480109 - Provide tests for SeaMonkey sessionstore

Status: RESOLVED FIXED

(SeaMonkey :: Session Restore, defect)

Description

[Misak Khachatryan]

I've decided to file separate bug for SM sessionstore tests and leave bug 448976 alone.

Comment 1

[Misak Khachatryan]

Attachment: tests

Port of all sessionstore tests. Currently failing only undo close tab related tests due SM specific implementation of that function.

Comment 2

[Misak Khachatryan]

Attachment: v2

It's been a while i didn't update this patch, now they ready.

Comment 3

[[e-mail]]

Comment on attachment 428212 [details] [diff] [review]
v2

> 	browser_bug524369.js \
> 	browser_bug524371.js \
> 	browser_bug524371_sample.html \
>+	browser_339445.js \
>+	browser_339445_sample.html \
>+	browser_345898.js \
Why the change in naming scheme?

>+	browser_476161.js \
>+	browser_476161_sample.html \
Missing?

Comment 4

[Misak Khachatryan]

Attachment: v3

Added missing test, renamed all files SeaMonkey way.

Comment 5

[[e-mail]]

Well, my tests build has been crashing lately, which may or may not have caused 339445 to fail; 346337 failed but maybe my tree wasn't up-to-date (it was last updated last week); 524365 threw an exception; 524369 had three failures.

Comment 6

[Misak Khachatryan]

346337 is definitely needs Bug 547710 patch, which landed yesterday, 524365 and 524369 are not a part of this patch. 339445 is failing for me too, investigating ...

Comment 7

[[e-mail]]

(In reply to comment #5)
> Well, my tests build has been crashing lately, which may or may not have caused
> 339445 to fail; 346337 failed but maybe my tree wasn't up-to-date (it was last
> updated last week); 524365 threw an exception; 524369 had three failures.
In a build from 40 hours ago, 524365 is still throwing an exception and 339445 is still failing. But didn't I see someone submit a patch for that exception?

Comment 8

[Misak Khachatryan]

I have bunch of patches with checkin-needed patch, waiting to be checked in. With them, 524365 and 339445 passing clean. I'll add a comment when they all land.

Comment 9

[Misak Khachatryan]

sorry with checkin-needed flag of course

Comment 10

[Serge Gautherie (:sgautherie)]

It looks like this patch needs a new naming scheme update: to sync' with FF. Thanks.

Comment 11

[Misak Khachatryan]

Of course, i'll also need to update some tests too. Patch coming ...

Comment 12

[Misak Khachatryan]

Attachment: v4

added new tests
renamed all tests to original ff bug numbers, mintioning in bug ported bug number, as agreed (except for bug 515006 - there is depended bug, so i decided to leave it for now)
all tests synced to ff ones.

Comment 13

[[e-mail]]

Comment on attachment 432509 [details] [diff] [review]
v4

>+	browser_bug423132.js \
>+	browser_bug423132_sample.html \
> 	browser_bug431826.js \
>+        browser_bug447951.js \
These should all use tabs, but some of them don't.

>+        tab3 = tabbrowser.undoCloseTab(0);
Nit: undeclared variable.

Comment 14

[Misak Khachatryan]

Attachment: for checkin

nits fixed, carrying forward r+ from Neil.

Comment 15

[Misak Khachatryan]

checkin ping

Comment 16

[Serge Gautherie (:sgautherie)]

(In reply to comment #15)

See bug 542946 comment 11...

Comment 17

[Misak Khachatryan]

Attachment: rename all related tests

This patch renames all related test files as Serge wants, should be applied over main tests patch. So main patch can be checked in meanwhile.

Comment 18

[[e-mail]]

(In reply to comment #17)
> This patch renames all related test files as Serge wants, should be applied
> over main tests patch. So main patch can be checked in meanwhile.
I'd prefer both patches pushed in one commit, to reduce history spew ;-)

Comment 19

[Robert Kaiser]

Pushed as http://hg.mozilla.org/comm-central/rev/9c1e65314027 and http://hg.mozilla.org/comm-central/rev/6bd4bb3b4b5d - I'll leave it to Misak to mark the bug fixed. :)

Comment 20

[Serge Gautherie (:sgautherie)]

(In reply to comment #19)

Nits, just ftr:
*In Makefile.in, I would have used 1-tab or 2-spaces rather than 8-spaces.
*Patches were not folded together per comment 18.

Anyway, thanks for these tests :-)

Comment 21

[Misak Khachatryan]

Thanks Serge, i'll fix it on next test commit.

No problems noted so far, marking FIXED.