| amaury_ | for example, someone had the good idea that PyWeakref_GetObject returns a borrowed reference | 00:00 |
|---|---|---|
| amaury_ | in pypy, this reference was *always* stale | 00:00 |
| basepi (~basepi@server.cmyers.net) left irc: Quit: ZNC - http://znc.sourceforge.net | 00:00 | |
| amaury_ | (refcount=0) | 00:00 |
| Alex_Gaynor | sigh | 00:01 |
| amaury_ | I added code so that the reference is kept alive for the duration of the C call | 00:01 |
| dittos (~ditto@203.236.100.15) joined #pypy. | 00:02 | |
| basepi (~basepi@174.127.114.26) joined #pypy. | 00:02 | |
| amaury_ | a bit like LocalReferences in JNI | 00:02 |
| espes (espes@2001:470:1f05:ca1::) joined #pypy. | 00:02 | |
| canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving. | 00:04 | |
| antocuni (~antocuni@host86-122-dynamic.6-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds | 00:05 | |
| Shinita (~John@osbk-4db06ce4.pool.mediaWays.net) joined #pypy. | 00:05 | |
| dittos (~ditto@203.236.100.15) left irc: Quit: bye | 00:08 | |
| mattip (chatzilla@bzq-109-64-118-235.red.bezeqint.net) left #pypy. | 00:08 | |
| tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 240 seconds | 00:15 | |
| tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy. | 00:15 | |
| amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 260 seconds | 00:17 | |
| Shinita (~John@osbk-4db06ce4.pool.mediaWays.net) left irc: Remote host closed the connection | 00:23 | |
| Moku (~John@osbk-4db06ce4.pool.mediaWays.net) joined #pypy. | 00:24 | |
| Nick change: Moku -> Guest91004 | 00:25 | |
| Nick change: Guest91004 -> Shinita | 00:25 | |
| derdon (~derdon@p5DE89701.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 00:28 | |
| jterrace (~jterrace@pu116187.student.Princeton.EDU) joined #pypy. | 00:29 | |
| tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 240 seconds | 00:35 | |
| Trundle (~andy@python/site-packages/trundle) left irc: Ping timeout: 240 seconds | 00:37 | |
| fzzzy (~donovan@nat/mozilla/x-zalwykhlbyeqkhii) left irc: Quit: fzzzy | 00:47 | |
| [mat^2] (~mathias@212.130.113.35) left irc: Ping timeout: 260 seconds | 00:49 | |
| mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep | 00:50 | |
| rekamso (~textual@67.51.82.66) left irc: Quit: bus | 00:53 | |
| aisipos (~aisipos@wc.lax.truecarcorp.com) left irc: Ping timeout: 260 seconds | 01:03 | |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy. | 01:06 | |
| tilgovi (~randall@couchdb/developer/tilgovi) left irc: Read error: Connection reset by peer | 01:10 | |
| Guest35710 (~jared@061093200075.ctinets.com) joined #pypy. | 01:11 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Read error: Connection reset by peer | 01:24 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 01:26 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Read error: Connection reset by peer | 01:26 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 01:28 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/712 | 01:28 |
| jterrace (jterrace@pu116187.student.Princeton.EDU) left #pypy. | 01:30 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/139 | 01:35 |
| aisipos (~aisipos@wc.lax.truecarcorp.com) joined #pypy. | 01:41 | |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1843 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/266 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1107 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/569 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/612 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/797 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/1439 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/703 | 01:44 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/264 | 01:44 |
| Guest35710 (~jared@061093200075.ctinets.com) left irc: Ping timeout: 256 seconds | 01:45 | |
| stakkars (~tismer@p5DDB5E98.dip.t-dialin.net) left irc: Ping timeout: 245 seconds | 01:52 | |
| whitelynx (~whitelynx@li117-47.members.linode.com) joined #pypy. | 02:07 | |
| Circlefusion (~cf@74-138-199-109.dhcp.insightbb.com) left irc: Read error: Connection reset by peer | 02:08 | |
| Circlefusion (~cf@74-138-199-109.dhcp.insightbb.com) joined #pypy. | 02:26 | |
| aat (~aat@c-67-188-9-79.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep. | 02:39 | |
| Shinita (~John@osbk-4db06ce4.pool.mediaWays.net) left irc: Ping timeout: 240 seconds | 02:48 | |
| Moku (~John@osbk-4d08b87e.pool.mediaWays.net) joined #pypy. | 02:48 | |
| Nick change: Moku -> Guest99702 | 02:48 | |
| Nick change: Gulaway -> Gulopine | 02:49 | |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) left irc: Remote host closed the connection | 02:59 | |
| AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) joined #pypy. | 03:01 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/941 | 03:08 |
| aat (~aat@c-67-188-9-79.hsd1.ca.comcast.net) joined #pypy. | 03:11 | |
| aisipos (~aisipos@wc.lax.truecarcorp.com) left irc: Ping timeout: 248 seconds | 03:12 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy. | 03:20 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Client Quit | 03:24 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy. | 03:33 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep | 03:40 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 256 seconds | 03:41 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 03:43 | |
| apanda (~apanda@c-24-7-101-62.hsd1.ca.comcast.net) joined #pypy. | 03:46 | |
| whitelynx (~whitelynx@li117-47.members.linode.com) left irc: Quit: Ex-Chat | 03:48 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds | 03:56 | |
| stakkars (~tismer@p5DDB7E68.dip.t-dialin.net) joined #pypy. | 04:07 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1843 | 04:08 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 04:11 | |
| jyeo (~voltorb@bb219-74-115-178.singnet.com.sg) joined #pypy. | 04:26 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy. | 04:28 | |
| AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) left irc: Quit: Leaving. | 04:34 | |
| aat (~aat@c-67-188-9-79.hsd1.ca.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/ | 04:36 | |
| jyeo (~voltorb@bb219-74-115-178.singnet.com.sg) left irc: Quit: Instantbird 1.1 | 04:36 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep | 04:41 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy. | 04:43 | |
| aurojit (~apanda@ucbvpn-209-125.VPN.Berkeley.EDU) joined #pypy. | 04:51 | |
| canta (~canta@77-20-123-240-dynip.superkabel.de) joined #pypy. | 04:51 | |
| apanda (~apanda@c-24-7-101-62.hsd1.ca.comcast.net) left irc: Ping timeout: 248 seconds | 04:54 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/797 | 05:06 |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/1439 | 05:06 |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/612 | 05:07 |
| kenaan | 12alex_gaynor jit-dynamic-getarrayitem 1125fd786beb8d 15/pypy/: merged default in | 05:10 |
| kenaan | 12alex_gaynor jit-dynamic-getarrayitem 11fc596321cc9b 15/pypy/jit/backend/model.py: make these not static | 05:10 |
| kenaan | 12alex_gaynor jit-dynamic-getarrayitem 114f3e2c9dda26 15/pypy/jit/backend/llgraph/: translation-ish fix, except it still breaks | 05:10 |
| aleksi (~aleksi@85.235.191.82) joined #pypy. | 05:15 | |
| apanda (~apanda@ucbvpn-209-161.VPN.Berkeley.EDU) joined #pypy. | 05:21 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1107 | 05:22 |
| aurojit (~apanda@ucbvpn-209-125.VPN.Berkeley.EDU) left irc: Ping timeout: 240 seconds | 05:22 | |
| Ademan (~dan@adsl-71-141-252-21.dsl.snfc21.pacbell.net) joined #pypy. | 06:03 | |
| Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 252 seconds | 06:06 | |
| Nick change: Gulopine -> Gulaway | 06:20 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Remote host closed the connection | 06:25 | |
| canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving. | 06:29 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/569 | 06:34 |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep | 06:39 | |
| Ademan (~dan@adsl-71-141-252-21.dsl.snfc21.pacbell.net) left irc: Quit: leaving | 06:42 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 06:47 | |
| amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy. | 06:58 | |
| hakanardo_ (~hakanardo@h59ec009e.seluldx.dyn.perspektivbredband.net) left irc: Ping timeout: 258 seconds | 06:59 | |
| amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Read error: Operation timed out | 07:03 | |
| kenaan | 12alex_gaynor default 117a416e643dc0 15/pypy/rpython/test/test_rtuple.py: a failing test | 07:05 |
| kenaan | 12alex_gaynor default 11cfb76a08edcb 15/pypy/module/cpyext/: merged upstream | 07:05 |
| kenaan | 12alex_gaynor default 11bf8bf8a5810e 15/pypy/rpython/test/test_rtuple.py: simplify failing test | 07:23 |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/703 | 07:32 |
| aurojit (~apanda@c-24-7-101-62.hsd1.ca.comcast.net) joined #pypy. | 07:39 | |
| apanda (~apanda@ucbvpn-209-161.VPN.Berkeley.EDU) left irc: Ping timeout: 260 seconds | 07:43 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/266 | 07:43 |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 07:51 | |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) left irc: Quit: fzzzy | 07:58 | |
| asmeurer (~asmeurer@dhcp-baca-230.resnet.nmt.edu) left irc: Quit: asmeurer | 07:59 | |
| Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) joined #pypy. | 08:03 | |
| Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) left irc: Changing host | 08:03 | |
| Trundle (~andy@python/site-packages/trundle) joined #pypy. | 08:03 | |
| Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy. | 08:06 | |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy. | 08:06 | |
| Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Read error: Connection reset by peer | 08:10 | |
| bivab (~david@fwstups.cs.uni-duesseldorf.de) joined #pypy. | 08:10 | |
| antocuni (~antocuni@host86-122-dynamic.6-79-r.retail.telecomitalia.it) joined #pypy. | 08:10 | |
| Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy. | 08:11 | |
| aurojit (~apanda@c-24-7-101-62.hsd1.ca.comcast.net) left irc: Remote host closed the connection | 08:31 | |
| tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy. | 08:38 | |
| ixokai (~ixokai@rrcs-24-199-8-246.west.biz.rr.com) joined #pypy. | 08:49 | |
| ixokai (~ixokai@rrcs-24-199-8-246.west.biz.rr.com) left irc: Changing host | 08:49 | |
| ixokai (~ixokai@pdpc/supporter/student/ixokai) joined #pypy. | 08:49 | |
| _habnabit (~habnabit@python/site-packages/habnabit) left irc: Quit: ZNC - http://znc.sourceforge.net | 08:54 | |
| _habnabit (~habnabit@2001:470:1f0e:bdd::2) joined #pypy. | 08:55 | |
| derdon (~derdon@pD9E1C71F.dip.t-dialin.net) joined #pypy. | 08:55 | |
| _habnabit (~habnabit@2001:470:1f0e:bdd::2) left irc: Changing host | 08:56 | |
| _habnabit (~habnabit@python/site-packages/habnabit) joined #pypy. | 08:56 | |
| EnCuKou (~encukou@a91-156-243-246.elisa-laajakaista.fi) left irc: Remote host closed the connection | 08:59 | |
| lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy. | 09:04 | |
| jperras (~jperras@li245-112.members.linode.com) left irc: Quit: ZNC - http://znc.sourceforge.net | 09:05 | |
| jperras (~jperras@li245-112.members.linode.com) joined #pypy. | 09:06 | |
| tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 240 seconds | 09:09 | |
| Squarepy (~Squarepy@wcw-staff-214-31.wireless.uva.nl) joined #pypy. | 09:10 | |
| Squarepy (~Squarepy@wcw-staff-214-31.wireless.uva.nl) left irc: Changing host | 09:11 | |
| Squarepy (~Squarepy@unaffiliated/squarepy) joined #pypy. | 09:11 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds | 09:16 | |
| antocuni_ (~antocuni@host200-125-dynamic.25-79-r.retail.telecomitalia.it) joined #pypy. | 09:17 | |
| kenaan | 12bivab ppc-jit-backend 11d1fa57a9cf80 15/pypy/jit/backend/ppc/ppcgen/: add a helper method to the register manager to allocate a scratch register | 09:19 |
| kenaan | 12bivab ppc-jit-backend 117a670ce597e7 15/pypy/jit/backend/ppc/ppcgen/: allocate and use a scratch register for get/set arrayitem in case the scale > 0 | 09:19 |
| antocuni (~antocuni@host86-122-dynamic.6-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds | 09:20 | |
| G2P (~G2P@fw-asn1.ornis.com) joined #pypy. | 09:23 | |
| Guest13217 (~marco@144.173.64.235) joined #pypy. | 09:24 | |
| Guest13217 | hi, I was trying to understand if it is possible to use my own swig-wrapped python extension with pypy, but from the docs I think it is not. Is this correct? | 09:25 |
| ojii (~ojii@40-34.60-188.cust.bluewin.ch) joined #pypy. | 09:33 | |
| hruske (~Gasper@2001:15c0:66e9:14:b08f:6cff:fe96:7699) joined #pypy. | 09:35 | |
| MostAwesomeDude | Guest13217: It might work, with a recompile. | 09:36 |
| Guest13217 | Gest13217, could you please point me to some documentation on how to do that? | 09:38 |
| Guest13217 | I actually compiled pypy from source | 09:38 |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) left irc: Quit: fzzzy | 09:38 | |
| MostAwesomeDude | Create a virtualenv with that pypy-c, then try to install your extension into that virtualenv the normal way. | 09:39 |
| MostAwesomeDude | It'll cause your C to be compiled against PyPy's headers. | 09:39 |
| Guest13217 | MostAwesomeDude, thanks, I'll check that | 09:39 |
| MostAwesomeDude | Gotta go; good luck. | 09:39 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 09:49 | |
| Squarepy (~Squarepy@unaffiliated/squarepy) left irc: Quit: Leaving | 10:09 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection | 10:24 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 10:25 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection | 10:25 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 10:26 | |
| lac (~quassel@2001:16d8:ffca:0:a288:b4ff:fe71:bcd0) joined #pypy. | 10:34 | |
| PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Quit: ZNC - http://znc.in | 11:07 | |
| PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy. | 11:10 | |
| amaury_ (~amaury_@74.125.57.34) joined #pypy. | 11:26 | |
| kenaan | 12hager ppc-jit-backend 112cc68464f249 15/pypy/jit/backend/ppc/ppcgen/ppc_assembler.py: Added regalloc_push and regalloc_pop => test_jump passes | 11:43 |
| kenaan | 12hager ppc-jit-backend 112a816abcb981 15/pypy/jit/backend/ppc/ppcgen/: merge | 11:43 |
| Guest35710 (~jared@061093200075.ctinets.com) joined #pypy. | 11:48 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds | 11:55 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 12:09 | |
| witulski (~stupsi@134.99.16.20) joined #pypy. | 12:32 | |
| witulski (stupsi@134.99.16.20) left #pypy. | 12:32 | |
| khs (~khs@128.41.202.84.customer.cdi.no) joined #pypy. | 12:41 | |
| amaury_ (~amaury_@74.125.57.34) left irc: Ping timeout: 244 seconds | 12:48 | |
| amaury_ (amaury_@nat/google/x-xkqrjkfungrgmmmt) joined #pypy. | 13:04 | |
| aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy. | 13:07 | |
| ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) joined #pypy. | 13:10 | |
| Rhy0lite (dje@nat/ibm/x-oiapmemjvuxksscd) joined #pypy. | 13:17 | |
| DanKluev (~root@81-086-adsl.vntc.ru) left irc: Ping timeout: 258 seconds | 13:19 | |
| fijal (~fijal@197.170.68.138) joined #pypy. | 13:24 | |
| Sho_ (~EHS1@kde/hein) joined #pypy. | 13:27 | |
| cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) joined #pypy. | 13:32 | |
| DanKluev (~root@83-116-adsl.vntc.ru) joined #pypy. | 13:34 | |
| hruske (~Gasper@2001:15c0:66e9:14:b08f:6cff:fe96:7699) left irc: Ping timeout: 240 seconds | 13:44 | |
| kenaan | 12hager ppc-jit-backend 11dd67675a28a9 15/pypy/jit/backend/ppc/ppcgen/: Started implementation of NEW, first test passes | 13:45 |
| hruske (~Gasper@2001:15c0:66e9:14:b08f:6cff:fe96:7699) joined #pypy. | 13:45 | |
| gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) joined #pypy. | 14:04 | |
| hruske (~Gasper@2001:15c0:66e9:14:b08f:6cff:fe96:7699) left irc: Quit: Leaving | 14:15 | |
| Nick change: Gulaway -> Gulopine | 14:23 | |
| fijal | Alex_Gaynor: ping | 14:37 |
| whitelynx (~whitelynx@li117-47.members.linode.com) joined #pypy. | 14:40 | |
| ojii_ (~ojii@40-34.60-188.cust.bluewin.ch) joined #pypy. | 14:42 | |
| ojii (~ojii@40-34.60-188.cust.bluewin.ch) left irc: Ping timeout: 258 seconds | 14:42 | |
| jell | exarkun: ok, i'll read this ;) | 14:47 |
| exarkun | jell: Oh, hi. :) | 14:48 |
| jell | exarkun: btw. if it's also in #3114, why i don't see any of this changes in trunk? | 14:49 |
| exarkun | jell: #3114 is still open | 14:49 |
| jell | ok | 14:49 |
| jell | probably my changes are too simple... | 14:49 |
| jell | and probably buggy as i see that now :/ | 14:50 |
| Nick change: Gulopine -> Gulaway | 14:51 | |
| Rhy0lite | bivab: ping | 14:55 |
| bivab | Rhy0lite: hi | 14:55 |
| Rhy0lite | bivab: hi | 14:55 |
| Rhy0lite | my other meeting was postponed to next week if you and Sven want to work now | 14:55 |
| bivab | sven is busy right now, maybe in about 30 minutes? | 14:57 |
| Rhy0lite | okay | 14:57 |
| Guest13217 (~marco@144.173.64.235) left irc: Remote host closed the connection | 15:12 | |
| aleksi (~aleksi@85.235.191.82) left irc: Remote host closed the connection | 15:17 | |
| arigato (~arigo@fwstups.cs.uni-duesseldorf.de) joined #pypy. | 15:22 | |
| cfbolz | arigato: hi! | 15:24 |
| cfbolz | arigato: got back safely? | 15:24 |
| arigato | hi | 15:28 |
| arigato | from next door | 15:28 |
| Alex_Gaynor | fijal: pong | 15:28 |
| arigato (~arigo@fwstups.cs.uni-duesseldorf.de) left irc: Quit: See you | 15:41 | |
| sven_hager (~sven@fwstups.cs.uni-duesseldorf.de) joined #pypy. | 15:45 | |
| bivab | Rhy0lite: sven_hager is back, should we meet in #pypy-ppc? | 15:47 |
| basepi (~basepi@174.127.114.26) left irc: Quit: ZNC - http://znc.sourceforge.net | 15:48 | |
| basepi (~basepi@server.cmyers.net) joined #pypy. | 15:49 | |
| Guest35710 (~jared@061093200075.ctinets.com) left irc: Ping timeout: 252 seconds | 15:51 | |
| Nick change: ojii_ -> ojii | 15:55 | |
| iori (~iori@110-133-45-54.rev.home.ne.jp) joined #pypy. | 15:57 | |
| AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) joined #pypy. | 16:06 | |
| fijal | Alex_Gaynor: what about the boston sprint? | 16:11 |
| AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) left irc: Client Quit | 16:11 | |
| Alex_Gaynor | fijal: what about it? | 16:11 |
| fijal | should we organize it? | 16:11 |
| Alex_Gaynor | Yes, but we should speak with someone in boston I guess :) | 16:11 |
| fijal | a good way would be to mail pypy-dev | 16:12 |
| exarkun | The Boston Python Meetup organization might be someone to coordinate that with. | 16:12 |
| antocuni_ | a boston sprint? When? | 16:12 |
| exarkun | They may be able to help with a venue | 16:12 |
| Alex_Gaynor | exarkun: and organizing I suspect, my all accounts Jessica McKellar (sp?) is good at that :) | 16:12 |
| exarkun | (They have some, though they typically use them for a few hours on evenings, so I don't know if they're suitable for multi-day sprint use) | 16:12 |
| exarkun | Indeed | 16:13 |
| Alex_Gaynor | fijal: in unrelated news, feel like taking a look at the failing test I pushed to rtuple? | 16:14 |
| Action: fijal ponders if he's able to organize his mind | 16:15 | |
| fijal | Alex_Gaynor: I'm after ~24h flying trip | 16:15 |
| Alex_Gaynor | fijal: no one else is around :) | 16:15 |
| fijal | not a very good excuse | 16:16 |
| gsnedders | So serializing the HTML5 spec using html5lib with etree is about half the speed of CPython& | 16:16 |
| fijal | but ok, maybe | 16:16 |
| fijal | gsnedders: good, submit a bug report | 16:16 |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection | 16:17 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 16:18 | |
| cfbolz | gsnedders: it's probably because pypy's etree is pure python, and cpython's is in C? | 16:19 |
| gsnedders | cfbolz: It's the serialization that's slower, and that's not in etree | 16:19 |
| Alex_Gaynor | cfbolz: feel like looking at a failing test I pushed to rtuple? | 16:20 |
| cfbolz | no, need to prepare my lecture for tomorrow | 16:20 |
| Alex_Gaynor | fun lecture? | 16:21 |
| cfbolz | it's really hard to come up with convincing short examples for multiple inheritance | 16:21 |
| cfbolz | Alex_Gaynor: if you know one, I can take a look :-) | 16:21 |
| Alex_Gaynor | cfbolz: if you allow for the case of mixins, there are some ok cases | 16:21 |
| cfbolz | yes, I did mixins. some nice examples for that count too | 16:22 |
| cfbolz | but I would also really like a useful diamond example | 16:22 |
| Alex_Gaynor | hmm, agood diamond example. | 16:22 |
| gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) left irc: Remote host closed the connection | 16:22 | |
| fijal | are there good diamond examples? | 16:23 |
| cfbolz | probably not | 16:23 |
| fijal | I am aware of a few bad ones | 16:23 |
| cfbolz | pairtype!!! | 16:23 |
| Alex_Gaynor | :D | 16:23 |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Ping timeout: 276 seconds | 16:23 | |
| Alex_Gaynor | cfbolz: FYI pairtype scares all the normal python programmers I showed it to | 16:23 |
| cfbolz | yes, I did that one year and lost half the students | 16:23 |
| cfbolz | fijal: which bad ones? | 16:24 |
| fijal | cfbolz: like exception hierarchy in Python's stdlib | 16:24 |
| cfbolz | ah | 16:24 |
| fijal | I think there is one that inherits from LookupError and something else | 16:24 |
| Alex_Gaynor | I don't think there's any multiple inheritance in exceptions, but maybe I'm wrong | 16:25 |
| Alex_Gaynor | Somehow the io module has no multiple inheritance. | 16:25 |
| cfbolz | anyway, I am taking compelling mixins too, any ideas? | 16:26 |
| chronitis (~chronos@88-104-113-187.dynamic.dsl.as9105.com) joined #pypy. | 16:26 | |
| Alex_Gaynor | cfbolz: django's class based views are built on mixins, to almost a scary degree | 16:26 |
| cfbolz | Alex_Gaynor: that's nothing I can program in a lecture | 16:26 |
| Alex_Gaynor | hmm, true | 16:26 |
| fijal | Alex_Gaynor: yes there is | 16:27 |
| fijal | >>> encodings.CodecRegistryError.__mro__ | 16:27 |
| fijal | (<class 'encodings.CodecRegistryError'>, <type 'exceptions.LookupError'>, <type 'exceptions.SystemError'>, <type 'exceptions.StandardError'>, <type 'exceptions.Exception'>, <type 'exceptions.BaseException'>, <type 'object'>) | 16:27 |
| fijal | >>> | 16:27 |
| cfbolz | Alex_Gaynor: how is that an argument? | 16:27 |
| cfbolz | fijal: meant you | 16:27 |
| Alex_Gaynor | fijal: why on earth does CodecRegistryError subclass SystemError | 16:27 |
| fijal | cfbolz: SystemError and LookupError have common superclass | 16:28 |
| fijal | but are not each others superclass | 16:28 |
| cfbolz | fijal: just saying you should have pasted this: | 16:28 |
| cfbolz | >>> encodings.CodecRegistryError.__bases__ | 16:28 |
| cfbolz | (<type 'exceptions.LookupError'>, <type 'exceptions.SystemError'>) | 16:28 |
| fijal | right | 16:28 |
| fijal | indeed, forgot about __bases__ | 16:29 |
| fijal | but I'm dead | 16:29 |
| fijal | "Please understand that static analysis and optimization are necessary to make complex web apps tractable to develop and efficient to load and run." | 16:30 |
| fijal | so they claim static analysis of javascript lets them optimize it | 16:30 |
| Alex_Gaynor | fijal: none of that is true | 16:30 |
| fijal | google+ folks | 16:30 |
| fijal | https://plus.google.com/u/0/115060278409766341143/posts/ViaVbBMpSVG | 16:30 |
| fijal | how can it be a lie? google said that | 16:30 |
| lucian | fijal: Closure does do dead code elimination and inlining | 16:31 |
| fijal | lucian: how does dead code elimination speed up anything? | 16:31 |
| Alex_Gaynor | you know what else does dead code elimination? not writing dead code, you know what does inlining? my compiler | 16:31 |
| fijal | and how does inlining change anything when you have a JIT? | 16:31 |
| lucian | fijal: network transfer, of course | 16:31 |
| fijal | yeah, not writing dead code sounds better though | 16:31 |
| lucian | fijal: and inlining is useful for things like IE | 16:31 |
| fijal | hm, ok, that might be valid argument | 16:32 |
| lucian | fijal: what about the library you're using? most of jQuery is dead code at any one time | 16:32 |
| fijal | but it's not quantified there | 16:32 |
| fijal | "which helps in corner cases" | 16:32 |
| fijal | lucian: how can you remove it? | 16:32 |
| Alex_Gaynor | how can you even *find* dead code in JS | 16:32 |
| lucian | Closure basically does the equivalent of static linking | 16:32 |
| lucian | Alex_Gaynor: it's a compiler | 16:32 |
| lucian | Alex_Gaynor: it can break things if you don't give it the right annotations | 16:33 |
| Alex_Gaynor | oh good, it can break things | 16:33 |
| durin42 | Alex_Gaynor: yeah, closure trusts the types you specify in special comment blocks | 16:33 |
| lucian | Alex_Gaynor: only if you're not careful | 16:33 |
| durin42 | Alex_Gaynor: and if you screw it up, things can break | 16:33 |
| durin42 | Alex_Gaynor: so the theory is you write good tests | 16:33 |
| Alex_Gaynor | so, only if you're perfect, it's the inverse type system | 16:33 |
| lucian | also, it does safe DCE by default | 16:33 |
| durin42 | That said, if you've got enough tests to trust your javascript codebase, you probably have enough tests that testing the closure-compiled js will be a good check of your type annotations. | 16:34 |
| lucian | so if you use just part of a library, just that part will be included, possibly inlined | 16:34 |
| durin42 | Right. | 16:34 |
| durin42 | I've heard that it's much more fun to work with than GWT. | 16:34 |
| durin42 | I have no direct experience with either too. | 16:35 |
| durin42 | s/too/tool/ | 16:35 |
| lucian | yes, GWT is just painful if you don't love Java | 16:35 |
| lucian | but Closure (library + compiler) also requires some Java love | 16:35 |
| durin42 | Huh? | 16:35 |
| lucian | type annotations, the library is very java-ish | 16:36 |
| durin42 | It requires java to run the jscompiler (IIRC), but that's all. | 16:36 |
| durin42 | Ah. | 16:36 |
| durin42 | That's not terribly surprising I guess. | 16:36 |
| lucian | Dart makes a lot of sense for google | 16:36 |
| lucian | types, java-ish | 16:36 |
| lucian | that's why it's so big, it inlines its stdlib and expects you to run Closure on it | 16:37 |
| lucian | google folks aren't entirely sane, methinks | 16:37 |
| durin42 | Dart honestly makes no sense to me. I'd rather just write javascript. | 16:38 |
| fijal | ok fine | 16:40 |
| fijal | but isn't that the real source of a problem? | 16:40 |
| fijal | now instead of one jquery I have bajillion small pieces of jquery | 16:40 |
| bivab (~david@fwstups.cs.uni-duesseldorf.de) left irc: Quit: bivab | 16:40 | |
| fijal | and each website has an incompatible one | 16:40 |
| fijal | and I can't cache it | 16:41 |
| fijal | no? | 16:41 |
| lucian | fijal: sure | 16:41 |
| lucian | but google don't care | 16:41 |
| durin42 | sure, but the reality is that everyone ships their own jquery anyway, right? | 16:41 |
| lucian | durin42: lots of people use the CDN ones | 16:41 |
| lac (~quassel@2001:16d8:ffca:0:a288:b4ff:fe71:bcd0) left irc: Read error: Connection reset by peer | 16:41 | |
| Alex_Gaynor | who does that? I just ship and cache, or CDN it. | 16:41 |
| durin42 | (I actually don't know if we do anything clever around reusing our CDN'd jqueries if it gets run through the closure meatgrinder.) | 16:42 |
| lucian | fijal: especially since few websites use Google Closure (the library) | 16:42 |
| Alex_Gaynor | people are almost starting to trust the CDNs, now that they haven't been down once in a long time | 16:42 |
| fijal | but once you load it you should be good | 16:42 |
| lucian | durin42: unless you use the "advanced" mode in closure, you're not | 16:42 |
| fijal | lucian: they apparently care because they wrote a blog post about infrastructure optimizations | 16:42 |
| lucian | fijal: they care about google plus being fast | 16:42 |
| fijal | and I can't quite wrap up my mind if they're doing something cool or just working around limitations | 16:42 |
| lucian | the second, mostly | 16:43 |
| Alex_Gaynor | fijal: limitations, JS sucks deeply | 16:43 |
| fijal | Alex_Gaynor: how's closure-compiled-to-js? | 16:43 |
| Alex_Gaynor | closure is just a JS lib, not a new language | 16:43 |
| lucian | fijal: Closure compiler or Closure library? | 16:43 |
| lucian | they're so fucking confusingly named | 16:43 |
| fijal | I think one of those is clojure | 16:43 |
| Alex_Gaynor | fijal: no, that's a lisp for the JVM | 16:44 |
| fijal | I suppose they love closure | 16:44 |
| lucian | the Closure compiler does minification of JavaScript, potentially using annotations | 16:44 |
| fijal | and | 16:44 |
| fijal | i thought clojure | 16:44 |
| Alex_Gaynor | fijal: no, closures are an abstraction over function | 16:44 |
| lucian | but it's funny, because ClojureScript uses the Google Closure Compiler | 16:44 |
| lucian | fijal: no, they don't use Clojure at all | 16:44 |
| Alex_Gaynor | ok I don't even know what the fuck that is | 16:44 |
| lucian | it'd be awesome if they did | 16:44 |
| lucian | Alex_Gaynor: ClojureScript? sounds straightforward to me. Clojure -> JavaScript compiler | 16:45 |
| fijal | lucian: ok, so this website: http://code.google.com/closure/ | 16:45 |
| fijal | clearly states that the resulting JS is faster | 16:45 |
| fijal | how? | 16:45 |
| lucian | fijal: it does do some optimisations sometimes | 16:45 |
| fijal | yes ok | 16:45 |
| lucian | like if( DEBUG ) | 16:45 |
| fijal | you said inlining and dead code removal | 16:45 |
| fijal | none of them speeds up the code | 16:45 |
| fijal | or inlining does, but the JIT does it anyway | 16:45 |
| lucian | it eliminates branches/loops based on the constantness of conditions | 16:46 |
| fijal | wouldn't jit do that as well? | 16:46 |
| gsnedders | Alex_Gaynor: You can find dead-code based upon type-profiling in JS. | 16:46 |
| fijal | even better? | 16:46 |
| lucian | fijal: sure, but not every browser has a JIT | 16:46 |
| Trundle (~andy@python/site-packages/trundle) left irc: Remote host closed the connection | 16:46 | |
| gsnedders | lucian: What browsers don't? | 16:46 |
| Alex_Gaynor | gsnedders: only if you know those types are stable in the actual run | 16:46 |
| lucian | gsnedders: IE | 16:47 |
| lucian | fijal: and most importantly, it's still smaller | 16:47 |
| Alex_Gaynor | maybe IE4, it's had one for several releases | 16:47 |
| gsnedders | lucian: IE9 does. | 16:47 |
| lucian | network trumps all | 16:47 |
| lucian | gsnedders: that's still too new | 16:47 |
| gsnedders | Alex_Gaynor: Only IE9 does. | 16:47 |
| fijal | I would stress that unqualified "high performance" means it's faster not on some obscure browsers | 16:47 |
| Alex_Gaynor | gsnedders: is IE10 not out? I've lost track. | 16:47 |
| lucian | gsnedders: they support IE7 and IE8 i believe | 16:47 |
| fijal | but pretty much everywhere | 16:47 |
| gsnedders | Alex_Gaynor: No, it isn't. | 16:47 |
| durin42 | IE6 is barely dead, much less 7 or 8. | 16:47 |
| fijal | anyway, why would I believe this website is not full of lies? | 16:47 |
| lucian | fijal: well, they're sort of lying about that | 16:47 |
| durin42 | And smartphones don't tend to have jit in their js interpreter | 16:47 |
| gsnedders | durin42: Fewer and fewer, though | 16:48 |
| gsnedders | durin42: The vast majority do now. | 16:48 |
| durin42 | plus, having less js to transmit and parse -> win | 16:48 |
| lucian | fijal: on the web smaller size = faster | 16:48 |
| durin42 | gsnedders: IIRC android doesn't until ICS | 16:48 |
| lucian | durin42: they sure do now | 16:48 |
| fijal | lucian: but not high-performance | 16:48 |
| gsnedders | durin42: Android always has. | 16:48 |
| lucian | durin42: android 2.3 has v8, before that it had javascriptcore's jit | 16:48 |
| gsnedders | durin42: Android has always used V8 which doesn't have an itnerpreter. | 16:48 |
| fijal | lucian: tell me you can read this website in a way that says "we make code smaller" | 16:48 |
| durin42 | Oh, then I'm thinking of iOS, which didn't until 4 or thereabouts | 16:48 |
| lucian | fijal: as i said "It runs quickly" means "It loads over the network, parses and then executes quickly" | 16:49 |
| lucian | gsnedders: actually no, just since 2.3 | 16:49 |
| fijal | this is not what the website says | 16:49 |
| lucian | gsnedders: or at least that's what my debugger says | 16:49 |
| gsnedders | lucian: `http://developer.android.com/sdk/android-2.2-highlights.html says 2.2 | 16:49 |
| fijal | The Closure Compiler compiles JavaScript into compact, high-performance code. | 16:49 |
| lucian | gsnedders: ah, ok | 16:49 |
| Alex_Gaynor | gsnedders: that's the dalvik JIT, isn't it? | 16:49 |
| Alex_Gaynor | not the JS one | 16:50 |
| gsnedders | Alex_Gaynor: Nope. | 16:50 |
| lucian | fijal: yeah. high-performance, which means it loads fast | 16:50 |
| fijal | no it does not | 16:50 |
| gsnedders | Alex_Gaynor: V8 and Dalvik landed in the same release. | 16:50 |
| lucian | Alex_Gaynor: it's shipped the webkit jit for a while | 16:50 |
| Alex_Gaynor | gsnedders: the only place I see JIT on that page specifically refers to dalvik | 16:50 |
| fijal | either it is deliberately lying or I have a very skewed vision | 16:50 |
| sven_hager (~sven@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving | 16:50 | |
| Alex_Gaynor | gsnedders: ah I see it says v8 there, but didn't you say it was just javascriptcore before that? | 16:50 |
| lucian | fijal: in the context of web development, it's only slightly lying | 16:51 |
| gsnedders | Alex_Gaynor: It was JSC, dunno if they used the JIT or not. | 16:51 |
| lucian | it also unrolls loops and stuff like that | 16:51 |
| lucian | anyway, it's only rarely useful | 16:51 |
| gsnedders | Most of what Closure does is becoming less and less important with optimizing compilers. | 16:51 |
| lucian | like for languages with no browser-side library (ClojureScript) | 16:51 |
| gsnedders | When we were still dealing with interpreters normally, it was fairly important. | 16:51 |
| Action: gsnedders doesn't acually know how far back JSC's ARM backend goes | 16:52 | |
| lucian | gsnedders: it's pretty good. iOS uses it | 16:52 |
| fijal | Alex_Gaynor: google names this shit | 16:52 |
| jterrace (~jterrace@aegis.CS.Princeton.EDU) joined #pypy. | 16:52 | |
| fijal | dart go | 16:52 |
| lucian | and it's used it for a few versions, i think | 16:52 |
| Alex_Gaynor | jterrace: you're fast :) | 16:52 |
| jterrace | hehe | 16:52 |
| lucian | fijal: except one's crap and the other's ok :) | 16:53 |
| gsnedders | lucian: Oh, yes, it's pretty good. Just don't know how long it's been around. | 16:53 |
| fijal | lucian: / | 16:53 |
| fijal | ? | 16:53 |
| lucian | gsnedders: i see | 16:53 |
| Action: lucian goes home. bye! | 16:53 | |
| jterrace | Alex_Gaynor: so when pycollada passes all tests in pypy, i will do a dance | 16:53 |
| gsnedders | June 2009 would appear to be the answer | 16:53 |
| fijal | jterrace: what's missing from numpy? | 16:54 |
| Alex_Gaynor | jterrace: so the float32 ones will be fixed on a branch I have, the readonly-ness of shape is fair game to file a bug (talk to fijal about tha tone I guess), and I guess file a bug for string as well | 16:54 |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy. | 16:54 | |
| jterrace | ok | 16:54 |
| Alex_Gaynor | fijal: numpy.float32, writing to the shape attr, and numpy.string_ | 16:54 |
| jterrace | does .reshape work instead of setting .shape directly? | 16:54 |
| fijal | jterrace: what happens if you change shape? | 16:54 |
| fijal | it reshapes the array? | 16:54 |
| Alex_Gaynor | jterrace: no, we don't have multidim on trunk anyways | 16:54 |
| fijal | I don't think there is reshape | 16:54 |
| jterrace | fijal: http://pastebin.com/zm2CJZDq | 16:54 |
| jterrace | oh ok | 16:54 |
| Alex_Gaynor | fijal: reshape with single dim is kind of boring :) | 16:55 |
| fijal | jterrace: we need multidim arrays first | 16:55 |
| jterrace | yeah | 16:55 |
| fijal | this is being-developed | 16:55 |
| jterrace | awesome | 16:55 |
| fijal | with a pace of a moderate walk if you ask me ;-) | 16:55 |
| Alex_Gaynor | jterrace: we have 2 big open branches ATM, one fixes a ton of dtype stuff and adds numpy.{float32, etc.} the other adds multidim | 16:56 |
| gsnedders | Alex_Gaynor: FWIW, you can DCE quite easily provided you have no objects (i.e., only boolean primitives, undefined, null, number primitives, and string primitives) | 16:56 |
| jterrace | great | 16:56 |
| cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving | 16:56 | |
| Alex_Gaynor | jterrace: the dtype one will be merged as soon as I finish yak shaving | 16:56 |
| fijal | same for the multidim one | 16:57 |
| fijal | there are two features missing and some yaks | 16:57 |
| jterrace | pycollada might be a nice testing case since it uses only a small subset of numpy functionality | 16:57 |
| kenaan | 12hager ppc-jit-backend 119b6289cbf90b 15/pypy/jit/backend/ppc/ppcgen/ppc_assembler.py: Use r0 as one-element stack | 16:57 |
| jterrace | if you need it | 16:57 |
| jterrace | but it uses multidim too | 16:57 |
| Alex_Gaynor | fijal: pff, your yaks don't involve needing to put optimizations in x86 backend or fixing bugs in rtuple | 16:58 |
| lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 240 seconds | 16:58 | |
| Alex_Gaynor | fijal: any opinions about what level my hash bug should be fixed at? | 17:00 |
| Alex_Gaynor | fijal: Does http://paste.pocoo.org/show/508250/ look ok, or wrong level? | 17:04 |
| asmeurer__ (~asmeurer@dhcp-baca-230.resnet.nmt.edu) joined #pypy. | 17:04 | |
| fijal | so we can hash an empty string? | 17:07 |
| fijal | would be cool if it comes with a test | 17:07 |
| Alex_Gaynor | fijal: a NULL string, yes | 17:07 |
| Alex_Gaynor | fijal: i checked in the test | 17:07 |
| Alex_Gaynor | fijal: so this is fixing it at an ok level? | 17:09 |
| fijal | I would think so | 17:09 |
| Alex_Gaynor | ok thanks | 17:09 |
| fijal | I'm a bit worried though about reading the hash field | 17:10 |
| fijal | do we have some shortcuts? | 17:10 |
| kenaan | 12alex_gaynor default 1189c328e7b0fa 15/pypy/rpython/: handle hashing a None rstr, fixes the test I checked in yesterday. | 17:10 |
| fijal | or it's gonna be guard_nonnull simply? | 17:10 |
| fijal | I think so | 17:10 |
| G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving. | 17:18 | |
| ojii (~ojii@40-34.60-188.cust.bluewin.ch) left irc: Remote host closed the connection | 17:18 | |
| khs (~khs@128.41.202.84.customer.cdi.no) left irc: Ping timeout: 248 seconds | 17:18 | |
| khs (~khs@128.41.202.84.customer.cdi.no) joined #pypy. | 17:20 | |
| fijal | Alex_Gaynor: another fun fact. You can't inherit from all combinations of exceptions | 17:20 |
| fijal | they have different C-level layouts | 17:20 |
| Alex_Gaynor | fijal: inconsistant base class stuff? | 17:20 |
| fijal | yeah | 17:20 |
| fijal | like OSError and LookupError | 17:21 |
| fijal | took us a while when we rewrote exceptions to interp-level | 17:21 |
| Alex_Gaynor | nice | 17:21 |
| fijal | no, this yuo can | 17:21 |
| fijal | but there are combinations you can't | 17:21 |
| Alex_Gaynor | LookupError just uses .args, IIRC | 17:21 |
| Alex_Gaynor | exceptions really need to have more semantic info on them | 17:21 |
| fijal | hm | 17:23 |
| aleksi (~aleksi@176.57.74.133) joined #pypy. | 17:23 | |
| fijal | I can't find a combination that does not work | 17:23 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds | 17:25 | |
| fijal (~fijal@197.170.68.138) left irc: Read error: Connection reset by peer | 17:26 | |
| kenaan | 12alex_gaynor jit-dynamic-getarrayitem 119733b32ba514 15/pypy/: merged default in | 17:28 |
| kenaan | 12alex_gaynor jit-dynamic-getarrayitem 11ee7c71b7e412 15/pypy/jit/: put tests in a sane subclass and only run the new ones in the x86 backend | 17:28 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/570 [12alex, jit-dynamic-getarrayitem] | 17:29 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/704 [12alex, jit-dynamic-getarrayitem] | 17:29 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1844 [12alex, jit-dynamic-getarrayitem] | 17:29 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1108 [12alex, jit-dynamic-getarrayitem] | 17:29 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 17:29 | |
| fijal (~fijal@197.168.232.138) joined #pypy. | 17:30 | |
| Alex_Gaynor | fijal: welcome back | 17:32 |
| fijal | I think that links to launchpad bugs in syslog are a bit of an exagerration | 17:32 |
| amaury_ | fijal: class C(UnicodeDecodeError, OSError): pass | 17:33 |
| amaury_ | does not work | 17:33 |
| fijal | amaury_: oh, ok | 17:33 |
| Alex_Gaynor | amaury_: whoever does this is crazy | 17:37 |
| amaury_ | yes, it's not necessary | 17:38 |
| amaury_ | C = (UnicodeDecodeError, OSError) has the same effect | 17:38 |
| amaury_ | at least in "except C" statements | 17:39 |
| Alex_Gaynor | anyone creating subclasses for except C statements is even crazier | 17:39 |
| amaury_ | with ABC there's probably a way to do it | 17:40 |
| amaury_ | __subclasscheck__ | 17:40 |
| etrepum (~bob@75-101-96-144.dsl.static.sonic.net) left irc: Quit: etrepum | 17:41 | |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) joined #pypy. | 17:42 | |
| fijal | pfff | 17:43 |
| fijal | Alex_Gaynor: do you remember which one is "C" order of storing matrices? | 17:43 |
| fijal | if I have [[1, 2], [3, 4]] | 17:43 |
| fijal | this would be 1 2 3 4 | 17:43 |
| fijal | or 1 3 2 4 | 17:43 |
| fijal | ? | 17:43 |
| Alex_Gaynor | 1 2 3 4 I think | 17:43 |
| Alex_Gaynor | because you'd have int[][] | 17:43 |
| Alex_Gaynor | so X[dim1] is a pointer to {1, 2} | 17:44 |
| fijal | then I think I got shard ordering wrong again -) | 17:45 |
| fijal | :-) | 17:45 |
| Action: Alex_Gaynor smacks fijal with a trout | 17:45 | |
| fijal | the obvious gain is that you can iterate one by one and be good | 17:45 |
| fijal | but it differs how do you initialize the thing | 17:46 |
| mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy. | 17:46 | |
| fijal | anyway, I guess I give up for today then | 17:46 |
| Alex_Gaynor | eventually we have to support both C and Fortran ordering I think | 17:46 |
| fijal | yes | 17:46 |
| kenaan | 12hager ppc-jit-backend 11b2ad6b915f48 15/pypy/jit/backend/ppc/ppcgen/: Implemented NEW_ARRAY | 17:46 |
| fijal | but since I changed it twice already I know how hard it would be right? | 17:46 |
| Alex_Gaynor | fijal: I'm sure there's other nonsense to getting it right | 17:47 |
| Alex_Gaynor | like doing broadcasting between arrays of different types | 17:47 |
| fschwidom (~fschwidom@46.115.26.122) joined #pypy. | 17:47 | |
| fijal | right | 17:49 |
| fijal | it's stupidly hard enough to get right | 17:50 |
| fijal | but now I know what to do | 17:50 |
| Alex_Gaynor | fijal: uhhh, why is one translation on tannit using 9GB of RAM? | 17:50 |
| aurojit (~apanda@dhcp-45-162.EECS.Berkeley.EDU) joined #pypy. | 17:51 | |
| Nick change: ronan_ -> ronan | 17:51 | |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) left irc: Read error: Connection reset by peer | 17:53 | |
| fijal (~fijal@197.168.232.138) left irc: Read error: Connection reset by peer | 17:55 | |
| kenaan | 12hager ppc-jit-backend 11991e2aba52b0 15/pypy/jit/backend/ppc/ppcgen/: Implemented NEWSTR and NEWUNICODE | 17:56 |
| gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) joined #pypy. | 17:58 | |
| mat^2 (~mathias@212.130.113.35) joined #pypy. | 18:03 | |
| bbot2 | 13Exception: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/570 [12alex, jit-dynamic-getarrayitem] | 18:04 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/571 [12alex, jit-dynamic-getarrayitem] | 18:08 |
| kenaan | 12edelsohn ppc-jit-backend 11760676c29f43 15/pypy/jit/backend/ppc/ppcgen/regalloc.py: Correct indentation of prepare_setarrayitem_raw. | 18:09 |
| etrepum (~bob@accessnat4.mochimedia.net) joined #pypy. | 18:10 | |
| Alex_Gaynor | Rhy0lite: out of curiosity, how do you find working on pypy compared to working on GCC? | 18:11 |
| Rhy0lite | PyPy reminds me of GCC 10-15 years ago | 18:11 |
| Alex_Gaynor | in what sense? | 18:11 |
| Rhy0lite | fun, bright, motivated developers working on exciting project | 18:12 |
| antocuni_ (~antocuni@host200-125-dynamic.25-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds | 18:13 | |
| Rhy0lite | Alex_Gaynor: PyPy has a lot of open, exciting projects to work on | 18:14 |
| Rhy0lite | and JITs and dynamic languages, like Python, are an open field | 18:14 |
| Alex_Gaynor | Rhy0lite: also read as: "Hairy yaks in strange places" :) | 18:15 |
| Rhy0lite | And you are an excellent yak hair stylist! | 18:15 |
| Alex_Gaynor | When I was a kid I never said that was what I wanted to do on career day. | 18:16 |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) left irc: Quit: fzzzy | 18:16 | |
| Rhy0lite | No one ever expects the Spanish Inquisition! | 18:16 |
| amaury_ (amaury_@nat/google/x-xkqrjkfungrgmmmt) left irc: Ping timeout: 244 seconds | 18:17 | |
| Alex_Gaynor | PyPy: The Running of the Hairy Yaks. | 18:17 |
| mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) left irc: Ping timeout: 260 seconds | 18:19 | |
| mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) joined #pypy. | 18:21 | |
| Rhy0lite | Does PyPy 1.7 branch translate or any builds from buildbot? | 18:24 |
| fzzzy (~donovan@nat/mozilla/x-qgfgkhbfdwcjflsd) joined #pypy. | 18:32 | |
| PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Ping timeout: 240 seconds | 18:32 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1108 [12alex, jit-dynamic-getarrayitem] | 18:38 |
| PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy. | 18:39 | |
| syntaxiko (~RosesEatP@cpe-67-252-73-238.stny.res.rr.com) joined #pypy. | 18:40 | |
| aleksi (~aleksi@176.57.74.133) left irc: Read error: Connection reset by peer | 18:41 | |
| kenaan | 12pjenvey py3k 11706419ee2d49 15/pypy/objspace/std/: backout 21b2914fdb96 pending type.name switching to unicode | 18:53 |
| aisipos (~aisipos@wc.lax.truecarcorp.com) joined #pypy. | 18:54 | |
| hruske (~Gasper@188-230-156-183.dynamic.t-2.net) joined #pypy. | 19:00 | |
| PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Ping timeout: 240 seconds | 19:02 | |
| PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy. | 19:03 | |
| aurojit (~apanda@dhcp-45-162.EECS.Berkeley.EDU) left irc: Remote host closed the connection | 19:12 | |
| nickik (~nick@217.150.239.67) joined #pypy. | 19:19 | |
| aleksi (~aleksi@176.57.74.133) joined #pypy. | 19:19 | |
| mattip (~chatzilla@109.64.118.235) joined #pypy. | 19:24 | |
| mattip | fijal: ping | 19:26 |
| Alex_Gaynor | ronny: ping | 19:27 |
| bbot2 | 13Exception: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/571 [12alex, jit-dynamic-getarrayitem] | 19:27 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/572 [12alex] | 19:28 |
| ronny | Alex_Gaynor: pong? | 19:31 |
| Alex_Gaynor | ronny: I was going to ask you if you could ssh into tannit and cheeck the memory use on that build, but I remembered buildbot has a timer and saw it was runnign way longer than it shoudl have been. Sorry about that. | 19:32 |
| ronny | k, np | 19:32 |
| mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep | 19:39 | |
| khs | Is it possible to import modules in RPython using __import__(...) like in Python? | 19:39 |
| exarkun | RPython is defined using normal Python. | 19:40 |
| exarkun | You can do anything that works in Python to construct the RPython. | 19:41 |
| exarkun | I don't know, but I would not expect __import__ to be RPython itself. | 19:41 |
| khs | Not for constructing RPython, but to dynamically import modules necessary during runtime | 19:42 |
| khs | ok | 19:42 |
| rguillebert | I think you can use it outside of functions but not inside | 19:42 |
| rguillebert | (which is just was exarkun said...) | 19:43 |
| EnCuKou (~encukou@a91-156-243-246.elisa-laajakaista.fi) joined #pypy. | 19:47 | |
| kenaan | 12hakanardo jit-targets 11520196bc50f9 15/pypy/jit/metainterp/compile.py: translation fix | 20:05 |
| Rhy0lite (dje@nat/ibm/x-oiapmemjvuxksscd) left irc: Quit: Leaving | 20:12 | |
| tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy. | 20:12 | |
| kkris (~kris@80-123-38-92.adsl.highway.telekom.at) joined #pypy. | 20:12 | |
| mattip (~chatzilla@109.64.118.235) left irc: Max SendQ exceeded | 20:16 | |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) joined #pypy. | 20:17 | |
| zain (~textual@c-67-160-201-63.hsd1.ca.comcast.net) left irc: Disconnected by services | 20:17 | |
| Nick change: zain2 -> zain | 20:17 | |
| zain_ (~textual@c-67-160-201-63.hsd1.ca.comcast.net) joined #pypy. | 20:18 | |
| [mat^2] (~mathias@212.130.113.35) joined #pypy. | 20:18 | |
| mattip (~chatzilla@109.64.118.235) joined #pypy. | 20:19 | |
| mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 240 seconds | 20:19 | |
| mattip (chatzilla@109.64.118.235) left #pypy. | 20:22 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/704 [12alex, jit-dynamic-getarrayitem] | 20:26 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/705 [12hakanardo, jit-targets] | 20:26 |
| aleksi (~aleksi@176.57.74.133) left irc: Remote host closed the connection | 20:38 | |
| djfroofy (~djfroofy@66-195-30-142.static.twtelecom.net) joined #pypy. | 20:47 | |
| djfroofy | some told me that with an interpretted language "it impossible to develop an app that can compare with a | 20:48 |
| djfroofy | native application" | 20:48 |
| djfroofy | didn't pypy have a one or two benchmarks which actually beat compiled c? | 20:49 |
| mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy. | 20:51 | |
| rguillebert | yes, there's one on the blog | 20:52 |
| rguillebert | about string formatting | 20:52 |
| rguillebert | djfroofy, when you say "native application" it doesn't include Java, does it ? | 20:53 |
| djfroofy | rguillebert: sort of. an internal flamewar thread at my office where our resident x86 expert is saying we should standaradize on one of native client, java or silverlight instead of html5 for our app development targeting web devices | 20:55 |
| djfroofy | so he's nearly grouping java and silverlight with "native application" | 20:56 |
| htoothrot | uhhh | 20:56 |
| rguillebert | so you're talking about statically typed languages rather than native applications | 20:56 |
| djfroofy | rguillebert: sorry i mispoke, this is the whole claim with one redaction http://paste.pocoo.org/show/508374/ | 20:58 |
| ronny | djfroofy: silverliht is dead, java fucks up stuff for the users and naive client is a mess of its own | 20:58 |
| MostAwesomeDude | djfroofy: Yeah, I wouldn't trust any of those. | 20:58 |
| rguillebert | "HTML5 depends on a scripting language" ?!? | 20:58 |
| MostAwesomeDude | HTML5 + JS is probably the best bet, but ugh. | 20:59 |
| Rhyolite | Dart is the future! | 20:59 |
| Rhyolite | ;-) | 20:59 |
| MostAwesomeDude | Pfft. | 20:59 |
| MostAwesomeDude | Did you know that Fx has Python support? They just don't ship it. I think they should. | 20:59 |
| ronny | dart is something you dont want in your eyes | 20:59 |
| djfroofy | yeah, the guy is a really smart dude. he's reversed a lot of hardware, wrote emulators for several game systems. but i think he's confused. and he's confusing my ceo now. | 20:59 |
| rguillebert | the first question is : do you need speed :) | 21:00 |
| djfroofy | in some instances yes: games | 21:01 |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) left irc: Remote host closed the connection | 21:01 | |
| rguillebert | well, games are a specific topic | 21:01 |
| rguillebert | you can really do games on pypy right now | 21:01 |
| rguillebert | because of the garbage collector | 21:01 |
| djfroofy | rguillebert: yes, i'm definitely not arguing we need to port assasin's creed (not my company's game for the record) to html5 | 21:02 |
| djfroofy | rguillebert: i know also eve most of eve online's client code is even python (albeit stackless, not pypy) | 21:02 |
| ronny | djfroofy: whats the complexity of the games? webgl datastructures + jit can be pretty close to c speed whise if used correct | 21:03 |
| rguillebert | djfroofy, afaik it's mostly the UI that is done in python | 21:03 |
| djfroofy | ronny: well here's kind of the funny catch. the only games we do in house are on par with "angry birds" in terms of complexity | 21:03 |
| rguillebert | ... | 21:04 |
| rguillebert | angry birds isn't written with silverlight | 21:05 |
| rguillebert | and it's doing pretty well performance wise | 21:05 |
| ronny | djfroofy: then anything will do | 21:05 |
| rguillebert | also "They all use virtual machines that don't model the real CPU at all" | 21:05 |
| rguillebert | but jitted code do :) | 21:06 |
| rguillebert | unless assembly doesn't model the real cpu | 21:06 |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) joined #pypy. | 21:07 | |
| kkris (~kris@80-123-38-92.adsl.highway.telekom.at) left irc: Quit: Leaving. | 21:07 | |
| timonator | well, to be fair, there is a lot of stuff the cpu does to assembly code that gets executed | 21:07 |
| rguillebert | but I think we're stuck with assembly :) | 21:08 |
| djfroofy | the other interesting argument and he and others are trying to make is that compiled code is better because the code is drm and your get better security through obscurity through compiled native code | 21:12 |
| rguillebert | djfroofy, I think you're facing a "real programmers write C code" kind of guy | 21:12 |
| djfroofy | rguillebert: oh yes | 21:12 |
| ronny | there are less and less reasons to take the c for the performance | 21:13 |
| rguillebert | maybe you should abandon the performance talk and speak about development cost, that kind of stuff | 21:13 |
| djfroofy | this is the guy i'm talking about: http://memegenerator.net/instance/11460195 | 21:14 |
| rguillebert | but the argument you've shown us is just FUD | 21:15 |
| djfroofy | rguillebert: yeah those are the points i'm trying to harp on. but he believes in that regard should be sticking to flash for making our apps b/c "RAD tools are better" | 21:16 |
| chronitis (~chronos@88-104-113-187.dynamic.dsl.as9105.com) left irc: Quit: Lost terminal | 21:16 | |
| exarkun | What did he say when Adobe announced that flash is over? | 21:16 |
| rguillebert | also RAD stuff are usually not very maintainable | 21:17 |
| rguillebert | (I've just realized that I'm doing FUD as well) | 21:17 |
| djfroofy | exarkun: yeah i pointed to that announcement about flash's view on the future and he hasn't responded yet. did have some things to say about apple: http://paste.pocoo.org/show/508381/ | 21:19 |
| djfroofy | s/flash's view/adobe's view/ | 21:19 |
| fschwidom (~fschwidom@46.115.26.122) left irc: Ping timeout: 252 seconds | 21:19 | |
| rguillebert | Flash and HTML5 are designed for desktop interfaces | 21:20 |
| djfroofy | rguillebert: his statement was absurd. like there's double-blind study that shows you can crank out better apps faster with rad tools than other alternatives. and again it could depend largely on the domain. | 21:20 |
| djfroofy | "It does NOT make building apps easier. Yes, it's a DOM, we have | 21:20 |
| djfroofy | millions of those, and Flash has a DOM and RAD tools that make Flash | 21:20 |
| djfroofy | developers consistantly outperform (in development time) HTML5 | 21:20 |
| djfroofy | developers" | 21:21 |
| rguillebert | Every new Apple mobile device and every new Mac — along with the latest version of Apple’s Safari web browser — supports web standards including HTML5, CSS3, and JavaScript. These web standards are open, reliable, highly secure, and efficient. They allow web designers and developers to create advanced graphics, typography, animations, and transitions. Standards aren’t add-ons to the web. They are the web. And you can st | 21:21 |
| rguillebert | art using them today. | 21:21 |
| rguillebert | apple.com/html5 | 21:21 |
| djfroofy | apologies for ranting off-topic, btw, here guys and distracting you from making python faster ;) | 21:22 |
| rguillebert | it's only half off-topic, "dynamically typed languages are slow' is what PyPy is fighting against | 21:23 |
| djfroofy | rguillebert: yeah, i figured there could be some sympatico there. | 21:23 |
| rguillebert | also | 21:25 |
| rguillebert | there's a port of quake2 in html5 | 21:25 |
| djfroofy | unrelated but more on-topic, has anyone done much serious with audio and pypy? googling leads me only to a few sketchy, half-cocked things. | 21:26 |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1844 [12alex, jit-dynamic-getarrayitem] | 21:26 |
| rguillebert | there has been video stuff but I don't know about audio | 21:26 |
| Ademan (~dan@197.sub-97-156-104.myvzw.com) joined #pypy. | 21:27 | |
| djfroofy | rguillebert: thanks, i'm hoping i don't have to roll up my sleeves and port portaudio to ctypes. | 21:28 |
| rguillebert | you can try cpyext | 21:29 |
| djfroofy | rguillebert: also do you think it would be a better strategy (and perf is key here) to do such a thing in rpython or just use ctypes? | 21:29 |
| rguillebert | ctypes | 21:29 |
| rguillebert | there's nice optimization for ctypes code | 21:29 |
| rguillebert | and you can really write rpython to make libraries | 21:30 |
| rguillebert | because you have to translated the whole interpreter each time you want to change something | 21:30 |
| djfroofy | so even in audio where the slightest crackle or pop in real time audio processing would be fail? (i know this is a vaguely stated problem) | 21:30 |
| khs (~khs@128.41.202.84.customer.cdi.no) left irc: Quit: Leaving | 21:31 | |
| rguillebert | I don't have enough experience in real-time computing to really help you | 21:32 |
| rguillebert | but as long as there's no concurrent garbage collector in PyPy you can't do real-time things | 21:32 |
| rguillebert | because the garbage collector can pause the entire application for quite some time | 21:33 |
| jterrace (jterrace@aegis.CS.Princeton.EDU) left #pypy. | 21:33 | |
| amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy. | 21:34 | |
| djfroofy | rguillebert: hmmm ... yeah i guess then it's just a question of whether the pause is long enough to cause cause buffer starvation | 21:34 |
| djfroofy | "quite some time" actually kind of scares me | 21:35 |
| exarkun | "Real time" audio is pretty easy | 21:35 |
| exarkun | I haven't tried anything on PyPy, but considering how easy it is even on CPython I'd be surprised if PyPy can't manage. | 21:36 |
| djfroofy | with just cpython i'm actually very good performance without any gc pause. albeit my trick is trying to make sure the garbage collector doesn't run: allocating objects before the start of a song and holding onto references | 21:37 |
| rguillebert | the edge-detection that has been presented at EuroPython was really good | 21:38 |
| exarkun | I don't remember any particular GC tricks in Shtoom | 21:38 |
| djfroofy | but this is for a pretty pretty course granularity: playing fluidsynth instruments at bpm < 300 | 21:38 |
| exarkun | A softphone only needs to play about 50 samples per second, though. | 21:39 |
| djfroofy | exarkun: i guess i should look at shtoom. things got dismal when i introduced pyaudio and tried to do real time effects over sound. | 21:39 |
| djfroofy | by "real time" i mean 44100 or whatever | 21:39 |
| exarkun | 44100 samples? :) Or 3 samples per second, containing 44100Hz audio? | 21:39 |
| rguillebert | djfroofy, http://morepypy.blogspot.com/2011/07/realtime-image-processing-in-python.html | 21:40 |
| Ademan (~dan@197.sub-97-156-104.myvzw.com) left irc: Quit: leaving | 21:41 | |
| djfroofy | rguillebert: thanks for the link | 21:41 |
| bbot2 | 13Exception: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/572 [12alex] | 21:43 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/573 [12hakanardo, jit-targets] | 21:43 |
| djfroofy | exarkun: x samples per second maybe. with regard to fluidsynth though i am able to easily get up to ~500 bpm (with 24 ticks per beat) withouth any stutter. and keep processes in sync. | 21:44 |
| kenaan | 12alex_gaynor jit-dynamic-getarrayitem 11ec49334c3989 15/: close for merge | 21:45 |
| kenaan | 12alex_gaynor default 11bd871afa3feb 15/pypy/: Merged jit-dynamic-getarrayitem. Added support for creating custom getarrayitems at jit-compile time. Steal... | 21:45 |
| basepi (~basepi@server.cmyers.net) left irc: Quit: ZNC - http://znc.sourceforge.net | 21:47 | |
| aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving | 21:49 | |
| basepi (~basepi@server.cmyers.net) joined #pypy. | 21:50 | |
| hruske (~Gasper@188-230-156-183.dynamic.t-2.net) left irc: Quit: Leaving | 21:50 | |
| mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) left irc: Read error: Operation timed out | 21:51 | |
| ericflo (~ericflo@75.103.8.110) joined #pypy. | 21:53 | |
| kenaan | 12alex_gaynor numpy-dtype-refactor 110b6ec862fb6e 15/pypy/: merged default in | 21:57 |
| kenaan | 12alex_gaynor numpy-dtype-refactor 11b1025f8b8ca2 15/pypy/module/micronumpy/: convert to use the new libffi support | 21:57 |
| ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) left irc: Quit: Leaving... | 21:58 | |
| __main__ (~main@adsl-99-173-15-158.dsl.pltn13.sbcglobal.net) left irc: Remote host closed the connection | 21:59 | |
| __main__ (~main@adsl-99-173-15-158.dsl.pltn13.sbcglobal.net) joined #pypy. | 21:59 | |
| syntaxiko (RosesEatP@cpe-67-252-73-238.stny.res.rr.com) left #pypy. | 22:00 | |
| amaury_ | testing a cpyext extension module on top of an *interpreted* pypy is fun | 22:00 |
| Action: Alex_Gaynor whacks the translator with a stick | 22:01 | |
| mattip (~chatzilla@109.64.118.235) joined #pypy. | 22:02 | |
| mattip | fijal: ping | 22:02 |
| rguillebert | amaury_, define "fun" :) | 22:03 |
| amaury_ | when it segfaults, you have 3 different stacks | 22:04 |
| amaury_ | one in C (the host python interpreter) | 22:04 |
| amaury_ | one in python (executing Rpython code, ll2ctypes &co) | 22:05 |
| amaury_ | and the script you run in pypy | 22:05 |
| rguillebert | ah ah | 22:05 |
| amaury_ | which calls back to C, since it's an extension module | 22:05 |
| amaury_ | with a lot of ctypes in between, of course | 22:06 |
| amaury_ | but at least, it *works* | 22:06 |
| amaury_ | well, except for the segfault :-) | 22:06 |
| Alex_Gaynor | define works :D | 22:06 |
| mattip | amaury_: ping | 22:08 |
| amaury_ | pong | 22:08 |
| amaury_ | how is wxPython doing? | 22:08 |
| mattip | Well, I only have ne machine that has enough memory to build pypy. | 22:08 |
| mattip | s/ne/one | 22:08 |
| mattip | and it has been too busy to build the pypy nightly. | 22:09 |
| mattip | But I saw you posted some screenshots. :) | 22:09 |
| amaury_ | yes :-) | 22:09 |
| mattip | Can you put the patch somewhere visible to the wxPython people? | 22:10 |
| amaury_ | mattip: did you try http://buildbot.pypy.org/nightly/trunk/ | 22:10 |
| amaury_ | to get a fresh pypy | 22:10 |
| mattip | Yes, connection from here is really slow, and timezone problems, blah blah blah. | 22:12 |
| chronitis (~chronitis@88-104-113-187.dynamic.dsl.as9105.com) joined #pypy. | 22:14 | |
| amaury_ | ok | 22:14 |
| amaury_ | yes, I'll try to clean the patch and post it soon | 22:15 |
| amaury_ | I'm also working on the next generation of wxPython | 22:15 |
| amaury_ | "phoenix" | 22:15 |
| mattip | "just when you thought it was safe to walk the streets again" | 22:15 |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) left irc: Ping timeout: 258 seconds | 22:16 | |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) joined #pypy. | 22:16 | |
| mattip | amaury_: Is your comment above about segfaults having to do with wxPython refcounting? | 22:19 |
| amaury_ | no, I'm trying the new phoenix version | 22:19 |
| amaury_ | which assumes a lot about the CPython API | 22:20 |
| amaury_ | so I'm trying to rewrite the "sip" layer | 22:20 |
| amaury_ | and avoid this metaclass with a custom allocator | 22:21 |
| amaury_ | crazy code | 22:21 |
| Action: mattip Reading about phoenix | 22:24 | |
| gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) left irc: Remote host closed the connection | 22:32 | |
| apanda (~apanda@dhcp-45-162.EECS.Berkeley.EDU) left irc: Remote host closed the connection | 22:34 | |
| _root_ (~Scalable@li252-14.members.linode.com) joined #pypy. | 22:34 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/705 [12hakanardo, jit-targets] | 22:36 |
| _root_ (~Scalable@li252-14.members.linode.com) left irc: Remote host closed the connection | 22:40 | |
| mattip (~chatzilla@109.64.118.235) left irc: Quit: ChatZilla 0.9.87 [Firefox 8.0a2/20110819142441] | 22:40 | |
| _root_ (~Scalable@li252-14.members.linode.com) joined #pypy. | 22:40 | |
| mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) joined #pypy. | 22:41 | |
| mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep | 22:42 | |
| nickik (~nick@217.150.239.67) left irc: Quit: Verlassend | 22:43 | |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/140 | 22:43 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/713 | 22:43 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/942 | 22:43 |
| _root_ (~Scalable@li252-14.members.linode.com) left irc: Read error: Connection reset by peer | 22:50 | |
| _root_ (~Scalable@li252-14.members.linode.com) joined #pypy. | 22:50 | |
| kenaan | 12alex_gaynor numpy-dtype-refactor 11b2bbe6d8f5fa 15/pypy/jit/backend/llgraph/llimpl.py: Teach llimpl about {get,set}interiorfield_raw with floats. | 22:51 |
| kenaan | 12alex_gaynor numpy-dtype-refactor 11b594ceafe738 15/pypy/rlib/objectmodel.py: Added a decorator for specialize:call_location | 22:51 |
| kenaan | 12alex_gaynor numpy-dtype-refactor 11eb398ab0ef00 15/pypy/rlib/libffi.py: Specialize these properly | 22:51 |
| kenaan | 12alex_gaynor numpy-dtype-refactor 11ea7461f576fa 15/pypy/module/micronumpy/test/test_zjit.py: Update these tests. | 22:51 |
| djfroofy (~djfroofy@66-195-30-142.static.twtelecom.net) left irc: Ping timeout: 276 seconds | 23:05 | |
| kenaan | 12mattip numpy-multidim-shards 11623b485bea06 15/pypy/module/micronumpy/interp_numarray.py: start to fix descr_repr | 23:15 |
| kenaan | 12mattip numpy-multidim-shards 11f508193f73f0 15/pypy/module/micronumpy/: split tests, test_repr passes | 23:15 |
| kenaan | 12mattip numpy-multidim-shards 11989579c2237f 15/pypy/module/micronumpy/: code cleanup | 23:15 |
| tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy. | 23:17 | |
| fzzzy (~donovan@nat/mozilla/x-qgfgkhbfdwcjflsd) left irc: Quit: fzzzy | 23:17 | |
| bobbyz (~bobbyz@184.232.81.149) joined #pypy. | 23:25 | |
| lac (~quassel@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy. | 23:27 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy. | 23:32 | |
| etrepum (~bob@accessnat4.mochimedia.net) left irc: Ping timeout: 276 seconds | 23:35 | |
| Squarepy (~Squarepy@ip82-139-118-42.lijbrandt.net) joined #pypy. | 23:36 | |
| mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Client Quit | 23:36 | |
| Squarepy (~Squarepy@ip82-139-118-42.lijbrandt.net) left irc: Changing host | 23:37 | |
| Squarepy (~Squarepy@unaffiliated/squarepy) joined #pypy. | 23:37 | |
| bobbyz (~bobbyz@184.232.81.149) left irc: Ping timeout: 240 seconds | 23:38 | |
| bobbyz (~bobbyz@184-229-210-255.pools.spcsdns.net) joined #pypy. | 23:39 | |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy. | 23:39 | |
| whitelynx (~whitelynx@li117-47.members.linode.com) left irc: Quit: Ex-Chat | 23:46 | |
| kenaan | 12amauryfa default 11dff1cac01f75 15/pypy/module/cpyext/presetup.py: Allows presetup.py to execute setup.py scripts which use the "if __name__ == '__main__'" idiom | 23:47 |
| kenaan | 12amauryfa default 11365410e9e95e 15/pypy/module/cpyext/include/modsupport.h: Update PyMODINIT_FUNC for C++ extensions. | 23:47 |
| chronitis | anyone on py3k around? | 23:53 |
| amaury_ | a couple of minutes | 23:55 |
| chronitis | I've been going through trying to clear up the remaining unittest fails for <bytes> | 23:56 |
| chronitis | which is mostly done, in a shallow way at least | 23:56 |
| chronitis | bytes is still named string in most places - is there any good reason not to do a wholesale rename to avoid the confusion this currently confuses? | 23:57 |
| amaury_ | regular merges from trunk, I think | 23:57 |
| chronitis | and what is the best way of handling the fact that most methods only support <bytes> and maybe <bytearray> at the moment but should handle anything buffer-like | 23:58 |
| chronitis | a specialised method__String_String... version and a duck-typed method__String_ANY...? | 23:58 |
| chronitis | (eg, some tests fail for not accepting <memoryview> as a bytes-like object) | 23:59 |
| amaury_ | yes, when the versions differ enough | 23:59 |
| --- Wed Nov 16 2011 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!