Config
Log for #openttd.dev on 11th January 2013:
Times are UTC Toggle Colours
07:38:26  *** DorpsGek changes topic to "OpenTTD Dev Channel || Latest SVN: r24905 || Logs: http://webster.openttdcoop.org/?channel=openttd.dev || Voice (talk-right) upon request via #openttd; make sure you are registered to NickServ before asking"
09:06:48  *** kamnet has joined #openttd.dev
09:17:58  *** kamnet has left #openttd.dev
09:39:47  *** Supercheese has left #openttd.dev
11:07:49  *** fonsinchen has joined #openttd.dev
11:07:49  *** ChanServ sets mode: +v fonsinchen
11:12:29  <fonsinchen> I've been thinking some more about the cargo loading algorithm.
11:12:49  <fonsinchen> Currently we have two settings governing that: improved and gradual loading.
11:13:31  <fonsinchen> Rubidium said that some people want improved loading to be switched off as that improved throughput in high throughput stations.
11:13:59  <fonsinchen> I assume that's because less cargo needs to be present in a station for all vehicles to start loading.
11:14:17  <fonsinchen> (in case of full load, that is)
11:15:21  <fonsinchen> Why do we keep the gradual loading setting, though?
11:15:45  <fonsinchen> I guess that is to equally distribute cargo among multiple vehicles which are non-full loading at the same station.
11:17:11  <fonsinchen> Is that actually right? Or is there some other story to those settings?
11:19:33  <fonsinchen> Basically the "gradual" setting should not matter when doing full load and the "improved" setting should not matter when doing non-full load.
11:53:12  *** DorpsGek changes topic to "OpenTTD Dev Channel || Latest SVN: r24906 || Logs: http://webster.openttdcoop.org/?channel=openttd.dev || Voice (talk-right) upon request via #openttd; make sure you are registered to NickServ before asking"
12:33:43  *** ChanServ sets mode: +v Maedhros
13:58:08  *** ntoskrnl has joined #openttd.dev
14:35:14  *** Belugas has quit IRC
15:07:04  *** Belugas has joined #openttd.dev
15:07:04  *** ChanServ sets mode: +v Belugas
15:47:27  *** fonsinchen has quit IRC
17:58:00  *** frosch123 has joined #openttd.dev
17:58:00  *** ChanServ sets mode: +v frosch123
18:47:01  *** DorpsGek changes topic to "OpenTTD Dev Channel || Latest SVN: r24907 || Logs: http://webster.openttdcoop.org/?channel=openttd.dev || Voice (talk-right) upon request via #openttd; make sure you are registered to NickServ before asking"
19:17:43  *** Supercheese has joined #openttd.dev
19:58:34  *** Alberth has joined #openttd.dev
19:58:34  *** ChanServ sets mode: +v Alberth
19:59:35  <planetmaker> what does it need to add a new language file to OpenTTD?
20:01:37  <michi_cc> Nothing except possible WT3 configuration for unfinished langs, otherwise a rerun of projects/generate.
20:03:04  <Alberth> o/
20:03:22  <Terkhen> hi Alberth
20:03:27  <planetmaker> it might actually go straight into the normal languages... 4681 lines and 4712 for english
20:03:28  <planetmaker> hi Alberth
20:03:43  <planetmaker> need to verify that all is actually translated...
20:03:58  <frosch123> planetmaker: there is a lot of strings we do not need translating
20:04:05  *** ntoskrnl has quit IRC
20:04:06  <frosch123> usually rb started new languages with them present
20:04:19  <planetmaker> verfication failed ;-)
20:05:36  <planetmaker> Well, that person takes english.txt and translates it string by string. He was asked to provide some initial translation before we add the language
20:06:29  <planetmaker> he seems to be at line 700 or so
20:07:46  <michi_cc> English strings shouldn't be commit though. And you have to check the meta-infomation (lang code, winlang, plural, cases etc).
20:09:34  <planetmaker> yes of course. With 700/4800 it's not going to get committed yet. Other language meta info are correct (and we had some mail exchange about that)
20:12:36  <planetmaker> I wonder though why I don't end up with a gaelic_scottish.lng... hm
20:14:34  <Alberth> or a swedish_muppet_cook.lng :p
20:15:13  <michi_cc> It can be committed (to unfinished) as long as you simply cull the english strings. WT should fix file formatting.
20:20:40  <planetmaker> the problem with adding all language request "just so" is that each consumes a newgrf language slot. Those which are finite
20:23:19  <Alberth> how finite are those?
20:23:24  <frosch123> there are only 136 iso codes (not counting country variants)
20:23:29  <frosch123> there are 127 newgrf languages
20:23:48  <frosch123> no idea how many window language codes there are
20:23:55  <frosch123> and whether they match the iso codes
20:23:58  <michi_cc> Sure, but I'd consider 700 strings good enough.
20:24:41  <planetmaker> to recap what I suggested in the other channel: let's commit it to unfinished. And make the policy that 'unfinished' doesn't mean final assignment of a grflangid
20:25:15  <planetmaker> (though of course a proper one can be chosen as long as we're fine with them)
20:25:48  <planetmaker> I'm talking in this context of adding gd (scottish gaelic)
20:26:06  <planetmaker> which is winlangid 0x043C - and I'd assign grflangid 0x43
20:28:54  <michi_cc> As good as any other, I can't discern any pattern to the id list.
20:29:07  <frosch123> something said it uses phone numbers
20:29:12  <frosch123> no idea whether that is true
20:29:21  <frosch123> but i thought pm wanted to leave it blank
20:29:32  <michi_cc> German is 02, so there's at least one exception :)
20:29:53  <frosch123> russian is correct
20:30:05  <Alberth> dutch too
20:30:06  <planetmaker> oh, phone numbers... then it should be 0x44
20:30:19  <planetmaker> or 0x4
20:30:29  <planetmaker> 0x04. Which... is taken. so 0x44?
20:30:33  <planetmaker> that's free, to
20:30:38  <Alberth> michi_cc: isn't german one of the original 6 or 7 languages?
20:30:41  <planetmaker> bah. spelling. bad is. today ;-)
20:30:45  <planetmaker> Alberth, yes
20:30:45  <frosch123> Alberth: 49
20:30:53  <frosch123> 7 is russia
20:31:02  <frosch123> the only other country i have been called from
20:31:22  <frosch123> hmm... i do not remember malaysia
20:31:36  <planetmaker> +44 is Britain. +49 is Germany. +32 is.. France? or was it +33?
20:31:45  <Alberth> frosch123: you should phone to it more often :p
20:32:13  <frosch123> it's just what the phone shows in its display at work when i am called
20:32:43  <frosch123> planetmaker: hex/dec?
20:33:07  <frosch123> @base 16 10 35
20:33:08  <DorpsGek> frosch123: 53
20:34:11  <frosch123> yeah, finnish is 35, but also uses 35h
20:34:34  <frosch123> @base 10 16 90
20:34:34  <DorpsGek> frosch123: 5A
20:34:54  <frosch123> yeah, some are just random :p
20:35:19  <frosch123> @calc 10 16 66
20:35:19  <DorpsGek> frosch123: Error: invalid syntax (<string>, line 1)
20:35:26  <frosch123> @base 10 16 66
20:35:26  <DorpsGek> frosch123: 42
20:35:35  <frosch123> some are correct :)
20:35:37  <planetmaker> yes, grflangIDs are... not all following a common system
20:35:58  <Rubidium> the grflandids are a mess
20:36:06  <frosch123> @base 10 16 44
20:36:06  <DorpsGek> frosch123: 2C
20:36:51  <Rubidium> but mostly following the phone number or something resembling that or just the first free number
20:37:05  <Alberth> nah, international phone codes are wrong, our numbers are good :p
20:37:05  <planetmaker> honestly it doesn't quite matter which is chosen as long as it's not yet taken
20:48:50  <planetmaker> hm... where are the plural cases defined (in the code)? It seems the wiki is not accurate... http://wiki.openttd.org/OpenTTDDevBlackBook/Format_of_langfiles#List_of_Plural_Types
20:49:06  <planetmaker> code value 4 requires 5 plural entries as opposed to 3 in the wiki
20:49:10  <frosch123> run strgen
20:49:17  <frosch123> it prints them to the console with some command line option
20:49:35  <frosch123> newgrf wiki likely also contains them
20:50:20  <frosch123> http://paste.openttdcoop.org/show/2045/
20:50:50  <frosch123> that's also the info shown in wt3
20:51:09  <planetmaker> and... what's the related code? Line number?
20:51:13  <frosch123> yes
20:51:43  <frosch123> http://newgrf-specs.tt-wiki.net/wiki/StringCodes#Using_plural_forms <- also there
20:52:00  <frosch123> hmm, let's match them
20:52:12  <frosch123> strgen output should be the best, we validated them some time ago
20:52:55  <planetmaker> strgen doesn't give the code numbers - something which would be helpful :D
20:52:57  <frosch123> yeah, form 3 seems to be shuffled on newgrf specs
20:53:14  <frosch123> planetmaker: it has the format that wt3 can parse :)
20:54:14  <planetmaker> :-)
20:54:27  <frosch123> wtf...
20:54:38  <frosch123> source code says even different for plural 3
20:54:49  <planetmaker> ok, anyhow, we agreed that it's worth that I prepare a patch to commit the lang to unfinished?
20:55:42  <planetmaker> you're fixing that? Or shall I?
20:55:43  <frosch123> newgrf say: [^1][1], 0, rest; strgen says 0, [^1]1, rest; code says: [^1]1, rest, 0
20:57:21  <frosch123> planetmaker: question is which one is correct :p
20:59:24  <planetmaker> ok. And *I* think that then the gaelic plural number code is wrong... he wants 12. I'll need to verify that with him, I guess
20:59:53  <frosch123> irish galic uses 4
21:07:01  <frosch123> ok, form 3 is the only one which is totally messed up
21:30:47  *** Alberth has left #openttd.dev
22:53:10  *** frosch123 has quit IRC

Powered by YARRSTE version: svn-trunk