Hideshow

[Hrvoje Niksic]

"Stephen J. Turnbull" <stephen at xemacs.org> writes:

>>>>>> "Klaus" == Klaus Berndl <klaus.berndl at sdm.de> writes:
>
>     Klaus> So: Why not moving the libs timer.el and thingatpt.el to
>     Klaus> another XEmacs-package?
>
> I agree about timer.el; APIs like run-at-time are going to be
> developed over and over if we don't provide them.

But there is a problem.  The last sentence of run-at-time's docstring
says:

    This function returns a timer object which you can use in
    `cancel-timer'.

This means that run-at-time is not standalone.  It requires at least
cancel-timer.  And if we have cancel-timer, people will expect the
rest of the timer functionality, which brings us back to fsf-compat
land.

How about creating a function similar to `run-at-time' that works
natively with itimers and returns an itimer?  We should, of course,
call it something other than `run-at-time', lest we confuse people.
Maybe `run-at-interval'?