On Fri, Aug 08, 2014 at 05:46:53AM -0400, Tyler Winter Mace wrote:
> Wow, this is surprising.
>
> If i think back to when we moved to Subversion, i remember how always it was
> to do these long urls. In CVS, there was defined a CVSROOT that handled any
> pathing like hostname or local directory.
On 08.08.2014 07:12, Ryan Schmidt wrote:
> On Aug 7, 2014, at 1:20 PM, Ben Reser wrote:
>> Our test suite does not test this scenario.
> Wouldn't that be a good first thing to do?
I'm sure we'd be happy to look at a patch to our test suite that allows
automated testing over HTTP on the server roo
Wow, this is surprising.
If i think back to when we moved to Subversion, i remember how always it was to
do these long urls. In CVS, there was defined a CVSROOT that handled any
pathing like hostname or local directory. We found the typing in a URL to be
error prone. To minimize error we wanted
On Aug 7, 2014, at 1:20 PM, Ben Reser wrote:
>
> Our test suite does not test this scenario.
Wouldn't that be a good first thing to do?
On 8/7/14 8:50 AM, Stefan Sperling wrote:
> Oh geez... the cobwebs...
>
> I honestly don't recall, but I strongly suspect that it has something to
> do with how the mirroring filters work. These things do essentially a
> global search-and-replace on the request and response bodies and
> headers,
On Wed, Aug 06, 2014 at 03:18:25PM -0400, Tyler Winter Mace wrote:
> I am adding a mirror repository, mirroring an http-based repository. That
> repository uses a shortest-possible naming scheme (""). For
> example:
> http://svnhost/repos/trunk/main.c
>
> Now that I'm trying to mirror it, I fi
