#jython IRC Log

Index

IRC Log for 2015-03-09

Timestamps are in GMT/BST.

[0:20] * xemdetia_ (~xemdetia@pool-96-230-53-83.bstnma.fios.verizon.net) Quit (Ping timeout: 272 seconds)
[5:13] * nickmbailey (~nickmbail@nickmbailey.com) Quit (Quit: Coyote finally caught me)
[8:50] * ebarrett (~edd@host-78-148-181-145.as13285.net) has joined #jython
[8:57] * mbooth (~mbooth@cpc11-shef10-2-0-cust659.barn.cable.virginm.net) has joined #jython
[8:57] * mbooth (~mbooth@cpc11-shef10-2-0-cust659.barn.cable.virginm.net) Quit (Changing host)
[8:57] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[9:04] * dustinm (~dustinm@105.ip-167-114-152.net) Quit (*.net *.split)
[9:04] * dustinm (~dustinm@2607:5300:100:200::160d) has joined #jython
[9:48] * Arfrever (~Arfrever@apache/committer/Arfrever) has joined #jython
[13:25] * enebo (~enebo@c-75-73-8-169.hsd1.mn.comcast.net) has joined #jython
[13:44] * xemdetia (xemdetia@nat/ibm/x-skvachskozzmtjdu) has joined #jython
[14:05] * nickmbailey (~nickmbail@nickmbailey.com) has joined #jython
[14:18] * mbooth (~mbooth@redhat/mbooth) Quit (Remote host closed the connection)
[14:27] * mbooth (~mbooth@cpc11-shef10-2-0-cust659.barn.cable.virginm.net) has joined #jython
[14:27] * mbooth (~mbooth@cpc11-shef10-2-0-cust659.barn.cable.virginm.net) Quit (Changing host)
[14:27] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[17:56] <jimbaker> agronholm, i filed a couple of bugs related to the web start email thread in jython-dev
[17:57] <jimbaker> so looking at the backlog, it looks you are suggesting that simply replacing jython-dev.jar with jython.jar in pycompile should have that work (and fix http://bugs.jython.org/issue2282). which makes sense. just need to add something to continue the support for jython-dev.jar, by testing for its existence
[17:58] <agronholm> jimbaker: I'm not sure I understand that issue
[17:58] <jimbaker> agronholm, you mean as i reported it?
[17:58] <agronholm> yeah
[17:59] <jimbaker> my understanding certainly may be incomplete! :)
[17:59] <agronholm> this has nothing to do with the problem I was talking about
[17:59] <jimbaker> it's not an issue i see. darn packaging ;)
[17:59] <agronholm> (afaik)
[18:00] <agronholm> the actual problem is the weird class loading error that I get when using the non-dev jython jar with the ant python compile task
[18:00] <agronholm> it makes no sense to me at all
[18:00] <agronholm> and it doesn't happen with the dev jar
[18:02] <agronholm> jimbaker: you'll see it yourself if you replace jython-dev.jar with the prepackaged jython.jar in the buildfile
[18:02] <agronholm> and run ant pycompile
[18:02] <jimbaker> agronholm, ok. i will give it a try
[18:04] <jimbaker> agronholm, i suspect that this pycompile task stopped working because of issues like this, was patched to be revisited later, and now we are finally revisiting :)
[18:04] <agronholm> oh?
[18:04] <agronholm> it did work before, that much I know
[18:04] <jimbaker> it's the most reasonable hypothesis
[18:05] <agronholm> I ran many builds on jython 2.5.3
[18:05] <agronholm> but suddenly it stopped working and started giving me class loading errors
[18:08] <jimbaker> i don't see class loading issues, but there's definitely something wrong in the build.xml file. i get /Users/jbaker/jythondev/jython27/dist/build.xml:522: destination directory "/Users/jbaker/jythondev/jython27/dist/tests/data/initializer" does not exist or is not a directory, fwiw - just noise i assume
[18:08] <agronholm> show me how you modified your build.xml
[18:12] <jimbaker> agronholm, <classpath path="${dist.dir}/jython.jar" />
[18:12] <agronholm> and you removed the entry to jython-dev.jar?
[18:15] <agronholm> can you get the pycompile step to work after removing jython-dev.jar from the task's classpath and replacing it with the stock jython.jar?
[18:15] <agronholm> I could not
[18:25] <paolo> jimbaker, agronholm: i hope it's not the case, but might the problem be related to this? http://bugs.jython.org/issue1672
[18:25] <paolo> especially
[18:25] <paolo> - cfile = fullname + (__debug__ and 'c' or 'o')
[18:25] <paolo> + cfile = fullname.replace('.py', '$py.class')
[18:25] <agronholm> no
[18:25] <agronholm> I've never had problems with -m compileall
[18:26] <agronholm> it's the ant task that's the prob
[18:26] <paolo> Compileall does not compile if .pyc file exists
[18:26] <agronholm> and the class loader error it gives
[18:26] <paolo> ah ok, sorry for the noise.
[18:36] <jimbaker> paolo, i think we are all set with bug fix for http://bugs.jython.org/issue1672
[18:36] <jimbaker> it's essentially a bootstrapping issue - we need a jython to run this ant task
[18:38] <jimbaker> anyway, i will turn my laser eyes onto this problem ;) - https://www.google.com/search?q=laser+eyes&es_sm=119&source=lnms&tbm=isch&sa=X
[18:40] <jimbaker> this urgent bug has been seen for a while, just has been worked around. time to excise it
[19:14] * xemdetia_ (xemdetia@nat/ibm/x-qqjrersiqlcvipgy) has joined #jython
[19:17] * xemdetia (xemdetia@nat/ibm/x-skvachskozzmtjdu) Quit (Ping timeout: 272 seconds)
[20:34] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 240 seconds)
[20:47] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[20:55] <nickmbailey> how long do we think we'll let beta4 stew before an rc? just until all 2.7 tickets marked urgent get fixed?
[21:59] <jimbaker> nickmbailey, i think that's pretty reasonable
[22:02] <jimbaker> the urgent tickets hit differences in platform (windows, now resolved linux); in packaging; and a fairly estoteric case (__dict__ support in __slots__, for proxying)
[22:02] <jimbaker> these are just the sort of issues we see as we complete a release cycle
[22:02] <jimbaker> so i feel comfortable. but famous last words? ;)
[22:02] <nickmbailey> haha fair enough
[22:03] <nickmbailey> yeah just wondering if the release cadence is planned to continue speeding up and sounds like it is
[22:04] <nickmbailey> having some success running the latest beta with twisted and considering jython, but "beta" is always scare
[22:04] <jimbaker> nickmbailey, we are close. i also really want to go to pycon and not have to say in my talk on jython, hey, still work on releasing it :)
[22:04] <nickmbailey> i guess i just assumed we'd go from beta to rc, is that the plan or is there an alpha in there
[22:04] <jimbaker> nickmbailey, no alphas until 2.7.1 :)
[22:05] <jimbaker> rc1 is our expectation of what should be a solid 2.7.0
[22:05] <nickmbailey> heh, good to know we have a goal of before mid april then
[22:05] <jimbaker> and i'm glad to hear about twisted. i think you brought this up earlier
[22:05] <jimbaker> you did some workaround, right?
[22:05] <nickmbailey> yeah i found a bug in beta3 :)
[22:06] <nickmbailey> yeah there are some issues i had to fix twisted side but those are pretty minor
[22:06] <jimbaker> nickmbailey, my memories of beta 3 have completely vanished i think :)
[22:06] <jimbaker> it's been too much work. but hey
[22:07] <jimbaker> so if we can resolve those twisted issues in 2.7.1, that would be awesome. but workarounds are not necessarily bad
[22:07] <jimbaker> i do not expect perfect for rc1. just solid and good
[22:08] <nickmbailey> definitely. well i'm happy to file bugs for the things we've found and go from there, but yeah i have no problems using the workarounds so far
[22:08] <nickmbailey> it's things like the way that twisted marks modules as deprecated
[22:09] <jimbaker> nickmbailey, sounds good, please file those bugs, it will be very helpful
[22:09] <nickmbailey> or twisted not liking the $py.class extension
[22:09] <jimbaker> that's twisted's fault. we are not going to change that :)
[22:09] <agronholm> twisted is so old it does things you wouldn't do today
[22:09] <agronholm> like how it manages plugins
[22:10] <jimbaker> it would be rather frowned upon if we changed $py.class to .pyc. although to be honest, $py.class is as meaningful - one cannot simply run those $py.class files without the jython runtime
[22:10] <nickmbailey> yeah… i didn't make the decision to use twisted, would go back in time and change it if i could
[22:12] <jimbaker> well, jython is a bit older than twisted - 17 years old! - but we also did do a few backwards breaking things for what will be 2.7.0
[22:37] <jimbaker> peke, speaking of windows bugs - can we close out the alt-gr bug? http://bugs.jython.org/issue2275
[22:44] * enebo (~enebo@c-75-73-8-169.hsd1.mn.comcast.net) Quit (Quit: enebo)
[22:59] * Arfrever (~Arfrever@apache/committer/Arfrever) Quit (Ping timeout: 245 seconds)
[23:18] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 256 seconds)
[23:30] * mbooth (~mbooth@cpc11-shef10-2-0-cust659.barn.cable.virginm.net) has joined #jython
[23:30] * mbooth (~mbooth@cpc11-shef10-2-0-cust659.barn.cable.virginm.net) Quit (Changing host)
[23:30] * mbooth (~mbooth@redhat/mbooth) has joined #jython

Index

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