Re: Question about test failure on Hydra

[Eli Zaretskii]

> From: Stephen Berman <address@hidden>
> Cc: address@hidden
> Date: Sun, 30 Jul 2017 22:23:32 +0200
> 
> On Sun, 30 Jul 2017 18:59:34 +0300 Eli Zaretskii <address@hidden> wrote:
> 
> > Doesn't dired-auto-revert-buffer by default use some kind of file
> > notifications?  If so, this could be some race between the
> > notification coming in and the rest.  You could introduce a wait to
> > try to solve that, or maybe try turning off file notifications.
> 
> I stepped through dired-revert and followed the call chain to
> insert-directory but didn't see anything that seemed related to file
> notifications.

No, you are right, this option has nothing to do with file
notifications: we always revert Dired buffers "by hand", whenever one
of them is visited.

So, back to the original issue:

> But the sexp (dired test-dir) should (and in all my test runs does)
> put point on the first "nontrivial" file (i.e. it skips "." and
> "..").

In your case, does (dired test-dir) reads the directory anew, or does
it revert an existing Dired buffer?  I think the place to look for the
problem depends on the answer to this question.