Re: [boost] [regression] run_tests.sh regression.py

From: Rene Rivera (grafik.list_at_[hidden])
Date: 2006-03-18 12:30:27

Roland Schwarz wrote:
> Rene Rivera wrote:
>> run_tests.sh is deprecated, it serves as an example of how one might run 
>> tests without Python.
> 
> Thank you.
> 
> Could you probably give me another hint please?
> 
> I am trying to set up kind of a small regression test for the 
> thread_rewrite branch.
> 
> I want to produce a html table for just the thread library. I am not 
> sure if this is supposed to work, or at least I cannot figure it out.
> 
> I have managed so far to get to run the process_jam_log tool 
> successfully. But now when I run the compiler_status tool all the 
> previous (full regression) tests are also included into the table. I 
> suspect I have to wipe out something before, but what? Can you give me a 
> hint please?

Well I can think of two ways to get only one set of results in the 
table. Delete all of the binaries, by default in boost-root/bin (or 
bin.v2 if it's BBv2), run the tests for your one library and then 
generate the tables. Or use the "-sALL_LOCATE_TARGET=<path>" when 
building and running the tests for the library, and tell process_jam_log 
and compiler_status where that bin location is. For the latter 
complier_status takes a "--locate-root=<path>" that you set to the 
ALL_LOCATE_TARGET location, and for process_jam_log the 
ALL_LOCATE_TARGET location is the last argument.

There are equivalent options in the regression.py script, I don't 
remember them offhand though.

--
-- Grafik - Don't Assume Anything
-- Redshift Software, Inc. - http://redshift-software.com
-- rrivera/acm.org - grafik/redshift-software.com
-- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo