IRC logs for #openttd on OFTC at 2020-05-29
⏴ go to previous day
00:04:41 <DorpsGek_III> [OpenTTD/OpenTTD] Yexo commented on issue #8092: Swedish translation issue in the performance rating window https://git.io/JfTUv
00:04:41 <DorpsGek_III> [OpenTTD/OpenTTD] Yexo closed issue #8092: Swedish translation issue in the performance rating window https://git.io/JfTUv
00:14:47 <Speeder_> is "produce" block mandatory?
00:17:37 <andythenorth> no, depends on the industry
00:18:05 <andythenorth> there are multiple ways to implement cargo production
00:20:30 <Speeder_> andythenorth, FIRS always use it?
00:20:57 <Speeder_> I am trying to figure out best way to add NML5 style stuff without changing the NML4.5 ones.
00:21:20 <andythenorth> FIRS probably doesn't use it for hotel
00:21:28 <andythenorth> or black hole industries maybe
00:22:36 <andythenorth> if you are using the actual FIRS codebase, and not just editing the nml, this is a hard way to go
00:22:46 <andythenorth> you'll have to engage with all the FIRS internal crap
00:22:57 <andythenorth> you'll need a new class of industry
00:23:10 <andythenorth> it's possible to do that btw
00:40:58 <Speeder_> andythenorth, FIRS4 has support for each cargo producing something specific?
00:41:02 <Speeder_> like the nml specs support?
00:41:27 <andythenorth> no, I think it's all 'produce' in FIRS 4
00:44:59 * andythenorth wonders about FIRS 3
00:45:49 <andythenorth> looks like FIRS 3 uses action 0 for tertiaries like hotel
00:46:39 <andythenorth> yeah it sets prod multiplier, I think it's action 0 for a few industries, produce for most
00:49:47 <Speeder_> why FIRS uses produce?
00:49:56 <Speeder_> just so I get what it does differently from action 0
00:51:15 <andythenorth> for primaries, it's the only sane way to adjust production to account for supplies
00:51:33 <andythenorth> for secondary, it's the only sane way to do combinatorial production
00:51:58 <andythenorth> for tertiary, I switched to produce so I'm only using one method not two
00:58:36 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz left a comment on commit: Cleanup: Correct typographic errors in code comments. https://git.io/JfoM5
01:18:32 <Speeder_> combinatorial production? as in... several inputs leading to only one output?
01:21:07 <andythenorth> as in different ratios depending which cargos are recently delivered
01:23:53 <Speeder_> andythenorth, that is the default now with nml5 no?
01:25:25 <andythenorth> there is a new syntax for action 0 which may or may not do that
01:25:33 <andythenorth> I don't use it so I don't know :)
01:25:51 <andythenorth> TL;DR most newgrf problems tend towards using the callback if one exists
01:26:01 <andythenorth> produce is a callback, action 0 props are proprs
01:26:10 <andythenorth> props are generally less flexible
01:26:26 <andythenorth> it's not universal advice, callback is not always needed
01:26:52 <Speeder_> andythenorth, the syntax example for nml5 has this: accept_cargo("IORE", produce_cargo("STEL", 1), produce_cargo("SLAG", 0.25))
02:43:45 *** Wrench_In_The_Plan_ has joined #openttd
02:46:18 *** Wrench_In_The_Plan__ has joined #openttd
02:47:49 *** Wrench_In_The_Plan has quit IRC
02:48:00 *** Wrench_In_The_Plan__ is now known as Wrench_In_The_Plan
02:52:48 *** Wrench_In_The_Plan_ has quit IRC
03:37:15 <Gadg8eer> Does anyone know why FIRS is incompatible with ECS?
03:38:08 <glx> you usually can't use more than one industry grf
03:39:01 <Speeder_> alright, attempting to use FIRS4 was just a time wasting disaster
03:39:22 <Gadg8eer> I don't have that problem with any of the GRFs I just tried. Is it a holdover from the old days?
03:39:46 <Gadg8eer> Because they added 16 cargos per industry recently.
03:39:49 <Speeder_> glx, you knwo if there is any industry newgrf that uses the new syntax?
03:40:46 <Gadg8eer> And iirc, the number of industries in GRFs was raised too?
03:48:14 <Gadg8eer> Basically, I'm asking if I could test a version of FIRS that doesn't disable itself, see if the issue is still there.
04:00:37 <glx> I think it's because FIRS defines a full economy with complex chains
04:01:34 <glx> FIRS4 compatibility check is more strict than FIRS3
04:04:22 <Eddi|zuHause> conflicting definitions of the same cargo label, and other stuff
04:16:41 <Gadg8eer> FIRS is coded in m4nfo, I assume?
04:22:19 <Eddi|zuHause> no, firs is coded in a custom templating framework that outputs something that resembles nml
04:25:34 <Gadg8eer> How easy would it be for me to modify FIRS to function in conjunction with ECS
04:31:08 <Gadg8eer> Is there another industry grf that you would recommend I try modifying?
04:40:38 <Speeder_> is there are a way to add cargo in a way that vehicle newgrfs will accept it automatically?
04:42:26 <Gadg8eer> Vehicle and industry newgrfs are supposed to use cargo classes so that new cargos are automatically compatible.
04:46:49 <Eddi|zuHause> Gadg8eer: modifying for what purpose?
04:48:01 <Eddi|zuHause> Speeder_: the vehicle set must be prepared properly to derive unknown cargos from their cargo classes. assume that most vehicle grfs won't do this properly
04:48:24 <Eddi|zuHause> Speeder_: in any case, they won't show correct graphics
04:49:52 <Gadg8eer> Eddi|zuHause: To remove industries and cargos that aren't compatible with ECS, and leave the rest intact.
04:49:59 <Speeder_> Eddi|zuHause, any suggestion on how I add leather as cargo?
04:53:12 <Eddi|zuHause> Gadg8eer: that doesn't make any sense
04:55:57 <Gadg8eer> I want to use ECS and FIRS (or a similar NewGRF) together, even if that means disabling the cargos that conflict.
04:56:16 <Gadg8eer> Eddi|zuHause: Does that make sense?
04:57:34 <Eddi|zuHause> Gadg8eer: about as much as "i want to strip out parts from this airplane so i can fly it while also driving a bus"
04:58:49 <Gadg8eer> That doesn't make sense to me. If ECS and FIRS both use cargo X, then why can't I make a version of FIRS that defaults to the ECS cargos if there's a conflict?
04:59:50 <Eddi|zuHause> Gadg8eer: both planes and busses have wheels. doesn't mean you can put bus wheels on a plane
05:00:44 <Gadg8eer> WTF? How different is coding industries from coding houses?
05:00:49 <Eddi|zuHause> Gadg8eer: my ethics forbid me to help you make frankensteins monster
05:02:03 <Gadg8eer> I would modify ECS but the license bans any changes from being distributed. If I want to use ECS, I have to Frankenstein anything I add.
05:03:27 <Gadg8eer> Eddi|zuHause: Please don't tell me that it's "wrong" to modify GPL NewGRFs I'm not completely satisfied with.
05:03:39 <Eddi|zuHause> maybe you should try make a completely new grf from scratch
05:04:35 <Gadg8eer> No. I can't sprite that we'll, and tbh I don't want to make a completely new GRF, I want to use the amazing graphics others have done better than I ever could.
05:05:10 <Eddi|zuHause> you can easily reuse sprites from FIRS without inheriting the giant mess of code
05:06:16 <Eddi|zuHause> but ECS and FIRS have so vastly different design philosophies that you'd need to strip out almost everything from FIRS anyway
05:06:34 <Gadg8eer> Eddi|zuHause: Ah. Fair enough. Can I do so in NML or do I have to code it in NFO or m4nfo?
05:06:57 <Eddi|zuHause> you can do that in NML
05:22:10 <Gadg8eer> Eddi|zuHause: Got a question you probably want me to ask. What are the differences in design between ECS economy and FIRS economy?
05:22:44 <Speeder_> just realized several of the stuff I want to "add" to OpenTTD already exists, FIRS just removed it
05:23:34 <Speeder_> Eddi|zuHause, suggestions on how to reuse FIRS sprites?
05:23:51 <Gadg8eer> Speeder_: Hmm. What are you trying to make?
05:24:22 <Gadg8eer> Oh! Good luck then. :)
05:24:24 <Speeder_> I was making a map in Brazil but got frustated none of the economies available in any newgrf or vanilla quite match it
05:24:29 <Speeder_> yet... all the parts needed exist
05:24:32 <Speeder_> they are just spread out
05:24:36 <Eddi|zuHause> Speeder_: FIRS develpment is very volatile, sometimes things just get removed on a whim
05:24:46 <Speeder_> Eddi|zuHause, I noticed D:
05:24:59 <Eddi|zuHause> that's what happens if you have a lunatic as lead developer :p
05:24:59 <Speeder_> I attempted to move from FIRS3 to FIRS4 today
05:25:02 <Speeder_> was a massive failure
05:25:08 <Speeder_> I mean, it did compiled
05:25:24 <Speeder_> but then several stuff was outright missing, lots of files with giant red crosses on them, etc...
05:25:43 <Speeder_> I want to use the nml5 features too, and FIRS codebase is not suited for that either
05:25:50 <Speeder_> but I have no idea how to make sprites work right
05:26:46 <Eddi|zuHause> Speeder_: there are two things to a sprite, the file containing the graphics, and a template containing the coordinates. it gets a bit more complicated if tiles are combined from several sprites
05:27:08 <Speeder_> Eddi|zuHause, I think that is the case with FIRS yes
05:27:22 <Speeder_> one of the things I want to use is aluminum plant
05:27:28 <Speeder_> and seemly it is heavily composite
05:27:49 <Eddi|zuHause> a typical industry tile will consist of a ground sprite and one or more building sprites
05:28:49 <Eddi|zuHause> each industry tile is a separate entity
05:29:35 <Speeder_> Eddi|zuHause, you know if there is any easy way to generate a nml file?
05:29:49 <Eddi|zuHause> generate from what?
05:29:52 <Speeder_> FIRS went with python+chameleon, that is a language I didn't like very much.
05:30:23 <Speeder_> I mean, it looks like editing nml file itself is even more insane than trying to edit FIRS
05:31:31 <Eddi|zuHause> i'd suggest first trying to build one from scratch using the tutorial
05:33:08 <Eddi|zuHause> any NML tutoral, preferably an industry one, dunno if that exists
05:33:58 <Eddi|zuHause> the point of the exercise is getting a clean, limited to one single industry, GRF.
05:34:40 <Eddi|zuHause> once you got that, "generating" mostly means copying that first industry, and filling out some of the properties differntly
05:37:29 <Speeder_> Eddi|zuHause, I only wanted to know if someone made tools for that already
05:37:43 <Speeder_> if not I will see if I can create the nml files using Lua
05:38:01 <Speeder_> that is a language I am way more confortable with than python (and allows some wackyer stuff)
05:40:30 <Gadg8eer> Speeder_: Where can I download FIRS4?
05:40:55 <Speeder_> Gadg8eer, I got it from github, but andy posted an alpha on the forums I believe
05:41:05 <Speeder_> Gadg8eer, I must warn you it is MESSY
05:41:15 <Speeder_> andy started and abandoned economies multiple times
05:41:27 <Speeder_> and seemly is redoing steeltown over and over again, adding and removing cargo again
05:42:34 <Eddi|zuHause> (that's the earliest version of that script... it "evolved" a bit from there)
05:44:15 <Speeder_> I am checking the source of opengfx+ industries
05:44:56 <Eddi|zuHause> pnml is files that must be assembled by a preprocessor first
05:45:30 <Eddi|zuHause> that'll resolve all the #include and #define directives
05:45:56 <Eddi|zuHause> it's very common to do that so you don't have one giant nml file
05:46:17 <Speeder_> oooh... hacky. but sounds interesting
05:46:26 <Speeder_> why invent your own preprocessor when you can use one that exists
05:47:22 <Eddi|zuHause> most people who write nml use the standardized openttdcoop makefile, which will automatically call gcc on pnml files
05:47:58 <Speeder_> I wonder what it takes to make it use graphviz too, I like that particular feature of FIRS
05:48:03 <Speeder_> great for debugging the production chains
05:48:31 <Eddi|zuHause> i don't know how updated the coop makefile is
05:49:30 <Eddi|zuHause> using graphviz is fairly easy. making it look good is fairly hard
05:49:55 <Speeder_> so where I should get the makefile for nml5?
05:50:29 <Eddi|zuHause> there shouldn't be any relevant changes to the makefile
05:50:59 <Eddi|zuHause> but i'm not the right person to ask about setting up the makefile, i've never done it myself
05:52:10 <Eddi|zuHause> but the basic gist would be that you clone it from the repo, and then make some custom setup with your own grf name and dependencies
06:08:49 *** Wrench_In_The_Plan has quit IRC
06:25:28 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8175: Fix #7970: Recursive faults in Windows post-crash due to event loop input https://git.io/Jfodu
07:32:48 *** andythenorth has joined #openttd
07:38:56 <andythenorth> oof sleeping fail
07:49:07 <andythenorth> everyone in my house woke up at 5am
08:12:20 *** sla_ro|master has joined #openttd
09:36:07 *** ChanServ sets mode: +v tokai
10:03:17 *** Flygon_ has joined #openttd
10:16:06 * andythenorth redesigns Unsinkable Sam
10:16:14 <andythenorth> good job I never finished it eh :P
11:21:08 <DorpsGek_III> [OpenTTD/OpenTTD] michicc commented on pull request #8175: Fix #7970: Recursive faults in Windows post-crash due to event loop input https://git.io/JfKeF
11:28:39 *** gelignite has joined #openttd
11:31:17 <Samu> hi yexo, how's the pathfinder?
11:42:08 *** jottyfan has joined #openttd
11:54:50 *** jottyfan has joined #openttd
12:28:26 <andythenorth> been playing Catan with my kids
12:28:36 <andythenorth> FIRS is basically Catan?
12:31:01 <Yexo> Samu: Haven't had time to work on it, too busy moving into my new house
12:40:33 <FLHerne> Gadg8eer: Making it not disable itself next to ECS is trivial, just delete most of incompatible_grfs.py
12:40:49 <FLHerne> (making it do something sane is probably harder)
12:54:36 <andythenorth> seems I should have stayed up for FIRS + ECS amusement?
13:13:44 <FLHerne> andythenorth: Gadg8eer wants to make it work, and Eddi|zuHause was trying to say it's a bad idea and not to do it
13:13:58 <FLHerne> I mean, it's obviously a bad idea, but why not do it anyway? :D
13:15:31 <Eddi|zuHause> yeah, i was debating in my head whether i should interpret his question in the trivial or non-trivial sense of "make it work"
13:15:45 <Eddi|zuHause> and went with the pure non-trivial way
13:16:30 <FLHerne> Eh, even in the less-trivial sense of "make it function but don't worry about balance" I don't think it would be /that/ bad?
13:16:31 <Eddi|zuHause> mostly for the reason that he wants to actually distribute his changes
13:16:45 <FLHerne> Mostly just delete stuff until it stops conflicting :p
13:17:23 <Eddi|zuHause> you probably wouldn't really see the conflicts
13:18:37 <Eddi|zuHause> also, you'd have to reduce it to one single economy or you'd go insane
13:19:47 <Gadg8eer> I've decided against modifying FIRS. If I ever do an industry set, I'll code it from scratch and use FIRS graphics.
13:20:08 <andythenorth> we need a cargo pool :P
13:20:17 <andythenorth> industries only get to use the cargos they've defined :P
13:20:26 * andythenorth is here all week
13:20:27 <Eddi|zuHause> Gadg8eer: yes, i'm fairly sure that's a much saner approach :)
13:20:51 <andythenorth> I am disappointed, FIRS forks might have got into double digits :P
13:22:29 <FLHerne> andythenorth: Why shouldnt ECS coal be the same as FIRS coal?
13:22:55 <Gadg8eer> The problem is that ECS can't be modified. Is it possible to use another NewGRF to override the ECS power plant? Maybe that way I can finally add the electricity lines to the Utility "Truck" Set.
13:23:09 <andythenorth> FLHerne I used to combine PBI with other industry sets, can't remember which
13:23:16 <andythenorth> sometimes you end up with e.g. Plastic twice
13:23:26 <Gadg8eer> I mean just the cargos, not the power plant graphics.
13:25:55 <Eddi|zuHause> i don't think you can disable other set's individual industries
13:26:16 <Gadg8eer> I don't want to disable it.
13:26:36 <Gadg8eer> I want to add ELEC to the production of an industry.
13:26:42 <Eddi|zuHause> well, that's what "override" is, you disable the other one and make a new one
13:27:53 <Gadg8eer> Is there a way to add ELEC to the power plant without overriding it?
13:28:14 <Eddi|zuHause> i don't think so
13:30:45 <andythenorth> just make your own industry grf?
13:30:48 <andythenorth> probably easiest
13:30:51 <andythenorth> that's what I would do
13:33:07 <Borg> Gadg8eer: you want to move electric by trucks? ;)
13:33:22 <andythenorth> invisible trucks?
13:33:39 <Borg> invisible ultra fast trucks ;)
13:34:24 <Gadg8eer> I want the graphics used in ECS, but those are locked behind a "no modifications" license. I can't sprite anything close to what ECS and FIRS did, so it's either tiptoe around an unchangeable GRF or abandon the best oil wells and coal mines in the community.
13:35:43 <Eddi|zuHause> or you could talk to George and ask him to re-license the sprites
13:35:56 <Gadg8eer> And yes, the trucks would be super fast and mostly invisible.
13:37:04 <Gadg8eer> I'll try asking later today. I was up way too late.
13:37:31 <andythenorth> some of the ECS sprites were Zimmlock's anyway
13:37:46 <andythenorth> if you dig through forums history, you might find them
13:37:58 <andythenorth> they might be licensed differently
13:38:30 <Eddi|zuHause> well he's not around anymore, i guess?
13:38:50 <Eddi|zuHause> licenses are complicated
13:39:24 <Borg> but they should be simple.. and I dont get ppl who do stuff for free for OSS game.. w/ picky licence.. its just fucking dumb
13:39:46 <Borg> licence should allow to do whatever you want... as far as 1) it stays free 2) opt - credits
13:40:13 <andythenorth> but historically that hasn't been the case in TTDPatch land
13:40:34 <andythenorth> it was really OpenTTD that drove GPL
13:40:50 <andythenorth> it's one of the reasons the community fragmented
13:42:37 <LordAro> i suppose it was ludde that picked GPL?
13:51:51 <Eddi|zuHause> yes-ish, maybe orudge was involved. but what introduced GPL to the newgrf world was much later, when the gpl-driven openttd community started to reimplement the grf specs and started to contribute their own grfs, plus synergies between devzone and bananas
13:53:40 <Eddi|zuHause> that basically created a generational split between the "old guard" and influx of new people
13:56:42 <andythenorth> it seemed to be a lot more about trust and artist's rights in the previous community
13:57:21 <andythenorth> and then unhelpful people like me turn up, churn out endless sprites and dump them all under GPL
13:57:27 <andythenorth> probably devalues artistry
13:57:32 <andythenorth> probably a bit annoying
13:57:48 <Timberwolf> Old-school modding communities always used to be like that (all about the licence terms like "don't upload elsewhere, don't post screenshots without crediting me, don't use this mod if you like POP or RAP music...")
13:58:05 <andythenorth> all seemed like social backscratching
13:58:17 <andythenorth> can be just sucking up and unspoken power structures
13:58:44 <andythenorth> wonder how many sprites I've drawn
13:58:54 <andythenorth> probably not as many as Zeph for opengfx :P
14:00:23 <Gadg8eer> I noticed how bad it was. I can understand Michael Blunck and George wanting to protect their hard work, but a lot of the stuff people were protecting was either overprotected or worthless.
14:01:20 * andythenorth has zero fear of reuse
14:01:23 <Borg> Gadg8eer: did they do that hard work for free?
14:01:35 <Timberwolf> It makes using things like newstats in a multiplayer game almost impossible unless you're a very close group or all super-dedicated players who'd have the grf anyway.
14:01:48 <andythenorth> we know that loss aversion is a classic human fallacy
14:02:25 <Gadg8eer> Simuscape angers me a lot. So many graphics that never got released because they clung to their sinking ship rather than return to tt-foruns.
14:02:43 <andythenorth> I thought tt-forums angered you to?
14:02:47 * andythenorth reads reddit most days
14:03:03 <andythenorth> I didn't ask as tt-forums drama is as old and boring as a sand dune
14:03:12 <Gadg8eer> No. Ufiby angers me, and the forums are on his side.
14:03:28 <Timberwolf> I've gone with GPLv2 or MIT for most of my things (including the tooling)
14:03:40 <andythenorth> the forums don't take a side usually
14:03:51 <andythenorth> it's very rare that Owen intervenes one way or another
14:03:54 <Timberwolf> Felt bad as I did get a PR to Villages a little while ago... for something I was already in the process of developing :(
14:04:20 <Gadg8eer> Ufiby abused a loophole in GPL to not release his GPL v2 source.
14:04:20 <andythenorth> I did try Villages :
14:04:33 <andythenorth> It amused me that it builds road to industries
14:05:47 <Timberwolf> Ooh, that's not in mine. Which script/AI is that?
14:08:04 <andythenorth> oh maybe that was Migrations?
14:16:45 <andythenorth> if I keep going with Sam, I think I end up with FISH
14:21:04 <Gadg8eer> Someone should make a ship set that uses the same scale as the land vehicles. It would of course mean that only small vessels would be possible to draw.
14:22:33 <Gadg8eer> I read somewhere that a realistically sized container ship in OpenTTD would be the size of a small town.
14:23:06 <Eddi|zuHause> yes, those are massive
14:23:51 <Eddi|zuHause> also, airports would be like 10 times the size if scaled the same as houses
14:24:36 <Eddi|zuHause> then there's issues about length vs. width
14:25:10 <Eddi|zuHause> most things in TTD are too wide for their length
14:28:13 <Timberwolf> TTD has charm in lieu of scale.
14:28:54 <Timberwolf> Ships aren't quite as bad as the mismatch between the internal dimensions of a depot vs its external dimensions.
14:29:22 <Timberwolf> TTD depots are actually vast underground bunkers and what you see is merely the surface entrance.
14:29:48 <Borg> I dont really get all that stuff you guys says.
14:30:03 <Borg> orginal TTD graphics is cute.. I dont see problems in scale.. its.. game.. GAME...
14:30:21 <Borg> also.. old one.. so there was limitations that needed to be addresses.. no problems..
14:30:40 <Borg> if you want to build true scale sim.. start from scratch..
14:30:54 <Eddi|zuHause> they might be stargate entrances
14:31:51 <Eddi|zuHause> well, at least with tunnels, the vehicles actually stay on the map :p
14:32:49 <andythenorth> Sam ships are close to train scale
14:33:02 <Borg> for example.. I never use any trains sets GRFs... because most of them use larger that 1/2 tile wagons size.. with.. look damn ugly..
14:33:03 <andythenorth> they're not in scale with RVS because RVS are slightly oversized
14:33:12 <Borg> its pity.. because it would be cool to use some other nice train sets..
14:33:25 <Borg> orginal graphics just looks good.. and fits..
14:33:55 <andythenorth> of all the reasons to not like Iron Horse, lack of 1/2 tile is not a valid one :P
14:35:08 <andythenorth> and yes, game art is game art
14:35:13 <andythenorth> depots etc never bother me
14:53:03 <Flygon_> <Timberwolf> TTD depots are actually vast underground bunkers and what you see is merely the surface entrance.
14:53:11 <Flygon_> I just assumed they were TARDIS's. :D
14:53:14 *** Flygon_ is now known as Flygon
14:55:35 <andythenorth> definitely tardis
14:55:43 <Eddi|zuHause> hey, now that you mention it, that would be a good GRF :p
14:56:49 <Eddi|zuHause> can we find a suitible backronym for DEPOT? :p
14:57:52 <Flygon> Depostory Enhanced Production of Trains...?
14:58:01 <andythenorth> hmm time to fix CHIPS makefile
14:58:06 * andythenorth never learnt how Make works
14:58:46 <Eddi|zuHause> lesson 1) it's "make" (lower case) and "Makefile" (upper case)
14:59:23 <Eddi|zuHause> lesson 2) tab characters are relevant syntax, not just decoration
14:59:47 <glx> andythenorth: basically replace hg with git
14:59:50 <Eddi|zuHause> lesson 3) everything else i have to google again whenever i touch it
15:00:09 <andythenorth> I think I'll drop the coop makefile
15:00:42 <glx> and I tried to look at chips nfo, it's very complex
15:01:08 <andythenorth> the makefile is 9 separate modules and > 20k
15:01:36 <andythenorth> I'll try the FIRS one, it's about 5k in a single script
15:03:00 <andythenorth> how hard can it be? :P
15:03:51 <glx> you just need to fix version detection in Makefile, no need to replace everything
15:04:00 <Eddi|zuHause> the complexity of a makefile comes from the fact that it must work for people other than yourself
15:05:14 <andythenorth> we no longer have anything like Jenkins for grfs
15:05:28 <andythenorth> I don't do anything to support 3rd parties compiling
15:06:38 * andythenorth wonders how make works
15:06:45 <andythenorth> this is a bi-annual question
15:06:52 <Eddi|zuHause> that's a very heavy bus factor
15:07:03 <andythenorth> if I am under the bus I don't mind, I'm dead :)
15:07:42 <andythenorth> what's good for a grf isn't what's good for OpenTTD
15:08:15 <andythenorth> trying to support weird 3rd party crap is way beyond my skillset
15:08:38 <Eddi|zuHause> makefile is actually very simple, the basic structure is "target: deps (commmand)*"
15:09:00 <andythenorth> does it build a tree and then recurse or something?
15:09:30 <Eddi|zuHause> for every target, it looks at the deps timestamp, and recurses if outdated, then runs the commands
15:09:58 <glx> but really the only broken stuff in chips Makefile is use of hg
15:10:04 <andythenorth> so it's just comparing A newer than B?
15:10:31 <andythenorth> glx think you could patch it? :)
15:10:46 <andythenorth> I have no idea what that makefile is supposed to do, so I don't know how to test it
15:10:53 <glx> well I worked around it with makefile_dist ;)
15:11:11 <Eddi|zuHause> forach dep: recurse; if dep newer than target: run commands
15:12:08 <glx> but if you want to rewrite makefile maybe switch to cmake ;)
15:13:34 <Eddi|zuHause> the tricky part of makefiles is getting the deps correctly, preferably without listing them manually
15:16:20 <FLHerne> Eddi|zuHause: tbh, my experience of the "standard" -coop makefile setup is that it doesn't really help
15:17:13 <FLHerne> Discovering all the variables and setting them appropriately is harder than just editing things :p
15:17:34 <Eddi|zuHause> dunno about it, i use a really outdated version that i didn't set up myself
15:17:43 <andythenorth> even the Iron Horse makefile is really overkill for CHIPS
15:18:09 <andythenorth> it's pretty much a shorter way to type 'grfcodec -e chips.grf'
15:18:14 <Eddi|zuHause> i also made some modifications 6 years ago that i barely understand now
15:18:16 <andythenorth> I could replace all of it with a symlink
15:30:57 *** greeter has joined #openttd
15:40:08 <andythenorth> eh lol :) I forgot CHIPS was CPP templated
15:40:14 <andythenorth> well that's retro
15:58:47 <milek7> tracking dependecies of #include is rather painful
16:16:58 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 opened pull request #8176: Fix #7970: [Win32] Disable event loop on crash to prevent recursive faults https://git.io/JfKt6
16:17:25 *** jottyfan has joined #openttd
16:20:34 <Speeder_> andythenorth, who makes FIRS art?
16:20:48 <andythenorth> mostly me though
16:20:57 <Speeder_> I think I will fork OpenGFX+ instead and use FIRS sprites, if you don't mind :)
16:21:17 <Speeder_> FIRS codebase is getting in the way instead of helping (although the graphviz thing is awesome)
16:22:06 <andythenorth> I really don't understand make
16:22:23 <andythenorth> $(NFO_FILE) should now be a dep for $(GRF_FILE)
16:22:49 <andythenorth> and $(NFO_FILE) should cause gcc or clang to build sprites/chips.nfo
16:24:15 *** virtualrandomnumber has joined #openttd
16:25:41 <andythenorth> oh it's the $(_E) "[CPP] $(NFO_FILE)" line that's failing
16:26:27 <andythenorth> it's looking for the actual file
16:26:30 <andythenorth> which is not found :P
16:28:55 <andythenorth> now I have sprites/chips.nfo but empty
16:28:58 <andythenorth> mystery continues
16:29:05 <andythenorth> CPP compile fails
16:29:19 <glx> ah no there's no NFO_FILE definition
16:30:39 <andythenorth> then fix the missing vars that sed wants
16:38:36 <andythenorth> oh this is lolz /bin/sh: sprites/chips.nfo: Permission denied
16:39:34 <andythenorth> ok I think I quit
16:39:38 <andythenorth> this is way too hard :)
16:40:04 <andythenorth> the nfo generated from the pnfo is not valid
16:40:17 <andythenorth> the problems are piling on top of the problems :D
16:41:04 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8176: Fix #7970: [Win32] Disable event loop on crash to prevent recursive faults https://git.io/JfKqa
16:41:38 <andythenorth> but if I just use clang to compile the pnfo myself, the nfo is then fine
16:41:52 <andythenorth> I'm using the same command as the Makefile :P
16:42:52 <andythenorth> I guess I should look up what these flags do :) clang -C -E -nostdinc -x c-header - < sprites/nfo/chips.pnfo
16:57:52 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #8176: Fix #7970: [Win32] Disable event loop on crash to prevent recursive faults https://git.io/JfKmg
17:03:15 * andythenorth deletes all the variables
17:03:35 <andythenorth> none of it works, burn it with fire
17:05:43 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #8176: Fix #7970: [Win32] Disable event loop on crash to prevent recursive faults https://git.io/JfKmA
17:08:46 <andythenorth> apparently the grfid for CHIPS is \"CHPS\"
17:08:56 <andythenorth> I have no idea how to get that into the nfo manually :D
17:09:03 <andythenorth> the sed substitutions no longer work
17:09:20 <andythenorth> if I paste it in, nforenum errors
17:10:00 <andythenorth> the grf does build fine, everything in it as expected
17:10:13 <andythenorth> only the compile is borked
17:10:25 <FLHerne> andythenorth: You probably don't want the slashes?
17:10:54 <FLHerne> I expect they're escaped to stop the shell eating them at some point during the make process
17:11:04 <andythenorth> I might have to resort to reading docs :P
17:11:04 *** Wormnest has joined #openttd
17:11:30 *** Wrench_In_The_Plan has joined #openttd
17:54:39 *** mindlesstux666 has quit IRC
17:55:23 <andythenorth> gets me make, make clean, make bundle_tar
17:55:37 <andythenorth> I haven't tested it detects deps correctly yet
18:02:07 <andythenorth> deps appear to work
18:02:53 *** virtualrandomnumber has quit IRC
18:10:38 <andythenorth> I'm not happy about commented out L69 and L71
18:10:57 <andythenorth> couldn't stop my shell trying to expand the filename in the var, instead of treating it as a string
18:11:33 <andythenorth> might just delete them
18:36:21 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #8176: Fix #7970: [Win32] Disable event loop on crash to prevent recursive faults https://git.io/JfKG4
18:51:59 <andythenorth> curious how sed works
18:52:14 <andythenorth> given this: REPLACE_REPO_VERSION
18:52:25 <andythenorth> in context: 2 * 67 08 07 "CHPS" "CHIPS Station Set" 00 "CHIPS Station Set {{REPLACE_REPO_VERSION}}" 00
18:52:47 <andythenorth> I have cargo culted this from old makefile
18:52:47 <andythenorth> sed -e "s/{{REPLACE_REPO_VERSION}}/$(REPO_VERSION)/"
18:53:23 <andythenorth> the replacement does not happen
18:53:55 <andythenorth> maybe the order is reversed
18:55:38 <Yexo> $(REPO_VERSION) should be ${REPO_VERSION} maybe?
18:55:46 <Yexo> Might depend on your shell, but that's how it works for me
18:56:11 <LordAro> Yexo: this is inside a makefile, so $(FOO) has already been replaced
18:56:24 <Yexo> oh, right, make replaces it
18:56:50 <Yexo> andythenorth: needs more context. how/where do you call sed?
18:56:58 <LordAro> andythenorth: by "does not happen", do you mean "{{FOO}}" remains, or that it's replaced with nothing?
18:57:25 <LordAro> try using single quotes?
18:57:50 <LordAro> shell might be trying to expand the {}
18:57:50 <Yexo> are you redirecting input/output correctly? If you are expecting sed to update a file in-place you need the -i flag
18:58:22 <LordAro> cat $< > sed ... > $@
18:58:39 <LordAro> except the first '>' should be a '|'
18:59:00 <LordAro> no, but i was keeping it simple
18:59:28 <LordAro> yeah that's a bit weird
18:59:32 <andythenorth> so far I have only cargo-culted from coop makefile, possibly missing something
18:59:36 <Yexo> andythenorth: line 69 redirects all output to $(NFO_FILE), so no output gets redirected to sed
18:59:52 <Yexo> Which also tries to write to the same file? That's definitely wrong
19:00:10 <andythenorth> the file output in L69 is a mistake
19:00:33 <andythenorth> just clown shoes :)
19:00:56 <andythenorth> oops wrong text box
19:12:42 <andythenorth> this nearly works
19:12:54 <andythenorth> just trying to remember where OpenTTD gets the display name of the grf from
19:15:31 *** arikover has joined #openttd
19:21:18 <andythenorth> sed is powerful eh
19:21:22 <andythenorth> and slightly odd
19:21:33 <andythenorth> and probably dangerous
19:21:42 <andythenorth> in untrained hands
19:22:38 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 updated pull request #8176: Fix #7970: [Win32] Disable event loop on crash to prevent recursive faults https://git.io/JfKt6
19:23:12 <glx> any tool is dangerous in wrong hands
19:24:44 <andythenorth> glx fwiw I think CHIPS should compile now without workarounds, I've pushed master
19:27:10 <glx> hmm seems to fail here, need to check
19:30:15 * andythenorth wonders about GH actions builds of newgrf
19:30:30 <andythenorth> CI might result in better makefiles
19:31:42 <andythenorth> no annotated tag?
19:31:50 <andythenorth> I had to tag -a test
19:37:40 *** sla_ro|master has joined #openttd
19:42:11 *** Progman has joined #openttd
21:26:03 *** jottyfan has joined #openttd
21:35:15 <Wolf01> Hmmm, japanese light novels don't help at all to fix my mood :|
21:50:07 <andythenorth> I haven't read many japanese novels
21:50:17 <andythenorth> but very few of the ones I read were uplifting
21:50:25 <andythenorth> some were aweseome though
21:59:58 <nielsm> light novels are quite a different style from "real" novels though
22:03:37 <nielsm> light novels are basically anime in novel form
22:03:58 <nielsm> except the barrier to get published is lower than the barrier to making animation
22:06:58 *** mindlesstux666 has joined #openttd
22:08:37 <nielsm> is that equivalent to walking into the door frame on the way out?
22:14:38 *** Wrench_In_The_Plan has quit IRC
22:23:39 <Wolf01> I should learn to not use the classic snippet tool while there is an UWP remote desktop active... recovering from that nasty situation is a pita
22:53:39 *** mindlesstux6662 has joined #openttd
22:53:47 <Speeder_> nmlc warning: "src/tiles/gold_mine_tiles.pnml", line 64: "src/gfx/gold_mine/gold_mine.png" at [x: 162, y: 258]: 240 of 1984 pixels (12%) are animated, but ANIM isn't in flags
22:54:02 <Speeder_> this happens when I do a "make" on opengfx+
22:54:36 <nielsm> it uses pixels with a palette index that is animated (colour cycling)
22:55:03 <nielsm> so the warning is that the sprite will have animation but you maybe did not intend that since you didn't put the ANIM flag on the sprite definition in the nml
22:55:05 <Speeder_> what editor people use for the png files?
22:55:08 <Speeder_> that won't break them?
22:57:01 <Speeder_> one of the worst itnerfaces I ever saw
22:57:02 <nielsm> you can also use other editors that support indexed colour mode
23:00:59 *** mindlesstux666 has quit IRC
23:01:20 <Speeder_> tried krita.. seemly it doesn't accept it :(
23:05:30 <nielsm> I don't think krita supports indexed modes
23:06:28 <FLHerne> If you're on Windows, I hear Paint.NET is pretty good for that
23:19:06 <Speeder_> anyone here work with pnml?
23:20:03 <glx> it's just nml with c preprocessor directives
23:21:52 <Speeder_> glx, can it output stuff that won't go on the nml itself?
23:22:01 <Speeder_> I wanted to see if I could write some kidn of graphviz documentation on it
23:22:56 <nielsm> sure you can extend it however you want
23:28:55 <supermop_Home> Speeder_ photoshop is easy
23:29:41 <supermop_Home> andythenorth I've mostly just read murakami
23:29:48 <supermop_Home> and then some older stuff
23:30:12 <supermop_Home> never made it through tale of genji yet, though that's not really a 'light' novel
23:46:25 <Speeder_> supermop_Home, I don't own photoshop, aaaand... I don't like it either
23:46:34 <Speeder_> I miss Macromedia Fireworks D:
23:51:06 <supermop_Home> blast from the past
23:51:42 *** Wrench_In_The_Plan has joined #openttd
continue to next day ⏵