From a2919ecd3bef14916efb949d7de61334e60b847c Mon Sep 17 00:00:00 2001
From: Klaus Leithoff <kl@stylite.de>
Date: Tue, 31 Aug 2010 14:23:58 +0000
Subject: [PATCH] improve convertHTMLToText, for toggle editor and mail2infolog
 process; better control of insert Signature at Top. Only insert the
 signature, when the compose window is opened the first time
 (compose/reply/forward), dont do it if sending failes, and the compose window
 is refreshed (in general when mySigID is set)

---
 infolog/inc/class.infolog_ui.inc.php   |  1 +
 phpgwapi/inc/class.html.inc.php        | 25 ++++++++++++++++++
 phpgwapi/inc/class.translation.inc.php | 35 +++++++++++++++++++++-----
 3 files changed, 55 insertions(+), 6 deletions(-)

diff --git a/infolog/inc/class.infolog_ui.inc.php b/infolog/inc/class.infolog_ui.inc.php
index 6fcf638514..9a0b520f17 100644
--- a/infolog/inc/class.infolog_ui.inc.php
+++ b/infolog/inc/class.infolog_ui.inc.php
@@ -1665,6 +1665,7 @@ class infolog_ui
 
 			if ($bodyParts[$i]['mimeType'] == 'text/html') {
 				// convert HTML to text, as we dont want HTML in infologs
+				$newBody = html::purify($newBody);
 				$newBody = $bofelamimail->convertHTMLToText($newBody,true);
 				$bofelamimail->getCleanHTML($newBody); // new Body passed by reference
 				$message .= $newBody;
diff --git a/phpgwapi/inc/class.html.inc.php b/phpgwapi/inc/class.html.inc.php
index 6148051de9..1fb6200132 100644
--- a/phpgwapi/inc/class.html.inc.php
+++ b/phpgwapi/inc/class.html.inc.php
@@ -1443,5 +1443,30 @@ class html
 			}
 		}
 	}
+
+	/**
+	 * split html by PRE tag, return array with all content pre-sections isolated in array elements
+	 * @author Leithoff, Klaus
+	 * @param string html
+	 * @return mixed array of parts or unaffected html 
+	 */
+	static function splithtmlByPRE($html)
+	{
+		if (($pos = stripos($html,'<pre>')) === false)
+		{
+			return $html;
+		}
+		$html2ret[] = substr($html,0,$pos);
+		while ($pos!==false)
+		{
+			$endofpre = stripos($html,'</pre>',$pos);
+			$length = $endofpre-$pos+6;
+			$html2ret[] = substr($html,$pos,$length);
+			$pos =  stripos($html,'<pre>', $endofpre+6);
+			$html2ret[] = ($pos ? substr($html,$endofpre+6,$pos-($endofpre+6)): substr($html,$endofpre+6));
+			//$pos=false;
+		}
+		return $html2ret;
+	}
 }
 html::_init_static();
diff --git a/phpgwapi/inc/class.translation.inc.php b/phpgwapi/inc/class.translation.inc.php
index de0a762e7d..6b7cc45db1 100644
--- a/phpgwapi/inc/class.translation.inc.php
+++ b/phpgwapi/inc/class.translation.inc.php
@@ -1139,6 +1139,7 @@ class translation
 			'@&(pound|#163);@i',              //   Pound
 			'@&(copy|#169);@i',               //   Copyright
 			'@&(reg|#174);@i',                //   Registered
+			'@&#39;@i',                       //   singleQuote
 		);
 		$Replace = array ('',
 			'"',
@@ -1151,11 +1152,26 @@ class translation
 			chr(163),
 			chr(169),
 			chr(174),
+			"'",
 		);
 		$_html = preg_replace($Rules, $Replace, $_html);
 
-		//   removing carriage return linefeeds
-		if ($stripcrl === true ) $_html = preg_replace('@(\r\n)@i',' ',$_html);
+		//   removing carriage return linefeeds, preserve those enclosed in <pre> </pre> tags
+		if ($stripcrl === true ) 
+		{
+			if (stripos($_html,'<pre>')!==false)
+			{
+				$contentArr = html::splithtmlByPRE($_html);
+				foreach ($contentArr as $k =>&$elem)
+				{
+					if (stripos($elem,'<pre>')===false) 
+					{
+						$elem = preg_replace('@(\r\n)@i',' ',$elem);
+					}
+				}
+				$_html = implode('',$contentArr);
+			}
+		}
 		$tags = array (
 			0 => '~<h[123][^>]*>\r*\n*~si',
 			1 => '~<h[456][^>]*>\r*\n*~si',
@@ -1168,6 +1184,9 @@ class translation
 			8 => '~<div[^>]*>\r*\n*~si',
 			9 => '~<hr[^>]*>\r*\n*~si',
 			10 => '/<blockquote type="cite">/',
+			11 => '/<blockquote>/',
+			12 => '~</blockquote>~si',
+			13 => '~<blockquote[^>]*>~si',
 		);
 		$Replace = array (
 			0 => "\r\n",
@@ -1181,6 +1200,9 @@ class translation
 			8 => "\r\n",
 			9 => "\r\n__________________________________________________\r\n",
 			10 => '#blockquote#type#cite#',
+			11 => '#blockquote#type#cite#',
+			12 => '#blockquote#end#cite#',
+			13 => '#blockquote#type#cite#',
 		);
 		$_html = preg_replace($tags,$Replace,$_html);
 		$_html = preg_replace('~</t(d|h)>\s*<t(d|h)[^>]*>~si',' - ',$_html);
@@ -1200,7 +1222,7 @@ class translation
 		#$_html = preg_replace('~^\s+~m','',$_html);
 		#$_html = preg_replace('~\s+$~m','',$_html);
 		// restoring the preserved blockquote
-		$_html = preg_replace('~#blockquote#type#cite#~s','<blockquote type="cite">',$_html);
+		//$_html = str_replace('#blockquote#type#cite#','<blockquote type="cite">',$_html);
 
 
 		$_html = html_entity_decode($_html, ENT_COMPAT, $displayCharset);
@@ -1214,14 +1236,15 @@ class translation
 			$indent = 0;
 			$indentString = '';
 
-			$quoteParts = preg_split('/<blockquote type="cite">/', $_html, -1, PREG_SPLIT_OFFSET_CAPTURE);
-
+			//$quoteParts = preg_split('/<blockquote type="cite">/', $_html, -1, PREG_SPLIT_OFFSET_CAPTURE);
+			$quoteParts = preg_split('/#blockquote#type#cite#/', $_html, -1, PREG_SPLIT_OFFSET_CAPTURE);
 			foreach($quoteParts as $quotePart) {
 				if($quotePart[1] > 0) {
 					$indent++;
 					$indentString .= '>';
 				}
-				$quoteParts2 = preg_split('/<\/blockquote>/', $quotePart[0], -1, PREG_SPLIT_OFFSET_CAPTURE);
+				//$quoteParts2 = preg_split('/<\/blockquote>/', $quotePart[0], -1, PREG_SPLIT_OFFSET_CAPTURE);
+				$quoteParts2 = preg_split('/#blockquote#end#cite#/', $quotePart[0], -1, PREG_SPLIT_OFFSET_CAPTURE);
 
 				foreach($quoteParts2 as $quotePart2) {
 					if($quotePart2[1] > 0) {