#pypy on 2020-08-14 — irc logs at freenode.irclog.whitequark.org

2019-08-29 19:33 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

00:41 infernix has joined #pypy

00:55 _whitelogger has joined #pypy

01:06 Rhy0lite has quit [Quit: Leaving]

01:28 <bbot2> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-linux-aarch64/builds/587 [default]

01:42 jcea has quit [Quit: jcea]

01:54 oberstet has quit [Quit: Leaving]

02:53 atomizer has quit [*.net *.split]

02:53 iko_ has quit [*.net *.split]

02:53 lastmikoi has quit [*.net *.split]

02:53 danilonc has quit [*.net *.split]

02:53 ulope has quit [*.net *.split]

02:53 runciter has quit [*.net *.split]

02:53 bogner has quit [*.net *.split]

02:53 epony has quit [*.net *.split]

02:54 epony has joined #pypy

02:55 epony has quit [Max SendQ exceeded]

02:56 epony has joined #pypy

02:57 fling has quit [Ping timeout: 246 seconds]

02:59 iko_ has joined #pypy

02:59 runciter has joined #pypy

02:59 danilonc has joined #pypy

02:59 atomizer has joined #pypy

02:59 lastmikoi has joined #pypy

02:59 ulope has joined #pypy

02:59 bogner has joined #pypy

03:40 epony has quit [Remote host closed the connection]

03:41 epony has joined #pypy

04:53 _whitelogger has joined #pypy

04:53 forgottenone has joined #pypy

05:00 <bbot2> Started: http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/3006 [default]

05:17 _whitelogger has joined #pypy

05:37 otisolsen70 has joined #pypy

05:42 <LarstiQ> mattip: do they fail on the buildbot and falsely claim to pass, or is there something else on the buildbot making them pass unlike a local env?

05:50 kbtr has quit [Ping timeout: 240 seconds]

05:51 Cheery has quit [Ping timeout: 240 seconds]

05:51 kbtr has joined #pypy

05:52 Cheery has joined #pypy

06:40 <bbot2> Failure: http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/3006 [default]

08:19 tsaka__ has joined #pypy

08:31 oberstet has joined #pypy

08:31 tsaka__ has quit [Ping timeout: 260 seconds]

08:35 * arigo fixes typo in sysconfig_pypy.py

08:40 <arigo> mattip, LarstiQ: unsure if that answers the question, but http://buildbot.pypy.org/summary?builder=pypy-c-jit-win-x86-32&category=win32&branch=py3.7 shows many failures too, including 12 in test_datetime

08:42 <arigo> ah sorry, you meant you're also getting many errors on linux64

08:42 <mattip> yup, but only when I rerun the test locally

08:53 <mattip> the implementation of utf8_encode_locale_surrogateescape is wrong

08:57 <mattip> or rather there is a problem in module/_codecs/locale_codec.c pypy_wchar2char?

09:03 dddddd_ is now known as dddddd

09:19 tsaka__ has joined #pypy

09:52 forgottenone has quit [Ping timeout: 264 seconds]

10:04 rubdos_ is now known as rubdos

10:36 <mattip> ahh, maybe the build machine is using a different locale than me, so linux64 utf8_encode_locale_surrogateescape passes on the buildbot and fails locally

11:08 travis-ci has joined #pypy

11:08 <travis-ci> arigo/ffi#97 (master - bc74f2c : Armin Rigo): The build is still failing.

11:08 <travis-ci> Build details : https://travis-ci.org/arigo/ffi/builds/717880253

11:08 <travis-ci> Change view : https://github.com/arigo/ffi/compare/e5017e49e964...bc74f2c265f8

11:08 travis-ci has left #pypy [#pypy]

11:19 <arigo> sorry for the above notification

11:20 <arigo> it's coming from an 8-years-old github project where I just pushed an update of the cffi web site in the README.md

11:20 <mattip> cool

11:40 tsaka__ has quit [Ping timeout: 240 seconds]

11:41 tsaka__ has joined #pypy

11:59 tsaka__ has quit [Ping timeout: 246 seconds]

12:13 Rhy0lite has joined #pypy

12:28 <mattip> hmm. running this from the command line raises

12:28 <mattip> python3 -c "import time; print(time.strftime('%y\ud800'))"

12:28 <mattip> but running inside the repl

12:29 <mattip> >>> import time

12:29 <mattip> >>> time.strftime('%y\ud800')

12:29 <mattip> prints `20\ud800'

12:31 <mattip> for me, on pypy3.7 HEAD both raise an rpython UnicoeEncodeError

12:32 <mattip> (on linux64)

12:33 <mattip> pypy3.6-v7.3.1 raises the same rpython UnicodeEncodeError

12:48 <mattip> I see what the solution is, but I would like to understand why cpython3 works inside the repl

13:06 tsaka__ has joined #pypy

13:35 dmalcolm__ has quit [Quit: Leaving]

13:36 dmalcolm has joined #pypy

13:53 _whitelogger has joined #pypy

13:55 <mattip> grr. calling C functions like strftime in tests untranslated on windows gets the function from the runtime of the python doing the hosting

13:56 <mattip> which is the wrong runtime for our FdValidator context manager that is supposed to setup a enter_suppress_iph handler

13:57 <mattip> so calling the function segfaults

13:57 <mattip> instead of returning -1

13:57 <mattip> s/segfaults/exits/

14:04 todda7 has joined #pypy

14:04 tsaka__ has quit [Ping timeout: 272 seconds]

14:13 marky1991 has joined #pypy

14:14 marky1991 has quit [Remote host closed the connection]

14:15 marky1991 has joined #pypy

14:17 <mattip> maybe I can tweak ll2ctypes get_libc_name with the platform version?

15:06 marky1991 has quit [Ping timeout: 256 seconds]

15:28 <LarstiQ> mattip: fwiw, my `Python 3.8.5 (default, Jul 20 2020, 00:00:00)` behaves the same as your cpython