Log for #openttdcoop.devzone on 6th February 2012:
Times are UTC Toggle Colours
06:35:46  *** JVassie has joined #openttdcoop.devzone
07:18:38  *** JVassie has quit IRC
07:19:50  <Brot6> Central European Train Set - Revision 629:c01a89b72287: adjust loading speed the same way as capa... (Eddi) @
07:28:34  <Brot6> cets: update from r628 to r629 done (251 warnings) -
08:28:51  <Rubidium> given he hasn't replied I feat he hasn't read it
08:28:58  <Rubidium> s/feat/fear/
08:35:24  <planetmaker> probably
08:39:18  <Brot6> Dutch Trains 2.0 - Support #3632: list of vehicles to be drawn (Voyager1) @
09:13:36  <Ammler> Rubidium: I will submit openttd to suse factory sometime this week, not sure how long the review will need, would be nice to get it into next release (12.2 July), you can wait for that or I will post another diff if needed
09:13:55  <Ammler> there are some tiny cleanups needed like removing suse copyright from header
09:18:31  <Ammler> also need to prepare opengfx, openmsx and nml, that's all?
09:21:45  <planetmaker> well. OpenSFX?
09:21:57  <planetmaker> in any case, both OpenSFX and OpenMSX didn't change
09:22:00  <Ammler> Rubidium: does debian commit minor updates during the year to the distro repo or CVE-patches only?
09:24:00  <Ammler> planetmaker: you think, I should try to push opensfx too and hope, the legal time might not see the "bad" license?
09:24:23  <planetmaker> no.
09:24:48  <planetmaker> I don't know what license is acceptable. If CCSP is not, then of course not
09:25:10  <Ammler> OSI capable
09:25:11  <planetmaker> for suse itself it might indeed not
09:25:23  <planetmaker> the commercial aspect might kill it
09:26:01  <Ammler> yes, it does, as I asked about it, some thought it is very in the grey that it even is allowed on the games repo
09:26:36  <Ammler> but if it is in the standard repo, it could be for example solded with SLE
09:26:44  <Rubidium> Ammler: it does, for unstable. For stable it's mostly security issues
09:26:54  <Rubidium> or "grave" bugs
09:26:55  <planetmaker> <-- dunno, Ammler
09:27:08  <planetmaker> ok, then clearly not :-)
09:30:37  <Ammler> afaik, not every factory package is finally on the DVD, that needs an additional submit/review afaik
09:31:16  <Ammler> but you should be able to install openttd without adding any additional repo
09:31:18  <planetmaker> well, DVD or not. It's part of a commercial endeavour intended to incur money. Thus I'd deem it not acceptable
09:31:55  <Ammler> ah yes, the packages should fit the legal therm for DVD
09:38:34  <Rubidium> a "missing" opensfx isn't that bad, likewise is a "missing" openmsx not a major issue
09:39:25  <Rubidium> opengfx missing is more of a problem, although... with 1.2.x it's not as big of a problem either as it allows downloading the first time you start it
09:41:19  <Ammler> maybe you could add a first time node and download opensfx too?
09:41:25  <Ammler> note*
09:42:26  <Ammler> openmsx is no big deal, clear license and the dependencies are already in the repo
09:44:13  <Ammler> (in my case timidity
10:11:08  <Ammler> if you split binary and data, you get pretty much 2 samesized packages, 1.8MB
10:11:28  <Ammler>
10:26:58  <Rubidium> so it saves 1.8 * (#architectures - 1) MiB per mirror
10:29:10  <Ammler> well, actually I waste 1.8M more
10:29:31  <Ammler> since I removed subpackage data which was noarch
10:41:45  <Rubidium> they required that?
10:42:12  <Rubidium> as in Debian it's somewhat the other way around, but then Debian has more architectures
11:29:05  <Ammler> if main package is noarch, you can't make subpackages with archs
11:29:40  <Rubidium> so don't make it the main package
11:29:53  <Rubidium> main package: arch openttd, sub package: noarch openttd-data
11:30:48  <Ammler> yes, but then the main package would be empty, but I need it for the dependencies
11:31:22  <Ammler> because the only arch file in openttd is the binary openttd, which needs to be in sub package
11:32:21  <Ammler> I could have made a package called something like openttd-source and then make specific subpackages openttd, openttd-gui and openttd-dedicated
11:32:40  <Ammler> but it is imo not worth the 2MB
11:33:18  <Ammler> maybe sometime, rpm will support noarch main with arch sub
11:44:07  <Ammler> planetmaker: any plan for opengfx bugfix release?
11:45:53  <Ammler> doesn't need any big annoucements :-)
11:53:51  <Ammler> I am not sure, if I shall submit gimp dependency
12:01:04  <planetmaker> what needs bug-fixing?
12:01:10  <planetmaker> we can do that anytime
12:03:56  <Ammler> .NOTPARALLEL
12:04:29  <Ammler> and not clean maintainer files if you clean gimp files
12:05:04  <Ammler> (already pushed)
12:07:17  <Rubidium> Ammler: did you see my request to "fix" the spec for grfcodec?
12:07:26  <Rubidium> I added a binary and man page
12:10:00  <planetmaker> ah that. Yes, please remind me Wednesday, if I don't find the time today
12:11:02  <Ammler> well, if I don't push gimp dep to factory, it isn't needed...
12:12:49  <Ammler> Rubidium: where did you make that request?
12:13:49  <Ammler> somehow devzone fails to assign compile issue tickets
12:14:59  <Rubidium> Ammler: #openttdcoop.devzone@201202051734UTC
12:18:22  <Brot6> GRFCodec - Bug #3647 (Closed): DevZone compile failed (compiler) @
12:18:22  <Brot6> GRFCodec - Revision 892:9a7567d7da15: Fix #3647: add grfstrip to rpm package (Ammler) @
12:18:22  <Brot6> GRFCodec - Bug #3647 (Closed): DevZone compile failed (Ammler) @
12:22:03  <Brot6> grfcodec: update from r876 to r892 done -
13:31:47  <Rubidium> thanks Ammler
13:32:44  <Ammler> yaw :-)
13:33:14  <Ammler> Rubidium: does Debian use gimp for opengfx?
13:36:30  <Rubidium> it doesn't even use nml yet
13:36:42  <Rubidium> but I guess it would be using gimp
14:28:24  <Ammler> planetmaker: "... Gimp not found. Graphis cannot be created. Terminating.", I would replace Terminating with Skipping
14:33:12  <planetmaker> Ammler, it can't skip that, if maintainer-clean was run
14:35:40  <Ammler> but the error apears also if not
14:38:18  <Ammler> planetmaker: also target check seems buggy, it has dependency of the md5 file
14:38:41  <Ammler> hmm, maybe I am wrong...
14:39:41  <planetmaker> is there a comprehensive summary of what I need to check and take care of?
14:40:04  <planetmaker> If not: can you create one? You seem to know what you need for building the release bundles better than I ;-)
14:40:14  <Ammler> well, basically I just need a release
14:40:33  <Ammler> other things are minor glitches if at all
14:41:41  <Ammler> and would not need to be in 0.4 anyway
14:42:15  <planetmaker> Anyway: I need a list of things I shall check / fix
14:42:42  <planetmaker> These "glitches" need a description so that I can reproduce. Or it won't happen as I'll forget or conclude 'works for me'
14:43:00  <Ammler> yes, that works for me :-)
14:43:25  <Ammler> some things I just need to understand...
14:43:39  <Ammler> so right now, I see nothing needed
14:44:07  <planetmaker> eh?
14:44:19  <Ammler> everything needed is already pushed
14:45:21  <Ammler> and the glitches are me interpreting it wrongly
15:14:56  <Ammler> I just wonder, what is the best way to tell people about sound
15:15:41  <Ammler> coolest would be if openttd has a kind "tip of the day" window :-)
15:53:17  <Rubidium> those windows annoy me
15:58:30  <V453000> well most games have such a thing when the player is waiting for something like a loading ... which isnt too common in openttd :)
16:08:16  <Ammler> planetmaker: did we ever search for the different md5sum issue?
16:08:29  <Ammler> (which is nml issue, not gimp)
16:11:00  <Ammler>
16:11:17  <Ammler>
16:21:57  <Ammler> Yexo: you found the issue, didn't you?
16:23:34  <Ammler> afaik, the issue is a difference between python 2.7 and older
16:31:14  <Ammler> and another big advantage we use so bleeding edge distro on devzone :-)
16:35:39  <Rubidium> interestingly... opengfx has to depend on the grfcodec package again, or... reimplement grfid -m ;)
16:51:34  *** frosch123 has joined #openttdcoop.devzone
17:21:55  <Brot6> bandit: update from r218 to r223 done (1 warnings) -
17:27:45  <Brot6> cets: update from r621 to r629 done (251 warnings) -
17:29:33  <Brot6> dutchtrains: update from r170 to r177 done (108 warnings) -
17:32:26  <Brot6> British Rail OpenTTD Set - Bug #3651 (New): DevZone compile failed (compiler) @
17:32:57  <Brot6> TransRapid Track Set - Bug #3652 (New): DevZone compile failed (compiler) @
17:33:15  *** LordAro has joined #openttdcoop.devzone
17:34:09  <Brot6> 2cc train set - Bug #3653 (New): DevZone compile failed (compiler) @
17:35:28  <Brot6> World Airliners Set - Bug #3654 (New): DevZone compile failed (compiler) @
17:39:23  *** [1]Mark has joined #openttdcoop.devzone
17:39:25  *** Mark is now known as Guest1718
17:41:20  <Brot6> Industrial Stations Renewal - Bug #3655 (New): DevZone compile failed (compiler) @
17:41:58  <Brot6> 32bpp-extra - Bug #3656 (New): DevZone compile failed (compiler) @
17:43:25  <Brot6> NewGRF build framework - Bug #3657 (New): DevZone compile failed (compiler) @
17:44:36  *** Guest1718 has quit IRC
17:45:17  <Brot6> Metro Track Set - Bug #3658 (New): DevZone compile failed (compiler) @
17:45:25  <LordAro> umm... has the compiler broken? or are all those actual broken? :)
17:47:01  <Brot6> FISH - Bug #3659 (New): DevZone compile failed (compiler) @
17:47:59  <Brot6> Total Town Replacement Set - Bug #3660 (New): DevZone compile failed (compiler) @
17:47:59  <Brot6> OpenGFX Trees - Bug #3661 (New): DevZone compile failed (compiler) @
17:48:36  <Brot6> SMITS - Bug #3662 (New): DevZone compile failed (compiler) @
17:49:48  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: heqs, basecosts, water-features, snowlinemod, chips (1 warnings)
17:50:09  <Brot6> Comic Style Houses - Bug #3663 (New): DevZone compile failed (compiler) @
17:50:12  <frosch123> LordAro: it's 2012. so likely the world ended over there
17:50:36  <LordAro> seems likely :)
17:52:55  <^Spike^> yeah our servers live 10 months ahead of time :D
17:53:00  <^Spike^> lives*
17:56:40  *** ODM has joined #openttdcoop.devzone
18:09:35  *** ODM has quit IRC
18:11:57  <michi_cc> reads like PCX reading is borked somehow.
18:18:44  <Rubidium> so far for stuff being recompiled every night
18:19:01  <Rubidium> if it were, then it should've long ago
18:23:16  <Yexo> <Ammler> [17:21:57] Yexo: you found the issue, didn't you? <- if it's not fixed and not an open issue in the nml issue tracker you can be pretty sure I didn't find a solution
18:24:43  <Rubidium> hi Yexo
18:24:48  <Brot6> GRFCodec - Revision 893:6a6461ca9515: -Fix: loading of pcx sprites failed (Rubidium) @
18:24:53  <Yexo> hello
18:25:22  <Yexo> Rubidium: with your patch applied nml was able to compile ogfx-trains, however the resulting grf can't be loaded by openttd (invalid size on sprite 33)
18:25:28  <Yexo> didn't have time to debug it yet
18:25:31  <Rubidium> <- new mostly working version
18:25:48  <Yexo> is that still the same as this morning?
18:26:05  <Rubidium> mostly yes
18:26:18  <Rubidium> regression fails on 26 and 27
18:26:28  <Rubidium> due to "empty sprites"
18:26:44  <Yexo> empty sprites are valid (or at least there were in the previous container format)
18:27:42  <Rubidium> I'm not sure where they are marked as valid in the specification
18:28:04  <Yexo> quite possible nowhere
18:33:23  <Rubidium> if grfcodec encodes the nfo nml exports it just looks like an action 0 (pseudo sprite + 0 as payload)
18:33:45  <Yexo> yep, it works like that
18:33:56  <Yexo> the game crashes as soon as it tries to load such a sprite
18:34:18  <Yexo> I can't find any reference, but I'm pretty sure it was used before nml
18:34:22  <Yexo> I haven't invented it
18:35:45  <planetmaker> <Rubidium> interestingly... opengfx has to depend on the grfcodec package again, or... reimplement grfid -m ;) <-- for stripping the "unneeded" zoom levels?
18:35:58  <planetmaker> hm, no. The md5sums?
18:36:06  <Rubidium> planetmaker: yes, the md5 checksums
18:36:15  <planetmaker> :-)
18:37:24  <Brot6> Dutch Trains 2.0 - Support #3632: list of vehicles to be drawn (Voyager1) @
18:38:11  <Yexo> <- andy uses it here
18:38:12  <Webster> Title: Transport Tycoon Forums View topic - Different speeds loaded & unloaded? (at
18:41:00  <Yexo> Rubidium: can't find any good reference. Not that important though, nml can simply encode it as a 1x1 transparent sprite
18:45:22  <Rubidium> that'd be better ;)
18:54:55  <frosch123> Yexo: nml is not supposed to set any default values of properties, is it?
18:55:14  <frosch123> i.e. is fs#5048 just that ng did not set "loading_speed" ?
18:56:53  <planetmaker> yes, you usually have to define the whole vehicle
18:57:03  <planetmaker> I also stumbled over that. More than once :-)
18:57:16  <planetmaker> Missing out on one essential property or so :-)
18:57:20  <frosch123> ok, i'll forward that answer :)
18:58:30  <planetmaker> maybe it'll be a good idea to actually offer a default for vehicles. Which could be explicitly included or so
19:04:30  <Yexo> frosch123: nml will start allocating IDs from directly above the maximum old id
19:04:45  <Yexo> so all vehicles that don't have an explicit id will be completely new vehicles without any defaults set
19:04:54  <Yexo> I see you've already answered :)
19:04:55  <Rubidium> then nml should warn about the missing props if they're reasonably important
19:05:03  <Yexo> yes, it should
19:05:09  <Yexo> it should not set defaults, but warn instead
19:05:25  <Yexo> however since properties can be split over multiple blocks that's not easy
19:05:49  <Yexo> unless we make a few more assumptions, like each ID being tied to exactly one name. We should do that anyway I think
19:09:28  <planetmaker> you can tie an ID to more than one name?
19:09:32  *** JVassie has joined #openttdcoop.devzone
19:10:02  <Yexo> item(name_1, 10) {...} item(name_2, 10) { ... } <- that works currently, it doesn't even generate a warning
19:11:07  <planetmaker> I guess I never tried as I always assumed that it a) won't work and b) is bad style ;-)
19:11:25  <Yexo> b) is certainly true, and it'll give unexpected problems if you're not aware of it
19:11:29  <planetmaker> and c) sufficient to define the ID once :-)
19:12:02  <Rubidium> fixed the regression with 26 and 27; all regressions are passed
19:12:11  <Rubidium> Yexo: maybe that test NML works now ;)
19:12:23  <Yexo> perhaps, but I have to go right now
19:12:31  <Yexo> will test maybe later this evening or tomorrow
19:12:33  <Yexo> same url?
19:12:37  <Rubidium> yes
19:14:07  *** [1]Mark is now known as Mark
19:16:50  <Brot6> NewGRF Meta Language - Feature Request #3664 (New): Use more sensible default for loading speed (... (oberhuemer) @
19:20:07  <Brot6> NewGRF Meta Language - Feature Request #3664: Use more sensible default for loading speed (and ot... (planetmaker) @
19:20:52  <planetmaker> eddi was btw, right, NG = oberhuemer. Interesting
19:22:46  <frosch123> collecting information from multiple networks? sounds evil :p
19:22:51  <Brot6> NewGRF Meta Language - Feature Request #3664: Use more sensible default for loading speed (and ot... (planetmaker) @
19:23:04  *** andythenorth has joined #openttdcoop.devzone
19:24:20  <planetmaker> well... :-P
19:24:36  <frosch123> actually, not all properties default to zero
19:24:55  <frosch123> those which were added later (after engine pool), do not :p
19:25:06  <planetmaker> not? What do they get?
19:25:18  <frosch123> whatever default value we considered useful
19:25:28  <planetmaker> umpf.
19:25:39  <frosch123> resp. which representend the behaviour before the property was added
19:26:19  <frosch123> e.g. all road vehicles have  TE of 0x4C by default
19:26:32  <frosch123> else all vehicles from old grfs would suddenlty stop without any TE :p
19:27:19  <frosch123> planetmaker: anyway, you do not have to document that
19:27:46  <frosch123> the rule is just: set all properties which are defined at the time of writing the grf :)
19:29:47  <planetmaker> :-)
19:30:51  <planetmaker> hm, but what would break to simply set the default for the "old" vehicles and old grfs which don't define it or define it as 0 to the 'default'?
19:31:04  <planetmaker> Thus applying the sane default always, if not set or set to 0
19:32:23  <frosch123> because 0 is a useful value in some cases
19:37:52  <Brot6> FIRS Industry Replacement Set - furniture_factory.png (andythenorth) @
19:46:46  *** Mark has quit IRC
19:50:36  *** [1]Mark has joined #openttdcoop.devzone
19:50:39  *** [1]Mark is now known as Mark
19:51:07  *** Mark has quit IRC
19:51:29  *** [1]Mark has joined #openttdcoop.devzone
19:51:51  *** [1]Mark is now known as Mark
20:23:26  *** Jupix has quit IRC
20:27:16  *** Jupix has joined #openttdcoop.devzone
21:24:23  <Brot6> BANDIT - Revision 224:175346dd6fe4: Feature: buy menu text now includes correct truck type info (... (andythenorth) @
21:24:23  <Brot6> BANDIT - Feature #3401 (Closed): Buy menu text (andythenorth) @
21:29:04  <Brot6> BANDIT - Revision 225:3b64100b209c: Codechange: use UPPERCASE identifiers for manually defined st... (andythenorth) @
21:29:32  <Brot6> BANDIT - Bug #3650: String identifiers should be upper case (andythenorth) @
21:29:32  <Brot6> BANDIT - Bug #3650 (Closed): String identifiers should be upper case (andythenorth) @
21:42:20  <Brot6> NewGRF Meta Language - Feature Request #3664: Use more sensible default for loading speed (and ot... (oberhuemer) @
21:43:21  *** Zuu has joined #openttdcoop.devzone
21:46:19  *** LordAro has quit IRC
21:49:58  *** Mark has quit IRC
21:53:52  *** frosch123 has quit IRC
21:54:30  *** [1]Mark has joined #openttdcoop.devzone
21:55:24  *** [1]Mark is now known as Mark
22:11:15  <Brot6> BANDIT - Feature #3665 (New): Drawbar and solo trucks need generated spritesets (andythenorth) @
22:13:02  *** andythenorth has quit IRC
22:47:59  <Brot6> NewGRF Meta Language - Feature Request #3664: Use more sensible default for loading speed (and ot... (planetmaker) @
23:09:00  <Brot6> NewGRF Meta Language - Feature Request #3664: Use more sensible default for loading speed (and ot... (yexo) @
23:16:48  *** JVassie has quit IRC
23:25:09  *** Zuu has quit IRC

Powered by YARRSTE version: svn-trunk