Python 3.x and IRIX, state of affairs - Printable Version
+- IRIX Network Forums (//forums.irixnet.org)
+-- Forum: SGI/MIPS (//forums.irixnet.org/forum-3.html)
+--- Forum: Development/Porting (//forums.irixnet.org/forum-9.html)
+--- Thread: Python 3.x and IRIX, state of affairs (/thread-2190.html)
+- IRIX Network Forums (//forums.irixnet.org)
+-- Forum: SGI/MIPS (//forums.irixnet.org/forum-3.html)
+--- Forum: Development/Porting (//forums.irixnet.org/forum-9.html)
+--- Thread: Python 3.x and IRIX, state of affairs (/thread-2190.html)
Python 3.x and IRIX, state of affairs - Raion - 06-04-2020
Since I've recently been exploring Python 2.7.x on IRIX, I encountered PEP 11: https://www.python.org/dev/peps/pep-0011/
Paraphrasing info here, we've got some problems:
1. IRIX 4.x support (not that this is a major concern) was removed in 2.x days.
2. All IRIX threading retired in 3.2; and all support for IRIX by 3.7.
Unlike Python 2.7, Python 3.x source appears to make a lot of GCC-ism related assumptions and there's major refactoring to be done.
Based on what I'm seeing, I think a fair target is Python 3.6. It should run python-related scons, ninja/meson etc. stuff for some time to come, and if we can get a decent build out, that'll probably pacify the immediate need.
Similar to my opinion on QT5, GTK3, etc. I don't think anything newer is worth pursuit. Latest and greatest is never been the aim of nekoware or other projects. We want a reliable base of software we can count on.
Once we get Python 2.7.18 sorted, my suggestion would be to get a Python 3.6 going and we can use that as our refpoint.