Mailing-List: contact flex-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: flex-dev@incubator.apache.org
Received-SPF: pass (athena.apache.org: domain of pent@adobe.com designates
 64.18.1.191 as permitted sender)
From: Peter Ent <pent@adobe.com>
To: "flex-dev@incubator.apache.org" <flex-dev@incubator.apache.org>
Date: Tue, 28 Aug 2012 06:29:38 -0700
Subject: Re: Getting Mustella to work.
Thread-Topic: Getting Mustella to work.
Thread-Index: Ac2FISyY95ck1hisSHqn4a90RZ0ogA==
Message-ID: <CC623EE0.2112%pent@adobe.com>
In-Reply-To: <503C95B4.9030504@gmail.com>
Accept-Language: en-US
Content-Language: en-US
user-agent: Microsoft-MacOutlook/14.2.3.120616
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

When you run any mustella test set for the first time, DO NOT use
-createImages. Just run it simply, such as:

./mini_run tests/components/Application

This will run the tests and compare them to the current baseline images.
If you get any compare-bitmap failures, use ImageDiff to look at the
bad.png (the one just created by your test run) and the baseline .png
files.=20

If you see differences in the fonts, which is to be expected, or a
baseline image had a graphic and the bad.png has a solid rectangle, that
is also expected. In these cases, new baselines need to be generated. I
tried to generate and save all new baselines, but with the thousands of
tests, I'm sure I missed some.

If you see minor differences, like I saw yesterday with a pixel that's
difference or something just slightly off, we should know about it and
then we can figure out what to do. I think we should at least get
confirmation of the same difference from several people so we can make an
informed decision.

--peter

On 8/28/12 5:56 AM, "Kiel" <kiel.s.sinfuego@gmail.com> wrote:

>Hi Carlos,
>
>I'll follow the steps mentioned.
>Got concerned since the number of Passes & Fails didn't match the lists
>on wiki.
>
>Thanks.
>
>On 8/28/2012 5:46 PM, Carlos Rovira wrote:
>> Hi Kiel,
>>
>> take into account that github is not synced with SVN right now, so
>> there you will not find the latest fixes to mustella for at least 2
>> weeks.
>>
>> the team is working with INFRA to get official GIT support. For the
>> latest news this will resolved in the following 1 to 3 weeks.
>>
>> In the meanwhile, we and others use the following path to work:
>>
>> 1) get SVN "develop" branch.
>> 2) work in your local copy.
>> 3) when you are done with a fix, extract the folder (I do this with
>> eclipse-subversive)
>>
>> now:
>>
>> 4) in github, fork apache flex
>> 5) in your machine clone the fork
>> 6) create a branch
>> 7) copy the extract folder created in 3) overlaying it (overwrite)
>> 8) commit to the new branch
>> 9) push to origin (github)
>> 10) create a JIRA ticket
>> 11) create a pull request
>> 12) link the jira ticket to the pull request and viceversa
>> 13) submit to this mailing list
>>
>> This works because the layout folder structure is the same and the
>> failing mustella folders are untouched at the moment so you can
>> overwrite with your changes and make a pull request only with the
>> changes.
>>
>> Alex, Peter and Carol are getting this pull request as patches without
>>problem.
>>
>> Best,
>>
>> Carlos
>>
>>
>> 2012/8/28 Kiel <kiel.s.sinfuego@gmail.com>:
>>> Greetings,
>>>
>>> How to do the Mustella tests on Git(Github)?
>>>
>>> Will I fork "unstable"?
>>>
>