On Nov 30, 2004, at 7:52 AM, Alexander Kitaev wrote:
> Hello Ben,
>
> Thank you very much I do understand it now. JavaSVN now behaves
> incorrectly,
> destroying entry that was deleted and commited immediately after
> commit,
> which results in obsolete "delete foo" in following update, that
> doesn't
> breaks anything but doesn't reflect the real state of the working copy
> as
> well. The problem begins when javaSVN reports those "deleted" entries
> created by real svn client as "missing" in status because I didn't new
> about
> this aspect of svn behaviour, and now I know how to fix it, thanks!
>
I don't want to discourage you, but this is just one *tiny* little
aspect of the complex behaviors of libsvn_wc. There are dozens and
dozens of complicated bookkeeping edge-cases like this, and they're all
covered by our python automated test suite.
This is reason we've been telling people for years that it would be
essentially impossible to reimplement Subversion from scratch in Java
(or other languages). It took us *years* to find all of these
edge-cases. I certainly admire your effort and resolve... but be
prepared for a very long road to compatibility and proper WC behavior.
;-)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Nov 30 15:11:11 2004