On 4/26/05, Karsten Ohme <[EMAIL PROTECTED]> wrote:
Ludovic Rousseau schrieb:
Hmm! Maybe the separation between pure pcsc and libmusclecard was not a good idea after all. They have too many source files in common.
Should we go back and have only one source archive as we had before 1.2.9-beta7?
Well, it is quite logical how it is now.
Yes, it is logical but not really practical.
It is reasonable that the functionality of libmusclecard and and pcsclite is quite different. The only difference for Windows systems is that some of the header files must be available as prerequisites.
This is also true for Unix, not just Windows. But the Unix header files are already installed in /usr/include/PCSC/ if pcsc-lite is already installed.
If this is mentioned, this is enough. But indeed also some sources files (I believe the tokenparser.c, maybe some more ... ) which in Unix are part of the pcsclite library, are necessary.
I plan to reorganise the source code to have something like: - pcsc-lite/src/ : pcsc API source code as it is now - pcsc-lite/libmusclecard/ : libmusclecard source code
libmusclecard will: - use some header files from pcsc-lite/src/ and pcsc-lite/src/PCSC/ using the -I../src preprocessor directive. I imagine it is possible to do the same with a Windows proprocesseor/compiler
Yes, should support each C compiler.
- use some source files from pcsc-lite/src/ using correct rules in Makefile.am. I also imagine it is possible to do that with a Windows compilation project.
This sounds like a autoconf/automake/make approach also for Windows?
I hope all this programs also exist for Windows ... I can make a try. I the beginning I tried cygwin, but it didn't compile. This would mean that under both systems everything could be build the same way. Would be great ...
Would this proposition be acceptable for the Windows developpers/users?
Regards,
_______________________________________________ Muscle mailing list [email protected] http://lists.drizzle.com/mailman/listinfo/muscle
