Re: [Cython] longintrepr.h issues with MinGW

2018-10-24 Thread Ian Henriksen
It's been a while since I've used MinGW to compile any extensions. The CRT incompatibility is a hassle to worry about unless you are using a custom Python build and MSVC 2015 and 2017 are pretty good. That said, there is a workaround for this one. See https://stackoverflow.com/a/30881190/1935144.

Re: [Cython] longintrepr.h issues with MinGW

2018-10-23 Thread Jeroen Demeyer
On 2018-10-24 07:58, Jeroen Demeyer wrote: If you care about MinGW, there is also this CPython issue which makes it basically impossible to compile any Cython extension on MinGW: https://bugs.python.org/issue11566 I forgot to say that this makes it impossible to compile any *C++* Cython exten

Re: [Cython] longintrepr.h issues with MinGW

2018-10-23 Thread Jeroen Demeyer
If you care about MinGW, there is also this CPython issue which makes it basically impossible to compile any Cython extension on MinGW: https://bugs.python.org/issue11566 ___ cython-devel mailing list cython-devel@python.org https://mail.python.org/mai

Re: [Cython] longintrepr.h issues with MinGW

2018-10-23 Thread Stefan Behnel
Robert Bradshaw schrieb am 24.10.2018 um 01:44: > On Tue, Oct 23, 2018 at 10:22 PM Stefan Behnel wrote: >> My guess is that MinGW is just fairly rarely used to build CPython >> extensions overall. >> >>> but perhaps we should even consider releasing bugfixes for previous >>> Cython versions. >> >>

Re: [Cython] longintrepr.h issues with MinGW

2018-10-23 Thread Robert Bradshaw
On Tue, Oct 23, 2018 at 10:22 PM Stefan Behnel wrote: > > Robert Bradshaw schrieb am 22.10.2018 um 11:03: > > Given that https://bugs.python.org/issue4709 results in extension > > modules that seem to work, but silently produce completely incorrect > > answers, I'm thinking we should either disabl

Re: [Cython] longintrepr.h issues with MinGW

2018-10-23 Thread Stefan Behnel
Robert Bradshaw schrieb am 22.10.2018 um 11:03: > Given that https://bugs.python.org/issue4709 results in extension > modules that seem to work, but silently produce completely incorrect > answers, I'm thinking we should either disable our long-unpacking code > on these platforms, or at the very le