Timestamps are in GMT/BST.
[0:29] * cschneid (~cschneid@c-76-120-122-74.hsd1.co.comcast.net) Quit (Remote host closed the connection)
[0:35] * cschneid (~cschneid@c-76-120-122-74.hsd1.co.comcast.net) has joined #jython
[2:34] <pjenvey> well, looking at this again, maybe we should rip out ipv6 support from getaddrinfo. the socket module docs show a getaddrinfo results -> socket.socket example
[2:35] <pjenvey> plus the code has to be fixed in not only httplib, but {ftp,pop,smtp,telnet}lib.py
[3:08] * lopex (lopex@89-76-44-27.dynamic.chello.pl) Quit ()
[4:37] * jimbaker (~jimbaker@75-151-81-169-Colorado.hfc.comcastbusiness.net) has joined #jython
[4:51] * jimbaker (~jimbaker@75-151-81-169-Colorado.hfc.comcastbusiness.net) Quit (Quit: jimbaker)
[6:27] * verterok (~ggonzalez@unaffiliated/verterok) Quit (Ping timeout: 265 seconds)
[6:31] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) has joined #jython
[6:47] * verterok (~ggonzalez@184-106-181-175.static.cloud-ips.com) has joined #jython
[6:50] * verterok (~ggonzalez@184-106-181-175.static.cloud-ips.com) Quit (Changing host)
[6:50] * verterok (~ggonzalez@unaffiliated/verterok) has joined #jython
[8:17] <agronholm> that's what I thought too
[8:42] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) Quit (Quit: jabley)
[11:34] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) has joined #jython
[11:36] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) Quit (Client Quit)
[11:46] * lopex (lopex@89-76-44-27.dynamic.chello.pl) has joined #jython
[11:52] * stakkars (~tismer@77-21-85-35-dynip.superkabel.de) Quit (Read error: Operation timed out)
[11:53] * stakkars (~tismer@77-21-85-35-dynip.superkabel.de) has joined #jython
[12:15] * seberg (~sebastian@134.76.0.204) has joined #jython
[12:20] * Oti (~ohumbel@adsl-89-217-43-96.adslplus.ch) has joined #jython
[12:29] <Oti> just read the discussion on socket / httplib
[12:30] <Oti> pjenvey - what if we leave everything as it is, except raising a socket.error instead of an AssertionError?
[12:30] <agronholm> doesn't sound like a very good idea
[12:31] <agronholm> what errno would you put there then
[12:31] <Oti> agronholm - well i suspected that answer :-)
[12:34] <Oti> no idea what number, my hope was that any socket.error is better than an AssertionError
[12:37] <agronholm> imho patching getaddrinfo will give the best practical results, as IPv6 does not work in Jython anyway
[12:37] <agronholm> well, not completely at least
[12:42] <Oti> that was pjenvey's last statement, and he also mentioned we should fix the callers: httplib, ftplib, poplib, smtplib, telnetlib
[12:42] <Oti> i am afraid i won't get that right
[12:42] <agronholm> that won't be necessary if we patch getaddrinfo
[12:43] <Oti> are you sure?
[12:44] <agronholm> the problem stems from getaddrinfo() returning ipv6 addresses and then the callers using said addresses with sockets
[12:45] <agronholm> but if getaddrinfo() is patched so it only returns ipv4 addresses, then there won't be a problem
[13:29] <Oti> can we return an IPv6 address a to the address info, if a.isIPv4CompatibleAddress() ?
[13:30] <agronholm> what's the point?
[13:30] <agronholm> wouldn't the same addresses be returned in an ipv4 address lookup too?
[13:31] <Oti> that's exactly the point i do not know: should i exclude every ipv6 address?
[13:31] <agronholm> just make it query for AF_INET addresses regardless of the address family parameter
[13:32] <agronholm> that should take care of the problem
[13:33] <Oti> ok, i'll try
[13:50] * jimbaker (~jimbaker@75-151-81-169-Colorado.hfc.comcastbusiness.net) has joined #jython
[14:01] * agronholm (~demigod@nblzone-211-37.nblnetworks.fi) Quit (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
[14:03] * agronholm (~demigod@nblzone-211-37.nblnetworks.fi) has joined #jython
[14:09] * jimbaker (~jimbaker@75-151-81-169-Colorado.hfc.comcastbusiness.net) Quit (Quit: jimbaker)
[14:56] <Oti> agronholm: what do you think of http://bugs.jython.org/file934/1697-patch2.txt?
[14:56] <Oti> this simply prevents IPv6 addresses from being added to the result of getaddrinfo()
[14:59] <agronholm> I thought we agreed on only querying for AF_INET addresses
[14:59] <agronholm> what is the point of this?
[14:59] <agronholm> why include AF_INET6 in the query if you're stripping away those addresses anyway?
[15:01] <Oti> because InetAddress.getAllByNames() returns all types anyway
[15:02] <Oti> and there is some magic in the filtering i do not understand - i was not able to skip Inet6Addresses there - my fault, most likely
[15:02] <agronholm> let me play around with it for a bit
[15:03] <Oti> sure, thanks!
[15:07] <agronholm> wouldn't it be simple to just do isinstance(x, Inet6Address) to filter out those?
[15:08] <agronholm> am I missing something?
[15:11] <Oti> but we have to tweak the family flag as well, because it goes into the result tuple
[15:11] <agronholm> not sure I understand
[15:12] <agronholm> what tweaks need to be done for ipv4 addresses?
[15:20] <agronholm> oti, so what exactly is the problem here?
[15:20] <agronholm> a more careful look at your patch shows that it's just what we need
[15:33] <Oti> thanks - i almost started doubting myself :-)
[15:35] <agronholm> not having ipv6 support blows but this is the best temporary fix I can think of
[15:43] <Oti> yes, it feels sad to pull the release trigger with known issues
[15:48] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) has joined #jython
[16:27] * MarderIII (~marderii@enneman.demon.nl) has joined #jython
[17:14] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) Quit (Quit: jabley)
[17:22] * MarderIII (~marderii@enneman.demon.nl) Quit (Quit: [IRSSI] Dont push that button. No.. NO DONT PU...)
[17:23] * MarderIII (~marderii@enneman.demon.nl) has joined #jython
[17:51] <pjenvey> cool, +1
[18:06] <agronholm> anything else missing from the final release?
[18:55] <pjenvey> amak just chimed in on #1697
[18:56] <agronholm> I wonder what he's planning to do differently
[18:58] <pr3d4t0r> pjenvey: Hej.
[18:58] <pr3d4t0r> pjenvey: I'm closer to agronholm these days than to you :(
[19:00] <agronholm> "One of the primary problems is this java bug, which effectively prevents the use of IPv6 in jython on Windows version older than Vista, i.e. it affects XP/2003."
[19:00] <agronholm> yeah well ipv6 support is a joke in winxp anyway
[19:08] <Oti> pjenvey - what should we do about #1697 regarding 2.5.2 ?
[19:08] <agronholm> amak is probably going to write some more comments
[19:12] <pjenvey> yea, it's up to amak at this point. sounds like more's coming
[19:46] * shanka (~shashank@c-98-245-87-8.hsd1.co.comcast.net) has joined #jython
[19:55] <Oti> ok, i'm going into standby mode...
[20:04] <pjenvey> alright, I'll harass amak some more if we don't hear anything from him in a bit =]
[20:18] * shanka (~shashank@c-98-245-87-8.hsd1.co.comcast.net) Quit (Ping timeout: 255 seconds)
[20:42] <agronholm> he just did
[20:42] <agronholm> reply
[20:43] <Oti> Alan asks for a day or two - for me that's fine
[20:43] <Oti> given the long time we already had...
[20:47] <pjenvey> cool
[20:48] <pjenvey> I've been tempted to just branch trunk to a 2.5-maint but we can just wait a couple more days
[20:48] <Oti> agreed!
[20:48] <pjenvey> looking back on 2.5, we should have just branched after a 2.5 release. we wanted to keep trunk 2.5 so we'd get more focus on it (mostly for like clamp) but that didn't happen anyway
[20:49] <pjenvey> so hopefully, couple more days, do a release, then we can begin 2.6 stuff. maybe we can get a fair amount of 2.6 work done during the pycon sprints
[20:51] <Oti> yes, i strongly hope so, too!
[21:03] * MarderIII (~marderii@enneman.demon.nl) Quit (Ping timeout: 240 seconds)
[21:19] * clajo04_ (~clajo04_@pool-74-108-95-175.nycmny.fios.verizon.net) Quit (Quit: clajo04_)
[21:25] * clajo04_ (~clajo04_@pool-74-108-95-175.nycmny.fios.verizon.net) has joined #jython
[21:48] * Oti (~ohumbel@adsl-89-217-43-96.adslplus.ch) Quit (Quit: Oti)
[21:51] * seberg (~sebastian@134.76.0.204) Quit (Quit: Ex-Chat)
[22:20] <fwierzbicki> +1 on a 2.5-maint! I've been dusting off some work I did on the 2.6 grammar changes lately
[22:22] <fwierzbicki> I think I've settled in at my newest job enough to start contributing again
[22:43] <agronholm> I wonder what else needs to be done for 2.6 besides grammar changes
[22:43] <agronholm> PySystemState/PyThreadState redesign?
[22:43] <agronholm> and clamp :))
[23:17] <pjenvey> fwierzbicki - cool, IIRC you had most of the 2.6 grammar changes already done
[23:17] <pjenvey> agronholm - ideally, off the top of my head, invokedynamic support, cjkcodecs, bz2
[23:17] <pjenvey> there's probably a million other things =]
[23:18] <agronholm> ipv6 support :)
[23:18] <pjenvey> there really isn't too much between 2.5 and 2.6, so who knows, if we can get most of the stdlib passing, some of those things might end up not making it
[23:19] <pjenvey> or could come in later point releases
These logs were automatically created by JythonLogBot_ on irc.freenode.net using a slightly modified version of the Java IRC LogBot (github).