Re: Slow to calculate correct indent column after changes above with 5.32.2

[Alan M. <ac...@mu...>]

Hi, John.

On Fri, Mar 30, 2012 at 12:03:52PM -0400, John Ruttenberg wrote:
> This one is intermittent and hard to duplicate, so I'd like to know if it
> sounds like an instance of a known bug before spending a lot of time on it.
> I am using 5.32.2 with FSF emacs 23.4.1 on linux, editing java.  Sometimes I
> make a change and it takes a while before newline-and-indent finds the
> right column.

As a matter of interest, how long is "a while"?  Slightly sluggish
(~0.25s), noticeable delay (1s), intolerable delay (4s), or even worse?

> Saving and reverting the buffer always solves this.  Going up in the
> file  and finding the last place that it is right by trial and error
> also seems to correct.

In otherwords, once an area of code gets into this state, it stays there
absent specific action to fix it.

> Does this sound at all familiar?

I've done a lot of optimisation, mainly for C++, since 5.32.2 was
released.  To be honest, 5.32.3 is somewhat overdue.  The problem you're
having sounds kind of vaguely familiar, but only vaguely.

What I'd like to do, if it's OK with you, is to send you an up-to-date
snapshot tarball of CC Mode and ask you to try that out.  It may be that
the bug you're getting could have been fixed already.

-- 
Alan Mackenzie (Nuremberg, Germany).