<?php
-# See diff.doc
+/**
+ * See diff.doc
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
+
+/** */
+require_once( 'Revision.php' );
+/**
+ * @todo document
+ * @access public
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class DifferenceEngine {
/* private */ var $mOldid, $mNewid;
- /* private */ var $mOldtitle, $mNewtitle;
+ /* private */ var $mOldtitle, $mNewtitle, $mPagetitle;
/* private */ var $mOldtext, $mNewtext;
/* private */ var $mOldUser, $mNewUser;
/* private */ var $mOldComment, $mNewComment;
function DifferenceEngine( $old, $new, $rcid = 0 )
{
- $this->mOldid = $old;
- $this->mNewid = $new;
+ global $wgTitle;
+ if ( 'prev' == $new ) {
+ # Show diff between revision $old and the previous one.
+ # Get previous one from DB.
+ #
+ $this->mNewid = intval($old);
+
+ $this->mOldid = $wgTitle->getPreviousRevisionID( $this->mNewid );
+
+ } elseif ( 'next' == $new ) {
+
+ # Show diff between revision $old and the previous one.
+ # Get previous one from DB.
+ #
+ $this->mOldid = intval($old);
+ $this->mNewid = $wgTitle->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".
+ $this->mNewid = 0;
+ }
+
+ } 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, $wgLang, $wgOnlySysopsCanPatrol, $wgUseRCPatrol;
- $fname = "DifferenceEngine::showDiffPage";
+ global $wgUser, $wgTitle, $wgOut, $wgContLang, $wgOnlySysopsCanPatrol, $wgUseRCPatrol;
+ $fname = 'DifferenceEngine::showDiffPage';
wfProfileIn( $fname );
+ # 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();
+ wfProfileOut( $fname );
+ return;
+ }
+
$t = $wgTitle->getPrefixedText() . " (Diff: {$this->mOldid}, " .
"{$this->mNewid})";
- $mtext = wfMsg( "missingarticle", $t );
+ $mtext = wfMsg( 'missingarticle', $t );
$wgOut->setArticleFlag( false );
if ( ! $this->loadText() ) {
- $wgOut->setPagetitle( wfMsg( "errorpagetitle" ) );
+ $wgOut->setPagetitle( wfMsg( 'errorpagetitle' ) );
$wgOut->addHTML( $mtext );
wfProfileOut( $fname );
return;
if( $oldTitle == $newTitle ) {
$wgOut->setPageTitle( $newTitle );
} else {
- $wgOut->setPageTitle( $oldTitle . ", " . $newTitle );
+ $wgOut->setPageTitle( $oldTitle . ', ' . $newTitle );
}
- $wgOut->setSubtitle( wfMsg( "difference" ) );
- $wgOut->setRobotpolicy( "noindex,follow" );
+ $wgOut->setSubtitle( wfMsg( 'difference' ) );
+ $wgOut->setRobotpolicy( 'noindex,follow' );
if ( !( $this->mOldPage->userCanRead() && $this->mNewPage->userCanRead() ) ) {
$wgOut->loginToUse();
}
$sk = $wgUser->getSkin();
- $talk = $wgLang->getNsText( NS_TALK );
- $contribs = wfMsg( "contribslink" );
+ $talk = $wgContLang->getNsText( NS_TALK );
+ $contribs = wfMsg( 'contribslink' );
$this->mOldComment = $sk->formatComment($this->mOldComment);
$this->mNewComment = $sk->formatComment($this->mNewComment);
- $oldUserLink = $sk->makeLinkObj( Title::makeTitle( NS_USER, $this->mOldUser ), $this->mOldUser );
- $newUserLink = $sk->makeLinkObj( Title::makeTitle( NS_USER, $this->mNewUser ), $this->mNewUser );
- $oldUTLink = $sk->makeLinkObj( Title::makeTitle( NS_USER_TALK, $this->mOldUser ), $talk );
- $newUTLink = $sk->makeLinkObj( Title::makeTitle( 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 = " <strong>[" . $sk->makeKnownLinkObj( $wgTitle, wfMsg( "rollbacklink" ),
- "action=rollback&from=" . urlencode($this->mNewUser) ) . "]</strong>";
+ $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->isAllowed('rollback') ) {
+ $rollback = ' <strong>[' . $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'rollbacklink' ),
+ 'action=rollback&from=' . urlencode($this->mNewUser) ) . ']</strong>';
} else {
- $rollback = "";
+ $rollback = '';
}
if ( $wgUseRCPatrol && $this->mRcidMarkPatrolled != 0 && $wgUser->getID() != 0 &&
- ( $wgUser->isSysop() || !$wgOnlySysopsCanPatrol ) )
+ ( $wgUser->isAllowed('rollback') || !$wgOnlySysopsCanPatrol ) )
{
- $patrol = " [" . $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'markaspatrolleddiff' ),
- "action=markpatrolled&rcid={$this->mRcidMarkPatrolled}" ) . "]";
+ $patrol = ' [' . $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'markaspatrolleddiff' ),
+ "action=markpatrolled&rcid={$this->mRcidMarkPatrolled}" ) . ']';
+ } else {
+ $patrol = '';
+ }
+
+ $prevlink = $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'previousdiff' ), 'diff=prev&oldid='.$this->mOldid );
+ if ( $this->mNewid == 0 ) {
+ $nextlink = '';
} else {
- $patrol = "";
+ $nextlink = $sk->makeKnownLinkObj( $wgTitle, wfMsg( 'nextdiff' ), 'diff=next&oldid='.$this->mNewid );
}
$oldHeader = "<strong>{$this->mOldtitle}</strong><br />$oldUserLink " .
- "($oldUTLink | $oldContribs)<br />" . $this->mOldComment;
+ "($oldUTLink | $oldContribs)<br />" . $this->mOldComment .
+ '<br />' . $prevlink;
$newHeader = "<strong>{$this->mNewtitle}</strong><br />$newUserLink " .
- "($newUTLink | $newContribs) $rollback<br />" . $this->mNewComment . $patrol;
+ "($newUTLink | $newContribs) $rollback<br />" . $this->mNewComment .
+ '<br />' . $nextlink . $patrol;
DifferenceEngine::showDiff( $this->mOldtext, $this->mNewtext,
$oldHeader, $newHeader );
- $wgOut->addHTML( "<hr /><h2>{$this->mNewtitle}</h2>\n" );
+ $wgOut->addHTML( "<hr /><h2>{$this->mPagetitle}</h2>\n" );
$wgOut->addWikiText( $this->mNewtext );
wfProfileOut( $fname );
}
- function showDiff( $otext, $ntext, $otitle, $ntitle )
+ # Show the first revision of an article. Uses normal diff headers in contrast to normal
+ # "old revision" display style.
+ #
+ function showFirstRevision()
{
- global $wgOut, $wgUseExternalDiffEngine;
+ global $wgOut, $wgTitle, $wgUser, $wgLang;
+
+ $fname = 'DifferenceEngine::showFirstRevision';
+ wfProfileIn( $fname );
- $otext = str_replace( "\r\n", "\n", htmlspecialchars( $otext ) );
- $ntext = str_replace( "\r\n", "\n", htmlspecialchars( $ntext ) );
+ $this->mOldid = $this->mNewid; # hack to make loadText() work.
- $wgOut->addHTML( "<table border='0' width='98%'
-cellpadding='0' cellspacing='4px' class='diff'><tr>
-<td colspan='2' width='50%' align='center' class='diff-otitle'>
-{$otitle}</td>
-<td colspan='2' width='50%' align='center' class='diff-ntitle'>
-{$ntitle}</td>
-</tr>\n" );
+ # Get article text from the DB
+ #
+ if ( ! $this->loadText() ) {
+ $t = $wgTitle->getPrefixedText() . " (Diff: {$this->mOldid}, " .
+ "{$this->mNewid})";
+ $mtext = wfMsg( 'missingarticle', $t );
+ $wgOut->setPagetitle( wfMsg( 'errorpagetitle' ) );
+ $wgOut->addHTML( $mtext );
+ wfProfileOut( $fname );
+ return;
+ }
+
+ # Check if user is allowed to look at this page. If not, bail out.
+ #
+ if ( !( $this->mOldPage->userCanRead() ) ) {
+ $wgOut->loginToUse();
+ $wgOut->output();
+ wfProfileOut( $fname );
+ exit;
+ }
+
+ # Prepare the header box
+ #
+ $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 );
+ $header = "<div class=\"firstrevisionheader\" style=\"text-align: center\"><strong>{$this->mOldtitle}</strong><br />$userLink " .
+ "($uTLink | $contribs)<br />" . $this->mOldComment .
+ '<br />' . $nextlink. "</div>\n";
+
+ $wgOut->addHTML( $header );
+
+ $wgOut->setSubtitle( wfMsg( 'difference' ) );
+ $wgOut->setRobotpolicy( 'noindex,follow' );
+
+
+ # Show current revision
+ #
+ $wgOut->addHTML( "<hr /><h2>{$this->mPagetitle}</h2>\n" );
+ $wgOut->addWikiText( $this->mNewtext );
+
+ wfProfileOut( $fname );
+ }
+
+ function showDiff( $otext, $ntext, $otitle, $ntitle )
+ {
+ global $wgOut;
+ $wgOut->addHTML( DifferenceEngine::getDiff( $otext, $ntext, $otitle, $ntitle ) );
+ }
+
+ function getDiff( $otext, $ntext, $otitle, $ntitle ) {
+ global $wgUseExternalDiffEngine, $wgContLang;
+ $out = "
+ <table border='0' width='98%' cellpadding='0' cellspacing='4' class='diff'>
+ <tr>
+ <td colspan='2' width='50%' align='center' class='diff-otitle'>{$otitle}</td>
+ <td colspan='2' width='50%' align='center' class='diff-ntitle'>{$ntitle}</td>
+ </tr>
+ ";
+ $otext = $wgContLang->segmentForDiff($otext);
+ $ntext = $wgContLang->segmentForDiff($ntext);
+ $difftext='';
if ( $wgUseExternalDiffEngine ) {
- dl("php_wikidiff.so");
- $wgOut->addHTML( wikidiff_do_diff( $otext, $ntext, 2) );
+ # 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 ) );
+ if( !function_exists( 'wikidiff_do_diff' ) ) {
+ dl('php_wikidiff.so');
+ }
+ $difftext = wikidiff_do_diff( $otext, $ntext, 2 );
} else {
- $ota = explode( "\n", $otext);
- $nta = explode( "\n", $ntext);
- $diffs = new Diff( $ota, $nta );
- $formatter = new TableDiffFormatter();
- $formatter->format( $diffs );
+ $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();
+ $difftext = $formatter->format( $diffs );
}
- $wgOut->addHTML( "</table>\n" );
+ $difftext = $wgContLang->unsegmentForDiff($difftext);
+ $out .= $difftext."</table>\n";
+ return $out;
}
# Load the text of the articles to compare. If newid is 0, then compare
function loadText()
{
global $wgTitle, $wgOut, $wgLang;
- $fname = "DifferenceEngine::loadText";
+ $fname = 'DifferenceEngine::loadText';
$dbr =& wfGetDB( DB_SLAVE );
- if ( 0 == $this->mNewid || 0 == $this->mOldid ) {
- $wgOut->setArticleFlag( true );
- $this->mNewtitle = wfMsg( "currentrev" );
- $id = $wgTitle->getArticleID();
-
- $s = $dbr->getArray( 'cur', array( 'cur_text', 'cur_user_text', 'cur_comment' ),
- array( 'cur_id' => $id ), $fname );
- if ( $s === false ) {
- return false;
- }
-
- $this->mNewPage = &$wgTitle;
- $this->mNewtext = $s->cur_text;
- $this->mNewUser = $s->cur_user_text;
- $this->mNewComment = $s->cur_comment;
+ if( $this->mNewid ) {
+ $this->newRev =& Revision::newFromId( $this->mNewid );
} 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 );
-
- if ( $s === false ) {
- return false;
- }
-
- $this->mNewtext = Article::getRevisionText( $s );
-
- $t = $wgLang->timeanddate( $s->old_timestamp, true );
- $this->mNewPage = Title::MakeTitle( $s->old_namespace, $s->old_title );
- $this->mNewtitle = wfMsg( "revisionasof", $t );
- $this->mNewUser = $s->old_user_text;
- $this->mNewComment = $s->old_comment;
+ $this->newRev =& Revision::newFromTitle( $wgTitle );
}
- 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( $this->newRev->isCurrent() ) {
+ $this->mPagetitle = htmlspecialchars( wfMsg( 'currentrev' ) );
+ $this->mNewPage = $wgTitle;
+ $newLink = $this->mNewPage->escapeLocalUrl();
+ $this->mNewtitle = "<a href='$newLink'>{$this->mPagetitle}</a>";
} 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
- );
+ $this->mNewPage = $this->newRev->getTitle();
+ $newLink = $this->mNewPage->escapeLocalUrl ('oldid=' . $this->mNewid );
+ $t = $wgLang->timeanddate( $this->newRev->getTimestamp(), true );
+ $this->mPagetitle = htmlspecialchars( wfMsg( 'revisionasof', $t ) );
+ $this->mNewtitle = "<a href='$newLink'>{$this->mPagetitle}</a>";
}
- if ( $s === false ) {
- return false;
+
+ if( $this->mOldid ) {
+ $this->oldRev =& Revision::newFromId( $this->mOldid );
+ } else {
+ $this->oldRev =& $this->newRev->getPrevious();
+ $this->mOldid = $this->oldRev->getId();
}
- $this->mOldPage = Title::MakeTitle( $s->old_namespace, $s->old_title );
- $this->mOldtext = Article::getRevisionText( $s );
-
- $t = $wgLang->timeanddate( $s->old_timestamp, true );
- $this->mOldtitle = wfMsg( "revisionasof", $t );
- $this->mOldUser = $s->old_user_text;
- $this->mOldComment = $s->old_comment;
+
+ $this->mOldPage = $this->oldRev->getTitle();
+
+ $t = $wgLang->timeanddate( $this->oldRev->getTimestamp(), true );
+ $oldLink = $this->mOldPage->escapeLocalUrl( 'oldid=' . $this->mOldid );
+ $this->mOldtitle = "<a href='$oldLink'>" . htmlspecialchars( wfMsg( 'revisionasof', $t ) ) . '</a>';
+
+ $this->mNewUser = $this->newRev->getUserText();
+ $this->mNewComment = $this->newRev->getComment();
+ $this->mNewtext = $this->newRev->getText();
+
+ $this->mOldUser = $this->oldRev->getUserText();
+ $this->mOldComment = $this->oldRev->getComment();
+ $this->mOldtext = $this->oldRev->getText();
return true;
}
define('USE_ASSERTS', function_exists('assert'));
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class _DiffOp {
var $type;
var $orig;
var $closing;
function reverse() {
- trigger_error("pure virtual", E_USER_ERROR);
+ trigger_error('pure virtual', E_USER_ERROR);
}
function norig() {
}
}
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class _DiffOp_Copy extends _DiffOp {
var $type = 'copy';
}
}
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class _DiffOp_Delete extends _DiffOp {
var $type = 'delete';
}
}
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class _DiffOp_Add extends _DiffOp {
var $type = 'add';
}
}
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class _DiffOp_Change extends _DiffOp {
var $type = 'change';
*
* @author Geoffrey T. Dairiki
* @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
*/
class _DiffEngine
{
function diff ($from_lines, $to_lines) {
+ $fname = '_DiffEngine::diff';
+ wfProfileIn( $fname );
+
$n_from = sizeof($from_lines);
$n_to = sizeof($to_lines);
elseif ($add)
$edits[] = new _DiffOp_Add($add);
}
+ wfProfileOut( $fname );
return $edits;
}
* of the portions it is going to specify.
*/
function _diag ($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);
+ $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; $chunk++) {
- if ($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; $x++) {
+ if ($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, $ylim);
-
- return array($this->lcs, $seps);
+ $seps[] = $flip ? array($ylim, $xlim) : array($xlim, $ylim);
+
+ 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]);
-
- $this->in_seq[$this->seq[$end]] = false;
- $this->seq[$end] = $ypos;
- $this->in_seq[$ypos] = 1;
- return $end;
+
+ USE_ASSERTS && assert($ypos != $this->seq[$end]);
+
+ $this->in_seq[$this->seq[$end]] = false;
+ $this->seq[$end] = $ypos;
+ $this->in_seq[$ypos] = 1;
+ wfProfileOut( $fname );
+ return $end;
}
/* Find LCS of two sequences.
* All line numbers are origin-0 and discarded lines are not counted.
*/
function _compareseq ($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 ($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 = $pt2;
+ if ($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( $fname );
}
/* Adjust inserts/deletes of identical lines to join changes
* This is extracted verbatim from analyze.c (GNU diffutils-2.7).
*/
function _shift_boundaries ($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])
- $j++;
- }
-
- if ($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;
+ $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]');
+ }
+
+ /*
+ * 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
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
*/
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;
}
/**
* @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 $lcs;
}
/**
* This is here only for debugging purposes.
*/
function _check ($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()))
}
$lcs = $this->lcs();
- trigger_error("Diff okay: LCS = $lcs", E_USER_NOTICE);
+ trigger_error('Diff okay: LCS = '.$lcs, E_USER_NOTICE);
+ wfProfileOut( $fname );
}
}
/**
* FIXME: bad name.
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
*/
-class MappedDiff
-extends Diff
+class MappedDiff extends Diff
{
/**
* Constructor.
*/
function MappedDiff($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));
$yi += sizeof($closing);
}
}
+ wfProfileOut( $fname );
}
}
* This class formats the diff in classic diff format.
* It is intended that this class be customized via inheritance,
* to obtain fancier outputs.
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
*/
class DiffFormatter
{
* @return string The formatted output.
*/
function format($diff) {
+ $fname = 'DiffFormatter::format';
+ wfProfileIn( $fname );
$xi = $yi = 1;
$block = false;
$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')
elseif ($edit->type == 'change')
$this->_changed($edit->orig, $edit->closing);
else
- trigger_error("Unknown edit type", E_USER_ERROR);
+ trigger_error('Unknown edit type', E_USER_ERROR);
}
$this->_end_block();
+ wfProfileOut( $fname );
}
function _start_diff() {
}
function _added($lines) {
- $this->_lines($lines, ">");
+ $this->_lines($lines, '>');
}
function _deleted($lines) {
- $this->_lines($lines, "<");
+ $this->_lines($lines, '<');
}
function _changed($orig, $closing) {
define('NBSP', ' '); // iso-8859-x non-breaking space.
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class _HWLDF_WordAccumulator {
function _HWLDF_WordAccumulator () {
$this->_lines = array();
function _flushGroup ($new_tag) {
if ($this->_group !== '') {
- if ($this->_tag == 'mark')
- $this->_line .= '<span class="diffchange">'.$this->_group.'</span>';
- else
- $this->_line .= $this->_group;
- }
+ if ($this->_tag == 'mark')
+ $this->_line .= '<span class="diffchange">' .
+ htmlspecialchars ( $this->_group ) . '</span>';
+ else
+ $this->_line .= htmlspecialchars ( $this->_group );
+ }
$this->_group = '';
$this->_tag = $new_tag;
}
function _flushLine ($new_tag) {
$this->_flushGroup($new_tag);
if ($this->_line != '')
- $this->_lines[] = $this->_line;
+ array_push ( $this->_lines, $this->_line );
+ else
+ # make empty lines visible by inserting an NBSP
+ array_push ( $this->_lines, NBSP );
$this->_line = '';
}
if ($word == '')
continue;
if ($word[0] == "\n") {
- $this->_group .= NBSP;
$this->_flushLine($tag);
$word = substr($word, 1);
}
}
}
+/**
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
+ */
class WordLevelDiff extends MappedDiff
{
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) {
- // FIXME: fix POSIX char class.
-# if (!preg_match_all('/ ( [^\S\n]+ | [[:alnum:]]+ | . ) (?: (?!< \n) [^\S\n])? /xs',
+ $fname = 'WordLevelDiff::_split';
+ wfProfileIn( $fname );
if (!preg_match_all('/ ( [^\S\n]+ | [0-9_A-Za-z\x80-\xff]+ | . ) (?: (?!< \n) [^\S\n])? /xs',
implode("\n", $lines),
$m)) {
+ wfProfileOut( $fname );
return array(array(''), array(''));
}
+ wfProfileOut( $fname );
return array($m[0], $m[1]);
}
function orig () {
+ $fname = 'WordLevelDiff::orig';
+ wfProfileIn( $fname );
$orig = new _HWLDF_WordAccumulator;
foreach ($this->edits as $edit) {
elseif ($edit->orig)
$orig->addWords($edit->orig, 'mark');
}
- 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 $edit) {
elseif ($edit->closing)
$closing->addWords($edit->closing, 'mark');
}
- return $closing->getLines();
+ $lines = $closing->getLines();
+ wfProfileOut( $fname );
+ return $lines;
}
}
/**
* Wikipedia Table style diff formatter.
- *
+ * @todo document
+ * @access private
+ * @package MediaWiki
+ * @subpackage DifferenceEngine
*/
class TableDiffFormatter extends DiffFormatter
{
}
function _block_header( $xbeg, $xlen, $ybeg, $ylen ) {
- $l1 = wfMsg( "lineno", $xbeg );
- $l2 = wfMsg( "lineno", $ybeg );
+ $l1 = wfMsg( 'lineno', $xbeg );
+ $l2 = wfMsg( 'lineno', $ybeg );
$r = '<tr><td colspan="2" align="left"><strong>'.$l1."</strong></td>\n" .
'<td colspan="2" align="left"><strong>'.$l2."</strong></td></tr>\n";
function _start_block( $header ) {
global $wgOut;
- $wgOut->addHTML( $header );
+ echo $header;
}
function _end_block() {
}
- function _lines( $lines, $prefix=' ', $color="white" ) {
+ function _lines( $lines, $prefix=' ', $color='white' ) {
}
+ # HTML-escape parameter before calling this
function addedLine( $line ) {
- return '<td>+</td><td class="diff-addedline">' .
- $line.'</td>';
+ return "<td>+</td><td class='diff-addedline'>{$line}</td>";
}
+ # HTML-escape parameter before calling this
function deletedLine( $line ) {
- return '<td>-</td><td class="diff-deletedline">' .
- $line.'</td>';
+ return "<td>-</td><td class='diff-deletedline'>{$line}</td>";
}
- function emptyLine() {
- return '<td colspan="2"> </td>';
+ # HTML-escape parameter before calling this
+ function contextLine( $line ) {
+ return "<td> </td><td class='diff-context'>{$line}</td>";
}
- function contextLine( $line ) {
- return '<td> </td><td class="diff-context">'.$line.'</td>';
+ function emptyLine() {
+ return '<td colspan="2"> </td>';
}
- function _added($lines) {
- global $wgOut;
+ function _added( $lines ) {
foreach ($lines as $line) {
- $wgOut->addHTML( '<tr>' . $this->emptyLine() .
- $this->addedLine( $line ) . "</tr>\n" );
+ echo '<tr>' . $this->emptyLine() .
+ $this->addedLine( htmlspecialchars ( $line ) ) . "</tr>\n";
}
}
function _deleted($lines) {
- global $wgOut;
foreach ($lines as $line) {
- $wgOut->addHTML( '<tr>' . $this->deletedLine( $line ) .
- $this->emptyLine() . "</tr>\n" );
+ echo '<tr>' . $this->deletedLine( htmlspecialchars ( $line ) ) .
+ $this->emptyLine() . "</tr>\n";
}
}
function _context( $lines ) {
- global $wgOut;
foreach ($lines as $line) {
- $wgOut->addHTML( '<tr>' . $this->contextLine( $line ) .
- $this->contextLine( $line ) . "</tr>\n" );
+ echo '<tr>' .
+ $this->contextLine( htmlspecialchars ( $line ) ) .
+ $this->contextLine( htmlspecialchars ( $line ) ) . "</tr>\n";
}
}
function _changed( $orig, $closing ) {
- global $wgOut;
+ $fname = 'TableDiffFormatter::_changed';
+ wfProfileIn( $fname );
+
$diff = new WordLevelDiff( $orig, $closing );
$del = $diff->orig();
$add = $diff->closing();
+ # Notice that WordLevelDiff returns HTML-escaped output.
+ # Hence, we will be calling addedLine/deletedLine without HTML-escaping.
+
while ( $line = array_shift( $del ) ) {
$aline = array_shift( $add );
- $wgOut->addHTML( '<tr>' . $this->deletedLine( $line ) .
- $this->addedLine( $aline ) . "</tr>\n" );
+ echo '<tr>' . $this->deletedLine( $line ) .
+ $this->addedLine( $aline ) . "</tr>\n";
}
- $this->_added( $add ); # If any leftovers
+ foreach ($add as $line) { # If any leftovers
+ echo '<tr>' . $this->emptyLine() .
+ $this->addedLine( $line ) . "</tr>\n";
+ }
+ wfProfileOut( $fname );
}
}