259480 - non-displayable "text/calendar" entries of multipart/alternative messages are completely hidden.

Status: RESOLVED DUPLICATE of bug 505024

(MailNews Core :: MIME, enhancement)

Description

[Keith Hopkins]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.2) Gecko/20040803
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.2) Gecko/20040803

Moz1.7.2 hides text/calendar of multipart/alternative messages (from MS-Exchange
200x) because it does not know how to display them.
"Unknown" or non-displayable types should not be hidden, but should be shown as
attachments.

Reproducible: Always
Steps to Reproduce:
example msg: (bits)

[beginning of msg]
[header]
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C4995B.04833D80"
Content-class: urn:content-classes:calendarmessage
[header]
This is a multi-part message in MIME format.

------_=_NextPart_001_01C4995B.04833D80
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
[plain text content]
------_=_NextPart_001_01C4995B.04833D80
Content-class: urn:content-classes:calendarmessage
Content-Type: text/calendar;
	method=REQUEST;
	name="meeting.ics"
Content-Transfer-Encoding: 8bit
[vcalendar entry]
------_=_NextPart_001_01C4995B.04833D80--
[end of msg]
Actual Results:  
"Content-Type: text/calendar;" is completely hidden if some other content is
displayable.

Expected Results:  
Fix: 
Display as attachment.
(alt fix:  see http://bugzilla.mozilla.org/show_bug.cgi?id=130119)

Enhance: 
If a text/calendar type is detected, and Calendar(Sunbird) is integrated into
Moz/Tbird, create a button/popup/icon in the window to export the text/calendar
part into Calendar.

Related items: http://bugzilla.mozilla.org/show_bug.cgi?id=130119#c4
and http://bugzilla.mozilla.org/show_bug.cgi?id=142092#c4

Comment 1

[Dan Mellem]

I can confirm that this happens, but isn't the point of alternative parts to
have a failback for browsers that don't support some of these functions? I do
see the RFE for recognizing and exporting vcalendar items and will confirm that.

Status->new
HW->All
OS->All
Severity->enhancement

Comment 2

[Keith Hopkins]

The problem is how Mozilla decides which part to display, and which part to
hide.  Is there some standard that says use first recognizable and ignore the
rest? or is it an internal design decision for Moz?

If I install a calendar plugin, can that plugin tell Moz to utilize
text/calendar; types instead of text/plain; types?  (If so, maybe I need to log
this bug against the calendar/sunbird project, because they are not doing it.)

Comment 3

[Dan Mellem]

It's supposed to chose the best form that it knows how to handle. It will choose
HTML over text, for example.

From http://www.w3.org/Protocols/rfc1341/7_2_Multipart.html, section 7.2.3:

'The user agent should either choose the "best" type based on the user's
environment and preferences, or offer the user the available alternatives. In
general, choosing the best type means displaying only the LAST part that can be
displayed.'

Comment 4

[Mike Cowperthwaite]

*** Bug 336436 has been marked as a duplicate of this bug. ***

Comment 5

[(not reading, please use [e-mail] instead)]

sorry for the spam.  making bugzilla reflect reality as I'm not working on these bugs.  filter on FOOBARCHEESE to remove these in bulk.

Comment 6

[:Dossy Shiobara]

I believe Bug 505024 is a duplicate of this bug, but it has seen a bit more detailed investigation.

Comment 7

[Ludovic Hirlimann [:Usul]]

Keith this might have been fixed by bug 351224. Could you take a few minutes and download the latest nightly ( http://ftp.mozilla.org/pub/mozilla.org/thunderbird/nightly/latest-comm-central/ ), backup your profile and test and let us know if this is fixed or not ?

Comment 8

[Mikhail T.]

I still have this problem using Thunderbird-3.1.10

Comment 9

[Mikhail T.]

Actually, once I uninstalled the old Lightning add-on, the problem disappeared...

I wish, Lightning actually worked with modern Thunderbird, but that's another story.

Comment 10

[Brian DeRocher]

The problem we are experiencing at work is the Exchange server generates the text/html and text/plain alternates for calendar event invitations and these alternatives are blank!

"<html>
...<meta name="Generator" content="Microsoft Exchange Server">
...<div>&nbsp;</div>
...</html>"

Since Thunderbird without Lightning will ignore text/calendar, it correctly displays this blank html message.

So i recommend that if Thunderbird sees text/calendar that at least it tells the user there is a preferred format.  This gives the user the option of installing extensions.  This could probably be added generically for any type of file.

Comment 11

[Brian DeRocher]

BTW, with the Lightning extension, the text/calendar alternative is displayed and appears to be working fine.

Comment 12

[Jim Porter (:squib)]

(In reply to comment #10)
> The problem we are experiencing at work is the Exchange server generates the
> text/html and text/plain alternates for calendar event invitations and these
> alternatives are blank!

Really, that sounds like an Exchange bug, since it's essentially lying by providing a blank part as an "alternative". However, this can be worked around via bug 602718. Personally, I think that's sufficient, since Exchange really needs to get their act together in this regard, and there's only so much that other clients should have to do to accommodate its broken behavior.

Comment 13

[Jan Hoogenraad]

Actually, often the alternative is NOT blank.
It is used by Outlook to display comments that accompany the invitation.

In this specific case, I suggest this kind of mails (even if they arguably do not conform to all standards, as Microsoft occasionally does) should not be handled as multipart/alternative but rather as multipart/mixed.

Comment 14

[Jan Hoogenraad]

I cannot modify the status, but I find it strange that a regression from a user's standpoint (not being able to see when you are invited to a meeting, while it was possible in earlier versions of mozilla) has importance "enhancement". For a user, this is a regression bug.
I cannot modify the importance. Who judges ?

Comment 15

[Jim Porter (:squib)]

Duping this to the bug where I'm fixing the issue (and which has test cases).