From: Paul Shirley <Paul AT chocolat DOT obvious DOT fake DOT foobar DOT co DOT uk>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Make "Clock Skew" problem.
Date: Sun, 10 May 1998 12:54:20 +0100
Organization: wot? me?
Message-ID: <qWScGCAsVZV1EwwZ@foobar.co.uk>
References: <SaZS1EAGs1U1Ewfk AT foobar DOT co DOT uk>
 <Pine DOT SUN DOT 3 DOT 91 DOT 980510115319 DOT 16252B-100000 AT is>
NNTP-Posting-Host: chocolat.foobar.co.uk
Mime-Version: 1.0
Lines: 32
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

In article <Pine DOT SUN DOT 3 DOT 91 DOT 980510115319 DOT 16252B-100000 AT is>, Eli Zaretskii
<eliz AT is DOT elta DOT co DOT il> writes
>Amazing!  So what you are telling me is that the time stamp of the
>file changes during its life even if you don't access it in any way,
>right?
>
>Could you please post the test progarm you used to show this behavior,
>if it is short enough to be posted?

The test program is w95jed with debugging calls added (compiled with VC4
so it won't work well with djgpp anyway. I don't think posting it would
be popular ;)
One additional fact: I have backup file creation enabled so conceivably
Windoze could be returning timestamps from the original renamed file and
then the new one. Of course that would be an even more frightening bug
and I'm inclined not to believe it.

>
>> This is a problem because Jed uses the timestamp to detect files changed
>> by other process's/users. Its bloody annoying being wrongly told all
>> your files have changed after every edit/compile cycle.
>
>Strange: Emacs does such checks as well, but I don't think I ever saw
>this test misfire on Windows.

It does not always happen with Jed, there's an annoying sense of
randomness to the problem. It does seem to happen consistently with some
files, the direct cause is the timestamp, what causes that is still a
mystery. Of course it could just be a problem with the VC libc.

---
Paul Shirley: my email address is 'obvious'ly anti-spammed