#jython IRC Log

Index

IRC Log for 2016-07-26

Timestamps are in GMT/BST.

[0:44] <patoff> stewori: cool man will check it out
[1:10] <stewori> :)
[1:11] <stewori> Note that you must use Jython 2.7.1 for this. For now either beta3 or building from repository will do
[1:20] <stewori> You can use sqlite3 bundled with your CPython then, however I would recommend not to simply ass usr/lib/python2.7 to Jython's sys.path, but instead copy the sqlite3-folder to a Jython-specific lib-folder which you can provide to Jython then
[1:20] <stewori> *add
[1:22] <stewori> otherwise this could mess up Jython's lib too much
[1:24] <stewori> on the other hand it's okay to simply add CPython's lib-dynload folder to provide _sqlite3.so
[1:28] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Remote host closed the connection)
[2:10] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[2:15] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 265 seconds)
[3:11] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[3:17] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 260 seconds)
[4:13] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[4:17] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 258 seconds)
[4:46] <jimbaker> stewori, wow, this is very impressive!
[4:47] <jimbaker> (i wonder if we could have just gotten socket/select/ssl to work this way...)
[4:48] <jimbaker> btw, i have made zero headway so far on that publication bug in PyType, but that's most likely due to not enough time to focus
[4:49] <jimbaker> (to reproduce on a fast system, simply remove the synchronized {} blocks - that just makes the race conditions surface easier)
[4:56] <stewori> jimbaker: Interesting idea with socket/select/ssl. I will test these when I find time. Still a port is always better, because of jit and platform support
[4:57] <stewori> JyNI is mainly intended to bridge the time until a port is made. And for extensions too complex to port.
[5:00] <stewori> Yesterday I visited a friend and we tried to reproduce the numpy-import on OSX, getting horrible problems. For some reason, Jython there wants to load all sorts of modules through JyNI from lib-dynload, e.g. grp and _io
[5:04] <stewori> grp and _io currently fail (at least on OSX), so the ported stuff is often better. Still figuring out why this issue appears to be OSX-specific
[5:15] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[5:20] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 244 seconds)
[6:11] * stewori (~stefan@5.146.128.93) Quit (Quit: Leaving.)
[6:16] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[6:21] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 258 seconds)
[7:17] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[7:21] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 240 seconds)
[8:17] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[8:22] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 276 seconds)
[8:45] * jimbaker (~jbaker@python/psf/jimbaker) Quit (Ping timeout: 264 seconds)
[9:18] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[9:23] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 276 seconds)
[10:19] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[10:23] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 240 seconds)
[11:20] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[11:24] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 240 seconds)
[12:20] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[12:24] * TomA (~TomA@c-68-32-46-223.hsd1.mi.comcast.net) has joined #jython
[12:25] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 264 seconds)
[13:21] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) has joined #jython
[13:26] * nickmbailey (~nickmbail@cpe-70-117-83-204.austin.res.rr.com) Quit (Ping timeout: 252 seconds)
[13:28] * patoff (~patoff@unaffiliated/patoff) has left #jython
[13:56] * opb94556 (~opb94556@193.62.223.241) has joined #jython
[13:57] * opb94556 (~opb94556@193.62.223.241) has left #jython
[14:14] * xemdetia (xemdetia@nat/ibm/x-lojgebcyjrntbgov) has joined #jython
[14:23] * fwierzbicki (~Adium@99-106-169-5.lightspeed.sntcca.sbcglobal.net) Quit (Quit: Leaving.)
[14:31] * nickmbailey (~nickmbail@aus.internal.datastax.com) has joined #jython
[15:00] * nickmbailey (~nickmbail@aus.internal.datastax.com) Quit (Quit: later)
[15:12] * jimbaker (~jbaker@python/psf/jimbaker) has joined #jython
[15:12] * ChanServ sets mode +o jimbaker
[15:16] * nickmbailey (~nickmbail@aus.internal.datastax.com) has joined #jython
[15:31] * TomA (~TomA@c-68-32-46-223.hsd1.mi.comcast.net) Quit (Remote host closed the connection)
[15:31] * TomA (~TomA@c-68-32-46-223.hsd1.mi.comcast.net) has joined #jython
[15:56] * fwierzbicki (~Adium@99-106-169-5.lightspeed.sntcca.sbcglobal.net) has joined #jython
[16:02] * jimbaker (~jbaker@python/psf/jimbaker) Quit (Quit: Textual IRC Client: www.textualapp.com)
[16:02] * jimbaker` (~jbaker@8.44.156.98) has joined #jython
[16:04] * jimbaker` is now known as jimbaker
[16:04] * jimbaker (~jbaker@8.44.156.98) Quit (Changing host)
[16:04] * jimbaker (~jbaker@python/psf/jimbaker) has joined #jython
[16:04] * ChanServ sets mode +o jimbaker
[16:10] * AndyBotwin (~Gustavo@177.204.192.227.dynamic.adsl.gvt.net.br) has joined #jython
[16:10] * AndyBotwin (~Gustavo@177.204.192.227.dynamic.adsl.gvt.net.br) Quit (Changing host)
[16:10] * AndyBotwin (~Gustavo@unaffiliated/andybotwin) has joined #jython
[16:11] * BillSussman (~Gustavo@177.204.192.227.dynamic.adsl.gvt.net.br) Quit (Ping timeout: 240 seconds)
[17:11] * nickmbailey (~nickmbail@aus.internal.datastax.com) Quit (Remote host closed the connection)
[17:38] * nickmbailey (~nickmbail@aus.internal.datastax.com) has joined #jython
[18:04] * jimbaker (~jbaker@python/psf/jimbaker) Quit (Ping timeout: 250 seconds)
[18:16] * jimbaker (~jbaker@8.44.156.98) has joined #jython
[18:16] * jimbaker (~jbaker@8.44.156.98) Quit (Changing host)
[18:16] * jimbaker (~jbaker@python/psf/jimbaker) has joined #jython
[18:16] * ChanServ sets mode +o jimbaker
[19:28] <jimbaker> nickmbailey, so i think http://bugs.jython.org/issue2508 is resolvable - there's enough in netty's api to get something reasonable
[19:28] * jimbaker (~jbaker@python/psf/jimbaker) has left #jython
[19:28] * jimbaker (~jbaker@python/psf/jimbaker) has joined #jython
[19:28] * ChanServ sets mode +o jimbaker
[19:30] <nickmbailey> awesome, thanks for taking a look at that
[19:30] <nickmbailey> what are your thoughts on the proposed solution of just blocking on the send future?
[19:30] <jimbaker> that's a bad idea
[19:31] <nickmbailey> bad correctness wise or performance wise
[19:31] <jimbaker> and violates c socket api semantics
[19:31] <nickmbailey> hmm
[19:31] <jimbaker> the send is immediate
[19:31] <jimbaker> if nonblocking
[19:31] <jimbaker> what you want to do is ensure data is sent before closing
[19:32] <jimbaker> i do think migrating to 4.1 is probably also helpful here, since we get some extra stuff to check about how many bytes can be written
[19:32] <jimbaker> presumably this is a straightforward change - i don't see anything in the 4.0 to 4.1 migration that should impact jython usage
[19:33] <nickmbailey> hmm
[19:33] <nickmbailey> i'm in a bit of a conundrum heh
[19:33] <jimbaker> obviously blocking semantics happen to be easier :)
[19:33] <nickmbailey> we're trying to get a fix for this out yesterday. since its something our customers are extremely likely to hit
[19:34] <nickmbailey> it looks easy but pushing a fix that involves a netty ugprade scares me a bit
[19:35] <jimbaker> nickmbailey, as a workaround, blocking on send can make sense for your case
[19:35] <nickmbailey> well we are testing a build with that fix now
[19:35] <nickmbailey> it isn't blowing up
[19:35] <jimbaker> but for 2.7.1 work, i think doing the upgrade will help us get the correct semantics
[19:35] <jimbaker> so both are probably valid
[19:35] * BillSussman (~Gustavo@177.204.192.227.dynamic.adsl.gvt.net.br) has joined #jython
[19:36] <nickmbailey> you didn't see anything in the 4.0 api that could enable doing it the *right* way though?
[19:36] <nickmbailey> yeah that could be fine
[19:36] <nickmbailey> we already have a custom build
[19:36] <jimbaker> well that additional bytes that can be written is new in 4.1
[19:36] <nickmbailey> and we can upgrade in our next major version
[19:36] <jimbaker> and that seems to be pretty important to communicate what can be sent with send
[19:36] <nickmbailey> but a netty upgrade in a patch release for us is pushing it
[19:36] <jimbaker> right
[19:37] <jimbaker> so i agree here
[19:37] * AndyBotwin (~Gustavo@unaffiliated/andybotwin) Quit (Ping timeout: 240 seconds)
[19:37] <jimbaker> with your decision
[19:37] <jimbaker> i think pushing this to a flush notification listener on close, which verifies every block has been flushed, gets the same behavior
[19:38] <jimbaker> but without blocking
[19:38] <jimbaker> at least as i understand the problem
[19:39] <nickmbailey> well we'd be happy to test out a build with that fix and a netty upgrade :) in addition to pushing out our own fix
[19:42] <jimbaker> nickmbailey, i certainly appreciate that
[19:42] <jimbaker> i will get the darn publication bug fixed, i know that, which is tying up the 2.7.1 release
[19:43] <jimbaker> i suspect that slow VMs test the race conditions better
[19:43] <jimbaker> but real testing will also be helpful
[19:44] <nickmbailey> definitely
[20:11] * nickmbailey (~nickmbail@aus.internal.datastax.com) Quit (Remote host closed the connection)
[20:12] * nickmbailey (~nickmbail@aus.internal.datastax.com) has joined #jython
[20:16] * nickmbailey (~nickmbail@aus.internal.datastax.com) Quit (Ping timeout: 258 seconds)
[20:34] * nickmbailey (~nickmbail@2605:6000:e8ce:3100:7c2f:d648:2218:d777) has joined #jython
[22:25] * stewori (~stefan@ip-95-222-164-13.hsi15.unitymediagroup.de) has joined #jython
[23:00] * AndyBotwin (~Gustavo@unaffiliated/andybotwin) has joined #jython
[23:00] * BillSussman (~Gustavo@177.204.192.227.dynamic.adsl.gvt.net.br) Quit (Ping timeout: 276 seconds)
[23:15] * TomA (~TomA@c-68-32-46-223.hsd1.mi.comcast.net) Quit (Remote host closed the connection)
[23:20] * xemdetia (xemdetia@nat/ibm/x-lojgebcyjrntbgov) Quit (Ping timeout: 250 seconds)
[23:39] * BillSussman (~Gustavo@177.204.192.227.dynamic.adsl.gvt.net.br) has joined #jython
[23:40] * AndyBotwin (~Gustavo@unaffiliated/andybotwin) Quit (Ping timeout: 264 seconds)

Index

These logs were automatically created by JythonLogBot on irc.freenode.net using a slightly modified version of the Java IRC LogBot.