Re: Optional support for GDI+ on Windows (emacs-28)

[Alan Third <alan@idiocy.org>]

On Tue, Mar 31, 2020 at 08:41:29PM +0300, Eli Zaretskii wrote:
> > Date: Tue, 31 Mar 2020 18:57:52 +0200 (CEST)
> > From: Alan Third <alan@idiocy.org>
> > Cc: Juan José García-Ripoll <juanjose.garciaripoll@gmail.com>,
> > 	emacs-devel@gnu.org
> > 
> > On Tue, Mar 31, 2020 at 07:47:38PM +0300, Eli Zaretskii wrote:
> > > 
> > > Like I said: this must not be a compile-time condition, we should
> > > decide whether GDI+ is supported at runtime, and we should provide
> > > variables to control whether GDI+ is used for each supported image
> > > format.  HAVE_GDIPLUS should guard code which uses GDI+, but it should
> > > NOT decide whether that code is actually used.
> > 
> > May I request that however the above is implemented, it can be
> > extended to the NS port’s use of NSImage vs libpng, etc. as well?
> 
> Can you elaborate?  I don't think I understand the request, probably
> because I know very little about how NS deals with images in Emacs.

Simply that since the proposed patch works largely the same as the NS
port, if possible leave any implementation generic. That is, rather
than create a variable called w32-png-use-gdi-plus, perhaps name it
something like image–png-use-native.

If it’s not possible then that’s fine.
-- 
Alan Third