http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5399





------- Additional Comments From [EMAIL PROTECTED]  2007-04-01 14:24 -------
I haven't spent a lot of time thinking about this, but why can't we take
anything that's not [EMAIL PROTECTED]/]+/[^/]+$@ and make it 'text/plain' ?

Also, keep in mind there are really 4 things in play here:

1) the actual Content-Type header
2) the Util code that tries to parse the CT header appropriately
3) the 'type' value in the Message/Message::Node object
4) the logic/RE used to determine what to render -- in 3.2, we aren't limited to
certain text parts -- it'll use anything that gets rendered, and there's a
"render-type" (may have a different name) type which specifies what the part was
rendered as.

So we could very easily let it all do the normal thing, except when rendering
assume that anything that doesn't match the above RE is really text/plain and
render it as text/plain.

(#4 was put in, btw, so that plugins could render an image or application/pdf or
whatever into text, and we'd use it for standard body rules...)



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to