FUDforum: Bug Reports » Maillist.php patch for pure html mails

Home » FUDforum Development » Bug Reports » Maillist.php patch for pure html mails

Show: Today's Messages :: Polls :: Message Navigator

Maillist.php patch for pure html mails [message #159246]

Fri, 15 May 2009 08:26

Peter Vendike

Messages: 65
Registered: February 2009
Location: Denmark

Karma: 0

Member
Translator

In (rare) cases where the imported mail has no 'text/plain' alternative and is pure html, the body-text showed up as one long line, barely readable (of cause only if 'Allow HTML in Mailing List Messages' is off).

This small patch saves the line breaks and the text formatting is kept for the imported message.

*** maillist.php	2009-05-14 17:39:33.000000000 +0200
--- maillist_pat.php	2009-05-15 09:44:54.000000000 +0200
***************
*** 121,126 ****
--- 121,129 ----
  
  			case 'text/html':
  				$this->decode_message_body();
+ 				// We need to keep line breaks for textformatting befor using 'strip_tags'
+ 				$l_breaks = array("<br />", "</p>", "<p>", "<br>" );
+ 				$this->body = str_replace($l_breaks, "\n", $this->body);
  				$this->body = (!$html ? strip_tags($this->body) : $this->body);
  				break;

[Updated on: Fri, 15 May 2009 08:55]

Report message to a moderator

Re: Maillist.php patch for pure html mails [message #159247 is a reply to message #159246]

Fri, 15 May 2009 10:21

naudefj

Messages: 3771
Registered: December 2004

Karma: 28

Senior Member
Administrator
Core Developer

Will this also work:

Index: maillist.php
===================================================================
RCS file: /forum21/install/forum_data/scripts/maillist.php,v
retrieving revision 1.78
diff -u -r1.78 maillist.php
--- maillist.php        3 May 2009 18:43:45 -0000       1.78
+++ maillist.php        15 May 2009 10:20:02 -0000
@@ -121,7 +121,11 @@

                        case 'text/html':
                                $this->decode_message_body();
-                               $this->body = (!$html ? strip_tags($this->body) : $this->body);
+                               if (!$html) {
+                                       $this->body = str_ireplace(array('<br />', '<br>'), "\n", $this->body);
+                                       $this->body = str_ireplace(array('<p>', '</p>'),  "\n\n", $this->body);
+                                       $this->body = strip_tags($this->body);
+                               }
                                break;

                        case 'multipart/parallel': // Apparently same as multipart/mixed but order of body parts does not matter

Report message to a moderator

Re: Maillist.php patch for pure html mails [message #159251 is a reply to message #159247]

Sat, 16 May 2009 06:55

naudefj

Messages: 3771
Registered: December 2004

Karma: 28

Senior Member
Administrator
Core Developer

Many thanks! Your patch was committed - see http://cvs.prohost.org/c/index.cgi/FUDforum/chngview?cn=11926

Report message to a moderator

Re: Maillist.php patch for pure html mails [message #159253 is a reply to message #159251]

Sat, 16 May 2009 13:02

Peter Vendike

Messages: 65
Registered: February 2009
Location: Denmark

Karma: 0

Member
Translator

Hej Frank

I tried your patch today, the "\n\n" for "<p>" looks more like the original html, so i think we could use your model, still keeping only 1 "\n" for "</p>" , and i also don't believe its necessary to ask for html=true ones again.

Peter

				$this->decode_message_body();
				$this->body = str_replace("<p>", "\n\n", $this->body);
				$this->body = str_replace(array("<br />", "</p>", "<br>" ), "\n", $this->body);
				$this->body = strip_tags($this->body);
				break;

[Updated on: Sat, 16 May 2009 13:05]

Report message to a moderator

Re: Maillist.php patch for pure html mails [message #159254 is a reply to message #159253]

Sat, 16 May 2009 13:47

naudefj

Messages: 3771
Registered: December 2004

Karma: 28

Senior Member
Administrator
Core Developer

Are you sure we can remove the HTML test? The function can be called with html=0 or 1.

Also, note the str_ireplace instead of str_replace.

Best regards.

Frank

Report message to a moderator

Re: Maillist.php patch for pure html mails [message #159255 is a reply to message #159254]

Sat, 16 May 2009 14:01

Peter Vendike

Messages: 65
Registered: February 2009
Location: Denmark

Karma: 0

Member
Translator

Yes I was just thinking the same, if that is used for the multipart bodies? I just can't deduct that.

Just to say, it work for some multipart test messages without that test:

$this->body = (!$html ? strip_tags($this->body) : $this->body);

else one could do something like $html => break earlier?

also, i don't now if these tags are sometimes used in capitals, so we need the case-insensitive version?

Peter

[Updated on: Sat, 16 May 2009 14:35]

Report message to a moderator

Re: Maillist.php patch for pure html mails [message #159256 is a reply to message #159255]

Sat, 16 May 2009 14:47

Peter Vendike

Messages: 65
Registered: February 2009
Location: Denmark

Karma: 0

Member
Translator

So I think I accept this version:

+                               if (!$html) {
+                                       $this->body = str_ireplace(array('<br />', '<br>', '</p>'), "\n", $this->body);
+                                       $this->body = str_ireplace('<p>', "\n\n", $this->body);
+                                       $this->body = strip_tags($this->body);
+                               }
                                break;

[Updated on: Mon, 18 May 2009 14:10]

Report message to a moderator