Bert and I independently arrived at this patch, which he and Paul are testing:
Index: subversion/svnserve/serve.c
===================================================================
--- subversion/svnserve/serve.c (revision 1064439)
+++ subversion/svnserve/serve.c (working copy)
@@ -2998,9 +2998,9 @@
svn_path_remove_components(url_buf,
svn_path_component_count(b->fs_path->data));
b->repos_url = url_buf->data;
- b->authz_repos_name = svn_fspath__is_child(root, repos_root, pool);
+ b->authz_repos_name = svn_dirent_is_child(root, repos_root, pool);
if (b->authz_repos_name == NULL)
- b->repos_name = svn_fspath__basename(repos_root, pool);
+ b->repos_name = svn_dirent_basename(repos_root, pool);
else
b->repos_name = b->authz_repos_name;
b->repos_name = svn_path_uri_encode(b->repos_name, pool);
On 01/28/2011 10:09 AM, C. Michael Pilato wrote:
> "We'll put our best man on it."
>
> /me makes sad puppy dog eyes at pburba.
>
> On 01/28/2011 07:33 AM, Philip Martin wrote:
>> Philip Martin <philip.martin_at_wandisco.com> writes:
>>
>>> cmpilato_at_apache.org writes:
>>>
>>>> Author: cmpilato
>>>> Date: Fri Jan 28 02:34:40 2011
>>>> New Revision: 1064436
>>>>
>>>> URL: http://svn.apache.org/viewvc?rev=1064436&view=rev
>>>> Log:
>>>> Reintegration the 'uris-as-urls' branch into trunk.
>>>> (Merged /subversion/branches/uris-as-urls:r1060426-1064427.)
>>>>
>>>> This should wrap up issue #2028.
>>>
>>> The Windows buildbot is failing over ra_svn and it seems to have started
>>> after this commit:
>>>
>>> http://ci.apache.org/waterfall?show=svn-slik-w2k3-x64-ra
>>>
>>> Not much other info is available. Anyone got a Windows environment to
>>> test?
>>
>> I think the first test to use the server, an import, is causing svnserve
>> to crash:
>>
>> http://ci.apache.org/builders/svn-slik-w2k3-x64-ra/builds/1617/steps/Test%20fsfs+svn/logs/testlog
>>
>
>
--
C. Michael Pilato <cmpilato_at_collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on 2011-01-28 16:26:47 CET