Timestamps are in GMT/BST.
[0:03] * slashfoo (~slashfoo@12.191.61.1) has joined #jython
[0:03] <shashank> Copyright (C) 2007 Vladimir Sizikov <vsizikov@gmail.com>
[0:03] * slashfoo is now known as Guest85272
[0:08] * Guest85272 (~slashfoo@12.191.61.1) Quit (Quit: leaving)
[0:10] * lheuer (~Adium@unaffiliated/lheuer) has joined #jython
[0:40] * Arfrever (~Arfrever@apache/committer/Arfrever) has joined #jython
[0:51] <pjenvey> jimbaker - yea, i think so
[0:56] * [Arfrever] (~Arfrever@apache/committer/Arfrever) has joined #jython
[1:03] * Rondom (~rondom@2a01:488:66:1000:b24d:4f2f:0:1) Quit (Remote host closed the connection)
[1:04] * Rondom (~rondom@2a01:488:66:1000:b24d:4f2f:0:1) has joined #jython
[1:14] * shashank (~shashank@12.191.60.109) Quit (Ping timeout: 252 seconds)
[1:16] * fwierzbicki (~frank@12.43.173.66) Quit (Quit: fwierzbicki)
[1:17] * agronholm (~Demi-god@12.43.173.16) Quit (Ping timeout: 252 seconds)
[1:48] * Oti (~ohumbel@adsl-89-217-185-43.adslplus.ch) Quit (Ping timeout: 260 seconds)
[1:50] * Oti (~ohumbel@adsl-89-217-161-104.adslplus.ch) has joined #jython
[2:00] * lheuer (~Adium@unaffiliated/lheuer) Quit (Quit: Leaving.)
[2:09] * ohumbel (~ohumbel@adsl-89-217-60-52.adslplus.ch) has joined #jython
[2:10] * Oti (~ohumbel@adsl-89-217-161-104.adslplus.ch) Quit (Ping timeout: 244 seconds)
[2:10] * ohumbel is now known as Oti
[2:11] * stakkars_ (~tismer@12.191.61.143) Quit (Quit: schnarch)
[2:24] * agronholm (~Demi-god@12.43.172.10) has joined #jython
[2:26] <agronholm> the sprint continues...tomorrow? or what
[3:42] * juneau001 (~juneau@50.44.13.106) has joined #jython
[3:42] * Arfrever (~Arfrever@apache/committer/Arfrever) Quit (Quit: Ex??re)
[4:06] * stakkars_ (~tismer@12.191.61.143) has joined #jython
[4:43] * juneau001 (~juneau@50.44.13.106) Quit (Quit: juneau001)
[4:47] * shashank (~shashank@209.119.68.98) has joined #jython
[4:54] <pjenvey> agronholm - yep
[4:55] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[4:56] <agronholm> pjenvey: still getting permission denied... :/
[5:28] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Quit: fwierzbicki)
[5:29] <pjenvey> agronholm - yea, hopefully it'll be fixed in the morning. george brandl is probably fast asleep still
[5:29] <agronholm> k
[5:42] * shashank (~shashank@209.119.68.98) Quit (Quit: Leaving.)
[6:04] * stakkars_ (~tismer@12.191.61.143) Quit (Quit: schnarch)
[6:33] * stakkars_ (~tismer@12.43.172.10) has joined #jython
[6:35] * randomlurker (402f6d12@gateway/web/freenode/ip.64.47.109.18) Quit (Ping timeout: 245 seconds)
[6:36] <pjenvey> agronholm - you should be good now? try it
[6:36] * JStoker (jstoker@unaffiliated/jstoker) Quit (Ping timeout: 244 seconds)
[6:37] * JStoker_ (jstoker@unaffiliated/jstoker) Quit (Ping timeout: 244 seconds)
[6:37] <agronholm> worked
[6:37] <agronholm> I'm surprised it went through w/o having to rebase
[6:37] <pjenvey> cool
[6:38] <agronholm> I bet the bytearray problems are still masking a great deal of issues in it
[6:38] <agronholm> but there's not much I can do until it's been implemented
[6:49] * JStoker (jstoker@unaffiliated/jstoker) has joined #jython
[6:57] * JStoker_ (jstoker@unaffiliated/jstoker) has joined #jython
[7:36] * ohumbel (~ohumbel@adsl-89-217-60-52.adslplus.ch) has joined #jython
[7:36] * Oti (~ohumbel@adsl-89-217-60-52.adslplus.ch) Quit (Read error: Connection reset by peer)
[7:36] * ohumbel is now known as Oti
[7:37] * ohumbel (~ohumbel@adsl-89-217-60-52.adslplus.ch) has joined #jython
[7:37] * Oti (~ohumbel@adsl-89-217-60-52.adslplus.ch) Quit (Read error: Connection reset by peer)
[7:37] * ohumbel is now known as Oti
[7:38] * ohumbel (~ohumbel@adsl-89-217-60-52.adslplus.ch) has joined #jython
[7:38] * Oti (~ohumbel@adsl-89-217-60-52.adslplus.ch) Quit (Read error: Connection reset by peer)
[7:38] * ohumbel is now known as Oti
[7:38] * Oti (~ohumbel@adsl-89-217-60-52.adslplus.ch) Quit (Client Quit)
[7:58] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Read error: Connection reset by peer)
[7:59] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[9:39] * r0bby (~wakawaka@guifications/user/r0bby) has joined #jython
[9:42] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Ping timeout: 244 seconds)
[9:50] * JStoker_ (jstoker@unaffiliated/jstoker) Quit (*.net *.split)
[9:55] * JStoker is now known as 31NAA1G4I
[9:55] * JStoker (jstoker@unaffiliated/jstoker) has joined #jython
[11:11] * juneau001 (~juneau@131.225.24.53) has joined #jython
[11:52] * wainersm (wainersm@nat/ibm/x-ykettychvhgdcmzh) has joined #jython
[12:14] * agronholm (~Demi-god@12.43.172.10) Quit (Ping timeout: 244 seconds)
[12:18] * r0bby (~wakawaka@guifications/user/r0bby) Quit (Ping timeout: 246 seconds)
[13:09] * enebo (~enebo@65-128-218-235.mpls.qwest.net) has joined #jython
[15:24] * shashank (~shashank@209.119.68.98) has joined #jython
[15:43] * agronholm (~Demi-god@12.43.172.10) has joined #jython
[16:03] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[16:03] <fwierzbicki> good morning all!
[16:04] * rwalter (~chatzilla@64.209.43.161) has joined #jython
[16:04] <fwierzbicki> I'm sprinting remotely today - I told some of you this, but my wife has an all day interview so I need to be around to pick up Lily from school etc
[16:06] <fwierzbicki> shashank: I was wondering - are you putting in some kind of flag so that invokedynamic can be turned off even if you are on JDK7? I'd want that to be an option at least for the short term (would be better for testing etc)
[16:06] * shashank (~shashank@209.119.68.98) Quit (Ping timeout: 260 seconds)
[16:06] <fwierzbicki> oops I missed him
[16:30] * rwalter (~chatzilla@64.209.43.161) Quit (Quit: ChatZilla 0.9.88.1 [Firefox 10.0.2/20120215223356])
[16:36] * stakkars_ (~tismer@12.43.172.10) Quit (Quit: schnarch)
[16:52] * agronholm (~Demi-god@12.43.172.10) Quit (Ping timeout: 248 seconds)
[16:58] <fwierzbicki> If someone sees shashank, I confirmed that ~35 new failures occur between r6329 and r6331
[16:59] <fwierzbicki> I can supply the list I got if he needs it
[17:13] <fwierzbicki> it looks like namespace packages will be more like 402 -- see http://mail.python.org/pipermail/import-sig/2012-March/000421.html which looks to me like it is good for us
[17:18] * agronholm (~Demi-god@12.43.173.16) has joined #jython
[17:22] * shashank (~shashank@12.43.173.31) has joined #jython
[17:57] <fwierzbicki> agronholm: oops, I sent a congrats on your first checkin but failed to trim it so it got held for moderation - I'll leave it unsent there and congratulate you here - so congrats!
[17:57] <agronholm> ah thx
[17:57] <agronholm> fwierzbicki: I'll have to redo a lot of it for 2.7 because there is no _fileio module in 2.7
[17:57] <agronholm> instead there is an _io module which has plenty more contents
[17:59] <fwierzbicki> agronholm: ah I see - if some of it is just re-implemented in C we can just use the python stuff to make things quicker
[17:59] <agronholm> fwierzbicki: also, is "from __future__ import print_function" supposed to work?
[17:59] <agronholm> it should raise a syntax error for the old-style print statement, but it doesn't
[18:00] <fwierzbicki> agronholm: yeah it is supposed to :)
[18:00] <fwierzbicki> agronholm: oh I see - that part may not work
[18:00] <agronholm> some of the tests are failing because of that
[18:00] <agronholm> 1 or 2 at most
[18:00] <agronholm> (I'm looking for a new task)
[18:01] <fwierzbicki> agronholm: ah sure, if you want to have a look at that that would be great!
[18:01] <fwierzbicki> agronholm: or you could go straight into making _io for 2.7 if you want
[18:01] <fwierzbicki> agronholm: since the plan is to go right for that
[18:01] <agronholm> writing _fileio was a traumatic experience
[18:01] <fwierzbicki> ha fair enough
[18:01] <agronholm> I was looking for something easier :)
[18:02] <fwierzbicki> I suspect print will involve looking at ANTLR
[18:02] <fwierzbicki> which can be quite tramatic
[18:02] <agronholm> fwierzbicki: remember my ERP app? I used antlr in that
[18:02] <agronholm> though it was antlr v2
[18:02] <fwierzbicki> oh you are familiar - then please go for it!
[18:02] <fwierzbicki> antlr 3 is quite different - but it's the same idea
[18:03] <agronholm> fwierzbicki: http://pastebin.com/vjd2igBN
[18:03] <fwierzbicki> oh drat that's not good at all
[18:03] <agronholm> seems the future import does not work at all
[18:05] <fwierzbicki> agronholm: ah - I have not made the partial parsing work with __future__
[18:05] <fwierzbicki> so it should work in a file
[18:05] <fwierzbicki> but not from >>>
[18:05] <agronholm> let's see
[18:05] <fwierzbicki> it's on my todo list
[18:06] <agronholm> it did work for a file...but why did the stdlib unit test fail then?
[18:06] <agronholm> s/for/from/
[18:06] <fwierzbicki> it *could* be doing an eval
[18:06] <fwierzbicki> which might have the same problem
[18:06] <agronholm> ahh
[18:06] <agronholm> that it does
[18:07] <fwierzbicki> I haven't gone through all of that to be sure it works
[18:08] <fwierzbicki> agronholm: if you'd like to look at those things they should be more analysis and less trauma
[18:08] <agronholm> I don't know how to make future imports work from eval() if that's what you mean
[18:09] <fwierzbicki> making future work with partial parsing is probably reasonably doable - let me see what file it would be
[18:11] * r0bby (~wakawaka@guifications/user/r0bby) has joined #jython
[18:15] <fwierzbicki> agronholm: hmmm not sure if I will figure this out quickly
[18:15] <agronholm> I'll look at something else in the meantime then
[18:15] <fwierzbicki> agronholm: ok
[18:15] <agronholm> the utf-7 codec is broken
[18:15] <agronholm> that looks like a low hanging fruit
[18:16] * r0bby is now known as robbyoconnor
[18:16] * stakkars_ (~tismer@12.191.60.185) has joined #jython
[18:24] * seletz_ (~seletz@HSI-KBW-078-042-237-238.hsi3.kabel-badenwuerttemberg.de) has joined #jython
[18:25] <agronholm> fwierzbicki: another issue: a = lambda: (yield) fails
[18:25] <agronholm> SyntaxError: 'return' with argument inside generator
[18:27] <fwierzbicki> agronholm: ah if you want to look at that I can give you pointers
[18:27] <agronholm> ok
[18:27] <fwierzbicki> first
[18:27] <fwierzbicki> get a small version of that in a file
[18:27] <fwierzbicki> .py file
[18:27] <fwierzbicki> then run:
[18:28] <fwierzbicki> ./dist/bin/jython ast/astview.py thatfile.py
[18:28] <fwierzbicki> then
[18:28] <fwierzbicki> python ast/astview.py thatfile.py
[18:28] <fwierzbicki> and see if they are different
[18:28] <fwierzbicki> the sad one is if jython gives you (SyntaxError)
[18:29] <fwierzbicki> then you'd just need to figure out how to get grammar/Python.g to accept that
[18:29] <fwierzbicki> or make it create the right AST
[18:29] <agronholm> the outputs are identical
[18:29] <fwierzbicki> oh weird then this is a compiler issue
[18:29] <fwierzbicki> so that might be more than you want unless you want to do some of that :)
[18:30] <fwierzbicki> jimbaker has some idea there
[18:30] <fwierzbicki> he was considering having a look at it
[18:30] <fwierzbicki> jimbaker: it seems that modifying the compiler is right for the lambda/yield thing
[18:30] <fwierzbicki> agronholm: any interest? there is surely trauma there though :)
[18:31] <agronholm> I don't know where to look
[18:31] <fwierzbicki> is jimbaker around?
[18:31] <agronholm> yes
[18:31] <fwierzbicki> he would be able to point you to it if you want to look at that
[18:32] <fwierzbicki> otherwise I'll keep thinking about how we'd get __future__ to work so I can point you there
[18:33] <jimbaker> fwierzbicki, just catching up now
[18:33] <fwierzbicki> jimbaker: no worries - turns out you where right about the lambda/yield thing
[18:33] <fwierzbicki> it's not AST
[18:33] <agronholm> and it's broken even in 2.5
[18:34] <fwierzbicki> oh ouch
[18:34] <agronholm> compiler issues may be beyond me for now
[18:34] <agronholm> there's a doctest problem I can probably work on, that should eliminate a lot of failures
[18:35] <fwierzbicki> agronholm: I just tried to eval a print function after "from __future__ import print_function" and it worked hmmm
[18:35] <fwierzbicki> agronholm: ok sounds good
[18:35] <jimbaker> fwierzbicki, ok just caught up
[18:35] <agronholm> fwierzbicki: what exactly did you try
[18:35] <fwierzbicki> it errors when I try old print - but now I see it errors when I try new print
[18:35] <jimbaker> sounds good about taking a look at this from the compiler side of things, good to see my compiler intuition holds here :)
[18:36] <fwierzbicki> jimbaker: :)
[18:37] <fwierzbicki> agronholm: oops I'm not getting it right at all
[18:37] <jimbaker> i think there's a quick fix; more importantly, it should work well within the scope of some new generator compiler support i may have mentioned
[18:37] <jimbaker> eg, why shouldn't generators lose some unnecessary overhead and inline much better
[18:38] <fwierzbicki> agronholm: which test is failing on print?
[18:38] <agronholm> lemme check
[18:39] <agronholm> test_multiple_features and test_parserhack in test_future.py
[18:44] <fwierzbicki> agronholm: ah cool test_future.py is probably the right one
[18:45] <agronholm> the failing doctests are failing with 2.5 too
[18:45] <agronholm> but it involves a deprecated Tester class so not such a big deal
[18:45] <fwierzbicki> ah right
[18:48] <fwierzbicki> agronholm: hmmm actually it looks like it's just some dumb mistake with print_function
[18:49] <agronholm> oh?
[18:49] <fwierzbicki> agronholm: probably in grammar/Python.g I'm not handling comma delimited from __future__ very well
[18:49] <fwierzbicki> or some such - let me check something
[18:49] <agronholm> those tests did not have commas in them
[18:50] * seletz__ (~seletz@HSI-KBW-078-042-237-238.hsi3.kabel-badenwuerttemberg.de) has joined #jython
[18:50] <fwierzbicki> I mean the "from __future__ import unicode_literals, print_function"
[18:50] * seletz__ is now known as seletz_home
[18:51] <seletz_home> dang, sorry. Just wanted to say a big "Thank you!" for your work on Jython.
[18:51] * SunSpot (~Adium@177.80.81.31) has joined #jython
[18:51] * SunSpot (~Adium@177.80.81.31) has left #jython
[18:52] * seletz_ (~seletz@HSI-KBW-078-042-237-238.hsi3.kabel-badenwuerttemberg.de) Quit (Ping timeout: 272 seconds)
[18:52] <fwierzbicki> seletz_home: hey no problem :)
[18:53] <seletz_home> Jython saves my live all day -- don't want to get too much into java insanity, you know ;)
[18:53] <fwierzbicki> agronholm: if I reverse the order of the imports in test_future5.py that test works - so I did something dumb in Python.g
[18:54] <fwierzbicki> ah I bet I know -
[18:54] <fwierzbicki> I bet I shouldn't have used "break" statements - premature optimization for the loss
[18:55] <fwierzbicki> checking that theory
[18:55] <fwierzbicki> seletz_home: jimbaker likes to say "we write Java so you don't have to"
[18:55] <fwierzbicki> seletz_home: I almost wrote "we right java so you don't have to" that might have worked too :)
[18:56] <seletz_home> heh -- yeah. I *do* write java if i must, though.
[18:56] <fwierzbicki> seletz_home: sure :)
[18:56] <jimbaker> fwierzbicki, seletz_home indeed that's one of two slogans i tend to use about jython
[18:57] <jimbaker> seletz_home, ideally you won't have to write as much java in the future - but this relies on the invokedynamic, gradual typing work landing
[18:57] <seletz_home> but for sure I'd went insane if It wasn't for python.. I'm working on a big enterprise tomcat/oracle stack ...
[18:57] <seletz_home> jimbaker: but isn't the invoke dynamic stuff only available in java 7?
[18:58] <seletz_home> unfortunately my customers live in the past -- or rather, the software vendor is ...
[19:00] <fwierzbicki> agronholm: ok I fixed one part of test_future, the other will probably be a little harder - I'll see if it makes sense for you to try
[19:01] <fwierzbicki> oh they caught me cheating I think!
[19:01] <jimbaker> seletz_home, yeah, it's going to require java 7
[19:02] <fwierzbicki> the put parens around "print_function" in the future import - ouch
[19:02] <jimbaker> seletz_home, but will your customers will be buying support from oracle when they end of life java 6 this fall?
[19:03] <agronholm> fwierzbicki: I figured out why the doctests are failing, but that led me to an even more bizarre problem
[19:04] <fwierzbicki> agronholm: uh oh
[19:04] <agronholm> fwierzbicki: I deleted the entire dist/Lib directory and reran ant
[19:04] <jimbaker> http://www.oracle.com/technetwork/java/eol-135779.html
[19:04] <agronholm> the strange thing is that doctest.py in dist differs from the one in CPythonlib/
[19:04] <jimbaker> nov 2012 is end of life for java 6, although extended support can be purchased
[19:04] <agronholm> not sure where it gets that doctest.py
[19:07] <fwierzbicki> agronholm: aha the second from future bug is pretty trivial too - are you happy with what you are doing or do you want to modify Python.g a little?
[19:07] <agronholm> I'll finish this and then look at Python.g
[19:07] <fwierzbicki> agronholm: ok let me know and I can tell you exactly what needs doing to get test_future.py passing
[19:08] <fwierzbicki> s/let me know/let me know when you are done/
[19:09] <seletz_home> jimbaker: oh lol -- my customers stimm require me to install RHEL 4.5
[19:10] <seletz_home> s/stimm/still/g
[19:10] * juneau001 (~juneau@131.225.24.53) Quit (Quit: juneau001)
[19:11] <seletz_home> jimbaker: but you're right of course. Can't blame anyone other than my customer for their decisions ;)
[19:11] * seletz_home stops ranting around
[19:19] <fwierzbicki> agronholm: actually it's too trivial -- I don't see an easy way to avoid a simple cut and paste, so I'll just do it. Maybe if you'd like to see if you can get rid of the cut and paste I'll point you at it :)
[19:19] <agronholm> ah, no need
[19:20] <agronholm> I discovered that jython has its own version of doctest.py
[19:20] <fwierzbicki> agronholm: ok
[19:20] <fwierzbicki> agronholm: oh I did not even know that :)
[19:20] <agronholm> so I need to find out why, and what alterations were made to it
[19:20] <fwierzbicki> (though maybe I did once - I wouldn't be surprised if you find my name in the blame)
[19:23] <agronholm> fwierzbicki: what version of cpythonlib were the Lib/ stuff altered against?
[19:23] <agronholm> I need to know that so I can do a diff and figure out why they're different
[19:23] <fwierzbicki> agronholm: should be 2.5 if it's old
[19:23] <agronholm> 2.5.0?
[19:23] <fwierzbicki> that's a tough question - if we did it properly it would be in the logs - let me see
[19:25] <fwierzbicki> darn how do you get the whole comment out of hg?
[19:25] <fwierzbicki> is "log" not the right command?
[19:25] <agronholm> which changeset are you looking at?
[19:26] <fwierzbicki> 3948 - that's when doctest was checked in last
[19:26] <fwierzbicki> from cpython I mean
[19:26] <agronholm> I think that "from:" is all that is in the changeset
[19:26] <agronholm> I don't think it's hiding anything
[19:27] <fwierzbicki> agronholm: I know that's just the first line of a two line comment
[19:27] <fwierzbicki> unless pjenvey made a mistake
[19:28] <fwierzbicki> which I doubt
[19:28] <fwierzbicki> aha it's -v
[19:28] <agronholm> ahh
[19:29] <fwierzbicki> http://svn.python.org/projects/python/branches/release25-maint/Lib/doctest.py@59174
[19:29] <fwierzbicki> so revision 59174 of release25-maint so we'd need to look at python svn
[19:32] <fwierzbicki> and I guessed the correct url
[19:32] <fwierzbicki> http://svn.python.org/projects/python/branches/release25-maint/Lib/doctest.py?p=59174
[19:33] <fwierzbicki> agronholm: the above should be the exact version we copied
[19:33] <fwierzbicki> oh duh of course r3948 in our hg repo is also exactly that
[19:41] <agronholm> ok I think I got it
[19:42] <agronholm> I'm seeing pjenvey's changes and I can forward-port them
[19:44] <fwierzbicki> agronholm: I wonder if you should do the forward porting to a 2.7 version - we'll need that soon anyway
[19:45] <agronholm> I'll do it in the 2.7 sandbox then?
[19:45] <fwierzbicki> agronholm: yeah do that
[19:55] <pjenvey> r3948 in hg may not be someone elses r3948 =]
[20:18] * AGreyCrane (~AGreyCran@99-184-22-167.lightspeed.rswlga.sbcglobal.net) Quit (Read error: Connection reset by peer)
[20:26] <agronholm> now the suite gets stuck at the pdb doctests
[20:26] <agronholm> (where pdb is launched from doctests)
[20:32] <agronholm> fwierzbicki: I pushed the doctest update which fixes a lot of tests, only 5 of 454 failing now
[20:33] <agronholm> the failures are rather annoying though since they launch pdb and it gets stuck there
[20:33] <agronholm> fixing those is probably beyond me right now
[20:55] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[20:56] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[20:57] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[20:57] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[20:57] <agronholm> there are a lot of modules that need updating
[20:58] <agronholm> doing so will probably break a lot more tests
[20:58] <agronholm> but it has to be done
[21:00] <fwierzbicki> agronholm: sounds fine - and yeah as far as I'm concerned "unexpected skip" is the same as broken - so that's probably most of what will happen
[21:00] * wainersm (wainersm@nat/ibm/x-ykettychvhgdcmzh) Quit (Quit: Ex-Chat)
[21:01] <agronholm> fwierzbicki: so do you still need me to look at Python.g?
[21:01] <agronholm> since I'm done with the doctests for my part
[21:04] <fwierzbicki> agronholm: nah I got it - it was cut and paste anyway
[21:04] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[21:05] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[21:07] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[21:08] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[21:12] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[21:12] * robbyoconnor (~wakawaka@146.95.62.197) has joined #jython
[21:12] * robbyoconnor (~wakawaka@146.95.62.197) Quit (Changing host)
[21:12] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[21:14] <fwierzbicki> agronholm: I noticed you changed partial.derived so it doesn't have a __dict__ -- that breaks test_functools.py is that something that will be ok in 2.7?
[21:14] <fwierzbicki> just curious as to why...?
[21:14] <agronholm> wait, what? I didn't change partial
[21:14] <agronholm> not to my knowledge at least
[21:14] <fwierzbicki> oh weird maybe it was a merge thing...
[21:14] <fwierzbicki> I see it in your checkin
[21:15] <fwierzbicki> if you look at your checkin email and grep on "partial.derived" you'll see it
[21:15] <agronholm> damn, I'm usually very careful about my commits
[21:15] <fwierzbicki> agronholm: maybe it was an experiment?
[21:16] <fwierzbicki> it's only a one line change minus the code gen :)
[21:16] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[21:16] <agronholm> what checkin email
[21:17] <fwierzbicki> oh maybe you aren't subscribed to jython-checkins :)
[21:17] <fwierzbicki> I'll see if I can fine a link
[21:18] <agronholm> which file is this?
[21:18] <fwierzbicki> http://mail.python.org/pipermail/jython-checkins/2012-March/000136.html
[21:18] <fwierzbicki> the file is partial.derived
[21:18] <fwierzbicki> in src/templates
[21:18] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[21:19] <fwierzbicki> specifically:
[21:19] <fwierzbicki> -want_dict: false
[21:19] <fwierzbicki> +want_dict: true
[21:19] <fwierzbicki> in that file - I can go ahead and change it back if you'd like
[21:19] <agronholm> waitaminute
[21:20] <agronholm> I changed that so that partial DOES have a __dict__
[21:20] <agronholm> that's what want_dict: true means, right?
[21:20] <fwierzbicki> oh drat hmmm
[21:21] <fwierzbicki> that is indeed what it means, weird
[21:21] <fwierzbicki> agronholm: so you did make that change on purpose - let me look at the test again
[21:21] <agronholm> no, but it fixes yet another broken thing in jython
[21:22] <agronholm> because partial has a __dict__ in cpython
[21:22] <fwierzbicki> yeah, weird
[21:22] <agronholm> but yeah that change was unintentional
[21:23] <fwierzbicki> oh it let __dict__ get deleted - probably it was passing before for the wrong reason
[21:24] <agronholm> so what I need to do with the tests is generate a list of patches against the original CPython version
[21:24] <agronholm> then update the files and reapply the patches
[21:24] <agronholm> basically the same thing I did with doctest.py, except on a massive scale
[21:24] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[21:25] <fwierzbicki> agronholm: that would be really great!
[21:25] <agronholm> yeah I need to write a script to produce said patches
[21:27] * r0bby (~wakawaka@guifications/user/r0bby) has joined #jython
[21:29] <fwierzbicki> agronholm: aha - it's just that del fails with an AttributeError instead of a TypeError - maybe that's fixable
[21:29] <agronholm> I see
[21:31] * seletz_home (~seletz@HSI-KBW-078-042-237-238.hsi3.kabel-badenwuerttemberg.de) Quit (Quit: seletz_home)
[21:31] * r0bby (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[21:32] * r0bby (~wakawaka@guifications/user/r0bby) has joined #jython
[21:34] <fwierzbicki> nope I'm wrong again -- need to investigate more, weirdness
[21:36] * r0bby (~wakawaka@guifications/user/r0bby) Quit (Read error: Connection reset by peer)
[21:37] * r0bby_ (~wakawaka@guifications/user/r0bby) has joined #jython
[21:41] * r0bby_ (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[21:42] * r0bby_ (~wakawaka@guifications/user/r0bby) has joined #jython
[21:42] <fwierzbicki> so weird - it does throw a TypeError - what the heck...?
[21:44] <jimbaker> fwierzbicki, i do have the beginning of a fix for yield expr inside of lambda
[21:44] <fwierzbicki> nice!
[21:45] <fwierzbicki> jimbaker: just to be sure you know (can't remember who said this) but it's a problem in 2.5 too
[21:45] <jimbaker> fwierzbicki, we can backport if this makes sense
[21:46] <jimbaker> fwierzbicki, this is such an edge case, i'm not certain it matters for 2.5
[21:46] <fwierzbicki> jimbaker: ok
[21:48] * 31NAA1G4I (jstoker@unaffiliated/jstoker) Quit (Quit: JStoker is gone :()
[21:51] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) Quit (Ping timeout: 244 seconds)
[21:53] * r0bby_ (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[21:54] * r0bby_ (~wakawaka@guifications/user/r0bby) has joined #jython
[21:57] * r0bby_ (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[21:57] * r0bby_ (~wakawaka@146.95.62.197) has joined #jython
[21:57] * r0bby_ (~wakawaka@146.95.62.197) Quit (Changing host)
[21:57] * r0bby_ (~wakawaka@guifications/user/r0bby) has joined #jython
[21:59] * fwierzbicki (~frank@99-106-170-105.lightspeed.sntcca.sbcglobal.net) has joined #jython
[22:08] * r0bby_ (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[22:09] * r0bby_ (~wakawaka@guifications/user/r0bby) has joined #jython
[22:09] <jimbaker> fwierzbicki, i committed the fix
[22:10] <fwierzbicki> jimbaker: cool!
[22:10] <jimbaker> i really need to use hudson to see any delta, but it should be safe :)
[22:10] <fwierzbicki> speaking of deltas - has shashank figured out what happened with the asm 4.0 update?
[22:10] <shashank> fwierzbicki: Yeah I've fixed a bunch of tests
[22:11] <fwierzbicki> shashank: oh cool
[22:11] <shashank> there's a weird test failure in test_chdir.py, specifically the test_symlink_src_is_relative which I cannot figure out
[22:12] <fwierzbicki> hmmm strange
[22:12] <shashank> fwierzbicki: the number of tests passing right now is about 120, so there are 10 tests I need to fix, but I can't figure out how my changes cause those tests to fail.
[22:13] * r0bby_ (~wakawaka@guifications/user/r0bby) Quit (Read error: Connection reset by peer)
[22:13] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[22:14] <shashank> _very_ strange! I've taken a break from those for now and working on the Java 7 and Java 6 living in the same branch, but I'll get to it later in the evening
[22:14] <fwierzbicki> shashank: very strange indeed - sounds it's going well though :)
[22:15] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[22:15] <shashank> fwierzbicki: yup, I'll get the others fixed soon too
[22:15] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[22:16] <fwierzbicki> shashank: great!
[22:18] <jimbaker> i'm going to add sys.flags.optimize (=0), not certain how our optimization would be reflected by jython, but for now it's user visible in 2.7 so we need something there
[22:18] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[22:19] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[22:23] <fwierzbicki> BTW I'm following the namespace package discussion closely now - it looks like things are going in a direction that is good for Jython
[22:23] <jimbaker> another easy fix, done
[22:23] <jimbaker> fwierzbicki, this is very good, thanks
[22:24] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[22:24] <fwierzbicki> it looks like __init__.py will be considered obsolete eventually (though backwards compatibility will remain) -- it could be a little tougher for us to tell the difference between python packages and java packages :)
[22:25] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[22:25] <fwierzbicki> but in general it looks like it will be good
[22:27] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Max SendQ exceeded)
[22:31] <shashank> fwierzbicki: In line with jimbaker's optimize flag, I'm thinking of adding another option ENABLE_INVOKEDYNAMIC which should be enabled if the user wants to enable indy
[22:34] <shashank> so the user would run jython like this: dist/bin/jython -Duse_invokedynamic=true
[22:34] <shashank> and that would imply that they get all the indy optimizations and calling conventions
[22:34] <fwierzbicki> shashank: that would be excellent - I'd really like it if folks can opt out
[22:34] <shashank> fwierzbicki: yeah here it's more of an opt-in
[22:35] <jimbaker> shashank, that sounds like a better way to do it
[22:35] <shashank> but we can make indy default at a later point of time (after we deem that it's sufficiently stable)
[22:35] <jimbaker> and fits in to the current settings mechanism
[22:35] <shashank> at which point a user can opt out by dist/bin/jython -Duse_invokedynamic=false
[22:35] <fwierzbicki> ha sure - opt in is even better for now
[22:35] <jimbaker> sys.flags.optimize is just a compability shim
[22:36] <shashank> fwierzbicki: jimbaker: ok sounds good, I'll add that right away
[22:46] * enebo (~enebo@65-128-218-235.mpls.qwest.net) Quit (Quit: enebo)
[22:53] <fwierzbicki> shashank: cool, thanks!
[22:54] * lopex (u4272@gateway/web/irccloud.com/x-ayubqulshvdvlyvt) Quit (Ping timeout: 252 seconds)
[22:58] * lopex (u4272@gateway/web/irccloud.com/x-uhltqzbeqaymlswu) has joined #jython
[23:04] <agronholm> fwierzbicki: https://gist.github.com/802efd12c0eaffddf207
[23:04] <agronholm> seems to work...
[23:14] <fwierzbicki> agronholm: looking
[23:15] <fwierzbicki> agronholm: interesting, that's pretty cool :)
[23:18] <agronholm> fwierzbicki: not all files have such a commit so I'm not sure what to do with them
[23:18] <fwierzbicki> agronholm: not everyone (myself included) is as careful as pjenvey - also some predate his practice
[23:19] <agronholm> so do we have to go through the files manually? :/
[23:19] <fwierzbicki> agronholm: I'd say skip those - if you could get me a list of them I'll look at them by hand - regrtest.py for example is a mess and hasn't had that discipline applied
[23:19] <fwierzbicki> agronholm: yeah they probably will need to be manual
[23:20] <fwierzbicki> agronholm: but the non manual step will be a great help
[23:20] <agronholm> fwierzbicki: https://gist.github.com/e34514e6e66cffe816d6
[23:20] <fwierzbicki> agronholm: and I'll be sure to be more careful going forward - I believe I've already gone wrong in the 2.7 branch (but was planning to cherry pick back into default
[23:21] <fwierzbicki> oh that's not too bad - I'm willing to take a look at those
[23:21] <fwierzbicki> at least several of those are my fault :)
[23:21] <agronholm> actually I'm going to make it include all the stdlib modules, not just the test suite
[23:22] <fwierzbicki> agronholm: oh ok that's cool.
[23:25] <agronholm> gist updated
[23:29] <agronholm> updated the script gist too
[23:29] <agronholm> should I commit this script? might be useful in a future migration too
[23:53] <jimbaker> so i'm now looking at memoryview
[23:53] <jimbaker> just trying out something small for now
[23:55] <fwierzbicki> agronholm: yeah please do commit it - probably Misc would be the right place
[23:55] <agronholm> yup that's where I put it
[23:55] <agronholm> I just finished the second script, the one that copies the files from CPythonLib and applies the patches on top of them
[23:57] <agronholm> let's see if it works
These logs were automatically created by JythonLogBot_ on irc.freenode.net using a slightly modified version of the Java IRC LogBot (github).