Log for #openttdcoop.devzone on 19th May 2011:
Times are UTC Toggle Colours
00:03:35  *** KenjiE20 has quit IRC
00:16:17  *** Lakie has quit IRC
06:50:54  <Brot6> OpenGFX - Revision 667:53bb70be54c9: -Fix: House sprite 1497 is a construction stage, not a groun... (planetmaker) @
07:25:15  <Brot6> OpenGFX+ Industries - Revision 105:a4c680efba21: Add: Translations for Greek (Tafidis) and Russia... (planetmaker) @
07:36:54  <Brot6> OpenGFX - Revision 668:c0cf06bd9146: Fix (r667): Use the correct sprite numbers explicitly (planetmaker) @
08:17:31  <planetmaker> @calc 85*12
08:17:31  <Webster> planetmaker: 1020
10:40:54  *** DayDreamer has joined #openttdcoop.devzone
10:41:15  <Ammler> XeryusTC: what a stupid license you have for mtss, could you at least change that
10:41:17  *** KenjiE20 has joined #openttdcoop.devzone
10:46:10  <Brot6> Modern Station Set - Bug #2638 (New): current license does violate DevZone Rules (Ammler) @
10:47:50  <planetmaker> :-)
10:48:21  <planetmaker> I considered that, too. Good that we think alike
10:48:21  <Ammler> Some people just shouldn't listen to pikka :-)
10:49:11  <Ammler> pikka might be a good author, but he is silly about licenses
10:49:54  <Ammler> well, bananas shouldn't allow it...
10:52:46  <Ammler> if XeryusTC won't answer until next sunday, I will tar the repo, upload to tt-forums and delete it here
10:53:16  <planetmaker> Just un-public it
10:53:38  <planetmaker> Don't delete it
10:53:45  <Ammler> it is already in zzz-pending
10:53:50  <planetmaker> I know
10:54:08  <planetmaker> But deleting it would destroy still valuable history
10:54:17  <Ammler> I think, it should be deleted,
10:54:26  <Ammler> no, I will tar the repo
10:54:31  <planetmaker> yes, eventually. Not that quickly
10:54:48  <Ammler> the tar will keep .hg
10:55:11  <planetmaker> hm, right
10:55:32  <Ammler> there is also no history
10:56:02  <Ammler>
10:56:12  <Ammler> just docs :-)
10:56:19  <Terkhen> what license it is using?
10:56:25  <Ammler> none
10:56:37  <Ammler> well, explicitly none
10:56:55  <Ammler> pikka like
10:57:37  <Terkhen> oh
10:57:58  <Ammler> if it wouldn't have no license, I could add GPL, but so it isn't possible
10:58:06  <Ammler> -n't
10:58:12  <planetmaker> well, we rised the issue already when he worked on it. And gave the repo in good faith that he'll use one according to the devzone rules
10:58:29  <Terkhen> are you even allowed to host the source then?
10:58:35  <Ammler> also as long as he is working on it, the license didn't matter much :-)
10:58:41  <planetmaker> Terkhen: he commited it himself
10:58:43  <Ammler> Terkhen: he uploaded it
10:58:53  <Ammler> yes, we are
10:58:56  <Ammler> everyone is
10:58:59  <Ammler> it is opensource
10:59:15  <Ammler> you just aren't allowed to make a derivate
10:59:35  <Terkhen> oh, that's quite bad for the set then
10:59:44  <planetmaker> yep
10:59:52  <Ammler> indeed :-)
11:00:02  <Ammler> s/:-)/:'-(/
11:00:15  <planetmaker> that won't parse ;-)
11:00:50  <planetmaker> s/:-\)/:-\(/ :'-P
11:01:08  <Ammler> hehe, also the tear might need escaping
11:01:09  <planetmaker> I clearly played around too much with regex
11:01:14  <planetmaker> yes, probably
11:01:57  <Ammler> not sure, if default sed is advanced regex
11:02:09  <planetmaker> -e or -E enables it
11:02:19  <Ammler> -e is default isn't?
11:02:33  <planetmaker> and it depends whether you talk about sed or gnu sed, too, I think ;-)
11:03:10  <planetmaker> it needs -E here for extended regex
11:03:13  <Ammler> there is a non gnu sed?
11:03:22  <planetmaker> bsd sed
11:03:37  <Ammler> GNU sed version 4.2.1
11:05:40  <planetmaker>      The sed utility is expected to be a superset of the IEEE Std 1003.2 (``POSIX.2'') specification.     The -E, -a and -i options are non-standard FreeBSD extensions and may not be available on other operating systems.
11:05:53  <planetmaker> :-)
11:05:57  <planetmaker> from my sed man page
11:06:23  <planetmaker> SED(1)                    BSD General Commands Manual                   SED(1)
11:06:33  <planetmaker> shameless copy it seems ;-)
11:06:38  <Ammler> yes, no -E here
11:06:40  <Ammler> but -r
11:07:38  <Ammler> oh, btw. did you ask, why gimp is calling gnome-vfs?
11:07:51  <planetmaker> nope
11:08:38  <planetmaker> <-- but given that it's probably gimp's means to operate on files
11:08:39  <Webster> Title: GnomeVFS - Wikipedia, the free encyclopedia (at
11:09:20  <planetmaker> after all gimp and gnome are closely linked
11:10:15  <Ammler> yes, both are gtk based
11:10:25  <planetmaker> gtk was basically written for gimp
11:10:29  <Ammler> on suse, gimp is developed as gnome app
11:11:51  <planetmaker> others then adopted gtk(+) afterwards
11:12:01  <planetmaker> gtk = gimp tool kit ;-)
11:12:16  <Ammler> hmm, and you are sure, the arguments you use are the best?
11:12:30  <planetmaker> hm?
11:13:22  <Ammler> e.q.  -c, --console-messages              Send messages to console instead of using a dialog
11:13:44  <planetmaker> ah... maybe worth a try :-)
11:14:35  <Ammler> well, I would expect -i should include those
11:17:30  <planetmaker> no effect
11:17:51  <planetmaker> t1140:~/ogfx-rv> gimp -i -c -b - < src/gfx/bulk_truck/bulk_truck_2_lumber.gimp.png.scr
11:17:53  <planetmaker> (file-uri:11587): Gtk-WARNING **: cannot open display:
11:17:54  <planetmaker> (gimp:11578): LibGimpBase-WARNING **: gimp: gimp_wire_read(): error
11:18:07  *** DayDreamer has left #openttdcoop.devzone
11:41:04  *** Lakie has joined #openttdcoop.devzone
11:46:31  *** ODM has joined #openttdcoop.devzone
12:52:15  <Ammler> [13:17] <planetmaker> t1140:~/ogfx-rv> gimp -i -c -b - < src/gfx/bulk_truck/bulk_truck_2_lumber.gimp.png.scr <-- you should know, not to run commands as root :-P
13:08:48  <planetmaker> it was run as planetmaker
13:09:02  <planetmaker> in my home dir
13:09:49  <planetmaker> there's just little point to copy a lengthy command prompt to IRC
13:09:50  <Ammler> hmm
13:10:27  <planetmaker> :-)
13:10:39  <Ammler> I should have seen the > instead #
15:06:24  <Terkhen> one of these days I'll finish with #2578
15:06:25  <Brot6> Terkhen: #2578 is "OpenGFX+ Industries - Bug #2578: Power plants can close even if serviced - #openttdcoop Development Zone"
15:06:34  <Terkhen> banks are even mentioned in the original task and I forgot about them
15:07:54  <Brot6> OpenGFX+ Industries - Bug #2578 (Reopened): Power plants can close even if serviced (Terkhen) @
15:17:05  <planetmaker> Terkhen: don't ask andy how long it's an issue for him ;-)
15:18:05  <Terkhen> what, forgetting stuff? :P
15:18:54  <Terkhen> the gold mine problem is probably a poor serviced mine or bad luck, though
15:19:18  <Terkhen> IIRC we did no changes to it, besides ground awareness and maybe animation
15:24:22  <planetmaker> nope, not forgetting. But the pain the closure issue can be
15:24:44  <planetmaker> I agree with your gold mine issue assessment. That was my first thought there, too
15:29:01  <Terkhen> besides gold/diamonds banks, is there any other black hole industry? water tower, power plant and shop are already covered
15:40:30  <planetmaker> hm...
15:40:53  <planetmaker> I've said that probably before, but not that I recall now ;-)
15:42:00  <Terkhen> neither do I :P
15:42:44  <Terkhen> IMO banks closing sometimes is not something game shattering enough to warrant a new release, so no hurry :)
15:44:53  <planetmaker> hm, yes
17:06:57  *** frosch123 has joined #openttdcoop.devzone
17:18:29  <Brot6> fish: update from r628 to r633 done -
17:18:52  <Brot6> ogfx-industries: update from r104 to r105 done -
17:19:31  <Brot6> ogfx-landscape: update from r67 to r68 done -
17:21:01  <Brot6> opengfx: update from r666 to r668 done -
17:21:07  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r750), 32bpp-extra (r40), ai-admiralai (r75), ai-aroai (r39), ailib-common (r21), ailib-direction (r17), ailib-list (r32), ailib-string (r29), ailib-tile (r16), airportsplus (r73), basecosts (r25), belarusiantowns (r8), bros (r52), chips (r141), comic-houses (r71), firs (r1990), frenchtowns (r6), german-townnames (r33), grfcodec (r829), grfpack (r279), heqs (r605),
17:21:07  <Brot6> indonesiantowns (r41), manindu (r7), metrotrackset (r56), narvs (r37), newgrf_makefile (r289), nml (r1335), nutracks (r186), ogfx-rv (r103), ogfx-rv.clone (r103), ogfx-trains (r241), ogfx-trees (r42), openmsx (r97), opensfx (r97), smts (r19), snowlinemod (r49), spanishtowns (r10), sub-landscape (ERROR r66), sub-opengfx (ERROR r666), swedishrails (r202), swisstowns (r22), transrapidtrackset (r15), ttdviewer (r34), ttrs (r36),
17:21:09  <Brot6> worldairlinersset (r671)
17:21:57  <Brot6> sub-landscape: compile of r66 still failed (#2616) -
17:22:46  <Brot6> sub-opengfx: compile of r666 still failed (#2586) -
17:46:55  *** andythenorth has joined #openttdcoop.devzone
17:47:47  *** andythenorth has quit IRC
17:48:09  *** andythenorth has joined #openttdcoop.devzone
17:48:41  *** frosch123 has quit IRC
17:54:33  <andythenorth> how do I configure the repo for my new project in redmine?
17:56:03  <Brot6> repository /home/hg/bandit registered in Redmine with url /home/hg/bandit
17:56:03  <Brot6> repository /home/hg/bandit created
17:56:03  <planetmaker> basically just create a new project on the DevZone and check the box for the module 'repository'
17:56:09  <planetmaker> he. like that
17:56:37  <planetmaker> so... configure... do you want nfo or nml? ;-)
17:56:55  <andythenorth> dunno
17:56:58  <andythenorth> which should I want?
17:57:41  <andythenorth> I don't *not* want nml
17:57:56  <andythenorth> will I gain much from it though?
17:58:08  <planetmaker> <-- choose or
17:58:56  <planetmaker> and unzip that into your new repository's dir. Edit Makefile.config accordingly and the readme.ptxt. And of course the main source file
17:59:39  <andythenorth> am I allowed to whine about everything that will confuse me in nml?
18:00:39  <Ammler> there is already a nml rv set
18:01:15  <planetmaker> andythenorth: *if* you code in NML, there's no way we could stop you whining ;-)
18:03:55  <andythenorth> nml seems the better choice
18:03:58  <andythenorth> although I'll learn less
18:04:05  <andythenorth> I learn more about programming by using nfo
18:04:18  <andythenorth> but I'm not planning to migrate HEQS
18:07:12  <andythenorth> hmm
18:07:16  * andythenorth needs to amend default push
18:07:50  <Brot6> BANDIT - Revision 0:dd4a18975b60: Add: default newgrf project files (andythenorth) @
18:09:39  <Brot6> BANDIT - Revision 1:9952ee4ac4a6: Add: hgignore (andythenorth) @
18:14:39  <planetmaker> hmpf. I need to get a monitor. A 13" laptop is no good for graphical work
18:21:18  <andythenorth> yeah it is :)
18:21:24  <andythenorth> or at least, I manage
18:21:50  <andythenorth> I used to have multiple screens, but I got bored of the side-effects
18:29:05  * andythenorth how does nml work
18:29:06  <andythenorth> ?
18:29:48  <Ammler> start with ogfx-rv and narvs
18:30:02  <andythenorth> are there docs?
18:30:09  <planetmaker> yes
18:30:14  <Terkhen>
18:30:21  <planetmaker>
18:30:24  <Terkhen> :P
18:30:28  <planetmaker> :-)
18:30:37  <Terkhen> ogfx-rv is quite simple, yes
18:30:55  <Terkhen> no callbacks: only properties and sprites
18:30:59  <Terkhen> and parameters :P
18:31:06  <andythenorth> frick
18:31:10  <andythenorth> do I need a working PIL?
18:31:17  <Terkhen> yup
18:31:18  <planetmaker> yes
18:31:21  <andythenorth> grr
18:31:25  <Ammler> well
18:31:35  <planetmaker> sudo ports install ...
18:31:44  <planetmaker> works here
18:31:51  <andythenorth> I probably have it
18:31:52  <Ammler> or else use our test suse :-P
18:31:59  <Terkhen> PIL was a PITA to install in windows
18:32:03  <andythenorth> but possibly configured in a specific way for zope
18:32:35  <Ammler> run make in regression to test your nml
18:32:47  <planetmaker> py-pil @1.1.7 (python, graphics)
18:32:48  <planetmaker>     Python Imaging Library
18:32:56  <planetmaker> sudo port install py-pil
18:34:03  <andythenorth> I need to run
18:34:16  <Ammler> I wouldn't
18:34:19  * andythenorth reads docs
18:34:26  <Ammler> just symlink nmlc to ~/bin
18:34:59  <andythenorth> why / why not?
18:35:04  <planetmaker> andythenorth: with NML the best choice is: get the repo via hg checkout. And then symlink nmlc to /usr/bin
18:35:21  <planetmaker> why? Because you then update nml by a simple hg pull -u
18:35:25  <andythenorth> ok
18:35:25  <planetmaker> why not? No reason
18:35:33  <andythenorth> no buildout?
18:35:49  <Ammler> what's buildout?
18:35:55  <planetmaker> nothing to build there
18:36:01  <planetmaker> it's a script language
18:36:02  <andythenorth> ok
18:36:13  <Terkhen> I always use " build" and " install" to reinstall nml
18:36:16  <andythenorth> wouldn't buildout fetch the dependencies and build them?
18:36:26  <Terkhen> what is the right way to do it?
18:36:35  <Ammler> Terkhen: that works too
18:36:49  <Ammler> it is like you install openttd or not
18:36:56  <Ammler> most people run openttd without install
18:37:05  <planetmaker> Terkhen: really? why?
18:37:20  <Ammler> hmm, maybe on windows it is best way as you don't have symlink
18:37:30  <Terkhen> no, I'm talking about linux
18:37:44  <Ammler> on linux, symlink to ~/bin is easiest, imo
18:37:48  <Terkhen> hmm... I could do the symlink thingie, yes
18:37:59  <planetmaker>
18:38:16  <Terkhen> but you still need to build
18:38:33  <Ammler> planetmaker: that is abit silly, linking to root space from user space
18:38:56  <planetmaker> ?
18:39:11  <planetmaker> it's vice versa
18:39:14  <andythenorth> hmm
18:39:20  <andythenorth> fails anyway :P
18:39:31  <Terkhen> andythenorth: with what error?
18:39:36  <andythenorth> permissions
18:39:43  <andythenorth> do I need to run it as sudo?
18:39:57  <Terkhen> if you are only doing build it shouldn't need sudo
18:40:00  <andythenorth> running unknown installers as sudo is not usual
18:40:00  <planetmaker> IF you want to install it, of course
18:40:02  <Ammler> planetmaker: use ingo could sabotage root space
18:40:06  <Ammler> user*
18:40:13  <planetmaker> ?
18:40:27  <Ammler> a user should not have root access
18:40:30  <Ammler> oh well
18:40:36  <andythenorth> does nml prefer any particular python version?
18:40:41  <planetmaker> 2.4 ... 2.7
18:40:52  <planetmaker> maybe not 2.7. Not sure
18:41:04  <andythenorth> fails with 2.5 and 2.6
18:41:05  <Ammler> andythenorth: just run make in regression and you see if it works
18:41:14  <andythenorth> how?
18:41:19  <Ammler> how to run make?
18:41:26  <planetmaker> cd regressions
18:41:27  <Ammler> make<enter>
18:41:27  <planetmaker> make
18:41:32  <Terkhen> I use 2.7
18:41:34  <andythenorth> no targets
18:41:41  <Terkhen> but nml won't run in 3.x
18:41:47  <planetmaker> ok, we're clearly not talking about the same dirs
18:41:59  <planetmaker> what do you install. Where. What do you run
18:42:04  <planetmaker> 'fails' is much too vague
18:42:12  <andythenorth> :)
18:42:16  <Ammler> andythenorth: in the repo nml
18:42:18  <Ammler> :-)
18:42:20  <andythenorth> too many instructions at once :P
18:42:25  <andythenorth> ok
18:42:25  <planetmaker> yes. Clearly
18:42:30  <andythenorth> lets go the make route
18:42:37  <andythenorth> so what dir should I be in?
18:42:46  <planetmaker> first: what do you want?
18:42:51  <planetmaker> what do you already have?
18:42:57  <andythenorth> I have a clone
18:43:01  <planetmaker> do you have py-pil?
18:43:05  <planetmaker> do you have py-lex?
18:43:22  <planetmaker> err.. .ply
18:43:25  <Ammler> planetmaker: how do you know that?
18:43:39  <Ammler> easiest is to run make in regression, imo
18:44:01  <planetmaker> Ammler: same as you: check your repo manager
18:44:03  <andythenorth> I have pil for python 2.4
18:44:20  <andythenorth> I'll install py-lex now
18:44:30  <Ammler> planetmaker: nah, that wouldn't tell me if it is working :-)
18:45:08  <Ammler> andythenorth: py-lex not needed
18:45:13  <andythenorth> hmm
18:45:13  <Ammler> py-ply
18:45:19  <andythenorth> confusing
18:45:32  <Ammler> yep, pm corrected right after :-)
18:45:34  <planetmaker>
18:45:50  <Ammler> anyway, run make in regression before installing anything
18:46:00  <planetmaker> maybe python 2.4 is wrong... hm
18:46:25  <andythenorth> maybe I just need another pil
18:46:35  <planetmaker> I can only confirm 2.5 and 2.6. Terkhen obviously 2.7
18:46:51  <planetmaker> and the docs say 2.5 .. 2.7 ;-)
18:47:29  <planetmaker> thus you'll need probably a new(er) python
18:47:35  <Ammler> (also make in regression would tell that)
18:47:43  <andythenorth> py-ply has installed to my python 2.4 as well
18:47:51  <andythenorth> how do I tell it to use one of my other pythons?
18:48:11  <Ammler> check shebang
18:48:26  <planetmaker> python_select
18:48:46  <planetmaker> is a package... hm
18:49:02  <andythenorth> so I'll need to switch to 2.6, then back to 2.4?
18:49:08  <Ammler> hmm, then you might need to write your own nmlc wrapper?
18:49:08  <andythenorth> (I need 2.4 as default)
18:49:15  <Ammler> why do you use such old python?
18:49:31  <andythenorth> that's what the frameworks I use need
18:50:23  <planetmaker> <-- see answer 5
18:50:24  <Webster> Title: What is the most compatible way to install python modules on a Mac? - Stack Overflow (at
18:50:35  <planetmaker> err, answer two
18:51:01  <planetmaker> oh, you need 2.4 as default? Hm... that'll not help necessarily
18:51:46  * andythenorth is checking if default can be switched
18:52:00  <andythenorth> normally we just use a combination of easy_install, and buildout
18:52:04  <andythenorth> mostly buildout
18:52:13  * planetmaker never heart of 'buildout'
18:52:25  <andythenorth> so I would do python2.6
18:52:38  <andythenorth> which would fetch all the eggs and other dependencies
18:52:45  <andythenorth> then I would run /bin/buildout
18:52:50  <andythenorth> which would build all deps
18:52:52  <Ammler> andythenorth: copy nmlc instead symlink
18:53:00  <Ammler> and modify it
18:53:01  <planetmaker> Ammler: we're not there yet...
18:53:07  <andythenorth> ^ the order I described above was a bit wrong
18:53:11  <andythenorth> but anyway
18:53:40  <andythenorth>
18:53:41  <Webster> Title: Buildout - software build system reloaded! (at
18:53:53  <Ammler> [20:52] <andythenorth> so I would do python2.6 <-- sounds like you are already farer ;-)
18:54:14  <andythenorth> well I'm not because it fails
18:54:15  <andythenorth> :P
18:54:17  <Ammler> (I would not recommend installing)
18:54:38  <andythenorth> I suspect if I don't, it won't find the deps
18:54:54  <andythenorth> there are going to be python version problems
18:54:57  <Ammler> [20:52] <Ammler> andythenorth: copy nmlc instead symlink <-- solves it :-)
18:55:20  <andythenorth> hmm
18:55:24  <planetmaker> Ammler: but not with a wrong python version. Or how do you see that solve it?
18:55:31  <Ammler> then you can use another shebang and maybe modify exports like PYTHONLIB etc.
18:55:32  <andythenorth> this is confusing
18:55:37  * andythenorth will do something else
18:56:03  <Ammler> but you should be more expert with python, sorry :-P
18:57:28  <Ammler> hmm, you need symlink, else the wrapper might not work...
18:58:34  <planetmaker> andythenorth: what you *could* do: edit nmlc in nml's main dir
18:58:44  <planetmaker> let the first line point to a proper python version
18:59:04  <Ammler> pythonlib?
19:00:29  <Ammler> cp nmlc nmlc-wrapper, modify nmlc-wrapper and symlink that to ~/bin :-)
19:01:16  <planetmaker> whatever python you choose, andythenorth : you'll need to have pil and ply for that version installed
19:01:49  <Brot6> BANDIT - Revision 2:a414a5e7a57b: Cleanup: remove nml newgrf project (andythenorth) @
19:01:49  <Brot6> BANDIT - Revision 3:a8034e6db26c: Add: nfo newgrf project (andythenorth) @
19:02:02  <planetmaker> right
19:02:27  <planetmaker> I do also something else now, I guess
19:02:42  <andythenorth> sorry :P
19:02:54  <andythenorth> I just want to make a newgrf, not spend my days fighting python
19:03:06  <planetmaker> I have the feeling you didn't try. But your choice
19:03:16  <planetmaker> and two people explaining was the error
19:03:26  <andythenorth> it seems the methods aren't known
19:03:29  <planetmaker> that's too confusing
19:03:47  <planetmaker> the methods are easy. You *just* need a working python 2.5 ... 2.7
19:03:47  <andythenorth> I can't follow instructions when the instructions are not agreed on
19:03:50  <planetmaker> and nmlc
19:04:06  <andythenorth> and a PIL in the right python
19:04:11  <andythenorth> and PLY in the right python
19:04:12  <planetmaker> and that linked to ... ach. I won't waste more time on futile explanations
19:04:16  <andythenorth> :)
19:04:27  * planetmaker pissed
19:04:40  <Ammler> planetmaker: well, usually the lines are prefixed with nicknames, so it shouldn't hurt
19:04:44  <planetmaker> spend an hour on bullshit
19:04:50  <Ammler> else you should replace your client
19:05:08  <planetmaker> good night
19:05:47  <andythenorth> good night planetmaker
19:06:05  <planetmaker> but I'm not impressed, andythenorth :-(
19:06:11  <planetmaker> you didn't even try
19:06:34  <planetmaker> try to follow one path. Both at the same time never work
19:06:45  <andythenorth> I'm not sure what path I'm supposed to follow
19:06:50  <andythenorth> there is conflicting advice
19:06:51  <Ammler> you should be able to filter infos, don't tell me you never had too
19:07:01  <planetmaker> that's what I say: follow ONE advice. Not two paths
19:07:03  <andythenorth> hmm
19:07:07  <andythenorth> forget BANDIT
19:07:17  <planetmaker> nah. don't let this distract you
19:07:21  <andythenorth> lets see if I can get nml working for opengfx vehicles
19:07:35  <Ammler> andythenorth: don't try any project first
19:07:42  <Ammler> just try make in regression
19:07:49  <andythenorth> what does that mean
19:07:50  <andythenorth> ?
19:08:03  <planetmaker> cd nml
19:08:05  <Ammler> those are little grfs and test everything
19:08:07  <planetmaker> cd regression
19:08:09  <planetmaker> make
19:08:25  <andythenorth> it fails
19:08:29  <andythenorth> for obvious reasons
19:08:30  <andythenorth> pdq2s-macbook-3:regression andy$ make
19:08:30  <andythenorth> Running test 001_action8
19:08:31  <andythenorth> Traceback (most recent call last):
19:08:31  <andythenorth>   File "../nmlc", line 3, in ?
19:08:31  <andythenorth>     from nml import main
19:08:32  <andythenorth>   File "/Users/andy/Documents/OTTD_graphics/tools/nml/nml/", line 2, in ?
19:08:32  <andythenorth>     from nml import generic, grfstrings, parser, version_info, output_base, output_nml, output_nfo, output_grf, palette
19:08:34  <andythenorth>   File "/Users/andy/Documents/OTTD_graphics/tools/nml/nml/", line 236
19:08:34  <andythenorth>     start = cur + 1 if quoted else cur
19:08:36  <andythenorth>                      ^
19:08:36  <andythenorth> SyntaxError: invalid syntax
19:08:38  <andythenorth> make: *** [001_action8] Error 1
19:08:38  <andythenorth> oops
19:08:40  <andythenorth> sorry
19:08:40  <andythenorth> should have used pastebin :(
19:08:45  <andythenorth>
19:08:55  <andythenorth> I have no PIL and no PLY
19:09:02  <andythenorth> so it ain't going to work :P
19:09:31  <andythenorth> although that doesn't seem to be the actual error
19:10:25  <Ammler> looks more like wrong python
19:13:23  <Ammler> does nmlc also test python version or just
19:15:39  <Yexo> only
19:15:47  <Yexo> andythenorth: do you by any chance have python 2.4 (or maybe 2.5)?
19:15:58  <andythenorth> I have 2.4, 2.5 and 2.6
19:16:06  <andythenorth> I've just switched the default to 2.6
19:16:21  <andythenorth> I'm also installing PIL and PLY for 2.6
19:16:34  <andythenorth> hmm
19:16:42  <Yexo> you'll need at least 2.5
19:17:28  <andythenorth> I now have PIL and PLY for 2.6
19:18:23  <andythenorth> 'make' in nml/regression still fails with the error I pasted above
19:18:43  <andythenorth> that line doesn't look obviously wrong
19:19:10  <Yexo> what happens if you do "python26 nmlc regression/001_action8.nml" ?
19:19:11  <Ammler> python --version tells?
19:19:48  <andythenorth> there's nothing wrong with that python
19:19:48  <andythenorth>
19:20:21  <Ammler> just wondering, if you really switched python version
19:20:45  <andythenorth> Ammler: python --version doesn't work
19:20:54  <andythenorth> however it tells me I haven't switched
19:20:59  <andythenorth> due to the error ;)
19:21:10  <andythenorth> that means my python_select is broken
19:21:17  <Ammler> I know, python --version works for 2.6.5 :-)
19:21:45  <Ammler> and 2.7
19:22:34  <andythenorth> which python
19:22:42  <andythenorth> shows 2.4
19:23:09  <Ammler> are you able to call python like yexo tells
19:24:27  <andythenorth> yes
19:24:31  <andythenorth> that's no problem
19:24:39  <andythenorth> that's how I usually select which python version is needed
19:24:51  <andythenorth> result is same:
19:25:39  <andythenorth> oops
19:25:41  <andythenorth> my bad
19:25:47  <Ammler> and if you edit the shebang of nmlc?
19:25:56  <andythenorth> ignore that
19:25:59  <andythenorth> it was my mistake
19:26:09  <andythenorth> I now have a more valid error
19:26:20  <andythenorth> hmm
19:26:28  <andythenorth> maybe I can build PIL and PLY from source
19:26:32  <andythenorth> instead of port install
19:26:45  <Yexo> ply is python only, no need to build anything
19:27:16  <andythenorth> so I just put it on the filesystem?
19:27:34  <Yexo> run "python install"
19:27:38  <Yexo> or python2.6 in your case
19:27:59  <andythenorth> for ply?
19:28:01  <andythenorth> or for nml?
19:32:06  <andythenorth> hmm
19:32:14  <andythenorth> it's not usual to run as sudo?
19:32:34  <Yexo> you can run it as sudo for a system-wide install
19:32:47  <Yexo> the above works for both ply and nml
19:32:58  <Yexo> and it might even work for pil too, as it'll build from source in that case
19:33:08  <andythenorth> my python is in /opt/local, so I guess it wants sudo
19:33:10  <andythenorth> works anyway
19:34:24  * andythenorth just needs to get PIL
19:37:22  <andythenorth> pdq2s-macbook-3:nml andy$ python2.6 nmlc regression/001_action8.nml
19:37:22  <andythenorth> nmlc: Default language file "lang/english.lng" doesn't exist
19:38:00  <Yexo> try "cd regression; ../nmlc 001_action8.nml"
19:38:21  <Yexo> or "nmlc regression/001_action8.nml --lang-dir=regression/lang/"
19:41:19  <andythenorth> Yexo: it only raises if there's errors?
19:41:30  <andythenorth> pdq2s-macbook-3:regression andy$ python2.6 ../nmlc 001_action8.nml appears to run silently
19:41:41  <Yexo> in that case it worked :)
19:42:09  <andythenorth> it's irritating that I can't switch my default python version
19:42:13  <Ammler> edit shebang of nmlc, then you don't need to prefix
19:42:21  <Yexo> if you want to run all regression tests you could try: make NMLC="python2.6 ../nmlc"
19:43:03  <andythenorth> ok
19:43:20  * andythenorth wonders how buildout deals with this
19:43:34  <andythenorth> maybe it changes the shebang for me
19:45:49  <Ammler> just change python to python2.6 :-)
19:46:17  <andythenorth> Ammler (1) that's possibly a problem for about 25 other projects, and if it is, I don't want to find out the hard way
19:46:21  <andythenorth> (2) I don't seem to be able to
19:46:32  <andythenorth> the recommended methods on my OS are failing
19:46:59  <Yexo> andythenorth: I think Ammler was suggesting to change the first line of nmlc
19:47:12  <Yexo> from "#! /usr/bin/env python" to "#! /usr/bin/env python2.6" or so
19:47:38  <andythenorth> is that wise?
19:47:42  <andythenorth> then I'm using a non-standard nmlc
19:47:53  <Yexo> it's a hack, but it'll work
19:48:16  <Yexo> standard or non-standard nmlc doesn't matter for the output, as long as you're not changing any relevant code
19:48:51  <andythenorth> OT but for the projects I usually work on, first you checkout, then run, which looks like this:
19:49:33  <andythenorth> that then fetches deps
19:49:38  <andythenorth> etc
19:51:16  <andythenorth> anyway...changing the shebang worked
19:51:34  <andythenorth> 22 tests pass
19:59:08  * andythenorth ponders learning how to create a buildout
20:01:03  <andythenorth>
20:01:05  <Webster> Title: Use Case - A Single Module Buildout v1.2.1 documentation (at
20:01:09  <andythenorth> Yexo: interested? ^
20:01:25  <Yexo> just looked at it
20:01:36  <Yexo> I'm not sure what the advantages of it are
20:02:22  <andythenorth> repeatable deployments with a simpler process
20:02:31  <andythenorth> user doesn't have to fetch PIL, PLY et
20:02:42  <andythenorth> it's not a big saver, but might make life easier for some
20:02:55  <Yexo> if it's not too much work, sure, I'm interested
20:03:14  <andythenorth> I'm not sure yet
20:03:36  <andythenorth> maintaining them can be a PITA, due to upstream dependencies (remote repos)
20:03:38  <Terkhen> andythenorth: can you compile nml projects already? :)
20:03:45  <andythenorth> not sure
20:03:48  <andythenorth> the tests pass though
20:04:29  <andythenorth> Yexo: it can also run the tests and raise if they fail
20:04:50  <andythenorth> basically, it's a lot fewer instructions for user
20:05:31  <Yexo> sounds good
20:07:05  * andythenorth will read a bit more
20:08:53  <Brot6> Nutracks - Bug #2566: High speed track graphics (oberhuemer) @
20:09:15  * andythenorth reads
20:09:32  <Brot6> Nutracks - Bug #2354 (Closed): loading speed on wagons (oberhuemer) @
20:09:51  <andythenorth> ply and pil are on pypi
20:11:06  <Brot6> OpenGFX+ Road Vehicles - Bug #2636 (Closed): DevZone compile failed (Terkhen) @
20:11:06  <Brot6> OpenGFX+ Road Vehicles - Bug #2637 (Closed): DevZone compile failed (Terkhen) @
20:11:28  <Terkhen> so... do we know what is causing the warnings with the xcf files?
20:11:47  <Brot6> Nutracks - Feature #2639 (New): Better transitions between track types (oberhuemer) @
20:16:41  <andythenorth> hmm
20:16:55  <andythenorth> buildout *does* set the shebang for all the deps it installs
20:17:01  <andythenorth> that's why I haven't had to do that ever
20:17:22  <andythenorth> anyone want to play along with this?  there's an interesting tutorial exercise..
20:17:37  <andythenorth> svn co svn://
20:24:25  *** frosch123 has joined #openttdcoop.devzone
20:27:51  <andythenorth> kaboom
20:27:57  * andythenorth has pil added to the tutorial buildout
20:28:34  <andythenorth> also ply
20:28:38  <andythenorth> not sure how to version pin them
20:28:51  <andythenorth> Yexo: there's no reason to version pin the deps at the moment?
20:29:40  <Yexo> every version of PIL I've seen works, and ply 3.3 also works (which is from september 2009
20:29:57  <andythenorth> ok
20:29:58  <andythenorth> hmm
20:30:05  <andythenorth> seems we could also install python with buildout
20:30:14  <andythenorth> dunno if that would be better or worse
20:30:36  <Yexo> how does that work? without python can't be run
20:30:58  * andythenorth will read that tutorial
20:31:31  <andythenorth> ok
20:31:36  <andythenorth> maybe I misunderstood
20:31:59  <andythenorth> you can use it to install a specific version of python, using the python you already have
20:32:06  <andythenorth> and it can be non-system-wide
20:32:22  <Yexo> don't think that's a good idea for nml
20:32:25  <andythenorth> nope
20:32:26  <andythenorth>
20:32:27  <Webster> Title: buildout - using different python version - Stack Overflow (at
20:33:07  <andythenorth> not sure how we create eggs yet
20:41:30  <andythenorth>
20:41:31  <Webster> Title: Deploy using zc.buildout and PythonPaste | Flask (A Python Microframework) (at
20:42:42  <Brot6> OpenGFX+ Road Vehicles - Revision 104:e3515e2678bb: Doc: Update readme. (Terkhen) @
20:42:42  <Brot6> OpenGFX+ Road Vehicles - Revision 105:bfe13debc7d6: Doc: Update changelog. (Terkhen) @
20:42:42  <Brot6> OpenGFX+ Road Vehicles - Bug #2640 (New): DevZone compile failed (compiler) @
20:42:42  <Brot6> OpenGFX+ Road Vehicles - Revision 106:ffd99dec94d3: Change: Update the NewGRF name. (Terkhen) @
20:42:45  <Brot6> OpenGFX+ Road Vehicles - Revision 107:93ca616d0b1d: Added tag 0.2.0 for changeset ffd99dec94d3 (Terkhen) @
20:44:27  <Brot6> ogfx-rv: update from r103 to r107 done (180 errors) -
20:47:00  <Terkhen> hmmm... no compiled release?
20:47:07  <Terkhen> am I too impatient? :P
20:47:50  <andythenorth> compared to who?
20:47:51  <andythenorth> me?
20:48:04  <andythenorth> Terkhen: read a nice tutorial while you wait :P
20:48:05  <Webster> Title: Developing Django apps with zc.buildout (at
20:48:25  <Terkhen> nobody, I was expecting a release compile, not a nightly compile
20:48:33  <Terkhen> but maybe it is compiling the release now
20:52:02  <Ammler> hmm
21:03:25  <andythenorth> Yexo: I was thinking wrong
21:03:43  <andythenorth> it could be more useful to have nml project use a buildout
21:03:54  <andythenorth> then they would fetch nml, pil, ply, etc
21:04:19  <Yexo> wouldn't both make sense? nml as buildout to easily install nml, nml projects as buildout to install nml + project?
21:04:20  <andythenorth> means that each project is about as self-contained as can be
21:04:45  <andythenorth> not sure that a buildout can install something which itself uses a buildout
21:04:48  <andythenorth> I just asked :P
21:04:55  <andythenorth> "but buildouts can extend each other (so a.cfg can depend on b.cfg)"
21:04:59  <andythenorth> I was told
21:37:13  *** ODM has quit IRC
21:41:40  *** andythenorth has quit IRC
21:57:35  *** frosch123 has quit IRC
23:41:54  *** KenjiE20 has quit IRC

Powered by YARRSTE version: svn-trunk