[00:29] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) joined #pypy.
[00:35] jriehl (~jriehl@sidewinder.cs.uchicago.edu) left irc: "Download Gaim [http://gaim.sourceforge.net/]";
[00:42] pedronis (~Samuele_P@c-358b70d5.022-54-67626719.cust.bredbandsbolaget.se) left irc: "Chatzilla 0.9.67 [Firefox 1.0.2/20050317]"
----- silence for 31 minutes ----- [01:13] nufu (~seanj@d207-6-164-187.bchsia.telus.net) joined #pypy.
----- silence for 16 minutes ----- [01:29] nufu (~seanj@d207-6-164-187.bchsia.telus.net) left irc: "Lost terminal"
----- silence for 1 hr and 37 minutes ----- [03:06] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) left irc: "Leaving"
[03:16] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) joined #pypy.
----- silence for 6 hr and 8 minutes ----- [09:24] arigo (~arigo@134.99.18.225) joined #pypy.
----- silence for 41 minutes ----- [10:05] <arigo> hpk: morning
[10:08] Action: arigo visits hpk for a 2.5 days mini-sprint
[10:08] <arigo> hpk: I'm going for the 11:38-14:57 train.
[10:18] arigo (~arigo@134.99.18.225) left irc: "see you"
----- silence for 23 minutes ----- [10:41] arigo (~arigo@134.99.112.244) joined #pypy.
[10:43] arigo (~arigo@134.99.112.244) left irc: Client Quit
----- silence for 6 hr and 15 minutes ----- [16:58] arigo (~arigo@mail.trillke.de) joined #pypy.
[16:58] <hpk> arigo: hi
[17:01] <arigo> hi!
----- silence for 4 hr and 9 minutes ----- [21:10] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) left irc: Success
[21:11] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) joined #pypy.
----- silence for 38 minutes ----- [21:49] <arigo> warning: merging the pypy branch "py-collect"! no check-ins please...
----- silence for 26 minutes ----- [22:15] pedronis (~Samuele_P@c-1d8b70d5.022-54-67626719.cust.bredbandsbolaget.se) joined #pypy.
[22:16] <arigo> hi!
[22:16] <pedronis> hi
[22:16] <arigo> "svn up" in the root dist...
[22:16] <pedronis> I did that
[22:16] <arigo> we merged the py lib's py-collect branch back, and the corresponding pypy branch.
[22:17] <arigo> ok.
[22:17] <pedronis> in fact test_all is not working for me
[22:17] <pedronis> because on the Mac I have no tkinter
[22:17] <pedronis> and so its import fails
[22:21] <arigo> oups
[22:24] <arigo> hpk fixed it in the py lib
[22:24] <arigo> we have a problem in builtin/importing.py
[22:25] <arigo> load_part() can receive None as w_path, but isn't prepared to handle it
[22:31] <pedronis> arigo: in what situation?
[22:31] <arigo> well, that's kind of the hard question
[22:32] <pedronis> because there's a guard in the code right now
[22:32] <arigo> I mean, w_path = try_getattr(...) could give None
[22:32] <arigo> which can get passed to load_path() on the next loop iteration
[22:32] <arigo> the hard question is why hpk was running into the problem -- now he no longer is
[22:35] <pedronis> I see, well you just need a package and remove its __path__
[22:36] <arigo> yes, I guess so
[22:36] <arigo> we should make a test for it
[22:37] <arigo> btw, I forgot to update llvm, which uses the s_item attribute of SomeList quite a bit
[22:37] <arigo> of course the tests were just skipped on my machine
[22:38] <pedronis> it's seems that guarding the for in unpack w_path with a test about w_path is not None should fix the problem
[22:38] <pedronis> you would get an import error
[22:39] <pedronis> the question is also what exactly does CPython in such a case
[22:41] <arigo> of course...
[22:44] <arigo> pedronis: you get an ImportError too, it seems. I cannot import anything from such a package...
[22:48] <pedronis> mmh, very strange
[22:48] <pedronis> if I run test_all translator alone it works
[22:48] <pedronis> but if I run all the tests all compilation tests fail
[22:57] <hpk> with a particular message?
[22:58] <pedronis> something about linker flags
[22:59] <pedronis> also all combinations of module|objspace|... alone + translator work also
[22:59] <pedronis> so it is very very strange
[23:00] <pedronis> hpk: btw, it seems that doing python test_all.py in the pypy directory
[23:00] <pedronis> you get a test_all.py [0]
[23:00] <pedronis> that means test_all itself is tried has a test
[23:01] <arigo> well, yes but it doesn't really matter
[23:01] <hpk> but it shouldn't matter indeed
[23:01] <hpk> because there is a __name__ == '__main__ guard
[23:01] <hpk> do you think your problem is related to this?
[23:02] <hpk> it was like this before, only that it didn't show the line at all
[23:02] <pedronis> ah
[23:02] <hpk> it's probably possible to change it back to the old behaviour :-)
[23:03] <hpk> although armin argues that it makes sense to see the [0] empty test files
[23:04] <pedronis> anyway I'm trying to reduce the set of tests necessary to fail but the fact that picking one of the other packages + translator also work is not helping
[23:06] <hpk> ok, i guess i am going to try to reproduce the problem on my OSX
[23:06] <hpk> armin and me need to get upstairs to grab some food, anyway ...
[23:09] <hpk> if you find something then please mail me
[23:09] <arigo> see you later
[23:09] arigo (~arigo@mail.trillke.de) left irc: Remote closed the connection
----- silence for 18 minutes ----- [23:27] hpk_ (~holger_kr@mail.trillke.de) joined #pypy.
[23:38] <pedronis> hpk: discovered what is going on and sent a mail
[00:00] --- Sun Apr 17 2005