Re: file modifications get lost due to FAT 2s time resolution

Ben Reser wrote:
> On Thu, Jul 15, 2004 at 12:46:36PM -0400, Mark Phippard wrote:
>
> But it's the same thing that
> CVS does and I suspect nobody had a beter solution at the time.
>

cvs records the time when it modifies a file
and sleeps until that time passes:

code snippet from cvs commit.c:

int
commit (argc, argv)
[...]
     /* see if we need to sleep before returning to avoid time-stamp
races */
     if (last_register_time)
     {
        sleep_past (last_register_time);
     }
[...]

and 'last_register_time' is updated whenever
a relevant file is modified.

but the rest of the cvs sourcecode
really gives me the creeps...
it even seems to have the same flaw on FAT,
but does not trigger it easily because
its 'slow enough' anyway.

:-)
======
c.a.t.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Jul 15 21:57:42 2004