From: David Abrahams (dave_at_[hidden])
Date: 2006-08-17 09:59:42
"Andrei Melnikov" <andy.melnikov_at_[hidden]> writes:
> On 16/08/06, David Abrahams <dave_at_[hidden]> wrote:
>> So -DBOOST_PYTHON_STATIC_LIB is not getting applied. This looks like a
>> pretty serious BBv2 bug to me. Volodya?
> I moved the defines to RSP files. But SOME defines are applied, so I
> don't think the bug is inside the code I patched. It's more likely
> that the bug is inside Jamfile.v2 for Boost.Python.
I think the bug is inside the guts of Boost.Build v2, but this is all
> The rationale of putting all switches to response files was to enhance
> debugging, not to make it worse.
It makes it worse when you have access to the command-lines, but not
> Some switches were inside the rsp files, some switches were on command
> lines. So I had to look in two different places to find the problem.
> Now all switches are in a single place.
Great for you if you're not looking at the online build logs to figure
out what happens.
> If we still want to use -d switch to trace the actual command lines, I
> think it's a good idea to display the parameters for :E variable
> modifier at higher debug levels.
> What do you think?
I don't know what -d and :E do.
-- Dave Abrahams Boost Consulting www.boost-consulting.com
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