X
xeno fears
I work with the Mingw-w64 Project, and am the project owner of WPG
System64 (you can find out about both at http://www.cadforte.com),
which contains Python.org Python 2.6.2. I haven't used Python 3.1.1
because I have seen errors with "print" occur in several places, but
that is another topic (I am just now adding Python to my portfolio.)
Python support is becoming critical. For numerous reasons, I want to
be able to build Python via the standard Posix method (configure,
make, make install), but for Windows, which means using gcc and in my
case MSys.
The barriers to success are quite small, it is simply a matter of
getting the right configuration, which the provided configure (and
configure.ac) are unable to produce. It is missing some key elements
that, while it builds just fine, produces a broken python (no module
os, another missing module, site_import failure.)
Mingw-w64 has created a branch that is to be labeled stable with the
release of gcc 4.4.2. It is an excellent import crt for msvc using
gcc. It is almost too good, making it all the way through the build
process except for a few small changes here and there (mainly to the
posixmodule, as it has no configuration there. I have configured it.)
I am also unclear on what the MS_WINDOWS directive is for vs the
standard WIN32, I don't know whether to use it or not. I have gotten
through the build without it.
Is there any chance of getting some of the devs or anyone familiar
enough with the source code to make this possibility become reality?
MinGW(32) seems to have given up after 2.5. We are quite active and
have complete VC support back to pre-2005 even.
Thanks
Xenofears (Peter)
System64 (you can find out about both at http://www.cadforte.com),
which contains Python.org Python 2.6.2. I haven't used Python 3.1.1
because I have seen errors with "print" occur in several places, but
that is another topic (I am just now adding Python to my portfolio.)
Python support is becoming critical. For numerous reasons, I want to
be able to build Python via the standard Posix method (configure,
make, make install), but for Windows, which means using gcc and in my
case MSys.
The barriers to success are quite small, it is simply a matter of
getting the right configuration, which the provided configure (and
configure.ac) are unable to produce. It is missing some key elements
that, while it builds just fine, produces a broken python (no module
os, another missing module, site_import failure.)
Mingw-w64 has created a branch that is to be labeled stable with the
release of gcc 4.4.2. It is an excellent import crt for msvc using
gcc. It is almost too good, making it all the way through the build
process except for a few small changes here and there (mainly to the
posixmodule, as it has no configuration there. I have configured it.)
I am also unclear on what the MS_WINDOWS directive is for vs the
standard WIN32, I don't know whether to use it or not. I have gotten
through the build without it.
Is there any chance of getting some of the devs or anyone familiar
enough with the source code to make this possibility become reality?
MinGW(32) seems to have given up after 2.5. We are quite active and
have complete VC support back to pre-2005 even.
Thanks
Xenofears (Peter)