Re: [boost] Handling testcases that are known to fail?

From: Douglas Gregor (gregod_at_[hidden])
Date: 2002-01-25 08:45:58

On Wednesday 23 January 2002 11:45 pm, you wrote:
> You're making the assumption that having errors show up in the regression
> tests will force a maintainer to make library X work on some compiler.
> This assumption is incorrect. Boost is a volunteer organization and can't
> force anyone to do anything. 

Yes, but even volunteers respond to prodding. I personally have been working 
on the _huge_ number of problems under GCC 2.95.3 on FreeBSD, with the hope 
of finally attaining the holy grail of nightly testing - a green dashboard. 
(Green = no errors, Red = errors). I understand that nobody but me is 
actually using the dashboard (shameless plug : 
http://public.kitware.com/dashboard.php?name=boost) at this point, and it is 
completely red. However, if one looks at, e.g., the CMake dashboard (click 
"CMake" at the top of the Boost dashboard page), one sees all green. Were you 
to wake up one morning and see that your change made it red, wouldn't you try 
to fix the bug you introduced? 

        Doug