#jython IRC Log (v0.9)

Index

IRC Log for 2011-05-04

Timestamps are in GMT/BST.

[0:04] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Quit: Ex-Chat)
[0:10] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[1:51] * plankton (~kvirc@201-14-52-39.cscgo701.dsl.brasiltelecom.net.br) has joined #jython
[1:51] <plankton> :)
[2:09] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) Quit (Read error: Connection reset by peer)
[2:10] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) has joined #jython
[2:26] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has joined #jython
[2:28] * gqlewis_ (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has joined #jython
[2:31] * plankton (~kvirc@201-14-52-39.cscgo701.dsl.brasiltelecom.net.br) Quit (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
[2:45] * thobe (~Adium@83.249.235.76) Quit (Read error: Connection reset by peer)
[3:10] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[3:11] * lopex (lopex@84-10-213-252.dynamic.chello.pl) Quit ()
[3:21] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) Quit (Ping timeout: 252 seconds)
[3:49] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) Quit (Quit: gqlewis)
[3:50] * gqlewis_ (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) Quit (Remote host closed the connection)
[3:50] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Read error: Operation timed out)
[5:15] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[6:57] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has joined #jython
[7:15] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Ping timeout: 240 seconds)
[7:16] * lahwran (~lahwran@unaffiliated/lahwran) has joined #jython
[7:16] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[7:27] * kilon (~kilon@adsl-30.109.242.89.tellas.gr) has joined #jython
[7:37] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Ping timeout: 240 seconds)
[7:39] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[7:50] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) has joined #jython
[8:09] * kilon is now known as kilon_away
[8:11] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) Quit (Quit: jabley)
[8:36] * jabley (~jabley@78-105-125-211.zone3.bethere.co.uk) has joined #jython
[9:43] * kilon_away is now known as kilon
[11:08] * juneau001 (~juneau@131.225.24.148) has joined #jython
[11:36] * jabley (~jabley@78-105-125-211.zone3.bethere.co.uk) Quit (Quit: jabley)
[11:59] * jabley (~jabley@78-105-125-211.zone3.bethere.co.uk) has joined #jython
[12:45] * lopex (531346b2@gateway/web/freenode/ip.83.19.70.178) has joined #jython
[13:16] * pr3d4t0r (~zhenya01@varenka.cime.net) Quit (Quit: WeeChat 0.3.0)
[13:46] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) has joined #jython
[13:54] * mhagger (~mhagger@ssh.berlin.jpk.com) has joined #jython
[14:11] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 260 seconds)
[14:14] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[14:28] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 240 seconds)
[14:34] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[14:39] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 246 seconds)
[14:55] * epoitras (~eddie@69-196-138-130.dsl.teksavvy.com) has joined #jython
[15:03] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) Quit (Ping timeout: 260 seconds)
[15:06] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) has joined #jython
[15:08] <mhagger> We've had strange ImportErrors when we commingle *.class files and *.py files in the same directories, when some of the intermediate paths have no *.class files.
[15:10] <mhagger> For example, if I create the following files then "import a.b.Foo" fails: a/b/Foo.java a/__init__.py a/b/__init__.py a/b/Bar.py
[15:10] <mhagger> (Of course this is after compiling a/b/Foo.java to a/b/Foo.class)
[15:11] <mhagger> If I add a/Foobar.java, then the problem goes away and I can import both Java and Python files.
[15:12] <mhagger> And strangely, if I delete one or both of the __init__.py files, then I can suddenly import the Java file (though now I can't import the Python files because they are not seen to be part of a Python package).
[15:13] <mhagger> It seems like Jython is deciding that the directory is either a Java package or a Python package but gets confused if it is both.
[15:13] <mhagger> Does this ring any bells?
[15:14] <mhagger> The problem is the same with Jython 2.2.1 and Jython 2.5.1
[15:24] * kilon (~kilon@adsl-30.109.242.89.tellas.gr) Quit (Quit: Leaving)
[15:25] * ohumbel (5390f2dc@gateway/web/freenode/ip.83.144.242.220) Quit (Ping timeout: 252 seconds)
[15:27] <agronholm> mhagger: rings bells
[15:27] <agronholm> mhagger: you can't put both java classes and python modules in the same directory
[15:28] <mhagger> Is there a fundamental reason for this limitation or is it a bug?
[15:28] <agronholm> apparently it's hard to support
[15:29] <agronholm> not a trivial problem
[15:29] <mhagger> Um, but I *can* put java classes and python modules in the same directories, *provided* all of the directories contain java classes.
[15:30] <mhagger> In fact we have been using this workaround for a while here
[15:38] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[15:40] * lopex (531346b2@gateway/web/freenode/ip.83.19.70.178) Quit ()
[15:41] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Quit: Ex-Chat)
[15:43] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[15:50] <agronholm> but there are problems loading classes from the same directory where you have python modules
[15:50] <agronholm> I wish I could find the exact posts explaining why
[15:52] <agronholm> I tried doing the same before, and failed
[15:52] * thobe1 (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[15:53] <jimbaker> agronholm, there was a bug on this
[15:53] <jimbaker> it may have been fixed. i may have in fact fixed it ;)
[15:53] <agronholm> I see
[15:53] <jimbaker> but i cannot recall its status right now
[15:54] <mhagger> I think I see the relevant code in PathPackageManager.PackageExistsFileFilter
[15:54] <jimbaker> i'm pretty certain it was fixed, because it was behavior that was allowed in the past
[15:54] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Ping timeout: 258 seconds)
[15:54] <jimbaker> with jython 2.2 and earlier and there was no good reason not to allow it
[15:54] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 276 seconds)
[15:55] <mhagger> I observe the same problem in jython 2.2.1
[15:58] <mhagger> jimbaker: Maybe I'm talking about a corner case of a worse problem that has been fixed.
[15:58] <mhagger> I have no problem importing Python and Java files in the same directory
[15:58] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[15:59] <jimbaker> mhagger, that would make sense to me. i did a quick scan of NEWS for jython, i didn't see anything relevant. but again it was certainly discussed around this time last year
[15:59] <mhagger> I only have a problem in the case that one directory contains *only* Python, but a subdirectory contains Java -- in this case I cannot import the Java
[16:00] <jimbaker> mhagger, i think the case that was fixed was python and java in the same dir
[16:00] <jimbaker> 2.5 and 2.5.1 iirc did not allow that
[16:00] <mhagger> I suppose that the problem is the following code PackageExistsFileFilter.packageExists():
[16:00] <mhagger> if (this.python && !this.java) {
[16:00] <mhagger> return false;
[16:00] <mhagger> }
[16:01] <mhagger> Hmm, I cannot swear that I have 2.5.1 installed; "jython -V" shows "2.5.1+"
[16:02] <mhagger> I just built trunk; I'll try to test with that (if I can figure out how)
[16:02] <jimbaker> mhagger, definitely try upgrading to 2.5.2 then, that 2.5.1+ means from source after we tagged it to 2.5.1
[16:04] <jimbaker> mhagger, trunk is a partial 2.6, this could be problematic. depends on what your code is doing
[16:04] <mhagger> Right now I'm just using trivial empty classes and files
[16:05] <mhagger> It looks like trunk has the same problem.
[16:06] <mhagger> (I'm assuming that if I run dist/bin/jython then I get a proper trunk environment)
[16:13] <mhagger> Would it be worth submitting a test case? If so, where should I submit it and in what format?
[16:14] <jimbaker> mhagger, bugs.jython.org
[16:15] <jimbaker> for now, it will only be addressed in trunk, although there's a small chance that a 2.5.3 might be done to backport bug fixes
[16:15] <jimbaker> but i don't know anyone who is planning to work on that, it is just possible
[16:19] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) Quit (Quit: gqlewis)
[16:23] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Quit: Ex-Chat)
[16:25] * lopex (lopex@84-10-213-252.dynamic.chello.pl) has joined #jython
[16:25] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[16:26] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has joined #jython
[16:29] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has left #jython
[16:29] <mhagger> This patch fixes the problem for me, but I don't know what it might break: http://pastebin.com/3yaRgLYv
[16:30] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has joined #jython
[16:30] <mhagger> Is there a test suite that tests hairy Java+Python imports?
[16:34] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has left #jython
[16:34] <jimbaker> mhagger, our convention is to put all the tests related to jython specifics in Lib/tests/*_jy.py
[16:34] <jimbaker> there are a few that break that, mostly stuff seen in *_java_*.py
[16:35] <jimbaker> (in those cases they don't have python equivalents anyway)
[16:36] <mhagger> This seems to be the bug: http://bugs.jython.org/issue1878866
[16:37] <jimbaker> not the bug we were looking at, however
[16:37] <mhagger> No, this seems to be a more general problem that appears (in my simple tests) to be solved already.
[16:38] <jimbaker> also it was listed as priority=normal, but severity=critical
[16:38] <jimbaker> roundup basically sucks for finding issues
[16:38] <jimbaker> we must move off it asap
[16:46] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Quit: Ex-Chat)
[16:47] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) has joined #jython
[16:49] * gqlewis (~gqlewis@h246.70.82.166.dynamic.ip.windstream.net) Quit (Remote host closed the connection)
[16:50] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[16:50] * jabley (~jabley@78-105-125-211.zone3.bethere.co.uk) Quit (Quit: jabley)
[17:09] <mhagger> jimbaker: I just added a test case and a patch to the bug quoted above. If there's anything else I can do to help, please let me know.
[17:09] * epoitras (~eddie@69-196-138-130.dsl.teksavvy.com) Quit (Quit: Leaving)
[17:11] <jimbaker> mhagger, if you were able to add this to an existing test case (something like test_import_jy.py is probably the right one), this would be very helpful in getting this patch on a fast track into trunk
[17:14] <jimbaker> in my experience, a patch w/o a test case will languish in our bugs system - in fact it is often more useful for us to have a good test case (in our standard format) than any other submission :)
[17:19] <jimbaker> mhagger, also, it looks we are still waiting on hg, for now, the best way to prep the patch is to use svn diff
[17:29] <mhagger> jimbaker: I seem some tests in bugtests/ that seem more along the lines of what I need to test for this issue. For example, I need to compile some Java and have the class files land in the same directory as some Python files. Is bugtests/ appropriate for this test?
[17:35] * MarderIII (~marderii@enneman.demon.nl) has joined #jython
[17:36] <jimbaker> mhagger, that may well be the case
[17:37] <jimbaker> biab
[17:58] <jimbaker> back
[18:10] <mhagger> http://bugs.jython.org/issue1878866 now also includes a patch to the test suite. The earlier patch does not seem to cause any new regressions in the test suite or in bugtests.
[18:12] * shashank1 (~shashank@cpe-75-185-124-139.insight.res.rr.com) has joined #jython
[18:12] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) Quit (Read error: Connection reset by peer)
[18:25] <jimbaker> mhagger, i should have thought about this more closely - bugtests are really the legacy tests for jython, we have been trying to move away from putting stuff in there
[18:26] <jimbaker> i was thinking "javatests", as seen in tests/java/javatests when you said bugtests
[18:27] * thobe1 (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 240 seconds)
[18:27] <jimbaker> in the past, we wouldn't support testing like support.compileJava("test403p/b/Foo.java"), but now that's part of java 6, we could do that
[18:29] <jimbaker> other than that, it looks good to me (other than seeing the eol-style tags in http://bugs.jython.org/file970/test.patch, i guess that should be fine)
[18:33] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[18:34] <pjenvey> if you want the test on the 2.5 branch though, can't really do that
[18:36] <mhagger> 2.5 is not so critically important to me
[18:37] <mhagger> We're just now migrating from 2.2 to 2.5 so we'll have enough benefits already. We can leave the old Foobar.java workaround in our code.
[18:41] <mhagger> I didn't see any *.java files under Lib/tests/ and it is not clear to me how to place them, compile them, and ensure that they land in a directory alongside the *.py files.
[18:42] <pjenvey> we just recently moved to 2.6 development so we don't have a 'test.test_support.compileJava' (which is what jim meant), but we're intending to built a function like that using the java 6 apis that lets us compile java to bytecode
[18:42] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 258 seconds)
[18:48] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[18:59] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) Quit (Ping timeout: 252 seconds)
[19:05] * thobe (~Adium@c83-249-235-76.bredband.comhem.se) has joined #jython
[19:17] * juneau001 (~juneau@131.225.24.148) Quit (Quit: juneau001)
[19:35] * lopex (lopex@84-10-213-252.dynamic.chello.pl) Quit (Ping timeout: 260 seconds)
[19:40] * lopex (lopex@84-10-213-252.dynamic.chello.pl) has joined #jython
[19:44] * shashank1 (~shashank@cpe-75-185-124-139.insight.res.rr.com) Quit (Quit: Leaving.)
[19:45] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) has joined #jython
[19:49] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) Quit (Ping timeout: 260 seconds)
[19:57] * mhagger (~mhagger@ssh.berlin.jpk.com) Quit (Quit: Leaving.)
[20:13] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Ping timeout: 260 seconds)
[20:26] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) has joined #jython
[20:30] * shashank (~shashank@cpe-75-185-124-139.insight.res.rr.com) has joined #jython
[20:30] * MarderIII (~marderii@enneman.demon.nl) Quit (Quit: [IRSSI] Hanging on to resentment is like letting someone you despise live rent free in your head)
[21:27] * Torpeo is now known as zz_Torpeo
[21:31] * zz_Torpeo is now known as Torpeo
[21:42] * Torpeo is now known as zz_Torpeo
[22:17] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) Quit (Quit: jabley)
[22:31] * jabley (~jabley@cpc1-farn4-0-0-cust318.6-2.cable.virginmedia.com) 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 (github).