00:41
Rhy0lite has quit [Quit: Leaving]
01:13
jvesely has quit [Quit: jvesely]
01:13
jvesely has joined #pypy
01:17
jvesely has quit [Read error: Connection reset by peer]
02:40
jcea has quit [Quit: jcea]
04:25
jerith has quit [Ping timeout: 240 seconds]
04:25
_habnabit has quit [Ping timeout: 256 seconds]
04:31
_habnabit has joined #pypy
04:32
oberstet has quit [Quit: Leaving]
04:38
_habnabit has quit [Ping timeout: 240 seconds]
04:39
jerith has joined #pypy
04:41
_habnabit has joined #pypy
05:26
_whitelogger has joined #pypy
06:10
<
mattip >
annotation does not check comparisons with rffi.int_REAL, so annotation passed but translation failed :(
06:41
fling has quit [Remote host closed the connection]
06:42
fling has joined #pypy
07:41
Taggnostr has quit [Quit: Switching to single player mode.]
07:57
Taggnostr has joined #pypy
08:15
<
mgorny >
mattip: if my 3.6 MR is fine, i'll do 3.7 today
08:26
<
mattip >
ok. I am only lightly checking this before merging, so the CI will test it
08:27
<
mattip >
for 3.7, let's hold off. I think it makes more sense to update the stdlib in a future release
08:28
<
mattip >
much of this will be picked up when I merge py3.6 into py3.7 anyway
08:28
<
mgorny >
well, i wanted to add 3.7 to gentoo too, and i'd really like to avoid adding vulnerable code
08:29
Taggnostr has quit [Quit: Switching to single player mode.]
08:29
<
mattip >
are there new things specifically for 3.7?
08:29
<
mattip >
that won't get picked up by the merge?
08:58
<
mattip >
mgorny: is there an easy way to backport the last set of changes to 2.7 (default) ?
09:00
<
mgorny >
mattip: i don't understand. are you talking of changes in 3.7?
09:00
<
mgorny >
mattip: all applicable changes are already there
09:01
<
mgorny >
my previous MR took care of it
09:02
<
mgorny >
plus some of the earlier fixes were already there; i guess stdlib was updated more recently than for 3.6
09:02
<
mattip >
cool, thanks
09:06
<
mgorny >
mattip: so, is it fine if i prepare backports for 3.7 too?
09:09
Taggnostr2 has joined #pypy
10:01
dddddd_ is now known as dddddd
10:04
YannickJadoul has joined #pypy
10:05
<
YannickJadoul >
Great job again, everyone :-)
10:05
<
mattip >
YannickJadoul: is there a way to test the python3.7 releases?
10:06
<
YannickJadoul >
mattip: I also added that
10:06
<
YannickJadoul >
Wheels get built and seem to be getting the correct name, etc :-
10:06
<
YannickJadoul >
(i.e. 37 wherever 36 was before)
10:06
<
mattip >
mgorny,: I am merging py3.6 into py3.7, only base_events.py and server.py have some merge conflicts
10:07
<
mattip >
so if you could take a look and see what else needs doing that would be great
10:11
rubdos has quit [*.net *.split]
10:11
omasanori[m] has quit [*.net *.split]
10:11
nopf has quit [*.net *.split]
10:11
whitewolf has quit [*.net *.split]
10:16
omasanori[m] has joined #pypy
10:17
nopf has joined #pypy
10:17
whitewolf has joined #pypy
10:17
rubdos has joined #pypy
10:48
<
mattip >
there is a new failure in lib-python.3.test.test_multiprocessing_fork, but I don't think mgorny's changes touched that
10:52
<
mattip >
the new failures may be due to the recent rpython changes
10:52
<
mattip >
but before I merge default to py3.6 let's try a full buildbot run
11:33
<
arigato >
mattip: are there still some failures coming from my rpython changes? the fix is in default since yesterday but I think default wasn't merged to py3.6 yet
11:44
<
mattip >
arigato: right, I wanted to test default before merging to py3.6,
11:44
<
mattip >
but ended up breaking translation last night
11:44
<
mattip >
so I fixed translation and ran a py3.6 build before realizing I should have tested default first
12:18
<
mgorny >
mattip: lots of new test failures with pypy3.7 :-(
12:20
<
mattip >
ahh. bummer. Sorry. Is it fixable or should I revert the merge?
12:26
<
mattip >
restarted since heptapod was undergoing maintenance so the hg pull failed
12:39
omnipotententity has quit [Read error: Connection reset by peer]
12:40
<
arigato >
mattip: ah sorry, my fault for the first one
12:40
omnipotententity has joined #pypy
12:40
omnipotententity has quit [Changing host]
12:40
omnipotententity has joined #pypy
12:46
marky1991 has joined #pypy
12:46
marky1991 has quit [Remote host closed the connection]
12:47
marky1991 has joined #pypy
12:53
lritter has joined #pypy
12:54
<
arigato >
hah, the test_rsocket.py failures are failing because tmpdir is not '/tmp' but '/build_dir/tmp'
12:54
<
arigato >
that's 10 characters more
12:54
<
arigato >
it pushes the length of the full path beyond the limit of 108 characters
12:57
<
arigato >
(wat? yes, on linux, unix "filename"-like sockets can be created but only with full paths that are up to 108 bytes)
13:04
<
arigato >
mattip: all these linux32 failures should be fixed now
13:25
marky1991 has quit [Ping timeout: 240 seconds]
13:55
marky1991 has joined #pypy
14:12
marky1991 has quit [Ping timeout: 265 seconds]
14:18
<
mattip >
mgorny: I am trying an update to 3.7.9 to gauge the breakage
14:24
andr_ has left #pypy [#pypy]
15:02
oberstet has joined #pypy
15:50
<
mgorny >
mattip: i see a merge mistake in lib-python/3/asyncio/base_events.py ; async def create_datagram_endpoint -> default should be reuse_address=_unset insetad of None
15:51
<
mgorny >
that's all i've found
15:57
YannickJadoul has quit [Quit: Leaving]
16:00
slavfox has quit [Quit: Idle for 30+ days]
16:14
jacob22_ has quit [Read error: Connection reset by peer]
16:18
jacob22_ has joined #pypy
16:32
rubdos has quit [Ping timeout: 240 seconds]
16:32
rubdos has joined #pypy
20:26
jacob22_ has quit [Read error: Connection reset by peer]
20:29
jacob22_ has joined #pypy
20:33
lritter has quit [Ping timeout: 260 seconds]
20:33
lritter has joined #pypy
21:35
jcea has joined #pypy
22:29
_whitelogger has joined #pypy
23:08
epony has quit [Quit: reconf]
23:09
epony has joined #pypy
23:25
lritter has quit [Quit: Leaving]