#jython IRC Log

Index

IRC Log for 2015-01-08

Timestamps are in GMT/BST.

[0:09] * AndyBotwin (~botwin@unaffiliated/andybotwin) has joined #jython
[0:13] * newbie72 (cb0e4634@gateway/web/cgi-irc/kiwiirc.com/ip.203.14.70.52) has joined #jython
[0:23] * newbie72 (cb0e4634@gateway/web/cgi-irc/kiwiirc.com/ip.203.14.70.52) Quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
[2:14] * xemdetia (xemdetia@nat/ibm/x-xppvxltozhoibpok) Quit (Ping timeout: 264 seconds)
[2:36] * AndyBotwin (~botwin@unaffiliated/andybotwin) Quit (Read error: Connection reset by peer)
[6:11] * Oti (~ohumbel@adsl-178-38-248-15.adslplus.ch) Quit (Quit: Oti)
[8:22] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Excess Flood)
[8:23] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[9:54] * lanox (~lanox@111.67.21.197) Quit (Ping timeout: 245 seconds)
[9:55] * lanox (~lanox@111.67.21.197) has joined #jython
[11:40] * Paladiamors (~justin@pa9b391.tokynt01.ap.so-net.ne.jp) has joined #jython
[12:35] * lupulcelbun (~andrei@79.115.61.97) has joined #jython
[12:36] <lupulcelbun> Hello
[12:43] <lupulcelbun> I have a complex problem that boils down to a failed "import json" in jython2.5.4.rc1. I don't know if I am doing the import properly. Where in the documentation is this treated? I feel that I am missing something vital
[12:44] <agronholm> lupulcelbun: the json module is new to Python 2.6
[12:45] <agronholm> as such it's not included in Jython 2.5
[12:45] <agronholm> try Jython 2.7 instead
[12:47] <lupulcelbun> Trying again, thanks
[13:01] <lupulcelbun> Down to two errors. It seems that I am on the right track
[13:17] <peke> agronholm: jimbaker: The pull request has been tested on Windows and Linux and colleague tested an earlier version on OSX.
[13:19] <peke> Specifically, the pr was tested on Win7. I can also test on WinXP but don't have access to Win8. One reason I'd like the fix to be included in b4 is that it would get more testing on various environments.
[13:23] <agronholm> practical question -- is there a quick way to merge this?
[13:23] <agronholm> or do I have to resort to manually exporting to git diffs and importing to hg?
[13:25] <lupulcelbun> According to
[13:25] <lupulcelbun> http://www.jython.org/jythonbook/en/1.0/ModulesPackages.html#module-search-path-and-loading
[13:25] <lupulcelbun> I have to set CLASSPATH or PYTHONPATH to the modules.
[13:25] <lupulcelbun> This seems to be working but only half way through.
[13:25] <lupulcelbun> Initially I have both CLASSPATH and PYTHONPATH empty.
[13:25] <lupulcelbun> ~/opt/jython2.7b3/jython
[13:26] <lupulcelbun> Jython 2.7b3 (default:e81256215fb0, Aug 4 2014, 02:39:51)
[13:26] <lupulcelbun> [IBM J9 VM (IBM Corporation)] on java1.7.0
[13:26] <lupulcelbun> Type "help", "copyright", "credits" or "license" for more information.
[13:26] <lupulcelbun> >>> import RequestsLibrary
[13:26] <lupulcelbun> Traceback (most recent call last):
[13:26] <lupulcelbun> File "<stdin>", line 1, in <module>
[13:26] <lupulcelbun> ImportError: No module named RequestsLibrary
[13:26] <lupulcelbun> >>>
[13:26] <lupulcelbun> I fix the paths and try again.
[13:26] <lupulcelbun> 0 ??? andrei@oc2435728566 ~/opt/jython2.7b3 $ export PYTHONPATH=/usr/lib/python2.6/site-packages/
[13:26] <lupulcelbun> 0 ??? andrei@oc2435728566 ~/opt/jython2.7b3 $ export CLASSSPATH=/usr/lib/python2.6/site-packages/
[13:26] <lupulcelbun> 0 ??? andrei@oc2435728566 ~/opt/jython2.7b3 $ ./jython
[13:26] <lupulcelbun> Jython 2.7b3 (default:e81256215fb0, Aug 4 2014, 02:39:51)
[13:26] <lupulcelbun> [IBM J9 VM (IBM Corporation)] on java1.7.0
[13:26] <lupulcelbun> Type "help", "copyright", "credits" or "license" for more information.
[13:26] <lupulcelbun> >>> import RequestsLibrary
[13:26] <lupulcelbun> Traceback (most recent call last):
[13:26] <lupulcelbun> File "<stdin>", line 1, in <module>
[13:26] <lupulcelbun> File "__pyclasspath__/RequestsLibrary/__init__.py", line 1, in <module>
[13:26] <lupulcelbun> File "__pyclasspath__/RequestsLibrary/keywords.py", line 3, in <module>
[13:26] <lupulcelbun> File "__pyclasspath__/vcr/__init__.py", line 2, in <module>
[13:26] <lupulcelbun> File "__pyclasspath__/vcr/config.py", line 6, in <module>
[13:26] <lupulcelbun> File "__pyclasspath__/vcr/cassette.py", line 5, in <module>
[13:26] <lupulcelbun> ImportError: No module named wrapt
[13:26] <agronholm> lupulcelbun: you mistyped CLASSPATH as "CLASSSPATH"
[13:26] <lupulcelbun> >>> import sys
[13:26] <lupulcelbun> >>> sys.path
[13:26] <lupulcelbun> ['', '/home/andrei/opt/jython2.7b3/Lib', '__classpath__', '__pyclasspath__/', '/home/andrei/opt/jython2.7b3/Lib/site-packages']
[13:26] <lupulcelbun> >>>
[13:27] <agronholm> lupulcelbun: PLEASE do not spam the channel
[13:27] <lupulcelbun> This is most strange given that wrapt is side by side with RequestsLibrary
[13:27] <lupulcelbun> I am doing something horribly wrong and I can't see what.
[13:27] <agronholm> there are pastebin services for this purpose
[13:28] <agronholm> lupulcelbun: besides, as the traceback indicates, the RequestsLibrary package was found by the importer
[13:28] <agronholm> it was that "wrapt" module that was missing
[13:28] <peke> agronholm: I think the easiest way to merge the pr is downloading a diff from GitHub and applying it manually. Cc: jimbaker
[13:29] <lupulcelbun> wrapt is there
[13:29] <peke> agronholm: https://github.com/jythontools/jython/pull/12.diff
[13:29] <lupulcelbun> http://pastebin.com/9PGVYBSp
[13:29] <agronholm> lupulcelbun: according to the error message, it's not
[13:30] <agronholm> lupulcelbun: did you fix the CLASSPATH problem I mentioned?
[13:33] <lupulcelbun> http://pastebin.com/mL6qhGFj
[13:34] <lupulcelbun> fixed the path, same behavior
[13:34] <agronholm> try unsetting CLASSPATH entirely
[13:34] <agronholm> can you still import anything from site-packages?
[13:36] <lupulcelbun> apparently not
[13:36] <lupulcelbun> http://pastebin.com/6RXVBPxV
[13:36] <agronholm> try removing the / from the end of the path
[13:40] <peke> lupulcelbun: You are using https://github.com/bulkan/robotframework-requests, right?
[13:41] <peke> It seems that the latest code doesn't use vcr module anymore. Perhaps you should clone and install it directly from the version control and see does it work.
[13:41] <lupulcelbun> peke: I took it with pip, I think that they are identical.
[13:41] <lupulcelbun> afronholm: http://pastebin.com/6jpLZ98R
[13:46] <peke> lupulcelbun: No, vcr was definitely removed some time ago: https://github.com/bulkan/robotframework-requests/pull/47
[13:49] <lupulcelbun> I have update the package, for all intents and purposes what is in git is pulled via pip as well.
[13:50] <lupulcelbun> It seems to be working.
[13:51] <lupulcelbun> good catch peke!
[14:34] * lupulcelbun (~andrei@79.115.61.97) Quit (Ping timeout: 264 seconds)
[14:37] <agronholm> has hg.python.org's SSH key changed recently?
[14:37] <agronholm> I get a warning
[14:38] <agronholm> can someone verify its RSA fingerprint? I get a0:12:52:50:4a:4b:db:43:ac:65:26:b6:6f:0a:f7:b8
[14:52] * enebo (~enebo@c-75-73-8-169.hsd1.mn.comcast.net) has joined #jython
[14:58] <agronholm> you can get that by running ssh-keygen -l -F hg.python.org
[14:58] * xemdetia (xemdetia@nat/ibm/x-fistppaczpulcsyo) has joined #jython
[14:59] <paolo> agronholm: to me it gives empty output
[14:59] <agronholm> that only applies to people who have already accessed it via ssh
[15:00] <paolo> can i help in any way?
[15:00] <agronholm> I don't think so :/
[15:01] <agronholm> I recall there was a server change in the recent months
[15:02] <paolo> agronholm: https://mail.python.org/pipermail/python-committers/2014-September/003146.html
[15:03] <agronholm> paolo: ah thanks
[15:03] <agronholm> yes, the fingerprint matches
[15:03] <paolo> np
[15:40] <jimbaker> agronholm, sounds good about the commits
[15:40] <jimbaker> at this point, i just use the diffs for PRs, there really is no other workable way given how hg.python.org is set up
[15:41] <jimbaker> (that the PR process is so difficult is just one more bit of friction we face)
[15:44] * Paladiamors (~justin@pa9b391.tokynt01.ap.so-net.ne.jp) Quit (Ping timeout: 264 seconds)
[16:55] * xemdetia_ (xemdetia@nat/ibm/x-muomsejcqvkkwzor) has joined #jython
[16:58] * xemdetia_ (xemdetia@nat/ibm/x-muomsejcqvkkwzor) Quit (Remote host closed the connection)
[16:58] * xemdetia_ (xemdetia@nat/ibm/x-fmbblbebomhbowuh) has joined #jython
[16:58] * xemdetia (xemdetia@nat/ibm/x-fistppaczpulcsyo) Quit (Ping timeout: 265 seconds)
[17:47] * xemdetia_ (xemdetia@nat/ibm/x-fmbblbebomhbowuh) Quit (Ping timeout: 264 seconds)
[17:51] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 264 seconds)
[18:33] * xemdetia (xemdetia@nat/ibm/x-rrdpfyyakujcissz) has joined #jython
[18:55] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[19:56] * AndyBotwin (~botwin@177.40.28.139) has joined #jython
[19:56] * AndyBotwin (~botwin@177.40.28.139) Quit (Changing host)
[19:56] * AndyBotwin (~botwin@unaffiliated/andybotwin) has joined #jython
[20:06] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 244 seconds)
[20:08] * AndyBotwin (~botwin@unaffiliated/andybotwin) Quit (Ping timeout: 255 seconds)
[20:28] * AndyBotwin (~botwin@unaffiliated/andybotwin) has joined #jython
[20:38] <peke> agronholm: thanks for merging my Popen.pid fixes! I closed the PR on GitHub.
[20:42] <peke> jimbaker: fwierzbicki: others: Is there anything preventing moving Jython to GitHub fully? IronPython moved there recently, and there has been discussion on python-dev about moving at leasts some CPython repos either there or BitBucket.
[20:43] <peke> With CPython there's so much infrastructure that moving everything would be a huge task, and then there are people who object using a service that is not implemented using Python and/or is closed source. I don't see anything holding back Jython.
[20:44] <jimbaker> we are certainly not dogmatic, but it seemed like a good idea to follow cpython in terms of its infrastructure
[20:44] <peke> My experience from moving Robot Framework from Google Code to GitHub last summer has been extremely positive. We've got a lot more contributions afterwards.
[20:45] <jimbaker> right, there's certainly that
[20:45] <jimbaker> i know there's been discussion on python-dev about improving infrastructure, but it seemed that it was unlikely to get very far - lots of bikeshedding potential!
[20:46] <peke> Yeah, a lot of bikeshedding already on python-dev. Apparently that's going to continue on PyCon.
[20:46] <jimbaker> it would be a big move. it might make sense. pycon is also the right time to decide on any change
[20:46] <jimbaker> peke, are you planning to attend?
[20:48] <peke> jimbaker: Not this year. Kids are still a bit too small to allow extra travel. Definitely want to come some year.
[20:48] <jimbaker> peke, makes sense
[20:49] <jimbaker> so there's no question that a streamlined process would be great. i feel like the current process is extraordinarily cramped and time consuming
[20:49] <jimbaker> in terms of commit workflow
[20:49] <jimbaker> and that's from my perspective! i can only imagine people trying to submit patches
[20:50] <peke> Agreed. Is there anything else preventing Jython to move to GitHub (or BitBucket) than following CPython?
[20:51] <jimbaker> peke, none
[20:51] <peke> Personally I don't see that as a very strong argument. Jython needs are so much simpler than CPython needs. CPython is also so much bigger and more well known that it is likely to attract contributors even if the contribution process isn't that modern.
[20:52] <jimbaker> ideally we could port all open bugs to github, retaining bug numbers
[20:52] <peke> That's what we did when we moved RF.
[20:52] <jimbaker> the other thing is just get all PRs tested against at least windows and some unix variant upon the pull request
[20:52] <jimbaker> and of course subsequent commits
[20:53] <jimbaker> that would be a huge time saver
[20:54] <peke> Yeah, it is. And it's not that complicated to set up.
[20:55] <jimbaker> peke, hah! you might be volunteering before you know it :)
[20:55] <peke> I'm definitely willing to help.
[20:57] <peke> Moving code so that history is preserved is pretty easy. As a bonus it's possible to remove big commits (jars and other compiled files) to make a repository smaller.
[20:57] <jimbaker> sounds good, that's usually what it takes to go beyond, it would be nice... to actually getting it done
[20:57] <jimbaker> peke, yeah, mavenification would also be a good thing
[20:58] <jimbaker> our build process in build.xml is good for 2007, needs to be updated
[20:58] <peke> Cannot help much with maven. But I agree it would likely be a good idea.
[20:59] <peke> With issues the problem would be retaining ids of the issues originating from Source Forge.
[20:59] <jimbaker> about the only thing is, i can tolerate cramped for getting 2.7.0 out. maybe others can too. but after that, we need to make some changes. i think github is a great idea. i would love to completely abandon the completely broken aspects of hg for anything large
[21:00] <peke> I agree getting 2.7 out is more important.
[21:01] <jimbaker> fwierzbicki, pjenvey, any thoughts here? also, i think it's about time to push out the beta 4
[21:03] <jimbaker> right now, i'm working on set/dict comprehensions rewritten in code compiler so they don't link the iteration variable; and a few misc things for stabilizing regrtest. but nothing worth postponing the beta
[21:03] <jimbaker> leak the iteration variable
[21:04] <peke> I could actually ask colleagues with whom we moved Robot Framework to GitHub would they be willing to help. If they are interested, and there's a consensus that GitHub move is a good idea, we could possibly have a local sprint trying to get it done.
[21:05] <peke> The move itself should be automated as much as possible. The actual work is writing and testing those automation scripts. Testing obviously can be done in temporary repository.
[21:06] <pjenvey> i'm pretty much for moving from bitbucket to github, bitbucket's been nice to us and supports git but github might increase contributor activity or make the workflow easier for contributors
[21:06] <pjenvey> converting the repo should be really easy
[21:06] <pjenvey> moving the issues would be nice but wouldn't have to happen immediately, really
[21:07] <agronholm> jimbaker: what aspects of hg do you see as completely broken?
[21:07] <pjenvey> we piggybacked onto cpython/psf infrastructure but github lets us piggyback onto theirs, i was also hoping cpython devs might pay a little more attention to jython having the repo there
[21:07] <pjenvey> but they don't at all =]
[21:07] <jimbaker> pjenvey, no they don't :)
[21:08] <pjenvey> someone would have to do the work of moving it though =]
[21:09] <jimbaker> agronholm, anything with bookmarks and branches basically cannot be merged
[21:09] <agronholm> I'm confused
[21:09] <jimbaker> because it violates the hg.python.org gate
[21:09] <agronholm> gate?
[21:10] <jimbaker> there is a postcommit hook in hg.python.org that rejects any changesets that reference bookmarks/branches
[21:11] <agronholm> sounds more like a problem of that thing rather than a flaw in hg itself
[21:11] <jimbaker> so it means that most anything submitted through bitbucket PRs has to be committed by extracting the changeset as a patch. which is undocumented, but there's a sort-of-rest API that can be used
[21:11] <jimbaker> to extract the diff
[21:12] <jimbaker> agronholm, that's why i call the process *cramped*
[21:12] <agronholm> whatever the case, a system where PRs can be directly accepted would be great
[21:13] <jimbaker> i can make the process work, but it's tedious, manual (although presumably scriptable, but not easily, see "sort-of-REST API" above)
[21:13] <jimbaker> agronholm, exactly
[21:13] <jimbaker> and peke has basically volunteered to do the work ;) well i'm sure it should be done more equitably than that
[21:15] <jimbaker> fwierzbicki, do you own https://github.com/jython ?
[21:15] <pjenvey> was just going to ask the exact same question jim =] noticed he's the only follower
[21:15] <jimbaker> i know we have a mirror at https://github.com/jythontools/jython
[21:16] <jimbaker> pjenvey, yeah, otherwise we know people at github who can probably give it to us, if it's just some cybersquatting
[21:21] <peke> I agree with pjenvey that BitBucket is nice but GitHub has won.
[21:23] <peke> jimbaker: pjenvey: How good is https://github.com/jythontools/jython?
[21:24] <jimbaker> peke, it certainly mirrors better than the bitbucket mirror
[21:24] <jimbaker> peke, we have that mirror go down at least twice, once for a least a couple of months
[21:24] <peke> Would moving the code to GitHub require much more than start using that as the master?
[21:25] <peke> Perhaps after moving it to https://github.com/jython.
[21:25] <jimbaker> peke, right, i think that the nature of dvcs means it can be done at any time - it becomes master when we stop making it a mirror
[21:25] <peke> Exactly.
[21:26] <peke> Moving the master makes it possible to clean up the repository, though.
[21:27] <jimbaker> peke, right. and no risk, other than having to clean up the clean up
[21:27] <peke> Could make sense if there are big jars or other compiled files committed at some point.
[21:27] <jimbaker> the only big jar we have is icu4j, at about 10M
[21:27] <peke> Yeah, especially if the idea is to move the repo under jython account.
[21:28] <jimbaker> which makes the most sense - it's what people would look for
[21:28] <peke> If history is edited after the repository is taken into use, everyone needs to re-clone. Not good.
[21:29] <jimbaker> peke, got it. so we are good as far as that is concerned. fresh slate and all
[21:29] <peke> If it's moved under jython account, there's a good change to do all kind of clean-up. Removing big commits is one possibility, updating contributor info to match their GitHub user names is another.
[21:30] <peke> Yes.
[21:30] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Quit: Konversation terminated!)
[21:32] <jimbaker> peke, presumably contributor info can dangle, given that this project is about 17 years old or so
[21:33] <peke> jimbaker: Pro Git book both explains why removing big objects may be a good idea and shows how to do it: http://git-scm.com/book/en/v2/Git-Internals-Maintenance-and-Data-Recovery#Removing-Objects
[21:36] <jimbaker> peke, ok, so we will want to do in concert with adding maven support
[21:37] <peke> With Maven support you mean not having dependencies committed to version control but automatically downloaded by Maven?
[21:40] <peke> Anyway, if we agree moving to GitHub is a good idea, moving source code isn't a blocker. It's already there, but there's a change to clean it up if we want to.
[21:42] <peke> That cleaning up is probably better done by someone who knows the source code better than I. I can look at moving issues from the current tracker to GitHub, though.
[21:42] <jimbaker> peke, right, i want to get those dependencies out and managed using standard pom support, which we can set up in conjunction with our build.xml
[21:43] <jimbaker> peke, i think we can just have extlibs populated by a maven step - probably dist/extlibs. but i need to work on my maven fu first
[21:44] <chrisseaton> lots of activity in Jython these days it seems - I saw you listed investigating Truffle via Zippy as an option on your yearly plan - has anyone looked at that yet
[21:48] <jimbaker> chrisseaton, i have not followed up on that. i know the zippy folks have made some progress, but we have been deep in basic compatibility issues around standard library and stuff like import semantics
[21:48] <jimbaker> stuff that zippy should be orthogonal too :)
[21:48] <chrisseaton> if anyone is going to FOSDEM I'll be there and can let you know how we've found integrating Truffle into JRuby
[21:49] <jimbaker> chrisseaton, but it sounds like a much more fun way to spend time on jython
[21:49] <chrisseaton> you will probably have identical challenges if you try it
[21:49] <jimbaker> darn, FOSDEM is a conference i would love to go to. i assume you plan to be at the JVM lang summit this summer?
[21:50] <chrisseaton> me or someone from the team - we've got three people working on JRuby+Truffle full-time now, with a fourth on the way
[21:51] * AndyBotwin (~botwin@unaffiliated/andybotwin) Quit (Quit: Leaving)
[21:51] * AndyBotwin (~Kevin@unaffiliated/andybotwin) has joined #jython
[21:53] <fwierzbicki> Moving to github has a lot going for it - I think we would get more contributors
[21:54] * robbyoconnor (~wakawaka@cpe-68-174-58-54.hvc.res.rr.com) has joined #jython
[21:55] * robbyoconnor (~wakawaka@cpe-68-174-58-54.hvc.res.rr.com) Quit (Changing host)
[21:55] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[21:55] <fwierzbicki> I do not own github.com/jython - at least I don't remember registering it and my typical password patterns don't work
[21:55] <offbyone> If I want to subclass a Java class in Jython, where the Java class has an overloaded constructor, how would I do that and still be able to call the superclass constructor in __init__?
[22:00] <jimbaker> chrisseaton, awesome, that's 3 more full-time people than are working on jython as a whole. i will be going back to parttime status my self pretty soon
[22:00] <jimbaker> but that's the way it works out
[22:01] <jimbaker> will be moving to focusing on kafka and spark and that ecosystem real soon now
[22:05] <jimbaker> (but i believe jython will be very helpful in this work. scala as well of course)
[22:06] <jimbaker> offbyone, you should be able select the proper overloaded constructor. sometimes that requires wrapping one or more of the args so it doesn't get improperly coerced by jython's matching algorithm
[22:06] <jimbaker> which is weak, naive, and usually works
[22:07] <jimbaker> ;)
[22:08] <jimbaker> eg you might have to do something like Foo.__init__(java.lang.Boolean(x), java.lang.Integer(y), z)
[22:09] <jimbaker> offbyone, i assume characteristic is working fine for you now?
[22:09] * Paladiamors (~justin@pa9b391.tokynt01.ap.so-net.ne.jp) has joined #jython
[22:10] <offbyone> I expect so
[22:10] <offbyone> I got swamped with handoff work :)
[22:10] <offbyone> and have stalled on that initiative
[22:10] <offbyone> Thanks
[22:12] <jimbaker> offbyone, certainly keep us posted, that's a pretty cool test project for jython - or any python implementation - but its own testing seems to be comprehensive
[22:19] * siel (~siel@unaffiliated/motley) Quit (Quit: leaving)
[22:21] * siel (~siel@unaffiliated/motley) has joined #jython
[22:24] * siel (~siel@unaffiliated/motley) Quit (Read error: Connection reset by peer)
[22:24] * siel (~siel@2605:6400:10:c07c:4231:4321:2341:2031) has joined #jython
[22:25] * siel (~siel@2605:6400:10:c07c:4231:4321:2341:2031) Quit (Changing host)
[22:25] * siel (~siel@unaffiliated/motley) has joined #jython
[22:40] * Paladiamors (~justin@pa9b391.tokynt01.ap.so-net.ne.jp) Quit (Ping timeout: 245 seconds)
[23:19] <offbyone> jimbaker, another odd Jython question: if I have a Java class with an instance child class (class Foo { class Bar {} })
[23:19] <offbyone> how would I subclass class Bar in a subclass of Foo?
[23:19] <offbyone> (hell, I'm not even sure how to formulate that question, nor how to answer it even for Java :) )
[23:19] <jimbaker> offbyone, that's actually static class Bar {}
[23:19] <offbyone> oh, I wish...
[23:19] <offbyone> it's not
[23:20] <jimbaker> but just by Foo.Bar()
[23:21] <jimbaker> offbyone, oh right, this obscure case
[23:21] <offbyone> let me whip up a gist
[23:21] <offbyone> one sec.
[23:21] <jimbaker> offbyone, that would be handy
[23:25] <offbyone> here
[23:25] <offbyone> https://gist.github.com/anonymous/a52759e840a4e5f722da
[23:25] <offbyone> I think that describes what I'm aiming for fairly well
[23:25] <offbyone> I'd show you the code itself, but it's one of those NDA things :/
[23:34] * AndyBotwin (~Kevin@unaffiliated/andybotwin) Quit (Quit: Leaving)
[23:38] <offbyone> Did that gist make sense?

Index

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