On Wed, 18 Sep 2013, Peter Stuge wrote:
Daniel only cares about what he himself uses.
That's not what I said.
"I rather defer all the build stuff on build other systems than my
own to others"
Exactly. To me, there's a big difference between those two statements. I
_care_ about all builds, but I prefer to let people who actually use and know
the other build systems to do the work on them.
We could also have a single C file with all of the code, full of #ifdef, and
not have any build system at all. That is even more ridiculous than
compiling files which are *known* to be *empty*..
Yes, and we can do silly remarks here forever and never get anywhere. That's
also an option.
This sentence doesn't make sense to me. I've implemented a solution which
is independent of autotools. Did you look at the patch I sent today?
No, I missed that. But yeah, I figure using a set of different .inc files like
that is perfectly fine for non-autotools builds as then they can include the
correct set. It still means slightly more work for the lazy MSVC "include all
files into a project" people but at least we offer a fairly easy solution.
Do you prefer that I apply it on top of the existing master, or that I
revert d512b25 and squash the new commit into d512b25 so that the entire
change is in a single commit and that we don't have the broken Makefile.inc
state in-between?
I'm not sure I understand. What you've already pushed cannot be changed, so
just go ahead apply on master.
--
/ daniel.haxx.se
_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel