> > Just a side question: If i run wrap_c on one of the larger system > libraries (e.g. X11.h) and get an error; is it more likely it uses a > rather esoteric cpp features (and therefore isn't covered in the current > wrap_c) OR is it more likely that i miss some compiler flags?
Hard to say, depends on the error... Probably 60/40 chance it is a compiler flag issue. I'm guessing that X11 (like most system libraries) has a pretty gnarly set of ifdefs. At some point it is almost unavoidable to require integrating compiler-adjacent tools like this into the proper build system like SWIG does. I use linux too, so I can have a look if you want to file an issue and link to your wrapper script. Matching compiler options is going to be a tricky issue for distribution of some Cxx.jl based wrappers as well, and I don't know what the answer is. I suspect we may end up with a build-the-world approach for the big "Julia Environment" distribution that is envisioned. This is basically the solution that Kitware and various other groups have adopted to the problem of "first class everywhere" cross-platform applications, after many years of experience. On Mon, May 18, 2015 at 8:21 AM, Andreas Lobinger <lobing...@gmail.com> wrote: > On Sunday, May 17, 2015 at 7:06:34 PM UTC+2, Isaiah wrote: >> >> >> (please file issues...) >> > > Just a side question: If i run wrap_c on one of the larger system > libraries (e.g. X11.h) and get an error; is it more likely it uses a > rather esoteric cpp features (and therefore isn't covered in the current > wrap_c) OR is it more likely that i miss some compiler flags? >
