RE: MinGW-w64 Compatibility

[Hoyt, David]

Bruno Haible wrote:
> Can you please provide a minimal reproducible sample? That is, can you
> give an input file that, when compiled with gnulib's <sys/socket.h>, yields
> an error? Also, please show the compiler command-line, and the error that
> you get.

Everyone's blaming someone else, so I honestly don't know who to believe. I've 
had the same issue compiling gnutls (despite what Ozkan said -- see 
http://comments.gmane.org/gmane.network.gnutls.general/2124), however, so it 
may be that the problem's in 2 places. Could be there's something wrong w/ my 
setup (but I don't think so). I don't know gnulib and don't have time to put 
together an example testing it. Perhaps for now you can consider this an issue 
to look into when (if) you have the time. If I get more time, I'd be happy (of 
course) to put together an example.

> Among the first things I would try then is to see whether the error also
> occurs on mingw.
> 
> I'm asking also because Ozkan Sezer blamed neon, not gnulib, for the error
> you were seeing [1].

It may or may not work w/ mingw.org (I haven't tried). I do know they don't 
have near the same amount of Windows API coverage that MinGW-w64 has (and 
therefore potentially fewer conflicts/issues - not b/c what they have is 
necessarily right, they just haven't added enough coverage for the issues to 
arise). And it's been my experience that their headers are often very much not 
in line with Microsoft's (IOW, theirs is highly modified and inconsistent with 
platform SDK headers). These days I avoid it as much as possible.

I do appreciate the effort in thinking about the issue and researching it.

BTW, I apologize if the word "BLOCKED" appears in any URL I provide -- our 
email filter does that to prevent people where I work from inadvertently 
clicking on bad links.

Thanks,
- David Hoyt