tonyair/silverstripe-reports
1
0
Fork 0
mirror of https://github.com/silverstripe/silverstripe-reports synced 2024-10-22 11:05:53 +02:00
Code Issues Projects Releases Wiki Activity

FEATURE: Add HTMLCleaner abstract class, and Diff::cleanHTML() (from r98469)

Browse Source 
git-svn-id: svn://svn.silverstripe.com/silverstripe/open/modules/cms/trunk@105863 467b73ca-7a2a-4603-9d3b-597d59a354a9
This commit is contained in:
Ingo Schommer 2010-05-28 02:38:19 +00:00
parent d98e20f5e6
commit b27f6649da
2 changed files with 48 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
code/Diff.php
View File

@ -529,6 +529,8 @@ class _DiffEngine
*/
class Diff
{
public static $html_cleaner_class = null;
var $edits;
/**
@ -660,6 +662,36 @@ class Diff
/**
* Attempt to clean invalid HTML, which messes up diffs.
* This checks for various methods and cleans code if possible.
*
* NB: By default, only extremely simple tidying is performed,
* by passing through DomDocument::loadHTML and saveXML
* You will either need to install the php_tidy module
* See: http://www.php.net/manual/en/tidy.installation.php
* See also: http://htmlpurifier.org
*
* @param string $content HTML content
* @param object $cleaner Optional instance of a HTMLCleaner class to
* use, overriding self::$html_cleaner_class
*/
static function cleanHTML($content, $cleaner=null) {
if (!$cleaner)) {
if (class_exists(self::$html_cleaner_class)) {
$cleaner = new self::$html_cleaner_class;
}
}
if ($cleaner) {
$content = $cleaner->cleanHTML($content);
} else {
// At most basic level of cleaning, use DOMDocument to save valid XML.
$doc = new SS_HTMLValue($content);
$content = $doc->getContent();
}
return $content;
}
static function compareHTML($from, $to) {
// First split up the content into words and tags
$set1 = self::getHTMLChunks($from);
@ -750,7 +782,7 @@ class Diff
}
}
// echo "<p>" . htmlentities($content) . "</p>";
return $content;
return self::cleanHTML($content);
}
static function getHTMLChunks($content) {
$content = str_replace(array("&nbsp;","<", ">"),array(" "," <", "> "),$content);

14
code/HTMLCleaner.php Normal file
View File

@ -0,0 +1,14 @@
<?php
/**
* Base class for HTML cleaning classes.
*/
abstract class HTMLCleaner extends Object {
/**
* Passed $content, return HTML that has been tidied.
* @return string $content HTML, tidied
*/
public abstract function cleanHTML($content);
}
?>