Rob Savoye wrote:
Jens Petersen wrote:
Hmmm, pkgconfig doesn't seem an unreasonable dependency to me,
considering that libxml2, gtk2, cairo, pango, glib2, and gtkglext all
support it.
Right. But those are new dependencies for the GTK version of the
player. pkgconfig isn't supported on all platforms though, for instance
it would make cross-compiling with MingW or VC++ impossible, or other
embedded platforms like eCOS or VxWorks. Gnash isn't just for the
desktop...
Ok, fair point. :)
Anyway the main problem with current configure file seems to be for
the gtk family of libs. This is what configure gives on my FC5 x86_64
box:
[snip]
Configured paths for x86_64-redhat-linux-gnu are:
[examples of misconfig snipped]
The simple pkgconfig hack patch I posted in my original mail at least
makes it build on x86_64 for me on both FC4 and FC5 (and on i386 too
of course).
All this has added is the path to the config files, which I've never
needed to do to build Gnash on any platforms.
Sorry, not sure what you mean here? (I was going to comment in the
previous mail that x86_64 Linux seems an important arch for gnash on
the desktop.:)
They should probably be
included though, so I'll add them to the macros. I don't mind a hack
that'll use pkgconfig if present, and then drop into my own tests if
not. I'll look into it when I add these additional include paths.
Anyway, I agree using pkgconfig if available and otherwise falling
back to something like the current configure tests should make most
people happy. :)
Thanks, Jens
_______________________________________________
Gnash mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnash
