JUnit sampler

[sebb <se...@gmail.com>]

I think it would be better to use the Sampler name as the sample
label, and save the junit class + method as the sampler data.

That way, one can aggregate multiple samples with different  class/method names.

==

Also, I'm not sure that many existing JMeter JUnit tests would make
sense to test, so maybe jmeter classes should be excluded from the
list?

Perhaps better would be to move all the JMeter tests to a separate
source tree and/or separate jar file. This could then be excluded from
the normal binary distribution, as the tests are only really needed by
developers.

S.

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org

Re: JUnit sampler

[sebb <se...@gmail.com>]

On 21/07/05, Peter Lin <wo...@gmail.com> wrote:
> that's a good idea. I'll make that change later today along with some
> other enhancements.

e.g. some basic docn for component reference?

Also, it might be an idea to provide default values for the
success/fail codes/messges.
 
> I thought about excluding the jmeter unit tests, but I couldn't
> decide. basically, I took the easy way out and decided not to filter
> them. I can always add them later.  I like the idea of slowly moving
> the unit tests to some other jar, since it is just for the developers.
> 
> should we make the 2.1 RC and post it?
> 
> I can give it a try if others are busy.  It might take me a few hours
> to figure out how to go about doing that.
>
> peter
> 
> 
> On 7/21/05, sebb <se...@gmail.com> wrote:
> > I think it would be better to use the Sampler name as the sample
> > label, and save the junit class + method as the sampler data.
> >
> > That way, one can aggregate multiple samples with different  class/method names.
> >
> > ==
> >
> > Also, I'm not sure that many existing JMeter JUnit tests would make
> > sense to test, so maybe jmeter classes should be excluded from the
> > list?
> >
> > Perhaps better would be to move all the JMeter tests to a separate
> > source tree and/or separate jar file. This could then be excluded from
> > the normal binary distribution, as the tests are only really needed by
> > developers.
> >
> > S.
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> >
> >
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org

Re: JUnit sampler

[Peter Lin <wo...@gmail.com>]

that's a good idea. I'll make that change later today along with some
other enhancements.

I thought about excluding the jmeter unit tests, but I couldn't
decide. basically, I took the easy way out and decided not to filter
them. I can always add them later.  I like the idea of slowly moving
the unit tests to some other jar, since it is just for the developers.

should we make the 2.1 RC and post it?

I can give it a try if others are busy.  It might take me a few hours
to figure out how to go about doing that.

peter


On 7/21/05, sebb <se...@gmail.com> wrote:
> I think it would be better to use the Sampler name as the sample
> label, and save the junit class + method as the sampler data.
> 
> That way, one can aggregate multiple samples with different  class/method names.
> 
> ==
> 
> Also, I'm not sure that many existing JMeter JUnit tests would make
> sense to test, so maybe jmeter classes should be excluded from the
> list?
> 
> Perhaps better would be to move all the JMeter tests to a separate
> source tree and/or separate jar file. This could then be excluded from
> the normal binary distribution, as the tests are only really needed by
> developers.
> 
> S.
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org

Re: JUnit sampler

[sebb <se...@gmail.com>]

Yes, some code would need to be split out of the existing classes -
which might mean some recoding to get round access to private methods.

There are are also a few stand-alone test classes that could just be
moved to the test source directory tree.

S.
On 21/07/05, Michael Stover <ms...@apache.org> wrote:
> Eliminating the unit tests wouldn't be so easy - the one's under the
> test/ directory are easy, but there's still  a lot done as static inner
> classes that, IMO, should be moved.
> 
> -Mike
> 
> On Thu, 2005-07-21 at 12:19 +0100, sebb wrote:
> > I think it would be better to use the Sampler name as the sample
> > label, and save the junit class + method as the sampler data.
> >
> > That way, one can aggregate multiple samples with different  class/method names.
> >
> > ==
> >
> > Also, I'm not sure that many existing JMeter JUnit tests would make
> > sense to test, so maybe jmeter classes should be excluded from the
> > list?
> >
> > Perhaps better would be to move all the JMeter tests to a separate
> > source tree and/or separate jar file. This could then be excluded from
> > the normal binary distribution, as the tests are only really needed by
> > developers.
> >
> > S.
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org

Re: JUnit sampler

[Michael Stover <ms...@apache.org>]

Eliminating the unit tests wouldn't be so easy - the one's under the
test/ directory are easy, but there's still  a lot done as static inner
classes that, IMO, should be moved.

-Mike

On Thu, 2005-07-21 at 12:19 +0100, sebb wrote:
> I think it would be better to use the Sampler name as the sample
> label, and save the junit class + method as the sampler data.
> 
> That way, one can aggregate multiple samples with different  class/method names.
> 
> ==
> 
> Also, I'm not sure that many existing JMeter JUnit tests would make
> sense to test, so maybe jmeter classes should be excluded from the
> list?
> 
> Perhaps better would be to move all the JMeter tests to a separate
> source tree and/or separate jar file. This could then be excluded from
> the normal binary distribution, as the tests are only really needed by
> developers.
> 
> S.
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org