cfbolz changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://quodlibet.duckdns.org/irc/pypy/latest.log.html#irc-end ) | use cffi for calling C | if a pep adds a mere 25-30 [C-API] functions or so, it's a drop in the ocean (cough) - Armin
<mattip>
arigato: rather than sys.is_untranslated, I used an environment variable in 42f538e60a6b
<mattip>
does that seem reasonable?
oberstet has joined #pypy
jacob22 has quit [Read error: Connection reset by peer]
jacob22 has joined #pypy
todda7 has quit [Ping timeout: 265 seconds]
<agronholm>
how far along is pypy's 3.7 support?
<arigato>
mattip: sched_param(sched_priority=0): I think we should just support that for all lib_pypy/_structseq objects, there shouldn't be an issue, right?
<arigato>
mattip: 42f538e60a6b: this is obscure and a "security risk", please just use monkey-patching
<arigato>
mattip: in fact, a better idea would be to have WE_ARE_TRANSLATED = True in the module, and temporarily monkey-patch it to False both in test_app_main.py and in "if __name__=='__main__'" in app_main.py
<arigato>
for test_app_main.py, use monkeypatch.setattr()
todda7 has joined #pypy
* arigato
tries to fix this
Ai9zO5AP has joined #pypy
andi- has quit [Ping timeout: 260 seconds]
<mattip>
thanks
<mattip>
agronholm: basically ready for a release I think
<mattip>
the win32 support is still not running lib-python tests to completion, I would like to try to clear that problem
<mattip>
and I am hoping the hpy support will be merged before the release
<agronholm>
that's great to hear
<agronholm>
I set the minimum Python version requirement according to the latest pypy3 release in my projects
andi- has joined #pypy
andi- has quit [Excess Flood]
andi- has joined #pypy
<antocuni>
mattip: do you have an ETA for the release?
<antocuni>
depending on when it is, we might be able to include some hpy support
<antocuni>
ronan: are you still actively working on the branch? If not, I can do it
todda7 has quit [Ping timeout: 272 seconds]
<antocuni>
another question is whether we WANT to include hpy in the upcoming release; the API is still changing relatively fast, so it is likely that the _hpy_universal included in the release would become obsolete very soon
<mattip>
whatever you guys feel
<mattip>
I may just give up on win32, but I would like to understand why the py3.7 lib-python run times out
<mattip>
so that is why I am trying to get own tests working on top of pypy
<mattip>
arigato: 1443744a7f56 won't work for TestInteraction and TestNonInteractive
<mattip>
since they are run as "hostpython path/to/app_main.py <extra args>"
<mattip>
or did I miss something?
<antocuni>
mattip: I think it's a good idea to merge the hpy branch into py3.6, and continue the development there (possibily with more sub-branches). This way, we can have continuous nighlies with latest hpy support
<mattip>
ahh, yes I missed the assignment in __main__ at the bottom
<mattip>
+1
lritter has joined #pypy
<mattip>
it's not like there are tons of hpy c-extensions out there that will break because of an old version of hpy :)
<antocuni>
yes, having hpy included in the next official release is probably a no-op. If you want to try hpy, you will very likely need a nightly; if you don't care about hpy, its presence is harmless
<antocuni>
so, I vote for: let's do the release whenever it's ready; if hpy has already been merged, that's good. If not, too bad
Rhy0lite has joined #pypy
otisolsen70 has joined #pypy
epsilonKNOT has quit [Ping timeout: 240 seconds]
epsilonKNOT_ has joined #pypy
jcea has joined #pypy
epsilonKNOT_ has quit [Read error: Connection reset by peer]
epsilonKNOT has joined #pypy
<ronan>
antocuni: I haven't done anything since Thursday, but I'll continue today
<mattip>
I would like it to be in for the release so we can generate some buzz around HPy, even if it is a no-op
<mattip>
optics are important
todda7 has joined #pypy
YannickJadoul has joined #pypy
<YannickJadoul>
Hi all! Do you know about CMake's problem with generating correct extension module filenames?
<YannickJadoul>
Cfr. https://gitlab.kitware.com/cmake/cmake/-/issues/21070; "Here, I think the pypy implementation is wrong. As clearly stated by PEP 3149, SOABI is the module extension without the shared library suffix specific to the platform (i.e. .so or whatever)."
Smigwell has joined #pypy
dddddd has quit [Ping timeout: 240 seconds]
Alex_Gaynor has quit [Ping timeout: 240 seconds]
altendky has quit [Ping timeout: 240 seconds]
Taggnostr has quit [Remote host closed the connection]
<YannickJadoul>
mattip, LarstiQ: Something else I just thought of. Does this just need to be coordinated with wheel? Isn't it setuptools which determines the name of the .so file?