> Uh... maybe this is useful, maybe not.. anyway here it goes: > What i do is edit message as new, then delete the attachment, and then > save in drafts, and move back to the inbox. > Of course, composer re-writes the header, and all sender data is > therefore lost. But the email looks just like the original without the > attachment.
Yep, that's the currently supported option for people who don't mind if they lose the header info - including the message id, which causes a loss of threading. The point of bug 2920 is to have a way to accomplish this without losing any of the header info. The only existing way to do it is to edit the mailbox file by hand, and delete the corresponding .msf file, which isn't convenient and which most users wouldn't willing to do; it also carries the risk of corrupting the mailbox, if someone edits the file improperly. > Could it be parsed by the email composer? <--- (i dont know what im > talking about here) > > ---8<--*snip* *snip*---8<--- > >> > I expect it wouldn't be >> > practical to use code from the html editor to read and parse the >> > structure of the html, insert text in the appropriate place in the > > ---8<--*snip* *snip*---8<--- AFAIK, the same code used for editing and viewing web pages is also used for editing and viewing html e-mail. So having it be parsed by the e-mail composer would be pretty much the same as having it parsed by the html editor (I think). At this point, neither one appears to be practical or reliable for appending text info to existing messages. But it might not matter much because I'm starting to lean away from appending text info to the message body. That approach may be too problematic, with the potential of it being totally unworkable (as far as appending text info to non-plain-text versions of the message body - html or otherwise). The leading candidate right now is to replace the attachment with a text-info attachment. This might be awkward, but it may be less problematic to deal with. It still needs to be investigated thoroughly, and everything for bug 2920 is generally up in the air. -- Matt Coughlin [EMAIL PROTECTED] <remove "sp4mless_" from the e-mail address to reply>
