Hi, The patch below has been floating in dev@ for about a month. It looks looks like an improvement to me, but it is way out of my comfort zone.
@Joe Orton <jor...@redhat.com> are you able to test this? Cheers, Daniel Den tors 19 juni 2025 kl 14:06 skrev Yasuhito FUTATSUKI < futat...@yf.bsdclub.org>: > Hello, > > On 2025/06/19 20:30, Branko Čibej wrote: > > On 19. 6. 25 12:42, Joe Orton wrote: > >> On Thu, Jun 19, 2025 at 10:26:43AM +0200, Branko Čibej wrote: > >>> On 19. 6. 25 10:22,jor...@apache.org wrote: > >>>> Author: jorton > >>>> Date: Thu Jun 19 08:22:56 2025 > >>>> New Revision: 1926575 > >>>> > >>>> URL:http://svn.apache.org/viewvc?rev=1926575&view=rev > >>>> Log: > >>>> Fix SWIG test cases for Python 3.14. > >>>> > >>>> The interpreter internally avoids some reference count modifications > when > >>>> loading objects onto the operands stack by borrowing references when > >>>> possible. This can lead to smaller reference count values compared to > >>>> previous Python versions. > >>>> > >>>> https://docs.python.org/3.14/whatsnew/3.14.html#whatsnew314-refcount > >>> How come all the other getrefcount()s in repository.py aren't affected? > >> I have no idea, sorry, but all the tests pass under 3.14 with this > >> applied. > > > > Yes, I verified that (3.14.0b3 via pyenv). Well, as long as it works ... > > Then, I've tweak repository.py without using magic number of refcount > and adding comments. Although I didn't test it on Python 3.14, but > it is expected to work. > > -- > Yasuhito FUTATSUKI <futat...@yf.bsdclub.org> >
