==== Channel ##pypy: 05/26/05 ====

[00:10] arigo (~arigo@c-3a8b70d5.022-54-67626719.cust.bredbandsbolaget.se) left irc: "good night"

[00:16] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) left irc: "BitchX sucks"

----- silence for 22 minutes -----

[00:38] fredrik (fredrik@c83-248-135-181.bredband.comhem.se) left irc: "http://fredrikj.net";

----- silence for 46 minutes -----

[01:24] dialtone (~dialtone@host27-8.pool80116.interbusiness.it) left irc: "Leaving"

[01:25] GoGi (~u@p549CD51F.dip.t-dialin.net) joined #pypy.

[01:25] <GoGi> Hi

[01:29] dialtone (~dialtone@host27-8.pool80116.interbusiness.it) joined #pypy.

----- silence for 1 hr and 31 minutes -----

[03:00] <stakkars> """arigo: aleale: yes, but this is only if the import statement is at the global level in the app helper"""

[03:00] <stakkars> arigo: I fear this is not true.

[03:00] <stakkars> by importing at the global level, but not doing the imports immediately,

[03:00] <stakkars> a global (but useless) reference to the module is created, and we get

[03:01] <stakkars> an dependancy. This was the cause for the release crash, btw.

[03:01] <stakkars> I could do something against it by wiping unreferenced stuff from the globals, I think.

[03:09] <stakkars> arigo: the import *is* seen, because it puts the module into the globals

----- silence for 4 hr -----

[07:09] idnar (mithrandi@idnar.user) left irc: Read error: 110 (Connection timed out)

----- silence for 20 minutes -----

[07:29] Rhamphoryncus (~rhamph@rhamphoryncus.user) left irc: ""What do you mean, how many hit points do I have?" || Support ISO 8601! http://www.cl.cam.ac.uk/~mgk25/iso-time.html";

----- silence for 46 minutes -----

[08:15] adim (~adim@logilab.net2.nerim.net) joined #pypy.

[08:22] arre (ac@ratthing-b3fa.strakt.com) joined #pypy.

----- silence for 49 minutes -----

[09:11] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) joined #pypy.

[09:23] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) left irc: "*monkey grin*"

[09:23] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) joined #pypy.

[09:32] <aleale> stakkars: thanks for clearing that issue (I thought I was going crazy)

[09:33] <aleale> We have found a better solution for the problem at hand. We mimic CPythons behaviour

----- silence for 1 hr and 48 minutes -----

[11:21] <stakkars> aleale: hi!

[11:21] <stakkars> aleale: did you check it in? What was the solution?

[11:22] <stakkars> btw. I think the import behavior of geninterp is insane, now. We should revert to having the extra parameter.

[11:27] <aleale> The problem was that to get the encodings regitered in the codecs registry I had to import encodings.

[11:28] <aleale> I did that in a modified codecs.py (there was something there to help the modulefinder get to know encodings)

[11:29] <aleale> but the modified codecs.py wasn't picked up because it was in modified-2.3.4 and not in lib (where geninterplevel looked)

[11:30] <aleale> The solution is for the lookup function to import encodings if no search functions are found

[11:31] <aleale> it is checked in as lib/inprogress__codecs.py

[11:31] <aleale> Armin would run the regression tests, because they do not work on windows

[11:32] <aleale> I am running some them now (not with py.test). It takes a long time to finish

[11:46] stakkars (hpxhaomw@dsl-62-220-11-192.berlikomm.net) left irc: Read error: 113 (No route to host)

----- silence for 21 minutes -----

[12:07] Nick change: hpk_away -> hpk

[12:07] yuuh (ore@dsl-62-220-11-192.berlikomm.net) joined #pypy.

----- silence for 1 hr and 19 minutes -----

[13:26] arre (ac@ratthing-b3fa.strakt.com) left irc: Remote closed the connection

[13:37] stakkars (~tismer@rosine232.inf.fu-berlin.de) joined #pypy.

----- silence for 16 minutes -----

[13:53] yuuh (ore@dsl-62-220-11-192.berlikomm.net) left irc: Read error: 110 (Connection timed out)

[14:00] froosch2_ (~frosch@a81-14-145-253.net-htp.de) joined #pypy.

----- silence for 19 minutes -----

[14:19] ericvrp (~chatzilla@ericvrp.demon.nl) left irc: "Chatzilla 0.9.68.3 [Firefox 1.0.4/20050516]"

[14:20] ericvrp (~chatzilla@ericvrp.demon.nl) joined #pypy.

[14:34] ludal (~ludal@logilab.net2.nerim.net) joined #pypy.

[14:37] arre (~ac@1-1-5-33a.gfa.gbg.bostream.se) joined #pypy.

----- silence for 37 minutes -----

[15:14] <hpk> ericvrp: hi eric

[15:20] cfbolz (~cfbolz@hdlb-d9b95eb4.pool.mediaWays.net) joined #pypy.

[15:20] <cfbolz> hi all!

[15:21] <hpk> hi carl and bye :-)

[15:21] Nick change: hpk -> hpk_busy

[15:21] <cfbolz> ok :-)

[15:32] <ericvrp> hi all

----- silence for 21 minutes -----

[15:53] cfbolz (~cfbolz@hdlb-d9b95eb4.pool.mediaWays.net) left irc: Read error: 113 (No route to host)

----- silence for 40 minutes -----

[16:33] Rhamphoryncus (~rhamph@rhamphoryncus.user) joined #pypy.

----- silence for 50 minutes -----

[17:23] cfbolz (~cfbolz@hdlb-d9b946b6.pool.mediaWays.net) joined #pypy.

[17:24] cfbolz (~cfbolz@hdlb-d9b946b6.pool.mediaWays.net) left irc: Client Quit

----- silence for 16 minutes -----

[17:40] arigo (~arigo@c-3a8b70d5.022-54-67626719.cust.bredbandsbolaget.se) joined #pypy.

----- silence for 27 minutes -----

[18:07] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) left irc: "Eject! Eject! Eject!"

[18:11] arre (~ac@1-1-5-33a.gfa.gbg.bostream.se) left irc: "using sirc version 2.211+KSIRC/1.3.11"

[18:11] idnar (mithrandi@idnar.user) joined #pypy.

[18:15] stakkars (~tismer@rosine232.inf.fu-berlin.de) left irc: Read error: 60 (Operation timed out)

[18:21] Nick change: hpk_busy -> hpk

[18:22] Nick change: hpk -> hpk_away

[18:25] cfbolz (~cfbolz@hdlb-d9b946bb.pool.mediaWays.net) joined #pypy.

[18:25] <cfbolz> hi!

[18:25] senra (~chatzilla@200.172.8.47) joined #pypy.

[18:26] <arigo> hi!

[18:26] <cfbolz> arigo: you greatly confused me with your comment about the 'graph' attribute

[18:27] <cfbolz> this is not yet implemented, it seems :-)

[18:27] <arigo> oups oups

[18:27] <arigo> yes, I see now that it actually uses _callable=spec_function

[18:27] <cfbolz> yes, I found that

[18:28] <arigo> sorry about that

[18:28] <cfbolz> no problem, it gave me a nice tour through lltype

[18:28] <arigo> :-)

[18:29] <cfbolz> I started rewriting genllvm -- no most tests break, because I call RPythonTyper.specialize

[18:29] <cfbolz> and this raises exceptions for most functions genllvm uses

[18:29] senra (~chatzilla@200.172.8.47) left irc: "Chatzilla 0.9.68.5 [Firefox 1.0.4/20050511]"

[18:30] <arigo> specialize() should leave alone operations it doesn't know about

[18:30] <arigo> that would be very useful for now

[18:30] <cfbolz> not now :-)

[18:31] <cfbolz> I mean, it doesn't do so now

[18:31] <arigo> yes, sorry (that was a wish)

[18:31] <cfbolz> this is all no problem -- I have enough to do to implement lltypes in genllvm

[18:36] <cfbolz> is there an explicit "cast" operation planned for the low level flow graphs?

[18:37] <arigo> yes, between pointer types

[18:37] <arigo> I guess we need one between primitive types, too

[18:38] <arigo> what does LLVM require?

[18:38] <cfbolz> explicit casting between unsigned and signed

[18:38] <cfbolz> and float as well

[18:38] <arigo> I see

[18:38] <cfbolz> -- no implicit conversion

[18:38] <cfbolz> maybe that's the use case for the first typer option :-)

[18:39] <arigo> there are exact equivalants of int_add and uint_add operations?

[18:39] <cfbolz> what do you mean?

[18:39] GoGi (u@p549CD51F.dip.t-dialin.net) left #pypy ("Client exiting").

[18:39] <arigo> when you add, you must either add two signed, or two unsigned, explicitely?

[18:39] <cfbolz> yes

[18:39] <arigo> ok

[18:39] <arigo> I guess we can just use this as a rule for the rtyper

[18:39] <cfbolz> it gives very funny errors if you don't

[18:40] <cfbolz> llvm complains about a variable not found: int %a and uint %a are different variables

[18:40] <cfbolz> so if you do %r = add int %a, %b

[18:40] <cfbolz> and %b is actually uint, it still looks for int %b

[18:40] <arigo> I see

[18:41] <arigo> explicit casts are fine for C too, so it's not a problem

[18:41] <cfbolz> ok. this would go into rtype_convert_from_to?

[18:41] <arigo> in fact C would require either explicit casts, or implementing "int_add" as "c = (signed)a + (signed)b", which is less elegant anyway

[18:42] <arigo> yes

[18:42] <cfbolz> ok. slowly I'm getting the picture :-)

[18:44] <cfbolz> I'm going again. See you.

[18:44] <arigo> see you!

[18:44] cfbolz (~cfbolz@hdlb-d9b946bb.pool.mediaWays.net) left irc: "Verlassend"

[18:54] arigo (~arigo@c-3a8b70d5.022-54-67626719.cust.bredbandsbolaget.se) left irc: "[x]chat"

----- silence for 18 minutes -----

[19:12] adim (adim@logilab.net2.nerim.net) left #pypy.

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

[19:48] ludal (~ludal@logilab.net2.nerim.net) left irc: "Download Gaim: http://gaim.sourceforge.net/";

----- silence for 17 minutes -----

[20:05] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) joined #pypy.

----- silence for 20 minutes -----

[20:25] fredrik (fredrik@c83-248-135-181.bredband.comhem.se) joined #pypy.

----- silence for 39 minutes -----

[21:04] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) left irc: "Eject! Eject! Eject!"

----- silence for 25 minutes -----

[21:29] fredrik (fredrik@c83-248-135-181.bredband.comhem.se) got netsplit.

[21:29] mwh (~user@pc150.maths.bris.ac.uk) got netsplit.

[21:29] tic (~vision@c-a86e73d5.019-35-67626717.cust.bredbandsbolaget.se) got netsplit.

[21:29] fredrik (fredrik@c83-248-135-181.bredband.comhem.se) returned to #pypy.

[21:29] tic (~vision@c-a86e73d5.019-35-67626717.cust.bredbandsbolaget.se) returned to #pypy.

[21:29] mwh (~user@pc150.maths.bris.ac.uk) returned to #pypy.

[21:40] yuuh (ore@82.140.25.212) joined #pypy.

[21:47] hpk_away (~hpk@merlinux.de) left irc: Read error: 110 (Connection timed out)

[21:47] hpk_away (~hpk@merlinux.de) joined #pypy.

----- silence for 33 minutes -----

[22:20] fr (BR@host183-87.pool8256.interbusiness.it) joined #pypy.

[22:21] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) joined #pypy.

[22:24] fr (BR@host183-87.pool8256.interbusiness.it) left #pypy.

[22:33] aleale (~redorlik@cpe.atm0-0-0-129140.0x3ef2fa3a.bynxx3.customer.tele.dk) left irc: "Eject! Eject! Eject!"

----- silence for 1 hr and 12 minutes -----

[23:45] yuuh (ore@82.140.25.212) left irc: "utz utz utz"

[00:00] --- Fri May 27 2005