>
> I've dived into the CVS code and see where the error is coming from,
> and it's not a non-printing character as I originally suspected.
> There's a note that the original author thought about reporting
> precisely where the offending character is but considered portability
> issues and rejected that idea.
>
> I'm in the process of trying to use the CVS RCS code to parse an RCS
> ",v" file to see the individual elements and hope to find the culprit
> that way (and I'm a captive audience and don't immediately have to
> think portability, so I'll add some debug stuff to the CVS RCS code).
> If I have any luck I'll report here
My experience was:
1) made a copy of the actual CVS repository to another location, so I
didn't affect the original .v files
2) execute the cvs2svn, finding the offending files. Some where
invalid end of line characters (which I removed manually). The other
problem I had was a .v file actually had 2 initial versions. So,
somehow, someone used CVS to have two initial revision records in the
.v file. The comments and dates were exactly the same, and I don't
know how it was done. But, it was easy for me to remove the duplicate
record, and then I could progress.
So, I was able to progress without macking the cvs code itself. So I
can't help too much in that regards.
Good luck!
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: users-help_at_subversion.tigris.org
Received on 2008-01-07 04:18:33 CET