| |
| |
On 3/30/22 17:21 imacarthur wrote:
On Wednesday, 30 March 2022
at 15:51:57 UTC+1 Lauri wrote:
On
Wed, 30 Mar 2022 01:48:40 -0700 (PDT)
Manolo wrote:
> Indeed, the hybrid Wayland/X11 library could use
cairo-only rendering for
> its X11 backend.
> Let's see what other developers think.
I personally think that having the *option* to use Cairo rendering
for the X11 backend would be great (and Manolo did already post a
prototype) but I never thought of removing the "plain old X11"
option for the upcoming FLTK 1.4.x releases.
Is it any large maintenance overhead, the X drawing? In the
interest of
size, we'd like to keep X drawing available and cairo not
mandatory.
(We'd of course also disable wayland)
FLTK 1.4.0 must be a straight forward step from 1.3.x (currently
1.3.8 which will likely be the last 1.3.x release unless we pick up
the few additional commits and make a 1.3.9 final release or we need
one for serious bug fixes or whatever).
Therefore 1.4.0 will consequently have to be as much backward
compatible (API wise) as possible. All main additional features, for
instance Wayland support and Cairo rendering on X11 *must* IMHO be
options that can be switched on or off selectively but we need to
make a decision about each option's default setting.
I'm thinking about making the release default to the hybrid
Wayland/X11 system but then users would still be able to select X11
only. The reason I would like to have this as the default is that I
believe that Linux distributions would be "guided" to use the hybrid
system automatically (all options == default) which is what we're
discussing here. But the final decision about this is to be
discussed when it is available.
Lauri makes a fair point - I too have a handful of
"embedded" systems running some or other cut-down Linux, and
in general they only have X11 as a render option (at least one
actually has an xf86 server, albeit now running on arm, but I
think that gives a sense of how "old" some of these
environments are...)
So losing the X11 backend would not be ideal, though of
course I still have 1.3...
I think there's no need to worry about losing the standard X11
backend, at least for the 1.4.x series. There's no real reason to
strip it off or replace it. In fact to do that would take much more
time and testing than we have to release 1.4.0. Sure, we still need
some time. The schedule to release 1.4.0 during April 2022 was
probably a bit too optimistic, but anyway: 1.4.0 should be released
ASAP and I'm working on it as much as I can do. Having the Wayland
port in 1.4 is IMHO worth a delay.
For the near future I hope that we can release more often but I
don't think that we should change or remove any options during the
1.4.x release series.
However, I'm also suggesting to make the current Pango option the
default on Linux/X11 because it is the best text rendering we can
have and it would reduce questions about fonts not rendering asian
and other text properly. IIRC this would also be what the Wayland
platform uses and would thus give us the most compatibility between
the two Linux platforms.
Additionally I could imagine to remove the (mostly 1.x) option to
build FLTK w/o Xft which is really ugly from today's point of view
that I think we can leave this option to 1.3.x users.
But sorry, this is going OT here, although it's slightly related to
the build options of a "hybrid" FLTK library. Too many options make
the maintenance much more difficult and I believe it's OK to remove
old, probably not often used options in favor of the new (Pango,
Cairo, Wayland) variants.
--
You received this message because you are subscribed to the Google Groups "fltk.coredev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fltkcoredev+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fltkcoredev/40409d83-62d4-4f4c-27b0-04d0a021e7d9%40online.de.
[ Direct Link to Message ] | |
|
| |
