#pypy IRC log for Tuesday, 2007-04-17

braintoneafa - r42108 - in pypy/dist/pypy/rlib: . test - rsocket progress: implementation of getnameinfo() and getaddrinfo() on Windows. Some basic tests added, but no IPv6 support.00:05
braintoneafa - r42109 - in pypy/dist/pypy/rlib: . test - CPython2.3 on Windows did perform a reverse DNS lookup in getaddrinfo. Implement 2.4 behaviour, which is consistent with Unix platforms.00:25
dialtone (n=dialtone@81-208-31-208.ip.fastwebnet.it) left irc: 00:50
froosch_ (n=frosch@ joined #pypy.00:54
braintoneafa - r42111 - pypy/dist/pypy/rlib - rsocket progress on Windows: implement timeouts and _select()00:55
froosch (n=frosch@ left irc: Read error: 110 (Connection timed out)00:58
braintoneafa - r42112 - pypy/dist/pypy/rlib - Argh. fd_set structure exists only on win32.00:59
braintoneafa - r42113 - pypy/dist/pypy/rlib/test - Skip some rsocket tests not relevant on win3201:08
huangyi (n=huangyi@ joined #pypy.01:56
Rhamphoryncus (n=rhamph@unaffiliated/rhamphoryncus) left irc: "Valid HTML! http://validator.w3.org/ | Support ISO 8601! http://www.cl.cam.ac.uk/~mgk25/iso-time.html"02:10
joe_mp_ (n=joe@p549C7AEA.dip.t-dialin.net) joined #pypy.02:13
joe_mp__ (n=joe@p549C7AEA.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out)02:28
joe_mp__ (n=joe@p549C7AEA.dip.t-dialin.net) joined #pypy.02:28
joe_mp_ (n=joe@p549C7AEA.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out)02:35
simonpy (n=simon@ left irc: Read error: 110 (Connection timed out)03:06
rhettinger (n=rhetting@ left irc: Read error: 110 (Connection timed out)03:06
joe_mp__ (n=joe@p549C7AEA.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out)03:47
lene (n=chatzill@central.merlinux.de) left irc: Read error: 110 (Connection timed out)03:59
joe_mp__ (n=joe@p549C721F.dip.t-dialin.net) joined #pypy.04:01
hpk (n=hpk@central.merlinux.de) left irc: Read error: 110 (Connection timed out)04:01
Rhamphoryncus (n=rhamph@unaffiliated/rhamphoryncus) joined #pypy.04:23
huangyi (n=huangyi@ left irc: "ChatZilla [Firefox]"04:44
Nick change: headtone -> wristtone06:35
Nick change: __off__ -> __doc__06:47
joe_mp_ (n=joe@p549C6196.dip.t-dialin.net) joined #pypy.07:01
hpk (n=hpk@central.merlinux.de) joined #pypy.07:05
joe_mp__ (n=joe@p549C721F.dip.t-dialin.net) left irc: Read error: 60 (Operation timed out)07:09
dialtone (n=dialtone@81-208-31-208.ip.fastwebnet.it) joined #pypy.07:37
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Read error: 110 (Connection timed out)07:55
Rhamphoryncus (n=rhamph@unaffiliated/rhamphoryncus) left irc: "Valid HTML! http://validator.w3.org/ | Support ISO 8601! http://www.cl.cam.ac.uk/~mgk25/iso-time.html"08:08
fijal (n=fijal@host-53-19.compi.net.pl) joined #pypy.08:30
arre (n=ac@alchemist.openend.se) joined #pypy.08:31
antocuni (n=antocuni@ joined #pypy.08:44
fijalhi mwh, hi anto08:47
fijalmwh: feedback from Michael Foord was also quite positive08:50
mwhfijal: cool08:51
cfbolz (n=cfbolz@fwstups.cs.uni-duesseldorf.de) joined #pypy.08:51
mwhhe's interested in the extcompiler for .net, right?08:51
mwhhi cf08:51
fijalmwh: yes08:51
fijalcfbolz: hi carl08:51
mwhsounds like work to me :)08:51
antocunihi michael, carl, maciek!08:51
cfbolzby now I really should know better08:52
fijalmwh: he would like to have parts of RPython compiled as libraries to .NET08:52
fijalI don't think it would be that hard to achieve08:52
antocuniit's not hard, but neither trivial08:52
cfbolzmy thesis was not accepted because I have a very slight variation on the thesis than what the uni thinks the title is08:52
mwhfijal: i guess that's actually easier than making CPython extensions, at least in some ways08:53
fijalcfbolz: what?08:53
mwhcfbolz: ouch :/08:53
fijalmwh: surely08:53
mwhbut also :)08:53
antocunicfbolz: :-(08:53
cfbolzwell, not too bad08:53
fijalcfbolz: what now?>08:53
cfbolzI have to get a letter from michaeL stating that the titles are equivalent08:53
fijalheh :)08:54
antocunibtw, about the compiler for .NET08:54
mwhmichaeL is a very useful man :)08:54
antocuniI'm not sure if the best way to go in that direction is to write extend the extcompiler08:55
cfbolzantocuni: why not?08:55
lene (n=chatzill@central.merlinux.de) joined #pypy.08:55
mwhhttp://weblog.lonelylion.com/2007/04/16/first-pycon-now-the-olympics/ :)08:55
antocuniprobably it's easier to provide a general compiler for .NET08:55
antocunii.e., equivalent to C$08:55
antocuniIronPython can already uses that kind of libraries seamlessy08:56
Action: antocuni can't really type correctly this morning :-(08:56
cfbolzah, I see08:56
fijalantocuni: well, what's the difference?08:56
cfbolzantocuni: ironpython does this kind of wrapping already, so the extcompiler does not have to do it, right?08:56
fijalI mean you need completely different hacks than extcompiler for C08:57
antocunimore or less yes08:57
fijalah cool08:57
antocuniin the C#-like case you don't have to know nothing about the ironpython api08:57
cfbolzsomething similar is true for jython too, I guess08:57
fijalcool :)08:57
fijalso it would be very easy than08:57
antocuniwell, maybe an extcompiler-like tool would be useful as well08:57
antocuniyou could mix app-level and interp-level code in the same module08:58
antocunibut I think that at the moment it's much more usefult the general compiler for .NET08:58
cfbolzantocuni: well, you can create some python code that interfaces nicely08:58
antocunithis task will probably be the first I work on as soon as I restart to code :-)08:59
antocuniand Michael Foord told me he's interested in trying it at work08:59
fijalantocuni: he definitely is09:00
fijalI think he's quite scared by maturity of all of the pypy09:00
fijalbut is eager to try anyway09:00
cfbolzfijal: you mean by lack of maturity, I guess?09:01
fijalcfbolz: yes09:01
fijalI've tried to emphasis that parts, ie extcompiler are not worth trying in production environment yet09:01
fijal (n=fijal@host-53-19.compi.net.pl) left irc: "Leaving"09:05
xorAxAxcfbolz: hmm, what kind of difference was there that could be made such a fuzz about?09:06
cfbolzxorAxAx: they said they would make a fuzz about "every" difference09:07
xorAxAxhow can they judge the difference without having read the thesis? did they just look at the abstract?09:08
cfbolzonly at the title page09:08
xorAxAxhow can they make out the difference if they just look at one side?09:08
cfbolzthere is a title that I was registered with, and it is a slightly different one on the title page09:09
Action: cfbolz -> prüfungsamt09:11
jkakar (n=jkakar@yttrium.canonical.com) joined #pypy.09:15
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.09:16
arigo (n=arigo@d83-180-250-6.cust.tele2.ch) joined #pypy.09:22
antocunihi armin!09:23
xorAxAxhi arigo 09:29
mwhmorning armin09:31
arigoargh, CPython HEAD still contains strange new checks about what kind of metaclass is ok and what kind is not?09:38
arigoI thought I reverted this09:39
cfbolzhi armin!09:40
cfbolzso, my thesis is officially handed in just in time09:40
cfbolz(and I have a form to prove it)09:41
arigo(oh well, I just meant to revert it but never did it, apparently)09:42
arigocfbolz: fiesta tonight! :-)09:42
cfbolzarigo: something like this :-)09:43
antocunicfbolz: congrats09:44
antocuniwhen are you graduating?09:44
pedronis (i=pedronis@vitaly.openend.se) joined #pypy.09:44
cfbolzantocuni: there are some things missing still09:48
cfbolzhi samuele09:48
cfbolzantocuni: a professor didn't tell the results of one of my exams to the office that's responsible yet. will have to chase him to do it09:50
cfbolzantocuni: and I still have to give a talk about my thesis09:50
antocuniah, so that kind of things happen also in germany :-)09:51
antocuniit was the same for me09:51
dialtone (n=dialtone@81-208-31-208.ip.fastwebnet.it) left irc: 09:59
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Read error: 110 (Connection timed out)10:07
arigo (n=arigo@d83-180-250-6.cust.tele2.ch) left irc: Remote closed the connection10:28
jiwon_ (n=jiwon@rescomp-05-80325.Stanford.EDU) joined #pypy.10:34
jiwon_ (n=jiwon@rescomp-05-80325.Stanford.EDU) left irc: 10:40
zx64 (n=zx64@freecnc/zx64) left irc: Remote closed the connection10:50
picxk (n=picxk@ joined #pypy.11:04
fijal (n=fijal@genesilico.pl) joined #pypy.11:05
picxkwhat would be the implication of the stackless pypy on the rest of the system?11:06
picxkfor instance, scheduler, h/w11:06
picxkparallelism pattern..., etc.11:07
picxkfor instance, taskelets only(?) communicates through channels, so does this mean11:07
picxkthat snoopy cache coherence is not proper for stackless pypy?11:08
dialtone (n=dialtone@81-208-31-208.ip.fastwebnet.it) joined #pypy.11:10
mwhwell, as currently implemented it's all at rather a higher level than that11:27
picxkwhat about scheduler or gc?11:48
picxkdoes stackless pypy need some support from them?11:48
fijal (n=fijal@genesilico.pl) left irc: Remote closed the connection12:00
antocuni (n=antocuni@ left irc: "Leaving"12:01
mwhwhat do you mean by scheduler?12:07
picxkcould be os scheduler, could be scheduler for tasklets12:17
jcea2 (n=jcea@jabber.hst.ru) left irc: Remote closed the connection12:18
picxkit seems like stackless pypy gets benefits from gc framework in pypy... so12:18
mwhwell, pypy doesn't provide any kind of sophisticated scheduling itself12:18
mwhyou could probably write some, not really sure12:18
mwhthe logic space needs both stackless and the framework gc, but for different reasons12:19
picxkwhat reasons are they?12:21
mwhit uses coroutines for concurrency, and the framework gc for cloning computation spaces12:22
mwh(i think, i'm not exactly an expert on it)12:22
cfbolzmwh: sounds right12:23
picxkwhy does it need a special gc or gc framework for cloning computation spaces?12:23
cfbolzhow else would you clone something?12:24
Action: cfbolz --> off12:26
cfbolz (n=cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: "Leaving"12:26
braintoneafa - r42123 - pypy/dist/pypy/rlib/test - Adapt rsocket tests for win32: - windows often returns EWOULDBLOCK errors - local sockets seem to have a small buffer12:57
dialtone (n=dialtone@81-208-31-208.ip.fastwebnet.it) left irc: 12:59
fijal (n=fijal@genesilico.pl) joined #pypy.13:04
antocuni (n=antocuni@ joined #pypy.13:37
jcea3 (n=jcea@jabber.hst.ru) joined #pypy.13:45
Nick change: jcea3 -> jcea213:45
fijalantocuni: re-hi14:07
antocunire-hi maciek14:08
fijalantocuni: what do you think about ext-compiler for .NET? Do you plan to do one, or are rather waiting for someone to do it?14:08
stakkars (i=czolqi@i577B5B87.versanet.de) left irc: 14:10
antocuniI plan to write a general rpython compiler for .NET14:11
antocuniwhose output can be reused also from C#, VB.NET, etc.14:12
stakkars (i=bvdzkgt@i577B5B87.versanet.de) joined #pypy.14:14
stakkars (i=bvdzkgt@i577B5B87.versanet.de) left irc: Client Quit14:16
fijalantocuni: aha14:16
fijalantocuni: but than how easy would be to provide bindings for IronPython? Probably not hard14:16
antocuniIronPython can import any .NET library for free14:16
antocuniso the rpython compiler would produce code reusable also from IronPython14:17
stakkars (i=zvkexlbh@i577B5B87.versanet.de) joined #pypy.14:17
pedronisantocuni: but's what the speed difference between general bindings and specific bindings. In jython at the moment can be quite big14:18
pedronis(reflection vs. not)14:18
stakkars (i=zvkexlbh@i577B5B87.versanet.de) left irc: Client Quit14:18
antocuniI guess that there would not be any difference because of the way IronPython works14:19
antocunifor example14:19
antocunipython integers are represented as .NET boxed integers14:19
antocuniso I can't see how the "specific bindings for IronPython" would speed up things14:20
fijalI was thinking a bit about limitations of IronPython14:20
fijalie xxx.__class__ = stuff14:20
fijalwill not work at all14:20
fijalbut that doesn't come for free, ie there are penalties for that (which are the main difference I think)14:21
antocuniin which context?14:21
fijalin IronPython you cannot change __class__ at runtime14:21
fijalI think you cannot modify a class attribute14:21
fijalat all14:21
fijalyou cannot access frames, etc.14:21
antocuniok, but how is this related with the ext-compiler?14:22
fijalnot at all :)14:23
antocuniah, ok :-)14:23
fijalI was rather thinking how to optimize stuff, using default .NET frames and going back to frame simulation just in case14:23
fijalthis would need some notion of virtual xxx or sth14:24
antocunido you mean for pypy-cli?14:24
fijalbut this will need some pypy interpreter changes, which would be hardly general14:24
antocuniI think so14:25
fijala bit like virtual frames in jit14:25
antocuniwe could also think about generating .NET bytecode instead of python bytecode14:26
antocunias IronPython does14:26
fijalsounds like jit backend :)14:26
fijalthat would make more sense to do it just-in-time14:26
antocunino, I mean at compile time14:26
fijalif we have technology, why not do it14:27
antocuniyes, it's a possibility14:27
fijalwould need some style of ootypesystem jit14:27
fijalbut still I guess even easier to do than x86 one14:27
fijalcan you modify .NET bytecode at runtime?14:28
antocuniarmin told me that it would not be so hard to port the jit to ootype14:28
antocuniI don't know14:28
fijalcool :)14:28
antocunibut I guess not14:28
fijalbut anyway, sounds like a great opportunity14:28
fijalwhy not?14:28
fijalI would start from stackless though ;-)14:28
antocunithe main blocker now is my professor14:28
antocunihe wants me to work on the rpython-compiler14:29
fijalwhat is rpython-compiler?14:29
antocunithe rpython compiler for .net14:30
antocunilet's call it rpython# :-)14:30
fijalwhat's difference between rpython compiler and pypy's cli backend?14:31
antocuninot much14:31
antocunigenerally speaking, it must be more user-friendly14:31
antocunimoreover, we have to get rid of name mangling14:32
antocuniand find a way to specify explicit signatures for functions14:32
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.14:32
antocunibecause if we compile a library we don't have en entry-point14:32
fijalwell, that's easy14:33
antocuniI'll begin working on it very soon14:34
antocuninext, I have to polish the static bindings for .NET14:34
antocunibeacuse so far they are quite hackish14:34
fijal (n=fijal@genesilico.pl) left irc: "Leaving"15:15
User995 (n=User@lt-quet3.dfki.uni-sb.de) joined #pypy.15:16
jcea2 (n=jcea@jabber.hst.ru) left #pypy.15:16
User995 (n=User@lt-quet3.dfki.uni-sb.de) left irc: Client Quit15:16
danad (n=danad@lt-quet3.dfki.uni-sb.de) joined #pypy.15:16
stakkars (n=tismer@gprs-pool-1-020.eplus-online.de) joined #pypy.15:29
cfbolz (n=cfbolz@ip127.22.reserved.ish.de) joined #pypy.15:29
mwhsegfaulting tests?15:33
pedronisnot good15:34
mwhfijal's string formatting stuff?15:34
stakkars (n=tismer@gprs-pool-1-020.eplus-online.de) left irc: 15:35
pedronispossibly yes15:36
pedronisis not like there's been a lot of checkins recently15:36
pedronisbut there's been some other changes too here and there15:36
cfbolzmwh: armin planned to refactor it on the train15:37
mwhoh indeed15:37
mwhi wonder if he will fix the bugs too :)15:37
pedronissince when the tests are segfaulting15:37
mwh(if indeed this is where the bugs are)15:37
mwhseems around 4200015:37
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Read error: 110 (Connection timed out)15:40
pauldeg (n=paul@n1-29-169.dhcp.drexel.edu) joined #pypy.15:49
pauldeghi everyone16:04
xorAxAxhi pauldeg 16:04
xorAxAxgood morning :)16:04
pauldegjust an update where I am: finishing up Antonio's thesis16:04
pauldeghahaha good evening :) (Where are you located?)16:05
xorAxAx5 pm here16:05
pauldegawesome, I guess correct then!16:06
xorAxAxi usually just say moin, regardless of the actual daytim16:06
cfbolzxorAxAx: moin means good anyway16:07
pauldegcfbolz: hi carl, how's it going?16:08
xorAxAxcfbolz: not directly, but i think people dont agree on it - http://moinmoin.wikiwikiweb.de/MoinMoinEtymology16:08
antocunihi paul!16:09
pauldeghi antonio!16:09
pauldegbuona sera16:09
cfbolzpauldeg: hi paul16:10
cfbolzxorAxAx: who disagrees?16:10
jcea3 (n=jcea@jabber.hst.ru) joined #pypy.16:10
Nick change: jcea3 -> jcea216:10
antocunipauldeg: :-)16:10
xorAxAxcfbolz: well, read the page, non-scientists disagree there and e.g. the WP article also disagrees with that wikipage (providing only sparse references to sources)16:10
jcea2 (n=jcea@jabber.hst.ru) left #pypy.16:11
jcea2 (n=jcea@jabber.hst.ru) joined #pypy.16:12
antocunipauldeg: I won't be at home from tomorrow until saturday, I don't know how much I'll be able to be online16:14
antocunifeel free to send me an email if you need something16:14
cfbolzxorAxAx: then I am too lazy to dig through the whole discussion just to find some badly-reasoned disagreement :-)16:15
cfbolzpauldeg: you can probably also ask the other pypyers if you have problems16:15
pauldegantocuni: So far things have been making a lot of sense to me, in fact I found a place where I had the wrong mental model about something16:15
xorAxAxcfbolz: the above wikipage shows different oppinions and i havent found scientific sources for any of them (but i havent seriously digged either :))16:15
pauldegcfbolz: thanks!16:16
pauldegWhere does niko go to school?16:16
cfbolzxorAxAx: the moin = good theory is (fwliw) taught at universities: I got it from my girlfriend, who has it from some lecture16:16
cfbolzpauldeg: no, he is at ETH Zürich, I think16:17
pauldegohhh awesome16:17
xorAxAxcfbolz: oh well, maybe she has some sources16:17
pauldegThere's a Niko at Drexel, I just wanted to make sure it wasn't him hahaha16:17
cfbolzxorAxAx: I guess I could ask her, if you are really interested16:18
cfbolzhah. moinmoin.org does not give you what you would expect16:18
xorAxAxcfbolz: yeah, as much as i am studying a cultural science, i am :)16:18
cfbolzxorAxAx: ok, will do then :-)16:19
cfbolz(mind you, it will most likely to be some ethymology dictionary or something)16:20
xorAxAxperfectly ok :)16:26
jcea2 (n=jcea@jabber.hst.ru) left irc: Remote closed the connection16:26
jcea2 (n=jcea@jabber.hst.ru) joined #pypy.16:27
mwhpauldeg: hmm, i don't think i actually set up your codespeak account yet did i?16:44
pauldegmwh: I never got an email confirmation from you16:45
pauldegI was going to email you again today to check the status on that16:45
fijal (n=fijal@host-53-19.compi.net.pl) joined #pypy.16:47
mwhpauldeg: right, you should be able to log in now as pdg (paul was indeed already taken)16:47
pauldegmwh: awesome, thanks a lot!16:48
mwhnext step is to try "svn mkdir http://codespeak.net/svn/user/pdg"16:50
antocuni (n=antocuni@ left irc: "Leaving"16:50
fijalsegfaulting pypy-c16:53
fijalseems to be my fault16:53
fijalah, got it16:53
fijalspace.getitem(None, space.wrap('stuff'))16:54
cfbolzwhy should that segfault?16:54
fijalit's interp-level None16:54
fijalbut not None in a sense None, it's just an empty instance of sth16:55
cfbolza, None but not w_None16:56
fijalyes, I'm fixing16:56
pauldegmwh: my password for pdg isn't working, any ideas?16:57
Nick change: __doc__ -> __off__16:58
mwhyou can log in, but not commit to svn?16:58
mwhthis happened before, and i forgot which bit i had to thwack to get it to work16:58
pauldegI go to make a commit, approve the message, and when I enter the username pdg, and the password for the shh2 key i sent you, it doesn't authenticate16:58
pauldegis this also an ssh account? or just svn username and password?16:59
mwhit's an ssh account16:59
fijalwith methodcache patch, one unicode test seems to be failing17:00
mwhthe key you sent me doesn't have a password, that's associated with the private part you didn't send me :)17:00
braintonefijal - r42131 - pypy/dist/pypy/objspace/std - Use space.w_None instead of real None, bad fijal.17:00
mwhif you ssh pdg@codespeak.net 17:00
fijalmwh: newline issues?17:00
mwhyou'll find your unix password in ~/.ssh/password17:01
mwhfijal: i don't think so17:01
fijalmwh: the key must be pasted without newlines17:01
mwhfijal: um17:01
mwhfijal: where did you get that idea of a problem from?17:01
fijalhappened to me few times17:02
mwhok, but i don't think it's the issue here :)17:02
fijalmwh: I think this was it few tuatara keys or so, doesn't matter17:03
arre (n=ac@alchemist.openend.se) left irc: Remote closed the connection17:03
pauldegmwh: just the key part, or do i need the proc-type header too?17:08
mwhpauldeg: ?17:08
pauldegmwh: pasting the dsa private key in password17:08
pauldegmwh: ?17:08
mwhno no17:09
mwhyou now have a shell account on codespeak.net17:09
mwhwhich has the key you sent me in .ssh/authorized_keys17:09
pauldegcorrect, and I can ssh17:09
fijalmwh: segfault should be fixed by now17:10
mwhso you can use svn+ssh: to access the repo 17:10
mwhor if you want to use http:// urls you need your unix password17:10
mwhwhich is in ~pdg/.ssh/password17:11
pauldegohhhhh, ok, i get what you mean now17:11
mwhapologies for being unclear:)17:11
pauldeghahaha no worries at all :)17:12
mwhpauldeg: is it working?17:22
pauldegmwh: hahah yes it is.  I had some typos in my path, but I'm better now17:23
pauldegmwh: thanks for your help and patience17:24
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.17:27
danad (n=danad@lt-quet3.dfki.uni-sb.de) left irc: "Leaving"17:30
pauldeg_ (n=paul@n1-29-169.dhcp.drexel.edu) joined #pypy.17:32
pauldeg (n=paul@n1-29-169.dhcp.drexel.edu) left irc: Read error: 104 (Connection reset by peer)17:32
dialtone (n=dialtone@213-156-52-97.fastres.net) joined #pypy.17:39
pauldeg_dialtone: scott?17:42
cfbolzpauldeg_: no17:42
pauldeg_ohhh ok, just curious17:42
cfbolzthat is valentino17:42
pauldeg_ohh ok, thanks17:42
pauldeg_ (n=paul@n1-29-169.dhcp.drexel.edu) left irc: 17:43
pauldeg (n=paul@n1-29-169.dhcp.drexel.edu) joined #pypy.17:43
Action: pedronis -> home17:47
pedronis (i=pedronis@vitaly.openend.se) left irc: "Chatzilla 0.9.75 [Firefox]"17:47
lene (n=chatzill@central.merlinux.de) left irc: "ChatZilla 0.9.61 [Mozilla rv:1.7.12/20050915]"18:00
jkakar (n=jkakar@yttrium.canonical.com) left irc: Read error: 113 (No route to host)18:25
jcea2 (n=jcea@jabber.hst.ru) left #pypy.18:43
jcea2 (n=jcea@jabber.hst.ru) joined #pypy.18:44
jcea2 (n=jcea@jabber.hst.ru) left #pypy.18:45
jcea2 (n=jcea@jabber.hst.ru) joined #pypy.18:45
jcea2 (n=jcea@jabber.hst.ru) left #pypy.18:46
jcea2 (n=jcea@jabber.hst.ru) joined #pypy.18:47
Rhamphoryncus (n=rhamph@unaffiliated/rhamphoryncus) joined #pypy.18:54
zx64 (n=zx64@freecnc/zx64) joined #pypy.19:08
stakkars (n=tismer@gprs-pool-1-016.eplus-online.de) joined #pypy.19:24
fijalhi christian19:25
pauldegfijal: what's christian's alias19:26
stakkars (n=tismer@gprs-pool-1-016.eplus-online.de) left irc: Client Quit19:26
fijalpauldeg: stakkars19:28
pauldegohh cool19:28
fijalpauldeg: are you doing some development regarding pypy right now?19:40
pauldegfijal: jvm backend enhancements, jsr223 bindings19:40
pauldegat this very moment, no :)19:40
fijalI know *what* are you doing, was just asking about now19:41
fijalanyway, if you feel at any point we can help you somehow ask on channel19:41
pauldegfijal: thanks!  I appreciate that19:41
braintoneafa - r42135 - pypy/dist/pypy/rlib/test - rsocket intermittent failure: win32 sockets seem to need a few cycles to transfer data19:41
pauldegI'm finishing reading antonio's thesis19:41
fijalgood :)19:43
Nick change: froosch_ -> froosch20:16
__doc__ (n=doc_pyth@84-73-211-45.dclient.hispeed.ch) joined #pypy.20:31
cfbolz (n=cfbolz@ip127.22.reserved.ish.de) left irc: Read error: 110 (Connection timed out)20:42
braintoneafa - r42136 - in pypy/dist/pypy: module/rsocketnmodule/rsocket/test rlib rlib/test - rsocket progress on win32: hide socket.dup, add implementation of inet_aton20:52
pauldeg (n=paul@n1-29-169.dhcp.drexel.edu) left irc: 21:12
braintoneafa - r42137 - in pypy/dist/pypy: module/rsocketnmodule/rsocket/test rlib - Hide socket.inet_pton (inet_ntop does exist and works) on win32 + win32 may return errno.WSAEINVAL instead of EINVAL Only one failing test in rsocket module! But many functions are not tested at...21:13
braintoneafa - r42138 - pypy/dist/pypy/module/rsocket/test - This test should not run on win32, after all.21:18
braintoneafa - r42139 - in pypy/dist/pypy: module/rsocketnmodule/rsocket/test rlib - Correction of previous checkin: on win32 inet_ntop can detect correctly bogus args, but is unable to work in the normal case... :( Untested code never works!21:46
picxk (n=picxk@ left irc: "Lost terminal"22:06
braintoneafa - r42140 - in pypy/dist/pypy: module/rsocket rlib - Call WSAStartup() on module import. Also hide two functions invalid on win3222:28
jcea2 (n=jcea@jabber.hst.ru) left irc: Remote closed the connection22:28
jcea2 (n=jcea@jabber.hst.ru) joined #pypy.22:32
simonp1 (n=simon@ joined #pypy.22:39
simonp1i just presently wished that i could __extend__ multiple classes..22:40
simonp1it just works (TM)22:46
fijalsimonp1: :)22:58
stakkars (i=mlelqp@i577B58F5.versanet.de) joined #pypy.23:09
cfbolz (n=cfbolz@ip127.22.reserved.ish.de) joined #pypy.23:11
fijalre-hi carl23:13
fijalcfbolz: how's your thesis, submitted?23:18
fijalnice :)23:21
fijaland gn :)23:21
idnar (i=mithrand@unaffiliated/idnar) left irc: Connection timed out23:28
jacob22|home (n=jacob@c-3cc6e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Read error: 110 (Connection timed out)23:34
cfbolz (n=cfbolz@ip127.22.reserved.ish.de) left irc: "Leaving"23:54
--- Wed Apr 18 200700:00

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