On Fri, Apr 04, 2008 at 07:47:36PM +0300, Alon Bar-Lev wrote: > > The complexity issue works both ways. By doing cross compiling > > you have now introduced additional packages to build it, using > > derived header files. > > I don't understand.
I think he means that in the current state which allows cross-compilation the build no longer works with MSVC which is the only thing many Windevelopers know and thus the change can be perceived as a regression. This may not have to hold true in the future. > > > Alternatively, I can produce .rc out of rc.in when we distribute > > > the package, so Windows build will find already prepared .rc file > > > with correct version. > > > > I like that better. I don't believe there is any requirement > > to build from SVN using Windows only. > > I don't understand... > Do you agree to generate this files into the package tarball, and > not build directly from svn? I take it he agrees. I like this much better too, the more "complete" a tarball is the better! > > Actually I don't like two build systems. I would prefer the > > Windows based build over the MinG approach. > > Two = different for Windows and none Windows. To be clear; I think the ideal is to be able to build natively using gcc on as many platforms as possible. On *ix this is nothing out of the ordinary. On Windows this means using MinGW. When this works properly, we get cross-compilation (build using i*86-mingw32-gcc on Linux to generate Windows binaries) for free! One thing to keep in mind is that a MinGW install on Windows is only a few megabytes, while MSVC probably adds at least two zeroes to the end of that. Since there is interest for MSVC building I think that should be welcome too, but as was hinted to a new SCB binary is the ideal for Windows environments. It'll get there I'm sure. :) //Peter _______________________________________________ opensc-devel mailing list [email protected] http://www.opensc-project.org/mailman/listinfo/opensc-devel
