IRC logs for #openttd on OFTC at 2013-07-20
⏴ go to previous day
00:33:05 *** MatrixCL has joined #openttd
00:35:44 *** HerzogDeXtEr1 has joined #openttd
00:42:55 *** SamanthaD has joined #openttd
01:39:03 <tswett> Ahoy. Is it still possible to reach the difficulty screen in 1.3.1?
01:49:10 *** roboboy has joined #openttd
01:50:58 <Supercheese> I think that was integrated into the advanced options panel
01:51:27 <Supercheese> Things like interest rate, max loan, subsidy multiplier are now under Economy
02:31:40 <tswett> All right, gotta go. See you guys.
03:15:06 *** amiller has joined #openttd
04:56:16 *** Eddi|zuHause has joined #openttd
05:42:24 *** kais58|A1K has joined #openttd
05:57:32 *** Tom_Soft has joined #openttd
06:04:15 *** Tom_Soft_backup has joined #openttd
07:04:27 *** sla_ro|master has joined #openttd
07:16:47 <Zuu> Do anyone remember where the utility is to convert openttd.exe to a console application? (so that I can direct the log output to a file)
07:17:17 <Rubidium> somewhere devs/~glx ?
07:17:52 *** Progman has joined #openttd
07:23:54 <Zuu> Hmm, and I already got a copy of it where I wanted to save it :-)
07:24:52 <Rubidium> *you* need to wake up ;)
07:26:52 <Zuu> I need my coffe from my coffe maker
07:40:48 *** andythen_ has joined #openttd
07:41:45 *** andythenorth has joined #openttd
07:43:26 *** andythenorth has joined #openttd
07:54:42 *** frosch123 has joined #openttd
08:26:20 *** Tom_Soft_backup has quit IRC
08:47:09 * Zuu found a possible quite old bug in the AI/GS library initialization. Question is, will any library on bananas break if I fix it? :-)
08:48:04 <frosch123> 90% are maintained by you :)
08:52:56 *** Devroush has joined #openttd
08:55:16 <Zuu> The binary heap library do rely on this bug. Its a core library used by many libraries, so fixing this bug will break most AIs/GSs :p
08:57:51 <Zuu> So I think I will just add a task, note the work around and then mark it as "won't fix"
09:00:04 <gynter> g. changes newlines from .txt files to "DOS" (\r\n) newlines.
09:00:16 <gynter> bananas repackaging no longer does that?
09:07:31 <frosch123> it does that for all license, readme and changelog files
09:07:42 <frosch123> it does the reverse for .nut files
09:09:28 <gynter> forgot to add -b to vim
09:09:53 <gynter> a. keeps files called "readme", "license" and "copying" with .txt or .pdf as extension or without an extension.
09:10:12 <gynter> I had "license" and "readme", repackaing renamed those to license.txt and readme.txt
09:11:11 <gynter> also changelog should be added to that list
09:18:01 <frosch123> using GetName will be hard anyway, since it contains spaces
09:18:07 <frosch123> how should libraries refer to it?
09:21:36 <frosch123> Zuu: if a script uses a library, it needs to know the classnames to call any function, doesn't it?
09:23:48 <Zuu> CreateInstance() in library.nut will return the name of the class. That is used by OpenTTD to load the main class of a library. In our import() command we also have a parameter that allows scripts to decide under what name the main class should be available as. Eg. so that you can rename the library main class to the name that you prefere.
09:24:48 <Zuu> Eg. imort("Library", "SCPLib", "MySCPLib", 1) will look for a library with the script info string "library.scplib.1" and make its main class available in the global scope as "MySCPLib"
09:25:18 <Zuu> There is no problem if GetName() returns a string with spaces as far as I can see.
09:26:16 <Zuu> Other than that it breaks all imports of that library when "fixing" this problem
09:27:00 <frosch123> can we fix it via the api version?
09:27:53 <frosch123> hmm, libaries do not specify an api version
09:28:29 <Zuu> In this scope we can get the API version reported by the library and act differently depending on that.
09:29:42 <Zuu> Libraries don't need to specifiy it as it isn't used by OpenTTD as far as I know. I usually include it for documentation purposes. However the gs/ai will decide which compat.nut to load as libraries live in the global scope.
09:30:12 <frosch123> well, but can't libraries import other libraries?
09:30:59 <Zuu> Anyway, I think fixing the bug will cause more problems than keeping it as is.
09:31:03 <frosch123> so no ai could use libraries anymore which use the old binary heap
09:31:46 <frosch123> well, can we adjust some documentation?
09:31:48 <Zuu> and updating all libraries as well as AIs/GSs won't happen very quickly
09:32:03 <Zuu> I'm working on documentation adjustments :-)
09:32:08 <frosch123> generally i didn't like ottd identifying ais via their name, it should rather use the 4-byte unique id
09:32:53 <frosch123> so, if libs use category + instance, that would be fine anyway
09:33:00 <frosch123> name would be a user represenation
09:33:34 <Zuu> Only that it forces libraries to use a unique name of their main class (or use an unique category)
09:36:09 <frosch123> apparenlty everyone was fine with that up to now :p
09:40:06 *** sla_ro|master has joined #openttd
09:52:04 <Zuu> frosch123: Until I tried to create a library and spent 2-3 hours trying to figure out why OpenTTD didn't load my library that looked perfectly fine.
09:52:20 *** Ristovski has joined #openttd
09:53:30 <Zuu> Currently [AI|GS]Library do not have an item of its own in the doxygen docs. Instead, in [AI|GS]Info we tell that the documentation there also applies to libraries.
09:54:37 <frosch123> "+ "." + ScriptInfo::GetVersion()." <- hmm, i thought the version is the last parameter of the import?
09:55:40 <Zuu> hmm, yes you are correct. In import() used by script authors, version is a separate parameter (probably to enforce that it is an integer)
10:26:26 *** MatrixCL has joined #openttd
10:36:33 <gynter> For some reason one pieces part is 256 records
10:38:21 *** Devroush has joined #openttd
10:47:42 <lugo> Zuu, i'm doing a german translation for NAI and have a question: Per town growth on/off switch using signs
10:47:42 <planetmaker> gynter, hard to tell. You use a modified NML
10:47:51 <lugo> What kind of signs are meant here?
10:48:10 <gynter> planetmaker: i use the lastest upstream + the patch from yesterday
10:48:32 <gynter> but i think i found the bug already
10:49:23 <Zuu> lugo: Its documented in the readme. If you go to the settings window, it will refere you there for documentation on how to use that. I expect maximum 5 % of the users to be interested in this.
10:49:50 <Zuu> When enabled you can put a sign on the town tile with a special code word to enable/disable town growth.
10:49:58 <Zuu> Much like the sign commands in TTDPatch
10:54:06 *** gelignite has joined #openttd
10:58:59 <lugo> Zuu: and "Congested towns will grow slower or not at all.."
10:59:12 <lugo> congested in the sense of too much traffic?
10:59:31 <Zuu> There is a section about this in the readme
10:59:43 <lugo> Where should i upload the translated file? forum thread?
10:59:53 <Zuu> Upload it to the f orum thread.
10:59:58 <lugo> yeah, the readme isn't in the devzone-project archive :)
11:13:24 *** andythenorth has joined #openttd
11:15:16 <Rubidium> no, it has not yet recovered/fixed the HDD corruption from over two weeks ago
11:15:25 <andythenorth> well we'll wait a bit longer then
11:15:43 <Rubidium> with it at my work, you can better hibernate
11:16:13 <Rubidium> although, it's not really HDD corruption, just corruption of the file system by a stupid action
11:17:27 <Rubidium> or didn't you mean it as in it crowd?
11:19:12 <planetmaker> hello andythenorth
11:19:32 <planetmaker> I hacked firs a bit last night, updating the makefile
11:20:51 <andythenorth> so /scripts got smaller
11:21:13 <planetmaker> there were tons of obsolete stuff there
11:21:26 <andythenorth> it did make searching harder :)
11:23:36 <planetmaker> I added one big hack there... you find it in Makefile.in
11:24:00 <planetmaker> I left that for you to fix, if you should feel like
11:24:07 <Eddi|zuHause> should i dare updating nml? :)
11:24:22 <andythenorth> planetmaker: I see the line. What needs to change?
11:24:32 <andythenorth> change the filename?
11:24:51 <planetmaker> andythenorth, the and location, yes. Basically the mv command should not be needed and be done by the script, I think
11:26:59 <Eddi|zuHause> raise YaccError("Unable to build parser") ... uh... what?
11:27:25 <andythenorth> I didn't see that locally ;)
11:28:23 <andythenorth> I solved 30% of the hack
11:28:26 <andythenorth> I'll sort the rest out in a bit
11:30:24 <Eddi|zuHause> "blah: foo" instead of "blah : foo"
11:30:45 *** Alberth has joined #openttd
11:30:45 *** ChanServ sets mode: +o Alberth
11:35:13 <Alberth> there is a bug in the program would be my first guess :)
11:37:21 <planetmaker> so you didn't solve it despite the first impression, gynter ?
11:37:39 *** amiller has joined #openttd
11:37:47 <gynter> I have currently debugged the code, and the problematic part seems to be nml/ast/townnames.py TownNamesPart::move_pieces() line 197
11:37:59 <gynter> the issue is that some parts are larger than 255
11:38:21 <Alberth> the filenames of your paste service are horrible :)
11:38:36 <gynter> this is easy to paste thou :P
11:39:53 <gynter> Alberth: also the issue resolvs if I replace all 1 possibilities with 2
11:40:03 <Alberth> do you also have a language file?
11:40:57 <Alberth> gynter: I vaguely remember it throwing out common factors
11:42:53 <Alberth> nmlc ERROR: "lang/english.lng", line 2: Undefined command "VERSION" <-- how do I fix this?
11:45:05 <Alberth> ok, reproduced the error
11:46:23 <Eddi|zuHause> even if it was, it shouldn't be an "internal error"
11:46:26 <gynter> Rubidium: yes, "NML tries to make it fit by creating sub-blocks. If that fails, you will get an error."
11:46:37 <gynter> The sub blocks are too big :)
11:47:10 <gynter> Eddi|zuHause: it does since assert checks the size of the sub-block
11:48:15 <Alberth> assert is a paranoia check that should never fail
11:48:24 <Eddi|zuHause> gynter: i mean: if it is a proper error (and not just a mistake) it should generate a "user error"
11:48:44 <Alberth> ie the splitting code is failing in some way
11:48:58 <Eddi|zuHause> in either case, it's a bug in nml
11:49:30 <Eddi|zuHause> no user should ever see "internal errors"
11:51:31 <gynter> Alberth: line 197, sub[0] + prob, since I have lots of 1 possibility towns it creates an incremental parts therefore some parts have more members than others
11:53:04 <Alberth> iirc that should not be a problem, a part gets its own probability (I think, but I need to check that against the newgrf spec)
11:59:20 <Alberth> hmm, should not be needed? 16*255=4080 > 4014 (=number of pieces)
12:00:20 <gynter> it should not be needed indeed
12:00:27 <gynter> 17 works fine, but 16 should work too iimho
12:17:25 <Alberth> ie prevent parts from getting more than 255 members
12:18:57 *** zeknurn has joined #openttd
12:20:18 <krinn> hi guys, is there a property to see if an engine is dual head ?
12:26:35 <planetmaker> a property, yes. a variable... possibly not
12:29:42 <Eddi|zuHause> at least in TTO, dual-headed engines could be built single headed if there was no free vehicle slot left
12:30:27 <Eddi|zuHause> so just because the engine is dual headed doesn't mean the vehicle is part of a dual headed consist, in some corner cases (like loading such an old "broken" savegame)
12:30:53 <Eddi|zuHause> krinn: maybe you should explain better what you're trying to do
12:33:15 <krinn> i'm trying to detect if i should use twice that engine to pull cargo
12:33:41 <krinn> it makes sense because of cargo freight weight
12:34:09 <krinn> it makes far less sense if it took 4 "spaces" and not 2 in my vehicle length
12:35:05 <Eddi|zuHause> krinn: i don't think there's a way to get the length of a vehicle before you build it
12:35:27 <planetmaker> yeah. there's no possibility but to build it to know the length
12:35:28 <krinn> building it isn't a problem, i need at least 1 (lol or who will pull my wagons)
12:35:53 <Eddi|zuHause> krinn: that wasn't the point...
12:36:03 <krinn> do you mean i could detect dual head because of length of the engine ?
12:36:11 <Eddi|zuHause> i mean "build one of each vehicle, record its total length, and store that in a table"
12:36:29 <Eddi|zuHause> vehicles may be shortened, or articulated, or ...
12:36:29 <krinn> Eddi|zuHause, done already, keep track length, refit... properties
12:37:05 <krinn> but as length is not a fixed size, i don't see how i could say that engine length is bigger because dual head
12:37:25 <Eddi|zuHause> why would you need to know that?
12:37:43 <krinn> again: to not build twice time that engine in the same vehicle
12:37:45 <Alberth> while not enough power: add engine :)
12:38:14 <Eddi|zuHause> there could be a dual-headed engines which are shortened so they are shorter than a single normal engine
12:38:38 <Eddi|zuHause> why do you care if the engine is dual head, if only the total power and length of the consist matters?
12:38:41 <krinn> Alberth, can't do: some 1 head engine have more power than a dual head one
12:38:43 <planetmaker> uh... missing an important thing Alberth : the break condition ;-)
12:38:58 <planetmaker> I'm sure NewGRFs would otherwise hit that
12:39:45 <Eddi|zuHause> krinn: so you calculate power/length
12:39:53 <Eddi|zuHause> and sort the engines by that value
12:40:00 <Alberth> planetmaker: NewGRFs are too weird
12:40:18 <krinn> power/length what a funky filter
12:40:27 <Eddi|zuHause> krinn: and exclude engines that are longer than <limit> before
12:41:10 <Eddi|zuHause> krinn: 1 tile = 16 length units
12:42:25 <krinn> that's the key, if i need to calc anything base on unit length, with a unit length that mean nothing as an engine could be smaller or bigger that 16 base, it's sure failure
12:42:57 <Eddi|zuHause> an engine can't be shorter than 1 length unit
12:43:31 <Alberth> krinn: why are you so worried about building the same dual headed engine in the same vehicle?
12:43:33 <Eddi|zuHause> so if you have a train "8 tiles, 1 tile max should be engines", then you filter out all engines that are longer than 16lu
12:44:14 <krinn> Alberth, to not build twice time the same engine to keep vehicle length for wagons
12:44:38 <Eddi|zuHause> and then you make the above sort, and pick the "best" engine by that filter, reduce the length by that amount, and repeat the step, until your engine length is filled
12:44:39 <krinn> Eddi|zuHause, so for you i could then say "if engine length > 1 tile it's a dual head" ?
12:44:53 <Eddi|zuHause> krinn: no, we don't care if it's dual headed
12:45:00 <Eddi|zuHause> krinn: we only care about the total length
12:45:17 <Alberth> krinn: so what to do if 2 dual headed engines == 7 normal engines?
12:45:17 <Eddi|zuHause> "dual headed" doesn't tell us what length it is
12:45:46 <Alberth> ie to get the power, you either build 2 dual headed engines, or 7 notmal engines
12:45:57 <krinn> Alberth, a pitty in that case, i suppose it could be done by newGRF, hopefuly didn't saw one that crazy yet
12:46:47 <Eddi|zuHause> krinn: there are mighty crazy newgrf sets out there
12:46:50 <Alberth> ie it could be that 2 dual headed engines is shorter than the equivalent in normal engines
12:46:58 <Eddi|zuHause> and i'm not necessarily talking about NUTS :p
12:47:17 <krinn> it's easy to get : using two engines makes the overall vehicle working better, with a little space taken (1 wagon sacrifice), but with two heads engine, that would kill 2 space for wagons, so in that case i would just build 1xdual head to save space for wagons
12:47:27 * Alberth wonders about crazy nuts newgrf sets :)
12:47:33 <Eddi|zuHause> krinn: power/length is really not as crazy as you think :p
12:47:51 <krinn> Alberth, yes i know, that's the problem there, i can't really use so length to detect dual head as i said
12:48:19 <Alberth> krinn: so if you use total length of the engines, the dual head fall out by themselves if there are shorter alternatives
12:49:22 <Alberth> no need to special case them
12:50:19 <frosch> for scripts there is no difference in a dual headed engine or an articulated one
12:50:34 <Eddi|zuHause> krinn: let's have an example: we have 3 engines, one is 24lu with 4000hp, one is 16lu with 2500hp and one is 8lu with 1000hp
12:50:47 <krinn> ah finally a nice suggest frosch maybe dual head are state as articulated ! i will check
12:51:04 <Eddi|zuHause> we said in the above example that we only want 16lu of engines total, so we remove the first from the candidate list
12:51:18 <Eddi|zuHause> means we have 2500/16 and 1000/8 as candidates
12:51:45 <Eddi|zuHause> the first one has a higher power/length ratio than the second, since it fills the whole 16lu, we can buy one of those for our train
12:51:48 <gynter> Alberth: works, thanks
12:51:50 <frosch> rear heads should be invisible to ais just like articulated parts
12:52:01 <frosch> an ai should not have to know about dualheadedness or similar
12:52:12 <frosch> if you need that info, then something else is wrong
12:52:17 <Alberth> ok, thanks for reporting
12:52:21 <Eddi|zuHause> the second has the lower ratio, since it's 8lu we could buy two of those for our train, resulting in 2000hp, which is lower than the 2500 of the other
12:52:47 <Eddi|zuHause> krinn: that way, we find out that 1x2500/16 is the best choice for our train
12:53:08 <Eddi|zuHause> it gets more problematic with "odd" lengths
12:53:13 <Alberth> frosch: hopefully you can query the total length
12:53:20 <Eddi|zuHause> but at no point we care whether it's dual-headed or not
12:53:28 <frosch> Alberth: afaik you can *only* query the total length
12:53:54 <krinn> Eddi|zuHause, all is base if power have a real meaning, and it's not, take any engine and put two of them the vehicle will works better, even if power is too much
12:54:03 <Alberth> I'll take that as a "yes" :)
12:54:25 <Eddi|zuHause> krinn: that sentence did not make any sense
12:55:41 <krinn> Eddi|zuHause, to say i don't care the power, i will just add two engine if length isn't too big and single head, and 1 engine if dual head
12:56:24 <Eddi|zuHause> i don't think you understood ANYTHING i said this whole time
13:00:50 *** LordAro has joined #openttd
13:01:00 <krinn> to sum up there's no way to detect dual head so
13:01:43 <krinn> except some assumption about length of engine
13:02:21 <Eddi|zuHause> krinn: i tried explaining to you that you're solving the wrong problem
13:03:11 <andythenorth> can't the length of the engine be measured?
13:03:18 <Eddi|zuHause> krinn: like "i want to buy a truck for transporting wood, what size front window do they come in?"
13:04:28 <krinn> andythenorth, yes, after creation i could get that info, but lenght of the engine is a weak info to base assumption if engine is single or dual head because newGRF can build a single head engine with same size as a dual head one
13:04:55 <andythenorth> I'll trust you on that
13:05:17 <andythenorth> I am not understanding the issue, but then I don't know the AI API
13:05:29 <krinn> andythenorth, so if length are fixed to 16px, having a engine built with 32 you could say it's a dual head, easy, but that 16px is not a standard
13:05:32 <Eddi|zuHause> krinn: again, if two dual headed fit into the same space as two single headed, why care if it's dual headed?
13:06:11 <krinn> Eddi|zuHause, i won't care of course
13:06:20 <Eddi|zuHause> krinn: then why try to check it?
13:07:16 <krinn> Eddi|zuHause, because it's not the case in real, dual heads are nearly always taking much more length than 1 head engine
13:07:52 <Eddi|zuHause> yes, so length is the only thing you really care about
13:08:05 <Eddi|zuHause> so why worry about anything else but the length directly?
13:08:09 <andythenorth> hp per unit length
13:08:19 <Eddi|zuHause> andythenorth: that's what i said
13:08:20 <krinn> in real, no, i don't care about length, but only in that special case when i attach two same engine
13:08:35 <andythenorth> Eddi|zuHause: how is your nml fork going?
13:08:57 <Alberth> it's at least 2 revisions behind :)
13:08:58 <Eddi|zuHause> andythenorth: which fork?
13:09:19 <krinn> Eddi|zuHause, that would makes the engine picking too complicate (and it's really really enough like it is now)
13:09:19 <andythenorth> there are multiple?
13:09:35 <Eddi|zuHause> andythenorth: no, but maybe you mean something different :)
13:10:11 * andythenorth misread a devzone attribution
13:10:20 <krinn> Eddi|zuHause, don't care about length and pickup best engine for your need : done. Now when you add two of the pickup engine : this time length matter, but i don't fancy redo everything at the pickup part
13:10:21 * andythenorth is somewhat lacking sleep
13:10:21 <Eddi|zuHause> krinn: well, just try to run your AI with CETS, it has all sorts of weird vehicle lengths and articulations :)
13:10:45 <andythenorth> krinn: it sounds like a tmwftlb problem
13:10:49 <andythenorth> just do less? :)
13:11:26 <Alberth> andythenorth: we tried talking him into doing that, but he doesn't want to
13:11:44 <Eddi|zuHause> krinn: either you do duplicated engines properly with an algorithm like i said, or you don't bother at all
13:12:26 <Eddi|zuHause> andythenorth: the only "fork" of nml i did was adding actionC support
13:12:27 <krinn> Eddi|zuHause, that's why the pickup engine, if not dual head build two, else build one, is the solve i seek, but if i cannot see if it's dual or not...
13:12:43 <andythenorth> so who has forked nml, and why?
13:13:01 <Eddi|zuHause> andythenorth: it's found in the devzone as eddi-nml
13:13:46 <Eddi|zuHause> it's a requirement for my partial compilation hack in CETS
13:14:09 <andythenorth> I thought that was yours
13:14:17 * andythenorth is not very awake
13:14:30 <andythenorth> when did the ship speed limit get lifted?
13:26:16 <Eddi|zuHause> hm, you lost your numbers, now you have a weird colour :p
13:26:23 <peter1139> i have a patch for that...
13:37:07 *** EyeMWing has joined #openttd
13:41:14 <krinn> can we get that dual head state in the API ?
13:42:21 <krinn> frosch, the IsArticulated return false when engine is dual head, but it was a nice suggest
13:42:46 <frosch> that function is mainly meant for roadvehcles
13:42:53 <frosch> to decide wehther regular roadstops work
13:43:17 <krinn> it is handle by train too, but i have no idea what an articulated train could be :)
13:45:03 <peter1139> i think maybe it fell out of favour
13:45:11 <krinn> so some exist single head and articulated ?
13:45:13 <Eddi|zuHause> most of my engines are 3-part articulated
13:50:03 <krinn> yes, but IsArticulated properties for train would makes more sense to return dual head state than the engine is one head made of 3 parts no ?
13:50:39 <krinn> who care how many parts the engine is for a train station that don't care (not like road vehicle that have a station limit to accept them)
13:54:20 <Alberth> I think it should return dual headedness for all engine with an even number of parts, except at Wednesday and perhaps Sunday
13:55:05 <Xaroth|Work> and friday the 13th
13:55:17 <Eddi|zuHause> krinn: articulated vs. dual headed have different effects wrt train composition
13:55:31 <Alberth> Xaroth|Work: oh, good point, I forgot about that one
13:55:45 <Xaroth|Work> how's opentrct going, Alberth?
13:56:06 <krinn> Eddi|zuHause, how (in openttd i see where), in noai i don't see how the ai could compose its vehicle, as openttd attach the second head itself
13:56:07 <Xaroth|Work> s/open/free/ even
13:56:08 <Eddi|zuHause> besides, in the future there might be a way to make engines both articulated AND dual headed
13:56:38 <Alberth> but the message is not to change meaning of properties if they happen to not fit in your idea of use
13:57:10 <krinn> Eddi|zuHause, even with that, i don't see a problem if noai doesn't answer articulated for train but only return dual head state : even dual head + articulated, noai won't care about articulated state
13:57:34 <Alberth> Xaroth|Work: forward :) working on being able to build a roller coaster track in the world
13:57:44 <Eddi|zuHause> krinn: if you don't care, don't change it.
13:57:50 <krinn> well, for noai IsArticulated return value for trains has no usage
13:58:23 <krinn> only usable for road/tram where station to use depend on that
13:58:30 <Alberth> I prepared the world for it, added track pieces, and finished the state machine handling the building. Now it just needs high level UI control, and actual adding of pieces in the world :)
13:58:55 <Alberth> once that works, it may be nice to have a car following the track :)
13:59:03 <Xaroth|Work> that.. would be useful yes :P
13:59:12 <Eddi|zuHause> krinn: maybe no usage for YOUR ai, but maybe some other AI does depend on this fine difference?
13:59:40 <krinn> Eddi|zuHause, to do what with that ?
14:00:23 <Alberth> but it feels like it is starting to come together now, I spent time refactoring existing code rather than adding new stuff all the time
14:05:17 *** HerzogDeXtEr has joined #openttd
14:06:17 *** MrShell has joined #openttd
14:16:02 <Eddi|zuHause> krinn: maybe this will finally settle this silly discussion? www.informatik.uni-halle.de/~krause/ai_multihead.diff
14:16:24 <Eddi|zuHause> needs exporting to ai, need to figure that out yet
14:17:40 <Eddi|zuHause> @devs: is there a reason why IsWagon() returns power==0 instead of railveh_type == RAILVEH_WAGON?
14:18:18 <frosch> likely copied it from script_engine
14:19:18 <TrueBrain> pfft, cheap shot; I will get you for that next weekend :P
14:19:39 <krinn> Eddi|zuHause, thank you
14:21:18 <Eddi|zuHause> krinn: www.informatik.uni-halle.de/~krause/ai_multihead2.diff <-- now with squirrel export stuff
14:24:56 <frosch> btw. is there meanwhile a reason, why ais need that function?
14:25:35 <peter1139> well, power == 0 was the definitive way to define a wagon..
14:26:05 <krinn> frosch, if you really want one : aesthetic one : having a vehicle with 3 wagons and 4 heads doesn't look nice
14:26:14 <Eddi|zuHause> peter1139: that may be so, but this looks like duplicating some logic
14:27:19 <krinn> frosch, this will also tell the AI because of multihead, last wagon is an engine, and this may cause problem if you move a wagon to end of queue and look if it is there and don't see it...
14:27:26 <frosch> krinn: ok, earlier it sounded like you tried to use it instead of getmaxtractiveeffort
14:27:45 <krinn> frosch, for this too, but there's more than just that reason
14:28:19 <krinn> right now, it's not really a problem to have 4 heads in the vehicle, i lost a bit of length for wagons, but it also look a bit odd
14:28:59 <krinn> just build a vehicle with two tgv in it, you'll see what i mean (except for a big big vehicle) it looks strange
14:29:11 <Eddi|zuHause> krinn: well it looks better if you put half the wagons between each set of dual heads
14:29:58 <krinn> yes, but i'll could get mad doing such a function
14:30:39 <Eddi|zuHause> frosch: in general, i see more reasons to include the function than to omit it
14:32:11 <Eddi|zuHause> just not for krinn's original problem :=)
14:32:30 <krinn> :) but it will solve mine too
14:47:19 *** sla_ro|master has joined #openttd
15:36:43 <DorpsGek> frosch: OpenTTD uses TCP and UDP port 3979 for server <-> client communication, UDP port 3978 for masterserver (advertise) communication (outbound), and TCP port 3978 for content service, a.k.a. BaNaNaS (outbound)
16:14:02 *** amiller has joined #openttd
16:24:11 <Zuu> Also, soon there will be a CluelessPlus with actual NoCarGoal support. (I though it was already done, but it wasn't)
16:25:58 <Zuu> (oh and sory for Swedish strings. It has grain, iron ore and steel as goal cargos in the screenshot)
16:28:48 <Rubidium> maybe refer to {WHITE}Gold instead of GOLD in the goals of bronze and silver
16:29:38 <Rubidium> and... feature request: multi lining strings, so those goals can also be read on 640x480 ;)
16:29:48 <Eddi|zuHause> maybe {GOLD}Gold? :p
16:33:11 <frosch> somewhen ottd will get a html renderer
16:33:11 <Zuu> I was about to remove the global goals and replace with a short line of text "Complete all three company goals to score GOLD". However, I decided to keep them even though I don't like the bloat of them because a spectator on a server without any companies will not see the company specific part.
16:34:49 <Zuu> That said, the information about what goal cargos there are is included in the front story page (which is global), so those spectators will still be able to see the information even if I remove it from the goal window.
16:35:36 <andythenorth> frosch: markdown :P
16:35:51 <andythenorth> if you do html, please ensure you support <blink>
16:37:56 <Eddi|zuHause> anyone remember wingdings? :p
16:38:20 <Rubidium> Eddi|zuHause: wingdings has no place in websites... use webdings instead!
16:42:52 <Xaroth|Work> fontawesome is epic
16:45:51 <V453000> all fonts are epic with beer
16:47:14 <frosch> Xaroth|Work: i remember i codes such things using javascript 15 years ago, since when is there a html tag for it?
16:48:03 <Rubidium> frosch: since like html 1
16:48:28 *** oskari89 has joined #openttd
16:48:41 <Xaroth|Work> frosch: fontawesome?
16:49:12 <Rubidium> oh, it was non-standard
16:49:47 <Xaroth|Work> Microsoft did marquee
16:49:50 <Xaroth|Work> netscape did blink
16:49:53 <Xaroth|Work> both not official
16:55:31 <Eddi|zuHause> i'm fairly sure marquee existed 15 years ago
16:56:23 <Xaroth|Work> it was just not an official tag
17:24:44 *** y2000rtc has joined #openttd
17:27:41 <peter1139> blinking marquees where the best
17:45:17 <DorpsGek> Commit by translators :: r25619 trunk/src/lang/turkish.txt (2013-07-20 17:45:09 UTC)
17:45:18 <DorpsGek> -Update from WebTranslator v3.0:
17:45:19 <DorpsGek> turkish - 7 changes by wakeup
18:07:49 *** Djohaal has joined #openttd
18:14:21 *** kais58|A1K is now known as kais58__6
18:17:57 *** SpComb^_ is now known as SpComb
18:25:30 <Rubidium> so... is there a game script that needs multiplayer testing this evening?
18:25:46 <andythenorth> interesting idea
18:27:47 *** frosch is now known as frosch123
18:44:12 <Zuu> While I do have a upcomming NoCarGoal release with some news that are useful in multiplayer, it is not completely ready yet.
18:44:44 <Zuu> And the keyword here is 'usefu', none of the news changes the actual gameplay other than that it becomes easier to keep track of your enemies :-)
18:50:53 *** Devroush has joined #openttd
18:56:15 *** montalvo has joined #openttd
19:06:37 <krinn> hi Terkhen Zuu & one i miss
19:08:33 <andythenorth> I don't want to provide too much info
19:08:38 <andythenorth> play the game to see that :)
19:08:59 * krinn will not suggest crew members names so
19:09:08 <andythenorth> interesting idea :)
19:09:34 <andythenorth> Ship 58 sank: crew members Bob and Sam are dead
19:09:34 <krinn> lol specially if edward smith is one of them
19:10:24 <krinn> but speed and cargo type would be basic to get no ?
19:11:24 <andythenorth> cargo type - slightly trickier, depends on available cargos in game
19:11:30 <andythenorth> can show freight / pax / mail capacities
19:12:07 <andythenorth> I can render any property into the docs pretty much
19:13:29 <krinn> i think saying speeed and if (passenger or cargo) type
19:17:15 <Alberth> I'd add the intended purpose of the ships
19:17:35 <Alberth> speed is less relevant, I'd say
19:17:49 <andythenorth> speed in Squid is a bit complicated
19:17:58 <andythenorth> depends on canal, ocean and load amount
19:18:39 <Alberth> if you want speed, say <min speed> or faster :)
19:19:04 <Alberth> or, in commercialism, upto <max speed> :)
19:19:29 <krinn> speed would tell if you will use a thing so slow that you will die waiting it to goes and come back, for ships i think it's important
19:19:40 <andythenorth> game will tell you though, no?
19:19:59 <Alberth> krinn: you need to see what other ships are available too, which is hard in this list
19:20:07 <andythenorth> mm, also the speed parameter affects speed :)
19:20:40 <krinn> well, in game checking a vehicle is a pain when you have that much amount to choose
19:20:52 <Zuu> andythenorth: So you need to render a javascript page where you can set the parameters at the top :-p
19:21:53 <andythenorth> somewhat overkill :)
19:21:56 <Alberth> upload the openttd.cfg file, and I'll tell you the speed of the ship :p
19:23:36 <andythenorth> if I put in capacities, I have to provide the units
19:23:39 <andythenorth> that's a bit meh
19:23:54 <krinn> would be cool to see ship picture
19:24:05 <andythenorth> might add that later
19:24:13 <andythenorth> many are not drawn yet :)(
19:24:19 <andythenorth> doing docs helps me avoid drawing
19:24:27 <krinn> that why the "?" picture was made for
19:26:37 <krinn> so it's a webpage to say : look at the ships you cannot get :)
19:27:30 <krinn> ^^ you should do teaser for dumb movie
19:27:57 <andythenorth> V453000 draw me some ships?
19:28:19 <V453000> or doing my things :P
19:28:23 <V453000> also I cant really draw nice ships
19:28:29 <V453000> dont know enough details
19:28:43 <V453000> slash has not enough experience to imagine details
19:29:31 <krinn> when everyone speak about ships, i never see a submarine, can't you do one ? even unrealistic it would be cool to have one usable to carry passenger or something :)
19:30:18 <krinn> we only have the event one, and you must die looking at water to see it
19:30:37 <V453000> my next ship is going to be turtle convoy
19:30:56 <krinn> well, as long as no dwarfs ride it, it's ok
19:31:43 <andythenorth> andrew350 did a sub
19:31:53 <andythenorth> you probably won't see it
19:34:42 <krinn> lol thx albert nice catch
19:35:12 <Alberth> they are utterly useless as for transport
19:35:32 <krinn> just like zepplin, but it's funny to get one
19:35:47 <Alberth> they load missiles, drive around for a long time, and then return them to the same harbour
19:39:28 <Alberth> and last but not least, they are insanely expensive
19:40:17 <krinn> generally it's not a problem, it's not like you cannot get a huge load of money to waste
19:41:57 <Alberth> stock holders won't agree :)
19:44:25 <krinn> lol i wish a kick ass yatch with 1 passenger capacity that cost an awful lot of money : stock holder love shown their hands are full
19:46:28 <krinn> while looking at sub in the wiki i see the ufo pic of rails getting blown by the aircraft
19:54:50 *** Frank-BTPro has joined #openttd
19:56:22 <Frank-BTPro> o/ Hello Everyone :)
19:57:18 <Alberth> ai are just another player, so yeah, they should be
19:57:42 <Frank-BTPro> We are currently in the process of setting up a "inter community" tournament between the BTPro OpenTTD Community, N-Ice OpenTTD Community and Novapolis OpenTTD Community
19:58:07 <Frank-BTPro> we were wondering if there would be people interested here to join in or maybe add OpenTTDCoop Community to join in with a team!
19:58:36 <Frank-BTPro> the Tournament will include CityBuilder games + Company Value Goal games!
19:58:42 <Alberth> Frank-BTPro: you better post at the forum then, you'll reach a much wider audience
19:58:55 <Frank-BTPro> maybe that's a good idea Alberth :)
19:59:07 <Frank-BTPro> it's just a heads up for now, we are in the process of organizing it :)
19:59:12 * andythenorth considers refactoring a bit more Squid
19:59:31 <andythenorth> eventually I run out of code to fix, and have to start drawing stuff :(
19:59:56 <Frank-BTPro> sorry, thought it was an answer to what I was saying ;-)
19:59:56 <Alberth> Frank-BTPro: improve code without changing functionality
20:00:09 <Frank-BTPro> right, I get it now :)
20:00:28 <Alberth> squid still has too many tentacles? :)
20:01:07 <andythenorth> Alberth: (no context sorry), I have a case where I'm creating a new instance of a class, and it seems correct to have to declare 'foo = somevalue' or 'foo = None' in the __init__
20:01:23 <andythenorth> is requiring an explicit 'foo = None' weird?
20:01:43 <andythenorth> the __init__ could obviously just handle the case of the param being absent
20:02:08 <andythenorth> sorry - should have said
20:02:25 <andythenorth> I want to know that it's intended None and not just forgotten
20:03:05 <andythenorth> smells odd though
20:03:09 <Alberth> python needs a value for every parameter at each call, so if the user doesn't supply one, you need to, in the formal parameter list
20:03:25 <krinn> i'm not sure i get it, but having a default value on init of a class instance isn't a bad idea
20:03:38 <Rubidium> andythenorth: just create a fake value ("invalid" or so) and pass that as default?
20:03:55 <Alberth> automatically picking of None for omitted values would disable checking that you pass all parameters
20:04:12 <andythenorth> that's ok in many cases for this context
20:04:16 <andythenorth> but maybe not this one
20:04:21 * andythenorth is probably overthinking it
20:04:21 <Alberth> ie, None could be a legal value, that you want to distinguish from omitting a value
20:04:51 <krinn> None is the null classic ?
20:05:21 <Alberth> krinn mostly, but it's quite universal
20:05:25 <andythenorth> I want to remove the get() and have it fail if not a valid value or None
20:05:48 <andythenorth> just seems right
20:05:52 <andythenorth> definitely overthinking it
20:06:38 <Alberth> kwargs['graphic_variations_by_date'] does that
20:06:48 <andythenorth> it's more of a style question than 'how?'
20:07:18 <Alberth> I'd remove the "**" :)
20:07:40 <Alberth> let the user figure out how to make the dict
20:07:52 <Alberth> iirc dict(a=b) works too
20:08:34 <Alberth> but I thought you split all those values into some mixins?
20:09:15 <andythenorth> I'm pretty happy with the structure, just got obsessed about this style thing
20:09:44 <Alberth> good sign for needing to go to bed? :)
20:10:59 <Alberth> I wonder whether you cannot make that dict available in one simple method
20:11:15 <Alberth> it seems such a waste to expand it all
20:11:49 <Alberth> self.buy_menu_bb_xy = kwargs.get('buy_menu_bb_xy') <-- line 38 is missing a default
20:11:58 <andythenorth> I could have just passed ship_props in as a dict :P
20:12:46 <andythenorth> fixed l38 thanks
20:13:10 <Eddi|zuHause> i don't use kwargs in the function code, but i put them as explicit parameters if the function uses it (you can still pass them as kwargs, but this way it complains if it's missing)
20:13:44 <Eddi|zuHause> as in def function(params, that, i, actually, use, **don_t_care)
20:14:15 <Eddi|zuHause> then i can call function(params, **kwargs)
20:14:39 <Eddi|zuHause> and it'll extract "that" "i" "actually" and "use" from the kwargs
20:15:24 <Alberth> andy cares about them, there just toooooo many to list explicitly
20:15:53 *** Frank-BTPro has left #openttd
20:16:33 <Alberth> you can test the set keys being a superset of the expected arguments
20:19:36 <Eddi|zuHause> and i only extract the columns that i actually use
20:20:21 <Eddi|zuHause> which sometimes are few, and other times a lot
20:24:22 <frosch123> oh my... would i have thought 5 years ago, that i would run 3 virtual machines in parallel on my personal computer
20:24:52 <andythenorth> then you're special :)
20:41:50 *** mindlesstux has joined #openttd
21:43:59 <Xaroth|Work> kwargs has it's use
22:11:43 *** ChanServ sets mode: +v tokai
22:19:40 *** Extrems has joined #openttd
22:19:44 <gynter> Yay, I has my server online now with my town names, good :)
22:26:38 *** LordAro has joined #openttd
22:28:52 <Zuu> SCPLib_NoCarGoal or SCPClient_NoCarGoal?
22:29:29 <Zuu> (for an AI library that contains a SCP client to talk to NoCarGoal GS)
22:30:57 <krinn> I think the nocargoal should goes first
22:31:23 <krinn> NoCarGoal_SCPClient or something like that
22:32:55 <krinn> i really didn't think about making a lib to encapsulate SCP in :)
22:33:06 <krinn> but it's a great idea, it will help a lot AI users handling it
23:10:13 *** TrueBrain has joined #openttd
23:40:15 *** Extrems has joined #openttd
continue to next day ⏵
