Config
Log for #openttdcoop.devzone on 24th July 2011:
Times are UTC Toggle Colours
05:50:04  *** andythenorth has joined #openttdcoop.devzone
05:50:45  *** andythenorth has joined #openttdcoop.devzone
06:08:19  *** andythenorth is now known as Guest3394
06:08:19  *** andythenorth has joined #openttdcoop.devzone
06:08:46  *** Guest3394 has quit IRC
07:29:18  *** ODM has joined #openttdcoop.devzone
07:35:34  *** bodis has joined #openttdcoop.devzone
08:06:05  *** frosch123 has joined #openttdcoop.devzone
08:20:46  *** andythenorth has quit IRC
08:20:59  <Brot6> FISH - Revision 665:5a7eb8ef3457: Fix: set climate for Tow Boats to none rather than removing the... (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/5a7eb8ef3457
08:53:06  <planetmaker> frosch123: wrt the global vars page: we should start making use of it in order to avoid duplication. Should I just link it somehow or try to find a way how to include it?
08:55:12  <frosch123> i think just remove all info from pages 6,7,9 regarding vars 80+ and link to the page instead
08:55:29  <frosch123> same for vars 00-3f of varact2
08:55:47  <frosch123> but i am not sure whether all information is already on the new page
08:55:58  <frosch123> esp. about the stuff on actiond page
09:00:16  <planetmaker> Yes, that needs review along this process, not a blind 'delete everything'
09:09:53  <planetmaker> hm... how to consistently reference the variables in their details description on the 'global vars' page? by varaction2 name? by action7 name? a combination?
09:10:08  <planetmaker> probably it'll have to be some combination...
09:24:32  <frosch123> something like "Current year (01/81)" ?
09:28:49  <planetmaker> I know nothing better either; that's what I started with for now, yes.
09:30:15  <frosch123> looks like most vars with detailed description are not suitable for varact2 anyway
09:30:31  <planetmaker> though it interfers with parallel handling of serveral vars. But that can be handled by spaces
09:31:14  <frosch123> i guess a list of vars should use "," as delimeter
09:31:32  <planetmaker> or comman and /, so yes
09:31:36  <planetmaker> they actually do
10:13:48  *** andythenorth has joined #openttdcoop.devzone
10:21:05  <andythenorth> Terkhen: you wanted to help with some FIRS? :)
10:21:11  <andythenorth> but your nml won't build?
10:56:01  <planetmaker> frosch123: in your opinion, do 5F and 7B...7F belong also in the global var list?
10:56:18  <planetmaker> I tend to say 'yes' - even when they can only be used in varaction2
10:56:54  <frosch123> imo definitely not :)
10:57:08  <frosch123> they are feature specific variables, nothing global
10:58:10  <frosch123> only varact2 variables 00-3f are global
10:58:27  <frosch123> they are coded in both ttdp and ottd to match the same variables as those from act679d
10:58:35  <frosch123> the other variables are entirely different
10:58:56  <frosch123> (though var 25 is an exception, we assigned a wrong number to it back then)
10:59:09  <planetmaker> ok, then I leave them there where they are now
10:59:26  <Terkhen> andythenorth: it seems that nmlc.exe can't handle pcx files
10:59:37  <Terkhen> I'll need to do the whole setup on windows: python, pil, ply, etc
10:59:49  <Terkhen> as I don't plan to install linux for now
10:59:54  <planetmaker> not sure whether that helps
11:00:02  <andythenorth> Terkhen: once you have python, the buildout I added *should* help you
11:00:06  <planetmaker> (PIL on win)
11:00:23  <Terkhen> andythenorth: true, I'll try that :)
11:00:23  <andythenorth> it's quite plausible to mass-convert FIRS to png
11:00:34  <Terkhen> not now though, I have to go in a few minutes
11:00:45  <Terkhen> the buildout was included on nml trunk right?
11:00:49  <andythenorth> I think so
11:00:55  <Terkhen> great :)
11:00:56  <andythenorth> I have been manually converting pcx -> png for months
11:01:08  <planetmaker> what happend to your linux, Terkhen?
11:01:42  <Terkhen> I changed some pieces on my computer and it did strange things afterwards :P
11:01:51  <planetmaker> meh
11:02:07  <Terkhen> windows just dies when you change some stuff so I supposed that it's best to reinstall
11:02:25  <Terkhen> I don't plan on coding much for a bit so I don't need linux for now :P
11:04:29  <planetmaker> hm, ho :-(
11:05:25  <Terkhen> don't worry I'll still help with the nml conversion :P
11:08:58  <Terkhen> see you later
11:09:57  <planetmaker> enjoy :-)
11:27:24  *** andythenorth has quit IRC
11:27:43  *** andythenorth has joined #openttdcoop.devzone
11:49:24  <Brot6> FISH - Bug #2888 (New): Set mail capacity for hydrofoils (andythenorth) @ http://dev.openttdcoop.org/issues/2888
11:51:40  <Brot6> FISH - Revision 666:03d0ba5e6f9f: Change: set props on Sungei Catamaran (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/03d0ba5e6f9f
12:05:44  *** andythenorth has quit IRC
12:07:13  *** andythenorth has joined #openttdcoop.devzone
12:07:29  <Brot6> FISH - Revision 667:2558c945fc49: Feature: Sungei Catamaran fast ferry (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/2558c945fc49
13:08:49  <Brot6> GRFCodec - Feature Request #2889 (New): Special command to start new png file (George) @ http://dev.openttdcoop.org/issues/2889
13:09:21  *** andythenorth has left #openttdcoop.devzone
13:11:06  <planetmaker> strange request
13:54:34  <frosch123> it's the art of a decoded grf being better than the original source
13:58:11  <frosch123> "note, this is experimental and may be removed if it's useless" <- patchman' changelog comment wrt. the first newgrf stuff
14:22:50  <Brot6> Central European Train Set - Revision 87:d8142145e3bc: duplicate the gfx-template files to prepar... (Eddi) @ http://dev.openttdcoop.org/projects/cets/repository/revisions/d8142145e3bc
14:22:50  <Brot6> Central European Train Set - Revision 88:62f9960e0790: use the correct length gfx-template (Eddi) @ http://dev.openttdcoop.org/projects/cets/repository/revisions/62f9960e0790
15:45:23  <frosch123> hg exercises finished
15:45:53  <planetmaker> hm?
15:46:38  <frosch123> http://hg.openttd.org/developers/frosch/newgrfspecs.hg/annotate/dc3675690391/newgrf.txt <- i did some archaeology
15:47:54  <planetmaker> he... nice :-)
15:48:20  <frosch123> looking at the very beginning (1.8 alpha) explains quite some parts :p
16:05:54  *** andythenorth has joined #openttdcoop.devzone
16:19:59  <Brot6> FISH - Revision 668:451cfc622ddd: Fix: Hydrofoils weren't using correct capacities for Mail (andythenorth) @ http://dev.openttdcoop.org/projects/fish/repository/revisions/451cfc622ddd
16:29:15  *** andythenorth has quit IRC
16:42:08  *** Lakie has joined #openttdcoop.devzone
16:47:58  <Brot6> SuperLib - Revision 1:f697be50d160: current state (Leif) @ http://dev.openttdcoop.org/projects/superlib/repository/revisions/f697be50d160
16:59:52  <Brot6> SuperLib - Revision 2:a7db7820719f: adding the make_tar.rb script (Leif) @ http://dev.openttdcoop.org/projects/superlib/repository/revisions/a7db7820719f
17:00:14  *** andythenorth has joined #openttdcoop.devzone
17:05:18  <Brot6> FISH - Bug #2888 (Closed): Set mail capacity for hydrofoils (andythenorth) @ http://dev.openttdcoop.org/issues/2888
17:05:18  <Brot6> FISH - Bug #2888 (Closed): Set mail capacity for hydrofoils (andythenorth) @ http://dev.openttdcoop.org/issues/2888#change-7132
17:20:41  <Brot6> cets: update from r84 to r88 done (482 warnings) - http://bundles.openttdcoop.org/cets/nightlies/r88
17:22:24  <Brot6> fish: update from r662 to r668 done (4 warnings) - http://bundles.openttdcoop.org/fish/nightlies/r668
17:22:49  <Brot6> Following repos didn't need a nightlies update: ogfx-trains (r245), narvs (r37), bros (r52), ogfx-industries (r121), firs (r2183), sub-landscape (r72), opengfx (r681), ailib-tile (r16), transrapidtrackset (r15), 2cctrainset (r750), ailib-list (r32), opensfx (r97), ttdviewer (r34), worldairlinersset (r672), heqs (r605), openmsx (r97), basecosts (r25), nutracks (r202), nml (r1527), 32bpp-extra (r40), manindu (r7), newgrf_makefile (r305),
17:22:49  <Brot6> ailib-direction (r17), ailib-common (r21), snowlinemod (r49), dutchtramset (r84), ai-admiralai (r75), swisstowns (r22), metrotrackset (r56), spanishtowns (r10), frenchtowns (r6), grfpack (r279), ogfx-rv (r107), ogfx-landscape (r70), ttrs (r36), source-test (r2), ogfx-trees (r51), swedishrails (r203), grfcodec (r832), ai-aroai (r39), german-townnames (r34), smts (r19), chips (r143), belarusiantowns (r8), indonesiantowns (r41),
17:22:51  <Brot6> ailib-string (r29), airportsplus (r107), comic-houses (r71)
17:23:30  <Brot6> narvs: compile of r37 still failed (#2789) - http://bundles.openttdcoop.org/narvs/nightlies/ERROR/r37
17:24:06  <Brot6> ogfx-industries: compile of r121 still failed (#2792) - http://bundles.openttdcoop.org/ogfx-industries/nightlies/ERROR/r121
17:26:20  *** Zuu has joined #openttdcoop.devzone
17:26:26  <planetmaker> what repo do you need?
17:26:32  <Zuu> hg
17:26:45  <planetmaker> uhm... but you commited already to superlib, did you?
17:26:53  <Zuu> eg. /home/hg/tutorial-ai
17:27:24  <planetmaker> ah, ok
17:27:34  <Zuu> yes, superlib worked fine, but I guess someone has done something with the respority of that project (eg. fixed with path on server) while I was away on vacation.
17:29:01  <Ammler> you should prefix ai projects with ai- and ailib-
17:29:24  <Ammler> that is how yexo started those
17:29:28  <Zuu> oh
17:29:35  <planetmaker> it's not done consistently anyway, Ammler...
17:29:55  <planetmaker> not even his own AI does that :-P
17:30:02  <Ammler> which one?
17:30:10  <planetmaker> hm, actually. it does
17:30:40  <Ammler> I do not care, superlib is still there, I just moved it to libs
17:30:56  <planetmaker> http://dev.openttdcoop.org/projects/ai-tutorialai <-- Zuu
17:31:05  <planetmaker> just wait a few minutes then the repo will be created
17:31:13  <Ammler> why ai twice? :-)
17:31:35  <planetmaker> because otherwise it'd be an AI tutorial and not an tutorial AI ;-)
17:31:41  <Zuu> planetmaker: Ok, and you will remove the other one I created?
17:31:48  <planetmaker> oh, there was one?
17:31:54  <planetmaker> sorry, missed that
17:32:03  <Ammler> ah, it's not a ai?
17:32:07  <Zuu> http://dev.openttdcoop.org/projects/tutorial-ai <-- this one
17:32:09  <planetmaker> Ammler: it is.
17:32:14  <Ammler> then maybe you should not prefix with ai
17:33:14  <planetmaker> Ammler: the prefixing with ai- exactly causes that as many AI have AI in the name
17:33:37  <Ammler> yeah, just noticed
17:33:46  <planetmaker> doesn't hurt
17:33:53  <Zuu> Although TutorialAI is the first one of my AIs to have that. :-p
17:33:55  <Ammler> well, it does not need the prefix for technical reason
17:34:06  <Ammler> so if you prefer without prefix, do it without
17:34:16  <Zuu> I don't care really.
17:34:41  <Ammler> it just mentioned it because yexo made it that way
17:35:03  <Brot6> repository /home/hg/ai-tutorialai registered in Redmine with url /home/hg/ai-tutorialai
17:35:03  <Brot6> repository /home/hg/ai-tutorialai created
17:35:28  <planetmaker> well... I don't quite care... maybe we try to keep it. Doesn't hurt and only alters the repo address
17:35:45  <planetmaker> locally you can call the dir anyway however you want
17:36:10  <Zuu> indeed
17:36:26  <Zuu> and the repo adress I will store in .hg/hgrc anyway.
17:37:08  <Brot6> FISH - Feature #2886 (New): Complete sprites + stats Freshney freighter (andythenorth) @ http://dev.openttdcoop.org/issues/2886
17:38:02  <planetmaker> yeah, it's something you hardly need ;-)
17:38:16  <planetmaker> I copied the description from your previous project
17:38:23  <Zuu> So did I :-)
17:39:06  <Zuu> You are welcome to remove http://dev.openttdcoop.org/projects/tutorial-ai  :-)
17:39:26  <planetmaker> ok. will do now. Last chance to save anything ;-)
17:40:00  <planetmaker> too late :-P
17:40:15  <Zuu> I still have it open in my browser here ;-)
17:40:25  <planetmaker> so do I actually
17:42:12  <Brot6> Central European Train Set - Revision 89:2eda03c3e712: some erroneous pixels in the 8lu template (Eddi) @ http://dev.openttdcoop.org/projects/cets/repository/revisions/2eda03c3e712
17:42:12  <Brot6> Central European Train Set - Revision 90:cb1a2ce81053: attempt to fix offsets for 8lu vehicles. t... (Eddi) @ http://dev.openttdcoop.org/projects/cets/repository/revisions/cb1a2ce81053
17:44:35  <Brot6> Tutorial AI - Revision 1:caa0d87a01c1: Add bus chapter file (Leif) @ http://dev.openttdcoop.org/projects/ai-tutorialai/repository/revisions/caa0d87a01c1
17:44:52  *** andythenorth has quit IRC
17:45:17  <Zuu> Oh, my commits show up here so you can monitor them :-p
17:45:43  <Brot6> Tutorial AI - Revision 2:b2e3d148f5e8: Add make_tar.rb script (Leif) @ http://dev.openttdcoop.org/projects/ai-tutorialai/repository/revisions/b2e3d148f5e8
17:46:06  <Ammler> ruby :-)
17:46:40  <Zuu> I don't write any new scripts in it, but I don't see any reason to convert existing scripts for no good reason.
17:48:08  *** andythenorth has joined #openttdcoop.devzone
17:48:52  <Rubidium> :O the second Ruby script I've seen ;)
17:52:13  <planetmaker> Zuu: indeed we see most commits here ;-)
17:52:27  <planetmaker> consider this channel also something like devzone.notice ;-)
17:52:47  <Ammler> brot6 is also ruby :-)
17:52:50  <Terkhen> hi
17:52:54  <Ammler> devzone is ruby :-)
17:53:09  <planetmaker> hi Terkhen
17:53:15  <Zuu> hello Terkhen
17:57:16  * andythenorth wonders how much god_object actually remains in FIRS
17:57:36  <andythenorth> 1337 lines
17:57:48  <andythenorth> probably oil rig / oil wells / steel mill I guess
17:57:53  <andythenorth> but I haven't actually worked it out
17:59:42  <andythenorth> maybe sawmill as well
18:01:43  * andythenorth stops looking
18:02:00  <andythenorth> the amount of FIRS conversion work is a little bit daunting
18:53:06  *** Lakie` has joined #openttdcoop.devzone
18:59:59  *** Lakie has quit IRC
19:05:16  <Brot6> clientpatches: update from r22683 to r22687 done (6 warnings) - http://bundles.openttdcoop.org/clientpatches/testing/r22687
19:10:05  <Brot6> openttd-vehiclevars: update from r22683 to r22687 done - http://bundles.openttdcoop.org/openttd-vehiclevars/testing/r22687
19:14:57  <Brot6> serverpatches: update from r22683 to r22687 done (10 warnings) - http://bundles.openttdcoop.org/serverpatches/testing/r22687
19:17:10  <Brot6> 32bpp-ez-patches: compile of r22687 still failed (#2446) - http://bundles.openttdcoop.org/32bpp-ez-patches/testing/ERROR/r22687
19:21:43  <Brot6> openttd-vehiclevars-demo: update from r22683 to r22687 done - http://bundles.openttdcoop.org/openttd-vehiclevars-demo/testing/r22687
19:28:06  <Terkhen> andythenorth: what steps should I follow to use the buildout for nml?
19:39:41  <Ammler> ./setup.py install
19:46:01  <Ammler> Terkhen: or "pip install nml"
19:47:53  <Terkhen> that fails because it is missing setuptools, which IIRC is installed automatically (along with the rest of dependencies) using the buildout
19:48:14  <Ammler> how should that be possible?
19:48:24  <Ammler> it needs setuptools to run buildout, doesn't?
19:51:56  <Terkhen> I remember testing it on windows
19:52:02  <Terkhen> I only needed to install python
19:52:10  <Terkhen> but I don't remember the steps to use the buildout
19:53:38  <Ammler> buildout just prepares setup
19:58:40  <Ammler> hmm, maybe not
19:58:57  <Ammler> check #2651
19:58:57  <Brot6> Ammler: #2651 is http://dev.openttdcoop.org/issues/show/2651 "NewGRF Meta Language - Patch #2651: buildout for nml - #openttdcoop Development Zone"
20:00:01  <Ammler> but then, we should change the spec too
20:01:48  <Ammler> http://dev.openttdcoop.org/projects/nml/repository/entry/docs/readme.txt#L62 <-- readme would be wrong too in that case, please test :-)
20:06:48  <Terkhen> both methods (the one in the task and the one in the readme) end with an internal error
20:11:00  <andythenorth> Terkhen: normally 'bootstrap.py'
20:11:12  <andythenorth> possibly specifying a python version
20:12:02  <Ammler> andythenorth: but shouldn't setup.py do that?
20:12:09  <Ammler> according to the readme
20:12:16  <andythenorth> Ammler: buildout is an alternative to setup.py
20:12:20  <Ammler> nana
20:12:26  <andythenorth> they are competing frameworks
20:12:29  <Ammler> do you know that or is that a guess :-)
20:12:35  <andythenorth> I wrote the buildout
20:12:37  <andythenorth> so yes
20:12:45  <Ammler> :-o
20:13:15  <Ammler> then we should also fix the readme
20:13:16  <andythenorth> the buildout doesn't 100% work unfortunately
20:13:32  <Ammler> and the spec, they should use buildout
20:13:37  <andythenorth> Ammler: currently the situation is confusing
20:13:53  <andythenorth> we have setup.py, and we have buildout and you've also added pip support
20:14:04  <andythenorth> so there are potentially three ways to install :P
20:14:12  <Ammler> well, for me buildout was just a extension to setup
20:14:31  <andythenorth> yes, it builds on setup tools / easy install
20:14:47  <andythenorth> can't remember all the details of which
20:15:04  <andythenorth> anyway, it works to fetch the deps, but it actually failed to build nml last time I checked it
20:15:24  <Ammler> http://www.buildout.org/thanks.html <-- you wrote buildout?
20:15:25  <Webster> Title: Thanks Buildout v1.2.1 documentation (at www.buildout.org)
20:15:36  <andythenorth> no I wrote the buildout script ;)
20:15:44  <andythenorth> buildout is jim fulton, the zope pope iirc
20:16:14  <Ammler> ah ok, was a bit surprised :-)
20:16:27  <andythenorth> I wrote the buildout script - with help ;)
20:16:43  <Ammler> you mean, you wrote the buildout script for nml?
20:16:46  <andythenorth> yes
20:17:17  <Ammler> well, I still am not sure you are sure :-P
20:17:45  <Ammler> e.g. yexo also changed setup.py as he added builout
20:17:58  <andythenorth> yes, he cleaned something up there
20:18:32  <andythenorth> Terkhen: what python version do you have?
20:19:03  <Ammler> it would not make sense to use setuptools and buildout
20:19:55  <Ammler> but if buildout works without setuptools, you are right
20:20:03  <Ammler> Terkhen: can confirm that
20:21:18  <andythenorth> iirc, buildout needs setuptools available
20:21:21  <Terkhen> I have 2.7
20:21:39  <Terkhen> if it needs setuptools then maybe that's what I'm missing
20:22:34  <andythenorth> iirc you'll need it whether you use the buildout or whether you install deps yourself.  But I'm not sure
20:22:37  <andythenorth> my OS ships with setuptools :P
20:23:10  <Terkhen> it's strange, I don't remember installing anything
20:23:34  <Ammler> Terkhen: if you need setuptools, then do it like the readme
20:24:13  <Terkhen> if I do it as in the readme it will be missing PIL and PLY
20:24:18  <Ammler> ok
20:24:30  <Ammler> so the readme is wrong
20:24:40  <Terkhen> no, it's just missing information
20:24:45  <Ammler> it does not use buildout
20:24:56  <Terkhen> "To install NML run" <-- it will install nml, not its dependencies
20:25:00  <Ammler> well, the readme tells, that you can use buildout, but it doesn't
20:25:25  <planetmaker> now let's see... competition with 17 AI started on devzone ;-)
20:25:28  <Ammler> that is why I guessed, setup.py does care about buildout config
20:25:42  <andythenorth> Ammler: the source of bootstrap.py is quite interesting
20:25:43  <Terkhen> hmmm...
20:25:46  <Terkhen> wait
20:25:48  <andythenorth> I haven't read it for a while
20:25:52  <Terkhen> maybe I should not be using x64 version of python
20:25:53  <Terkhen> :P
20:26:54  <Rubidium> ia64 ftw ;)
20:27:24  <andythenorth> trying to work out if bootstrap.py also fetches setuptools
20:27:24  <Rubidium> reminds me, should I run my s390 OpenTTD server again? ;)
20:28:33  <Terkhen> heh, I had to google s390 :P
20:30:35  <Rubidium> it's the platform to test big endian code on as it doesn't provide a BIG_ENDIAN like preprocessor definition ;)
20:31:02  <Rubidium> which means assuming something is big endian based on the existance of such a definition fails on that platform
20:31:23  <planetmaker> he
20:32:05  <Rubidium> it takes quite a while to compile OpenTTD though
20:32:15  <Rubidium> in the emulated s390 that is
20:38:11  <planetmaker> Ammler: the games VM has more than two cores at its disposal, right?
20:39:30  <Terkhen> nah, it also fails with 32 bits, I'll install setuptools
20:39:43  <Ammler> planetmaker: I am not sure, if the cpu limit works at all
20:39:49  <Ammler> check with top
20:40:49  <planetmaker> well, top clearly shows me that the ai tournament eats 100% while stable eats 7% ;-)
20:42:34  <frosch123> night
20:42:38  *** frosch123 has quit IRC
20:45:29  <Ammler> https://munin.openttdcoop.org/openttdcoop.org/games.openttdcoop.org/cpu.html
20:46:39  <andythenorth> Terkhen: I'm testing a clean clone of nml
20:46:56  <Terkhen> hmmm... with setuptools I get an error too
20:46:57  <Ammler> I guess, the cpu usage is for all vz vms the same
20:47:21  <Terkhen> http://paste.openttdcoop.org/show/368/ <--- this one
20:47:47  <Terkhen> I also had to copy buildout.cfg to ./bin, as buildout.exe expected it there
20:48:45  <andythenorth> hmm
20:48:48  <andythenorth> sadly
20:48:50  <andythenorth> works for me
20:48:51  <andythenorth> :P
20:48:59  <andythenorth> clean clone, python 2.6, OS X
20:49:12  <andythenorth> 'python bootstrap.py'
20:49:16  <andythenorth> 'bin/buildout'
20:49:21  <andythenorth> or 'bin/buildout -v'
20:50:03  <Ammler> Terkhen: did you run boostrap first?
20:51:13  <Terkhen> yes
20:51:23  <andythenorth> what a pita
20:51:36  <Terkhen> buildout.exe is generated by it
20:52:22  <Terkhen> I'm using python 2.7
20:52:31  <Terkhen> but that shouldn't be causing errors right?
20:53:02  <planetmaker> hm... maybe...
20:53:10  <planetmaker> I'm not sure whether 2.7 is tested
20:54:26  <andythenorth> I'll get it now
20:54:38  * andythenorth hopes this doesn't screw with python 2.4 :P
20:54:39  <Ammler> well, 2.7 is used here
20:54:46  <Ammler> and on devzone
20:54:57  <planetmaker> oh, ok
20:55:04  <Ammler> but not for the nmlc.exe
20:55:56  <Ammler> maybe it would be easier to teach me how to get nmlc.exe to support pcx
20:56:18  <planetmaker> Ammler: iirc that's a PIL for windows issue
20:56:32  <planetmaker> and with PIL development at a near or total halt...
20:56:49  <Ammler> planetmaker: no, it works if you install nml the usual way
20:57:00  <planetmaker> hm, ok
20:57:03  <Ammler> else Terkhen's current work would be useless :-)
20:57:27  <Ammler> as that is the reason he wants to install
21:07:00  <andythenorth> hmm
21:07:09  <andythenorth> bootstrap and buildout work for me with 2.7
21:07:13  <andythenorth> but more of the tests fail
21:07:17  <andythenorth> (for nml)
21:07:20  <andythenorth> dunno if that's relevant
21:07:36  <Terkhen> hmmm... which tests? regression?
21:07:49  <andythenorth> yes
21:08:05  <andythenorth> more fail with 2.7 than 2.6
21:08:17  <andythenorth> in fact
21:08:27  <andythenorth> they look to be failing on module 'Image' which implies PIL?
21:08:58  <Terkhen> for me it seems to fail in the first steps of the buildout itself
21:09:04  <Terkhen> and regression should not be failing
21:09:14  <Terkhen> unless it fails because it is missing PIL/PLY and so on
21:09:28  <Ammler> test is ok, but it might fail how you run test
21:12:37  <Ammler> Terkhen: pip install nml should work in any case
21:12:38  <Ammler> :-)
21:13:26  <Ammler> if that doesn't install dependencies, then also run "pip install <dependecy>"
21:14:49  <andythenorth> sidestepping all this for a minute - I was under the impression that windows nmlc.exe was 100% standalone
21:14:56  <andythenorth> to avoid all this
21:15:04  <andythenorth> is that not the case?
21:15:36  <Ammler> yes it is
21:15:45  <Ammler> but without pcx
21:16:07  <andythenorth> hmm
21:16:11  *** ODM has quit IRC
21:16:26  <andythenorth> I would try building FIRS with python2.7
21:16:35  <andythenorth> but (a) I have to go to sleep :P
21:17:07  <Terkhen> what's pip?
21:17:17  <andythenorth> (b) it's a bit of a hassle to be sure that nml is actually using python 2.7
21:17:24  <andythenorth> (I"d have to read shebangs and such)
21:17:44  <andythenorth> Terkhen: pip is an alternative to easy_install + buildout
21:17:49  <andythenorth> it's a form of package manager
21:17:57  <andythenorth> you'd need to install it
21:18:05  <andythenorth> it's well liked by those who use it
21:18:28  <Terkhen> too many ways of installing, this is too confusing
21:18:44  <Terkhen> I'll just install everything by hand tomorrow
21:18:52  <andythenorth> that's why I wrote the buildout - there were too many conflicting options :P
21:19:01  <andythenorth> I sadly think I failed to make it better :(
21:19:24  <Terkhen> I don't know what's missing; I can help you with further tests if you need it
21:19:30  <andythenorth> PIL is known to be a complete pain
21:19:37  <andythenorth> so most likely you have a PIL problem
21:19:49  <andythenorth> if you run verbose buildout, that might help
21:19:51  <Terkhen> but I want to start doing some conversions on FIRS
21:20:17  <andythenorth> bin/buildout -v
21:20:19  <andythenorth> is verbose
21:20:20  <Ammler> oh well, easy_install... should also work
21:20:28  <Ammler> I thought pip is easier
21:20:44  <Terkhen> http://paste.openttdcoop.org/show/369/ <--- looks about the same with -v
21:20:48  <andythenorth> Ammler: it's most likely that PIL is failing, in which case the choice of package manager is moot :)
21:21:11  <andythenorth> ok that's beyond me :(
21:21:18  <andythenorth> manual install beckons...
21:21:20  <Ammler> how can windows survive with such complicated things
21:21:36  <Ammler> Terkhen: you should just install linux :-)
21:21:59  <Ammler> the only good thing windows is for is gaming
21:22:09  <andythenorth> having bin/bin looks suspicious
21:22:39  <andythenorth> it looks like path errors
21:22:58  <andythenorth> it can't find setup.py
21:23:01  <andythenorth> is my thinking
21:23:11  <Terkhen> that's why I'm on windows
21:23:23  <andythenorth> it's looking in bin for setup.py
21:23:28  <andythenorth> hmm
21:23:30  <Terkhen> I can't be bothered to install linux right now :)
21:24:04  <Terkhen> this is the only coding stuff I'm planning to do for some weeks, and I should not need to install a new OS just for that
21:24:09  <andythenorth> vmware is mighty handy :P
21:24:16  <andythenorth> with 'appliances'
21:24:17  <Terkhen> it's also a PITA
21:24:21  <planetmaker> mingw ;-)
21:24:26  <Terkhen> mingw is what I'm using
21:24:31  <planetmaker> oh
21:24:58  <andythenorth> Terkhen: I don't know why you you have nml/bin/bin
21:25:03  <andythenorth> the last /bin is one too many
21:25:17  <andythenorth> possibly a mistake by bootstrap.py, dunno
21:25:48  <andythenorth> nml/bin should contain 'buildout' or maybe 'buildout.exe'
21:25:53  <planetmaker> @calc 5*50*1.5
21:25:53  <Webster> planetmaker: 375
21:25:59  <planetmaker> @calc 5*50*1.5/60
21:25:59  <Webster> planetmaker: 6.25
21:26:04  <andythenorth> and buildout.cfg should be in nml/
21:26:06  <planetmaker> ok... :-P
21:26:11  <andythenorth> as should setup.py
21:26:43  <Terkhen> strange
21:26:48  <Ammler> installing a whole linux is easier as installing one app on windows ;-)
21:26:50  <Terkhen> I do have nml/bin/bin, but it is empty
21:27:10  <andythenorth> well the buildout seems to be using it as the . for the paths
21:27:24  <andythenorth> if I read your error paste correctly :P
21:27:30  <Terkhen> probably related to why I had to copy the buildout.cfg to ./bin
21:27:50  <Terkhen> Ammler: not really IMO
21:28:14  <andythenorth> Terkhen: this seems resolvable - but hard without being able to see your filesystem ;)
21:28:26  <Terkhen> heh, it works if I run it from /nml
21:28:39  <Terkhen> ./bin/buildout.exe instead of cd bin && ./buildout.exe
21:28:50  <andythenorth> yes
21:29:03  <andythenorth> that's the correct method ;)
21:29:21  <Terkhen> it should fail more gracefully if you don't run it with the correct path IMO
21:29:31  <andythenorth> buildout is not a graceful thing :P
21:29:36  <Terkhen> I wouldn't have noticed the extra /bin by myself
21:29:36  <andythenorth> it's great, but it also has issues
21:29:56  <andythenorth> buildout saves a lot of time, but when it goes wrong....it wastes a lot of time
21:30:02  <andythenorth> it's a bit too magic for most people's taste
21:30:11  <andythenorth> but installers are one place magic is needed
21:30:29  <andythenorth> anyway, that should have fetched all the deps
21:30:53  <andythenorth> n.b. the deps will be local to your project
21:30:59  <andythenorth> PIL etc won't be system-wide
21:31:17  <andythenorth> that's another purpose of buildout - let you fetch specific deps without breaking the rest of your system
21:32:00  <andythenorth> Terkhen: did buildout also run the tests?
21:32:07  <andythenorth> (nml regression)
21:32:08  <Terkhen> it's running 027 now
21:32:15  <Terkhen> so it's looking fine so far
21:33:08  <andythenorth> ok, time for sleep in my case then
21:33:16  <andythenorth> anything else, I probably can't help with
21:33:28  <andythenorth> I have absolutely no idea how nmlc works :)
21:33:34  <andythenorth> good night
21:34:14  <Terkhen> good night andythenorth
21:34:18  <Terkhen> and thanks ;)
21:34:21  *** andythenorth has quit IRC
21:36:43  <planetmaker> ho, sweet.
21:37:02  <planetmaker> Now it would be interesting to actually get this procedure simplified to 'download this' :-P
21:37:14  <Terkhen> at first I just downloaded nmlc.exe
21:37:22  <Terkhen> it works as long as your project does not have any pcx files
21:37:32  <planetmaker> :-D
21:37:37  <planetmaker> sounds easy enough
21:37:41  <Terkhen> yes
21:38:03  <Terkhen> but I did not want to convert all pcx files, so... :P
21:38:08  <Terkhen> *in FIRS
21:38:39  <planetmaker> :-)
21:39:14  <Ammler> Terkhen: I could give you VM for a time
21:39:21  <planetmaker> if my estimate is roughly correct around tomorrow noon I'll know the results of a 17AI, 50 year game length AI competition ;-)
21:39:31  <planetmaker> maybe already tomorrow morning
21:39:31  <Ammler> which you could work on suse over nx
21:43:16  <Terkhen> Ammler: don't worry, I could also be using a linux server I have around, but I'm more confortable using my own computer :)
21:43:18  <Terkhen> thanks :)
21:43:47  <Terkhen> now I get a "cannot identify image type" while trying to compile FIRS
21:43:51  <Terkhen> but that's for tomorrow
21:43:55  <Terkhen> good night
21:43:57  <Ammler> :-)
21:44:00  <Ammler> good night
21:48:53  <planetmaker> good night
21:49:02  <planetmaker> also from here :-)
22:12:47  *** bodis has quit IRC
23:34:43  *** Zuu has quit IRC
23:52:20  *** Lakie` has quit IRC

Powered by YARRSTE version: svn-trunk