1075446 - [email/UI] Message reader progress spinner shows forever on messages with no usable body parts, such as PGP/MIME multipart/encrypted parts.

Status: RESOLVED WONTFIX

(Firefox OS Graveyard :: Gaia::E-Mail, defect)

Description

[Jan Jansen]

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
Build ID: 20140930030202

Steps to reproduce:

Send encrypted PGP Mail to the Firefox OS email app and try to open the mail with Firefox OS Mail app.
Tested on Simulator 2.2.


Actual results:

The loading circle start and never stop.


Expected results:

For example a Message: Error can't open encrypted mails.

Comment 1

[Jan Jansen]

I use PGP/MIME to send encrypted Mails

Comment 2

[Andrew Sutherland [:asuth] (he/him)]

PGP/MIME implies that the message body will be a multipart/encrypted part, which we will ignore.  I would expect what's happening in your case is that there is only a multipart/encrypted part.  All our bodyRepsDownloaded() checks use Array.every() so they are fine.  I see in message_reader.js that we do check reps.length and assume that a loaded message would have at least one part.

I think there are really 2 things to be done in general:

1) If there are no body parts, show a special box that says something like "This message has no body parts that can be displayed".  It's conceivable we could have no body but still have attachments, or there could be a malformed message with parts no mail client could display.

2) For messages including encrypted parts, maybe make a note in some support structure and use that to alter the previous string or to just use a different string like "This message contains encrypted contents that this application cannot display."

I would propose we just address the first case here and spin the second case off to be a separate bug.  Arguably the first case would be a marked improvement for PGP users.  Since it's unlikely anyone is using PGP and is not aware of it, they'd probably be able to infer what's happening from the first case and the second case is really just an improvement.


Looking at the code I don't think there's any functional breakage that occurs, just the failure to remove the progress spinner.

Comment 3

[BMO Automation]

Firefox OS is not being worked on