else { + wfIncrStats( 'diff_uncacheable' ); + } + // Replace line numbers with the text in the user's language + if ( $difftext !== false ) { + $difftext = $this->localiseLineNumbers( $difftext ); + } wfProfileOut( $fname ); + return $difftext; } - function showDiff( $otext, $ntext, $otitle, $ntitle ) - { - global $wgOut, $wgUseExternalDiffEngine; + /** + * Generate a diff, no caching + * $otext and $ntext must be already segmented + */ + function generateDiffBody( $otext, $ntext ) { + global $wgExternalDiffEngine, $wgContLang; + $fname = 'DifferenceEngine::generateDiffBody'; + + $otext = str_replace( "\r\n", "\n", $otext ); + $ntext = str_replace( "\r\n", "\n", $ntext ); + + if ( $wgExternalDiffEngine == 'wikidiff' ) { + # For historical reasons, external diff engine expects + # input text to be HTML-escaped already + $otext = htmlspecialchars ( $wgContLang->segmentForDiff( $otext ) ); + $ntext = htmlspecialchars ( $wgContLang->segmentForDiff( $ntext ) ); + if( !function_exists( 'wikidiff_do_diff' ) ) { + dl('php_wikidiff.so'); + } + return $wgContLang->unsegementForDiff( wikidiff_do_diff( $otext, $ntext, 2 ) ); + } + + if ( $wgExternalDiffEngine == 'wikidiff2' ) { + # Better external diff engine, the 2 may some day be dropped + # This one does the escaping and segmenting itself + if ( !function_exists( 'wikidiff2_do_diff' ) ) { + wfProfileIn( "$fname-dl" ); + @dl('php_wikidiff2.so'); + wfProfileOut( "$fname-dl" ); + } + if ( function_exists( 'wikidiff2_do_diff' ) ) { + wfProfileIn( 'wikidiff2_do_diff' ); + $text = wikidiff2_do_diff( $otext, $ntext, 2 ); + wfProfileOut( 'wikidiff2_do_diff' ); + return $text; + } + } + if ( $wgExternalDiffEngine !== false ) { + # Diff via the shell + global $wgTmpDirectory; + $tempName1 = tempnam( $wgTmpDirectory, 'diff_' ); + $tempName2 = tempnam( $wgTmpDirectory, 'diff_' ); + + $tempFile1 = fopen( $tempName1, "w" ); + if ( !$tempFile1 ) { + wfProfileOut( $fname ); + return false; + } + $tempFile2 = fopen( $tempName2, "w" ); + if ( !$tempFile2 ) { + wfProfileOut( $fname ); + return false; + } + fwrite( $tempFile1, $otext ); + fwrite( $tempFile2, $ntext ); + fclose( $tempFile1 ); + fclose( $tempFile2 ); + $cmd = wfEscapeShellArg( $wgExternalDiffEngine, $tempName1, $tempName2 ); + wfProfileIn( "$fname-shellexec" ); + $difftext = wfShellExec( $cmd ); + wfProfileOut( "$fname-shellexec" ); + unlink( $tempName1 ); + unlink( $tempName2 ); + return $difftext; + } + + # Native PHP diff + $ota = explode( "\n", $wgContLang->segmentForDiff( $otext ) ); + $nta = explode( "\n", $wgContLang->segmentForDiff( $ntext ) ); + $diffs = new Diff( $ota, $nta ); + $formatter = new TableDiffFormatter(); + return $wgContLang->unsegmentForDiff( $formatter->format( $diffs ) ); + } + + + /** + * Replace line numbers with the text in the user's language + */ + function localiseLineNumbers( $text ) { + return preg_replace_callback( '//', + array( &$this, 'localiseLineNumbersCb' ), $text ); + } + + function localiseLineNumbersCb( $matches ) { + global $wgLang; + return wfMsgExt( 'lineno', array('parseinline'), $wgLang->formatNum( $matches[1] ) ); + } + + + /** + * If there are revisions between the ones being compared, return a note saying so. + */ + function getMultiNotice() { + if ( !is_object($this->mOldRev) || !is_object($this->mNewRev) ) + return ''; + + if( !$this->mOldPage->equals( $this->mNewPage ) ) { + // Comparing two different pages? Count would be meaningless. + return ''; + } + + $oldid = $this->mOldRev->getId(); + $newid = $this->mNewRev->getId(); + if ( $oldid > $newid ) { + $tmp = $oldid; $oldid = $newid; $newid = $tmp; + } + + $n = $this->mTitle->countRevisionsBetween( $oldid, $newid ); + if ( !$n ) + return ''; + + return wfMsgExt( 'diff-multi', array( 'parseinline' ), $n ); + } - $wgOut->addHTML( " + + /** + * Add the header to a diff body + */ + function addHeader( $diff, $otitle, $ntitle, $multi = '' ) { + global $wgOut; + + if ( $this->mOldRev && $this->mOldRev->isDeleted(Revision::DELETED_TEXT) ) { + $otitle = ''.$otitle.''; + } + if ( $this->mNewRev && $this->mNewRev->isDeleted(Revision::DELETED_TEXT) ) { + $ntitle = ''.$ntitle.''; + } + $header = " - " ); + "; - if ( $wgUseExternalDiffEngine ) { - # For historical reasons, external diff engine expects - # input text to be HTML-escaped already - $otext = str_replace( "\r\n", "\n", htmlspecialchars ( $otext ) ); - $ntext = str_replace( "\r\n", "\n", htmlspecialchars ( $ntext ) ); - dl('php_wikidiff.so'); - $wgOut->addHTML( wikidiff_do_diff( $otext, $ntext, 2) ); + if ( $multi != '' ) + $header .= ""; + + return $header . $diff . "
{$otitle} {$ntitle}
{$multi}
"; + } + + /** + * Use specified text instead of loading from the database + */ + function setText( $oldText, $newText ) { + $this->mOldtext = $oldText; + $this->mNewtext = $newText; + $this->mTextLoaded = 2; + } + + /** + * Load revision metadata for the specified articles. If newid is 0, then compare + * the old article in oldid to the current article; if oldid is 0, then + * compare the current article to the immediately previous one (ignoring the + * value of newid). + * + * If oldid is false, leave the corresponding revision object set + * to false. This is impossible via ordinary user input, and is provided for + * API convenience. + */ + function loadRevisionData() { + global $wgLang; + if ( $this->mRevisionsLoaded ) { + return true; } else { - $ota = explode( "\n", str_replace( "\r\n", "\n", $otext ) ); - $nta = explode( "\n", str_replace( "\r\n", "\n", $ntext ) ); - $diffs = new Diff( $ota, $nta ); - $formatter = new TableDiffFormatter(); - $formatter->format( $diffs ); - } - $wgOut->addHTML( "\n" ); - } - - # Load the text of the articles to compare. If newid is 0, then compare - # the old article in oldid to the current article; if oldid is 0, then - # compare the current article to the immediately previous one (ignoring - # the value of newid). - # - function loadText() - { - global $wgTitle, $wgOut, $wgLang; - $fname = 'DifferenceEngine::loadText'; - - $dbr =& wfGetDB( DB_SLAVE ); - if ( 0 == $this->mNewid || 0 == $this->mOldid ) { - $wgOut->setArticleFlag( true ); - $newLink = $wgTitle->getLocalUrl(); - $this->mPagetitle = wfMsg( 'currentrev' ); - $this->mNewtitle = "{$this->mPagetitle}"; - $id = $wgTitle->getArticleID(); - - $s = $dbr->getArray( 'cur', array( 'cur_text', 'cur_user_text', 'cur_comment' ), - array( 'cur_id' => $id ), $fname ); - if ( $s === false ) { - wfDebug( "Unable to load cur_id $id\n" ); - return false; - } + // Whether it succeeds or fails, we don't want to try again + $this->mRevisionsLoaded = true; + } + + // Load the new revision object + if( $this->mNewid ) { + $this->mNewRev = Revision::newFromId( $this->mNewid ); + } else { + $this->mNewRev = Revision::newFromTitle( $this->mTitle ); + } + + if( is_null( $this->mNewRev ) ) { + return false; + } + + // Set assorted variables + $timestamp = $wgLang->timeanddate( $this->mNewRev->getTimestamp(), true ); + $this->mNewPage = $this->mNewRev->getTitle(); + if( $this->mNewRev->isCurrent() ) { + $newLink = $this->mNewPage->escapeLocalUrl(); + $this->mPagetitle = htmlspecialchars( wfMsg( 'currentrev' ) ); + $newEdit = $this->mNewPage->escapeLocalUrl( 'action=edit' ); + $newUndo = $this->mNewPage->escapeLocalUrl( 'action=edit&undo=' . $this->mNewid ); + + $this->mNewtitle = "{$this->mPagetitle} ($timestamp)" + . " (" . htmlspecialchars( wfMsg( 'editold' ) ) . ")" + . " (" . htmlspecialchars( wfMsg( 'editundo' ) ) . ")"; - $this->mNewPage = &$wgTitle; - $this->mNewtext = $s->cur_text; - $this->mNewUser = $s->cur_user_text; - $this->mNewComment = $s->cur_comment; } else { - $s = $dbr->getArray( 'old', array( 'old_namespace','old_title','old_timestamp', 'old_text', - 'old_flags','old_user_text','old_comment' ), array( 'old_id' => $this->mNewid ), $fname ); + $newLink = $this->mNewPage->escapeLocalUrl( 'oldid=' . $this->mNewid ); + $newEdit = $this->mNewPage->escapeLocalUrl( 'action=edit&oldid=' . $this->mNewid ); + $newUndo = $this->mNewPage->escapeLocalUrl( 'action=edit&undo=' . $this->mNewid ); + $this->mPagetitle = htmlspecialchars( wfMsg( 'revisionasof', $timestamp ) ); + + $this->mNewtitle = "{$this->mPagetitle}" + . " (" . htmlspecialchars( wfMsg( 'editold' ) ) . ")" + . " (" . htmlspecialchars( wfMsg( 'editundo' ) ) . ")"; + } - if ( $s === false ) { - wfDebug( "Unable to load old_id {$this->mNewid}\n" ); - return false; + // Load the old revision object + $this->mOldRev = false; + if( $this->mOldid ) { + $this->mOldRev = Revision::newFromId( $this->mOldid ); + } elseif ( $this->mOldid === 0 ) { + $rev = $this->mNewRev->getPrevious(); + if( $rev ) { + $this->mOldid = $rev->getId(); + $this->mOldRev = $rev; + } else { + // No previous revision; mark to show as first-version only. + $this->mOldid = false; + $this->mOldRev = false; } + }/* elseif ( $this->mOldid === false ) leave mOldRev false; */ - $this->mNewtext = Article::getRevisionText( $s ); - - $t = $wgLang->timeanddate( $s->old_timestamp, true ); - $this->mNewPage = Title::MakeTitle( $s->old_namespace, $s->old_title ); - $newLink = $wgTitle->getLocalUrl ('oldid=' . $this->mNewid); - $this->mPagetitle = wfMsg( 'revisionasof', $t ); - $this->mNewtitle = "{$this->mPagetitle}"; - $this->mNewUser = $s->old_user_text; - $this->mNewComment = $s->old_comment; - } - if ( 0 == $this->mOldid ) { - $s = $dbr->getArray( 'old', - array( 'old_namespace','old_title','old_timestamp','old_text', 'old_flags','old_user_text','old_comment' ), - array( /* WHERE */ - 'old_namespace' => $this->mNewPage->getNamespace(), - 'old_title' => $this->mNewPage->getDBkey() - ), $fname, array( 'ORDER BY' => 'inverse_timestamp', 'USE INDEX' => 'name_title_timestamp' ) - ); - if ( $s === false ) { - wfDebug( 'Unable to load ' . $this->mNewPage->getPrefixedDBkey . " from old\n" ); + if( is_null( $this->mOldRev ) ) { + return false; + } + + if ( $this->mOldRev ) { + $this->mOldPage = $this->mOldRev->getTitle(); + + $t = $wgLang->timeanddate( $this->mOldRev->getTimestamp(), true ); + $oldLink = $this->mOldPage->escapeLocalUrl( 'oldid=' . $this->mOldid ); + $oldEdit = $this->mOldPage->escapeLocalUrl( 'action=edit&oldid=' . $this->mOldid ); + $this->mOldtitle = "" . htmlspecialchars( wfMsg( 'revisionasof', $t ) ) + . " (" . htmlspecialchars( wfMsg( 'editold' ) ) . ")"; + } + + return true; + } + + /** + * Load the text of the revisions, as well as revision data. + */ + function loadText() { + if ( $this->mTextLoaded == 2 ) { + return true; + } else { + // Whether it succeeds or fails, we don't want to try again + $this->mTextLoaded = 2; + } + + if ( !$this->loadRevisionData() ) { + return false; + } + if ( $this->mOldRev ) { + // FIXME: permission tests + $this->mOldtext = $this->mOldRev->revText(); + if ( $this->mOldtext === false ) { return false; } - } else { - $s = $dbr->getArray( 'old', - array( 'old_namespace','old_title','old_timestamp','old_text','old_flags','old_user_text','old_comment'), - array( 'old_id' => $this->mOldid ), - $fname - ); - if ( $s === false ) { - wfDebug( "Unable to load old_id {$this->mOldid}\n" ); + } + if ( $this->mNewRev ) { + $this->mNewtext = $this->mNewRev->revText(); + if ( $this->mNewtext === false ) { return false; } } - $this->mOldPage = Title::MakeTitle( $s->old_namespace, $s->old_title ); - $this->mOldtext = Article::getRevisionText( $s ); - - $t = $wgLang->timeanddate( $s->old_timestamp, true ); - $oldLink = $this->mOldPage->getLocalUrl ('oldid=' . $this->mOldid); - $this->mOldtitle = "" . wfMsg( 'revisionasof', $t ) . ''; - $this->mOldUser = $s->old_user_text; - $this->mOldComment = $s->old_comment; + return true; + } + /** + * Load the text of the new revision, not the old one + */ + function loadNewText() { + if ( $this->mTextLoaded >= 1 ) { + return true; + } else { + $this->mTextLoaded = 1; + } + if ( !$this->loadRevisionData() ) { + return false; + } + $this->mNewtext = $this->mNewRev->getText(); return true; } + + } // A PHP diff engine for phpwiki. (Taken from phpwiki-1.3.3) @@ -337,7 +655,8 @@ define('USE_ASSERTS', function_exists('assert')); /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class _DiffOp { var $type; @@ -359,7 +678,8 @@ class _DiffOp { /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class _DiffOp_Copy extends _DiffOp { var $type = 'copy'; @@ -378,7 +698,8 @@ class _DiffOp_Copy extends _DiffOp { /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class _DiffOp_Delete extends _DiffOp { var $type = 'delete'; @@ -395,7 +716,8 @@ class _DiffOp_Delete extends _DiffOp { /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class _DiffOp_Add extends _DiffOp { var $type = 'add'; @@ -412,7 +734,8 @@ class _DiffOp_Add extends _DiffOp { /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class _DiffOp_Change extends _DiffOp { var $type = 'change'; @@ -445,13 +768,20 @@ class _DiffOp_Change extends _DiffOp { * closingly, some ideas (subdivision by NCHUNKS > 2, and some optimizations) * are my own. * - * @author Geoffrey T. Dairiki - * @package MediaWiki - * @access private + * Line length limits for robustness added by Tim Starling, 2005-08-31 + * + * @author Geoffrey T. Dairiki, Tim Starling + * @private + * @addtogroup DifferenceEngine */ class _DiffEngine { + const MAX_XREF_LENGTH = 10000; + function diff ($from_lines, $to_lines) { + $fname = '_DiffEngine::diff'; + wfProfileIn( $fname ); + $n_from = sizeof($from_lines); $n_to = sizeof($to_lines); @@ -464,32 +794,34 @@ class _DiffEngine // Skip leading common lines. for ($skip = 0; $skip < $n_from && $skip < $n_to; $skip++) { - if ($from_lines[$skip] != $to_lines[$skip]) + if ($from_lines[$skip] !== $to_lines[$skip]) break; $this->xchanged[$skip] = $this->ychanged[$skip] = false; } // Skip trailing common lines. $xi = $n_from; $yi = $n_to; for ($endskip = 0; --$xi > $skip && --$yi > $skip; $endskip++) { - if ($from_lines[$xi] != $to_lines[$yi]) + if ($from_lines[$xi] !== $to_lines[$yi]) break; $this->xchanged[$xi] = $this->ychanged[$yi] = false; } // Ignore lines which do not exist in both files. - for ($xi = $skip; $xi < $n_from - $endskip; $xi++) - $xhash[$from_lines[$xi]] = 1; + for ($xi = $skip; $xi < $n_from - $endskip; $xi++) { + $xhash[$this->_line_hash($from_lines[$xi])] = 1; + } + for ($yi = $skip; $yi < $n_to - $endskip; $yi++) { $line = $to_lines[$yi]; - if ( ($this->ychanged[$yi] = empty($xhash[$line])) ) + if ( ($this->ychanged[$yi] = empty($xhash[$this->_line_hash($line)])) ) continue; - $yhash[$line] = 1; + $yhash[$this->_line_hash($line)] = 1; $this->yv[] = $line; $this->yind[] = $yi; } for ($xi = $skip; $xi < $n_from - $endskip; $xi++) { $line = $from_lines[$xi]; - if ( ($this->xchanged[$xi] = empty($yhash[$line])) ) + if ( ($this->xchanged[$xi] = empty($yhash[$this->_line_hash($line)])) ) continue; $this->xv[] = $line; $this->xind[] = $xi; @@ -535,9 +867,21 @@ class _DiffEngine elseif ($add) $edits[] = new _DiffOp_Add($add); } + wfProfileOut( $fname ); return $edits; } + /** + * Returns the whole line if it's small enough, or the MD5 hash otherwise + */ + function _line_hash( $line ) { + if ( strlen( $line ) > self::MAX_XREF_LENGTH ) { + return md5( $line ); + } else { + return $line; + } + } + /* Divide the Largest Common Subsequence (LCS) of the sequences * [XOFF, XLIM) and [YOFF, YLIM) into NCHUNKS approximately equally @@ -556,102 +900,111 @@ class _DiffEngine * of the portions it is going to specify. */ function _diag ($xoff, $xlim, $yoff, $ylim, $nchunks) { - $flip = false; - - if ($xlim - $xoff > $ylim - $yoff) { - // Things seems faster (I'm not sure I understand why) - // when the shortest sequence in X. - $flip = true; - list ($xoff, $xlim, $yoff, $ylim) - = array( $yoff, $ylim, $xoff, $xlim); - } - - if ($flip) - for ($i = $ylim - 1; $i >= $yoff; $i--) - $ymatches[$this->xv[$i]][] = $i; - else - for ($i = $ylim - 1; $i >= $yoff; $i--) - $ymatches[$this->yv[$i]][] = $i; - - $this->lcs = 0; - $this->seq[0]= $yoff - 1; - $this->in_seq = array(); - $ymids[0] = array(); - - $numer = $xlim - $xoff + $nchunks - 1; - $x = $xoff; - for ($chunk = 0; $chunk < $nchunks; $chunk++) { - if ($chunk > 0) - for ($i = 0; $i <= $this->lcs; $i++) - $ymids[$i][$chunk-1] = $this->seq[$i]; - - $x1 = $xoff + (int)(($numer + ($xlim-$xoff)*$chunk) / $nchunks); - for ( ; $x < $x1; $x++) { + $fname = '_DiffEngine::_diag'; + wfProfileIn( $fname ); + $flip = false; + + if ($xlim - $xoff > $ylim - $yoff) { + // Things seems faster (I'm not sure I understand why) + // when the shortest sequence in X. + $flip = true; + list ($xoff, $xlim, $yoff, $ylim) + = array( $yoff, $ylim, $xoff, $xlim); + } + + if ($flip) + for ($i = $ylim - 1; $i >= $yoff; $i--) + $ymatches[$this->xv[$i]][] = $i; + else + for ($i = $ylim - 1; $i >= $yoff; $i--) + $ymatches[$this->yv[$i]][] = $i; + + $this->lcs = 0; + $this->seq[0]= $yoff - 1; + $this->in_seq = array(); + $ymids[0] = array(); + + $numer = $xlim - $xoff + $nchunks - 1; + $x = $xoff; + for ($chunk = 0; $chunk < $nchunks; $chunk++) { + wfProfileIn( "$fname-chunk" ); + if ($chunk > 0) + for ($i = 0; $i <= $this->lcs; $i++) + $ymids[$i][$chunk-1] = $this->seq[$i]; + + $x1 = $xoff + (int)(($numer + ($xlim-$xoff)*$chunk) / $nchunks); + for ( ; $x < $x1; $x++) { $line = $flip ? $this->yv[$x] : $this->xv[$x]; - if (empty($ymatches[$line])) - continue; - $matches = $ymatches[$line]; + if (empty($ymatches[$line])) + continue; + $matches = $ymatches[$line]; reset($matches); - while (list ($junk, $y) = each($matches)) - if (empty($this->in_seq[$y])) { - $k = $this->_lcs_pos($y); - USE_ASSERTS && assert($k > 0); - $ymids[$k] = $ymids[$k-1]; - break; + while (list ($junk, $y) = each($matches)) + if (empty($this->in_seq[$y])) { + $k = $this->_lcs_pos($y); + USE_ASSERTS && assert($k > 0); + $ymids[$k] = $ymids[$k-1]; + break; } - while (list ($junk, $y) = each($matches)) { - if ($y > $this->seq[$k-1]) { - USE_ASSERTS && assert($y < $this->seq[$k]); - // Optimization: this is a common case: - // next match is just replacing previous match. - $this->in_seq[$this->seq[$k]] = false; - $this->seq[$k] = $y; - $this->in_seq[$y] = 1; - } - else if (empty($this->in_seq[$y])) { - $k = $this->_lcs_pos($y); - USE_ASSERTS && assert($k > 0); - $ymids[$k] = $ymids[$k-1]; + while (list ( /* $junk */, $y) = each($matches)) { + if ($y > $this->seq[$k-1]) { + USE_ASSERTS && assert($y < $this->seq[$k]); + // Optimization: this is a common case: + // next match is just replacing previous match. + $this->in_seq[$this->seq[$k]] = false; + $this->seq[$k] = $y; + $this->in_seq[$y] = 1; + } else if (empty($this->in_seq[$y])) { + $k = $this->_lcs_pos($y); + USE_ASSERTS && assert($k > 0); + $ymids[$k] = $ymids[$k-1]; } } } + wfProfileOut( "$fname-chunk" ); } - $seps[] = $flip ? array($yoff, $xoff) : array($xoff, $yoff); - $ymid = $ymids[$this->lcs]; - for ($n = 0; $n < $nchunks - 1; $n++) { - $x1 = $xoff + (int)(($numer + ($xlim - $xoff) * $n) / $nchunks); - $y1 = $ymid[$n] + 1; - $seps[] = $flip ? array($y1, $x1) : array($x1, $y1); + $seps[] = $flip ? array($yoff, $xoff) : array($xoff, $yoff); + $ymid = $ymids[$this->lcs]; + for ($n = 0; $n < $nchunks - 1; $n++) { + $x1 = $xoff + (int)(($numer + ($xlim - $xoff) * $n) / $nchunks); + $y1 = $ymid[$n] + 1; + $seps[] = $flip ? array($y1, $x1) : array($x1, $y1); } - $seps[] = $flip ? array($ylim, $xlim) : array($xlim, $ylim); + $seps[] = $flip ? array($ylim, $xlim) : array($xlim, $ylim); - return array($this->lcs, $seps); + wfProfileOut( $fname ); + return array($this->lcs, $seps); } function _lcs_pos ($ypos) { - $end = $this->lcs; - if ($end == 0 || $ypos > $this->seq[$end]) { - $this->seq[++$this->lcs] = $ypos; - $this->in_seq[$ypos] = 1; - return $this->lcs; + $fname = '_DiffEngine::_lcs_pos'; + wfProfileIn( $fname ); + + $end = $this->lcs; + if ($end == 0 || $ypos > $this->seq[$end]) { + $this->seq[++$this->lcs] = $ypos; + $this->in_seq[$ypos] = 1; + wfProfileOut( $fname ); + return $this->lcs; } - $beg = 1; - while ($beg < $end) { - $mid = (int)(($beg + $end) / 2); - if ( $ypos > $this->seq[$mid] ) - $beg = $mid + 1; - else - $end = $mid; + $beg = 1; + while ($beg < $end) { + $mid = (int)(($beg + $end) / 2); + if ( $ypos > $this->seq[$mid] ) + $beg = $mid + 1; + else + $end = $mid; } - USE_ASSERTS && assert($ypos != $this->seq[$end]); + USE_ASSERTS && assert($ypos != $this->seq[$end]); - $this->in_seq[$this->seq[$end]] = false; - $this->seq[$end] = $ypos; - $this->in_seq[$ypos] = 1; - return $end; + $this->in_seq[$this->seq[$end]] = false; + $this->seq[$end] = $ypos; + $this->in_seq[$ypos] = 1; + wfProfileOut( $fname ); + return $end; } /* Find LCS of two sequences. @@ -666,48 +1019,51 @@ class _DiffEngine * All line numbers are origin-0 and discarded lines are not counted. */ function _compareseq ($xoff, $xlim, $yoff, $ylim) { - // Slide down the bottom initial diagonal. - while ($xoff < $xlim && $yoff < $ylim + $fname = '_DiffEngine::_compareseq'; + wfProfileIn( $fname ); + + // Slide down the bottom initial diagonal. + while ($xoff < $xlim && $yoff < $ylim && $this->xv[$xoff] == $this->yv[$yoff]) { - ++$xoff; - ++$yoff; + ++$xoff; + ++$yoff; } - // Slide up the top initial diagonal. - while ($xlim > $xoff && $ylim > $yoff + // Slide up the top initial diagonal. + while ($xlim > $xoff && $ylim > $yoff && $this->xv[$xlim - 1] == $this->yv[$ylim - 1]) { - --$xlim; - --$ylim; + --$xlim; + --$ylim; } - if ($xoff == $xlim || $yoff == $ylim) - $lcs = 0; - else { - // This is ad hoc but seems to work well. - //$nchunks = sqrt(min($xlim - $xoff, $ylim - $yoff) / 2.5); - //$nchunks = max(2,min(8,(int)$nchunks)); - $nchunks = min(7, $xlim - $xoff, $ylim - $yoff) + 1; - list ($lcs, $seps) - = $this->_diag($xoff,$xlim,$yoff, $ylim,$nchunks); - } - - if ($lcs == 0) { - // X and Y sequences have no common subsequence: - // mark all changed. - while ($yoff < $ylim) - $this->ychanged[$this->yind[$yoff++]] = 1; - while ($xoff < $xlim) - $this->xchanged[$this->xind[$xoff++]] = 1; - } - else { - // Use the partitions to split this problem into subproblems. - reset($seps); - $pt1 = $seps[0]; - while ($pt2 = next($seps)) { - $this->_compareseq ($pt1[0], $pt2[0], $pt1[1], $pt2[1]); - $pt1 = $pt2; + if ($xoff == $xlim || $yoff == $ylim) + $lcs = 0; + else { + // This is ad hoc but seems to work well. + //$nchunks = sqrt(min($xlim - $xoff, $ylim - $yoff) / 2.5); + //$nchunks = max(2,min(8,(int)$nchunks)); + $nchunks = min(7, $xlim - $xoff, $ylim - $yoff) + 1; + list ($lcs, $seps) + = $this->_diag($xoff,$xlim,$yoff, $ylim,$nchunks); + } + + if ($lcs == 0) { + // X and Y sequences have no common subsequence: + // mark all changed. + while ($yoff < $ylim) + $this->ychanged[$this->yind[$yoff++]] = 1; + while ($xoff < $xlim) + $this->xchanged[$this->xind[$xoff++]] = 1; + } else { + // Use the partitions to split this problem into subproblems. + reset($seps); + $pt1 = $seps[0]; + while ($pt2 = next($seps)) { + $this->_compareseq ($pt1[0], $pt2[0], $pt1[1], $pt2[1]); + $pt1 = $pt2; } } + wfProfileOut( $fname ); } /* Adjust inserts/deletes of identical lines to join changes @@ -723,117 +1079,121 @@ class _DiffEngine * This is extracted verbatim from analyze.c (GNU diffutils-2.7). */ function _shift_boundaries ($lines, &$changed, $other_changed) { - $i = 0; - $j = 0; - - USE_ASSERTS && assert('sizeof($lines) == sizeof($changed)'); - $len = sizeof($lines); - $other_len = sizeof($other_changed); - - while (1) { - /* - * Scan forwards to find beginning of another run of changes. - * Also keep track of the corresponding point in the other file. - * - * Throughout this code, $i and $j are adjusted together so that - * the first $i elements of $changed and the first $j elements - * of $other_changed both contain the same number of zeros - * (unchanged lines). - * Furthermore, $j is always kept so that $j == $other_len or - * $other_changed[$j] == false. - */ - while ($j < $other_len && $other_changed[$j]) - $j++; - - while ($i < $len && ! $changed[$i]) { - USE_ASSERTS && assert('$j < $other_len && ! $other_changed[$j]'); - $i++; $j++; - while ($j < $other_len && $other_changed[$j]) - $j++; + $fname = '_DiffEngine::_shift_boundaries'; + wfProfileIn( $fname ); + $i = 0; + $j = 0; + + USE_ASSERTS && assert('sizeof($lines) == sizeof($changed)'); + $len = sizeof($lines); + $other_len = sizeof($other_changed); + + while (1) { + /* + * Scan forwards to find beginning of another run of changes. + * Also keep track of the corresponding point in the other file. + * + * Throughout this code, $i and $j are adjusted together so that + * the first $i elements of $changed and the first $j elements + * of $other_changed both contain the same number of zeros + * (unchanged lines). + * Furthermore, $j is always kept so that $j == $other_len or + * $other_changed[$j] == false. + */ + while ($j < $other_len && $other_changed[$j]) + $j++; + + while ($i < $len && ! $changed[$i]) { + USE_ASSERTS && assert('$j < $other_len && ! $other_changed[$j]'); + $i++; $j++; + while ($j < $other_len && $other_changed[$j]) + $j++; } - if ($i == $len) - break; - - $start = $i; - - // Find the end of this run of changes. - while (++$i < $len && $changed[$i]) - continue; - - do { - /* - * Record the length of this run of changes, so that - * we can later determine whether the run has grown. - */ - $runlength = $i - $start; - - /* - * Move the changed region back, so long as the - * previous unchanged line matches the last changed one. - * This merges with previous changed regions. - */ - while ($start > 0 && $lines[$start - 1] == $lines[$i - 1]) { - $changed[--$start] = 1; - $changed[--$i] = false; - while ($start > 0 && $changed[$start - 1]) - $start--; - USE_ASSERTS && assert('$j > 0'); - while ($other_changed[--$j]) - continue; - USE_ASSERTS && assert('$j >= 0 && !$other_changed[$j]'); + if ($i == $len) + break; + + $start = $i; + + // Find the end of this run of changes. + while (++$i < $len && $changed[$i]) + continue; + + do { + /* + * Record the length of this run of changes, so that + * we can later determine whether the run has grown. + */ + $runlength = $i - $start; + + /* + * Move the changed region back, so long as the + * previous unchanged line matches the last changed one. + * This merges with previous changed regions. + */ + while ($start > 0 && $lines[$start - 1] == $lines[$i - 1]) { + $changed[--$start] = 1; + $changed[--$i] = false; + while ($start > 0 && $changed[$start - 1]) + $start--; + USE_ASSERTS && assert('$j > 0'); + while ($other_changed[--$j]) + continue; + USE_ASSERTS && assert('$j >= 0 && !$other_changed[$j]'); } - /* - * Set CORRESPONDING to the end of the changed run, at the last - * point where it corresponds to a changed run in the other file. - * CORRESPONDING == LEN means no such point has been found. - */ - $corresponding = $j < $other_len ? $i : $len; - - /* - * Move the changed region forward, so long as the - * first changed line matches the following unchanged one. - * This merges with following changed regions. - * Do this second, so that if there are no merges, - * the changed region is moved forward as far as possible. - */ - while ($i < $len && $lines[$start] == $lines[$i]) { - $changed[$start++] = false; - $changed[$i++] = 1; - while ($i < $len && $changed[$i]) - $i++; - - USE_ASSERTS && assert('$j < $other_len && ! $other_changed[$j]'); - $j++; - if ($j < $other_len && $other_changed[$j]) { - $corresponding = $i; - while ($j < $other_len && $other_changed[$j]) - $j++; + /* + * Set CORRESPONDING to the end of the changed run, at the last + * point where it corresponds to a changed run in the other file. + * CORRESPONDING == LEN means no such point has been found. + */ + $corresponding = $j < $other_len ? $i : $len; + + /* + * Move the changed region forward, so long as the + * first changed line matches the following unchanged one. + * This merges with following changed regions. + * Do this second, so that if there are no merges, + * the changed region is moved forward as far as possible. + */ + while ($i < $len && $lines[$start] == $lines[$i]) { + $changed[$start++] = false; + $changed[$i++] = 1; + while ($i < $len && $changed[$i]) + $i++; + + USE_ASSERTS && assert('$j < $other_len && ! $other_changed[$j]'); + $j++; + if ($j < $other_len && $other_changed[$j]) { + $corresponding = $i; + while ($j < $other_len && $other_changed[$j]) + $j++; } } } while ($runlength != $i - $start); - /* - * If possible, move the fully-merged run of changes - * back to a corresponding run in the other file. - */ - while ($corresponding < $i) { - $changed[--$start] = 1; - $changed[--$i] = 0; - USE_ASSERTS && assert('$j > 0'); - while ($other_changed[--$j]) - continue; - USE_ASSERTS && assert('$j >= 0 && !$other_changed[$j]'); + /* + * If possible, move the fully-merged run of changes + * back to a corresponding run in the other file. + */ + while ($corresponding < $i) { + $changed[--$start] = 1; + $changed[--$i] = 0; + USE_ASSERTS && assert('$j > 0'); + while ($other_changed[--$j]) + continue; + USE_ASSERTS && assert('$j >= 0 && !$other_changed[$j]'); } } + wfProfileOut( $fname ); } } /** * Class representing a 'diff' between two sequences of strings. * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class Diff { @@ -864,12 +1224,12 @@ class Diff * original diff. */ function reverse () { - $rev = $this; + $rev = $this; $rev->edits = array(); foreach ($this->edits as $edit) { $rev->edits[] = $edit->reverse(); } - return $rev; + return $rev; } /** @@ -893,12 +1253,12 @@ class Diff * @return int The length of the LCS. */ function lcs () { - $lcs = 0; + $lcs = 0; foreach ($this->edits as $edit) { if ($edit->type == 'copy') $lcs += sizeof($edit->orig); } - return $lcs; + return $lcs; } /** @@ -943,6 +1303,8 @@ class Diff * This is here only for debugging purposes. */ function _check ($from_lines, $to_lines) { + $fname = 'Diff::_check'; + wfProfileIn( $fname ); if (serialize($from_lines) != serialize($this->orig())) trigger_error("Reconstructed original doesn't match", E_USER_ERROR); if (serialize($to_lines) != serialize($this->closing())) @@ -964,13 +1326,15 @@ class Diff $lcs = $this->lcs(); trigger_error('Diff okay: LCS = '.$lcs, E_USER_NOTICE); + wfProfileOut( $fname ); } } /** * FIXME: bad name. * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class MappedDiff extends Diff { @@ -999,6 +1363,8 @@ class MappedDiff extends Diff */ function MappedDiff($from_lines, $to_lines, $mapped_from_lines, $mapped_to_lines) { + $fname = 'MappedDiff::MappedDiff'; + wfProfileIn( $fname ); assert(sizeof($from_lines) == sizeof($mapped_from_lines)); assert(sizeof($to_lines) == sizeof($mapped_to_lines)); @@ -1019,6 +1385,7 @@ class MappedDiff extends Diff $yi += sizeof($closing); } } + wfProfileOut( $fname ); } } @@ -1029,7 +1396,8 @@ class MappedDiff extends Diff * It is intended that this class be customized via inheritance, * to obtain fancier outputs. * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class DiffFormatter { @@ -1056,6 +1424,8 @@ class DiffFormatter * @return string The formatted output. */ function format($diff) { + $fname = 'DiffFormatter::format'; + wfProfileIn( $fname ); $xi = $yi = 1; $block = false; @@ -1108,10 +1478,14 @@ class DiffFormatter $y0, $yi - $y0, $block); - return $this->_end_diff(); + $end = $this->_end_diff(); + wfProfileOut( $fname ); + return $end; } function _block($xbeg, $xlen, $ybeg, $ylen, &$edits) { + $fname = 'DiffFormatter::_block'; + wfProfileIn( $fname ); $this->_start_block($this->_block_header($xbeg, $xlen, $ybeg, $ylen)); foreach ($edits as $edit) { if ($edit->type == 'copy') @@ -1126,6 +1500,7 @@ class DiffFormatter trigger_error('Unknown edit type', E_USER_ERROR); } $this->_end_block(); + wfProfileOut( $fname ); } function _start_diff() { @@ -1187,7 +1562,8 @@ define('NBSP', ' '); // iso-8859-x non-breaking space. /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class _HWLDF_WordAccumulator { function _HWLDF_WordAccumulator () { @@ -1199,9 +1575,12 @@ class _HWLDF_WordAccumulator { function _flushGroup ($new_tag) { if ($this->_group !== '') { - if ($this->_tag == 'mark') - $this->_line .= '' . - htmlspecialchars ( $this->_group ) . ''; + if ($this->_tag == 'ins') + $this->_line .= '' . + htmlspecialchars ( $this->_group ) . ''; + elseif ($this->_tag == 'del') + $this->_line .= '' . + htmlspecialchars ( $this->_group ) . ''; else $this->_line .= htmlspecialchars ( $this->_group ); } @@ -1244,57 +1623,96 @@ class _HWLDF_WordAccumulator { /** * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class WordLevelDiff extends MappedDiff { + const MAX_LINE_LENGTH = 10000; + function WordLevelDiff ($orig_lines, $closing_lines) { + $fname = 'WordLevelDiff::WordLevelDiff'; + wfProfileIn( $fname ); + list ($orig_words, $orig_stripped) = $this->_split($orig_lines); list ($closing_words, $closing_stripped) = $this->_split($closing_lines); - $this->MappedDiff($orig_words, $closing_words, $orig_stripped, $closing_stripped); + wfProfileOut( $fname ); } function _split($lines) { - if (!preg_match_all('/ ( [^\S\n]+ | [0-9_A-Za-z\x80-\xff]+ | . ) (?: (?!< \n) [^\S\n])? /xs', - implode("\n", $lines), - $m)) { - return array(array(''), array('')); + $fname = 'WordLevelDiff::_split'; + wfProfileIn( $fname ); + + $words = array(); + $stripped = array(); + $first = true; + foreach ( $lines as $line ) { + # If the line is too long, just pretend the entire line is one big word + # This prevents resource exhaustion problems + if ( $first ) { + $first = false; + } else { + $words[] = "\n"; + $stripped[] = "\n"; + } + if ( strlen( $line ) > self::MAX_LINE_LENGTH ) { + $words[] = $line; + $stripped[] = $line; + } else { + $m = array(); + if (preg_match_all('/ ( [^\S\n]+ | [0-9_A-Za-z\x80-\xff]+ | . ) (?: (?!< \n) [^\S\n])? /xs', + $line, $m)) + { + $words = array_merge( $words, $m[0] ); + $stripped = array_merge( $stripped, $m[1] ); + } + } } - return array($m[0], $m[1]); + wfProfileOut( $fname ); + return array($words, $stripped); } function orig () { + $fname = 'WordLevelDiff::orig'; + wfProfileIn( $fname ); $orig = new _HWLDF_WordAccumulator; foreach ($this->edits as $edit) { if ($edit->type == 'copy') $orig->addWords($edit->orig); elseif ($edit->orig) - $orig->addWords($edit->orig, 'mark'); + $orig->addWords($edit->orig, 'del'); } - return $orig->getLines(); + $lines = $orig->getLines(); + wfProfileOut( $fname ); + return $lines; } function closing () { + $fname = 'WordLevelDiff::closing'; + wfProfileIn( $fname ); $closing = new _HWLDF_WordAccumulator; foreach ($this->edits as $edit) { if ($edit->type == 'copy') $closing->addWords($edit->closing); elseif ($edit->closing) - $closing->addWords($edit->closing, 'mark'); + $closing->addWords($edit->closing, 'ins'); } - return $closing->getLines(); + $lines = $closing->getLines(); + wfProfileOut( $fname ); + return $lines; } } /** * Wikipedia Table style diff formatter. * @todo document - * @package MediaWiki + * @private + * @addtogroup DifferenceEngine */ class TableDiffFormatter extends DiffFormatter { @@ -1304,17 +1722,13 @@ class TableDiffFormatter extends DiffFormatter } function _block_header( $xbeg, $xlen, $ybeg, $ylen ) { - $l1 = wfMsg( 'lineno', $xbeg ); - $l2 = wfMsg( 'lineno', $ybeg ); - - $r = ''.$l1."\n" . - ''.$l2."\n"; + $r = '\n" . + '\n"; return $r; } function _start_block( $header ) { - global $wgOut; - $wgOut->addHTML( $header ); + echo $header; } function _end_block() { @@ -1343,32 +1757,31 @@ class TableDiffFormatter extends DiffFormatter } function _added( $lines ) { - global $wgOut; foreach ($lines as $line) { - $wgOut->addHTML( '' . $this->emptyLine() . - $this->addedLine( htmlspecialchars ( $line ) ) . "\n" ); + echo '' . $this->emptyLine() . + $this->addedLine( htmlspecialchars ( $line ) ) . "\n"; } } function _deleted($lines) { - global $wgOut; foreach ($lines as $line) { - $wgOut->addHTML( '' . $this->deletedLine( htmlspecialchars ( $line ) ) . - $this->emptyLine() . "\n" ); + echo '' . $this->deletedLine( htmlspecialchars ( $line ) ) . + $this->emptyLine() . "\n"; } } function _context( $lines ) { - global $wgOut; foreach ($lines as $line) { - $wgOut->addHTML( '' . + echo '' . $this->contextLine( htmlspecialchars ( $line ) ) . - $this->contextLine( htmlspecialchars ( $line ) ) . "\n" ); + $this->contextLine( htmlspecialchars ( $line ) ) . "\n"; } } function _changed( $orig, $closing ) { - global $wgOut; + $fname = 'TableDiffFormatter::_changed'; + wfProfileIn( $fname ); + $diff = new WordLevelDiff( $orig, $closing ); $del = $diff->orig(); $add = $diff->closing(); @@ -1378,13 +1791,14 @@ class TableDiffFormatter extends DiffFormatter while ( $line = array_shift( $del ) ) { $aline = array_shift( $add ); - $wgOut->addHTML( '' . $this->deletedLine( $line ) . - $this->addedLine( $aline ) . "\n" ); + echo '' . $this->deletedLine( $line ) . + $this->addedLine( $aline ) . "\n"; } foreach ($add as $line) { # If any leftovers - $wgOut->addHTML( '' . $this->emptyLine() . - $this->addedLine( $line ) . "\n" ); + echo '' . $this->emptyLine() . + $this->addedLine( $line ) . "\n"; } + wfProfileOut( $fname ); } }