Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Thunderbird QA - Litmus, do you ?

0 views
Skip to first unread message

Wayne Mery

unread,
Jul 21, 2008, 10:40:27 PM7/21/08
to
Litmus tests for Thunderbird
http://wiki.mozilla.org/Thunderbird:Testing#Litmus_Tests are in need of
your LOVE.

We want these in better shape for Thunderbird v3 testing. So if you have
a bit of time and gumption (there's a word) the following tests need to
be updated for Thunderbird 2, and the corresponding test for Thunderbird
3. (I've only listed the v2 tests below)

Just pick a test or result, research it to determine what needs to be
changed. URLs to get you to testcase and result
test# https://litmus.mozilla.org/show_test.cgi?id=nnn
result# https://litmus.mozilla.org/single_result.cgi?id=nnnn

You don't need a litmus account to view - if you can't edit testcases
just send me an email with the details of your findings. But if you have
an account with admin privileges then you can, carefully of course, make
the required changes to a testcase.


test subgroup ver result
---- --------- --- --- ------
2171 attach 2.0 BFT 142494 clarify test / find bug#
144175
122847
2183 replace 2.0 BFT 74777
2164 IMAP 2.0 BFT 90173
2195 save Draft 2.0 BFT 55127
2198 Global: 2.0 "Return Receipts..." button changed in TB2
button now located @ Options | Advanced | General.
2234 DL Themes 2.0 BFT 53883
2236 Help 2.0 BFT 90205


2.0 BFT with comments, no bug# (may be some overlap with list above)
http://tinyurl.com/6lv4zq


If you finish the above list, feel free to ask me about more that need LOVE.

Thanks in advance for your help.

Ron K.

unread,
Jul 22, 2008, 1:56:55 AM7/22/08
to
Wayne Mery keyboarded, On 7/21/2008 10:40 PM :

#2195 Save Draft lead me to conclude that Tb 2.0.0.14 has a bug because
the Edit Draft button is hidden within the collapsed headers bar. I
normally resort to the Message > Edit as new... menu option, or its
right click context menu equivalent. To me the bug is a case of poor UI
design which I predict will be seen in Shredder since the landing of the
Header scrollbar patch to the default theme.

--
Ron K.
Who is General Failure, and why is he searching my HDD?
Kernel Restore reported Major Error used BSOD to msg the enemy!

Wayne Mery

unread,
Jul 22, 2008, 4:27:09 PM7/22/08
to Ron K.
On 7/22/2008 1:56 AM, Ron K. wrote:
> Wayne Mery keyboarded, On 7/21/2008 10:40 PM :
>
> #2195 Save Draft lead me to conclude that Tb 2.0.0.14 has a bug because
> the Edit Draft button is hidden within the collapsed headers bar. I
> normally resort to the Message > Edit as new... menu option, or its
> right click context menu equivalent. To me the bug is a case of poor UI
> design which I predict will be seen in Shredder since the landing of the
> Header scrollbar patch to the default theme.

adjusted testcase #2195(BFT), 2037(FFT) for 2.0 and 5689(BFT), 5531(FFT)
for 3.0


Ron K.

unread,
Jul 22, 2008, 10:14:45 PM7/22/08
to
Wayne Mery keyboarded, On 7/21/2008 10:40 PM :

Took a look at #2234 and it needs update for Tb 2. Since upgrading to
the Add-on Mgr from the Extensions/Theme Mgr the ability to Drag&Drop
onto the A-Mgr is a user option for Windows platform, perhaps others.
Will need relook with Shredder because of the upgrades to the A-Mgr
introduced by Fx 3 (Plugins tab for example)

Do We need a Litmus test for the Plugins tab of A-Mgr for Shredder
builds? Generaly the Tb builds have Plugins turned off, much to my
displeasure. Also on the same track, what about the AMO preview tab?

Is there a Litmus for the Extensions Tab?

Wayne Mery

unread,
Jul 24, 2008, 12:35:42 PM7/24/08
to
> [see better, newer queries below] http://tinyurl.com/6lv4zq

>
> If you finish the above list, feel free to ask me about more that need
> LOVE.
>
> Thanks in advance for your help.


Someone asked "Any special tasks to do or should i go to litmus and just
pick a test for improvement?".

Great enthusiasm. Great question. Short answer is yes, pick any
testresult.


More specifically:

a) Pick your testresult(s) to work. Queries which list non-PASSing
results, with commencements:
2.0+3.0 FAILED http://tinyurl.com/6npmq4
2.0+3.0 UNCLEAR http://tinyurl.com/6cb382
You'll want to check here in newsgroup to see which testresults people
have posted as already reviewed. Also, rerun the query about once per
hour so it excludes testsresults that got marked 'vetted'.

b) Research testresult(s). Post your findings** with a comment in the
testresult. Then list the testresult#s you worked here in the newsgroup
so everyone can see which have been worked. Then ping me in irc #maildev
at wsmwk, or email, so someone can vet the results.


** Findings - What to put in testresult comments, and reporting results:
- if a comment states the results were reviewed by QA then the result
wasn't vetted - otherwise it wouldn't have appeared in the query of
results. So, a) is the QA review correct and complete, and testcase
corrected if needed? and b) report the testresult#
- is tester's comment accurate? improve if not accurate.
- is status correct? (eg. should it be PASS but it's set to FAIL)
- is bug# cited if a bug exists or should have been filed? cite bug#
- does testcase need to change, based on your findings and tester's?


A note about "duplication" of tests. Testcases are arranged as ordered
lists into (functional) subgroups, and multiple subgroups comprise a
(large) testgroup such as a BFT, FFT, smoke test, etc. In the queries
cited above, i.e. the list of testresults, you will see testresults with
same title but different testcase #. This is because a) each release has
it's own testcases and b) often BFT, FFT and smoketests each have their
own testcase for the exact same test. Obviously, a testcase against the
same version, with identical steps, should fail or pass in the same way.
In other words, we probably shouldn't see several pass results if
someone else correctly identified a failure. So when you see an
identical title, if results are same and correct just cite the
testresult# and and point to a prior finding/review. But if someone's
result differs for the same set of steps, you might want to take a
closer look at the results and the testcase.

0 new messages