Abdelrazak Younes <[EMAIL PROTECTED]> writes: | Abdelrazak Younes a écrit : | > Bo Peng a écrit : | >> On 5/8/06, Abdelrazak Younes | >> <[EMAIL PROTECTED]> wrote: | >>> Where do I find this win32file? | >> | >> If you use official python, you need to install win32 module from | >> http://starship.python.net/crew/skippy/win32/Downloads.html . This is | >> for mingw only. Cygwin does not seem to need this. | > OK, thanks. | | Got this. With the QtCore4 hack (until you do something about that) | the compilation was going on until Scons tried to compile coloritem.C | which is not there anymore. Please find attached a patch that remove | coloritem.C and QBrowseBox.C from SConsscript, please review and apply. | | After that, relaunching Scons display quickly a lot of cached test and | then stay for a while at this point: | | scons: done reading SConscript files. | scons: Building targets ... | | And then went on with the compilation and stopped at the linking stage | of libqt4.a | | Is it already quicker because it does not need than the Makefile based | compilation because it does not go through the shell libtool command | step before each call to gcc. I am talking about this: | | /bin/sh ../../../../libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H ...
I had the opposite feeling... scons being slower... On Linux process creation is usually very fast, I don't think that is the case on win. | I have a question: Is it necessary to copy the complete tree inside | "win32-qt4/"? I reckon the object files is enought, isn't it? That is a scons faq. (kindo) http://www.scons.org/doc/HTML/scons-user/x1948.html http://www.scons.org/doc/HTML/scons-user/x1965.html -- Lgb
