You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Daniel Shahaf <d....@daniel.shahaf.name> on 2008/10/02 17:28:42 UTC
Re: svn commit: r33396 - trunk/subversion/tests/libsvn_subr
lgo@tigris.org wrote on Thu, 2 Oct 2008 at 07:57 -0700:
> Author: lgo
> Date: Thu Oct 2 07:57:22 2008
> New Revision: 33396
>
> Log:
> Manually merged the remaining part of r33211 from the windows_dirents branch
> to trunk.
>
...which finishes the windows_dirents work. Thanks!
> * subversion/tests/libsvn_subr/dirent_uri-test.c
> (svn_dirent_is_absolute): Add some testcases.
^
s/absolute/ancestor/
But more importantly, r33211 also made copy_tests 74 pass on Windows.
(The test is XFail on trunk, but passes on the branch. My fault for not
making this clearer at the time I committed the fix.) I think the
proper fix now is just:
Index: subversion/libsvn_wc/copy.c
===================================================================
--- subversion/libsvn_wc/copy.c (revision 33373)
+++ subversion/libsvn_wc/copy.c (working copy)
@@ -334,8 +334,8 @@ get_copyfrom_url_rev_via_parent(const char *src_pa
/* Don't look for parent_path in src_access if it can't be
there... */
- if (svn_path_is_ancestor(svn_wc_adm_access_path(src_access),
- parent_path))
+ if (svn_dirent_is_ancestor(svn_wc_adm_access_path(src_access),
+ parent_path))
{
SVN_ERR(svn_wc_adm_retrieve(&parent_access, src_access,
parent_path, pool));
but I don't have time to look further now. (I'll commit it myself in a
few days, if I'm not beaten to it.)
Daniel
>
> Modified:
> trunk/subversion/tests/libsvn_subr/dirent_uri-test.c
>
> Modified: trunk/subversion/tests/libsvn_subr/dirent_uri-test.c
> URL: http://svn.collab.net/viewvc/svn/trunk/subversion/tests/libsvn_subr/dirent_uri-test.c?pathrev=33396&r1=33395&r2=33396
> ==============================================================================
> --- trunk/subversion/tests/libsvn_subr/dirent_uri-test.c Thu Oct 2 07:49:09 2008 (r33395)
> +++ trunk/subversion/tests/libsvn_subr/dirent_uri-test.c Thu Oct 2 07:57:22 2008 (r33396)
> @@ -773,8 +773,10 @@ test_dirent_is_ancestor(const char **msg
> { "X:/foo", "X:/", FALSE},
> { "X:/", "X:/foo", TRUE},
> { "X:", "X:foo", TRUE},
> + { SVN_EMPTY_PATH, "C:/", FALSE},
> #else /* WIN32 or Cygwin */
> { "X:", "X:foo", FALSE},
> + { SVN_EMPTY_PATH, "C:/", TRUE},
> #endif /* non-WIN32 */
> };
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
RE: Adding svn_dirent to the Windows build (was: Re: svn commit:
r33396 - trunk/subversion/tests/libsvn_subr)
Posted by Daniel Shahaf <d....@daniel.shahaf.name>.
> > > > With the above patch, the libsvn_wc dll fails to link, even
> > > > after I add the missing '#include "svn_dirent_uri.h"' and add
> > > > svn_dirent_uri.h to build.conf. How do I fix this?
> > >
> svn_dirent_uri.h was added to msvc-export in r33460, but editing the
> build.conf file with gen_make.py didn't recreate the .def
>
> This build issue is resolved in r33461, by adding a input dependency on
> build.conf for the .def files.
>
> Bert
And the original patch was committed in r33462.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
RE: Adding svn_dirent to the Windows build (was: Re: svn commit: r33396 - trunk/subversion/tests/libsvn_subr)
Posted by Bert Huijben <be...@vmoo.com>.
> -----Original Message-----
> From: Daniel Shahaf [mailto:d.s@daniel.shahaf.name]
> Sent: zondag 5 oktober 2008 18:54
> To: Bert Huijben
> Cc: dev@subversion.tigris.org
> Subject: RE: Adding svn_dirent to the Windows build (was: Re: svn
> commit: r33396 - trunk/subversion/tests/libsvn_subr)
>
> > > With the above patch, the libsvn_wc dll fails to link, even after I
> add
> > > the missing '#include "svn_dirent_uri.h"' and add svn_dirent_uri.h
> to
> > > build.conf. How do I fix this?
> >
> > Adding the #include is probably enough and regenerating the makefiles
> (via
> > gen_make.py) should handle the rest of the dependency tracking.
> >
> > What compilation/linking error do you get?
> >
>
> See: http://paste.lisp.org/display/67979
svn_dirent_uri.h was added to msvc-export in r33460, but editing the
build.conf file with gen_make.py didn't recreate the .def
This build issue is resolved in r33461, by adding a input dependency on
build.conf for the .def files.
Bert
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
RE: Adding svn_dirent to the Windows build (was: Re: svn commit:
r33396 - trunk/subversion/tests/libsvn_subr)
Posted by Daniel Shahaf <d....@daniel.shahaf.name>.
> > With the above patch, the libsvn_wc dll fails to link, even after I add
> > the missing '#include "svn_dirent_uri.h"' and add svn_dirent_uri.h to
> > build.conf. How do I fix this?
>
> Adding the #include is probably enough and regenerating the makefiles (via
> gen_make.py) should handle the rest of the dependency tracking.
>
> What compilation/linking error do you get?
>
See: http://paste.lisp.org/display/67979
> (The dirent code is part of libsvn_subr which is linked to libsvn_wc)
>
> Bert
>
>
Thanks,
Daniel
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
RE: Adding svn_dirent to the Windows build (was: Re: svn commit: r33396 - trunk/subversion/tests/libsvn_subr)
Posted by Bert Huijben <be...@vmoo.com>.
> -----Original Message-----
> From: Daniel Shahaf [mailto:d.s@daniel.shahaf.name]
> Sent: zondag 5 oktober 2008 18:34
> To: dev@subversion.tigris.org
> Subject: Adding svn_dirent to the Windows build (was: Re: svn commit:
> r33396 - trunk/subversion/tests/libsvn_subr)
>
> Daniel Shahaf wrote on Thu, 2 Oct 2008 at 20:28 +0300:
> > But more importantly, r33211 also made copy_tests 74 pass on Windows.
> > (The test is XFail on trunk, but passes on the branch. My fault for
> not
> > making this clearer at the time I committed the fix.) I think the
> > proper fix now is just:
> >
> > Index: subversion/libsvn_wc/copy.c
> >
> ===================================================================
> > --- subversion/libsvn_wc/copy.c (revision 33373)
> > +++ subversion/libsvn_wc/copy.c (working copy)
> > @@ -334,8 +334,8 @@ get_copyfrom_url_rev_via_parent(const char
> *src_pa
> >
> > /* Don't look for parent_path in src_access if it can't be
> > there... */
> > - if
> (svn_path_is_ancestor(svn_wc_adm_access_path(src_access),
> > - parent_path))
> > + if
> (svn_dirent_is_ancestor(svn_wc_adm_access_path(src_access),
> > + parent_path))
> > {
> > SVN_ERR(svn_wc_adm_retrieve(&parent_access,
> src_access,
> > parent_path, pool));
>
> With the above patch, the libsvn_wc dll fails to link, even after I add
> the missing '#include "svn_dirent_uri.h"' and add svn_dirent_uri.h to
> build.conf. How do I fix this?
Adding the #include is probably enough and regenerating the makefiles (via
gen_make.py) should handle the rest of the dependency tracking.
What compilation/linking error do you get?
(The dirent code is part of libsvn_subr which is linked to libsvn_wc)
Bert
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Adding svn_dirent to the Windows build (was: Re: svn commit: r33396
- trunk/subversion/tests/libsvn_subr)
Posted by Daniel Shahaf <d....@daniel.shahaf.name>.
Daniel Shahaf wrote on Thu, 2 Oct 2008 at 20:28 +0300:
> But more importantly, r33211 also made copy_tests 74 pass on Windows.
> (The test is XFail on trunk, but passes on the branch. My fault for not
> making this clearer at the time I committed the fix.) I think the
> proper fix now is just:
>
> Index: subversion/libsvn_wc/copy.c
> ===================================================================
> --- subversion/libsvn_wc/copy.c (revision 33373)
> +++ subversion/libsvn_wc/copy.c (working copy)
> @@ -334,8 +334,8 @@ get_copyfrom_url_rev_via_parent(const char *src_pa
>
> /* Don't look for parent_path in src_access if it can't be
> there... */
> - if (svn_path_is_ancestor(svn_wc_adm_access_path(src_access),
> - parent_path))
> + if (svn_dirent_is_ancestor(svn_wc_adm_access_path(src_access),
> + parent_path))
> {
> SVN_ERR(svn_wc_adm_retrieve(&parent_access, src_access,
> parent_path, pool));
With the above patch, the libsvn_wc dll fails to link, even after I add
the missing '#include "svn_dirent_uri.h"' and add svn_dirent_uri.h to
build.conf. How do I fix this?
Daniel
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: r33396 - trunk/subversion/tests/libsvn_subr
Posted by Lieven Govaerts <sv...@mobsol.be>.
Daniel Shahaf wrote:
> lgo@tigris.org wrote on Thu, 2 Oct 2008 at 07:57 -0700:
>> Author: lgo
>> Date: Thu Oct 2 07:57:22 2008
>> New Revision: 33396
>>
>> Log:
>> Manually merged the remaining part of r33211 from the windows_dirents branch
>> to trunk.
>>
>
> ...which finishes the windows_dirents work. Thanks!
>
>> * subversion/tests/libsvn_subr/dirent_uri-test.c
>> (svn_dirent_is_absolute): Add some testcases.
> ^
> s/absolute/ancestor/
>
> But more importantly, r33211 also made copy_tests 74 pass on Windows.
> (The test is XFail on trunk, but passes on the branch. My fault for not
> making this clearer at the time I committed the fix.) I think the
> proper fix now is just:
>
> Index: subversion/libsvn_wc/copy.c
> ===================================================================
> --- subversion/libsvn_wc/copy.c (revision 33373)
> +++ subversion/libsvn_wc/copy.c (working copy)
> @@ -334,8 +334,8 @@ get_copyfrom_url_rev_via_parent(const char *src_pa
>
> /* Don't look for parent_path in src_access if it can't be
> there... */
> - if (svn_path_is_ancestor(svn_wc_adm_access_path(src_access),
> - parent_path))
> + if (svn_dirent_is_ancestor(svn_wc_adm_access_path(src_access),
> + parent_path))
> {
> SVN_ERR(svn_wc_adm_retrieve(&parent_access, src_access,
> parent_path, pool));
>
> but I don't have time to look further now. (I'll commit it myself in a
> few days, if I'm not beaten to it.)
Right. I wanted to ask you what that change was for, couldn't really
find it.
FYI, I'm currently creating all svn_dirent and svn_uri functions and
once that API is finished I'll start to migrate libsvn_wc,
libsvn_client, the ra layers one by one. Just want to make sure first
that the new API is correct before committing to it in 1.6.
Lieven
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org