<?xml version="1.0" encoding="UTF-8"?> 
<?xml-stylesheet href="https://dev.horde.org/themes/horde//default/feed-rss.xsl" type="text/xsl"?> 
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"> 
 <channel> 
  <title>&quot;forward&quot; handles multipart/alternative incorrectly</title> 
  <pubDate>Fri, 10 Apr 2026 13:09:12 +0000</pubDate> 
  <link>https://bugs.horde.org/ticket/3381</link> 
  <atom:link rel="self" type="application/rss+xml" title="&quot;forward&quot; handles multipart/alternative incorrectly" href="https://bugs.horde.org/ticket/3381/rss" /> 
  <description>&quot;forward&quot; handles multipart/alternative incorrectly</description> 
 
   
   
  <item> 
   <title>Hi,



We&#039;ve noticed a problem with the way IMP forwards mes</title> 
   <description>Hi,



We&#039;ve noticed a problem with the way IMP forwards messages, and wanted to report it here so that you could look at the situation and consider changing IMP.



The problem:

- When forwarding a multipart/alternative message, IMP puts the text/plain part into the text editing field, and encourages the user to edit that text.  However, IMP adds the text/html part as an attachment to the message!

- This causes problems: we have users who will edit the text section (adding or deleting text), and send the message on.  When the message is received, a recipient&#039;s mail program will often show the text/html part instead of the text/plain part -- showing the recipient text that the sender did not intend for the recipient to see.



Users don&#039;t understand multipart/alternative and text/plain text/html distinctions.  They only understand &quot;I edited the text and I forwarded the message -- but my correspondant saw things he wasn&#039;t supposed to see!  Stop this from happening!&quot;



We can train our users to check the Attachments section on the composition page, deleting text/html attachments before they send the forwarded message.  But it would be easier for us if IMP were improved to be even more user-friendly.



Possible solutions:

I can think of two ways to adjust IMP to avoid this problem. You may be able to think of ways yourself.

- Option 1: when forwarding a multipart/alternative message, IMP should choose the text/plain part and should DISCARD the text/html part.  The user can edit the text/plain, and forward that, and they will send exactly what they thought they were sending.

- Option 2: when forwarding a multipart/alternative message, IMP should not allow the user to edit the original text.  IMP should attach the original text as a message/rfc822 attachment, and should allow the user to compose a preamble message.



The main problem is the text/plain and text/html parts of a message should never be out of sync.  Either of these options will avoid that problem; you may be able to think of your own solutions.





Thank you,



Jacob Morzinski

MIT IS&amp;T: Customer Support Services

jmorzins@mit.edu</description> 
   <pubDate>Wed, 01 Feb 2006 18:16:54 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16112</link> 
  </item> 
   
  <item> 
   <title>Damn - I just wrote a page for this ticket and my session ti</title> 
   <description>Damn - I just wrote a page for this ticket and my session timed out :(



Quick synopsis then: I have been wrestling with this issue for way too long now and finally have decided to at least try forwarding always as a message/rfc822 part instead of trying to be fancy in figuring out what parts should be forwarded.  I have implemented (sort of) #2 with the addition that we still put the text of the first viewable text part into the message body window.  This is because I (at the least) find it very useful when forwarding messages to others in order to point out details of the forwarded message for example.  But since the original text part is in the rfc822 part, the concerns raised below about differing text is alleviated.



This will just be in HEAD for awhile but if the reaction is good there, this should probably be backported before we release 4.1.0.</description> 
   <pubDate>Fri, 03 Feb 2006 01:54:21 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16216</link> 
  </item> 
   
  <item> 
   <title>&gt; I have implemented 

&gt; (sort of) #2 with the addition that</title> 
   <description>&gt; I have implemented 

&gt; (sort of) #2 with the addition that we still put the text of the 

&gt; first viewable text part into the message body window. 



Hi,



Thanks for looking at this!



I have one fear about your proposed solution: is the text that you put in the message body window editable?  Please make that text not editable!  If the text is editable, then two version of the text will still be sent -- the edited text (in the message body), plus the original text (in the attachment).  My goal is to make sure that only one version of text is sent.



If you&#039;d like to compare to another mail program, I can say how the Pine mail reader does it:

1. By default, forward only the text/plain part.  During the sending process, insert the text/plain into the editing buffer, so that the user can edit it.

2. If the user choose an optional forwarding command, pine will attach the forwarded message as a message/rfc822 attachment, and will not let the user edit the text of the attachment.  (The user can still type a preamble, that will be put in to the first part of the message, before the attachment.)



This algorithm simplifies down to something like:

1. If the user edits the text of the forwarded message, make sure that ONLY that text is sent, and that no other text from the original message is sent.  Send only the edited text.

2. If the user user does not edit the text of the forwarded message, then it is safe to forward anything that was in the original message.





Thank you,



 -Jacob Morzinski</description> 
   <pubDate>Fri, 03 Feb 2006 21:02:18 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16272</link> 
  </item> 
   
  <item> 
   <title>&gt; Quick synopsis then: I have been wrestling with this issue</title> 
   <description>&gt; Quick synopsis then: I have been wrestling with this issue for way 

&gt; too long now and finally have decided to at least try forwarding 

&gt; always as a message/rfc822 part instead of trying to be fancy in 

&gt; figuring out what parts should be forwarded.  I have implemented 

&gt; (sort of) #2 with the addition that we still put the text of the 

&gt; first viewable text part into the message body window.  This is 

&gt; because I (at the least) find it very useful when forwarding messages 

&gt; to others in order to point out details of the forwarded message for 

&gt; example.  But since the original text part is in the rfc822 part, the 

&gt; concerns raised below about differing text is alleviated.



I&#039;m not really in favor of this. For one thing, wasn&#039;t whether or not to forward as attachment a preference before? I also like having the ability to pick and choose which attachments get forwarded when I forward a message.



Seems to me like we should be able to, if forwarding a multipart/alternative message inline, just not add as attachments any other alternatives to that part. Is it not that simple?</description> 
   <pubDate>Sun, 05 Feb 2006 17:53:59 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16320</link> 
  </item> 
   
  <item> 
   <title>Having used the &quot;new&quot; way for a few days, I agree with Chuck</title> 
   <description>Having used the &quot;new&quot; way for a few days, I agree with Chuck that this is not an ideal solution.  Pros/cons:



Pros: This is the only way to reliably forward the message leaving the old message (and its attachments) unaltered.  Easy.  Clean.

Cons: Can&#039;t delete individual parts from forwarded message.



Pros/cons of an approach where we simply attach all MIME_Parts from the original message into the new message:



Pros: Allows us to pick/choose parts we actually want to forward.

Cons: We end up with complicated MIME structure information that the enduser should really not be dealing with (i.e. part 2 may be of type multipart/mixed and part 3 may be of type multipart/related, which probably mean about zero to the enduser).  Additionally, we end up forwarding information (i.e. PGP signature information) that has absolutely no use to the receiving user - since it does not pertain to the forwarded message at all.



Another alternative: only attach parts that have a &quot;filename&quot; defined in the MIME headers.



Pros: We are assured these parts can be &quot;downloaded&quot;.

Cons: We lose the ability to forward all kinds of good stuff (e.g. multipart/related) that are possible via MIME.  People can&#039;t think of a MIME Part as a single file (i.e. MIME Part = 1file on local filesystem).  MIME Parts allow much richer presentation than this.  This option is completely not feasible.



So you can see my dilemna.</description> 
   <pubDate>Wed, 08 Feb 2006 21:33:24 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16514</link> 
  </item> 
   
  <item> 
   <title>Brainstorming some more: maybe this should become a preferen</title> 
   <description>Brainstorming some more: maybe this should become a preference.  Thereby allowing those that want to attach via rfc822 part and those that want individual parts to both be happy.</description> 
   <pubDate>Thu, 09 Feb 2006 03:39:33 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16528</link> 
  </item> 
   
  <item> 
   <title>I really thought this _did_ used to be a preference?</title> 
   <description>I really thought this _did_ used to be a preference?</description> 
   <pubDate>Thu, 09 Feb 2006 16:14:13 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16568</link> 
  </item> 
   
  <item> 
   <title>&gt; I really thought this _did_ used to be a preference?



No</title> 
   <description>&gt; I really thought this _did_ used to be a preference?



Not for forwards AFAIK.  Especially considering, up until a week or so ago, it was not possible to forward all parts as a single attachment from the message screen before since there was no code setup to do this.</description> 
   <pubDate>Fri, 10 Feb 2006 07:08:24 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16581</link> 
  </item> 
   
  <item> 
   <title>How do other mail client handle this?</title> 
   <description>How do other mail client handle this?</description> 
   <pubDate>Wed, 15 Feb 2006 01:25:55 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16751</link> 
  </item> 
   
  <item> 
   <title>&gt; How do other mail client handle this?



All over the boar</title> 
   <description>&gt; How do other mail client handle this?



All over the board.  But it is important to note that we really should not be looking at how *desktop* clients handle forwarding (i.e. thunderbird/outlook).  They have the advantage of showing the content of the forwards in the text body themselves - and the user doesn&#039;t have a concept that they are even attachments.



Obviously, we are more limited by what we can do in how we display attachment information since we are dealing with both browser and platform limitations.</description> 
   <pubDate>Wed, 15 Feb 2006 01:33:14 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16752</link> 
  </item> 
   
  <item> 
   <title>Another idea - what about having the &#039;Forward&#039; button be a j</title> 
   <description>Another idea - what about having the &#039;Forward&#039; button be a javascript popup menu that would allow the user to choose from &#039;Forward entire message&#039; and &#039;Forward attachments only&#039;?  This would also be useful for the Reply button - i.e. having a single reply button that may contain the options &#039;Reply&#039;, &#039;Reply to All&#039;, and (if necessary) &#039;Reply to List&#039;.  We could default to having all these options simply displaying on the action bar if javascript is not available.</description> 
   <pubDate>Wed, 15 Feb 2006 01:35:30 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16754</link> 
  </item> 
   
  <item> 
   <title>That sounds like a smart solution, though I personally would</title> 
   <description>That sounds like a smart solution, though I personally would hate to loose the direct keyboard shortcuts for the several reply types.</description> 
   <pubDate>Wed, 15 Feb 2006 08:38:07 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16776</link> 
  </item> 
   
  <item> 
   <title>Since we really *need* this for forwardings only, how about </title> 
   <description>Since we really *need* this for forwardings only, how about the much simpler, though not as beautiful, solution, to show a confirm dialog instead. The most often used selection should be attached to the OK button. Something along &quot;Click OK to forward message&#039;s attachments separately, otherwise the complete message will be attached.&quot;</description> 
   <pubDate>Wed, 15 Feb 2006 08:41:44 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16778</link> 
  </item> 
   
  <item> 
   <title>&gt; Since we really *need* this for forwardings only, how abou</title> 
   <description>&gt; Since we really *need* this for forwardings only, how about the much 

&gt; simpler, though not as beautiful, solution, to show a confirm dialog 

&gt; instead. The most often used selection should be attached to the OK 

&gt; button. Something along &quot;Click OK to forward message&#039;s attachments 

&gt; separately, otherwise the complete message will be attached.&quot;



If I understand you correctly, the alternative would be bound to the &quot;Cancel&quot; button, right? I think that&#039;s very unintuitive. I&#039;d rather have two links, or put the options into a dropdown (we can still bind shortcut keys with JavaScript then, and at least it can be made to work without js. And it can be enhanced with dhtml, etc, if it&#039;s a &lt;select&gt; dropdown.</description> 
   <pubDate>Wed, 15 Feb 2006 16:07:11 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t16807</link> 
  </item> 
   
  <item> 
   <title>What about using drop down menus using CSS code such as:

ht</title> 
   <description>What about using drop down menus using CSS code such as:

http://www.htmldog.com/articles/suckerfish/dropdowns/</description> 
   <pubDate>Fri, 10 Mar 2006 05:08:56 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t17479</link> 
  </item> 
   
  <item> 
   <title>This specific one doesn&#039;t work with IE, but something like t</title> 
   <description>This specific one doesn&#039;t work with IE, but something like this would be nice. Hm, maybe we should take the opportunity and turn the whole action bar into an unordered list?</description> 
   <pubDate>Fri, 10 Mar 2006 09:31:33 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t17484</link> 
  </item> 
   
  <item> 
   <title>&gt; This specific one doesn&#039;t work with IE, but something like</title> 
   <description>&gt; This specific one doesn&#039;t work with IE, but something like this would 

&gt; be nice. Hm, maybe we should take the opportunity and turn the whole 

&gt; action bar into an unordered list?



Sounds good to me. Also, the javascript is there in this implementation specifically so it does work in IE - emulates the :hover stuff.</description> 
   <pubDate>Fri, 10 Mar 2006 17:26:39 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t17539</link> 
  </item> 
   
  <item> 
   <title>Added option in CVS to forward without adding the body text </title> 
   <description>Added option in CVS to forward without adding the body text to the compose message&#039;s body.</description> 
   <pubDate>Thu, 13 Apr 2006 19:58:07 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18896</link> 
  </item> 
   
  <item> 
   <title>&gt; Added option in CVS to forward without adding the body tex</title> 
   <description>&gt; Added option in CVS to forward without adding the body text to the 

&gt; compose message&#039;s body.



I&#039;m not clear what the purpose of this is - it&#039;s trivial to delete the text if you don&#039;t want it. The original reporter was complaining about the original text still being in the _attachment_.</description> 
   <pubDate>Fri, 14 Apr 2006 02:25:32 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18900</link> 
  </item> 
   
  <item> 
   <title>&gt;&gt; Added option in CVS to forward without adding the body te</title> 
   <description>&gt;&gt; Added option in CVS to forward without adding the body text to the

&gt;&gt; compose message&#039;s body.

&gt;

&gt; I&#039;m not clear what the purpose of this is - it&#039;s trivial to delete 

&gt; the text if you don&#039;t want it. The original reporter was complaining 

&gt; about the original text still being in the _attachment_.



I would like to comment on this issue from the perspective of a simple user.



A user gets a message, which contains a body and may  be one or more attachments. When I forwar this message, I expect to see the body in the editor, so that I can comment it or just add my own text. Iexpect eventual attachments from the original message to be attached again with the possibility to remove them one by one. That&#039;s it.



I can see that there may be reasons to do more sophisticated things with message parts, but please let that not be the default behavior.



As it is now compared to how it used to be in older versions:

- Processing mail gets more complicated, I have to go through a menu instead of just click the action I want(or use a short cut key).

- After two times forwarding a message(common practice here) the end recipient gets a message with two extra attachments containing message sources and two &#039;unnamed&#039; links which if followed show a piece of the message already on screen. This is the &#039;without body text&#039; option. In the standard behavior things are even worse, it ends up with three copies of the original body text as an extra.



Please keep it simple for ordinary users.</description> 
   <pubDate>Fri, 14 Apr 2006 13:09:21 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18903</link> 
  </item> 
   
  <item> 
   <title>I am an ordinary user too - I loved the possibilty to remove</title> 
   <description>I am an ordinary user too - I loved the possibilty to remove single attachments</description> 
   <pubDate>Fri, 14 Apr 2006 13:13:51 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18904</link> 
  </item> 
   
  <item> 
   <title>&gt; I am an ordinary user too - I loved the possibilty to remo</title> 
   <description>&gt; I am an ordinary user too - I loved the possibilty to remove single 

&gt; attachments



Please see the comment dated 2/8/2006.  The issue is it is very difficult (if not impossible) to code an algorithim that 100% correctly identifies what &quot;should&quot; be forwarded and what shouldn&#039;t.  IMP&#039;s previous algorithim was not 100% reliable - that is why I have been playing around with other methods.



My goal is to create a forward drop-down (like the current reply drop-down) that has (at least) 2 options: Forward as RFC822 part and Forward attachments only.  Also, to add a preference for default forwarding behavior - this behavior will be used when the &#039;Forward&#039; navbar action is clicked (rather than any of the options in the drop down).  I just haven&#039;t had a whole heck of a lot of time lately, which is why this ticket has stalled lately.  But this is at the top of my list of things to do (this obviously needs to be resolved before IMP 4.2 so I am bumping the milestone).</description> 
   <pubDate>Fri, 14 Apr 2006 15:24:21 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18909</link> 
  </item> 
   
  <item> 
   <title>sounds great (in the meantime I also learned to like the rep</title> 
   <description>sounds great (in the meantime I also learned to like the reply drop-down)</description> 
   <pubDate>Fri, 14 Apr 2006 17:23:25 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18911</link> 
  </item> 
   
  <item> 
   <title>Michael, I think the plan of having a choice between &quot;forwar</title> 
   <description>Michael, I think the plan of having a choice between &quot;forward as message/rfc822&quot; and &quot;forward text and attachments&quot; is a good solution.



I&#039;ve been supporting mail programs for 12 years but don&#039;t know your own experience, so please forgive me if the following comment is something you already know: I hope you are aware that sub-parts of multipart/alternative are not attachments.



This is crucially important, and bears repeating: sub-parts of multipart/alternative are not attachments. They are alternate versions of a single source. The &quot;forward text and attachments&quot; command should only forward one of the sub-parts of the multipart/alternative bundle.





I&#039;m sorry if I&#039;m wasting your time by repeating things that you already know, but I&#039;d hate to see you spend a lot of time coding a feature that doesn&#039;t addess the core of this bug report.



Multipart/alternative is the core of this report.  IMP&#039;s current behavior has two bugs: (1) the &quot;forward&quot; command converts multipart/alternative into multipart/mixed, and (2) the compose window allows a user to edit one of the sub-parts without propagating those changes into the other sub-parts.



1) Converting multipart/alternative into multipart/mixed violates the intent of multipart/alternative, because mail clients will automatically display all subparts of multipart/mixed to the user.  See RFC 2046: &quot;What is most critical, however, is that the user not automatically be shown multiple versions of the same data.&quot;



2) Editing one body part without editing the others also violates the intent of multipart/alternative: see RFC 2046: &quot;In particular, each of the body parts is an &quot;alternative&quot; version of the same information.&quot;





Summing up: yes, I like your plan of offering a choice between &quot;forward as MIME attachment&quot; and &quot;forward text + attachments&quot;.</description> 
   <pubDate>Fri, 14 Apr 2006 17:35:50 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18913</link> 
  </item> 
   
  <item> 
   <title>&gt; Please see the comment dated 2/8/2006.  The issue is it is</title> 
   <description>&gt; Please see the comment dated 2/8/2006.  The issue is it is very 

&gt; difficult (if not impossible) to code an algorithim that 100% 

&gt; correctly identifies what &quot;should&quot; be forwarded and what shouldn&#039;t.  

&gt; IMP&#039;s previous algorithim was not 100% reliable - that is why I have 

&gt; been playing around with other methods.



I understand where you&#039;re coming from here, but I think the previous behavior was more intuitive for most users. The only real issue was with multipart/alternative parts, thus the original issue in this ticket.



&gt; My goal is to create a forward drop-down (like the current reply 

&gt; drop-down) that has (at least) 2 options: Forward as RFC822 part and 

&gt; Forward attachments only.  Also, to add a preference for default 

&gt; forwarding behavior - this behavior will be used when the &#039;Forward&#039; 

&gt; navbar action is clicked (rather than any of the options in the drop 

&gt; down).  I just haven&#039;t had a whole heck of a lot of time lately, 

&gt; which is why this ticket has stalled lately.  But this is at the top 

&gt; of my list of things to do (this obviously needs to be resolved 

&gt; before IMP 4.2 so I am bumping the milestone).



I think those 2 are what is needed, and a preference matches what I remember clients like Thunderbird providing.



With multipart/alternative + forwarding inline, I&#039;d also be fine with just taking the text part from the alternative part and dropping the rest. If at some point we can be smarter and take the HTML part if we support HTML composition, that&#039;s a bonus.</description> 
   <pubDate>Sat, 15 Apr 2006 17:11:09 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t18944</link> 
  </item> 
   
  <item> 
   <title>&gt; With multipart/alternative + forwarding inline, I&#039;d also b</title> 
   <description>&gt; With multipart/alternative + forwarding inline, I&#039;d also be fine with 

&gt; just taking the text part from the alternative part and dropping the 

&gt; rest. If at some point we can be smarter and take the HTML part if we 

&gt; support HTML composition, that&#039;s a bonus.



I haven&#039;t looked into the source code about this yet, but I imagine just pulling out the text/plain or text/html depending on if the user selects plain or html editor should be simple enough, and not forward the other alternative parts.</description> 
   <pubDate>Wed, 19 Apr 2006 17:03:44 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t19188</link> 
  </item> 
   
  <item> 
   <title>&gt; With multipart/alternative + forwarding inline, I&#039;d also b</title> 
   <description>&gt; With multipart/alternative + forwarding inline, I&#039;d also be fine with 

&gt; just taking the text part from the alternative part and dropping the 

&gt; rest. If at some point we can be smarter and take the HTML part if we 

&gt; support HTML composition, that&#039;s a bonus.



Not trying to be difficult here :) but here&#039;s the big issue w/ multipart/alternative parts.  Say you receive this message (which I do, on a fairly consistent basis, from a known contact):



+ multipart/alternative

  + text/plain

  + text/html

  + application/pdf



What I *really* want out of these three is the application/pdf.  Those other 2 are useful for displaying inline (since inline PDF display isn&#039;t happening - at least very easily w/web based clients).  But when I forward this kind of message (which I often do), what I really want is that application/pdf part to make it to the next user.  the text/plain and text/html parts are worthless when it comes to formatting of that PDF part.  So simply &quot;picking&quot; one part of the multipart is not that easy...</description> 
   <pubDate>Wed, 19 Apr 2006 17:16:07 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t19193</link> 
  </item> 
   
  <item> 
   <title>&gt; + multipart/alternative

&gt;   + text/plain

&gt;   + text/html</title> 
   <description>&gt; + multipart/alternative

&gt;   + text/plain

&gt;   + text/html

&gt;   + application/pdf



An excellent example!  Thank you for including it.



&gt; But when I forward this kind of message (which I often do), what I 

&gt; really want is that application/pdf part to make it to the next user. 



I suggest that you could choose to &quot;Forward entire message as MIME attachment&quot;.  The message will be forwarded as a message/rfc822 attachment, successfully passing the entire multipart/alternative tree onward to the next recipient.  (You can do this from the folder index -- put a checkmark on the line with the message, and click the &quot;forward&quot; command from above the list of messages.  And yes, it would be nice if there were a way to invoke the &quot;forward as MIME&quot; command directly while viewing a message.)



Most people won&#039;t be as sophisticated as you, and will be content with a simpler scheme, where IMP picks just one part of the multipart/alternative bundle, and forwards that one part.</description> 
   <pubDate>Wed, 19 Apr 2006 17:37:26 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t19194</link> 
  </item> 
   
  <item> 
   <title>Forwarding attachments has been reimplemented in HEAD.  Help</title> 
   <description>Forwarding attachments has been reimplemented in HEAD.  Help would be great in tweaking the attachment recognition algorithim.</description> 
   <pubDate>Fri, 05 May 2006 00:43:46 +0000</pubDate> 
   <link>https://bugs.horde.org/ticket/3381#t19840</link> 
  </item> 
   
   
 
 </channel> 
</rss> 
