From 35086b850f61cef9730f940e9c6cf0991809c020 Mon Sep 17 00:00:00 2001 From: ryan Date: Fri, 18 Apr 2008 23:38:21 +0000 Subject: [PATCH] Post revisions from mdawaffe. see #6775 git-svn-id: http://svn.automattic.com/wordpress/trunk@7747 1a063a9b-81f0-0310-95a4-ce76da25c4cd --- wp-admin/admin-ajax.php | 2 + wp-admin/css/colors-classic.css | 17 +- wp-admin/css/colors-fresh.css | 17 +- wp-admin/css/ie.css | 3 + wp-admin/edit-form-advanced.php | 10 + wp-admin/page.php | 4 +- wp-admin/post.php | 4 +- wp-admin/revision.php | 132 +++++++ wp-admin/wp-admin.css | 39 ++ wp-includes/Text/Diff.php | 413 ++++++++++++++++++++ wp-includes/Text/Diff/Engine/native.php | 437 ++++++++++++++++++++++ wp-includes/Text/Diff/Engine/shell.php | 164 ++++++++ wp-includes/Text/Diff/Engine/string.php | 234 ++++++++++++ wp-includes/Text/Diff/Engine/xdiff.php | 63 ++++ wp-includes/Text/Diff/Renderer.php | 237 ++++++++++++ wp-includes/Text/Diff/Renderer/inline.php | 170 +++++++++ wp-includes/default-filters.php | 1 + wp-includes/link-template.php | 16 +- wp-includes/pluggable.php | 64 ++++ wp-includes/post-template.php | 134 ++++++- wp-includes/post.php | 228 +++++++++++ wp-includes/wp-diff.php | 318 ++++++++++++++++ 22 files changed, 2698 insertions(+), 9 deletions(-) create mode 100644 wp-admin/revision.php create mode 100644 wp-includes/Text/Diff.php create mode 100644 wp-includes/Text/Diff/Engine/native.php create mode 100644 wp-includes/Text/Diff/Engine/shell.php create mode 100644 wp-includes/Text/Diff/Engine/string.php create mode 100644 wp-includes/Text/Diff/Engine/xdiff.php create mode 100644 wp-includes/Text/Diff/Renderer.php create mode 100644 wp-includes/Text/Diff/Renderer/inline.php create mode 100644 wp-includes/wp-diff.php diff --git a/wp-admin/admin-ajax.php b/wp-admin/admin-ajax.php index aa9165222..4e8716440 100644 --- a/wp-admin/admin-ajax.php +++ b/wp-admin/admin-ajax.php @@ -460,6 +460,8 @@ case 'add-user' : $x->send(); break; case 'autosave' : // The name of this action is hardcoded in edit_post() + define( 'DOING_AUTOSAVE', true ); + $nonce_age = check_ajax_referer( 'autosave', 'autosavenonce'); global $current_user; diff --git a/wp-admin/css/colors-classic.css b/wp-admin/css/colors-classic.css index 3a8f1380c..08010ee2c 100644 --- a/wp-admin/css/colors-classic.css +++ b/wp-admin/css/colors-classic.css @@ -2,7 +2,7 @@ a.page-numbers:hover { border-color: #999; } -body { +body, .form-table pre { background-color: #fff; color: #333; } @@ -713,3 +713,18 @@ div#media-upload-error, .file-error, abbr.required, .widget-control-remove:hover background-color: #ddd; color: #333; } + +/* Diff */ + +table.diff .diff-deletedline { + background-color: #ffdddd; +} +table.diff .diff-deletedline del { + background-color: #ff9999; +} +table.diff .diff-addedline { + background-color: #ddffdd; +} +table.diff .diff-addedline ins { + background-color: #99ff99; +} diff --git a/wp-admin/css/colors-fresh.css b/wp-admin/css/colors-fresh.css index a0d45fb19..6819cf973 100644 --- a/wp-admin/css/colors-fresh.css +++ b/wp-admin/css/colors-fresh.css @@ -2,7 +2,7 @@ a.page-numbers:hover { border-color: #999; } -body { +body, .form-table pre { background-color: #fff; color: #333; } @@ -684,3 +684,18 @@ div#media-upload-error, .file-error, abbr.required, .widget-control-remove:hover background-color: #ddd; color: #333; } + +/* Diff */ + +table.diff .diff-deletedline { + background-color: #ffdddd; +} +table.diff .diff-deletedline del { + background-color: #ff9999; +} +table.diff .diff-addedline { + background-color: #ddffdd; +} +table.diff .diff-addedline ins { + background-color: #99ff99; +} diff --git a/wp-admin/css/ie.css b/wp-admin/css/ie.css index 39ef25463..e4d035548 100644 --- a/wp-admin/css/ie.css +++ b/wp-admin/css/ie.css @@ -111,7 +111,10 @@ div#dashboard-widgets { .tablenav-pages { display: block; margin-top: -3px; +} +table.ie-fixed { + table-layout: fixed; } #post-search .button, #widget-search .button { diff --git a/wp-admin/edit-form-advanced.php b/wp-admin/edit-form-advanced.php index 366206977..8ddfc1137 100644 --- a/wp-admin/edit-form-advanced.php +++ b/wp-admin/edit-form-advanced.php @@ -6,6 +6,7 @@ $messages[1] = sprintf( __( 'Post updated. Continue editing below or

@@ -336,6 +337,15 @@ if ( $authors && count( $authors ) > 1 ) : + +
+

+
+ +
+
+ + diff --git a/wp-admin/page.php b/wp-admin/page.php index 03c129680..e3aac547e 100644 --- a/wp-admin/page.php +++ b/wp-admin/page.php @@ -70,8 +70,8 @@ case 'edit': if ( empty($post->ID) ) wp_die( __("You attempted to edit a page that doesn't exist. Perhaps it was deleted?") ); - if ( 'post' == $post->post_type ) { - wp_redirect("post.php?action=edit&post=$post_ID"); + if ( 'page' != $post->post_type ) { + wp_redirect( get_edit_post_link( $post_ID, 'url' ) ); exit(); } diff --git a/wp-admin/post.php b/wp-admin/post.php index 66e0ab761..d9a799fba 100644 --- a/wp-admin/post.php +++ b/wp-admin/post.php @@ -77,8 +77,8 @@ case 'edit': if ( empty($post->ID) ) wp_die( __("You attempted to edit a post that doesn't exist. Perhaps it was deleted?") ); - if ( 'page' == $post->post_type ) { - wp_redirect("page.php?action=edit&post=$post_ID"); + if ( 'post' != $post->post_type ) { + wp_redirect( get_edit_post_link( $post->ID, 'url' ) ); exit(); } diff --git a/wp-admin/revision.php b/wp-admin/revision.php new file mode 100644 index 000000000..18bfcc4a3 --- /dev/null +++ b/wp-admin/revision.php @@ -0,0 +1,132 @@ +post_type ? get_post( $revision->post_parent ) : get_post( $revision_id ); + $left_revision = get_post( $diff ); + + // Don't allow reverse diffs? + if ( strtotime($revision->post_modified_gmt) < strtotime($left_revision->post_modified_gmt) ) { + wp_redirect( add_query_arg( array( 'diff' => $revision->ID, 'revision' => $diff ) ) ); + exit; + } + + $h2 = __( 'Compare Revisions of “%1$s”' ); + $right = $revision->ID; + $left = $left_revision->ID; + + if ( + // They're the same + $left_revision->ID == $revision->ID + || + // They don't have a comment parent (and we're not comparing a revision to it's post) + ( $left_revision->ID != $revision->post_parent && $left_revision->post_parent != $revision->ID && $left_revision->post_parent != $revision->post_parent ) + || + // Neither is a revision + ( !wp_get_revision( $left_revision->ID ) && !wp_get_revision( $revision->ID ) ) + ) { + wp_redirect( get_edit_post_link( $revision->ID, 'url' ) ); + exit(); + } +} else { + $revision = wp_get_revision( $revision_id ); + $post = get_post( $revision->post_parent ); + $h2 = __( 'Post Revision for “%1$s” created on %2$s' ); + $right = $post->ID; + $left = $revision->ID; +} + +if ( !$revision || !$post ) { + wp_redirect("edit.php"); + exit(); +} + +if ( $restore && current_user_can( 'edit_post', $revision->post_parent ) ) { + check_admin_referer( "restore-post_$post->ID|$revision->ID" ); + wp_restore_revision( $revision->ID ); + wp_redirect( add_query_arg( array( 'message' => 5, 'revision' => $revision->ID ), get_edit_post_link( $post->ID, 'url' ) ) ); + exit; +} + +add_filter( '_wp_revision_field_post_author', 'get_author_name' ); + +$title = __( 'Post Revision' ); + +require_once( 'admin-header.php' ); + +$post_title = '' . get_the_title() . ''; +$revision_time = wp_post_revision_time( $revision ); +?> + +
+ +

+ + + + + + + + + + $field_title ) : + if ( !$diff ) + add_filter( "_wp_revision_field_$field", 'htmlspecialchars' ); + $content = apply_filters( "_wp_revision_field_$field", $revision->$field, $field ); + if ( $diff ) { + $left_content = apply_filters( "_wp_revision_field_$field", $left_revision->$field, $field ); + if ( !$content = wp_text_diff( $left_content, $content ) ) + continue; + } + $identical = false; + ?> + + + + + + + + + + + + +
+ +

+ +
+ +

+ + 'form-table', 'exclude' => $revision->ID, 'parent' => true, 'right' => $right, 'left' => $left ) ); + + require_once( 'admin-footer.php' ); diff --git a/wp-admin/wp-admin.css b/wp-admin/wp-admin.css index 7a9283ca1..c29a81ec3 100644 --- a/wp-admin/wp-admin.css +++ b/wp-admin/wp-admin.css @@ -894,6 +894,21 @@ html, body { margin-right: 5px } +.form-table pre { + padding: 8px; + margin: 0; + /* http://www.longren.org/2006/09/27/wrapping-text-inside-pre-tags/ */ + white-space: pre-wrap; /* css-3 */ + white-space: -moz-pre-wrap !important; /* Mozilla, since 1999 */ + white-space: -pre-wrap; /* Opera 4-6 */ + white-space: -o-pre-wrap; /* Opera 7 */ + word-wrap: break-word; /* Internet Explorer 5.5+ */ +} + +table.form-table td .updated { + font-size: 13px; +} + /* Post Screen */ #tagsdiv #newtag { @@ -1464,3 +1479,27 @@ td.action-links, th.action-links { .hide-if-no-js { display: none; } + +/* Diff */ + +table.diff { + width: 100%; +} + +table.diff col.content { + width: 50%; +} + +table.diff tr { + background-color: transparent; +} + +table.diff td, table.diff th { + padding: .5em; + font-family: monospace; + border: none; +} + +table.diff .diff-deletedline del, table.diff .diff-addedline ins { + text-decoration: none; +} diff --git a/wp-includes/Text/Diff.php b/wp-includes/Text/Diff.php new file mode 100644 index 000000000..f8a974c1c --- /dev/null +++ b/wp-includes/Text/Diff.php @@ -0,0 +1,413 @@ +, and is used/adapted with his permission. + * + * $Horde: framework/Text_Diff/Diff.php,v 1.26 2008/01/04 10:07:49 jan Exp $ + * + * Copyright 2004 Geoffrey T. Dairiki + * Copyright 2004-2008 The Horde Project (http://www.horde.org/) + * + * See the enclosed file COPYING for license information (LGPL). If you did + * not receive this file, see http://opensource.org/licenses/lgpl-license.php. + * + * @package Text_Diff + * @author Geoffrey T. Dairiki + */ +class Text_Diff { + + /** + * Array of changes. + * + * @var array + */ + var $_edits; + + /** + * Computes diffs between sequences of strings. + * + * @param string $engine Name of the diffing engine to use. 'auto' + * will automatically select the best. + * @param array $params Parameters to pass to the diffing engine. + * Normally an array of two arrays, each + * containing the lines from a file. + */ + function Text_Diff($engine, $params) + { + // Backward compatibility workaround. + if (!is_string($engine)) { + $params = array($engine, $params); + $engine = 'auto'; + } + + if ($engine == 'auto') { + $engine = extension_loaded('xdiff') ? 'xdiff' : 'native'; + } else { + $engine = basename($engine); + } + + require_once 'Text/Diff/Engine/' . $engine . '.php'; + $class = 'Text_Diff_Engine_' . $engine; + $diff_engine = new $class(); + + $this->_edits = call_user_func_array(array($diff_engine, 'diff'), $params); + } + + /** + * Returns the array of differences. + */ + function getDiff() + { + return $this->_edits; + } + + /** + * Computes a reversed diff. + * + * Example: + * + * $diff = new Text_Diff($lines1, $lines2); + * $rev = $diff->reverse(); + * + * + * @return Text_Diff A Diff object representing the inverse of the + * original diff. Note that we purposely don't return a + * reference here, since this essentially is a clone() + * method. + */ + function reverse() + { + if (version_compare(zend_version(), '2', '>')) { + $rev = clone($this); + } else { + $rev = $this; + } + $rev->_edits = array(); + foreach ($this->_edits as $edit) { + $rev->_edits[] = $edit->reverse(); + } + return $rev; + } + + /** + * Checks for an empty diff. + * + * @return boolean True if two sequences were identical. + */ + function isEmpty() + { + foreach ($this->_edits as $edit) { + if (!is_a($edit, 'Text_Diff_Op_copy')) { + return false; + } + } + return true; + } + + /** + * Computes the length of the Longest Common Subsequence (LCS). + * + * This is mostly for diagnostic purposes. + * + * @return integer The length of the LCS. + */ + function lcs() + { + $lcs = 0; + foreach ($this->_edits as $edit) { + if (is_a($edit, 'Text_Diff_Op_copy')) { + $lcs += count($edit->orig); + } + } + return $lcs; + } + + /** + * Gets the original set of lines. + * + * This reconstructs the $from_lines parameter passed to the constructor. + * + * @return array The original sequence of strings. + */ + function getOriginal() + { + $lines = array(); + foreach ($this->_edits as $edit) { + if ($edit->orig) { + array_splice($lines, count($lines), 0, $edit->orig); + } + } + return $lines; + } + + /** + * Gets the final set of lines. + * + * This reconstructs the $to_lines parameter passed to the constructor. + * + * @return array The sequence of strings. + */ + function getFinal() + { + $lines = array(); + foreach ($this->_edits as $edit) { + if ($edit->final) { + array_splice($lines, count($lines), 0, $edit->final); + } + } + return $lines; + } + + /** + * Removes trailing newlines from a line of text. This is meant to be used + * with array_walk(). + * + * @param string $line The line to trim. + * @param integer $key The index of the line in the array. Not used. + */ + function trimNewlines(&$line, $key) + { + $line = str_replace(array("\n", "\r"), '', $line); + } + + /** + * Determines the location of the system temporary directory. + * + * @static + * + * @access protected + * + * @return string A directory name which can be used for temp files. + * Returns false if one could not be found. + */ + function _getTempDir() + { + $tmp_locations = array('/tmp', '/var/tmp', 'c:\WUTemp', 'c:\temp', + 'c:\windows\temp', 'c:\winnt\temp'); + + /* Try PHP's upload_tmp_dir directive. */ + $tmp = ini_get('upload_tmp_dir'); + + /* Otherwise, try to determine the TMPDIR environment variable. */ + if (!strlen($tmp)) { + $tmp = getenv('TMPDIR'); + } + + /* If we still cannot determine a value, then cycle through a list of + * preset possibilities. */ + while (!strlen($tmp) && count($tmp_locations)) { + $tmp_check = array_shift($tmp_locations); + if (@is_dir($tmp_check)) { + $tmp = $tmp_check; + } + } + + /* If it is still empty, we have failed, so return false; otherwise + * return the directory determined. */ + return strlen($tmp) ? $tmp : false; + } + + /** + * Checks a diff for validity. + * + * This is here only for debugging purposes. + */ + function _check($from_lines, $to_lines) + { + if (serialize($from_lines) != serialize($this->getOriginal())) { + trigger_error("Reconstructed original doesn't match", E_USER_ERROR); + } + if (serialize($to_lines) != serialize($this->getFinal())) { + trigger_error("Reconstructed final doesn't match", E_USER_ERROR); + } + + $rev = $this->reverse(); + if (serialize($to_lines) != serialize($rev->getOriginal())) { + trigger_error("Reversed original doesn't match", E_USER_ERROR); + } + if (serialize($from_lines) != serialize($rev->getFinal())) { + trigger_error("Reversed final doesn't match", E_USER_ERROR); + } + + $prevtype = null; + foreach ($this->_edits as $edit) { + if ($prevtype == get_class($edit)) { + trigger_error("Edit sequence is non-optimal", E_USER_ERROR); + } + $prevtype = get_class($edit); + } + + return true; + } + +} + +/** + * @package Text_Diff + * @author Geoffrey T. Dairiki + */ +class Text_MappedDiff extends Text_Diff { + + /** + * Computes a diff between sequences of strings. + * + * This can be used to compute things like case-insensitve diffs, or diffs + * which ignore changes in white-space. + * + * @param array $from_lines An array of strings. + * @param array $to_lines An array of strings. + * @param array $mapped_from_lines This array should have the same size + * number of elements as $from_lines. The + * elements in $mapped_from_lines and + * $mapped_to_lines are what is actually + * compared when computing the diff. + * @param array $mapped_to_lines This array should have the same number + * of elements as $to_lines. + */ + function Text_MappedDiff($from_lines, $to_lines, + $mapped_from_lines, $mapped_to_lines) + { + assert(count($from_lines) == count($mapped_from_lines)); + assert(count($to_lines) == count($mapped_to_lines)); + + parent::Text_Diff($mapped_from_lines, $mapped_to_lines); + + $xi = $yi = 0; + for ($i = 0; $i < count($this->_edits); $i++) { + $orig = &$this->_edits[$i]->orig; + if (is_array($orig)) { + $orig = array_slice($from_lines, $xi, count($orig)); + $xi += count($orig); + } + + $final = &$this->_edits[$i]->final; + if (is_array($final)) { + $final = array_slice($to_lines, $yi, count($final)); + $yi += count($final); + } + } + } + +} + +/** + * @package Text_Diff + * @author Geoffrey T. Dairiki + * + * @access private + */ +class Text_Diff_Op { + + var $orig; + var $final; + + function &reverse() + { + trigger_error('Abstract method', E_USER_ERROR); + } + + function norig() + { + return $this->orig ? count($this->orig) : 0; + } + + function nfinal() + { + return $this->final ? count($this->final) : 0; + } + +} + +/** + * @package Text_Diff + * @author Geoffrey T. Dairiki + * + * @access private + */ +class Text_Diff_Op_copy extends Text_Diff_Op { + + function Text_Diff_Op_copy($orig, $final = false) + { + if (!is_array($final)) { + $final = $orig; + } + $this->orig = $orig; + $this->final = $final; + } + + function &reverse() + { + $reverse = &new Text_Diff_Op_copy($this->final, $this->orig); + return $reverse; + } + +} + +/** + * @package Text_Diff + * @author Geoffrey T. Dairiki + * + * @access private + */ +class Text_Diff_Op_delete extends Text_Diff_Op { + + function Text_Diff_Op_delete($lines) + { + $this->orig = $lines; + $this->final = false; + } + + function &reverse() + { + $reverse = &new Text_Diff_Op_add($this->orig); + return $reverse; + } + +} + +/** + * @package Text_Diff + * @author Geoffrey T. Dairiki + * + * @access private + */ +class Text_Diff_Op_add extends Text_Diff_Op { + + function Text_Diff_Op_add($lines) + { + $this->final = $lines; + $this->orig = false; + } + + function &reverse() + { + $reverse = &new Text_Diff_Op_delete($this->final); + return $reverse; + } + +} + +/** + * @package Text_Diff + * @author Geoffrey T. Dairiki + * + * @access private + */ +class Text_Diff_Op_change extends Text_Diff_Op { + + function Text_Diff_Op_change($orig, $final) + { + $this->orig = $orig; + $this->final = $final; + } + + function &reverse() + { + $reverse = &new Text_Diff_Op_change($this->final, $this->orig); + return $reverse; + } + +} diff --git a/wp-includes/Text/Diff/Engine/native.php b/wp-includes/Text/Diff/Engine/native.php new file mode 100644 index 000000000..aa5bca5b0 --- /dev/null +++ b/wp-includes/Text/Diff/Engine/native.php @@ -0,0 +1,437 @@ + 2, and some optimizations) are from + * Geoffrey T. Dairiki . The original PHP version of this + * code was written by him, and is used/adapted with his permission. + * + * Copyright 2004-2008 The Horde Project (http://www.horde.org/) + * + * See the enclosed file COPYING for license information (LGPL). If you did + * not receive this file, see http://opensource.org/licenses/lgpl-license.php. + * + * @author Geoffrey T. Dairiki + * @package Text_Diff + */ +class Text_Diff_Engine_native { + + function diff($from_lines, $to_lines) + { + array_walk($from_lines, array('Text_Diff', 'trimNewlines')); + array_walk($to_lines, array('Text_Diff', 'trimNewlines')); + + $n_from = count($from_lines); + $n_to = count($to_lines); + + $this->xchanged = $this->ychanged = array(); + $this->xv = $this->yv = array(); + $this->xind = $this->yind = array(); + unset($this->seq); + unset($this->in_seq); + unset($this->lcs); + + // Skip leading common lines. + for ($skip = 0; $skip < $n_from && $skip < $n_to; $skip++) { + if ($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]) { + 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[$xi]] = 1; + } + for ($yi = $skip; $yi < $n_to - $endskip; $yi++) { + $line = $to_lines[$yi]; + if (($this->ychanged[$yi] = empty($xhash[$line]))) { + continue; + } + $yhash[$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]))) { + continue; + } + $this->xv[] = $line; + $this->xind[] = $xi; + } + + // Find the LCS. + $this->_compareseq(0, count($this->xv), 0, count($this->yv)); + + // Merge edits when possible. + $this->_shiftBoundaries($from_lines, $this->xchanged, $this->ychanged); + $this->_shiftBoundaries($to_lines, $this->ychanged, $this->xchanged); + + // Compute the edit operations. + $edits = array(); + $xi = $yi = 0; + while ($xi < $n_from || $yi < $n_to) { + assert($yi < $n_to || $this->xchanged[$xi]); + assert($xi < $n_from || $this->ychanged[$yi]); + + // Skip matching "snake". + $copy = array(); + while ($xi < $n_from && $yi < $n_to + && !$this->xchanged[$xi] && !$this->ychanged[$yi]) { + $copy[] = $from_lines[$xi++]; + ++$yi; + } + if ($copy) { + $edits[] = &new Text_Diff_Op_copy($copy); + } + + // Find deletes & adds. + $delete = array(); + while ($xi < $n_from && $this->xchanged[$xi]) { + $delete[] = $from_lines[$xi++]; + } + + $add = array(); + while ($yi < $n_to && $this->ychanged[$yi]) { + $add[] = $to_lines[$yi++]; + } + + if ($delete && $add) { + $edits[] = &new Text_Diff_Op_change($delete, $add); + } elseif ($delete) { + $edits[] = &new Text_Diff_Op_delete($delete); + } elseif ($add) { + $edits[] = &new Text_Diff_Op_add($add); + } + } + + return $edits; + } + + /** + * Divides the Largest Common Subsequence (LCS) of the sequences (XOFF, + * XLIM) and (YOFF, YLIM) into NCHUNKS approximately equally sized + * segments. + * + * Returns (LCS, PTS). LCS is the length of the LCS. PTS is an array of + * NCHUNKS+1 (X, Y) indexes giving the diving points between sub + * sequences. The first sub-sequence is contained in (X0, X1), (Y0, Y1), + * the second in (X1, X2), (Y1, Y2) and so on. Note that (X0, Y0) == + * (XOFF, YOFF) and (X[NCHUNKS], Y[NCHUNKS]) == (XLIM, YLIM). + * + * This function assumes that the first lines of the specified portions of + * the two files do not match, and likewise that the last lines do not + * match. The caller must trim matching lines from the beginning and end + * of the portions it is going to specify. + */ + function _diag ($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 is 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++) { + $line = $flip ? $this->yv[$x] : $this->xv[$x]; + if (empty($ymatches[$line])) { + continue; + } + $matches = $ymatches[$line]; + reset($matches); + while (list(, $y) = each($matches)) { + if (empty($this->in_seq[$y])) { + $k = $this->_lcsPos($y); + assert($k > 0); + $ymids[$k] = $ymids[$k - 1]; + break; + } + } + while (list(, $y) = each($matches)) { + if ($y > $this->seq[$k - 1]) { + 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; + } elseif (empty($this->in_seq[$y])) { + $k = $this->_lcsPos($y); + assert($k > 0); + $ymids[$k] = $ymids[$k - 1]; + } + } + } + } + + $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, $seps); + } + + function _lcsPos($ypos) + { + $end = $this->lcs; + if ($end == 0 || $ypos > $this->seq[$end]) { + $this->seq[++$this->lcs] = $ypos; + $this->in_seq[$ypos] = 1; + return $this->lcs; + } + + $beg = 1; + while ($beg < $end) { + $mid = (int)(($beg + $end) / 2); + if ($ypos > $this->seq[$mid]) { + $beg = $mid + 1; + } else { + $end = $mid; + } + } + + assert($ypos != $this->seq[$end]); + + $this->in_seq[$this->seq[$end]] = false; + $this->seq[$end] = $ypos; + $this->in_seq[$ypos] = 1; + return $end; + } + + /** + * Finds LCS of two sequences. + * + * The results are recorded in the vectors $this->{x,y}changed[], by + * storing a 1 in the element for each line that is an insertion or + * deletion (ie. is not in the LCS). + * + * The subsequence of file 0 is (XOFF, XLIM) and likewise for file 1. + * + * Note that XLIM, YLIM are exclusive bounds. All line numbers are + * origin-0 and discarded lines are not counted. + */ + function _compareseq ($xoff, $xlim, $yoff, $ylim) + { + /* Slide down the bottom initial diagonal. */ + while ($xoff < $xlim && $yoff < $ylim + && $this->xv[$xoff] == $this->yv[$yoff]) { + ++$xoff; + ++$yoff; + } + + /* Slide up the top initial diagonal. */ + while ($xlim > $xoff && $ylim > $yoff + && $this->xv[$xlim - 1] == $this->yv[$ylim - 1]) { + --$xlim; + --$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 = $pt2; + } + } + } + + /** + * Adjusts inserts/deletes of identical lines to join changes as much as + * possible. + * + * We do something when a run of changed lines include a line at one end + * and has an excluded, identical line at the other. We are free to + * choose which identical line is included. `compareseq' usually chooses + * the one at the beginning, but usually it is cleaner to consider the + * following identical line to be the "change". + * + * This is extracted verbatim from analyze.c (GNU diffutils-2.7). + */ + function _shiftBoundaries($lines, &$changed, $other_changed) + { + $i = 0; + $j = 0; + + assert('count($lines) == count($changed)'); + $len = count($lines); + $other_len = count($other_changed); + + while (1) { + /* Scan forward to find the 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]) { + 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]) { + $start--; + } + assert('$j > 0'); + while ($other_changed[--$j]) { + continue; + } + 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]) { + $i++; + } + + assert('$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; + assert('$j > 0'); + while ($other_changed[--$j]) { + continue; + } + assert('$j >= 0 && !$other_changed[$j]'); + } + } + } + +} diff --git a/wp-includes/Text/Diff/Engine/shell.php b/wp-includes/Text/Diff/Engine/shell.php new file mode 100644 index 000000000..7e9629f50 --- /dev/null +++ b/wp-includes/Text/Diff/Engine/shell.php @@ -0,0 +1,164 @@ + + * @package Text_Diff + * @since 0.3.0 + */ +class Text_Diff_Engine_shell { + + /** + * Path to the diff executable + * + * @var string + */ + var $_diffCommand = 'diff'; + + /** + * Returns the array of differences. + * + * @param array $from_lines lines of text from old file + * @param array $to_lines lines of text from new file + * + * @return array all changes made (array with Text_Diff_Op_* objects) + */ + function diff($from_lines, $to_lines) + { + array_walk($from_lines, array('Text_Diff', 'trimNewlines')); + array_walk($to_lines, array('Text_Diff', 'trimNewlines')); + + $temp_dir = Text_Diff::_getTempDir(); + + // Execute gnu diff or similar to get a standard diff file. + $from_file = tempnam($temp_dir, 'Text_Diff'); + $to_file = tempnam($temp_dir, 'Text_Diff'); + $fp = fopen($from_file, 'w'); + fwrite($fp, implode("\n", $from_lines)); + fclose($fp); + $fp = fopen($to_file, 'w'); + fwrite($fp, implode("\n", $to_lines)); + fclose($fp); + $diff = shell_exec($this->_diffCommand . ' ' . $from_file . ' ' . $to_file); + unlink($from_file); + unlink($to_file); + + if (is_null($diff)) { + // No changes were made + return array(new Text_Diff_Op_copy($from_lines)); + } + + $from_line_no = 1; + $to_line_no = 1; + $edits = array(); + + // Get changed lines by parsing something like: + // 0a1,2 + // 1,2c4,6 + // 1,5d6 + preg_match_all('#^(\d+)(?:,(\d+))?([adc])(\d+)(?:,(\d+))?$#m', $diff, + $matches, PREG_SET_ORDER); + + foreach ($matches as $match) { + if (!isset($match[5])) { + // This paren is not set every time (see regex). + $match[5] = false; + } + + if ($match[3] == 'a') { + $from_line_no--; + } + + if ($match[3] == 'd') { + $to_line_no--; + } + + if ($from_line_no < $match[1] || $to_line_no < $match[4]) { + // copied lines + assert('$match[1] - $from_line_no == $match[4] - $to_line_no'); + array_push($edits, + new Text_Diff_Op_copy( + $this->_getLines($from_lines, $from_line_no, $match[1] - 1), + $this->_getLines($to_lines, $to_line_no, $match[4] - 1))); + } + + switch ($match[3]) { + case 'd': + // deleted lines + array_push($edits, + new Text_Diff_Op_delete( + $this->_getLines($from_lines, $from_line_no, $match[2]))); + $to_line_no++; + break; + + case 'c': + // changed lines + array_push($edits, + new Text_Diff_Op_change( + $this->_getLines($from_lines, $from_line_no, $match[2]), + $this->_getLines($to_lines, $to_line_no, $match[5]))); + break; + + case 'a': + // added lines + array_push($edits, + new Text_Diff_Op_add( + $this->_getLines($to_lines, $to_line_no, $match[5]))); + $from_line_no++; + break; + } + } + + if (!empty($from_lines)) { + // Some lines might still be pending. Add them as copied + array_push($edits, + new Text_Diff_Op_copy( + $this->_getLines($from_lines, $from_line_no, + $from_line_no + count($from_lines) - 1), + $this->_getLines($to_lines, $to_line_no, + $to_line_no + count($to_lines) - 1))); + } + + return $edits; + } + + /** + * Get lines from either the old or new text + * + * @access private + * + * @param array &$text_lines Either $from_lines or $to_lines + * @param int &$line_no Current line number + * @param int $end Optional end line, when we want to chop more + * than one line. + * + * @return array The chopped lines + */ + function _getLines(&$text_lines, &$line_no, $end = false) + { + if (!empty($end)) { + $lines = array(); + // We can shift even more + while ($line_no <= $end) { + array_push($lines, array_shift($text_lines)); + $line_no++; + } + } else { + $lines = array(array_shift($text_lines)); + $line_no++; + } + + return $lines; + } + +} diff --git a/wp-includes/Text/Diff/Engine/string.php b/wp-includes/Text/Diff/Engine/string.php new file mode 100644 index 000000000..1a0bd3f37 --- /dev/null +++ b/wp-includes/Text/Diff/Engine/string.php @@ -0,0 +1,234 @@ + + * $patch = file_get_contents('example.patch'); + * $diff = new Text_Diff('string', array($patch)); + * $renderer = new Text_Diff_Renderer_inline(); + * echo $renderer->render($diff); + * + * + * $Horde: framework/Text_Diff/Diff/Engine/string.php,v 1.7 2008/01/04 10:07:50 jan Exp $ + * + * Copyright 2005 Örjan Persson + * Copyright 2005-2008 The Horde Project (http://www.horde.org/) + * + * See the enclosed file COPYING for license information (LGPL). If you did + * not receive this file, see http://opensource.org/licenses/lgpl-license.php. + * + * @author Örjan Persson + * @package Text_Diff + * @since 0.2.0 + */ +class Text_Diff_Engine_string { + + /** + * Parses a unified or context diff. + * + * First param contains the whole diff and the second can be used to force + * a specific diff type. If the second parameter is 'autodetect', the + * diff will be examined to find out which type of diff this is. + * + * @param string $diff The diff content. + * @param string $mode The diff mode of the content in $diff. One of + * 'context', 'unified', or 'autodetect'. + * + * @return array List of all diff operations. + */ + function diff($diff, $mode = 'autodetect') + { + if ($mode != 'autodetect' && $mode != 'context' && $mode != 'unified') { + return PEAR::raiseError('Type of diff is unsupported'); + } + + if ($mode == 'autodetect') { + $context = strpos($diff, '***'); + $unified = strpos($diff, '---'); + if ($context === $unified) { + return PEAR::raiseError('Type of diff could not be detected'); + } elseif ($context === false || $context === false) { + $mode = $context !== false ? 'context' : 'unified'; + } else { + $mode = $context < $unified ? 'context' : 'unified'; + } + } + + // split by new line and remove the diff header + $diff = explode("\n", $diff); + array_shift($diff); + array_shift($diff); + + if ($mode == 'context') { + return $this->parseContextDiff($diff); + } else { + return $this->parseUnifiedDiff($diff); + } + } + + /** + * Parses an array containing the unified diff. + * + * @param array $diff Array of lines. + * + * @return array List of all diff operations. + */ + function parseUnifiedDiff($diff) + { + $edits = array(); + $end = count($diff) - 1; + for ($i = 0; $i < $end;) { + $diff1 = array(); + switch (substr($diff[$i], 0, 1)) { + case ' ': + do { + $diff1[] = substr($diff[$i], 1); + } while (++$i < $end && substr($diff[$i], 0, 1) == ' '); + $edits[] = &new Text_Diff_Op_copy($diff1); + break; + + case '+': + // get all new lines + do { + $diff1[] = substr($diff[$i], 1); + } while (++$i < $end && substr($diff[$i], 0, 1) == '+'); + $edits[] = &new Text_Diff_Op_add($diff1); + break; + + case '-': + // get changed or removed lines + $diff2 = array(); + do { + $diff1[] = substr($diff[$i], 1); + } while (++$i < $end && substr($diff[$i], 0, 1) == '-'); + + while ($i < $end && substr($diff[$i], 0, 1) == '+') { + $diff2[] = substr($diff[$i++], 1); + } + if (count($diff2) == 0) { + $edits[] = &new Text_Diff_Op_delete($diff1); + } else { + $edits[] = &new Text_Diff_Op_change($diff1, $diff2); + } + break; + + default: + $i++; + break; + } + } + + return $edits; + } + + /** + * Parses an array containing the context diff. + * + * @param array $diff Array of lines. + * + * @return array List of all diff operations. + */ + function parseContextDiff(&$diff) + { + $edits = array(); + $i = $max_i = $j = $max_j = 0; + $end = count($diff) - 1; + while ($i < $end && $j < $end) { + while ($i >= $max_i && $j >= $max_j) { + // Find the boundaries of the diff output of the two files + for ($i = $j; + $i < $end && substr($diff[$i], 0, 3) == '***'; + $i++); + for ($max_i = $i; + $max_i < $end && substr($diff[$max_i], 0, 3) != '---'; + $max_i++); + for ($j = $max_i; + $j < $end && substr($diff[$j], 0, 3) == '---'; + $j++); + for ($max_j = $j; + $max_j < $end && substr($diff[$max_j], 0, 3) != '***'; + $max_j++); + } + + // find what hasn't been changed + $array = array(); + while ($i < $max_i && + $j < $max_j && + strcmp($diff[$i], $diff[$j]) == 0) { + $array[] = substr($diff[$i], 2); + $i++; + $j++; + } + + while ($i < $max_i && ($max_j-$j) <= 1) { + if ($diff[$i] != '' && substr($diff[$i], 0, 1) != ' ') { + break; + } + $array[] = substr($diff[$i++], 2); + } + + while ($j < $max_j && ($max_i-$i) <= 1) { + if ($diff[$j] != '' && substr($diff[$j], 0, 1) != ' ') { + break; + } + $array[] = substr($diff[$j++], 2); + } + if (count($array) > 0) { + $edits[] = &new Text_Diff_Op_copy($array); + } + + if ($i < $max_i) { + $diff1 = array(); + switch (substr($diff[$i], 0, 1)) { + case '!': + $diff2 = array(); + do { + $diff1[] = substr($diff[$i], 2); + if ($j < $max_j && substr($diff[$j], 0, 1) == '!') { + $diff2[] = substr($diff[$j++], 2); + } + } while (++$i < $max_i && substr($diff[$i], 0, 1) == '!'); + $edits[] = &new Text_Diff_Op_change($diff1, $diff2); + break; + + case '+': + do { + $diff1[] = substr($diff[$i], 2); + } while (++$i < $max_i && substr($diff[$i], 0, 1) == '+'); + $edits[] = &new Text_Diff_Op_add($diff1); + break; + + case '-': + do { + $diff1[] = substr($diff[$i], 2); + } while (++$i < $max_i && substr($diff[$i], 0, 1) == '-'); + $edits[] = &new Text_Diff_Op_delete($diff1); + break; + } + } + + if ($j < $max_j) { + $diff2 = array(); + switch (substr($diff[$j], 0, 1)) { + case '+': + do { + $diff2[] = substr($diff[$j++], 2); + } while ($j < $max_j && substr($diff[$j], 0, 1) == '+'); + $edits[] = &new Text_Diff_Op_add($diff2); + break; + + case '-': + do { + $diff2[] = substr($diff[$j++], 2); + } while ($j < $max_j && substr($diff[$j], 0, 1) == '-'); + $edits[] = &new Text_Diff_Op_delete($diff2); + break; + } + } + } + + return $edits; + } + +} diff --git a/wp-includes/Text/Diff/Engine/xdiff.php b/wp-includes/Text/Diff/Engine/xdiff.php new file mode 100644 index 000000000..a39a4be17 --- /dev/null +++ b/wp-includes/Text/Diff/Engine/xdiff.php @@ -0,0 +1,63 @@ + + * @package Text_Diff + */ +class Text_Diff_Engine_xdiff { + + /** + */ + function diff($from_lines, $to_lines) + { + array_walk($from_lines, array('Text_Diff', 'trimNewlines')); + array_walk($to_lines, array('Text_Diff', 'trimNewlines')); + + /* Convert the two input arrays into strings for xdiff processing. */ + $from_string = implode("\n", $from_lines); + $to_string = implode("\n", $to_lines); + + /* Diff the two strings and convert the result to an array. */ + $diff = xdiff_string_diff($from_string, $to_string, count($to_lines)); + $diff = explode("\n", $diff); + + /* Walk through the diff one line at a time. We build the $edits + * array of diff operations by reading the first character of the + * xdiff output (which is in the "unified diff" format). + * + * Note that we don't have enough information to detect "changed" + * lines using this approach, so we can't add Text_Diff_Op_changed + * instances to the $edits array. The result is still perfectly + * valid, albeit a little less descriptive and efficient. */ + $edits = array(); + foreach ($diff as $line) { + switch ($line[0]) { + case ' ': + $edits[] = &new Text_Diff_Op_copy(array(substr($line, 1))); + break; + + case '+': + $edits[] = &new Text_Diff_Op_add(array(substr($line, 1))); + break; + + case '-': + $edits[] = &new Text_Diff_Op_delete(array(substr($line, 1))); + break; + } + } + + return $edits; + } + +} diff --git a/wp-includes/Text/Diff/Renderer.php b/wp-includes/Text/Diff/Renderer.php new file mode 100644 index 000000000..5d226b4e2 --- /dev/null +++ b/wp-includes/Text/Diff/Renderer.php @@ -0,0 +1,237 @@ + $value) { + $v = '_' . $param; + if (isset($this->$v)) { + $this->$v = $value; + } + } + } + + /** + * Get any renderer parameters. + * + * @return array All parameters of this renderer object. + */ + function getParams() + { + $params = array(); + foreach (get_object_vars($this) as $k => $v) { + if ($k[0] == '_') { + $params[substr($k, 1)] = $v; + } + } + + return $params; + } + + /** + * Renders a diff. + * + * @param Text_Diff $diff A Text_Diff object. + * + * @return string The formatted output. + */ + function render($diff) + { + $xi = $yi = 1; + $block = false; + $context = array(); + + $nlead = $this->_leading_context_lines; + $ntrail = $this->_trailing_context_lines; + + $output = $this->_startDiff(); + + $diffs = $diff->getDiff(); + foreach ($diffs as $i => $edit) { + /* If these are unchanged (copied) lines, and we want to keep + * leading or trailing context lines, extract them from the copy + * block. */ + if (is_a($edit, 'Text_Diff_Op_copy')) { + /* Do we have any diff blocks yet? */ + if (is_array($block)) { + /* How many lines to keep as context from the copy + * block. */ + $keep = $i == count($diffs) - 1 ? $ntrail : $nlead + $ntrail; + if (count($edit->orig) <= $keep) { + /* We have less lines in the block than we want for + * context => keep the whole block. */ + $block[] = $edit; + } else { + if ($ntrail) { + /* Create a new block with as many lines as we need + * for the trailing context. */ + $context = array_slice($edit->orig, 0, $ntrail); + $block[] = &new Text_Diff_Op_copy($context); + } + /* @todo */ + $output .= $this->_block($x0, $ntrail + $xi - $x0, + $y0, $ntrail + $yi - $y0, + $block); + $block = false; + } + } + /* Keep the copy block as the context for the next block. */ + $context = $edit->orig; + } else { + /* Don't we have any diff blocks yet? */ + if (!is_array($block)) { + /* Extract context lines from the preceding copy block. */ + $context = array_slice($context, count($context) - $nlead); + $x0 = $xi - count($context); + $y0 = $yi - count($context); + $block = array(); + if ($context) { + $block[] = &new Text_Diff_Op_copy($context); + } + } + $block[] = $edit; + } + + if ($edit->orig) { + $xi += count($edit->orig); + } + if ($edit->final) { + $yi += count($edit->final); + } + } + + if (is_array($block)) { + $output .= $this->_block($x0, $xi - $x0, + $y0, $yi - $y0, + $block); + } + + return $output . $this->_endDiff(); + } + + function _block($xbeg, $xlen, $ybeg, $ylen, &$edits) + { + $output = $this->_startBlock($this->_blockHeader($xbeg, $xlen, $ybeg, $ylen)); + + foreach ($edits as $edit) { + switch (strtolower(get_class($edit))) { + case 'text_diff_op_copy': + $output .= $this->_context($edit->orig); + break; + + case 'text_diff_op_add': + $output .= $this->_added($edit->final); + break; + + case 'text_diff_op_delete': + $output .= $this->_deleted($edit->orig); + break; + + case 'text_diff_op_change': + $output .= $this->_changed($edit->orig, $edit->final); + break; + } + } + + return $output . $this->_endBlock(); + } + + function _startDiff() + { + return ''; + } + + function _endDiff() + { + return ''; + } + + function _blockHeader($xbeg, $xlen, $ybeg, $ylen) + { + if ($xlen > 1) { + $xbeg .= ',' . ($xbeg + $xlen - 1); + } + if ($ylen > 1) { + $ybeg .= ',' . ($ybeg + $ylen - 1); + } + + // this matches the GNU Diff behaviour + if ($xlen && !$ylen) { + $ybeg--; + } elseif (!$xlen) { + $xbeg--; + } + + return $xbeg . ($xlen ? ($ylen ? 'c' : 'd') : 'a') . $ybeg; + } + + function _startBlock($header) + { + return $header . "\n"; + } + + function _endBlock() + { + return ''; + } + + function _lines($lines, $prefix = ' ') + { + return $prefix . implode("\n$prefix", $lines) . "\n"; + } + + function _context($lines) + { + return $this->_lines($lines, ' '); + } + + function _added($lines) + { + return $this->_lines($lines, '> '); + } + + function _deleted($lines) + { + return $this->_lines($lines, '< '); + } + + function _changed($orig, $final) + { + return $this->_deleted($orig) . "---\n" . $this->_added($final); + } + +} diff --git a/wp-includes/Text/Diff/Renderer/inline.php b/wp-includes/Text/Diff/Renderer/inline.php new file mode 100644 index 000000000..b547ff236 --- /dev/null +++ b/wp-includes/Text/Diff/Renderer/inline.php @@ -0,0 +1,170 @@ +'; + + /** + * Suffix for inserted text. + */ + var $_ins_suffix = ''; + + /** + * Prefix for deleted text. + */ + var $_del_prefix = ''; + + /** + * Suffix for deleted text. + */ + var $_del_suffix = ''; + + /** + * Header for each change block. + */ + var $_block_header = ''; + + /** + * What are we currently splitting on? Used to recurse to show word-level + * changes. + */ + var $_split_level = 'lines'; + + function _blockHeader($xbeg, $xlen, $ybeg, $ylen) + { + return $this->_block_header; + } + + function _startBlock($header) + { + return $header; + } + + function _lines($lines, $prefix = ' ', $encode = true) + { + if ($encode) { + array_walk($lines, array(&$this, '_encode')); + } + + if ($this->_split_level == 'words') { + return implode('', $lines); + } else { + return implode("\n", $lines) . "\n"; + } + } + + function _added($lines) + { + array_walk($lines, array(&$this, '_encode')); + $lines[0] = $this->_ins_prefix . $lines[0]; + $lines[count($lines) - 1] .= $this->_ins_suffix; + return $this->_lines($lines, ' ', false); + } + + function _deleted($lines, $words = false) + { + array_walk($lines, array(&$this, '_encode')); + $lines[0] = $this->_del_prefix . $lines[0]; + $lines[count($lines) - 1] .= $this->_del_suffix; + return $this->_lines($lines, ' ', false); + } + + function _changed($orig, $final) + { + /* If we've already split on words, don't try to do so again - just + * display. */ + if ($this->_split_level == 'words') { + $prefix = ''; + while ($orig[0] !== false && $final[0] !== false && + substr($orig[0], 0, 1) == ' ' && + substr($final[0], 0, 1) == ' ') { + $prefix .= substr($orig[0], 0, 1); + $orig[0] = substr($orig[0], 1); + $final[0] = substr($final[0], 1); + } + return $prefix . $this->_deleted($orig) . $this->_added($final); + } + + $text1 = implode("\n", $orig); + $text2 = implode("\n", $final); + + /* Non-printing newline marker. */ + $nl = "\0"; + + /* We want to split on word boundaries, but we need to + * preserve whitespace as well. Therefore we split on words, + * but include all blocks of whitespace in the wordlist. */ + $diff = new Text_Diff($this->_splitOnWords($text1, $nl), + $this->_splitOnWords($text2, $nl)); + + /* Get the diff in inline format. */ + $renderer = new Text_Diff_Renderer_inline(array_merge($this->getParams(), + array('split_level' => 'words'))); + + /* Run the diff and get the output. */ + return str_replace($nl, "\n", $renderer->render($diff)) . "\n"; + } + + function _splitOnWords($string, $newlineEscape = "\n") + { + // Ignore \0; otherwise the while loop will never finish. + $string = str_replace("\0", '', $string); + + $words = array(); + $length = strlen($string); + $pos = 0; + + while ($pos < $length) { + // Eat a word with any preceding whitespace. + $spaces = strspn(substr($string, $pos), " \n"); + $nextpos = strcspn(substr($string, $pos + $spaces), " \n"); + $words[] = str_replace("\n", $newlineEscape, substr($string, $pos, $spaces + $nextpos)); + $pos += $spaces + $nextpos; + } + + return $words; + } + + function _encode(&$string) + { + $string = htmlspecialchars($string); + } + +} diff --git a/wp-includes/default-filters.php b/wp-includes/default-filters.php index 857de94dd..d858b20d4 100644 --- a/wp-includes/default-filters.php +++ b/wp-includes/default-filters.php @@ -178,6 +178,7 @@ add_action('admin_print_scripts', 'wp_print_scripts', 20); add_action('init', 'smilies_init', 5); add_action( 'plugins_loaded', 'wp_maybe_load_widgets', 0 ); add_action( 'shutdown', 'wp_ob_end_flush_all', 1); +add_action( 'pre_post_update', 'wp_save_revision' ); add_action('publish_post', '_publish_post_hook', 5, 1); add_action('future_post', '_future_post_hook', 5, 2); add_action('future_page', '_future_post_hook', 5, 2); diff --git a/wp-includes/link-template.php b/wp-includes/link-template.php index 74da3e747..d9763c515 100644 --- a/wp-includes/link-template.php +++ b/wp-includes/link-template.php @@ -442,10 +442,15 @@ function get_search_comments_feed_link($search_query = '', $feed = '') { return $link; } -function get_edit_post_link( $id = 0 ) { +function get_edit_post_link( $id = 0, $context = 'display' ) { if ( !$post = &get_post( $id ) ) return; + if ( 'display' == $context ) + $action = 'action=edit&'; + else + $action = 'action=edit&'; + switch ( $post->post_type ) : case 'page' : if ( !current_user_can( 'edit_page', $post->ID ) ) @@ -459,6 +464,13 @@ function get_edit_post_link( $id = 0 ) { $file = 'media'; $var = 'attachment_id'; break; + case 'revision' : + if ( !current_user_can( 'edit_post', $post->ID ) ) + return; + $file = 'revision'; + $var = 'revision'; + $action = ''; + break; default : if ( !current_user_can( 'edit_post', $post->ID ) ) return; @@ -467,7 +479,7 @@ function get_edit_post_link( $id = 0 ) { break; endswitch; - return apply_filters( 'get_edit_post_link', get_bloginfo( 'wpurl' ) . "/wp-admin/$file.php?action=edit&$var=$post->ID", $post->ID ); + return apply_filters( 'get_edit_post_link', get_bloginfo( 'wpurl' ) . "/wp-admin/$file.php?{$action}$var=$post->ID", $post->ID ); } function edit_post_link( $link = 'Edit This', $before = '', $after = '' ) { diff --git a/wp-includes/pluggable.php b/wp-includes/pluggable.php index 6eac474f2..098eeb903 100644 --- a/wp-includes/pluggable.php +++ b/wp-includes/pluggable.php @@ -1350,4 +1350,68 @@ function wp_login($username, $password, $deprecated = '') { } endif; +if ( !function_exists( 'wp_text_diff' ) ) : +/** + * wp_text_diff() - compares two strings and outputs a human readable HTML representation of their difference + * + * Basically a wrapper for man diff(1) + * + * Must accept an optional third parameter, $args @see wp_parse_args() + * (string) title: optional. If present, titles the diff in a manner compatible with the output + * + * Must return the empty string if the two compared strings are found to be equivalent according to whatever metric + * + * @since 2.6 + * @uses Text_Diff + * @uses WP_Text_Diff_Renderer_Table + * + * @param string $left_string "old" (left) version of string + * @param string $right_string "new" (right) version of string + * @param string|array $args @see wp_parse_args() + * @return string human readable HTML of string differences. Empty string if strings are equivalent + */ +function wp_text_diff( $left_string, $right_string, $args = null ) { + $defaults = array( 'title' => '' ); + $args = wp_parse_args( $args, $defaults ); + + // PEAR Text_Diff is lame; it includes things from include_path rather than it's own path. + // Not sure of the ramifications of disttributing modified code. + ini_set('include_path', '.' . PATH_SEPARATOR . ABSPATH . WPINC ); + + if ( !class_exists( 'WP_Text_Diff_Renderer_Table' ) ) + require( ABSPATH . WPINC . '/wp-diff.php' ); + + // Normalize whitespace + $left_string = trim($left_string); + $right_string = trim($right_string); + $left_string = str_replace("\r", "\n", $left_string); + $right_string = str_replace("\r", "\n", $right_string); + $left_string = preg_replace( array( '/\n+/', '/[ \t]+/' ), array( "\n", ' ' ), $left_string ); + $right_string = preg_replace( array( '/\n+/', '/[ \t]+/' ), array( "\n", ' ' ), $right_string ); + + $left_lines = split("\n", $left_string); + $right_lines = split("\n", $right_string); + + $text_diff = new Text_Diff($left_lines, $right_lines); + $renderer = new WP_Text_Diff_Renderer_Table(); + $diff = $renderer->render($text_diff); + + ini_restore('include_path'); + + if ( !$diff ) + return ''; + + $r = "\n"; + $r .= ""; + + if ( $args['title'] ) + $r .= "\n"; + + $r .= "\n$diff\n\n"; + $r .= "
$args[title]
"; + + return $r; +} +endif; + ?> diff --git a/wp-includes/post-template.php b/wp-includes/post-template.php index 47c1068ad..1aa81ae45 100644 --- a/wp-includes/post-template.php +++ b/wp-includes/post-template.php @@ -564,4 +564,136 @@ function is_page_template($template = '') { return false; } -?> +/** + * wp_post_revision_time() - returns formatted datetimestamp of a revision + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses wp_get_revision() + * @uses date_i18n() + * + * @param int|object $revision revision ID or revision object + * @return string i18n formatted datetimestamp or localized 'Corrent Revision' + */ +function wp_post_revision_time( $revision ) { + if ( !$revision = wp_get_revision( $revision ) ) { + if ( $revision = get_post( $revision ) ) + return __( 'Current Revision' ); + return $revision; + } + + $datef = _c( 'j F, Y @ G:i|revision date format'); + return date_i18n( $datef, strtotime( $revision->post_date_gmt . ' +0000' ) ); +} + +/** + * wp_list_post_revisions() - echoes list of a post's revisions + * + * Can output either a UL with edit links or a TABLE with diff interface, and restore action links + * + * Second argument controls parameters: + * (bool) parent : include the parent (the "Current Revision") in the list + * (string) format : 'list' or 'form-table'. 'list' outputs UL, 'form-table' outputs TABLE with UI + * (int) right : what revision is currently being viewed - used in form-table format + * (int) left : what revision is currently being diffed against right - used in form-table format + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses wp_get_post_revisions() + * @uses wp_post_revision_time() + * @uses get_edit_post_link() + * @uses get_author_name() + * + * @param int|object $post_id post ID or post object + * @param string|array $args see description @see wp_parse_args() + */ +function wp_list_post_revisions( $post_id = 0, $args = null ) { // TODO? split into two functions (list, form-table) ? + if ( !$post = get_post( $post_id ) ) + return; + + if ( !$revisions = wp_get_post_revisions( $post->ID ) ) + return; + + $defaults = array( 'parent' => false, 'right' => false, 'left' => false, 'format' => 'list' ); + extract( wp_parse_args( $args, $defaults ), EXTR_SKIP ); + + $titlef = _c( '%1$s by %2$s|post revision 1:datetime, 2:name' ); + + if ( $parent ) + array_unshift( $revisions, $post ); + + $rows = ''; + $class = false; + foreach ( $revisions as $revision ) { + $date = wp_post_revision_time( $revision ); + if ( $link = get_edit_post_link( $revision->ID ) ) + $date = "$date"; + $name = get_author_name( $revision->post_author ); + + if ( 'form-table' == $format ) { + if ( $left ) + $old_checked = $left == $revision->ID ? ' checked="checked"' : ''; + else + $old_checked = $new_checked ? ' checked="checked"' : ''; + $new_checked = $right == $revision->ID ? ' checked="checked"' : ''; + + $class = $class ? '' : " class='alternate'"; + + if ( $post->ID != $revision->ID && current_user_can( 'edit_post', $post->ID ) ) + $actions = 'ID|$revision->ID" ) . '">' . __( 'Restore' ) . ''; + else + $actions = ''; + + $rows .= "\n"; + $rows .= "\t\n"; + $rows .= "\t$date\n"; + $rows .= "\t$name\n"; + $rows .= "\t$actions\n"; + $rows .= "\n"; + } else { + $rows .= "\t
  • " . sprintf( $titlef, $date, $name ). "
    • \n"; + } + } + + if ( 'form-table' == $format ) : ?> + +
    + +
    +
    + +
    +
    + +
    + + + + + + + + + + + + + + + + + +
    + +\n"; + echo $rows; + echo ""; + endif; + +} diff --git a/wp-includes/post.php b/wp-includes/post.php index 27d346258..1fe9da464 100644 --- a/wp-includes/post.php +++ b/wp-includes/post.php @@ -2990,4 +2990,232 @@ function _get_post_ancestors(&$_post) { } } +/* Post Revisions */ + +/** + * _wp_revision_fields() - determines which fields of posts are to be saved in revisions + * + * Does two things. If passed a postn *array*, it will return a post array ready to be + * insterted into the posts table as a post revision. + * Otherwise, returns an array whose keys are the post fields to be saved post revisions. + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @param array $post optional a post array to be processed for insertion as a post revision + * @return array post array ready to be inserted as a post revision or array of fields that can be versioned + */ +function _wp_revision_fields( $post = null ) { + static $fields = false; + + if ( !$fields ) { + // Allow these to be versioned + $fields = array( + 'post_title' => __( 'Title' ), + 'post_author' => __( 'Author' ), + 'post_content' => __( 'Content' ), + 'post_excerpt' => __( 'Excerpt' ), + ); + + // WP uses these internally either in versioning or elsewhere - they cannot be versioned + foreach ( array( 'ID', 'post_name', 'post_parent', 'post_date', 'post_date_gmt', 'post_status', 'post_type', 'comment_count' ) as $protect ) + unset( $fields[$protect] ); + } + + if ( !is_array($post) ) + return $fields; + + $return = array(); + foreach ( array_intersect( array_keys( $post ), array_keys( $fields ) ) as $field ) + $return[$field] = $post[$field]; + + $return['post_parent'] = $post['ID']; + $return['post_status'] = 'inherit'; + $return['post_type'] = 'revision'; + $return['post_name'] = "$post[ID]-revision"; + $return['post_date'] = $post['post_modified']; + $return['post_date_gmt'] = $post['post_modified_gmt']; + + return $return; +} + +/** + * wp_save_revision() - Saves an already existing post as a post revision. Typically used immediately prior to post updates. + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses _wp_put_revision() + * + * @param int $post_id The ID of the post to save as a revision + * @return mixed null or 0 if error, new revision ID if success + */ +function wp_save_revision( $post_id ) { + // TODO: rework autosave to use special type of post revision + if ( @constant( 'DOING_AUTOSAVE' ) ) + return; + + if ( !$post = get_post( $post_id, ARRAY_A ) ) + return; + + // TODO: open this up for pages also + if ( 'post' != $post->post_type ) + retun; + + return _wp_put_revision( $post ); +} + +/** + * _wp_put_revision() - Inserts post data into the posts table as a post revision + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses wp_insert_post() + * + * @param int|object|array $post post ID, post object OR post array + * @return mixed null or 0 if error, new revision ID if success + */ +function _wp_put_revision( $post = null ) { + if ( is_object($post) ) + $post = get_object_vars( $post ); + elseif ( !is_array($post) ) + $post = get_post($post, ARRAY_A); + + if ( !$post || empty($post['ID']) ) + return; + + if ( isset($post['post_type']) && 'revision' == $post_post['type'] ) + return new WP_Error( 'post_type', __( 'Cannot create a revision of a revision' ) ); + + $post = _wp_revision_fields( $post ); + + if ( $revision_id = wp_insert_post( $post ) ) + do_action( '_wp_put_revision', $revision_id ); + + return $revision_id; +} + +/** + * wp_get_revision() - Gets a post revision + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses get_post() + * + * @param int|object $post post ID or post object + * @param $output optional OBJECT, ARRAY_A, or ARRAY_N + * @param string $filter optional sanitation filter. @see sanitize_post() + * @return mixed null if error or post object if success + */ +function &wp_get_revision(&$post, $output = OBJECT, $filter = 'raw') { + $null = null; + if ( !$revision = get_post( $post, OBJECT, $filter ) ) + return $revision; + if ( 'revision' !== $revision->post_type ) + return $null; + + if ( $output == OBJECT ) { + return $revision; + } elseif ( $output == ARRAY_A ) { + $_revision = get_object_vars($revision); + return $_revision; + } elseif ( $output == ARRAY_N ) { + $_revision = array_values(get_object_vars($revision)); + return $_revision; + } + + return $revision; +} + +/** + * wp_restore_revision() - Restores a post to the specified revision + * + * Can restore a past using all fields of the post revision, or only selected fields. + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses wp_get_revision() + * @uses wp_update_post() + * + * @param int|object $revision_id revision ID or revision object + * @param array $fields optional What fields to restore from. Defaults to all. + * @return mixed null if error, false if no fields to restore, (int) post ID if success + */ +function wp_restore_revision( $revision_id, $fields = null ) { + if ( !$revision = wp_get_revision( $revision_id, ARRAY_A ) ) + return $revision; + + if ( !is_array( $fields ) ) + $fields = array_keys( _wp_revision_fields() ); + + $update = array(); + foreach( array_intersect( array_keys( $revision ), $fields ) as $field ) + $update[$field] = $revision[$field]; + + if ( !$update ) + return false; + + $update['ID'] = $revision['post_parent']; + + if ( $post_id = wp_update_post( $update ) ) + do_action( 'wp_restore_revision', $post_id, $revision['ID'] ); + + return $post_id; +} + +/** + * wp_delete_revision() - Deletes a revision. + * + * Deletes the row from the posts table corresponding to the specified revision + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses wp_get_revision() + * @uses wp_delete_post() + * + * @param int|object $revision_id revision ID or revision object + * @param array $fields optional What fields to restore from. Defaults to all. + * @return mixed null if error, false if no fields to restore, (int) post ID if success + */ +function wp_delete_revision( $revision_id ) { + if ( !$revision = wp_get_revision( $revision_id ) ) + return $revision; + + if ( $delete = wp_delete_post( $revision->ID ) ) + do_action( 'wp_delete_revision', $revision->ID, $revision ); + + return $delete; +} + +/** + * wp_get_post_revisions() - Returns all revisions of specified post + * + * @package WordPress + * @subpackage Post Revisions + * @since 2.6 + * + * @uses get_children() + * + * @param int|object $post_id post ID or post object + * @return array empty if no revisions + */ +function wp_get_post_revisions( $post_id = 0 ) { + if ( ( !$post = get_post( $post_id ) ) || empty( $post->ID ) ) + return array(); + + if ( !$revisions = get_children( array( 'post_parent' => $post->ID, 'post_type' => 'revision' ) ) ) + return array(); + return $revisions; +} + ?> diff --git a/wp-includes/wp-diff.php b/wp-includes/wp-diff.php new file mode 100644 index 000000000..f60614ad8 --- /dev/null +++ b/wp-includes/wp-diff.php @@ -0,0 +1,318 @@ +$parent( $params ); + } + + function _startBlock( $header ) { + return ''; + } + + function _lines( $lines, $prefix=' ' ) { + } + + // HTML-escape parameter before calling this + function addedLine( $line ) { + return "+{$line}"; + } + + // HTML-escape parameter before calling this + function deletedLine( $line ) { + return "-{$line}"; + } + + // HTML-escape parameter before calling this + function contextLine( $line ) { + return " {$line}"; + } + + function emptyLine() { + return ' '; + } + + function _added( $lines, $encode = true ) { + $r = ''; + foreach ($lines as $line) { + if ( $encode ) + $line = htmlspecialchars( $line ); + $r .= '' . $this->emptyLine() . $this->addedLine( $line ) . "\n"; + } + return $r; + } + + function _deleted( $lines, $encode = true ) { + $r = ''; + foreach ($lines as $line) { + if ( $encode ) + $line = htmlspecialchars( $line ); + $r .= '' . $this->deletedLine( $line ) . $this->emptyLine() . "\n"; + } + return $r; + } + + function _context( $lines, $encode = true ) { + $r = ''; + foreach ($lines as $line) { + if ( $encode ) + $line = htmlspecialchars( $line ); + $r .= '' . + $this->contextLine( $line ) . $this->contextLine( $line ) . "\n"; + } + return $r; + } + + // Process changed lines to do word-by-word diffs for extra highlighting (TRAC style) + // sometimes these lines can actually be deleted or added rows - we do additional processing + // to figure that out + function _changed( $orig, $final ) { + $r = ''; + + // Does the aforementioned additional processing + // *_matches tell what rows are "the same" in orig and final. Those pairs will be diffed to get word changes + // match is numeric: an index in other column + // match is 'X': no match. It is a new row + // *_rows are column vectors for the orig column and the final column. + // row >= 0: an indix of the $orig or $final array + // row < 0: a blank row for that column + list($orig_matches, $final_matches, $orig_rows, $final_rows) = $this->interleave_changed_lines( $orig, $final ); + + + // These will hold the word changes as determined by an inline diff + $orig_diffs = array(); + $final_diffs = array(); + + // Compute word diffs for each matched pair using the inline diff + foreach ( $orig_matches as $o => $f ) { + if ( is_numeric($o) && is_numeric($f) ) { + $text_diff = new Text_Diff( 'auto', array( array($orig[$o]), array($final[$f]) ) ); + $renderer = new $this->inline_diff_renderer; + $diff = $renderer->render( $text_diff ); + + // If they're too different, don't include any or + if ( $diff_count = preg_match_all( '!(.*?|.*?)!', $diff, $diff_matches ) ) { + // length of all text between or + $stripped_matches = strlen(strip_tags( join(' ', $diff_matches[0]) )); + // since we count lengith of text between or (instead of picking just one), + // we double the length of chars not in those tags. + $stripped_diff = strlen(strip_tags( $diff )) * 2 - $stripped_matches; + $diff_ratio = $stripped_matches / $stripped_diff; + if ( $diff_ratio > $this->_diff_threshold ) + continue; // Too different. Don't save diffs. + } + + // Un-inline the diffs by removing del or ins + $orig_diffs[$o] = preg_replace( '|.*?|', '', $diff ); + $final_diffs[$f] = preg_replace( '|.*?|', '', $diff ); + } + } + + foreach ( array_keys($orig_rows) as $row ) { + // Both columns have blanks. Ignore them. + if ( $orig_rows[$row] < 0 && $final_rows[$row] < 0 ) + continue; + + // If we have a word based diff, use it. Otherwise, use the normal line. + $orig_line = isset($orig_diffs[$orig_rows[$row]]) + ? $orig_diffs[$orig_rows[$row]] + : htmlspecialchars($orig[$orig_rows[$row]]); + $final_line = isset($final_diffs[$final_rows[$row]]) + ? $final_diffs[$final_rows[$row]] + : htmlspecialchars($final[$final_rows[$row]]); + + if ( $orig_rows[$row] < 0 ) { // Orig is blank. This is really an added row. + $r .= $this->_added( array($final_line), false ); + } elseif ( $final_rows[$row] < 0 ) { // Final is blank. This is really a deleted row. + $r .= $this->_deleted( array($orig_line), false ); + } else { // A true changed row. + $r .= '' . $this->deletedLine( $orig_line ) . $this->addedLine( $final_line ) . "\n"; + } + } + + return $r; + } + + // Takes changed blocks and matches which rows in orig turned into which rows in final. + // Returns + // *_matches ( which rows match with which ) + // *_rows ( order of rows in each column interleaved with blank rows as necessary ) + function interleave_changed_lines( $orig, $final ) { + + // Contains all pairwise string comparisons. Keys are such that this need only be a one dimensional array. + $matches = array(); + foreach ( array_keys($orig) as $o ) { + foreach ( array_keys($final) as $f ) { + $matches["$o,$f"] = $this->compute_string_distance( $orig[$o], $final[$f] ); + } + } + asort($matches); // Order by string distance. + + $orig_matches = array(); + $final_matches = array(); + + foreach ( $matches as $keys => $difference ) { + list($o, $f) = explode(',', $keys); + $o = (int) $o; + $f = (int) $f; + + // Already have better matches for these guys + if ( isset($orig_matches[$o]) && isset($final_matches[$f]) ) + continue; + + // First match for these guys. Must be best match + if ( !isset($orig_matches[$o]) && !isset($final_matches[$f]) ) { + $orig_matches[$o] = $f; + $final_matches[$f] = $o; + continue; + } + + // Best match of this final is already taken? Must mean this final is a new row. + if ( isset($orig_matches[$o]) ) + $final_matches[$f] = 'x'; + + // Best match of this orig is already taken? Must mean this orig is a deleted row. + elseif ( isset($final_matches[$f]) ) + $orig_matches[$o] = 'x'; + } + + // We read the text in this order + ksort($orig_matches); + ksort($final_matches); + + + // Stores rows and blanks for each column. + $orig_rows = $orig_rows_copy = array_keys($orig_matches); + $final_rows = array_keys($final_matches); + + // Interleaves rows with blanks to keep matches aligned. + // We may end up with some extraneous blank rows, but we'll just ignore them later. + foreach ( $orig_rows_copy as $orig_row ) { + $final_pos = array_search($orig_matches[$orig_row], $final_rows, true); + $orig_pos = (int) array_search($orig_row, $orig_rows, true); + + if ( false === $final_pos ) { // This orig is paired with a blank final. + array_splice( $final_rows, $orig_pos, 0, -1 ); + } elseif ( $final_pos < $orig_pos ) { // This orig's match is up a ways. Pad final with blank rows. + $diff_pos = $final_pos - $orig_pos; + while ( $diff_pos < 0 ) + array_splice( $final_rows, $orig_pos, 0, $diff_pos++ ); + } elseif ( $final_pos > $orig_pos ) { // This orig's match is down a ways. Pad orig with blank rows. + $diff_pos = $orig_pos - $final_pos; + while ( $diff_pos < 0 ) + array_splice( $orig_rows, $orig_pos, 0, $diff_pos++ ); + } + } + + + // Pad the ends with blank rows if the columns aren't the same length + $diff_count = count($orig_rows) - count($final_rows); + if ( $diff_count < 0 ) { + while ( $diff_count < 0 ) + array_push($orig_rows, $diff_count++); + } elseif ( $diff_count > 0 ) { + $diff_count = -1 * $diff_count; + while ( $diff_count < 0 ) + array_push($final_rows, $diff_count++); + } + + return array($orig_matches, $final_matches, $orig_rows, $final_rows); + +/* + // Debug + echo "\n\n\n\n\n"; + + echo "-- DEBUG Matches: Orig -> Final --"; + + foreach ( $orig_matches as $o => $f ) { + echo "\n\n\n\n\n"; + echo "ORIG: $o, FINAL: $f\n"; + var_dump($orig[$o],$final[$f]); + } + echo "\n\n\n\n\n"; + + echo "-- DEBUG Matches: Final -> Orig --"; + + foreach ( $final_matches as $f => $o ) { + echo "\n\n\n\n\n"; + echo "FINAL: $f, ORIG: $o\n"; + var_dump($final[$f],$orig[$o]); + } + echo "\n\n\n\n\n"; + + echo "-- DEBUG Rows: Orig -- Final --"; + + echo "\n\n\n\n\n"; + foreach ( $orig_rows as $row => $o ) { + if ( $o < 0 ) + $o = 'X'; + $f = $final_rows[$row]; + if ( $f < 0 ) + $f = 'X'; + echo "$o -- $f\n"; + } + echo "\n\n\n\n\n"; + + echo "-- END DEBUG --"; + + echo "\n\n\n\n\n"; + + return array($orig_matches, $final_matches, $orig_rows, $final_rows); +*/ + } + + + // Computes a number that is intended to reflect the "distance" between two strings. + function compute_string_distance( $string1, $string2 ) { + // Vectors containing character frequency for all chars in each string + $chars1 = count_chars($string1); + $chars2 = count_chars($string2); + + // L1-norm of difference vector. + $difference = array_sum( array_map( array(&$this, 'difference'), $chars1, $chars2 ) ); + + // $string1 has zero length? Odd. Give huge penalty by not dividing. + if ( !$string1 ) + return $difference; + + // Return distance per charcter (of string1) + return $difference / strlen($string1); + } + + function difference( $a, $b ) { + return abs( $a - $b ); + } + +} + +// Better word splitting than the PEAR package provides +class WP_Text_Diff_Renderer_inline extends Text_Diff_Renderer_inline { + + function _splitOnWords($string, $newlineEscape = "\n") { + $string = str_replace("\0", '', $string); + $words = preg_split( '/([^\w])/', $string, -1, PREG_SPLIT_DELIM_CAPTURE ); + $words = str_replace( "\n", $newlineEscape, $words ); + return $words; + } + +} + +?>