Log for #openttdcoop.devzone on 27th October 2013:
Times are UTC Toggle Colours
06:01:53  *** andythenorth has joined #openttdcoop.devzone
06:26:45  *** Supercheese has quit IRC
08:04:01  <DevZone> Project Iron Horse build #348-push: SUCCESS in 1 min 25 sec:
08:04:39  *** andythenorth has quit IRC
08:57:09  *** Alberth has joined #openttdcoop.devzone
09:23:44  *** Alberth has quit IRC
09:25:43  *** Alberth has joined #openttdcoop.devzone
09:32:21  *** zooks has joined #openttdcoop.devzone
10:09:54  *** zooks has quit IRC
10:17:14  *** frosch123 has joined #openttdcoop.devzone
10:19:00  *** zooks has joined #openttdcoop.devzone
10:28:33  <Alberth> frosch123: I fixed the eints bug of not copying the language properties while uploading a language file.
10:33:30  <frosch123> updated the server, you can commit your fix to firs again :)
10:42:29  <Alberth> I also had a go at the {P} reference problem of NML, but am still struggling to understand the code
10:42:49  <Alberth> so far I only have some documentation text
10:44:35  <frosch123> the ides is to fix nml, right?
10:45:28  <Alberth> I think that would be the best solution, as translation input then matches with openttd conventions
10:46:35  <frosch123> yup :)
10:49:05  *** andythenorth has joined #openttdcoop.devzone
10:49:48  <Alberth>   is the doc stuff I made yesterday
10:57:24  <Alberth> firs plural form commit reverted
11:00:35  <DevZone> Project FIRS Industry Replacement Set build #38-push: SUCCESS in 6 min 45 sec:
11:15:09  *** zooks has quit IRC
13:51:13  *** andythenorth has quit IRC
13:52:11  *** andythenorth has joined #openttdcoop.devzone
13:55:50  *** andythenorth has quit IRC
15:17:47  *** oskari89 has joined #openttdcoop.devzone
15:31:51  *** Supercheese has joined #openttdcoop.devzone
15:42:42  *** hjsb8q4rggrt has joined #openttdcoop.devzone
15:43:47  *** hjsb8q4rggrt has quit IRC
15:44:09  *** hjsb8q4rggrt has joined #openttdcoop.devzone
15:44:29  <DevZone> Project grfcodec build #6-push: SUCCESS in 50 sec:
15:44:42  *** hjsb8q4rggrt has quit IRC
15:46:16  <DevZone> Project NML - NewGRF Meta Language build #103-push: SUCCESS in 2 min 46 sec:
16:30:07  <planetmaker> hm... should we adopt like the versioning scheme for nml like gcc does: development is odd numbers, stable is even?
16:30:46  <planetmaker> I need a new NML for OpenGFX... backporting is boring
16:34:21  <Rubidium> planetmaker: except that gcc, like the kernel, stopped with that practice
16:34:50  <planetmaker> :D
16:34:56  <frosch123> yeah, i do not like that schema either :)
16:35:08  <planetmaker> I'm not too much in favour either
16:36:18  <planetmaker> But I'll have some belly ache to call current NML default branch 'stable'. Maybe I just shouldn't care and tag it 0.3.0 and be done with it. And re-tag whenever I need
16:36:34  <Alberth> RC-x ?
16:37:25  <Alberth> or "0.x.y-alpha" or "0.x.y-beta"   :)
16:37:53  <planetmaker> :-) Yeah, maybe that's better then
16:38:09  <Alberth> hmm, not enough numbers with alpha or beta, better stick with -RCn
16:38:11  <planetmaker> But the linux distros will need to supply it, too
16:38:20  <frosch123> 0.3.0-alpha-1
16:38:24  <planetmaker> :D
16:38:33  <frosch123> there is always a number behind the alpha
16:39:01  <Alberth> planetmaker: ok, it's unlikely they will ship things labeled "alpha/beta/RC" imho
16:39:09  <DevZone> Project OpenGFX build #81-push: SUCCESS in 6 min 32 sec:
16:39:29  <planetmaker> they'll need it for OpenGFX
16:39:41  <planetmaker> otherwise I'll not need to tag it :D
16:40:00  <frosch123> well, still some months left :)
16:40:26  <planetmaker> well.... yes, no. It's very bad, if bananas has no decent release of OpenGFX
16:40:46  <planetmaker> IMHO adding a button to OpenTTD must be followed by uploading an updated OpenGFX to bananas
16:41:00  <planetmaker> Otherwise it's very bad game support and we leave 2/3 of our players with a broken base set
16:41:01  <frosch123> then also make an ogfx nightly
16:41:16  <frosch123> it does not need to be a ogfx release
16:41:32  <frosch123> who says it's the last sprite we add?
16:41:53  <planetmaker> no-one. But a new tag comes easy, I'm not concerned about that really
16:42:19  <frosch123> well, then make a new ogfx, but no new nml ?
16:43:17  *** andythenorth has joined #openttdcoop.devzone
16:44:28  <planetmaker> well, ok.
16:44:56  <planetmaker> just a nightly update on bananas for OpenGFX; that suffices
16:45:23  <planetmaker> yet still, the nml-opengfx dependency needs a decision... what to do with it at least around Christmas
16:45:41  <Rubidium> technically downstreams don't need to bother with opengfx releases until a proper release is needed
16:46:06  <Rubidium> needed as in: significant fix *or* stable release of openttd
16:47:15  <planetmaker> sure. But it will be helpful to know how to deal with it before it's too late
16:47:31  <Rubidium> what *could* be done is release nightlies to bananas with a minimum version of <current openttd trunk>
16:47:57  <planetmaker> It needs to be anyway >= 1.2.0
16:48:22  <Rubidium> isn't the nml trunk?
16:48:52  <Rubidium> oh... it ain't
16:48:58  <Rubidium> I guess
16:49:07  <planetmaker> what do you mean? The resulting OpenGFX, as it uses DOS palette, imposes that requirement
16:49:41  <Rubidium> was thinking about what versions of nml created grfs valid for what versions of openttd
16:49:56  <Rubidium> grf container v2 / grf v8
16:50:30  <planetmaker> v2/grf v8/DOS
16:50:35  <Rubidium> oh, those are 1.2.0
16:51:05  <Rubidium> so what's the big issue?
16:51:55  <Rubidium> that you need to mark nml 'trunk' as stable in the next 6 months for opengfx to be built by our downstreams?
16:52:54  <planetmaker> yes
16:53:01  <planetmaker> that exactly :-)
16:53:36  <Rubidium> is it so unworthy for release that it can't be brought into a releaseable state?
16:54:15  <frosch123> nml has even more active developera as openttd :)
16:55:30  <planetmaker> it does? how so?
16:55:47  <planetmaker> well, there's a few
16:56:29  <Rubidium> a mere 16 bugs
16:56:34  <planetmaker> 5513, 5294, 5246, 4667 at least
16:56:39  <Rubidium> compared to 60
16:56:47  <planetmaker> yeah :S
16:58:27  <Rubidium> 5513 seems like adding a 'cd pathof($exe)' in front of the hg call (or calling it with a path?)
17:00:05  <planetmaker> something along those lines, yes
17:00:20  *** andythenorth has quit IRC
17:02:37  <Rubidium> hmm...
17:02:50  <Rubidium> or os.path.realpath doesn't work on windows
17:09:09  <Rubidium> on unices argv[0] contains the full path (if just nmlc from the global scope) or a relative path
17:09:28  <Rubidium> I reckon that on windows argv[0] is merely nmlc, instead of the path to it
17:16:10  <Rubidium> planetmaker:
17:20:20  <DevZone> Project NML - NewGRF Meta Language build #104-nightlies: SUCCESS in 1 min 44 sec:
17:28:07  *** andythenorth has joined #openttdcoop.devzone
17:28:22  <planetmaker> hg push
17:28:25  <planetmaker> :D
17:31:14  <Rubidium> it's at least 66% untested ;)
17:41:02  <planetmaker> let's setup a nmlc dev environment in my windows VM :D
17:50:09  *** ODM has joined #openttdcoop.devzone
17:50:33  *** andythenorth has quit IRC
18:08:51  <DevZone> Project finnishtrams build #7-nightlies: SUCCESS in 2 min 14 sec:
18:24:31  <DevZone> Project eints build #9-nightlies: SUCCESS in 13 sec:
18:27:17  <DevZone> Project Finnish Rail Infrastructure - Rails build #126-nightlies: SUCCESS in 41 sec:
18:32:10  <DevZone> Project Nutracks build #53-nightlies: SUCCESS in 1 min 34 sec:
18:49:14  <DevZone> Project Iron Horse build #349-nightlies: SUCCESS in 1 min 37 sec:
18:51:55  <planetmaker> can someone remind me in which form I give bananas the nightly version of OpenTTD?
18:52:00  <planetmaker> s/the/a/?
18:52:13  <planetmaker> hex newgrf version?
18:52:15  <planetmaker> or svn?
18:52:23  <planetmaker> or decimal newgrf version?
18:58:20  <Rubidium> IIRC it's just the decimal svn version
18:58:49  <frosch123> yup, decimal newgrf version
18:59:22  <frosch123> at least when you select "custom"
18:59:32  <frosch123> though iirc there is also a "nightly" item, where you enter the revision
19:00:34  <planetmaker> yes, there's both. Then I'll select nightly and use your commit for the OpenGFX as minimum
19:01:44  <DevZone> Project xussrset - Trains from Russia build #71-push: SUCCESS in 4 min 30 sec:
19:01:57  <planetmaker> then at least nightly players have the option to update
19:02:42  <planetmaker> and I forgot to commit credits...
19:04:46  *** zooks has joined #openttdcoop.devzone
19:05:35  <DevZone> Project OpenGFX build #82-nightlies: SUCCESS in 6 min 58 sec:
19:12:25  *** gelignite has joined #openttdcoop.devzone
19:13:51  *** andythenorth has joined #openttdcoop.devzone
19:21:01  *** zooks has quit IRC
19:32:37  <planetmaker> hmpf... "Your Base Graphics pack contains a (New)GRF file of which the md5 does not match what the obg defines."
19:40:00  <planetmaker> ok, this must be a grfid version thing. On my machine the grfid -m output matches the one in the obg
19:40:34  <planetmaker> Rubidium, can you check which grfid version checks uploads on bananas?
19:41:05  <Rubidium> musa or the non-musa?
19:41:13  <planetmaker> web interface
19:41:23  <planetmaker> I actually should try and learn how to use musa
19:41:36  <Rubidium> 878?
19:42:03  <planetmaker> so... quite old?
19:42:14  <planetmaker> there were some bug fixes for that since, I think
19:43:57  <Rubidium> based on the hg changelog it's just 2 commits behind which both do something with corrupted files
19:44:36  <planetmaker> hm... what's the hash? My tip is like 968 here locally :D
19:45:09  <Rubidium> changelog of grfid.cpp
19:46:12  <planetmaker> 962/3 here the last. and 872 before
19:46:39  <Rubidium> anyhow, I've updated it to HEAD
19:48:47  <DevZone> Project grfcodec build #7-push: SUCCESS in 2 sec:
19:54:33  <planetmaker> hm, I can use musa to add more authors to one of the items?
19:56:54  <Rubidium> with musa you can set multiple authors for newly uploaded versions; you can't change anything that's already uploaded
19:58:06  <planetmaker> Where is that author information used?
19:58:28  <Rubidium> checking whether someone may upload a new version
19:58:51  <Rubidium> by the way... jenkins is doing odd things
19:59:08  <Rubidium> first of all, its time seems to be off by 60 minutes
20:00:07  <Rubidium> second, it seems to compile rev-1 or so
20:00:42  <Rubidium> hmm, never mind the off-by-one... now UTC is off-by-one from CET (woopsie)
20:00:48  <DevZone> Project xussrset - Trains from Russia build #72-push: SUCCESS in 3 min 12 sec:
20:01:22  <Rubidium> in any case at 19:48:40 UTC I committed and then directly pushed a grfcodec
20:01:36  <Rubidium> at 19:48:44 UTC jenkins ran, doing the previous version
20:02:18  <Rubidium> I think it's quite unlikely for a jenkins push build to be triggered so near a push but still be a delayed one from the previous push, which happened over 4 hours before that push
20:05:24  <planetmaker> Rubidium, we pushed within 5 minutes to grfcodec
20:05:44  <planetmaker> err.. and 4 hours :D
20:07:59  <Rubidium> the 15:43:38 build (today), built r967 from days ago (r968 was committed at 15:42:05)
20:08:25  <planetmaker> hm... not good
20:09:09  <Rubidium> unless I'm reading the logs wrong
20:11:31  <Rubidium> can't check is as it's not actually built
20:12:34  <Rubidium> though... the last build it says 19:48:46 + echo 'Revision: 968:c3a979f65db7 tip'
20:12:58  <Rubidium> the build before that... it said 15:44:27 + echo 'Revision: 968:c3a979f65db7 tip'
20:13:00  <planetmaker> yeah, that's right
20:13:07  <planetmaker> that's not supposed to look like that
20:13:23  <Rubidium> twice the same version is definitely not right
20:14:22  <planetmaker> 15:44:26 changeset:   967:92074a35acd1
20:14:29  <planetmaker> 19:48:45 changeset:   968:c3a979f65db7
20:14:37  <planetmaker> so it built different. But wrong
20:14:55  <planetmaker> hm... meh
20:15:04  <planetmaker> I need to change how it gets the version, I guess
20:15:18  <planetmaker> skip the commit hook. And simply poll repos
20:15:19  <Rubidium> those changesets you just showed are $(rev)-1 of what was pushed
20:15:24  <planetmaker> yes
20:15:42  <Rubidium> I wonder if the commit hook is pre or post commit
20:15:49  <Rubidium> I got the feeling it is pre commit
20:15:56  <planetmaker> so do I
20:16:23  <planetmaker> there's two different commit hooks... one pre and one post actually. Might have gotten that wrong
20:20:51  *** oskari89 has quit IRC
20:26:26  *** ODM has quit IRC
20:27:43  <planetmaker> andythenorth, I recon that guy is oskari89
20:31:15  <andythenorth> oic :)
20:37:53  <planetmaker> I miss some configuration...
20:41:19  <Rubidium> seems like the TCP/IP connection got cut off
20:43:41  <planetmaker> hm
20:43:48  <planetmaker> I wonder why
21:00:21  <DevZone> Project grfcodec build #8-push: SUCCESS in 40 sec:
21:41:49  *** Alberth has left #openttdcoop.devzone
22:19:45  *** andythenorth has left #openttdcoop.devzone
22:20:08  *** gelignite has quit IRC
23:05:34  *** frosch123 has quit IRC

Powered by YARRSTE version: svn-trunk