Mailman 3 June 2017 - cython-devel

0.25.2b0 with 32bit fixes: (sometimes) Interrupted system call in builtin_subtype_methods_T653
by Yaroslav Halchenko July 6, 2017

July 6, 2017

A fresh one I ran into while rebuilding with fixes for 32bit (thanks BTW! sorry for not prompt followup/gratitude) ====================================================================== ERROR: runTest (__main__.CythonRunTestCase) compiling (cpp) and running builtin_subtype_methods_T653 ---------------------------------------------------------------------- Traceback (most recent call last): File "runtests.py", line 998, in run self.run_tests(result, ext_so_path) File "runtests.py", line 1013, in run_tests self.run_doctests(self.module, result, ext_so_path) File "runtests.py", line 1023, in run_doctests run_forked_test(result, run_test, self.shortDescription(), self.fork) File "runtests.py", line 1076, in run_forked_test cid, result_code = os.waitpid(child_id, 0) OSError: [Errno 4] Interrupted system call ---------------------------------------------------------------------- Ran 581 tests in 188.225s FAILED (errors=1) it is not 100% reproducible. Rebuilding the package went smooth without this hiccup. Just wanted to let you know might be related that the tests were ran as set -e; for P in 2.7 3.5; do \ PYTHONPATH=`/bin/ls -d /build/cython-0.25.2~b0/build/lib.*-$P` \ /usr/bin/python$P runtests.py --shard_count=16 \ --no-refnanny -v -v --exclude="(parallel|Debugger|annotate_html)" --work-dir=build/work-dir 2>&1; \ done so testing in parallel to speed up the build of the package. This one was the only failure I have ran into but it built fine before (but there could have been some changes in the core packages since then) -- Yaroslav O. Halchenko Center for Open Neuroscience http://centerforopenneuroscience.org Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755 Phone: +1 (603) 646-9834 Fax: +1 (603) 646-1419 WWW: http://www.linkedin.com/in/yarik

2 2

Cython 0.26 release
by Robert Bradshaw July 3, 2017

July 3, 2017

I'd like to officially start off the release process with an alpha candidate. This does necessarily contain everything that will be in the final release, but given the amount of time that has passed I would recommend everyone test this out with their projects sooner rather than later. Download: https://github.com/cython/cython/archive/0.26.alpha0.zip Initial changelog: https://github.com/cython/cython/blob/0.26.alpha0/CHANGES.rst Please feel free to respond to this thread or file bugs as necessary. Thanks, Robert

5 14

Next Cython Release
by Julian Rüth June 27, 2017

June 27, 2017

Hello Cython developers, some people over at conda-forge would like to use features that are already in master but have not been released yet (https://github.com/conda-forge/cython-feedstock/pull/20). We would prefer not to patch and older release for Cython but just wait for the next release. However, since there has not been a release in 6 months, we're wondering if there are any plans for a release? julian PS: Not sure if this is the right list for this but the link to the bug-tracker on https://mail.python.org/mailman/listinfo/cython-devel goes to http://cython.org/trac/cython_trac/ which produces a 404.

6 8

On cdef and cpdef discrepencies
by Robert Bradshaw June 23, 2017

June 23, 2017

On Thu, Jun 22, 2017 at 1:25 AM, Jeroen Demeyer <jdemeyer(a)cage.ugent.be> wrote: > On 2017-06-21 21:51, Robert Bradshaw wrote: >> >> I'll see if I can put together a patch. > > > For what? For Sage? For Cython? I was thinking for Sage. I was assuming there was just an oddball collection of oversights accumulated over the yeras. > Anyway, I think that coping with this new error message is not particularly > hard, just a lot of work. That's why I would prefer a warning instead of an > error. That way, we can easily see how bad the problem is and what needs to > be fixed and we fix it gradually (say, before Cython 0.27 which could make > this a hard error). I was digging into this a bit more, and the issue seems to be entirely the arithmetic methods like _add_, and was introduced one year and one day ago: https://github.com/sagemath/sage/commit/e1d2ba46574273ec4d358e654ed195a82b8… . It'd be cleaner if these were in fact cpdef methods, but our category method injection framework requires that the class not have an _add_ defined at the Python level. But any overrides must be cpdef methods lest a (Python) call to x._add_(y) succeed but not find the override. Perhaps it'd be more clear to define cpdef Element._add_ whose implementation directly calls getattr_from_category rather than (<object>self)._add_. - Robert

1 0