==== Channel ##pypy: 04/17/05 ====

[02:21] pedronis (~Samuele_P@c-1d8b70d5.022-54-67626719.cust.bredbandsbolaget.se) left irc: "Chatzilla 0.9.67 [Firefox 1.0.2/20050317]"

[02:30] hpk_ (~holger_kr@mail.trillke.de) left irc: Read error: 148 (No route to host)

----- silence for 6 hr and 43 minutes -----

[09:13] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) left irc: "This computer has gone to sleep"

----- silence for 2 hr and 25 minutes -----

[11:38] arigo (~arigo@mail.trillke.de) joined #pypy.

----- silence for 3 hr -----

[14:38] pedronis (~Samuele_P@c-1d8b70d5.022-54-67626719.cust.bredbandsbolaget.se) joined #pypy.

----- silence for 30 minutes -----

[15:08] <pedronis> hpk: hi ?

----- silence for 1 hr and 1 minute -----

[16:09] <hpk> hi

[16:11] <pedronis> hpk: hi

[16:11] <pedronis> I have a problem with rest_test

[16:11] <pedronis> is there a problem if specify also an input_enconding and not only an output one

[16:12] <pedronis> s/if/if we

[16:12] <hpk> i don't think so

[16:12] <hpk> we should do it

[16:12] <pedronis> because the default choice of docuilts here is mac-roman, but we have a file with à that then cannot be encoded to latin-1

[16:13] <pedronis> if read in as mac-roman

[16:14] <pedronis> hpk: specifying latin-1 as input enconding in py.misc.rest.convert_to_html fix the problem

[16:18] <hpk> yes, but do we want latin1 as the default?

[16:19] <pedronis> well, the other solution is to pass dictionaries trough process and convert_rest_html

[16:19] <pedronis> hpk: well we utf8 or latin1 for our own file, howtosvn.txt has latin1 à

[16:19] <pedronis> we want

[16:19] <hpk> i was more thinking of trying to determine the encoding semi-automatically

[16:20] <pedronis> that's what docutils is trying to do but it fails

[16:20] <hpk> ah ok

[16:20] <hpk> so maybe we just assume latin1 always?

[16:21] <hpk> for the time being, at least

[16:21] <pedronis> well, for our purposes it should be ok

[16:21] <hpk> (the rest.process() helper is only used from pypy and py internally, basically)

[16:21] <pedronis> as I said you could dict or ** args to have the caller change the defaults

[16:24] <pedronis> hpk: the problem with determining the enconding automatically is that all single bytes encoding work with any inputs unless you start checking whether the result makes sense as text

[16:24] <pedronis> or try to encode back to something

[16:25] <hpk> ok, let's just assume latin1 for now in py/misc/rest.py, doesn't really hurt and can be enhanced later

[16:28] <hpk> i can check the changes in

[16:28] <hpk> ok?

[16:31] <pedronis> yes

[16:32] <hpk> done

[16:37] <pedronis> thanks

[16:38] <hpk> does it work for you, now?

[16:48] tic (~vision@c-996e73d5.019-35-67626717.cust.bredbandsbolaget.se) left irc: Read error: 113 (No route to host)

[16:48] tic (~vision@c-996e73d5.019-35-67626717.cust.bredbandsbolaget.se) joined #pypy.

[16:51] <pedronis> hpk: yes

----- silence for 36 minutes -----

[17:27] dialtone (~dialtone@host131-1.pool21345.interbusiness.it) joined #pypy.

----- silence for 1 hr -----

[18:27] dialtone (~dialtone@host131-1.pool21345.interbusiness.it) left irc: Read error: 110 (Connection timed out)

----- silence for 2 hr and 11 minutes -----

[20:38] _hannes (yjicqcik@i528C1104.versanet.de) joined #pypy.

----- silence for 24 minutes -----

[21:02] cfbolz (~cfbolz@frn3-d9b94a1a.pool.mediaWays.net) joined #pypy.

[21:02] <cfbolz> Hi!

[21:03] <cfbolz> Anyone there?

[21:06] <arigo> hi!

[21:06] <cfbolz> hi!

[21:07] <cfbolz> What I somehow forgot to mention in the mail: the class with the __nonzero__ method returning self is from test_bool :-)

[21:07] <cfbolz> So there actually is a test

[21:09] <arigo> ok ok :-)

[21:09] <arigo> should just be fixed then.

[21:09] <cfbolz> :-) test_bool should pass afterwards, together with the tiny fix I did yesterday

[21:10] <cfbolz> I tried some more of CPythons tests, should I add the passing ones to conftest.py?

[21:10] <arigo> you know it's been refactored, right? (svn up)

[21:10] <cfbolz> yes

[21:10] <arigo> ok

[21:10] <arigo> yes, please add them then

[21:11] <cfbolz> I meant changing False --> True, of course :-)

[21:11] <arigo> you know also that they must be run with py.test test_xyz.py and not py.py test_xyz.py

[21:11] <cfbolz> Yes

[21:11] <arigo> because the latter won't check the output of outputtests

[21:11] <arigo> ok

[21:11] <arigo> :-)

[21:12] <cfbolz> I restricted myself to UnitTests anyway

[21:12] <cfbolz> Looking for passing tests is something I can do during university work :-)

[21:12] <cfbolz> (whereas extensive hacking on genllvm is not)

[21:13] <arigo> :-)

[21:13] <cfbolz> BTW: what about the tests that are neither UnitTests, nor OutputTests?

[21:13] <cfbolz> Does py.test handle these properly?

[21:17] <arigo> a question for hpk :-) but I think so

[21:17] <cfbolz> Ok. I'll just try. Sorry, I have to go offline (dialup). See you.

[21:18] <hpk> see you

[21:18] <hpk> just dropped in ... see you later then

[21:18] <cfbolz> bye

[21:18] cfbolz (~cfbolz@frn3-d9b94a1a.pool.mediaWays.net) left irc: "Verlassend"

----- silence for 1 hr and 22 minutes -----

[22:40] pedronis (~Samuele_P@c-1d8b70d5.022-54-67626719.cust.bredbandsbolaget.se) left irc: "Chatzilla 0.9.67 [Firefox 1.0.2/20050317]"

[22:52] dialtone (~dialtone@host111-56.pool80117.interbusiness.it) joined #pypy.

----- silence for 28 minutes -----

[23:20] pedronis (~Samuele_P@c-1d8b70d5.022-54-67626719.cust.bredbandsbolaget.se) joined #pypy.

[00:00] --- Mon Apr 18 2005