bug#15614: [Sébastien Gross] Patch correctif bug#15614

bug#15614: [Sébastien Gross] Patch correctif bug#15614

[Stefan Monnier]

> (let ((read (read-event nil window-system seconds)))

That's an interesting workaround.  Tho you'd probably want the reverse:

 (let ((read (read-event nil (not window-system) seconds)))

after all, you're probably working in a GUI frame and you want nil.
The patch that fixes bug#14782 used t because that's what was needed
under a tty.

> Since window-system is nil on tty and set on others it should fix both
> bugs.  What do you think? I will do diligent test if you're OK with this.

It won't fix your test case under a tty, but it should fix your test
case in GUI frames while still keeping bug#14782 fixed (which only bit
in tty frames).

>> A good workaround for your particular problem would be to change
>> flyspell so as not to use sit-for (use an idle timer instead).
> I don't mind doing that. However it seems like a rubber band rather than
> a fix.

I did say "workaround".  But it would be a good thing in any case, since
the use of `sit-for' in there introduces various other problems.


        Stefan

bug#15614: [Sébastien Gross] Patch correctif bug#15614

[Ivan Kanis]

Hi Stefan,

Thanks for fixing it! I just checked it fixed #16701.

I was going to prepare a patch for you. I was reading #15614 and was
surprised to find out that you had done something.

Since the two bugs are merged, don't you think I should get some sort of
notification?

Take care,

Ivan

bug#15614: [Sébastien Gross] Patch correctif bug#15614

[Sébastien Gross]

Hi guys,

I just test it with a brand new emacs compilation and as far as I could
test, bug #15614 seems to be fixed.

Good work,

Thanks


Cheers,
S.

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> I was going to prepare a patch for you. I was reading #15614 and was
>> surprised to find out that you had done something.
>> Since the two bugs are merged, don't you think I should get some sort of
>> notification?
>
> Indeed, I think you should have, since #16701 is marked as "done",
> according to http://debbugs.gnu.org/cgi/bugreport.cgi?bug=16701.
> The mysteries of Debbugs, I guess.
>
>
>         Stefan
>

-- 
Sébastien Gross