David,
On 2004 May 20, , at 14.20, David Berry wrote:
> again and this time I just left it running. After about 20 mins it came
> back with some output and seems to have done its stuff. I confess I am
> only on a 54kb/s broadband connection but 20mins seems a bit long!
It does seem a bit long, yes. I think there's potentially quite a lot
of chatter between the CVS client and server, but I thought they tried
quite hard to avoid transferring more than they had to. Try adding a
-z<number> option to the CVS command. That turns on compression of
some type, but I've never actually played with it. I don't know what
<number> is best, but I've seen -z3 in documentation.
> Anyway, I have not managed to merge the branch into my checkout of the
> trunk. I noticed this in the output from cvs update...
>
>
> cvs server: warning: config.h.in is not (any longer) pertinent
>
> and
>
> cvs server: conflict: starconf.status is modified but no longer in the
> repository
It looks like it worked.
Both of these are generated files (the first from autoheader, called
from autoreconf, and the second from starconf), so I removed them when
I was doing the pointset.h fix the other day. The update has probably
just removed the files. Try doing 'cvs update' in the trunk checkout
(not mentioning any branch) -- that should bring your (trunk) checkout
up to date with respect to the repository, which should also tidy up
the CVS/Entries file.
In general, it seems, branch merges report more conflicts than are
really there. 'cvs -n update' should report the real conflicts, and of
course building things will expose any problems of that type.
See you,
Norman
--
------------------------------------------------------------------------
---
Norman Gray
http://www.astro.gla.ac.uk/users/norman/
Physics and Astronomy, University of Glasgow, UK
[log in to unmask]
|