00:03
Nizumzen has joined #pypy
00:29
Hotpot33 has quit [Ping timeout: 276 seconds]
01:00
kipras`away is now known as kipras
01:06
exarkun has quit [Ping timeout: 268 seconds]
01:06
exarkun has joined #pypy
01:18
kipras is now known as kipras`away
01:37
marr has quit [Ping timeout: 240 seconds]
01:55
tbodt has joined #pypy
02:04
astronavt has quit [Quit: Leaving...]
02:07
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
02:08
tbodt has joined #pypy
02:11
astronavt has joined #pypy
02:13
Nizumzen has joined #pypy
02:19
antocuni has quit [Ping timeout: 248 seconds]
02:36
ssbr` has joined #pypy
02:46
exarkun has quit [Ping timeout: 240 seconds]
02:46
exarkun has joined #pypy
02:55
jcea has quit [Quit: jcea]
02:58
adamholm_ has joined #pypy
03:00
adamholmberg has quit [Ping timeout: 240 seconds]
03:27
ArneBab_ has joined #pypy
03:31
adamholm_ has quit [Remote host closed the connection]
03:32
ArneBab has quit [Ping timeout: 248 seconds]
03:47
<
kenaan >
mjacob default e9597229c18d /pypy/doc/: Fix Sphinx build warnings.
04:03
Hotpot33 has joined #pypy
04:06
inhahe_ has quit [Ping timeout: 240 seconds]
04:06
inhahe_ has joined #pypy
04:07
tbodt has quit [Ping timeout: 260 seconds]
04:08
tbodt has joined #pypy
04:32
adamholmberg has joined #pypy
04:36
<
jneen >
what does "FrozenDesc has no attribute pycall" mean
04:36
adamholmberg has quit [Ping timeout: 240 seconds]
04:42
<
ronan >
jneen: it means that you're trying to call something that isn't callable according to RPython
04:43
<
jneen >
could it be dict.iteritems()
04:43
<
jneen >
i was a little sketchy on that one
04:44
<
ronan >
unimplemented builtin methods can give you that error, IIRC
04:44
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
04:48
<
jneen >
it's dict([...])
04:48
<
kenaan >
rlamy unicode-utf8 3e5aa507f585 /pypy/: Fix errorhandler use in utf8_encode_charmap()
04:49
dddddd has quit [Remote host closed the connection]
04:56
<
jneen >
alsoooo @staticmethod seems kind of... broken
04:56
<
jneen >
you have to call it with self.method_name(...)
04:56
<
jneen >
and the build thinks that means it's being sent an extra arg
04:57
<
jneen >
errrr, hang on i might be wrong
04:58
<
jneen >
yeah nvm pebkac nothing to see here
05:00
jimbaker has quit [Ping timeout: 248 seconds]
05:17
<
kenaan >
rlamy unicode-utf8 a4d68881a89d /: Handle errorhandlers that go backward
05:25
lritter has joined #pypy
05:25
<
kenaan >
rlamy unicode-utf8 0accab6d493c /pypy/interpreter/unicodehelper.py: fix
05:32
lritter has quit [Remote host closed the connection]
05:33
adamholmberg has joined #pypy
05:36
<
kenaan >
rlamy unicode-utf8-test 1665df77270e /: hg merge unicode-utf8
05:38
adamholmberg has quit [Ping timeout: 268 seconds]
06:19
<
arigato >
ronan: fwiw, the merge of unicode-utf8-test with unicode-utf8-re passes the module/_sre tests
06:34
adamholmberg has joined #pypy
06:39
adamholmberg has quit [Ping timeout: 260 seconds]
07:21
<
fijal >
FYI I removed all the parts that need newunicode
07:21
<
fijal >
so -test can be closed
07:22
<
fijal >
arigato: so, we can probably run the benchmark run
07:22
<
fijal >
(unclear hwo would we compare)
07:23
* fijal
fixes micronumpy
07:23
<
arigato >
if the results are "it's kind of similar in speed" then success, and the next step is to make reasonable jit traces
07:23
<
fijal >
that's the output I got so far
07:23
<
fijal >
everything I run was within 5% or so, except django that was a lot faster and jsonbench that was ~8% slower or so
07:23
<
fijal >
probably because of dicts
07:24
<
fijal >
maybe it makes sense to have a dict made out of unicode and recompute the lnegth
07:24
<
arigato >
can I merge unicode-utf8-test inside unicode-utf8-re? it seems to fix translation
07:24
<
arigato >
(but then I get translation errors from module/_sre, which is the point)
07:25
<
fijal >
how about killing -test and merging utf8-re into unicode-utf8?
07:25
<
fijal >
and we can fix the translation
07:26
<
fijal >
there is one more question about micronumpy, but I think I'll get a coffee first
07:29
<
fijal >
arigato: tornado benchmark spends most of it's time formatting tracebacks for closed connections or something like that
07:29
<
fijal >
I wonder how much I care about that
07:30
<
kenaan >
arigo unicode-utf8-re 78d73593a136 /pypy/module/_sre/interp_sre.py: Add sanity-checks
07:30
<
kenaan >
arigo unicode-utf8-re 80cb52b67f60 /pypy/module/_sre/interp_sre.py: in-progress: translation fixes, but I don't understand why I need that
07:30
<
kenaan >
arigo unicode-utf8-re 58b6fedc39bc /: hg merge unicode-utf8
07:31
<
LarstiQ >
fijal: on this branch only, or also normally?
07:31
<
fijal >
also normally
07:31
<
LarstiQ >
seems a weird place to spend majority of time
07:31
<
fijal >
but the amount varies wildly
07:32
<
fijal >
LarstiQ: well it's not TOO surprising for pypy to be slow at formatting tracebacks
07:32
<
fijal >
but it's surprising a basic http server hit by ab or siege does that a lot
07:32
<
fijal >
why are there futures to do something on a stream that's closed by then?
07:34
<
arigato >
there are r_uint issues
07:34
<
arigato >
it's unclear to me why prev_codepoint_pos() must take a r_uint, and also why half the calls don't care
07:35
adamholmberg has joined #pypy
07:36
<
fijal >
arigato: I don't know
07:36
<
arigato >
...ok, it doesn't take r_uint
07:36
<
fijal >
if anything I cargo culted r_uint
07:36
<
arigato >
none of the calls pass r_uint
07:36
* arigato
kills the comment
07:39
<
kenaan >
arigo unicode-utf8-re 698fcd50ff42 /pypy/module/_sre/interp_sre.py: translation fix
07:39
<
kenaan >
arigo unicode-utf8-re 0d4b7b9d28b5 /rpython/rlib/: Kill the wrong comment about r_uint in prev_codepoint_pos
07:40
adamholmberg has quit [Ping timeout: 255 seconds]
07:45
<
kenaan >
arigo unicode-utf8-re ec4522b42070 /rpython/rlib/rsre/rsre_core.py: More translation fixes
07:50
<
kenaan >
arigo unicode-utf8-re 363f54e90b19 /rpython/rlib/rsre/: More
07:54
<
arigato >
fijal: ok, now it seems to translate. feel free to merge!
08:10
inad922 has joined #pypy
08:13
inad922 has quit [Remote host closed the connection]
08:55
dddddd has joined #pypy
09:02
kenaan has quit [Ping timeout: 248 seconds]
09:33
adamholmberg has joined #pypy
09:35
antocuni has joined #pypy
09:37
marr has joined #pypy
10:27
mvantellingen has quit [Ping timeout: 240 seconds]
10:32
sn6uv has quit [Ping timeout: 240 seconds]
10:34
mvantellingen has joined #pypy
10:37
astronavt has quit [Ping timeout: 255 seconds]
10:41
drolando has quit [Remote host closed the connection]
10:42
drolando has joined #pypy
10:42
mvantellingen has joined #pypy
11:08
antocuni has quit [Ping timeout: 255 seconds]
11:33
Nizumzen has joined #pypy
12:55
jcea has joined #pypy
13:01
kipras`away is now known as kipras
13:13
Taggnostr has quit [Remote host closed the connection]
13:18
Taggnostr has joined #pypy
13:34
Guest53285 has quit [Ping timeout: 260 seconds]
13:49
marvin has joined #pypy
13:50
marvin is now known as Guest65612
14:26
exarkun has quit [Ping timeout: 268 seconds]
14:27
exarkun has joined #pypy
14:43
dddddd has quit [Ping timeout: 248 seconds]
14:56
dddddd has joined #pypy
15:46
Rotonen has quit [Quit: reboot]
15:48
Rotonen has joined #pypy
16:06
exarkun has quit [Ping timeout: 260 seconds]
16:07
exarkun has joined #pypy
16:09
Nizumzen has quit [Ping timeout: 255 seconds]
16:17
gclawes has quit [Remote host closed the connection]
16:51
amaury has joined #pypy
17:16
<
fijal >
arigato: I think I'm taking a full day off :)
17:31
amaury has quit [Ping timeout: 260 seconds]
17:34
citizen_stig has joined #pypy
17:36
zmt00 has quit [Ping timeout: 240 seconds]
17:36
Nizumzen has joined #pypy
17:39
citizen_stig has quit [Remote host closed the connection]
17:40
citizen_stig has joined #pypy
17:42
citizen_stig has quit [Remote host closed the connection]
17:43
zmt00 has joined #pypy
17:44
citizen_stig has joined #pypy
17:47
exarkun has quit [Ping timeout: 250 seconds]
17:47
exarkun has joined #pypy
17:48
citizen_stig has quit [Ping timeout: 248 seconds]
18:07
gclawes has joined #pypy
18:54
sn6uv has joined #pypy
19:00
citizen_stig has joined #pypy
19:03
sn6uv has quit [Ping timeout: 246 seconds]
19:04
forgottenone has joined #pypy
19:10
jamesaxl has quit [Quit: WeeChat 1.9.1]
19:21
astronavt has joined #pypy
19:27
exarkun has quit [Ping timeout: 255 seconds]
19:27
exarkun has joined #pypy
19:40
forgottenone has quit [Remote host closed the connection]
19:45
amaury has joined #pypy
20:05
kenaan has joined #pypy
20:05
<
kenaan >
mattip buildbot afea1ba1dd41 /bot2/pypybuildbot/master.py: need to think more about running benchmarks with onlyIfChanged
20:05
bbot2 has quit [Quit: buildmaster reconfigured: bot disconnecting]
20:06
bbot2 has joined #pypy
20:22
astronavt has quit [Remote host closed the connection]
20:23
astronavt has joined #pypy
20:28
astronav_ has joined #pypy
20:28
astronavt has quit [Ping timeout: 276 seconds]
20:35
astronav_ has quit [Remote host closed the connection]
20:41
<
LarstiQ >
unfortunately without saying which version was tried
20:41
* LarstiQ
confirms it's in 5.8
20:42
<
cfbolz >
LarstiQ: weeeeell, also it cleanly raises an rpython exception, which is not a segfault, thus not easily exploitable
20:42
<
LarstiQ >
cfbolz: cmp(float('nan'), float('nan'))?
20:42
<
cfbolz >
LarstiQ: no, sorry, I mean the other example
20:46
<
cfbolz >
LarstiQ: I am not sure cmp(nan, nan) can be wrong. any number you return makes no sense
20:46
astronavt has joined #pypy
20:47
<
LarstiQ >
cfbolz: return a non int? ;)
20:47
<
cfbolz >
return nan, I suppose :-P
20:48
<
LarstiQ >
numpy returns False for all three
20:49
<
LarstiQ >
i.e, np.nan <, ==, > np.nan
20:49
<
LarstiQ >
but cmp's interface doesn't allow that
20:49
astronavt has quit [Remote host closed the connection]
20:50
<
LarstiQ >
and with that, time to retire for the night
20:51
<
cfbolz >
LarstiQ: I suppose I should fix the multiprocessing thing
20:52
astronavt has joined #pypy
20:56
astronavt has quit [Remote host closed the connection]
21:05
exarkun has quit [Ping timeout: 276 seconds]
21:07
exarkun has joined #pypy
21:09
astronavt has joined #pypy
21:13
citizen_stig has quit [Remote host closed the connection]
21:27
adamholmberg has quit [Remote host closed the connection]
21:35
<
kenaan >
amauryfa py3.6 cadf41717a3f /pypy/interpreter/astcompiler/: The ast module now has a "Constant" node that AST optimizers can use. Use it in place of PyPy's specific "Const".
21:35
<
kenaan >
amauryfa py3.6 2d7ef69f48bd /pypy/module/posix/: Scandir: Add a context manager, and a ResourceWarning when it is not explictly closed.
21:48
tbodt has joined #pypy
21:52
jacob22__ has quit [Ping timeout: 260 seconds]
22:00
jacob22__ has joined #pypy
22:28
adamholmberg has joined #pypy
22:32
adamholmberg has quit [Ping timeout: 240 seconds]
22:46
exarkun has quit [Ping timeout: 240 seconds]
22:47
exarkun has joined #pypy
22:51
oberstet has quit [Ping timeout: 240 seconds]
23:08
inhahe_ has quit [Ping timeout: 240 seconds]
23:09
oberstet has joined #pypy
23:09
inhahe_ has joined #pypy
23:15
oberstet has quit [Ping timeout: 248 seconds]
23:29
adamholmberg has joined #pypy
23:34
adamholmberg has quit [Ping timeout: 248 seconds]
23:42
antocuni has joined #pypy