This is a regression introduced with bug 161488. To reproduce this bug, you need a specially prepared message, that is usually not produced by mail agents, but is valid (and might be sent by other mail agents). For example, a message, where the single content is an image. I can send such a message to anybody who wants to test this behaviour, let me know. Steps to reproduce: - receive a message that is (e.g.) of content type image/gif - forward this message to yourself - receive the message and look at it Actual behaviour: - the received message does not display the image - when doing view source, raw binary data is displayed Expected behaviour: The image should get forwarded as base64 and should display fine when received.
Created attachment 115333 [details] [diff] [review] Patch v1 This patch appears to fix the problem for me. I've tested a variety of messages and all seems to be fine. The idea of the patch is: When doing forwarding as attachment and having output_format set to Decrypt-Only, when dealing with an mimeExternalObject, do not allow to decode (be it base64, uue, etc.), but instead pass through the external object directly.
Comment on attachment 115333 [details] [diff] [review] Patch v1 Can you please review?
Comment on attachment 115333 [details] [diff] [review] Patch v1 sr=sspitzer, but if he as time (he may not), can you get ducarroz to review? my only request would be before you check in, please add a comment to the code like: // for forwarding binary messages, see bug #194636 (no need for a new patch)
Comment on attachment 115333 [details] [diff] [review] Patch v1 R=ducarroz
thanks for the review, ducarroz. (I know you are busy with other things)
Patch checked in, marking fixed.
Comment on attachment 115333 [details] [diff] [review] Patch v1 Suggesting to fix this regression for the 1.3 release.
Comment on attachment 115333 [details] [diff] [review] Patch v1 a=asa (on brhalf of drivers) for checkin to 1.3 branch. Please land this ASAP and put fixed1.3 in the status whiteboard. Thanks.
Fix checked in to 1.3 branch.