You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by cm...@collab.net on 2002/04/23 06:48:05 UTC
Re: svn commit: rev 1746 - trunk/subversion/include trunk/subversion/tests/clients/cmdline trunk/subversion/libsvn_ra_dav
philip@tigris.org writes:
> Author: philip
> Date: 2002-04-22 20:26 GMT
> New Revision: 1746
>
> Modified:
> trunk/subversion/include/svn_error_codes.h
> trunk/subversion/libsvn_ra_dav/commit.c
> trunk/subversion/libsvn_ra_dav/props.c
> trunk/subversion/libsvn_ra_dav/ra_dav.h
> trunk/subversion/tests/clients/cmdline/commit_tests.py
> Log:
> Fix issue 644, when adding a file ensure that no file with the same URL
> already exists.
Two questions:
* should I be expecting all the python tests to pass over ra_dav
right now (cuz they're not)? commit_test 5 is failing for me on
A/D/H/chi, which is supposed to be "replaced".
* does this change take into account the possibility of files being
"replaced" as opposed to "added"? it seems to me that a property
lookup on something that's being replaced will succeed, indicating
that the item does exist ... but i didn't see in the patch where we
might have noted that we *knew* it existed, and we just finished
deleting it prior to re-adding it.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: rev 1746 - trunk/subversion/include trunk/subversion/tests/clients/cmdline trunk/subversion/libsvn_ra_dav
Posted by cm...@collab.net.
cmpilato@collab.net writes:
> philip@tigris.org writes:
>
> > Author: philip
> > Date: 2002-04-22 20:26 GMT
> > New Revision: 1746
> >
> > Modified:
> > trunk/subversion/include/svn_error_codes.h
> > trunk/subversion/libsvn_ra_dav/commit.c
> > trunk/subversion/libsvn_ra_dav/props.c
> > trunk/subversion/libsvn_ra_dav/ra_dav.h
> > trunk/subversion/tests/clients/cmdline/commit_tests.py
> > Log:
> > Fix issue 644, when adding a file ensure that no file with the same URL
> > already exists.
>
> Two questions:
>
> * should I be expecting all the python tests to pass over ra_dav
> right now (cuz they're not)? commit_test 5 is failing for me on
> A/D/H/chi, which is supposed to be "replaced".
>
> * does this change take into account the possibility of files being
> "replaced" as opposed to "added"? it seems to me that a property
> lookup on something that's being replaced will succeed, indicating
> that the item does exist ... but i didn't see in the patch where we
> might have noted that we *knew* it existed, and we just finished
> deleting it prior to re-adding it.
I double-checked the patch, and yeah, it's broken with respect to
"Replace" operations, I believe. I've got a fix for it that I'm
running through 'make check' and python tests over DAV now.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org