Re: [Ming-dev] small patch for python extension

[Patrice D. <per...@fr...>]

On Fri, Dec 19, 2008 at 06:55:22PM +0100, strk wrote:
> On Fri, Dec 19, 2008 at 11:57:03AM +0100, Patrice Dumas wrote:
> 
> Pat, the current ming head seems to succeed in building python
> out of the source tree already. What am I missing ?
> Different python or automake versions ?

Here is th eprecise error. Even if ming_wrap.c is created, the one
from the source dir is used, and the ming_wrap.o file is created in the 
sourcedir:

gcc -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -g -O2 -Wall -DSWF_LITTLE_ENDIAN -fno-strict-aliasing -fPIC -I/usr/local/include -I/home/dumas/tmp/ming/py_ext/../src -I/usr/include/python2.5 -c ../../../src/ming/py_ext/ming_wrap.c -o build/temp.linux-i686-2.5/../../../src/ming/py_ext/ming_wrap.o
Assembler messages:
Fatal error: can't create build/temp.linux-i686-2.5/../../../src/ming/py_ext/ming_wrap.o: No such file or directory

That's why I proposed to always have ming_wrap.c in the build 
directory and use that file in setup.py.in.

It even seems to me that what I propose doesn't break in case of parallel 
make, since ming_wrap.c being BUILT_SOURCES is a dependency of all, and
all-am is run by a call of MAKE and therefore cannot be done in parallel
with ming_wrap.c.


--
Pat