You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@harmony.apache.org by Tim Ellison <t....@gmail.com> on 2009/11/17 16:38:34 UTC
[testing] pre-commit testing (was: Re: svn commit: r835212 - /harmony/enhanced/classlib/trunk/modules/luni/src/main/java/java/io/BufferedReader.java)
On 17/Nov/2009 03:50, Nathan Beyer wrote:
> 2009/11/16 Alexei Fedotov <al...@gmail.com>:
>> Funny. We don't have a requirement that a committer have to test the
>> code before committing it here [1].
>
> I suppose some things are just assumed.
>
>> [1] http://wiki.apache.org/harmony/NewCommitter
Of course, we are all grown-ups. Mistakes happen.
How close are we to being able to run all the tests via Hudson?
I recall that there was discussion and work on packaging the tests so
they can be run from the HDK. I don't recall it being declared 'done'?
Regards,
Tim
Re: [testing] pre-commit testing (was: Re: svn commit: r835212 - /harmony/enhanced/classlib/trunk/modules/luni/src/main/java/java/io/BufferedReader.java)
Posted by Mark Hindess <ma...@googlemail.com>.
In message <4B...@gmail.com>, Tim Ellison writes:
>
> On 17/Nov/2009 03:50, Nathan Beyer wrote:
> > 2009/11/16 Alexei Fedotov <al...@gmail.com>:
> >> Funny. We don't have a requirement that a committer have to test the
> >> code before committing it here [1].
> >
> > I suppose some things are just assumed.
> >
> >> [1] http://wiki.apache.org/harmony/NewCommitter
>
> Of course, we are all grown-ups. Mistakes happen.
>
> How close are we to being able to run all the tests via Hudson?
> I recall that there was discussion and work on packaging the tests so
> they can be run from the HDK. I don't recall it being declared 'done'?
"ant -f /path/to/hdk/build/test/build.xml" should work but there are
a few tests that fail when run this way (which is a bit odd since the
tests effectively run from there when you run "ant test" too. However,
I see consistently broken tests anyway so perhaps we should just start
running all the tests this way *and* investigate all of the failures
then start being stricter about keeping them working (or excluded).
Regards,
Mark.
Re: [testing] pre-commit testing (was: Re: svn commit: r835212 -
/harmony/enhanced/classlib/trunk/modules/luni/src/main/java/java/io/BufferedReader.java)
Posted by Alexei Fedotov <al...@gmail.com>.
Ok, done.
On Wed, Nov 18, 2009 at 6:11 AM, Nathan Beyer <nb...@gmail.com> wrote:
> 2009/11/17 Alexei Fedotov <al...@gmail.com>:
>> I would rather call this "misunderstanding".
>>
>> Jesse run one set of tests, Tim expected another set. One way to
>
> If you think it should be documented on the wiki, go ahead and do it -
> it's open and intended to be less formal than the proper web site.
>
> -Nathan
>
>> correct the problem is to agree that committers run a specific minimal
>> set of tests, for example, on one platform. I thought we had such
>> agreements documented, but failed to find a proper pointer. It seems
>> to me now that instead of formal pre-commit rules we had Stepan
>> Mishura who guided us informally.
>>
>>
>>
>>
>> On Tue, Nov 17, 2009 at 6:38 PM, Tim Ellison <t....@gmail.com> wrote:
>>> On 17/Nov/2009 03:50, Nathan Beyer wrote:
>>>> 2009/11/16 Alexei Fedotov <al...@gmail.com>:
>>>>> Funny. We don't have a requirement that a committer have to test the
>>>>> code before committing it here [1].
>>>>
>>>> I suppose some things are just assumed.
>>>>
>>>>> [1] http://wiki.apache.org/harmony/NewCommitter
>>>
>>> Of course, we are all grown-ups. Mistakes happen.
>>>
>>> How close are we to being able to run all the tests via Hudson?
>>> I recall that there was discussion and work on packaging the tests so
>>> they can be run from the HDK. I don't recall it being declared 'done'?
>>>
>>> Regards,
>>> Tim
>>>
>>
>>
>>
>> --
>> With best regards / с наилучшими пожеланиями,
>> Alexei Fedotov / Алексей Федотов,
>> http://www.telecom-express.ru/
>> http://harmony.apache.org/
>> http://www.expressaas.com/
>> http://openmeetings.googlecode.com/
>>
>
--
With best regards / с наилучшими пожеланиями,
Alexei Fedotov / Алексей Федотов,
http://www.telecom-express.ru/
http://harmony.apache.org/
http://www.expressaas.com/
http://openmeetings.googlecode.com/
Re: [testing] pre-commit testing (was: Re: svn commit: r835212 -
/harmony/enhanced/classlib/trunk/modules/luni/src/main/java/java/io/BufferedReader.java)
Posted by Nathan Beyer <nb...@gmail.com>.
2009/11/17 Alexei Fedotov <al...@gmail.com>:
> I would rather call this "misunderstanding".
>
> Jesse run one set of tests, Tim expected another set. One way to
If you think it should be documented on the wiki, go ahead and do it -
it's open and intended to be less formal than the proper web site.
-Nathan
> correct the problem is to agree that committers run a specific minimal
> set of tests, for example, on one platform. I thought we had such
> agreements documented, but failed to find a proper pointer. It seems
> to me now that instead of formal pre-commit rules we had Stepan
> Mishura who guided us informally.
>
>
>
>
> On Tue, Nov 17, 2009 at 6:38 PM, Tim Ellison <t....@gmail.com> wrote:
>> On 17/Nov/2009 03:50, Nathan Beyer wrote:
>>> 2009/11/16 Alexei Fedotov <al...@gmail.com>:
>>>> Funny. We don't have a requirement that a committer have to test the
>>>> code before committing it here [1].
>>>
>>> I suppose some things are just assumed.
>>>
>>>> [1] http://wiki.apache.org/harmony/NewCommitter
>>
>> Of course, we are all grown-ups. Mistakes happen.
>>
>> How close are we to being able to run all the tests via Hudson?
>> I recall that there was discussion and work on packaging the tests so
>> they can be run from the HDK. I don't recall it being declared 'done'?
>>
>> Regards,
>> Tim
>>
>
>
>
> --
> With best regards / с наилучшими пожеланиями,
> Alexei Fedotov / Алексей Федотов,
> http://www.telecom-express.ru/
> http://harmony.apache.org/
> http://www.expressaas.com/
> http://openmeetings.googlecode.com/
>
Re: [testing] pre-commit testing (was: Re: svn commit: r835212 -
/harmony/enhanced/classlib/trunk/modules/luni/src/main/java/java/io/BufferedReader.java)
Posted by Alexei Fedotov <al...@gmail.com>.
I would rather call this "misunderstanding".
Jesse run one set of tests, Tim expected another set. One way to
correct the problem is to agree that committers run a specific minimal
set of tests, for example, on one platform. I thought we had such
agreements documented, but failed to find a proper pointer. It seems
to me now that instead of formal pre-commit rules we had Stepan
Mishura who guided us informally.
On Tue, Nov 17, 2009 at 6:38 PM, Tim Ellison <t....@gmail.com> wrote:
> On 17/Nov/2009 03:50, Nathan Beyer wrote:
>> 2009/11/16 Alexei Fedotov <al...@gmail.com>:
>>> Funny. We don't have a requirement that a committer have to test the
>>> code before committing it here [1].
>>
>> I suppose some things are just assumed.
>>
>>> [1] http://wiki.apache.org/harmony/NewCommitter
>
> Of course, we are all grown-ups. Mistakes happen.
>
> How close are we to being able to run all the tests via Hudson?
> I recall that there was discussion and work on packaging the tests so
> they can be run from the HDK. I don't recall it being declared 'done'?
>
> Regards,
> Tim
>
--
With best regards / с наилучшими пожеланиями,
Alexei Fedotov / Алексей Федотов,
http://www.telecom-express.ru/
http://harmony.apache.org/
http://www.expressaas.com/
http://openmeetings.googlecode.com/