Re: emacs for pure Gtk3

[Jeff Walsh <fejfighter@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 4244 bytes --]

> 
*From*: Eli Zaretskii *Subject*: Re: emacs for pure Gtk3 *Date*: Mon, 
27 Apr 2020 18:03:54 +0300
> >/ Date: Mon, 27 Apr 2020 21:37:27 +0900 (JST)/
> >/ Cc: address@hidden/
> >/ From: Yuuki Harano <address@hidden>/
> >/ /
> >/ >   . You don't seem to have a copyright assignment on file.  This 
> would/
> >/ >     be a significant contribution to Emacs, for which we must 
> have/
> >/ >     such an assignment from you before bringing this code into 
> the/
> >/ >     Emacs repository.  Would you be willing to start the legal/
> >/ >     paperwork now?  If so, I will send you the form to fill./
> >/ /
> >/ Yes.  Please send me it./
> 
> Form sent off-list.
> 
> >/ By the way, this fork contains much code written by @fejfighter./
> >/ He said "for now: Yes I do agree to assign my code to the FSF." 
> here,/
> >/ <https://github.com/masm11/emacs/pull/11#issuecomment-600856858> ./
> >/ What to do?/
> 
> He should fill the form I sent to you, separately, and email it
> according to instructions.

No worries, happy to do so.

> 
> >/ Because I was not going to merge to mainline when I started 
> porting,/
> >/ older commit messages are in Japanese.  If you don't like Japanese/
> >/ messages, I can make one big commit instead of existing commits./
> 
> That's probably the best.  But there's time before that happens, and
> you can meanwhile keep the original log messages while the code is on
> the branch.
> 
> >/ Since pgtk emacs is configured with '--without-x', existing X code/
> >/ is disabled.  If configured with '--with-x', the existing X 
> support/
> >/ should be enabled as before./
> 
> Would configuring --with-x disable Pgtk support code?  That is, do the
> X and Pgtk support contradict each other, and cannot live in the same
> binary?  Or maybe I don't have a clear idea what exactly gets disabled
> when building with Pgtk -- can you elaborate?
> 
/I think this may need a little more work in configure.ac/
/In essence it's not that different to --with-ns or --with-w32./
/it just happens to re-use a chunk of the gtkutil.c code where 
possible./
/
/
/Effectively it selects pgtkterm.h instead of xterm.h/



> >/ Pgtk emacs supports X window system too through Gtk library./
> >/ It can handle Wayland, X window system, and TTY in the same 
> session./
> >/ But segmentation fault may occur when running on X and Wayland/
> >/ in the same session./
> 
> I guess those segfaults need to be fixed, because having a GUI Emacs
> that can only run on Wayland would be a limitation that users might be
> unhappy about?

it will run on wayland or xwayland or X11 from the same binary, but not 
on wayland and X11 concurrently.

I'm not sure of a use-case for this, but I'm hoping someone is able to 
provide one.

> 
> >/ I don't know about Lisp threads.  I have never supported it 
> explicitly./
> >/ Pgtk emacs may not support it./
> 
> Well, for starters see if test/src/thread-tests.el runs and succeeds
> in your Pgtk build.

I get:

make lisp/thread-tests
make[1]: Entering directory '/home/fejfighter/dev/emacs-gtk/test'
  GEN      lisp/thread-tests.log
Running 3 tests (2020-04-28 12:52:55+1000, selector `(not (tag 
:unstable))')
  skipped  1/3  thread-tests-list-threads-error-when-not-configured 
(0.000133 sec)
   passed  2/3  thread-tests-thread-list-send-error (0.000488 sec)
   passed  3/3  thread-tests-thread-list-show-backtrace (0.014913 sec)

Ran 3 tests, 2 results as expected, 0 unexpected, 1 skipped (2020-04-28 
12:52:55+1000, 0.015803 sec)

1 skipped results:
  SKIPPED  thread-tests-list-threads-error-when-not-configured

make[1]: Leaving directory '/home/fejfighter/dev/emacs-gtk/test'


which matches a master build checkout from 
e49d3a45cd4a0554aa98c45f0976ed513c500951 (approx 1300 Aus EST)



> 
> >/ Gtk supports w32, but I have never tested on w32./
> 
> This would be nice, but is much less important, IMO.  The most
> important task is to keep users of Posix systems happy with the Pgtk
> build.
> 
> Thanks.

FWIW, this seems to work fine with gtk-broadway, the in-browser 
implementation of GTK3, aside from keystroke-clashes with firefox.
it may be some indication of a minimally "X'd" emacs, but a w32 version 
would be better proof.

Regards,

Jeff Walsh
(fejfighter)






[-- Attachment #2: Type: text/html, Size: 7903 bytes --]