Re: Nested sit-for's

[Kim F. Storm]

Chong Yidong <address@hidden> writes:

> address@hidden (Kim F. Storm) writes:
>
>>> It would work to have ONE timer that does sit-for if we make a rule
>>> that no others can do so.  We could define jit-lock as this one
>>> exception.  (This has the advantage of not involving any change in the
>>> code, just comments and the Lisp Manual.)
>>
>> I agree with your analysis.  
>>
>> But, IMO, if we make it a rule that timers should generally not use
>> sit-for, then a central function like jit-lock should definitely not
>> use sit-for!
>
> If we simply document that "timers (and process filters) should avoid
> using sit-for", it should be clear to the reader that rare exceptions
> may exist (especially if we add a comment to jit-lock-stealth-fontify
> stating this).  After the release, we can probably rework
> jit-lock-stealth-fontify to avoid using sit-for, but I don't think the
> current situation is bad enough to block the release.
>
> OTOH, I don't remember any other timers or process filters in the
> Emacs tree that use a long sit-for or loop waiting for input.  Anyone
> know of any?

[We have discussed this pb before, but never found a solution].

I don't know if it is related, but from time to time (at least a few
times daily), the cursor disappears, and doesn't return until I
do "something" (e.g. click the mouse or type a key).

I use blinking cursor, but when this happens, it typically seems to
happen immediately after some event that updates part of the screen
(like buffer switching), i.e. not as an effect of first showing the
cursor and then "blinking it off".  Examples where this happens:

1) returning to the group buffer in Gnus (after closing a summary buffer).
[this is where I've seen this most]

2) returning to the summary buffer after responding to an article.

But I also think I've seen it hiding the cursor after typing some
characters...

I don't know if it is related to timers at all, but it seems to
happen (marginally) more often now than before the recent sit-for
changes ...

-- 
Kim F. Storm <address@hidden> http://www.cua.dk