[PATCH] Fix bug in svn_client_commit.

[Ben Reser <be...@reser.org>]

This looks correct to me.  But the only question I have is what
result_rev should be returned to the client if there are externals?
Might be worth mentioning in the documentation that it will only return
the revision of what you asked to check out from the url not any
externals that checkout might also get.

Ohh and this is definitely a 1.0 canidate.

[[[
Fix a bug that caused result_rev to never be set for calls to
svn_client_checkout.

* subversion/libsvn_client/checkout.c
  (svn_client__checkout_internal): Pass result_rev to the calls to 
    svn_client_update.
]]]

-- 
Ben Reser <be...@reser.org>
http://ben.reser.org

"Conscience is the inner voice which warns us somebody may be looking."
- H.L. Mencken

Re: [PATCH] Fix bug in svn_client_commit.

[Justin Erenkrantz <ju...@erenkrantz.com>]

--On Saturday, December 27, 2003 12:08 AM -0800 Ben Reser <be...@reser.org> 
wrote:

> This looks correct to me.  But the only question I have is what
> result_rev should be returned to the client if there are externals?

It'll only return the revnum stored in the .svn/entries file - that is, it'll 
be the revision of the current working directory.  This isn't svnversion, 
which will try to determine mixed-revision WCs.  ;-)

> Might be worth mentioning in the documentation that it will only return
> the revision of what you asked to check out from the url not any
> externals that checkout might also get.

Hmm.  I'd tend to say not because then we'd have to explain the mixed-version 
WC and I think that'd be too confusing to explain in a little docstring...

> Ohh and this is definitely a 1.0 canidate.

Anyhoo, committed in r8096 as I think C-Mike just overlooked this case when he 
did r7799.  And, I'll add r8096 to the 1.0 list.  -- justin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org