r10233 broke the swig-python build

[Max Bowsher <ma...@ukf.net>]

r10233 was a change to the perl bindings. However, some of it affected the
build of all swig bindings. The python bindings are now broken, because of a
circular dependency between libsvn_swig_py and swig_core.

One solution would be to build the swig runtime code into libsvn_swig_*, not
into swig_core.

Max.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: r10233 broke the swig-python build

[Max Bowsher <ma...@ukf.net>]

Russell Yanofsky wrote:
> Max Bowsher wrote:
>> r10233 was a change to the perl bindings. However, some of it
>> affected the build of all swig bindings. The python bindings are now
>> broken, because of a circular dependency between libsvn_swig_py and
>> swig_core.
>>
>> One solution would be to build the swig runtime code into
>> libsvn_swig_*, not into swig_core.
>
> r10236 should fix this problem. It does keep the runtime in swig_core
> though.

That's good enough for unix, but Cygwin's still broken.

Could you give a short summary of the reason for these re-designs, so I can
be sure not to work against your goals as I try to fix the bindings on
Cygwin?

Thanks,

Max.

Re: r10233 broke the swig-python build

[Russell Yanofsky <re...@columbia.edu>]

Max Bowsher wrote:
> r10233 was a change to the perl bindings. However, some of it
> affected the build of all swig bindings. The python bindings are now
> broken, because of a circular dependency between libsvn_swig_py and
> swig_core.
>
> One solution would be to build the swig runtime code into
> libsvn_swig_*, not into swig_core.

r10236 should fix this problem. It does keep the runtime in swig_core
though.

- Russ



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org