Re: cvs2svn.py: more co problems on Windows

On 04 Feb 2004 12:24:09 -0600
kfogel@collab.net wrote:

> Could you make an annotation in issue #1718 about this?

Done.

> I don't think
> we need a separate issue for it (it's really more a matter of
> documenting the prerequisites better)

Still, it's not only a matter of prerequisites for a co replacement. The
current cvs2svn.py does not work for CVSNT users... (it's an easy fix,
but only if you know a little Python and a little about CVS, so not all
users will be able to work around it).

About issue 1315 (cvs2svn.py does not detect the exit status of co and
other subprocesses it runs):

I ran cvs2svn.py. It ran "fine". Checked through ViewCVS. No differences
between different versions. Checked the contents. All files contained "0
files copied". Puzzlement galore, till I remembered my COMSPEC points to
4NT.EXE and I have a 4NT alias co=copy /r. Took care of that.

I ran cvs2svn.py. No contents. Double-ration puzzlement till I
remembered that in my system, CVSROOT=:sspi:mymachine:d:\my\path, and
since I switched to Subversion I don't run the CVSNT service by default
anymore. Took care of that.

OK, it worked fine. Did some things, posted the co -q bug report, went
to sleep.

This morning I tried to repeat the process... You bet: ten minutes lost
till I remembered to start CVSNT service again!

I'd sure like to get an error: "Sorry, when I tried to run co, it spat
on you and had nothing good to say about your ancestry, either", or
equivalent, so I could easily (and quickly) understand that co was
having a bad day.

                                                                Juanma

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Feb 5 10:58:08 2004