608202 - xpcshell: new test_attachment_size.js fails, "ASSERTION: nothing to write!: 'dest >= line && dest < src' ... mimeenc.cpp, line 757" in yenc test

Status: VERIFIED FIXED

(MailNews Core :: Composition, defect)

Description

[Serge Gautherie (:sgautherie)]

I would guess SeaMonkey misses the msgHdrViewOverlay.js update:
http://hg.mozilla.org/comm-central/diff/41df9e16a197/mail/base/content/msgHdrViewOverlay.js

Comment 1

[Serge Gautherie (:sgautherie)]

Fwiw,
http://brasstacks.mozilla.com/topfails/test/SeaMonkey?name=xpcshell/tests/mailnews/mime/test/unit/test_attachment_size.js

Comment 2

[Jim Porter (:squib)]

It looks like something is going wrong at a much lower level than msgHdrViewOverlay.js. That shouldn't affect the xpcshell tests (I certainly didn't intend for them to be affected by mail/ stuff).

Do you know if the load() calls in test_attachment_size.js would end up pointing to different versions of those scripts? For example:

  load("../../../resources/mailDirService.js");

Incidentally, at least on Linux, the failure[1] is with yEnc attachments which, as far as I know, hasn't been touched recently. Those tests are testing changes from bug 561851, but it's possible that yEnc attachments are just plain broken in Seamonkey for other reasons.

[1] http://mxr.mozilla.org/comm-central/source/mailnews/mime/src/mimeenc.cpp#757

Comment 3

[Serge Gautherie (:sgautherie)]

This test passes on my local Windows 2000, with an Opt tinderbox-build:
I get "Context: test.test:6 state: finished".

(I can't run Debug builds anymore.)

*****

(In reply to comment #2)

> Do you know if the load() calls in test_attachment_size.js would end up
> pointing to different versions of those scripts?

No, they don't.

> Incidentally, at least on Linux, the failure[1] is with yEnc attachments

Yes, test 6: on Windows too.

Comment 4

[Serge Gautherie (:sgautherie)]

(In reply to comment #2)
> it's possible that yEnc attachments are just plain broken
> in Seamonkey for other reasons.

Fwiw, the Windows assertion stack has nothing SeaMonkey-specific.
Does this test passes on Debug Thunderbird?

Comment 5

[Justin Wood (:Callek)]

Mark, Karsten, any specific ideas?

Comment 6

[Serge Gautherie (:sgautherie)]

As this is the only xpcshell SeaMonkey perma-orange atm, I would suggest to disable this test for (Debug) SeaMonkey ftb.
Karsten, would you accept that?

Comment 7

[Karsten Düsterloh]

Lemme have a deeper look first.
Shallow digging tonight showed that the yencoded test attachment is causing the problem, all other test messages are processed fine.

Comment 8

[Mark Banner (:standard8)]

This doesn't appear to affect Thunderbird -> unaffected.

I've no idea why this would be Seamonkey Debug specific, unless there's some pref that's tweaked differently.

Comment 9

[Mark Banner (:standard8)]

Ok, this test does fail for me when run locally. Doesn't affect tinderbox for Thunderbird at the moment, so this falls in the generally wanted category.

Comment 10

[Karsten Düsterloh]

Attachment: fix assertion

The assertion is broken: "dest < src" will only work if the yenc encoding contains escaped characters - which our test does not. 
(Furthermore, "dest >= line" is very unlikely to fail, but doesn't hurt.)

Comment 11

[Serge Gautherie (:sgautherie)]

(In reply to comment #0)
> I would guess SeaMonkey misses the msgHdrViewOverlay.js update:
> http://hg.mozilla.org/comm-central/diff/41df9e16a197/mail/base/content/msgHdrViewOverlay.js

Karsten, though unrelated to the test failure, you might still want to port this.

Comment 12

[Karsten Düsterloh]

(In reply to comment #11)
> > http://hg.mozilla.org/comm-central/diff/41df9e16a197/mail/base/content/msgHdrViewOverlay.js
> 
> Karsten, though unrelated to the test failure, you might still want to port
> this.

True, but that's totally out of scope here.

Comment 13

[Mark Banner (:standard8)]

Comment on attachment 490743 [details] [diff] [review]
fix assertion

r=Standard8, this looks fine. No sr required.

Comment 14

[Karsten Düsterloh]

Pushed to comm-central as <http://hg.mozilla.org/comm-central/rev/7f98b2c78c31>.

Comment 15

[Serge Gautherie (:sgautherie)]

V.Fixed, per
http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1290136847.1290139365.2471.gz
Linux comm-central-trunk debug test xpcshell on 2010/11/18 19:20:47
and
http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1290139855.1290143292.19291.gz
WINNT 5.2 comm-central-trunk debug test xpcshell on 2010/11/18 20:10:55