#pypy IRC log for Friday, 2009-12-25

victorpoluceno (n=victorgp@187.5.147.230) left irc: Read error: 60 (Operation timed out)00:06
victorpoluceno (n=victorgp@187.52.160.24) joined #pypy.00:06
damien_ (n=damien@damien.grassart.com) left irc: "Leaving"00:08
netto (n=netto@200.119.150.7) left irc: "Ex-Chat"00:08
gonsor (n=quassel@p57A89965.dip0.t-ipconnect.de) joined #pypy.00:12
tav (n=tav@89.243.205.194) joined #pypy.00:13
victorpoluceno (n=victorgp@187.52.160.24) left irc: Read error: 60 (Operation timed out)00:38
victorpoluceno (n=victorgp@187.5.177.149) joined #pypy.00:38
victorpoluceno (n=victorgp@187.5.177.149) left irc: Read error: 60 (Operation timed out)00:46
victorpoluceno (n=victorgp@189-31-87-141.bnut3700.dsl.brasiltelecom.net.br) joined #pypy.00:46
victorpoluceno (n=victorgp@189-31-87-141.bnut3700.dsl.brasiltelecom.net.br) left irc: Read error: 60 (Operation timed out)00:58
victorpoluceno (n=victorgp@189.72.84.9) joined #pypy.00:59
ebo^ (n=ebolad@233.195.116.85.dsl.manitu.net) left irc: "Verlassend"00:59
stakkars (n=tismer@i59F7E2C6.versanet.de) left irc: Read error: 104 (Connection reset by peer)01:08
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) left irc: Remote closed the connection01:10
stakkars (n=tismer@i59F7E2C6.versanet.de) joined #pypy.01:16
lux` (n=lux@151.95.189.174) left irc: Remote closed the connection01:46
Misto (n=Misto@ppp-63-89.32-151.iol.it) joined #pypy.02:11
magcius (n=jstpierr@c-24-61-243-212.hsd1.ma.comcast.net) left irc: Read error: 110 (Connection timed out)02:13
hikoz (n=hikoz@27.183.244.43.ap.yournet.ne.jp) joined #pypy.02:14
magcius (n=jstpierr@c-24-61-243-212.hsd1.ma.comcast.net) joined #pypy.02:22
gonsor (n=quassel@p57A89965.dip0.t-ipconnect.de) left irc: Read error: 104 (Connection reset by peer)02:23
Misto (n=Misto@ppp-63-89.32-151.iol.it) left irc: 02:34
hpk (n=hpk@HSI-KBW-095-208-109-209.hsi5.kabel-badenwuerttemberg.de) left irc: Read error: 60 (Operation timed out)03:08
hruske (n=Gasper@internet-95-159-203-30.narocnik.mobitel.si) left irc: Read error: 110 (Connection timed out)03:10
pemryan (n=pem@159.226.35.246) left irc: Remote closed the connection03:28
Cheshirc (n=tu@unaffiliated/cheshirc) left irc: Read error: 113 (No route to host)04:14
Cheshirc (n=tu@unaffiliated/cheshirc) joined #pypy.04:14
Cheshirc (n=tu@unaffiliated/cheshirc) left irc: Read error: 113 (No route to host)04:38
Cheshirc (n=tu@unaffiliated/cheshirc) joined #pypy.04:38
lux` (n=lux@151.95.189.174) joined #pypy.04:46
lux` (n=lux@151.95.189.174) left irc: Remote closed the connection04:49
Ademancan global dicts be modified?05:10
Ademanlike adding keys and such?05:10
gutworthin rpython?05:10
Ademanyeah05:10
gutworthno05:11
Ademani know that globals are constant, but i don't know if that makes them immutable as well05:11
Ademanah05:11
gutworthonly global instances can be changed05:11
Ademanso if i wanted to have a mutable global dictionary i need to make a class that holds the data?05:11
Ademanalso... generators aren't rpython right?05:12
gutworthno05:12
gutworthand yes05:12
gutworth[switch]05:12
Ademanah05:12
Ademan:-/ makes this a bit uglier, oh well05:13
gutworthwell, globalness is already ugly05:13
Ademangutworth: can i instantiate rlib.rarithmetic.singlefloat and use it like a normal rpython primitive?05:23
gutworthyes05:24
Ademanooh, awesome!05:29
bbits (n=blue@adsl-99-142-45-232.dsl.emhril.sbcglobal.net) joined #pypy.05:37
Ademangutworth: is the staticmethod decorator rpython?05:43
Cheshirc (n=tu@unaffiliated/cheshirc) left irc: Read error: 110 (Connection timed out)05:56
Cheshirc (n=tu@unaffiliated/cheshirc) joined #pypy.05:56
tav (n=tav@89.243.205.194) left irc: 06:01
victorpoluceno (n=victorgp@189.72.84.9) left irc: Read error: 60 (Operation timed out)06:55
victorpoluceno (n=victorgp@189.73.98.77) joined #pypy.06:55
Alex_Gaynor (n=alex@c-67-163-15-166.hsd1.il.comcast.net) left irc: "Ex-Chat"07:57
jewel (n=jewel@vc-41-28-248-149.umts.vodacom.co.za) joined #pypy.08:19
Misto (n=Misto@ppp-174-90.32-151.iol.it) joined #pypy.08:50
jewel (n=jewel@vc-41-28-248-149.umts.vodacom.co.za) left irc: Read error: 60 (Operation timed out)09:26
jewel (n=jewel@vc-41-28-232-243.umts.vodacom.co.za) joined #pypy.09:41
netto (n=netto@200.119.150.10) joined #pypy.09:41
hpk (n=hpk@HSI-KBW-095-208-109-209.hsi5.kabel-badenwuerttemberg.de) joined #pypy.09:50
lac_ (n=lac@c-96c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: "Konversation terminated!"10:17
lac_ (n=lac@c-96c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.10:17
lac (n=lac@c-96c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.10:18
hruske (n=Gasper@internet-213-229-212-33.narocnik.mobitel.si) joined #pypy.10:27
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) joined #pypy.10:27
ebo^ (n=ebolad@233.195.116.85.dsl.manitu.net) joined #pypy.10:31
Nick change: armtone -> nosetone10:31
jewel (n=jewel@vc-41-28-232-243.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)10:36
netto (n=netto@200.119.150.10) left irc: "Ex-Chat"10:37
jewel (n=jewel@vc-41-30-127-143.umts.vodacom.co.za) joined #pypy.10:42
scorchsaber (n=scorchsa@d72-39-232-122.home1.cgocable.net) left irc: Read error: 60 (Operation timed out)10:46
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) left irc: Remote closed the connection10:47
weltling (n=toll@belsky.info) joined #pypy.10:59
jewel (n=jewel@vc-41-30-127-143.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)11:10
jewel (n=jewel@vc-41-27-51-16.umts.vodacom.co.za) joined #pypy.11:11
stakkars (n=tismer@i59F7E2C6.versanet.de) left irc: 11:14
arigato (n=arigo@adsl-188-155-115-199.adslplus.ch) joined #pypy.11:17
hruskex (n=Gasper@internet-213-229-221-107.narocnik.mobitel.si) joined #pypy.11:36
gonsor (n=quassel@p57A899D6.dip0.t-ipconnect.de) joined #pypy.11:36
hikoz (n=hikoz@27.183.244.43.ap.yournet.ne.jp) left irc: "Leaving..."11:36
jewel (n=jewel@vc-41-27-51-16.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)11:44
nullpo (n=nullpo@221x252x46x83.ap221.ftth.ucom.ne.jp) joined #pypy.11:44
hruske (n=Gasper@internet-213-229-212-33.narocnik.mobitel.si) left irc: Read error: 60 (Operation timed out)11:44
jewel (n=jewel@vc-41-27-17-100.umts.vodacom.co.za) joined #pypy.11:46
pemryan (n=pem@159.226.35.246) joined #pypy.11:47
lesshaste (n=lesshast@87-194-206-189.bethere.co.uk) joined #pypy.12:07
arigato (n=arigo@adsl-188-155-115-199.adslplus.ch) left irc: "See you"12:07
lesshastehello all12:09
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) joined #pypy.12:12
gonsor (n=quassel@p57A899D6.dip0.t-ipconnect.de) left irc: Read error: 60 (Operation timed out)12:23
lesshastehi lucian 12:34
lesshasteI just tried using pypy on a short python script that calls networkx... it made no difference at all to the run time12:35
lesshasteI am new to this so any tips much appreciated12:35
lesshastehttp://pastebin.ca/1726673 is the entire program12:36
lux` (n=lux@151.95.189.174) joined #pypy.12:37
jewel (n=jewel@vc-41-27-17-100.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)12:41
lux` (n=lux@151.95.189.174) left irc: Remote closed the connection12:46
lesshaste (n=lesshast@87-194-206-189.bethere.co.uk) left irc: "Leaving"12:47
magcius (n=jstpierr@c-24-61-243-212.hsd1.ma.comcast.net) left irc: Read error: 110 (Connection timed out)12:57
Trundle (n=andy@p5B13317B.dip.t-dialin.net) joined #pypy.13:01
jewel (n=jewel@vc-41-30-87-153.umts.vodacom.co.za) joined #pypy.13:44
Nick change: hruskex -> zejn13:52
ElectronicRU (n=Electron@pppoe2478.mv.ru) joined #pypy.13:57
ElectronicRUAdeman: look in your private14:06
jewel (n=jewel@vc-41-30-87-153.umts.vodacom.co.za) left irc: Read error: 60 (Operation timed out)14:12
gonsor (n=quassel@p57A899D6.dip0.t-ipconnect.de) joined #pypy.14:54
ElectronicRUAdeman: attracting,...15:57
Last message repeated 2 time(s).16:47
scorchsaber (n=scorchsa@d72-39-232-122.home1.cgocable.net) joined #pypy.16:47
pemryan (n=pem@159.226.35.246) left irc: Remote closed the connection16:56
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) left irc: Remote closed the connection17:06
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) joined #pypy.17:07
jewel (n=jewel@vc-41-27-32-76.umts.vodacom.co.za) joined #pypy.17:19
tav (n=tav@89.243.205.194) joined #pypy.17:30
MrMeanie (n=Geoff@82.152.254.168) joined #pypy.17:47
MrMeanieIs the logical constraint solver still a part of PyPy?17:47
ElectronicRUFrom where could you advise me to start writing interpreter for small language?17:49
gutworth_ (n=benjamin@64-131-1-181.usfamily.net) joined #pypy.17:55
gutworth (n=benjamin@64-131-1-181.usfamily.net) left irc: Remote closed the connection17:55
Alex_Gaynor (n=alex@c-67-163-15-166.hsd1.il.comcast.net) joined #pypy.17:57
Misto (n=Misto@ppp-174-90.32-151.iol.it) left irc: 18:02
Misto (n=Misto@ppp-174-90.32-151.iol.it) joined #pypy.18:10
Misto (n=Misto@ppp-174-90.32-151.iol.it) left irc: Client Quit18:11
lesshaste (n=lesshast@87-194-206-189.bethere.co.uk) joined #pypy.18:14
lesshastehi 18:14
lesshasteI have some simple code that appears quite a bit slowed in pypy than standard python18:15
lesshasteis there somewhere I can post it?18:15
ousado (n=johnny@p5B3C0632.dip0.t-ipconnect.de) joined #pypy.18:22
lucianlesshaste: put it on a pastebin and ask someone with experience (fijal,arigato,etc.)18:23
lucianlesshaste: about what you asked me yesterday, i'm as newbie as you are18:23
jewel_ (n=jewel@vc-41-27-136-199.umts.vodacom.co.za) joined #pypy.18:26
jewel (n=jewel@vc-41-27-32-76.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)18:26
ousado__ (n=johnny@p5B3C06A6.dip0.t-ipconnect.de) left irc: Read error: 60 (Operation timed out)18:26
lux` (n=lux@151.95.176.190) joined #pypy.18:31
victorpoluceno (n=victorgp@189.73.98.77) left irc: Read error: 60 (Operation timed out)18:33
victorpoluceno (n=victorgp@187.5.177.200) joined #pypy.18:34
lux` (n=lux@151.95.176.190) left irc: Remote closed the connection18:42
lac_MrMeanie: I don't think anybody has been maintaining that, so maybe it works, but probably it doesn't.18:46
lac_ (n=lac@c-96c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) left #pypy ("Konversation terminated!").18:48
arigato (n=arigo@adsl-188-155-200-92.adslplus.ch) joined #pypy.18:50
jewel_ElectronicRU, I would advise you try a small lisp interpreter, if you're interested in lisp or scheme18:57
ElectronicRUTo explore its code?18:58
ElectronicRUI need certain lang.18:58
ElectronicRUAnd start point to explore.18:58
kenaan03arigo 10r7027114 07M(pypy/trunk/pypy/jit/metainterp/optimizeopt.py): is_always_pure() is a subset of has_no_side_effect(). ...19:00
jewel_ElectronicRU, you can try this: http://www.paulgraham.com/rootsoflisp.html19:00
ElectronicRUI. Need. Small. Language. Interpreter. Example. In. RPython.  I. Need. No. Scheme.19:01
arigatopypy/jit/tl/tl*.py ?19:02
arigatothese are such examples, designed for testing the jit generator but perfectly readable as small examples of interpreters19:03
TrundleIs barla an awful example for an interpreter?19:06
Action: arigato doesn't know about barla19:07
Trundlearigato: That small language I wrote some time ago.19:08
victorpoluceno (n=victorgp@187.5.177.200) left irc: Read error: 60 (Operation timed out)19:10
victorpoluceno (n=victorgp@187.52.161.95) joined #pypy.19:11
lesshastelucian: ok thanks19:11
Action: lesshaste will be back in a second19:12
lesshaste (n=lesshast@87-194-206-189.bethere.co.uk) left irc: "Leaving"19:12
lesshaste (n=lesshast@87-194-206-189.bethere.co.uk) joined #pypy.19:15
lesshastearigato, hi19:16
arigatohi19:16
lesshasteThis rather simple piece of code http://pastebin.ca/1726856 seems quite a lot slower using pypy than with standard python19:16
lesshasteassuming that I am testing it right of course19:16
lesshastei am just putting #!/dir/pypy-trunk/pypy/translator/goal/pypy-c for the pypy test19:17
lesshasteand #!/usr/bin/python for the python one19:17
lesshasteI compiled with --opt=jit19:17
arigatoa bit hard to follow without knowing what bidirectional_dijkstra() does,19:19
arigatobut I'll keep a copy of the link for later19:19
lesshasteI profiled the python run19:19
lesshasteand half the time is just in the add_edge part it seems19:19
lesshastehow do you profile the pypy run?19:20
arigatothe same, I suppose19:21
lesshastepypy -m cprofile?19:21
arigatoshould work19:21
lesshastetimings.. python 32 secs, pypy 4819:22
hpk (n=hpk@HSI-KBW-095-208-109-209.hsi5.kabel-badenwuerttemberg.de) left irc: Read error: 110 (Connection timed out)19:22
lesshasteok trying it now19:23
lesshasteif there is anything useful I can do to help make it easier to debug do let me know19:23
lesshasteI'll see if I can make an even more minimalist version that shows the timings difference if that helps19:24
lesshastepypy-c -m cProfile  test.py doesn't seem to have output anything meaningful19:25
lucianlesshaste: try profile instead of cProfile19:26
lesshastelucian, how? Can you give me an example command line?19:28
lesshastepypy-c -m profile  doesn't work19:28
lucianlesshaste: pypy-c -m profile. there's no cProfile for pypy since there's no C modules. it may be aliased19:29
lesshastejoin #python19:29
lucianlesshaste: and pypy might not have the -m switch19:29
luciantry import profile19:29
lucianlesshaste: i've got to go19:29
arigatolucian: it does19:29
lesshastepypy-c -m cProfile works in some sense but -m profile just gives me error messages19:29
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) left irc: Remote closed the connection19:29
lucian (n=lucian@cpc6-newc9-0-0-cust64.gate.cable.virginmedia.com) joined #pypy.19:30
lesshastearigato, ok.. how about this? http://pastebin.ca/172686319:35
arigatoas I said, I'll just write it down for later19:37
arigatobut thanks anyway19:37
lesshasteno problem. I thought a really small example with the dijkstra part might help19:38
lesshastewithout the dijkstra part19:38
arigatoyes, thanks19:42
victorpoluceno (n=victorgp@187.52.161.95) left irc: Read error: 110 (Connection timed out)19:47
victorpoluceno (n=victorgp@189.75.22.10) joined #pypy.19:48
jewel_ (n=jewel@vc-41-27-136-199.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)19:49
nullpo (n=nullpo@221x252x46x83.ap221.ftth.ucom.ne.jp) left irc: Read error: 110 (Connection timed out)19:50
jewel_ (n=jewel@vc-41-29-77-24.umts.vodacom.co.za) joined #pypy.19:52
ElectronicRUHello,20:09
ElectronicRUI'm reviewing py2.6 features.20:09
ElectronicRUWhen PyPy is going to migrate to 2.6?20:09
ElectronicRUNext year?20:10
ElectronicRUOr there is no people to do this?20:10
ElectronicRUOr no sense of it?20:10
arigatoit's very unclear20:10
arigato"a bit all of the above" is the best answer20:10
ElectronicRUOne day we'll need it. Py2.7 comes.20:12
ElectronicRUOr, (secret), project 'll go his own way? With its own improvements?20:13
ElectronicRUWhere is 'property' class in objspace.std?20:17
arigatono, we don't want to "improve" the language, we're rather on the general opinion that the Python language is good enough as it is20:19
arigato'property' is in pypy/module/__builtin__/20:19
ElectronicRUNot parser.20:19
ElectronicRUFor ex. unicode-byte dilemma appears in py2.620:20
ElectronicRUThanks for location.20:20
ElectronicRUCould I commit change adds prop.getter/setter decorators?20:20
gonsor (n=quassel@p57A899D6.dip0.t-ipconnect.de) left irc: Read error: 110 (Connection timed out)20:20
arigatoobviously, everyone has their favorite "Python is great but if only *that* were improved".  Please go discuss it elsewhere, though :-)20:20
ElectronicRUSo could I?20:22
arigatoElectronicRU: unlikely -- if you mean adding properties to the source code of pypy, then no because we try to some extent to remain 2.3-compatible20:22
MrMeanie (n=Geoff@82.152.254.168) left irc: Read error: 60 (Operation timed out)20:22
arigatoand if you mean adding a new functionality to the Python language, then no, definitely not20:22
ElectronicRUhttp://pastebin.com/f3ea53d7c20:23
ElectronicRUBetter than words.20:23
ElectronicRUThis 'll not break compatibility.20:23
arigatoyup, looks nice.  but as I said, no, not interested at all20:23
MrMeanie (n=Geoff@82.152.254.168) joined #pypy.20:24
arigatobut go to #python or python-dev and discuss it there if you feel like20:24
Alex_Gaynorarigato: that feature is already in py2.620:24
ElectronicRUarigato: I couldn't understand you.20:24
ElectronicRUI'm asking for adding *exisiting* functionality in PyPy.20:24
arigatoAlex_Gaynor: ah ha, thanks for the pointer20:24
ElectronicRUThis snippet is from ``What's New in Python 2.6'' document.20:25
arigatoElectronicRU: I see20:25
ElectronicRUSo may I?20:25
arigatoI don't know, we don't have a policy set up yet20:25
arigatothe point is that if we say we support Python 2.5, and you add some 2.6-only feature, then it's also annoying20:26
Alex_Gaynorso perhaps a branch for 2.6 support makes sense20:26
arigatoyes20:26
Alex_GaynorI suppose it would have to merge trunk in regularly to be helpful20:26
ElectronicRUOk, how do I make a branch if I have not whole repo checked out?20:27
arigatoyou can use  svn cp http://...  http://...20:27
ElectronicRUsvn+ssh://, I think.20:27
arigatoyes, depending on your preference20:28
ElectronicRUsvn cp svn+ssh://.../pypy/trunk svn+ssh://../pypy/branch/2.6ish -- so?20:28
arigatoyes20:28
jewel__ (n=jewel@vc-41-27-3-35.umts.vodacom.co.za) joined #pypy.20:29
jewel_ (n=jewel@vc-41-29-77-24.umts.vodacom.co.za) left irc: Connection timed out20:30
kenaan03electronicru 10r7027214 05A(pypy/branch/2.6ish/): Adding branch for Python 2.6 support. ...20:31
Cheshirc (n=tu@unaffiliated/cheshirc) left irc: Read error: 113 (No route to host)20:31
netto (n=netto@200.119.150.1) joined #pypy.20:32
Cheshirc (n=td@unaffiliated/cheshirc) joined #pypy.20:32
ElectronicRUJob done!20:34
arigato:-)20:37
lesshastewow that was quick :)20:37
ElectronicRUIt is checking out now. Svn is so smart that it just passed 'cp' to server, without transfering files.20:38
gutworth_ how would it transferr files?20:38
Nick change: gutworth_ -> gutworth20:38
ElectronicRUTo and from my computer :)20:39
ElectronicRUWhat a smart things get!..20:41
gutworthyou obviously haven't merged many svn branches20:41
ElectronicRUWhat a smart have things became!20:41
ElectronicRUgutworth: I know how svn works. I know that it'll never transfer files.20:42
ElectronicRUTruly.20:42
Alex_GaynorOTOH if you want to do an svn diff it hits the server *how many* times ;)20:42
gutworthdepends on whether it's svn+ssh or not20:42
kenaan03arigo 10r7027314 pypy/branch/jit-delayed-write/pypy/jit/metainterp 07M(effectinfo.py optimizeopt.py test_effectinfo.py test_optimizefindnode.py test_optimizeopt.py): Get optimizeopt.py to delay some writes.  All tests pass ...20:42
ElectronicRUOh, btw, what's the history of using svn for pypy?20:43
gutworthforever20:43
arigatoyes, we have been using it since forever20:44
ElectronicRULet's ask so: why is svn used for pypy?20:44
gutworthbecause it works well enough20:44
arigatoand because the other vcs systems did not work well enough when we started pypy20:44
gutworthand it's a billion times better than nothing20:44
arigatoor than cvs :-)20:45
gutworth:)20:45
ElectronicRUYes, than cvs - definitely! :)20:46
ElectronicRUBut I heard that Python have switched to bzr or somewhat, is it true?20:47
gutworthwe're in the agonizing process of switching to hg20:47
lesshasteis git suitable?20:48
gutworthlots of us hate git20:48
gutworthlike me :)20:48
lesshastegrumpy g.. :)20:48
gutworth(btw, by "we" I'm referring to cpython)20:48
Action: Ademan snarks20:49
Ademanmorning, merry christmas20:49
lesshastegutworth, by the way... did you see the failure of my first pypy test?20:49
ElectronicRUIt will be good test platform too :).20:49
gutworthbesides, how bad would it look if we passed up vcs written in python for a c one?20:49
gutworthlesshaste: I don't think so20:49
lesshastegutworth, http://pastebin.ca/1726863 is quite a lot slower in pypy than in normal python20:50
lesshasteabout 11/8 slower20:50
Ademanit's known to happen...20:50
gutworthhard to say what's slower with such a large example20:50
gutworthand 11/8 is about the usual20:50
lesshasteAdeman, yes but it's the only thing I have ever tested :)20:50
ElectronicRULet's write a new one -- in rpython&pypy! :)20:50
lesshastegutworth, do you mean it's a large example of code? 20:51
Ademansvn isn't written in python...20:51
ElectronicRUAdeman, hello.20:51
gutworthlesshaste: it involves external libraries etc20:51
ElectronicRUBut mercurial is.20:51
Ademanhey ElectronicRU 20:51
lesshastegutworth, ah yes there is one... let me find the source for the line that takes the time20:52
Ademanmy understanding is/was we're sticking with svn20:52
ElectronicRUAdeman, I have posted something about dot product in private...20:52
AdemanElectronicRU: yeah i saw that, good job with that20:53
ElectronicRUIt was two hours of my time and four paper sheets.20:53
Ademanthere's alot of celebrating going on i didn't get much done last night... hopefully this morning...20:54
Ademanalthough it's getting close20:54
Ademanand it supports 32 bit floats20:54
lesshastegutworth, well it's the add_edge line that takes all the time as far as I can tell.. actually I couldn't see how to profile the pypy run but that's what takes the time in normal python20:54
lesshastegutworth, do you know how to profile a pypy execution?20:54
gutworthyou'd have to see what the Graph.add_edge method does20:55
gutworththe profile module should work20:55
lesshastepypy-c -m profile test.py just returns an error message20:55
ElectronicRUOr, it's catholic Christmas today! Happy Christmas!20:56
gutworthgrumble, grubmle20:56
lesshaste-m cProfile  at least runs20:56
lesshaste:)20:56
Action: lesshaste is searching the networkx source code20:56
jewel__ (n=jewel@vc-41-27-3-35.umts.vodacom.co.za) left irc: Read error: 110 (Connection timed out)20:59
lesshastegutworth, http://www.pastebin.ca/172692820:59
lesshastehttp://www.pastebin.ca/1726928 is the whole graph.py file21:00
lesshastegrr http://pastebin.com/f2a334c43 I mean21:00
lesshastethe first one is just add_edge21:00
Nick change: ElectronicRU -> ElectronicRU[awa21:01
lesshastegutworth, http://pastebin.ca/1726930 is the generator for the graph21:02
lesshasteit's not much code in the end21:02
Cheshirc_ (n=td@unaffiliated/cheshirc) joined #pypy.21:04
Cheshirc (n=td@unaffiliated/cheshirc) left irc: Read error: 113 (No route to host)21:04
Ademanis @staticmethod rpython?21:05
gutworthyes21:06
Ademanawesome21:06
lesshastegutworth, is this just the wrong way to be reporting this?21:06
lesshasteshould I be posting to a group or somesuch?21:06
gutworthwell, a line or so which is much slower in pypy than cpython is better21:07
lesshasteyou mean a single line of python that isn't calling something?21:07
gutworthright21:08
gutworthor excerpt21:08
Ademantry and narrow it down to the offending code21:08
lesshasteAdeman, ok I did my best :) It's the add_edge function21:08
lesshastewhich I pasted above21:08
lesshasteit's only half a dozen lines21:09
gutworthwith at least 6 calls :)21:09
lesshastebut I see those might be annoying :)21:09
lesshastethis is why I am trying to profile!21:09
lesshasteany help much appreciated21:09
lesshaste When I do python -m profile ftnn.py I get http://www.pastebin.ca/1726936  any idea why it fails?21:15
Ademanlesshaste: i certainly don't21:24
lesshasteAdeman, ok thanks21:25
Ademangutworth might though21:25
lesshastegutworth is a good name of christmas gorging :)21:26
lesshastes/of/for21:26
lesshasteis profile preferable to cProfile in any case?21:27
Ademangutworth: how is the structure of an rpython class determined?21:28
Ademangutworth: I want to do something like this: http://paste.pocoo.org/show/159643/ is this rpython?21:30
Ademanand if i had a no-op construction helper, would it be optimized out? (or do you think i could at least trust GCC to optimize it out?)21:31
lesshasteit looks like a bug in the profiler to me21:33
lesshastewhere best to report such a thing?21:34
Ademanis it our implementation of the profiler?21:34
lesshasteno sorry21:36
lesshasteit seems to be a general python profile module failure21:36
lesshasteI found the bug tracker in any case21:36
gutworth (n=benjamin@64-131-1-181.usfamily.net) left irc: Read error: 60 (Operation timed out)21:37
lesshasteAdeman, well.. it works fine in python2.5 but not 2.621:39
lesshastethere's a thing21:39
Ademanlesshaste: interesting...21:39
lesshasteAdeman, pypy-c -m profile ftnn.py fails with a different error21:40
lesshasteso much fun21:40
lesshastehttp://pastebin.ca/1726961 is the pypy profile failure21:41
Ademanlesshaste: haha, alwas a pleasure...21:44
Ademanwow, that's an interpreter error...21:44
lesshasteAdeman, so it seems the first thing is a bug in python profile.. which was introduced in january 2008!! http://svn.python.org/view/python/trunk/Lib/pstats.py?r1=60149&r2=60148&pathrev=6014921:45
lesshasteI worry about python sometimes21:45
lesshasteall I did was try to profile one script!21:45
lesshasteAdeman, can I leave the interpreter error with you?  I should really go21:46
lesshaste (n=lesshast@87-194-206-189.bethere.co.uk) left irc: "Leaving"21:49
netto (n=netto@200.119.150.1) left irc: "Ex-Chat"21:52
Cheshirc_ (n=td@unaffiliated/cheshirc) left irc: Read error: 113 (No route to host)22:18
Cheshirc_ (n=td@unaffiliated/cheshirc) joined #pypy.22:18
kenaan03arigo 10r7027414 pypy/branch/jit-delayed-write/pypy/jit/metainterp 07M(optimizeopt.py test_optimizeopt.py): Step 1: force lazy fields on all guards. ...22:23
kenaan03arigo 10r7027514 pypy/branch/jit-delayed-write/pypy/jit/metainterp 07M(optimizeopt.py test_optimizeopt.py): force_all_lazy_setfields_of_nonvirtuals(). ...22:29
gonsor (n=quassel@p57A899D6.dip0.t-ipconnect.de) joined #pypy.22:46
Ademansorry lesshaste22:52
lac (n=lac@c-96c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Remote closed the connection22:53
victorpoluceno (n=victorgp@189.75.22.10) left irc: Read error: 60 (Operation timed out)23:04
victorpoluceno (n=victorgp@187.7.45.100) joined #pypy.23:04
victorpoluceno (n=victorgp@187.7.45.100) left irc: Read error: 60 (Operation timed out)23:12
nullpo (n=nullpo@221x252x46x83.ap221.ftth.ucom.ne.jp) joined #pypy.23:15
gutworth (n=benjamin@64-131-1-181.usfamily.net) joined #pypy.23:16
gutworthAdeman: um, well, FooType will be None in that case :)23:16
Cheshirc_ (n=td@unaffiliated/cheshirc) left irc: Read error: 113 (No route to host)23:20
Cheshirc__ (n=td@unaffiliated/cheshirc) joined #pypy.23:20
MrMeanie (n=Geoff@82.152.254.168) left irc: " HydraIRC -> http://www.hydrairc.com <- Nine out of ten l33t h4x0rz prefer it"23:44
arigato (n=arigo@adsl-188-155-200-92.adslplus.ch) left irc: "See you"23:55
--- Sat Dec 26 200900:00

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