pcl-cvs signaling obsolete dirs as "new-dir Need-Update"

pcl-cvs signaling obsolete dirs as "new-dir Need-Update"

[Daniel Bonniot]

Hi,

In version 22.1.1: when running cvs-examine, pcl-cvs reports
directories with only files in the Attic as "new-dir Need-Update".
This is wasn't the case in earlier versions (21?) and the new behavior
is very counter-productive.

This issue has already been discussed there (for 22.0.50):
http://www.gatago.org/gnu/emacs/help/9708831.html
It includes a patch, but it seems to be in the 22.1 base and the
problem is still there, and I actually don't see how it was related to
this behaviour.

Cheers,

Daniel

Re: pcl-cvs signaling obsolete dirs as "new-dir Need-Update"

[Stefan Monnier]

> In version 22.1.1: when running cvs-examine, pcl-cvs reports
> directories with only files in the Attic as "new-dir Need-Update".
> This is wasn't the case in earlier versions (21?) and the new behavior
> is very counter-productive.

> This issue has already been discussed there (for 22.0.50):
> http://www.gatago.org/gnu/emacs/help/9708831.html
> It includes a patch, but it seems to be in the 22.1 base and the
> problem is still there, and I actually don't see how it was related to
> this behaviour.

It includes more than a patch, it includes the solution:

   Yes, it's annoying, but it can be very useful if it really means what
   it says.  To get rid of the warning, add the entry to the .cvsignore
   file (with a final /).

The problem really is in CVS itself.


        Stefan

Re: pcl-cvs signaling obsolete dirs as "new-dir Need-Update"

[Daniel Bonniot]

>    Yes, it's annoying, but it can be very useful if it really means what
>    it says.  To get rid of the warning, add the entry to the .cvsignore
>    file (with a final /).
>
> The problem really is in CVS itself.

Right. I just checked, the output of cvs is exactly the same whether
this is a new directory with files in it or with only an Attic. So
unless there is an option to make cvs give more information, the only
option would be to make several requests, which would be unacceptable
latency-wise. Or to fix CVS :-O

So thanks, the .cvsignore workaround is fine. And thanks a lot for
your work on this mode.

Daniel