Re: [AD] Cygwin ports

[Sven Sandberg <svsa1977@xxxxxxxxxx>]

hstokset@xxxxxxxxxx wrote:
> >In that
> >case, wouldn't it be more logical to install headers to
> >/usr/include/mingw32/ (instead of /usr/local/include, as it currently
> >does)?
> 
> probably. (but only if ALLEGRO_USE_CYGWIN is defined)

The attached patch does this to makefile.mgw (tested with Dev-C++ and
Cygwin). I also attach a patch proposing some clarifications in
mingw32.txt.

Btw, is this -mwindows flag really needed? I can compile+link demo.c
without it...

> >Also, would it be worth trying to port the Unix version of
> >Allegro to "native" Cygwin?
> 
> i don't think so. why put extra (cygwin) licensing restrictions on the
> allegro library?

If I understand http://cygwin.com/licensing.html correctly (in
particular the last four paragraphs), it merely requires Allegro's
license to comply with the open source definition, not with GPL. And if
I understand http://www.opensource.org/docs/definition_plain.html
correctly, Allegro's license _is_ compliant (assuming that "compliant"
is weaker than "approved"; I don't know if this is what they mean).

> not to mention OS calls will be slower due to an
> extra emulation layer that is not really needed since there's already
> a native win32 port of allegro.

I think it would mainly be useful for porting Allegro programs requiring
Unix to windows.

> in the case where the user have a need for the cygwin DLL for his
> program he can still link with the allegro DLL (-mno-cygwin). no
> problem.

I'm not sure I understand this: If the user has a need for the cygwin
DLL, then (s)he'd rather link with the cygwin DLL :-). And if I
understand the first sentence of
http://www.xraylith.wisc.edu/~khan/software/gnu-win32/
mno-cygwin-howto.txt correctly, -mno-cygwin excludes this. Am I missing
something?

-- 
Sven Sandberg    svsa1977@xxxxxxxxxx    home.student.uu.se/svsa1977

Attachment: makefile.mgw.diff.gz
Description: GNU Zip compressed data

Attachment: mingw32.txt.diff.gz
Description: GNU Zip compressed data