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
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
jcea has quit [Ping timeout: 260 seconds]
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
alexge50 has quit [Quit: ZNC 1.8.2 - https://znc.in]
alexge50 has joined #pypy
phoe6 has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
<bbot2> Started: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-64/builds/49 [mattip: force build, py3.7]
<bbot2> Started: http://buildbot.pypy.org/builders/own-win-x86-64/builds/128 [mattip: force build, py3.7]
_whitelogger has joined #pypy
17WAAOU5H has joined #pypy
_whitelogger has joined #pypy
17WAAOU5H has quit [Quit: 17WAAOU5H]
_whitelogger has joined #pypy
<bbot2> Exception: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-64/builds/49 [mattip: force build, py3.7]
alexge50 has quit [Ping timeout: 240 seconds]
wleslie has joined #pypy
oberstet has joined #pypy
<bbot2> Failure: http://buildbot.pypy.org/builders/own-win-x86-64/builds/128 [mattip: force build, py3.7]
<cfbolz> mattip: probably not expected no
<mattip> strange. I think the tick across the board on benchmarks from about ten days ago is due to me changing the machine configuration
<mattip> I turned off turbo to see if it stabalized pyperf runs, but it had the side effect of slowing down the machine in general
<mattip> so I reset it to "use turbo"
<mattip> and today's runs are back to "normal"
<cfbolz> welcome to benchmarking :-(((((
isidentical has joined #pypy
<mattip> arigato: nice blog post. Keeping my fingers crossed someone will take up the challenge
_whitelogger has joined #pypy
lritter has joined #pypy
jcea has joined #pypy
<arigato> mattip: yes, we'll see
<isidentical> I was checking out how to start hacking on pypy, but apparently I am missing a step. Is there a way of interactively working on the pypy3 core, without actually re-building everything from scratch (takes hours)?
<antocuni> isidentical: yes, you can start ./bin/pyinteractive.py (probably you want to use -S to avoid importing site.py which is slow)
<antocuni> however, we usually develop by running "AppTests", which are the equivalent of pyinteractive but inside a test session
<antocuni> see e.g. pypy/objspace/std/test/test_boolobject.py
<antocuni> and also apptest_boolobject.py
<isidentical> Ah, thanks for the hints! Will check them out
<antocuni> they are two different styles of writing apptests, but they are mostly the same under the hood
<cfbolz> isidentical: yay! hi!
<cfbolz> isidentical: we usually don't rebuild very often
<cfbolz> mostly we write apptests
<isidentical> hey cfbolz!
<cfbolz> isidentical: any particular thing you're planning to hack on?
<isidentical> I first want to check the general layout out, but after that no not really.
<isidentical> I don't know the general workflow, but maybe I can help on some of the 3.8 features
<antocuni> isidentical: inside pyinteractive you can also press CTRL-C to switch from applevel to interplevel. See e.g. this example sessions: http://paste.openstack.org/show/801353/
<cfbolz> isidentical: sweet!
<cfbolz> isidentical: a relatively simple 3.8 feature to get started might be the int.as_as_integer_ratio() method
<isidentical> I was thinking something relevant to the parser/compiler part, maybe f-string debugging expressions?
<cfbolz> yes, good idea too!
<cfbolz> isidentical: did you make a heptapod account yet?
<isidentical> I did
<cfbolz> "isidentical"?
<isidentical> I believe I also requested access, though this is my first time in hg
<isidentical> so I'll just try to see how I can push after I complete some stuff
<cfbolz> ah, didn't get the access request somehow. but I just approved you, welcome to being a pypy dev ;-)
<isidentical> cfbolz: thanks!
lukasa has joined #pypy
<mattip> this compile-C-once-run-anywhere project popped up in my feed
<mattip> putting it here less for the run-anywhere aspect, but more for how to make memcpy trash fewer registers
<mattip> is that a known thing?
<cfbolz> no clue
<gutworth> how useful is that given most constant, small memcpys will be expanded inline by the compiler anyway?
<cfbolz> Oh hi Benjamin! Didn't see you in ages! How are you doing?
<gutworth> hi, cfbolz, happy new year
<antocuni> this looks like a nice pypy success story. Too bad that we don't know/collect/advertise them more :( https://news.ycombinator.com/item?id=25593147
<cfbolz> Nice
<isidentical> Hey hey everyone! First of all, happy new years to all (there is 4 hours left, in my timezone)! What I'd like to ask is, whether it would be good or not to backport the fstring debugging expressions to the 3.7 branch (f'{x=}' syntax)? It is fully backwards compatible (since it is a new feature) and the basic implementation is proposed at MR#789.
<cfbolz> I'd be happy with merging it (and can review the MR tomorrow)
<mattip> maybe we should open a py3.8 branch?
<isidentical> It might be cool, since there are also a couple of features that I can port to pypy (like typed_ast migration, so that black / mypy etc. can run [if they aren't already]
<cfbolz> I'm fine with that too
<bbot2> Started: http://buildbot.pypy.org/builders/own-win-x86-32/builds/2588 [mattip: force build, win32consoleio2]
<bbot2> Started: http://buildbot.pypy.org/builders/own-win-x86-64/builds/129 [mattip: force build, win32consoleio2]
<bbot2> Failure: http://buildbot.pypy.org/builders/own-win-x86-32/builds/2588 [mattip: force build, win32consoleio2]
<bbot2> Failure: http://buildbot.pypy.org/builders/own-win-x86-64/builds/129 [mattip: force build, win32consoleio2]
<mattip> grr. I thought the lock would be shared between the two buildbots on the same machine
oberstet has quit [Quit: Leaving]
isidentical has quit [Quit: isidentical]
<cfbolz> this is fairly interesting too: https://news.ycombinator.com/item?id=25595590
isidentical has joined #pypy
isidentical has quit [Quit: isidentical]
<mattip> I started a py3.8 branch with the stdlib-3.8.7
<mattip> it seems on of the first things to do is positional-only parameters, since they are now used in the stdlib
<mattip> the quick CI "rpython --annotate" fails because opcodes were removed
[Arfrever] has quit [Quit: leaving]