On Wed, Feb 7, 2018 at 8:53 PM Kenneth Porter wrote:
> --On Thursday, February 08, 2018 2:30 AM + Troy Curtis Jr
> wrote:
>
> > If that package was being imported, I'd definitely expect that kind of
> > behavior, but merely it's presence on your system should not be enough to
> > actually ca
--On Thursday, February 08, 2018 2:30 AM + Troy Curtis Jr
wrote:
If that package was being imported, I'd definitely expect that kind of
behavior, but merely it's presence on your system should not be enough to
actually cause your python scripts to use the "future" behavior within
python 2.
--On Thursday, February 08, 2018 2:27 AM + Troy Curtis Jr
wrote:
I believe that Kenneth is actually referring to putting the type(fp) test
in svn/fs.py, where it is doing the prefixing internally, not that it is
leaking out into his application. I think he was simply trying to debug
our di
On Wed, Feb 7, 2018 at 11:54 AM Kenneth Porter
wrote:
> --On Wednesday, February 07, 2018 9:35 AM -0800 Kenneth Porter
> wrote:
>
> > So there's a builtins package hiding in this system somewhere.
>
> Found it. The system has the python2-future package which is a dependency
> of the certbot pack
On Wed, Feb 7, 2018 at 1:06 PM Daniel Shahaf wrote:
> Kenneth Porter wrote on Wed, 07 Feb 2018 09:35 -0800:
> > The overload of type was happening due to the unprefix stuff in fs.py.
> Red
> > herring, though interesting.
>
> This should only happen with 1.7 and earlier, see
> https://subversion.
Julian Foad writes:
> It looks like the behaviour changes in 1.10 are intentional and Good.
>
> I think the test suite should be updated to accept the old error when
> the server is old, like this:
I suppose we could change the testsuite along these lines, but I don't
see it as something we have
Kenneth Porter wrote on Wed, 07 Feb 2018 09:35 -0800:
> The overload of type was happening due to the unprefix stuff in fs.py. Red
> herring, though interesting.
This should only happen with 1.7 and earlier, see
https://subversion.apache.org/docs/release-notes/1.8#swig-py-star
Cheers,
Daniel
--On Wednesday, February 07, 2018 9:35 AM -0800 Kenneth Porter
wrote:
So there's a builtins package hiding in this system somewhere.
Found it. The system has the python2-future package which is a dependency
of the certbot package. I think I got that from the epel repo.
Here's the metadata
The overload of type was happening due to the unprefix stuff in fs.py. Red
herring, though interesting.
I distilled the failure down to this foo.py program. Of interest is that
"import builtins" succeeds on a Python 2.5 system. It should be throwing
unless Python is v3. So there's a builtins p
Here's an interesting clue. I added "print type(fp)" after the temp file is
opened to see what kind of file object was returned by the open call and
got this traceback. How is the type function getting overloaded with
svn_fs_type? So far I haven't been able to recreate this with a short
hello-w
On 2/6/2018 7:56 PM, Troy Curtis Jr wrote:
Kenneth, I'm having trouble reproducing your issue. Any other hints
at what might causing the trouble in your environment that you can
think of? I've tried changing my locale, changing the files diffed to
being utf8, all with no luck. Regardless you
Julian Foad writes:
> Philip Martin wrote:
>> First I ran the 1.9 testsuite against 1.10 servers
>> over http:// and svn://. [...]
>
> Philip, please can you tell me how you did this?
>
> [...]
>> I also ran the 1.10 testsuite against 1.9 servers over http:// and
>> svn://. In order to run the
12 matches
Mail list logo
