If the Html to Markdown converter works so well, might it also help better identifying the breakpoint from where to strip previous posts?
In a recent reply-by-email post where it completely failed to strip the previous post, I noticed that in the text part of email, the break point looked like this:
username Full Name Title=20
September 25
Hi Name!
Whereas the html version looked like this:
<div></div>
<div style="margin-bottom:25px;">
<table cellspacing="0" cellpadding="0" border="0">
<tbody><tr>
<td style="vertical-align:top;width:55px;">
<img src="http://forum.mydomain.com/user_avatar/forum.mydomain.com/username/45/26_1.png" title="username" width="45" height="45">
</td>
<td>
<a href="http://forum.mydomain.com/u/username" target="_blank" style="text-decoration: none; font-weight: bold; color: #006699;; font-size:13px;font-family:'lucida grande',tahoma,verdana,arial,sans-serif;color:#3b5998;text-decoration:none;font-weight:bold">username</a>
<a href="http://forum.mydomain.com/u/username" target="_blank" style="text-decoration: none; font-weight: bold; color: #006699;; font-size:13px;font-family:'lucida grande',tahoma,verdana,arial,sans-serif;text-decoration:none;margin-left:7px;color: #3b5998;font-weight:normal;">Full Name</a>
<span style="font-size:13px;font-family:'lucida grande',tahoma,verdana,arial,sans-serif;text-decoration:none;margin-left:7px;color: #999;">Title</span>
<br>
<span style="text-align:right;color:#999999;padding-right:5px;font-family:'lucida grande',tahoma,verdana,arial,sans-serif;font-size:11px">September 25</span>
</td>
</tr>
</tbody></table>
<div style="padding-top:5px;">
<p style="margin-top:0; border: 0;">Hi Name</p>
In other words, I can see why discourse fails to identify the breakpoint based on the text version…