Re: Subversion / swig1.3
* John Lenz [Tue, 26 Oct 2004 23:02:43 +0000]:
> On 10/26/04 16:35:35, Frank Lichtenheld wrote:
> >On Tue, Oct 26, 2004 at 01:39:15PM +0200, Philipp Hug wrote:
> >> subversion depends on swig >= 1.3.22-2 which is in unstable. but it's
> >> in testing yet because swig cannot be put into testing because it would
> >> break subversion 1.0.6-2 ;-)
> >> did I miss something or is this just a bug in the testing script and
> >> manual hinting?
> >Yes, it needs manual hinting and we're aware of this. But it needs to
> >wait for libhid anyway (two days), so there is no hurry to install the
> >necessary hint.
> I am a swig developer lurking on this list. Building the SWIG runtime
> library (which is what causes subversion to depend on SWIG) has been
> depreciated since 1.3.20, which was released almost a year ago. About a
> week ago, I removed the ability to build the runtime libraries in SWIG CVS,
> and so in the next version (1.3.23) you will not be able to build them at
> all. We are planning for a release in a week or two.
I am forwarding this information to the subversion maintainer(s), just
to make sure it doesn't get lost.
> I am not sure how this will impact debian, but I would seriously encourage
> you to build the python swig wrappers and every other package that depends
> on libswigruntime so they don't require the runtime library, and then
> remove that library. The runtime library leads to several bugs and
> problems, and if those get reported after sarge is released... I am
> actually surprised to learn people are still using runtime libraries. You
> can completly remove libswig*.so and not lose any functionality, and avoid
> a whole bunch of potential bugs.
EM: asp16 [ykwim] alu.ua.es | PK: DA6AE621
Arguing with an engineer is like wrestling with a pig in mud: after a
while, you realize the pig is enjoying it.