|
|
Boost Testing : |
From: Roland Schwarz (roland.schwarz_at_[hidden])
Date: 2007-03-16 04:06:44
David Abrahams wrote:
> on Thu Mar 15 2007, Roland Schwarz <roland.schwarz-AT-chello.at> wrote:
>> http://engineering.meta-comm.com/boost-regression/CVS-RC_1_34_0/speedsnail-gcc-d.html
>
> That setup will not generate successful tests of Boost.Python with the
> cygwin flavor of GCC, since you haven't got a "using python" with
> <target-os>cygwin in the condition. Naturally, that would have to
> refer to a cygwin build of Python, not the same one that's used for
> your mingw tests.
I tried with user-config.jam:
using gcc
: 3.4.4
: "C:/Cygwin/bin/g++.exe"
: <flavor>cygwin
<root>"C:/Cygwin/"
;
using python
: 2.5
: "C:/Programme/Python25"
;
using python
: 2.4
: "C:/Cygwin/bin"
: : : <target-os>cygwin
;
And invoked with (from BoostRoot/libs/python/test) :
bjam gcc-3.4.4/toolset-gcc:flavor=cygwin/target-os=cygwin
The first few lines of output:
warning: No working Python interpreter found.
warning: falling back to "C:/Cygwin/bin\bin\python2.4"
Building Boost.Regex with the optional Unicode/ICU support disabled.
Please refer to the Boost.Regex documentation for more information
(don't panic: this is a strictly optional feature).
error: No best alternative for /python_for_extensions
next alternative: required properties: <python>2.5 <target-os>windows
not matched
next alternative: required properties: <python>2.4 <target-os>cygwin
not matched
I either still understand something wrong, or there is a bug?
[ ... ]
> note that the preferred way to configure Python is now to specify a
> command, not a prefix directory as you have done.
Sorry, I am lost now. I am doing exactly as you were showing in the
prvious mail, not?
One additional observation:
When trying to
C:\Cygwin\bin\python from a CMD.EXE shell I get a message box from
the 16bit windows subsystem, complaining about some illegal instructions
or such. I do not get this box when invoking bjam with the above
configuration though.
Roland