#pypy IRC log for Monday, 2012-01-02

Sweet-P (~Sweet-P@203.205.116.93) joined #pypy.00:19
nedbat (~nedbat@python/psf/nedbat) joined #pypy.00:31
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/76300:32
pybot joined #pypy.01:10
-ChanServ (ChanServ@services.)- [#pypy] Welcome!01:10
-ChanServ (ChanServ@services.)- [#stackless] Welcome to #stackless. This channel is for the discussion of matters relating to Stackless Python. Remember that others in this channel may not be in the same timezone as you, so stick around for an answer (even if only in your channel history) if you ask a question. The mailing list is also recommended as being a faster place to get help (http://www.stackless.com/mailman/listinfo/stackless).01:11
pybot joined #pypy.01:19
-ChanServ (ChanServ@services.)- [#pypy] Welcome!01:19
-ChanServ (ChanServ@services.)- [#stackless] Welcome to #stackless. This channel is for the discussion of matters relating to Stackless Python. Remember that others in this channel may not be in the same timezone as you, so stick around for an answer (even if only in your channel history) if you ask a question. The mailing list is also recommended as being a faster place to get help (http://www.stackless.com/mailman/listinfo/stackless).01:20
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/1901:22
mat^2 (~mathias@87.55.0.198) left irc: Ping timeout: 255 seconds01:28
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/19101:29
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/190802:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/66002:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/31402:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/32502:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/66302:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/117902:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/84602:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/149802:00
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/77302:00
verte-wleslie (~verte@python/site-packages/verte) left irc: Quit: ~~~ Crash in JIT!02:06
dracman (~draco@d151010.adsl.hansenet.de) left irc: Ping timeout: 252 seconds02:15
hakanardo (~hakan@h59ec009e.seluldx.dyn.perspektivbredband.net) left irc: Ping timeout: 252 seconds02:15
verte-wleslie (~verte@python/site-packages/verte) joined #pypy.02:17
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds02:20
dracman (~draco@g231146014.adsl.alicedsl.de) joined #pypy.02:20
mikefctimotimo: thanks.  I'm just trying to make more sense of the code. I want to add some features and didn't know which level would be more appropriate.02:38
voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace02:53
mikefc (~mikefc@60-242-240-196.static.tpgi.com.au) left irc: Quit: mikefc03:00
getxsick (xsx@unaffiliated/getxsick) left irc: Read error: Operation timed out03:01
getxsick (xsx@unaffiliated/getxsick) joined #pypy.03:01
Kaskuka (~John@osbk-4d08b9ee.pool.mediaWays.net) left irc: Ping timeout: 240 seconds03:03
Kaskuka (~John@osbk-4db15f17.pool.mediaWays.net) joined #pypy.03:04
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/100003:09
espes (espes@2001:470:1f05:ca1::) left irc: Ping timeout: 240 seconds04:03
espes (espes@2001:470:1f05:ca1::) joined #pypy.04:06
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/190804:11
jacob22 (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Ping timeout: 240 seconds04:16
jacob22 (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.04:17
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 245 seconds04:48
mikefc (~mikefc@60-242-240-196.static.tpgi.com.au) joined #pypy.04:48
nettok (~quassel@200.119.182.101) joined #pypy.04:49
nettok (~quassel@200.119.182.101) left irc: Ping timeout: 252 seconds04:54
nettok (~quassel@200.119.156.196) joined #pypy.05:09
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/149805:17
anish (anish@nat/redhat/x-uwpqcgcdquhzbdik) joined #pypy.05:18
Nick change: anish -> Guest4776505:18
Nick change: Guest47765 -> __anish__05:21
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/84605:23
gutworth (~benjamin@64-131-9-52.usfamily.net) left irc: Ping timeout: 240 seconds05:23
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/66305:32
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/117905:33
gutworth (~benjamin@64-131-9-52.usfamily.net) joined #pypy.05:34
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/66005:37
int3__ (~int3__@bb220-255-192-11.singnet.com.sg) joined #pypy.06:07
mattip (c1a94669@gateway/web/freenode/ip.193.169.70.105) joined #pypy.06:09
mattipAlex_Gaynor: ping06:09
mikefcmattip: what's the consensus on copying docstrings/code from numpy source?06:12
Action: mikefc is looking at .cross()06:12
mattipIANAL, but I have been doing it. Maybe ask on a numpy channel/mailing list06:13
mikefcwe could bring in big swathes at the app level if that was the case.06:14
mattipmake sure it all has tests06:16
mikefcbut doing more at the app level may not be what we're aiming for?06:16
mikefcI'm doing a test .cross() at app level with tests that i'll put up on bugs.pypy for discussion.06:16
mattipdepends on your goals.06:17
Action: mikefc current goal is just to learn some pypy 06:17
mikefcwhat goals should I have after that?06:18
mattipSo you would gain more by avoiding numpy altogether: try using the jitviewer or something06:18
mattipto see how jit and pypy interact06:19
mattipIf your goal is to enable a wider community to use numpypy (like me)06:19
Action: mikefc 's current goal is to learn some pypy - so that I can get numpypy to do what i want :)06:19
mattipthen you want scipy, matplotlib, wxPython to run on pypy06:20
mikefci.e. be compat with numpy06:20
mattipHow much of numpy do you need and what are your performance requirements06:21
mikefcmattip: I'm just a tinkerer. I don't *need* it for anything. Just doing it to scratch an itch.06:22
gutworth (~benjamin@64-131-9-52.usfamily.net) left irc: Ping timeout: 244 seconds06:24
mikefci understand that just adding an applevel cross() from numpy isn't probably where people want to go with numpypy, but it's my 2nd day coding in pypy, so i'm just doing what I can to learn how it's held together.06:26
mattipgo for it. If it causes performance problems that bother someone, then they should fix it. Simple is better.06:27
Action: mattip back to grindstone06:28
mattip (c1a94669@gateway/web/freenode/ip.193.169.70.105) left irc: Quit: Page closed06:28
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/77306:37
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) joined #pypy.07:21
zain (~textual@cpe-68-173-70-124.nyc.res.rr.com) joined #pypy.07:22
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/32507:23
GoronHow portable is library-neutral code from traditional Python to PyPy?07:25
mikefc"depends"07:25
verte-wleslieGoron: pypy implements not only what is considered standard python, but also many cpython implementation details.07:25
verte-wlesliesuch as internal apis like sys._getframe().07:26
JaRoel (~jaroel|4d@sink.jaroel.nl) left irc: Remote host closed the connection07:29
GoronIf I an American PyPy will it still plug into european outlets?07:30
GoronIfI have an*07:30
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) joined #pypy.07:39
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/31407:39
CIA-5303mikefc 07roundup * 10#977/numpypy ndim attribute (patch attached): 07:40
CIA-53[new] Currently numpypy lacks an ndim attribute, this patch adds it.07:40
CIA-53However, I couldn't get numpypy to throw the same error if you try doing a ... * 14https://bugs.pypy.org/issue97707:40
dracman (~draco@g231146014.adsl.alicedsl.de) left irc: Read error: Connection reset by peer07:55
ssbr (~ssbr@python/site-packages/ssbr) joined #pypy.07:59
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.08:16
Varriount (~Varriount@hnng.public-craft.com) left irc: Ping timeout: 240 seconds08:20
JaRoel (~jaroel|4d@office.fourdigits.nl) joined #pypy.08:23
fermianyon (~lane@c-71-229-21-197.hsd1.al.comcast.net) left irc: Ping timeout: 240 seconds08:24
mikefcnumpypy question for whenever anyone wakes up:  If I want to add a method to numpypy's ndarray class (e.g. a.cumsum()), can I only do it at the interp level?   I can't see how I'd do it at the app level.  All the applevel functions aren't instance methods e.g. numpy.max(), numpy.min()...08:25
bivab (~david@fwstups.cs.uni-duesseldorf.de) joined #pypy.08:25
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Ping timeout: 252 seconds08:26
Action: mikefc hasn't quite worded his question correctly. numpy.arange()/zeros()/identy() are better examples of the type of function available at applevel. 08:45
mikefcbut I can't see how/where i'd put cumsum at applevel, since it's an instance method08:49
Action: mikefc probably isn't using OO terminology correctly.08:50
kushal (~kdas@114.143.164.46) joined #pypy.08:53
kushal (~kdas@114.143.164.46) left irc: Changing host08:53
kushal (~kdas@fedora/kushal) joined #pypy.08:53
kushal (~kdas@fedora/kushal) left irc: Remote host closed the connection08:54
mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) left irc: Remote host closed the connection09:04
fijal (~fijal@197.105.7.74) joined #pypy.09:04
mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) joined #pypy.09:06
ltratt (~ltratt@87.114.17.162) joined #pypy.09:13
whyking (~quassel@p4FFB759A.dip.t-dialin.net) joined #pypy.09:14
fijalltratt: hi09:14
ltrattfijal: good morning09:15
ltrattthanks for your help yesterday09:16
Varriount (~Varriount@hnng.public-craft.com) joined #pypy.09:16
ltratti'm trying to work out why my traces are so huge09:16
ltrattand getting somewhere slowly but surely i think09:17
fijaldid you get jitviewer to run09:17
fijal?09:17
fijalit won't help you, but it'll help you see traces that are not huge09:17
fijallet me have a look09:17
ltrattat the moment i'm using PYPYLOG=jit-log-opt:<file>09:17
fijalyeah09:18
Alex_Gaynorltratt: if you use pygments it has a highlighter for .pypylog files09:18
Alex_Gaynoronly in the hg repo though, not a release09:19
ltrattok, i'll have a look09:19
fijalbut there is a program that will give you a better view than trying to poke in the file09:20
fijalgoogle jitviewer09:20
fijalmeh09:20
fijalmy internet sucks :/09:20
fijalAlex_Gaynor: jitviewer is still *much* better09:20
ltrattis jitviewer pypy only?09:20
Alex_Gaynorfijal: yes, in a little bit I'll be able to show you a preview of jitviewer 2.0 :)09:20
fijalltratt: as in python-interpreter only09:20
fijalno09:20
ltrattaha, ok09:21
fijalit won't show you converge code mixed with IR (as it does for python)09:21
ltratti can't remember if i could get it working or not on openbsd. if i can't, i will badger carl friedrich later today, as we're off to the same event in germany09:22
ltratthere's something interesting for you in the meantime.09:22
ltrattconsider this program: http://pastebin.com/hHi1H2HK09:22
ltrattit ends up with two traces http://pastebin.com/u2nP3iQ1, one before the yield and one after the yield09:23
ltrattso, at least as i see it, any RPython program with a yield in it will double its trace lengths. that sort-of seems reasonable in one way - i'm not sure how to get around it09:23
Alex_Gaynorperhaps you need should_unroll_one_iteration09:24
ltratti'm not familiar with that09:24
ltratthmm, interp_jit.py has it09:25
Alex_Gaynorask fijal about it, he added it :)09:25
Alex_Gaynoryes, it's used for generators in python, I'm not 100% sure how it works though09:25
ltratti wonder how it interacts with RPython generators?09:25
fijalltratt: generators are tricky09:25
ltrattagreed09:26
fijalI have no idea how RPython generators work honestly :)09:26
ltrattnor i :)09:26
Alex_GaynorI think (assume?) they're just a state machine09:26
ltratti think i might send a message to the list and see what armin thinks, as he created rpython generators09:27
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) left irc: Ping timeout: 260 seconds09:28
fijalltratt: but in general unroll_one_iteration means you would unroll one iteration of a loop :)09:29
fijalin this case the internal generator loop09:29
ltrattyes, that makes sense09:29
ltratti guess rpython might want to do that for its own generators somehow... but i don't know how!09:29
antocuni (~antocuni@host117-71-dynamic.58-82-r.retail.telecomitalia.it) joined #pypy.09:32
fijal_ (~fijal@41.53.95.237) joined #pypy.09:32
fijal (~fijal@197.105.7.74) left irc: Read error: Connection reset by peer09:32
Nick change: fijal_ -> fijal09:36
G2P (~G2P@fw-asn1.ornis.com) joined #pypy.09:37
ltratthmm, no, i'm being stupid. that yield at the converge level doesn't correspond to an RPython yield...09:38
ltrattfijal: what does should_unroll_one_iteration ask the JITDriver to do?09:39
fijalltratt: it receives the green key and should reply with "is this green key a generator"09:41
fijalltratt: I can't look anywhere, my internet is super fucked09:41
ltrattyes, i can see it checking for a generator. but that then tells the JIT to do something clever I assume?09:42
fijalit's not very clever09:42
fijalit unrolls one loop iteration09:42
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.09:42
fijalthat would achieve the effect you want for the very simple generators that have yield once-per-iteration09:42
ltrattok, i think i see09:43
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.09:43
ltratti will try that later09:43
ltrattat the moment i don't have a way to check if a function is a generator, so i'll have to hack it :)09:44
fijalltratt: I'm trying to check out converge in like 15 minutes :/09:44
fijalI might give up09:44
ltrattah :(09:44
ltrattdodgy internet connections are a pain, i know what it's like!09:45
fijaleh09:47
fijalsomeone knows how to change xchat from black to default?09:48
fijalI pressed some buttons...09:48
fijalanyway, I give up working it's impossible09:48
fijalI can't google anything09:48
fijal (~fijal@41.53.95.237) left irc: Read error: Connection reset by peer09:48
fijal (~fijal@41.53.95.237) joined #pypy.09:48
ltratt (~ltratt@87.114.17.162) left irc: Quit: Off to Germany09:54
Sweet-P (~Sweet-P@203.205.116.93) left irc: Remote host closed the connection09:57
danchrI haven't10:00
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) left irc: Quit: asmeurer10:01
whyking (~quassel@p4FFB759A.dip.t-dialin.net) left irc: Ping timeout: 252 seconds10:06
antocuniaccording to the past irc logs, 2008 was the year of pypy on the desktop, 2009 the year of the jit on the desktop, 2010 the "pypy's year"10:08
antocuniand nothing for 201110:08
antocuniso10:08
antocuni2012 will be the year of...?10:08
h0h0 (~k@141.79-161-28.customer.lyse.net) joined #pypy.10:08
antocuniworld domination? :-)10:08
Alex_Gaynorantocuni: jitviewer on the web!10:09
danchrOops, wrong channel&10:09
antocuniwould be cool :-)10:09
antocuniAlex_Gaynor: are you working on it?10:09
Alex_Gaynorantocuni: preview: http://tracebin.ep.io/trace/1000/compiled/1/10:09
Alex_Gaynorstill plenty of work to do10:10
antocuniAlex_Gaynor: looks nice10:10
antocunitimeline doesn't work, but I bet you know it10:11
Alex_Gaynoryes :)10:11
nettok (~quassel@200.119.156.196) left irc: Remote host closed the connection10:11
Alex_Gaynorlots's of stuff doesn't work,a nd it's all fake datga ATM10:11
antocuniAlex_Gaynor: although I like more the current graphic for showing the source code + the traces10:12
Alex_Gaynorit needs quite a bit of improvement, seperating out the bytecode markers, add the JS back10:13
antocuniin particular, the fact that the ops are indented wrt the source code10:13
kkris (~kris@93-82-33-167.adsl.highway.telekom.at) joined #pypy.10:17
fijalAlex_Gaynor: the style sucks btw10:17
Alex_Gaynor?10:18
fijalthe CSS10:21
fijalalso, youshould not scroll small boxes10:21
fijalthey should wrap or something10:22
Alex_Gaynorthey're super unreadable if you linewrap10:22
zain (~textual@cpe-68-173-70-124.nyc.res.rr.com) left irc: Quit: Computer has gone to sleep.10:22
antocunifijal: not sure. If you have e.g. a guard with a looong list of parameters, you don't want it to wrap10:22
fijalwell10:22
antocunior maybe you want to be able to switch the behavior dynamically10:22
Alex_Gaynorfor now my next things are making merge points display better, and adding the JS for highlighting back10:23
zain (~textual@cpe-68-173-70-124.nyc.res.rr.com) joined #pypy.10:23
h0h0 (~k@141.79-161-28.customer.lyse.net) left irc: Quit: leaving10:26
Action: Alex_Gaynor -> bed10:27
ssbrThere isn't quite anything like the fun of writing an interpreter. :)10:27
fijalthe last thing you want is a small box with a scroll bar and tons of free space10:27
antocunifijal: re 7abc703337f7: "bah"10:31
fijalantocuni: yes :/10:32
mimi (~mimi@97-88-28-179.dhcp.roch.mn.charter.com) joined #pypy.10:41
fijalantocuni: good question "why?"10:47
antocuniI suppose the answer is "just because"10:47
antocuniwho's the author?10:48
fijalauthor of what?10:52
fijalmy question was "why does locals() force the frame"10:53
fijalthey don't seem to have to10:53
kenaan12hager ppc-jit-backend 110b164b7fc20c 15/pypy/jit/backend/ppc/ppcgen/ppc_assembler.py: replace tabs with spaces10:53
antocuniaah10:54
antocuniI thought the question was "why does this code use locals()"10:54
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) joined #pypy.10:57
fijalthe code use locals to shorten the loop10:58
fijalit makes sense10:58
fijalshorten the __init__10:58
kenaan12hager ppc-jit-backend 115d2419018c35 15/pypy/jit/backend/ppc/ppcgen/ppc_assembler.py: remove some more tabs10:58
fijalbecause you have to access stuff through __dict__10:58
fijalantocuni: the bad part - there is more :/10:59
fijalantocuni: are you going to finish ffistruct some time?11:00
antocunifijal: "yes"11:00
antocuniwell, I'm almost done with my paid jobs11:00
antocuniso I should really have time for ffistruct soon11:00
pybot joined #pypy.11:10
-:#pypy- [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp11:10
-ChanServ (ChanServ@services.)- [#pypy] Welcome!11:10
-ChanServ (ChanServ@services.)- [#stackless] Welcome to #stackless. This channel is for the discussion of matters relating to Stackless Python. Remember that others in this channel may not be in the same timezone as you, so stick around for an answer (even if only in your channel history) if you ask a question. The mailing list is also recommended as being a faster place to get help (http://www.stackless.com/mailman/listinfo/stackless).11:11
zain (~textual@cpe-68-173-70-124.nyc.res.rr.com) left irc: Quit: Computer has gone to sleep.11:13
CIA-5303mikefc 07roundup * 10#976/Implement numpypy.cross(): 11:17
CIA-53[chatting] I've had a go at implementing this with applevel code. Patch attached.11:17
CIA-53Note that the whole things is really a cut and paste from act ... * 14https://bugs.pypy.org/issue97611:17
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 249 seconds11:30
CIA-5303mikefc 07roundup * 10#976/Implement numpypy.cross(): Ooops. "We're also missing an ndim attribute." * 14https://bugs.pypy.org/issue97611:32
__anish__ (anish@nat/redhat/x-uwpqcgcdquhzbdik) left irc: Read error: Connection reset by peer11:43
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.11:48
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds11:52
lizardo (~lizardo@189.2.128.130) joined #pypy.11:59
mat^2 (~mathias@87.55.0.198) joined #pypy.12:00
kenaan12arigo default 11a58d0c303bce 15/pypy/jit/backend/x86/test/test_runner.py: fix test12:01
hakanardo (~hakan@h59ec009e.seluldx.dyn.perspektivbredband.net) joined #pypy.12:06
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) joined #pypy.12:09
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 244 seconds12:18
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.12:19
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 244 seconds12:30
derdon (~derdon@p5DE8BC89.dip.t-dialin.net) joined #pypy.12:39
kenaan12hager ppc-jit-backend 115f4684c8251c 15/pypy/jit/backend/ppc/ppcgen/: factor out implementation of calls12:43
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.12:43
antocunifijal: I think that "remove_invalidates" is a very bad name12:44
antocunie.g., I cannot figure out what the hell it's doing, even looking at the doc string :-)12:44
verte-wleslieplease consider 'confuse_anto' instead.12:45
kenaan12hager ppc-jit-backend 1125b8f969eadb 15/pypy/jit/backend/ppc/ppcgen/codebuilder.py: remove typo in codebuilder12:51
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 252 seconds12:55
kenaan12hager ppc-jit-backend 11defdbd4220ca 15/pypy/jit/backend/ppc/test/test_runner.py: skip test test_cond_call_gc_wb_array_card_marking_fast_path12:56
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) left irc: Ping timeout: 260 seconds12:59
`fox` (~fox@host71-57-dynamic.48-82-r.retail.telecomitalia.it) joined #pypy.13:00
arigato (~arigo@adsl-84-227-72-198.adslplus.ch) joined #pypy.13:08
nedbat (~nedbat@python/psf/nedbat) joined #pypy.13:13
dracman (~draco@212.255.40.92) joined #pypy.13:37
stakkarshappy new year!13:42
arigatohi christian, happy new year!13:42
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) joined #pypy.13:42
Goron (LC@h171.163.191.173.dynamic.ip.windstream.net) left #pypy.13:42
stakkarsI had pretty much trouble with my new machine which had a hardware bug.13:42
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) joined #pypy.13:42
arigato:-(13:43
stakkarshad to fight the support since friday afternoon since sunday morning.13:43
stakkarsthey finally were convinced to replace the machine ( was becoming angry after they did ridiculous looong tests to keep me quiet)13:44
stakkarsbut after I claimed to be ready to replace the provider, I had a replacement server after 2 hours. :-)13:45
stakkarsthen I did the transfer last night and had a smooth transition13:45
stakkarswith maybe half an hour downtime.13:46
stakkarsrsyncing multiple times, then stopping all containers, a final rsync, then changing DNS,13:47
stakkarsand then the final reboot of the new machine. Yeech ;-)13:47
stakkarsarigato: my windows64 installation is not activated, yet.13:49
stakkarsI will install this in an extra virtual container, to avoid troubles with the host machine.13:50
stakkarsmay take a bit of time...13:50
stakkarss/since sunday/until sunday/13:51
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.13:51
GoronOh! Now it makes sense13:53
GoronI thought you were nesting whens13:53
sunoano (~sa@mail.heartinternet.co.uk) joined #pypy.13:55
sunoano (~sa@mail.heartinternet.co.uk) left irc: Changing host13:55
sunoano (~sa@unaffiliated/sunoano) joined #pypy.13:55
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 252 seconds13:55
voidspace (~anonymous@87-194-212-65.bethere.co.uk) joined #pypy.13:57
voidspace (~anonymous@87-194-212-65.bethere.co.uk) left irc: Changing host13:57
voidspace (~anonymous@python/psf/voidspace) joined #pypy.13:57
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.14:02
whyking (~quassel@p4FFB606C.dip.t-dialin.net) joined #pypy.14:03
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Client Quit14:03
verte-wleslie (~verte@python/site-packages/verte) left irc: Quit: ~~~ Crash in JIT!14:05
arigato (~arigo@adsl-84-227-72-198.adslplus.ch) left irc: Ping timeout: 252 seconds14:08
kenaan12hager ppc-jit-backend 11fa8e4ebd847b 15/pypy/jit/backend/ppc/ppcgen/: (bivab, hager):  fix bug in prepare_guard_call_release_gil14:11
gutworth (~benjamin@64-131-9-52.usfamily.net) joined #pypy.14:17
`fox` (~fox@host71-57-dynamic.48-82-r.retail.telecomitalia.it) left irc: Quit: Lost terminal14:19
`fox` (~fox@host71-57-dynamic.48-82-r.retail.telecomitalia.it) joined #pypy.14:20
voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace14:28
nickik (~nick@217.150.234.222) joined #pypy.14:29
pybot joined #pypy.14:40
-ChanServ (ChanServ@services.)- [#pypy] Welcome!14:40
-ChanServ (ChanServ@services.)- [#stackless] Welcome to #stackless. This channel is for the discussion of matters relating to Stackless Python. Remember that others in this channel may not be in the same timezone as you, so stick around for an answer (even if only in your channel history) if you ask a question. The mailing list is also recommended as being a faster place to get help (http://www.stackless.com/mailman/listinfo/stackless).14:41
fprimex (fprimex@wireless-dhcp200d088.fsu.edu) joined #pypy.14:56
fijalantocuni: this is a hack that will go away14:57
antocuniok14:58
fijalif you don't understand what it does, don't use it :)14:58
antocuniok, I was just complaining about the bad naming14:58
antocunibut if it goes away, it's fine for me14:59
fijalwell14:59
fijalit calls rpython-level function of the same name :)14:59
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.15:02
`fox` (~fox@host71-57-dynamic.48-82-r.retail.telecomitalia.it) left irc: Ping timeout: 248 seconds15:03
bivab (~david@fwstups.cs.uni-duesseldorf.de) left irc: Quit: bivab15:04
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) joined #pypy.15:05
nickik (~nick@217.150.234.222) left irc: Quit: Verlassend15:17
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) left irc: Ping timeout: 252 seconds15:34
moreati (~alex@5ad0d588.bb.sky.com) joined #pypy.15:53
arigato (~arigo@adsl-84-227-72-198.adslplus.ch) joined #pypy.15:59
mattip (~chatzilla@bzq-79-176-117-44.red.bezeqint.net) joined #pypy.16:14
mattipfijal: hi16:30
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 276 seconds16:35
fijalmattip: hi16:40
mattipAre you settled in enough to take a look at the numpypy-axisops branch?16:41
fijalgood question16:41
fijalI have not settled my internet yet :/16:41
fijalmattip: how about I'll give it a go tomorrow?16:41
mattipOK. thnks.16:42
mattipI thought I would take a look at reworking to_str() since I have learned alot since then.16:43
JaRoel (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection16:44
fijalcool, there is a bug somewhere how to_str is not working correctly16:44
mattipI don't like that it is recursive, and it also should use iterators.16:44
fijalI would appreciate if you can find it on the issue tracker16:44
mattipOK, I have been chatting with mikefs about the bug.16:44
fijalah ok16:45
fijalmattip: cool, I'm a bit behind my schedule with *everything*16:47
fijaland have tons and tons of paperwork to deal with16:47
mattipSorry to hear. At least at the end of the paper trail is a pot of gold?16:48
fijalI heavily doubt that16:51
mattip:(16:53
fijalAlex_Gaynor: ping16:56
kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/16:57
G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving.17:01
nedbat (~nedbat@python/psf/nedbat) joined #pypy.17:02
kkris1 (~kris@80-123-47-161.adsl.highway.telekom.at) joined #pypy.17:02
kkris (~kris@93-82-33-167.adsl.highway.telekom.at) left irc: Ping timeout: 240 seconds17:03
voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace17:14
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds17:15
Alex_Gaynorfijal: pong17:20
fermianyon (~lane@c-71-229-21-197.hsd1.al.comcast.net) joined #pypy.17:21
JaRoel|4d (~jaroel|4d@2001:980:7133:1:69cd:db5d:7ebe:3291) joined #pypy.17:30
nedbat (~nedbat@python/psf/nedbat) left irc: Quit: Leaving17:39
fijalAlex_Gaynor: I don't remember any more :)17:42
fijalah17:42
fijalwhat do you want from API on the python side?17:42
Alex_Gaynorfor jiteviewer?17:43
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.17:50
nedbat (~nedbat@python/psf/nedbat) joined #pypy.17:58
arigato (~arigo@adsl-84-227-72-198.adslplus.ch) left irc: Ping timeout: 252 seconds17:59
Guest (~textual@216.190.27.14) joined #pypy.18:08
Nick change: Guest -> rekamso18:08
fijalAlex_Gaynor: yeah18:10
fijalAlex_Gaynor: sorry for spotty presence :/18:10
fijalfor tracebin18:10
Alex_Gaynorfijal: the abort stuff would be cool18:11
Alex_Gaynoras would real parameters18:11
Alex_Gaynorand having all this stuff work with numpy and rsre as well :)18:11
fijalsorry?18:11
fijalyou mean having in-python on_compile hooks?18:11
fijaland on_abort hooks?18:11
fijalon_abort works with numpy/rsre18:11
fijalon_compile I'll move18:11
Alex_Gaynorit does?18:12
fijalyes18:12
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) left irc: Ping timeout: 248 seconds18:12
Alex_Gaynorok, cool18:12
fijalit's global now18:12
Alex_Gaynoroh this is the portal thing18:12
fijalyes :)18:12
Alex_Gaynorcool18:12
fijaland I want to move on_compile there18:12
Action: Alex_Gaynor wonders if global isn't a bit backwards from per-obj18:12
fijalthe only problem is that right now it cases greenkey to types18:12
fijalno it's not18:12
fijalbecause per-object is a bit arbitrary18:12
fijalsince you can have an obscure combination of jitdrivers in your trace18:13
Alex_GaynorRight18:13
fijalalso, before was not per-object, it was one-object-only :)18:13
Alex_Gaynorfijal: would be cool if trace in compile_hook was split more into objects, rather than just strings, but that's more long term I guess18:15
fijalit is18:17
fijalbut anyway18:17
fijallet's discuss this for a second18:17
fijalare we aiming at tracebin running from the log or from the on_compile hook?18:17
fijalbecause there are different tradeoffs18:17
Alex_GaynorI'd like it to work with either, but I think the hooks is more important18:18
fijalI don't think pursuing both ways is a good idea18:18
fijalwe should promote one and forget about the other IMO18:18
Alex_Gaynorhooks then18:18
Alex_Gaynorpypy -mtracebin is such a better UI18:18
fijalyes it is18:19
fijalyou can whack it to work with logs, but anyway18:19
fijalso the hooks *right now* split trace into resops18:19
fijaland expose resops18:19
Alex_Gaynorthat's a good start18:19
Alex_Gaynorwe need it for all loop types though :)18:19
fijalyes, I can do that tomorrow probably18:19
fijalbut you no longer can measure total time probably18:20
fijalbecause they're slightly more expensive18:20
fijalunless you do some consumer API or so18:20
fijalthat would just dump it to a pipe18:20
Alex_Gaynorsure I can measure total time, it's per-section time that' s hard18:20
fijalare you fine with -mtracebin taking say 20% more time?18:20
fijalor rather, prolonging warmup 2x?18:21
fijalfor example18:21
fijalthat's pretty extreme18:21
Alex_Gaynorwhy is it so large?18:21
fijalI didn't say it is18:24
fijalbut probably would be slightly larger that it is now18:24
fijalis it fine18:24
fijal?18:24
fijalyou can do tricks though18:25
Alex_Gaynoryes, I think so18:25
fijalok18:25
Action: Alex_Gaynor is a fn of tricks18:25
fijalwe can think later about making it RPython18:25
fijalbut let's get this going18:25
fijalso ok18:25
fijalnow, the other part - we're missing assembler and counters18:25
fijaldo you want a hook for that as well?18:26
Alex_Gaynorassembler could be a part of on_compile hook, I think?18:26
fijalyes18:26
Alex_Gaynorand then you can do lininig it up with resops better than +offsets18:26
Alex_Gaynor:)18:26
fijalI also want a hook that does "gimme all compiled loops for this code obj"18:26
fijalyes18:26
Alex_Gaynorcounters should be a single hook that gets called in the end, I guess?18:26
Alex_Gaynormaybe even get_counters18:27
fijalmy goal is phase out the logs completely or semi-completely18:27
Alex_Gaynorthat I just call at the end of the process18:27
fijalyes18:27
fijalatexit.register(...)18:27
Alex_Gaynordoesn't have to be that, -mtracebin is just try: execfile() finally: do_end_stuff() or so18:27
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) joined #pypy.18:29
Alex_GaynorI need to think about how to structure teh database now I guess18:29
rekamso (~textual@216.190.27.14) left irc: Ping timeout: 244 seconds18:31
danchr (~danchr@cl-848.chi-02.us.sixxs.net) left irc: Ping timeout: 252 seconds18:44
danchr (~danchr@cl-848.chi-02.us.sixxs.net) joined #pypy.18:44
nettok (~quassel@200.119.184.55) joined #pypy.18:51
ericflo (~ericflo@75.103.8.110) joined #pypy.18:56
fijalantocuni: ping19:00
antocunipong19:05
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) joined #pypy.19:08
fijalfigured out myself by now :)19:12
mattipfijal: ping19:26
fijalmattip: pong19:27
Alex_Gaynorfijal: i'll try to reivew some of mikefc's patches tonight19:27
mattipI fixed the repr() bugs by adding three tests and making it work.19:27
fijalmattip: cool19:28
mattipbased on mikefc's work19:28
fijalanyway, can you guys coordinate, I have to drive to the airport soonish19:28
mattipI'm tempted to commit directly to default since the fixes are small, it passes test_ztranslate and all other tests too.19:28
fijalmattip: can you post the diff, so I can have a quick look?19:28
Alex_GaynorI can also take a quick look now19:29
fijalor alex19:29
mattiphttp://paste.pocoo.org/show/528955/19:30
Alex_Gaynor(diff version: http://paste.pocoo.org/show/528956/ )19:30
Alex_Gaynorthis isn't correct19:32
Alex_Gaynor(it wasn't correct before either)19:32
Alex_Gaynorit's the platform sized integer that shouldn't be displayed19:32
Alex_Gaynornot the int64 ones19:32
mattipon my 32 bit linux19:33
mattipnumpy only hides the dtype if it is long or 'int64'19:33
Alex_Gaynormmm, that's not the behavior I'm seeing here19:34
voidspace (~voidspace@python/psf/voidspace) joined #pypy.19:34
Action: mattip check again19:34
Action: mattip that's "me check again"19:34
`fox` (~fox@host142-107-dynamic.45-79-r.retail.telecomitalia.it) left irc: Ping timeout: 253 seconds19:35
mattipAlex_Gaynor: it seems you are (again) correct. I wonder what I was seeing before?19:36
nettok_ (~quassel@200.119.185.249) joined #pypy.19:36
Alex_Gaynormattip: dunno, I find it really easy to mix up all the different int types19:36
mattipWhat's the quickest way to get the native int size?19:36
mattipplatform something?19:37
mfoord (~anonymous@python/psf/voidspace) joined #pypy.19:37
mfoord (~anonymous@python/psf/voidspace) left irc: Client Quit19:37
nettok (~quassel@200.119.184.55) left irc: Ping timeout: 248 seconds19:38
Alex_Gaynorit's anythign where .kind == INT and .get_element_size() == rffi.sizeof(lltype.Integer)19:38
Alex_Gaynor(I think)19:38
gutworthand by that you mean lltype.Signed?19:40
`fox` (~fox@host94-105-dynamic.54-82-r.retail.telecomitalia.it) joined #pypy.19:40
Alex_Gaynoryeah19:40
Alex_Gaynor:)19:40
mattiphow about dtype('i')19:40
Alex_Gaynorthat'll match the condition I suggested19:41
fijal (~fijal@41.53.95.237) left irc: Read error: Connection reset by peer19:44
lmoura (lmoura@gateway/indt/session) left irc: Quit: Leaving19:47
mattipAlex_Gaynor: so now the tests must also check the native int size, and adjust the expectaions accordingly.20:01
mattipexpectations*20:02
zain__ (~textual@cpe-68-173-70-124.nyc.res.rr.com) joined #pypy.20:03
Nick change: zain__ -> Guest1793020:04
Guest17930 (~textual@cpe-68-173-70-124.nyc.res.rr.com) left irc: Quit: Computer has gone to sleep.20:16
lesshastehi all20:16
`fox` (~fox@host94-105-dynamic.54-82-r.retail.telecomitalia.it) left irc: Ping timeout: 240 seconds20:16
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.20:19
int3__ (~int3__@bb220-255-192-11.singnet.com.sg) left irc: Remote host closed the connection20:19
int3__ (~int3__@bb220-255-192-11.singnet.com.sg) joined #pypy.20:19
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving20:29
lizardo (~lizardo@189.2.128.130) left irc: Quit: Leaving20:29
lmoura (lmoura@gateway/indt/session) joined #pypy.20:32
nettok (~quassel@200.119.150.91) joined #pypy.20:35
nettok_ (~quassel@200.119.185.249) left irc: Ping timeout: 240 seconds20:36
antocuni (~antocuni@host117-71-dynamic.58-82-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds20:37
mattipAlex_Gaynor: http://paste.pocoo.org/show/52898920:42
kenaan12alex_gaynor default 116a589f1a038a 15/pypy/module/sys/app.py: Update some copyright years.20:43
Alex_Gaynormattip: sorry I'm running out now, I don't have time to take a look ATM20:43
mcdonc_ (~mcdonc@ip72-209-213-54.dc.dc.cox.net) joined #pypy.20:44
mattipok, I'll put it on a branch.20:44
mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) left irc: Ping timeout: 240 seconds20:44
kenaan12mattip numpypy-repr-fix 1168dec1e17bd9 15/pypy/module/micronumpy/: add tests for issue 964 and more, make tests pass20:48
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) joined #pypy.20:53
CIA-5303mattip 07roundup * 10#964/numpypy.zeros([]) returns wrong result: 20:54
CIA-53[in-progress] Fixed in changeset 68dec1e17bd9149babafdc36eb109eb8d41c8a61, based on mikefc's20:54
CIA-53patch. At the same time fixed bug for showing dtype ... * 14https://bugs.pypy.org/issue96420:54
Tanriol (~dennis@77.41.27.17) joined #pypy.21:20
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) left irc: Remote host closed the connection21:26
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) joined #pypy.21:27
whyking (~quassel@p4FFB606C.dip.t-dialin.net) left irc: Ping timeout: 240 seconds21:34
fijal (~fijal@197.110.79.161) joined #pypy.21:51
fijalhi21:52
Tanriol (dennis@77.41.27.17) left #pypy.21:55
fschwidom (~fschwidom@46.115.12.41) joined #pypy.21:55
fschwidom (~fschwidom@46.115.12.41) left irc: Remote host closed the connection21:56
fijal (~fijal@197.110.79.161) left irc: Read error: Connection reset by peer21:56
ssbrHey guys, so, I'm giving writing a regular expression implementation a shot (not least because I want to try implementing scheme-style pattern matching in Python). What should I read to know how to get regexps JIT-compiled?22:00
ssbrthe structure is basically following that (so far) given by russ cox (RE2 author) of a compiler from regular expressions to a regexp bytecode, and then an interpreter for said bytecode22:01
moreatissbr: are you aware of the re2 pacakge and google regex library bindings?22:02
ssbrmoreati: Yes.22:02
ssbrThat said, I couldn't get the re2 package to work, but my goals are different22:02
ssbrI started writing this so that I could pattern match on Python objects22:02
kkris1 (~kris@80-123-47-161.adsl.highway.telekom.at) left irc: Quit: Leaving.22:03
ssbrBesides, I will probably need to add backtracking search for nontrivial patterns.22:03
ssbr(scheme's allows you to apply an arbitrary predicate to a match)22:04
ssbr(eh, Racket's, I should say, although I think the "usual" scheme pattern matching library does the same)22:04
fijal (~fijal@41.48.168.162) joined #pypy.22:11
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) left irc: 22:12
mikefcmattip: if I wanted to write a cumsum() method, is it possible to do it at the app level?  I can see where it'd go at the interp level (in the classes in intep_numarray), but if I wrote it at the applevel, how would i get it in there?22:12
mattipAlex_Gaynor: I found some corner cases for large arrays with few rows or columns and added a fix for repr().22:15
kenaan12mattip numpypy-repr-fix 11ee3e5819364d 15/pypy/module/micronumpy/: additional tests, fixes to pass them22:15
mattipI couldn't find a way to use iterators instead of directly accessing strides yet.22:16
ssbrthis looks useful: http://morepypy.blogspot.com/2011/04/tutorial-writing-interpreter-with-pypy.html22:16
mattipPlease review, critique, and hopefully merge.22:16
espes (espes@2001:470:1f05:ca1::) left irc: Read error: Connection timed out22:17
espes (espes@2001:470:1f05:ca1::) joined #pypy.22:18
int3__ (~int3__@bb220-255-192-11.singnet.com.sg) left irc: Remote host closed the connection22:19
mattipmikefc: I'm thinking.22:23
durin42 (~durin@adium/durin42) got netsplit.22:24
Count_Niedar (~nnscript@ip68-231-248-182.hr.hr.cox.net) got netsplit.22:24
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) got netsplit.22:24
htoothrot (~mux@unaffiliated/htoothrot) got netsplit.22:24
timotimo (~timo@gallifrey.wybt.net) got netsplit.22:24
exarkun (~exarkun@wordeology.com) got netsplit.22:24
Arnar (~arnarb@li187-148.members.linode.com) got netsplit.22:24
exarkun (~exarkun@wordeology.com) returned to #pypy.22:24
#pypy: mode change '+o exarkun' by ChanServ!ChanServ@services.22:24
mikefcok. sorry. i thought there might be something obvious I'm missing, but I can't see any examples of what I want to do in the current code.22:25
durin42 (~durin@adium/durin42) returned to #pypy.22:25
Count_Niedar (~nnscript@ip68-231-248-182.hr.hr.cox.net) returned to #pypy.22:25
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) returned to #pypy.22:25
htoothrot (~mux@unaffiliated/htoothrot) returned to #pypy.22:25
timotimo (~timo@gallifrey.wybt.net) returned to #pypy.22:25
Arnar (~arnarb@li187-148.members.linode.com) returned to #pypy.22:25
mattipmikefc: sorry, no idea.22:33
fijalmattip: I'll have a look (hopefully) tomorrow22:33
mikefcok. thanks. I'll post something to the mailing list and see if there are any ideas.22:33
Action: mikefc is lazy and wants to get big chunks of numpypy working by borrowing from numpy's code22:34
lahwranheavens, ssbr 22:36
fprimex (fprimex@wireless-dhcp200d088.fsu.edu) left irc: Quit: Leaving22:36
fijalmikefc: good :)22:36
lahwranyou're writing a whole language just for this pattern matching thing?22:36
lahwranyou're going to set some kind of record at this rate22:36
mikefcfijal: my question to mattip was:  if I wanted to write a cumsum() method, is it possible to do it at the app level?  I can see where it'd go at the interp level (in the classes in intep_numarray), but if I wrote it at the applevel, how would i get it in there?22:37
mikefcfijal: any ideas?22:37
ssbrlahwran: what22:37
fijalmikefc: there are two possible places22:37
ssbrOh.22:37
ssbrlahwran: no, I'm reimplementing a language22:37
lahwranoh, regexes22:37
fijalyou can put it in app_numpy.py22:37
lahwranI didn't scroll up very far :P22:38
fijalor you can do all the work of reusing numpy in lib_pypy/numpy22:38
fijalthat simply imports _numpy or micronumpy or so22:38
fijalhm22:40
mikefcfijal: for now I'd just like to "put it in app_numpy.py".   But all the functions in there are class level functions, when I really want to write an instance method (I hope the terminology is right).22:40
fijalhow bad would that be to be implemented at applevel btw?22:40
fijalah, you want array.cumsum as a method?22:40
mikefcfijal: yes :)22:40
fijalthen I fear you have to implement it at interp-level22:41
fijalwhich makes quite some sense, no?22:41
fijallike any other axis-based function22:41
fijaldon't you have a branch that does it?22:41
fijalor someone else?22:41
fijalI got a bit lost, I have to review those branches tomorrow :/22:42
mikefcI was hoping that I could write it at app level, because I'm not smart enough to write interp level stuff yet22:42
fijalmattip was definitely working on it22:42
fijalmikefc: don't worry it's not that hard :)22:42
fijalit should be at interp-level really22:42
fijalmikefc: I can help with that, just not tonight22:42
fijalanyway, /me off22:43
mikefcthanks22:43
Action: mikefc is off too. work :/22:43
fijalmikefc: I will look into axes-based ufuncs22:43
fijalbecause they're more or less similar22:43
fijalso sometime this week, ok?22:43
mikefcfijal: ok.  cumsum() was only an example. there are lots of methods i'd like, but don't necessarily want to write in interp level :)22:43
mikefcok. gotta go.22:43
mikefcafk22:43
fijalmikefc: they'll all be like 2-lines long22:44
fijallike sum right now22:44
fijalso don't worry really22:44
fijal (~fijal@41.48.168.162) left irc: Read error: Connection reset by peer22:50
kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) joined #pypy.22:56
mattip (~chatzilla@bzq-79-176-117-44.red.bezeqint.net) left irc: Quit: ChatZilla 0.9.88 [Firefox 8.0/20111115183158]22:56
bbot2Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/76423:00
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/19223:00
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/100123:00
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/2023:00
furrykef (furrykef@ip68-12-180-167.ok.ok.cox.net) joined #pypy.23:05
furrykefIs there anybody working on a py2exe-like thing for PyPy? I'd like to package my program with PyPy instead of CPython for performance, and I'd rather not start from scratch if there's another project I could contribute to.23:07
gutworthno23:09
kennethreitzfurrykef: I want that SO badly.23:10
kennethreitzbut nobody cares. at all. it's tragic.23:10
furrykefHmm, I wonder why.23:11
furrykefMy assembler runs 10x as fast under PyPy as under CPython (at least for certain inputs).23:11
Dulaksounds like you guys just found a project to collaborate on23:12
furrykefI dunno, I might just stick with Python 2.6 + psyco (which still runs about 2x as fast as CPython for my project)23:14
lahwranperhaps look at adapting py2exe itself23:14
lahwranor pyinstaller23:14
stakkarsall: can I change the IRC log headline?23:18
stakkarsI'd like to replace23:18
stakkarshttp://www.tismer.com/pypy/irc-logs/pypy/23:19
stakkarswith23:19
stakkarshttp://tismerysoft.net/pypy/irc-logs/pypy/23:20
stakkarsthe tismer.com thing works by accident.23:20
stakkarsor should I make it explicit? Then I'd toss the other one.23:21
stakkarsbut there should be only one linkt, I think.23:21
Niedar (~nnscript@ip68-231-248-182.hr.hr.cox.net) joined #pypy.23:26
kennethreitzunfortunately, it's quite complicated stuff23:28
Topic changed on #pypy by kennethreitz!kennethre@c-24-127-96-129.hsd1.va.comcast.net: PyPy, a fast and compliant python interpreter | IRC logs: http://tismerysoft.net/pypy/irc-logs/pypy/ | if you want it to be virtual, you should not force it23:28
kennethreitzstakkars: \o/23:28
kennethreitzthat doesn't work for me though23:29
Count_Niedar (~nnscript@ip68-231-248-182.hr.hr.cox.net) left irc: Ping timeout: 240 seconds23:29
stakkarskennethreitz: what does not work for you?23:30
kennethreitzstakkars: that link23:30
stakkarsplease more precisely. that link that is visible on IRC, or that I proposed?23:31
stakkarsah, sorry, I just saw the change.23:32
stakkarskennethreitz: ugh, yeah, it is "tismerysoft.de" that works right now.23:33
stakkarswould you mind to change it again?23:34
stakkarsand which one should be it, in the end?23:34
Topic changed on #pypy by kennethreitz!kennethre@c-24-127-96-129.hsd1.va.comcast.net: PyPy, a fast and compliant python interpreter | IRC logs: http://tismerysoft.de/pypy/irc-logs/pypy/ | if you want it to be virtual, you should not force it23:34
kennethreitzstakkars: done :)23:35
stakkarsok, that's fine so far. Although tismerysoft.net is nicer, no?23:35
stakkarsanyway, I can toss that old tismer.com shit now.23:36
stakkarsok, let's keep it that way...23:36
stakkarsdid somebody realize that the site was replaced completely, yesterday?23:37
stakkarspresumably not, which is a success.23:38
Goron (~LC@h171.163.191.173.dynamic.ip.windstream.net) joined #pypy.23:39
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 276 seconds23:48
mcdonc_Alex_Gaynor: fwiw, i got sick of untested servers, so http://pypi.python.org/pypi/waitress/ (continually tested on pypy)23:50
Alex_Gaynormcdonc_: the world needed a new HTTP server like it needed a hole in it's head, nice :)23:51
mcdonc_i know! ;-)23:51
mcdonc_but it did23:51
mcdonc_at least a new wsgi server that people actually could verify worked23:51
--- Tue Jan 3 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!