You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Joe Orton <jo...@redhat.com> on 2015/04/10 10:25:06 UTC
Re: [PATCH] Fix swig-py bindings with SWIG 3.0.4
On Wed, Feb 11, 2015 at 02:36:31PM +0100, Branko Čibej wrote:
> This is clearly a bug in Swig; I can't imagine it being anything else.
> The only workaround I could find was to remove the comments entirely.
Same here. Filed as https://github.com/swig/swig/issues/379
> Even then, the generated Python bindings don't work because of the use
> of the argument-less svn_pool_create() in the wrong context, but I think
> this is a different bug.
This is https://github.com/swig/swig/issues/377 but there are a number
of similar issues with default argument handling in recent SWIG 3.0.x so
that might be a dupe. Has anybody found a workaround for this?
Regards, Joe
Re: [PATCH] Fix swig-py bindings with SWIG 3.0.4
Posted by Branko Čibej <br...@wandisco.com>.
On 10.04.2015 17:44, Andreas Stieger wrote:
> Am 10.04.2015 10:25 schrieb Joe Orton <jo...@redhat.com>:
>> On Wed, Feb 11, 2015 at 02:36:31PM +0100, Branko Čibej wrote:
>>> This is clearly a bug in Swig; I can't imagine it being anything else.
>>> The only workaround I could find was to remove the comments entirely.
>> Same here. Filed as https://github.com/swig/swig/issues/379
>>
>>> Even then, the generated Python bindings don't work because of the use
>>> of the argument-less svn_pool_create() in the wrong context, but I think
>>> this is a different bug.
>> This is https://github.com/swig/swig/issues/377 but there are a number
>> of similar issues with default argument handling in recent SWIG 3.0.x so
>> that might be a dupe. Has anybody found a workaround for this?
> Unfortunately not. I tried some swig master patches that looked relevant but in the end decided to wait for a fixed upstream release, while disabling the swig-py tests meanwhile.
I went back to using the latest Swig 2.x for bindings tests.
-- Brane
Re: [PATCH] Fix swig-py bindings with SWIG 3.0.4
Posted by Andreas Stieger <an...@gmx.de>.
Am 10.04.2015 10:25 schrieb Joe Orton <jo...@redhat.com>:
>
> On Wed, Feb 11, 2015 at 02:36:31PM +0100, Branko Čibej wrote:
> > This is clearly a bug in Swig; I can't imagine it being anything else.
> > The only workaround I could find was to remove the comments entirely.
>
> Same here. Filed as https://github.com/swig/swig/issues/379
>
> > Even then, the generated Python bindings don't work because of the use
> > of the argument-less svn_pool_create() in the wrong context, but I think
> > this is a different bug.
>
> This is https://github.com/swig/swig/issues/377 but there are a number
> of similar issues with default argument handling in recent SWIG 3.0.x so
> that might be a dupe. Has anybody found a workaround for this?
Unfortunately not. I tried some swig master patches that looked relevant but in the end decided to wait for a fixed upstream release, while disabling the swig-py tests meanwhile.
Andreas