Re: [Boost-testing] Branch agnostic testing...

From: Beman Dawes (bdawes_at_[hidden])
Date: 2007-10-15 17:22:49

Rene Rivera wrote:
> I just checked in the changes to the regression.py script(s) to allow 
> easier testing of any branch. Along with them I did some minor file 
> rearrangement. But with all the changes the old version of regression.py 
> is still alive in the usual location. Such that we don't interrupt 
> testing. Now for the new testing...

Thanks, Rene! This should be a big help.

> If you are interested in doing some testing with the new scripts there 
> is one requirement (currently): it only supports getting code from SVN. 
> This *might* change when we can get the tarball working again, but I 
> didn't bother at the moment as it's impossible to test it :-( For now 
> here are some short docs on what to do, while I write up some longer docs:
> 
> * Create a new directory for the branch you want to test.
> * Download the new run.py script 
> <http://svn.boost.org/svn/boost/trunk/tools/regression/src/run.py> into 
> that directory.
> * Run the "python run.py [options] [commands]" as you would for 
> regression.py.
> * To test trunk use "--tag=trunk" (the default), and to test the release 
> use "--tag=branches/release". Or substitute any Boost tree of your choice.
> 
> There are a few option and command changes in the rework but overall the 
> options are the same:
> 
> usage: run.py [options] [commands]
> 
> commands: cleanup, collect-logs, get-source, get-tools, patch,
> regression, setup, show-revision, test, test-clean, test-process,
> test-run, update-source, upload-logs

It looks like the default if commands is omitted is to download the 
tools_regression_source directory, but do nothing else.

Shouldn't the default be 'regression'?

I'm running branch/release tests now with the command set to 
'regression' and it seems to be working.

Thanks,

--Beman