<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 2/8/2014 3:33 PM, <a class="moz-txt-link-abbreviated" href="mailto:neandr@gmx.de">neandr@gmx.de</a>
wrote:<br>
</div>
<blockquote cite="mid:52F6A291.8050207@gmx.de" type="cite">
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
But having in mind all the discussions and the brave work some
also best knowledged people made to improve or replace the current
composer, I don't feel it's a GSoC project idea to "improve" what
we have right now.<br>
<br>
Also the composer (current, improved or replaced) has to be
integrated very much with the rest of the TB code, <i>how about
using an existing HTML composer and make it work with TB. </i><br>
Maybe there are legal issues for a product integration, but I
could imagine a GSoC project to just bundle those pieces as a test
case. <br>
<br>
I remember some people tried for that already .. but don't
remember the roadblocks.<br>
</blockquote>
<br>
One of the major roadblocks is that our backend compose code is
horribly coupled to the current frontend cod and requires a lot of
cooperation to make things work. For example, you need to pass a lot
of nsIDOMWindows or nsIEditor stuff to the backend to make things
work properly. I'm working on rewriting large swathes of the backend
code which should eventually ameliorate this painful coupling, but I
don't expect it to reach fruition in the next several months.<br>
<pre class="moz-signature" cols="72">--
Joshua Cranmer
Thunderbird and DXR developer
Source code archæologist</pre>
</body>
</html>