|
Boost-Build : |
From: Vladimir Prus (ghost_at_[hidden])
Date: 2006-02-21 02:34:35
On Monday 20 February 2006 22:38, David Abrahams wrote:
> Vladimir Prus <ghost_at_[hidden]> writes:
> > It looks like Python tests remain the biggest issue for regression tests
> > on Windows. I've just fixed a technical issue that prevented all tests
> > from run, but I'd expect future issues.
> >
> > Dave, can you run Python tests with V2 and look into any issues that
> > surface? I expect that you'll do this much faster than Alexey and I
> > sending results and patches back and forth?
>
> Volodya, it looks like there was not any effort to set up the
> $(interpreter) variable for nt and cygwin, so of course none of the
> tests run.
Oh, I did not realize this.
> Did you expect me to figure all that stuff out on my own
> or did you intend to set it up in a configuration that had a chance of
> working and ask me to debug it? Either way is okay as long as I
> understand what I'm getting into.
Of course, if you could figure everything on your on, this will be excellent.
The more pragmatic course of action would be:
- since you know Python tests very well, you can immediately point out to
any issues
- I try to fix all issues that I can without access to Windows
- You continue from there to fix the remaining problems.
I've just comitted a change that tries to set the 'interpreter' for NT.
- Volodya
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk