IRC logs for #openttd on OFTC at 2019-03-11
⏴ go to previous day
00:00:05 <Samu> "some docks were removed after savegame load"
00:00:54 <Samu> but what are you trying to do?
00:01:29 <peter1138> I want an efficient way to detect docking tile of a dock.
00:03:34 <andythenorth> let's do more FIRS
00:03:45 <andythenorth> hmm, that's not docks though
00:04:06 <andythenorth> what's the inefficient way peter1138 ?
00:05:15 <peter1138> Currently it iterates the docking tiles of the destination station to compare it with the pathfinder's current tile.
00:05:28 <peter1138> And that... is for every tile the pathfinder explores.
00:06:23 <peter1138> If the docking tile becomes an actual station tile, then it's a simple "is this the right station" check.
00:06:33 <peter1138> but that has the 3rd-tile issue.
00:07:21 <andythenorth> nobody mentioned state machines yet?
00:07:32 <peter1138> andythenorth, no because that's irrelevant.
00:07:44 <peter1138> A state machine would be what you do when you're at the dock.
00:07:56 <peter1138> You still need to find it.
00:08:14 * nielsm thinks of RCT boat rides
00:08:22 <peter1138> The old old way that bus stops used to use was to preallocate a specific stop to use.
00:08:44 <peter1138> then you just sent a vehicle to a specific tile, and it's easy.
00:08:45 <andythenorth> hmm is cb 37 working with > 3 cargos?
00:09:47 <nielsm> there isn't any specific nml support for it but I don't think that's needed
00:09:53 <nielsm> but the bitstuffed format has changed
00:10:13 <peter1138> Yes, that uses cargo ids, not bitmaps.
00:10:42 <Samu> can't the pathfinder be given multiple tiles?
00:12:36 <Samu> when I was toying with road pathfinder on my ai, i could insert multiple destinations
00:12:40 <andythenorth> I might have it configured wrong though
00:12:52 <Samu> never actually used it though
00:13:05 <andythenorth> the supplied cargos should be manganese (correct) and oxygen
00:13:25 <andythenorth> it's not economy aware, and doesn't use labels
00:13:31 <andythenorth> just relies on positions
00:13:36 <andythenorth> so might be my error
00:14:17 <nielsm> you should first extract bits 0..7 and compare against 0 and 1 (input/output)
00:14:48 <nielsm> actually maybe compare bits 0..15 in one go if you only need to handle a few cases
00:15:32 <nielsm> then after selecting the display case, compare bits 16..23 against cargotype("COAL") or whatever
00:15:40 <nielsm> to select the final string
00:15:58 <andythenorth> ok that's quite an advance on the original approach
00:16:49 <nielsm> probably 0..15 against 0x0100 for accepted cargo in the industry's own status window
00:17:05 <nielsm> and the rest might not need to be handled at all?
00:18:04 <andythenorth> original FIRS code just checks
00:18:05 <andythenorth> getbits(extra_callback_info2, 0, 8)
00:18:22 <andythenorth> and then chooses a text if any cargo has been delivered for that cargo num
00:19:36 <andythenorth> what sets prop 1A flag 18
00:19:41 <peter1138> bits 0..7 used to contin the slot number.
00:20:13 *** Thedarkb1-T60 has joined #openttd
00:20:15 <peter1138> 0-2 were inputs, 3 & 4 were outputs
00:20:29 <andythenorth> yeah the new format replaces
00:21:00 <andythenorth> oof me eyes are going, it's bedtime
00:21:05 <peter1138> Now you don't get the slot, you get the grf-local cargo type in bits 16..23
00:21:18 <peter1138> and bits 0..7 contain the in vs out
00:21:31 <andythenorth> I need to set the flag on 1A first
00:22:21 <andythenorth> oh yes, that's not in the nml docs yet :)
00:22:29 <peter1138> Shouldn't that be set already if you're doing 16-in-out alreadY/
00:23:13 <andythenorth> IND_FLAG_LONG_CARGO_TYPE_LISTS
00:23:13 <nielsm> it's not strictly needed no
00:23:43 <nielsm> but maybe nml should just set it by default now
00:23:48 <peter1138> Yeah, if you're able to do 16-in/16-out, then nml should've just set it/
00:24:03 <nielsm> I think I need to get sleep (too)
00:24:18 <peter1138> If that bit isn't set, then itwill only do 3/2
00:25:04 <andythenorth> hmm FIRS has crappy support for spec_flags, never needed them in industries
00:25:08 <peter1138> Samu, yeah, the problem is testing multiple tiles every time it moves to a new tile.
00:25:11 <andythenorth> job for tomorrow
00:25:38 <peter1138> Anyway, shouldn't nml handle all this testing bits for you?
00:26:03 <andythenorth> it might be in nml now
00:26:07 <andythenorth> I haven't got that far yet
00:26:25 <andythenorth> 'old' FIRS had to check the bits, because nml wasn't maintained for this
00:27:50 <peter1138> Yeah, and this is why "nmlc should allow old stuff to work" just won't work.
00:28:46 <andythenorth> nmlc needs to be quite scorched earth
00:29:39 <andythenorth> I can't see an nml commit for cb 37 changes
00:32:29 <andythenorth> also super defo bedtime
00:34:18 <peter1138> Set if tile is docking tile.
01:29:51 <peter1138> Whew, that was a horrible performance regression there...
01:29:57 <peter1138> Turns out, debug-level 3 :/
01:50:45 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison updated pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/fh2Pg
01:52:06 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison commented on pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/fhjfs
02:20:12 *** Supercheese has joined #openttd
02:25:42 *** octernion has joined #openttd
02:32:45 *** supermop_Home_ has quit IRC
02:49:18 *** supermop_Home_ has joined #openttd
03:02:10 *** Beerbelott has left #openttd
03:04:16 <supermop_Home_> why do no servers have cdist on?
03:04:22 <supermop_Home_> or use newgrfs
03:09:46 <Eddi|zuHause> <peter1138> It's simple to add a 3rd tile that is actually a station tile. <-- i would probably do that
03:14:51 <supermop_Home_> i hope these other players are AIs
03:15:49 <Eddi|zuHause> supermop_Home_: apparently there's too little overlap between players that want to play online and those that can accept a certain set of newgrfs that they did not hand.pick themselves
03:19:29 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison opened pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fhjJc
03:19:34 <supermop_Home_> i would settle for almost anything other than tropic base set trains
03:20:10 <Samu> hey supermop_Home_ you're on my game?
03:23:19 <supermop_Home_> hard to find anywhere to fit tracks between the ai rails
03:24:23 <Samu> I could try cargodist sometime
03:26:05 <Samu> which newgrfs do you want? I don't usually try newgrfs much
03:26:29 <supermop_Home_> basic stuff is pretty good, like iron horse
03:27:45 <Samu> there's iron horse, iron horse quicksilver add on, iron horse 2, termite
03:32:20 <Samu> what about ships, aircraft, road vehicles?
03:32:32 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison opened pull request #7367: Change: Include _current_company in crashlog AI config line https://git.io/fhjJa
03:33:30 <Samu> so many newgrfs that I have a hard time to pick
03:35:41 <supermop_Home_> RVs i guess it doesn't matter as much, just something generic but that can be refitted, has better capacity than the defaults
03:36:46 <Samu> uhm, which ones? I'm very unfamiliar regarding newgrfs
03:39:36 <supermop_Home_> road hog is good, opengfx+ RVs is good
03:40:10 <supermop_Home_> all of those will let you use newgrf industries
03:40:41 <supermop_Home_> and they are all pretty similar to the original vehicles, just a little nicer looking and more flexible
03:41:09 <Samu> would be a nice test for AIs
03:41:37 <Samu> ships and planes, anything worth using? and which industry newgrf?
04:21:56 <Samu> supermop_Home_, you there?
04:30:08 <supermop_Home_> going for a walk
04:52:50 *** ChanServ sets mode: +v tokai
05:39:45 *** supermop_Home_ has quit IRC
06:33:35 *** octernion has joined #openttd
06:50:05 *** Supercheese has joined #openttd
07:26:08 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fhjkm
07:47:36 <peter1138> Eddi|zuHause, it's not simple :(
08:09:23 *** andythenorth has joined #openttd
08:13:09 <peter1138> No custom tiles yet.
08:13:36 <peter1138> Station tile layout or house/industrytile/object style?
08:14:19 <peter1138> red square is debugging for docking location for pathfinder.
08:14:20 <andythenorth> is there any reason to choose station tile layout?
08:14:28 <peter1138> andythenorth, only that it's already there.
08:14:39 <andythenorth> will this end up newgrf configurable?
08:15:01 <peter1138> But I think it may be easy to bolt on the "new style" for non-rail stations.
08:15:10 <peter1138> (Well, even rail stations but not tackling that)
08:15:17 <andythenorth> the old style never gained nml support
08:15:36 <andythenorth> but you probably want to add station support to nml, right?
08:15:52 <peter1138> I want to get this working.
08:16:01 <andythenorth> I lke the tile format for industrytile
08:16:06 <andythenorth> it works for me, makes sense
08:16:07 <peter1138> The red squares is just highlight debugging for docking tile
08:16:15 <peter1138> As I have to update it periodically if the tile gets changed.
08:16:34 <peter1138> Plan is the docking tile will be able to be adjacent to andy dock tile.
08:16:40 <peter1138> so single-tile docks.
08:16:48 <andythenorth> arbitrary slopes on the tile? o_O
08:16:51 <peter1138> Dock tile can then be placed on slope, or flat land.
08:17:00 <peter1138> yes, maybe any slope
08:17:05 <peter1138> direction isn't going to be important.
08:17:08 <andythenorth> arbitrary docking points in industries? o_O
08:17:15 <andythenorth> industry docks are weird :P
08:17:40 <peter1138> And it should be possible to place dock tiles... in water.
08:17:46 <peter1138> Which opens up a can of worms
08:18:07 <peter1138> maybe that won't be possible
08:18:15 <peter1138> you could just spam water tiles with docks then
08:18:47 <peter1138> YAPF pathfinding seems to work.
08:18:50 <peter1138> I haven't tackled NPF.
08:20:09 <peter1138> Hmm, SLV_MULTIPLE_DOCKS should... maybe SLV_MULTITILE_DOCKS
08:22:00 <peter1138> Hmm, should a buoy be able to be a docking tile?
08:23:17 <peter1138> What about a ship depot... I think that currently works but is kinda odd.
08:24:18 <peter1138> I haven't done anything with reserving slots so ships always take the nearest docking tile.
08:28:53 <andythenorth> a ship depot as dock?
08:29:41 <peter1138> Not a dock, a docking tile.
08:32:33 <peter1138> ^ That is currently an arrangement that works.
08:32:52 <peter1138> As long as the ship can reach the docking tile...
08:34:57 <andythenorth> what if towns build blocking the tiles? o_O
08:35:11 <andythenorth> docking tiles on aqueducts?
08:35:59 <peter1138> If towns block the tiles then Samu will tell us about it.
08:39:37 <andythenorth> docking tiles in locks!
08:39:45 <andythenorth> ship goes up and down while docked
08:45:50 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh dismissed a review for pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/fhhh5
08:48:17 <peter1138> Eh, not actually IN the lock.
08:48:22 <peter1138> But yes, either end is possible.
08:51:37 * andythenorth tries to think of any more silly ideas
08:52:08 <andythenorth> all out of water features
08:52:19 <andythenorth> dock tiles in ship tunnels? o_O
08:55:44 <andythenorth> shall we add minecarts?
08:57:17 <peter1138> I need to make lunch.
08:57:31 <peter1138> And maybe have more breakfast. Is one slice of toast sufficient?
09:00:00 <andythenorth> you need melon and some oranges
09:00:22 <peter1138> I have clementines but no melon.
09:00:56 <peter1138> Ok, m1 bit 7 is free for station tiles too, that's good.
09:01:03 <peter1138> And, indeed, tunnels and bridges.
09:04:58 <Flygon> Oh nuts, can't buy out a human competitor in OTTD. x.x
09:05:11 <Flygon> Promised a friend I'd merge the two companies so he'd have an easier introduction into MP.
09:16:05 <Flygon> He made his own company, but felt overwhelmed by infrastructure sharing.
09:17:34 <peter1138> So you're not playing OpenTTD. Traitor!
09:19:26 *** Gustavo6056 has joined #openttd
09:20:21 <peter1138> thats even worse ;)
09:23:35 *** Gustavo6056 is now known as Gustavo6046
09:39:41 <Flygon> So the only option is for them to totally disassemble their company?
09:39:46 <Flygon> No way to convert it into an AI player?
09:44:00 *** andythenorth is now known as Guest2995
09:44:00 *** andythenorth has joined #openttd
10:11:05 <TrueBrain> I have been thinking ... talking about cutting old: we should drop DOS support. It has no threads (not fixable), it has no network (somewhat fixable, but nobody is going to do it), it has no cmake (NOOOO), any "real" DOS machine won't have the memory, it has bugs nobody is going to fix, it bloats our codebase for "a joke" (you can say a lot, but I
10:11:06 <TrueBrain> strongly doubt we have a real DOS userbase :P) .. so .. how about we just remove it? All ENABLE_NETWORK statements can be removed, some other special-case code can just vanish, etc. Too drastic? Agree? :D
10:16:04 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain merged pull request #7363: Lot of small codefixes related to configure https://git.io/fhhdG
10:17:46 <DorpsGek_II> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fhjLF
10:20:19 <LordAro> TrueBrain: ask about it in the monthly dev blog :)
10:20:48 <TrueBrain> I guess that is fair
10:20:56 <TrueBrain> we can also ask about OS/2 and Haiku? :P
10:22:17 <LordAro> we could, but idk - i get dropping old and dead operating systems, but dropping little-used OSes because they're little-used is a bit self fulfilling
10:23:44 *** kiwitree has joined #openttd
10:29:00 <TrueBrain> LordAro: but there is little used, and never used :)
10:29:09 <TrueBrain> and we are not dropping it because it is little used
10:29:17 <TrueBrain> we are dropping it because we have a huge technical debt because of it
10:29:28 <TrueBrain> or you means OS/2 and Haiku?
10:29:36 <TrueBrain> sorry, my mind didn't change state :P
10:29:44 <TrueBrain> isn't OS/2 also dead?
10:29:50 <TrueBrain> and yeah, okay, Haiku is fine I guess :P
10:29:57 <TrueBrain> someone needs to fix the bugs in network, but .. that is minor
10:30:32 <Eddi|zuHause> how about adding android/iOS as supported platforms?
10:31:00 <TrueBrain> go for it Eddi|zuHause
10:31:10 <TrueBrain> hmm .. why is spelling check not working in GitHub editor :(
10:31:23 <LordAro> perhaps, but also Haiku & OS/2 are much lower maintrnance than DOS, as they do have networking, threads & cmake
10:32:39 <TrueBrain> LordAro: yeah, I am fine with Haiku
10:32:42 <peter1138> Eddi|zuHause, I've managed to get it compiled for Android.
10:32:43 <TrueBrain> OS/2 .. no clue what the state is
10:32:50 <peter1138> Eddi|zuHause, I've not tried running it yet though.
10:32:53 <TrueBrain> I haven't test it, and I doubt someone else has :D
10:33:03 <peter1138> Because I compiled it without a video driver :p
10:33:15 <TrueBrain> Eddi|zuHause: the biggest thing to fix in Android/iOS, is UI I guess :P
10:33:48 <TrueBrain> LordAro: but okay, asked in the post about DOS; we will see if people reply :)
10:33:49 <peter1138> It needs to be done in a nice generic way.
10:33:54 <peter1138> Unlike what the patches do.
10:34:21 <peter1138> Toolbar buttons down the side of the screen might work quite well on desktop too.
10:34:47 <TrueBrain> after porting to various of platforms in the old days, my personal opinion became that I do not believe OpenTTD is a playable game on a mobile device (including tablets)
10:35:01 <TrueBrain> I love to be proven wrong :D
10:35:15 <Eddi|zuHause> yeah, switching to a "fat finger/touchscreen/miniscreen" mode must be generic enough to work on every OS, with maybe a detection routine to set a sane default
10:36:13 <peter1138> TrueBrain, yeah it needs a mouse pointer, I think.
10:36:20 <dwfreed> because you could be running openttd on windows with a touchscreen display (like a surface pro)
10:36:22 <Eddi|zuHause> it could incorporate the gui zoom/font zoom features
10:36:54 <TrueBrain> Eddi|zuHause is the project owner
10:36:58 <TrueBrain> any questions can be directed to him
10:37:05 <Eddi|zuHause> that sounds like a horrible idea :p
10:37:14 <TrueBrain> that's what she said
10:37:14 <LordAro> dwfreed: x2 gui zoom, done
10:37:42 <Eddi|zuHause> ... and she was probably right
10:38:14 <planetmaker> there is an android port... it surely could be canibalized to make it proper
10:39:01 <peter1138> i have it imported heh
10:39:30 <peter1138> first step is to fix the build system
10:39:49 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fhjtB
10:40:42 <peter1138> gui changes are obviously required but come later
10:41:18 <peter1138> sdl 1.2 android port is clunky, sdl 2 might be better
10:43:57 <Eddi|zuHause> have we settled on the debate of whether sdl2 support should remove sdl1.2 support or be a separate driver?
10:44:44 <Eddi|zuHause> i see the potential for lots of fringe cases breaking with sdl2
10:45:32 <peter1138> at the moment its not ready
10:45:36 <Eddi|zuHause> might be helpful to tell those people to use "-v sdl1"
11:01:02 <TrueBrain> I am so happy I put a space in my folder name
11:01:12 <TrueBrain> once again glx broke support for that :P (it is either he or me doing it :D)
11:07:50 <Eddi|zuHause> good test case :)
11:08:20 <peter1138> Do we want the CI to build with fluidsynth suport?
11:08:55 <peter1138> It would be an extra library for users to install.
11:09:34 <planetmaker> for the end user?
11:10:11 <planetmaker> well. On linux it doesn't hurt, if you install via package manager. Or you compile it in statically
11:10:42 <TrueBrain> peter1138: we only publish Debian packages that users can use on Linux
11:10:43 <peter1138> That post would suggest it is. Hmm.
11:10:51 <TrueBrain> the CI doesn't produce downloadable binaries
11:11:15 <TrueBrain> so I guess your question is two-fold: should we change os/debian, os/rpm, etc to include fluidsynth
11:11:25 <TrueBrain> and should we add fluidsynth to the CI so it is validated it still works?
11:11:35 <TrueBrain> on that last note, guess it would be good to install allegro etc too
11:11:38 <Eddi|zuHause> i think enough people requested a generic linux binary that we should consider providing one
11:12:04 <TrueBrain> we considered it. The answer resulted in a ticket
11:12:07 <peter1138> Eddi|zuHause, we should, just needs someone™ to set it up :)
11:12:38 <peter1138> TrueBrain, sorry I was conflating the two, I wasn't sure if adding it to the CI build rules would also affect the release/nightly builds.
11:12:45 <TrueBrain> we can't keep going in this same circle ;) It is not a question IF we should do it, but WHEN someone is going to :P
11:13:14 <TrueBrain> exactly the reason I split them in two flows btw :)
11:14:42 *** andythenorth has joined #openttd
11:16:36 <peter1138> Yeah so allegro too? Or are we dropping that?
11:17:55 <TrueBrain> for now, lets keep it
11:18:04 <TrueBrain> CMake showed me a lot of these libraries that are missing on the CI
11:18:41 <TrueBrain> yeah, fluidsynth and allegro are not fixed yet on the CI
11:26:48 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fhjqP
11:27:14 <andythenorth> TrueBrain: drop DOS, it's showboating to keep it
11:27:23 <andythenorth> I kill jokes quite often in newgrf
11:27:33 <andythenorth> after a few years, they're no longer funny
11:33:21 *** synchris has joined #openttd
11:46:15 <peter1138> On branch multiple-docks
11:46:15 <peter1138> nothing to commit, working tree clean
11:48:26 <peter1138> TrueBrain, hmm, I guess Linux dependencies are done differently from OSX?
11:48:38 <peter1138> It was azure-pipelines for OSX
11:49:12 <peter1138> I wonder if fluidsynth works on OSX? :D
11:51:17 <peter1138> andythenorth, did you ever try the palette "fix" I made? (SDL 1.2 only, dunno if you can run that)
12:05:41 <peter1138> Night mode sounds particularly useless.
12:05:49 <peter1138> Seeing as a game day takes about 2.5 seconds...
12:11:26 <Eddi|zuHause> i was browsing through some patches i stored ages ago, and among those was one that looked like it applied some palette magic to switch to night mode
12:12:31 <Eddi|zuHause> it was called "rAncient" and referenced .c files :p
12:15:34 <peter1138> Yeah, palette magic is actually easy, especially now we're in 32bpp mode these days.
12:15:48 <peter1138> It's just the lights, and the practicality of it.
12:18:27 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh approved pull request #7367: Change: Include _current_company in crashlog AI config line https://git.io/fhjmD
12:28:45 <TrueBrain> peter1138: Linux are done via Docker, as we need to run on different "OSes" there
12:28:55 <TrueBrain> OSX only has one, as does Windows, so they are done directly via Azure Pipelines
12:29:04 <peter1138> Yup. That makes sense.
12:29:07 <TrueBrain> bit annoying, but what can you do
12:29:49 <peter1138> I'll have to figure out how to play with docker. No bad thing.
12:30:17 <TrueBrain> I have to figure out why my spinning disk is causing hickups on my system :(
12:30:28 <TrueBrain> to see how the Dockers are build (and run) you can check the Azure Pipelines
12:30:42 <TrueBrain> if you want more detailed walk-through, let me know :)
12:31:21 <TrueBrain> okay, my computer is acting REALLY weird now .. lol .. stupid Windows :(
12:31:47 <TrueBrain> only Chrome works .. rest of the windows are stuck :P
12:32:00 <TrueBrain> but only with mouse interactions
12:32:02 <TrueBrain> with keyboard it is fine
12:33:16 <peter1138> I hope your drive is not toast :/
12:33:48 <TrueBrain> rebooted the system, everything is fine again
12:35:16 <Eddi|zuHause> sounds like typical Windows :p
12:35:47 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
12:39:04 *** snail_UES_ has joined #openttd
12:43:41 <TrueBrain> owh, I fixed non-MSVC, but that broke MSVC :D
12:44:02 <Eddi|zuHause> sounds like a tuesday
12:53:13 <TrueBrain> okay ... cmake lost the ability to find msbuild all of a sudden
12:54:49 <TrueBrain> things on Windows work like MAGIC
13:00:57 <peter1138> The std::vector patch is looking a bit... conflicty :/
13:02:48 <peter1138> Can a GameScript take over map creation?
13:03:15 <peter1138> Well, not landscape but town and industry placement.
13:06:43 *** tycoondemon has joined #openttd
13:13:21 <TrueBrain> sounds to me nielsm was just being lazy :P :P :P
13:13:46 <TrueBrain> he even merged something in that repo recently :) So yeah, he for sure has write access :D
13:47:50 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro merged pull request #7367: Change: Include _current_company in crashlog AI config line https://git.io/fhjJa
14:00:34 *** octernion has joined #openttd
14:03:07 <DorpsGek_II> [OpenTTD/OpenTTD] GabdaZM closed pull request #7242: Codechange: Improve performance of town name refresh on viewports https://git.io/fh5iB
14:03:08 <DorpsGek_II> [OpenTTD/OpenTTD] GabdaZM commented on pull request #7242: Codechange: Improve performance of town name refresh on viewports https://git.io/fhjOp
14:13:09 <DorpsGek_II> [OpenTTD/OpenTTD] GabdaZM opened pull request #7368: Codechange: town sign update on population change https://git.io/fhj3k
14:32:55 *** synchris has joined #openttd
14:33:13 <DorpsGek_II> [OpenTTD/OpenTTD] GabdaZM updated pull request #7025: Add #6887: Option to show zone inside local authority boundary of towns https://git.io/fhsu0
14:39:09 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7025: Add #6887: Option to show zone inside local authority boundary of towns https://git.io/fhj31
14:39:29 *** andythenorth has joined #openttd
14:57:27 <peter1138> "These changes/fixes can obviously be controversial, so please discuss and ask aobut them."
14:58:59 <TrueBrain> clicking edit is REALLY easy, so I expect a commit in 20 seconds? :D
15:02:55 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
15:04:07 <TrueBrain> (I assumed you were talking about people's inability to write "about" correct :P)
15:04:43 <peter1138> TrueBrain, no, didn't even notice that.
15:05:01 <andythenorth> is it lunchtime?
15:05:02 <TrueBrain> jumping to the wrong conclusion there :D
15:05:09 <TrueBrain> andythenorth: +/- 2 hours late, sure
15:05:15 <peter1138> andythenorth, oh I see, he's using group liveries to ... do livery changes.
15:05:19 <peter1138> TrueBrain, that's my job :D
15:05:41 <TrueBrain> peter1138: again, my apologies :D
15:05:43 <peter1138> I had lunch. It was low calorie until I ate the roll, crisps, and scone.
15:05:51 <andythenorth> I had: half a packet of cookies, some hula hoops, a tomato, and a lasagne
15:06:04 <TrueBrain> why is MSBuild not on my search path :(
15:06:07 <peter1138> It sounds like you had it.
15:06:15 <andythenorth> that was a snack
15:06:21 <peter1138> I had a lotus biscuit. Not quite half a packet.
15:06:31 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7025: Add #6887: Option to show zone inside local authority boundary of towns https://git.io/fhjs1
15:06:38 <TrueBrain> owh, that error happens because I have VS2019 installed, it says
15:07:30 *** m3henry has joined #openttd
15:09:34 * m3henry is wondering how many merge-fixups are needed when he gets back
15:10:13 <DorpsGek_II> [OpenTTD/OpenTTD] GabdaZM commented on pull request #7025: Add #6887: Option to show zone inside local authority boundary of towns https://git.io/fhjsS
15:15:08 <TrueBrain> okay, CMake is working again \o/
15:15:24 <TrueBrain> baseset generation is also working (tnx to glx)
15:15:27 <TrueBrain> source.list is no longer needed
15:15:32 <TrueBrain> all in all, looking good :)
15:15:58 <TrueBrain> biggest thing left, is regression :)
15:16:13 *** supermop_work has joined #openttd
15:18:56 <peter1138> TrueBrain, last three commits there might be interesting
15:19:08 <TrueBrain> yippie, peter1138 is going to port those to CMake :D \o/ :)
15:19:10 <peter1138> TrueBrain, but i believe that should be done sseparate afterwards
15:19:15 <LordAro> andythenorth: yes please
15:19:18 <peter1138> TrueBrain, yup, that's partly why i stopped developing it.
15:19:51 <peter1138> I have 2 Oreo eggs. They're not quite the same.
15:20:03 <TrueBrain> with CMake it becomes so much easier .. 1 script to do all platforms with
15:20:09 <TrueBrain> instead of all this shell/vbs shit constantly :P
15:20:13 <peter1138> TrueBrain, hopefully
15:20:16 <TrueBrain> but yeah, that would be really great to just generate those files on build
15:20:20 <TrueBrain> instead of storing them in repo
15:20:24 <TrueBrain> which is a bit iffy .. always has been :P
15:20:36 <peter1138> Yeah, and we keep missing these files, as glx tells us :)
15:20:52 <TrueBrain> one of the things we also should work for, is using more folders :P
15:20:56 <TrueBrain> 'src' really has too many files
15:21:08 <TrueBrain> silly enough, in MSVC we do use more folders
15:21:14 <TrueBrain> but ONLY MSVC does :P
15:21:18 <TrueBrain> (virtual folders, that is)
15:21:29 <peter1138> Yeah, these files should perhaps not going in src as they are genearted.
15:21:39 <TrueBrain> I meant outside the generated files
15:21:43 <peter1138> Also there's an odd one in there which is autogenerated, but not completely.
15:22:02 <TrueBrain> source.list has categories for MSVC projects
15:22:10 <TrueBrain> but most of the files in those categories are in 'src'
15:22:18 <TrueBrain> so .. all non-MSVC people are punished :P
15:22:28 <TrueBrain> possibly 'src' should only contain folders ..
15:24:02 <peter1138> All my C# stuff is one-file-per-class, and mostly namespaces-as-directories
15:24:18 <TrueBrain> THE SANITY! STOP IT! :P
15:24:49 <peter1138> Maybe we should use namespaces...
15:25:24 <peter1138> andythenorth, I've got a kiwi fruit. Not quite a creme egg :(
15:28:22 *** octernion has joined #openttd
15:34:12 *** JacobD88 has joined #openttd
15:44:45 <TrueBrain> nielsm: minor request, in markdown, please hit that enter key after every end-of-sentence marker; makes diffs a lot easier :) (minor thingy :P)
15:45:05 <nielsm> your change afterwards reminded me :)
15:57:12 <nielsm> so, does anyone remember where the Wentbound save originates from? :)
16:00:04 <peter1138> <peter1138>Testing with a massive savegame I found on the forum that makes my i7 struggle.
16:00:27 *** Alberth has joined #openttd
16:00:27 *** ChanServ sets mode: +o Alberth
16:03:13 *** supermop_work has joined #openttd
16:05:12 *** Wormnest has joined #openttd
16:06:40 <Alberth> not so terribly busy here atm :)
16:07:24 <Alberth> or did everyone agreed on what they should do, and are busy doing that now, perhaps
16:07:56 <andythenorth> we're all tapping our fingers waiting for nml2
16:08:02 <andythenorth> all work stopped :)
16:11:12 <LordAro> Alberth: you should review some PRs to pass the time ;)
16:14:00 <peter1138> Hmm, should I have the second lotus biscuit?
16:16:58 <Alberth> andy: I did ponder on nfo linking, in particular how to do an experiment that's not too much work
16:19:25 <peter1138> orudge, is there a way to search on upload filename on tt-f?
16:22:44 <TrueBrain> funny .. os2_m.cpp is never compiled
16:22:50 <TrueBrain> when is the last time os2 was tested? :P
16:34:23 <peter1138> os2 was orudge's thing, wasn't it?
16:41:08 <TrueBrain> it used to be that coop provided the most extreme savegames :P
16:45:20 <nielsm> well ProZone 13 is still also a test case :)
16:47:04 <TrueBrain> meh, SSE fails again with CMake .. setting properties on files is difficult :D
16:49:09 <TrueBrain> set_source_files_properties only works after you added the file to an executable
16:49:22 <TrueBrain> so .. I guess it is not marked as source file before that
16:50:54 <TrueBrain> 'Due to historical development reasons, source file properties
16:50:54 <TrueBrain> are scoped only in the directory where they are set.'
17:03:22 *** Beerbelott has joined #openttd
17:08:46 *** sla_ro|master has joined #openttd
17:08:49 <TrueBrain> glx: as a FYI, I am doing some massive work in CMake; just in case you were planning to do the same :D
17:09:09 <glx> I'm working on generate_widget :)
17:09:34 <glx> I guess you are replacing add_files with target_sources
17:09:35 <peter1138> Did you look at y branch?
17:10:17 <TrueBrain> glx: not currently, but that is a good idea yes :)
17:11:09 <peter1138> That is about all I did on that one :p
17:11:24 <glx> btw I have 2 allegro files only in SOURCE_LIST2
17:11:53 <TrueBrain> source.list was weird there
17:12:26 *** supermop_work has joined #openttd
17:14:01 <glx> that's what I have for now
17:14:10 *** supermop_work_ has joined #openttd
17:14:32 <glx> and I keed the original script_window.hpp to compare my output
17:15:15 <TrueBrain> spend 20 minutes figuring out I misspelled CFLAGS :P
17:15:19 <Samu> I created a game with newgrfs, supermop_work
17:16:47 * glx like configure_file() cmake function :)
17:18:41 *** supermop_work__ has joined #openttd
17:19:33 <glx> TrueBrain: I guess you noticed all the ugly stuff in src/os/* just to keep source list parsing output
17:20:28 <glx> mainly for headers, as source files where properly selected in source.list
17:20:49 <glx> but most headers were just added even when unneeded
17:21:54 <andythenorth> vehicle variants then?
17:22:10 <andythenorth> Horse 2 is nearly finished, so I need a reason to delay it
17:22:37 <nielsm> did you make the nuclear steam-electric loco yet?
17:22:59 <TrueBrain> glx: fixed that already :)
17:23:45 <andythenorth> I did not make nuclear electric steam engine no
17:23:47 <glx> but my first step was perfect matching
17:24:06 <andythenorth> nor did I add the parameter to scale intro dates by atrbitrary amount
17:24:16 <andythenorth> Eddi said I wasn't allowed :(
17:25:50 *** supermop_work_ has quit IRC
17:27:13 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7369: Add: k-d tree element checker function https://git.io/fhjcN
17:27:19 <peter1138> andythenorth, oh, variants... Hmm.
17:27:35 <andythenorth> think of the liveries I could draw
17:27:38 <peter1138> andythenorth, multitle docks is going well, so... after.
17:27:41 <andythenorth> I could delay Horse for years
17:27:49 <andythenorth> I love multititle docs
17:28:04 <andythenorth> is it a branch yet?
17:28:10 <peter1138> Also, should we add rgb company colours, or just add a fixed set of more (but 8bpp) colours?
17:28:23 <andythenorth> then we can look up the real liveries
17:28:26 <TrueBrain> glx: I now wrote something that makes sure we can set the sse flags locally, and that it still all works .. its a bit weird :P
17:28:34 <peter1138> RGB might be "too much choice" ?
17:29:16 <peter1138> Although it's possible to offer fixed colours which happen to be RGB.
17:29:48 *** Maarten has joined #openttd
17:29:59 *** supermop_work has joined #openttd
17:32:57 <TrueBrain> glx: and now I want to look if add_files() can have a parameter to do the filter, removing this if() debt, which is horrible to read :D
17:33:18 <andythenorth> so I've added spec_flag IND_FLAG_LONG_CARGO_TYPE_LISTS
17:33:33 <andythenorth> now I need to read different bits, and match up the cargos to the industry cargos in the economy
17:34:06 * andythenorth wonders, can we merge 16-cargos yet?
17:34:14 <andythenorth> and have some nml docs? o_O
17:34:38 <glx> TrueBrain: it's easy to add a parameter
17:34:51 *** supermop_work__ has quit IRC
17:35:15 <glx> but the foreach ARGV will need some care
17:36:16 <TrueBrain> yeah, that I now figured out how that should work :)
17:37:13 <glx> cmake language is quite powerful, but sometimes need ugly solutions
17:39:08 <peter1138> andythenorth, oh yeah, need to fix that bug, eh?
17:39:37 <TrueBrain> they also have a bad case of "backwards compatibility"
17:40:07 <peter1138> I used to use CMake but got annoyed ;(
17:40:18 <peter1138> It was better than ./waf though.
17:40:27 <TrueBrain> because, for example, with target_sources you can add sources to a target; awesome
17:40:32 <TrueBrain> you cannot create a target without at least a single source
17:49:13 *** supermop_work_ has joined #openttd
17:49:27 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #7351: Add: Warn players that company passwords are not truly secure https://git.io/fhhrD
17:50:46 *** supermop_work has joined #openttd
17:52:46 <TrueBrain> so video/CMakeLists.txt is somewhat more readable :P
17:52:53 <LordAro> purely for interest purposes, i looked up the first reference to "wentbourne" in my logs:
17:53:31 <LordAro> [21:07:08] <peter1138> Testing with a massive savegame I found on the forum that makes my i7 struggle.
17:53:35 <TrueBrain> glx: FILTER a good word?
17:53:40 <TrueBrain> LordAro: peter1138 said so already, yes :)
17:53:57 <glx> so you make a temporary list then append it depending on filter test
17:57:15 *** supermop_work_ has quit IRC
18:14:13 <TrueBrain> w00p, this seems to work :)
18:14:22 <TrueBrain> slowly we are creeping to cleaner and cleaner code :)
18:14:32 <nielsm> can we make Window::nested_array a std::vector or something?
18:15:43 <andythenorth> is this like a slow controlled bonfire? :P
18:16:06 <TrueBrain> I like bonfires too, as long as they are not inside my house :D
18:18:25 <TrueBrain> lets see what I broke ...
18:18:25 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
18:22:36 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #7351: Add: Warn players that company passwords are not truly secure https://git.io/fhhrD
18:24:05 <TrueBrain> ha, it works :D Sweet :)
18:25:46 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7351: Add: Warn players that company passwords are not truly secure https://git.io/fhjW8
18:26:01 <TrueBrain> except that it doesnt ..
18:26:08 <TrueBrain> I selected too many files for OSX :D
18:26:26 <TrueBrain> yeah, okay, I already was a bit: huh? .. but so that is correct :P
18:30:16 <LordAro> nielsm: get the smallvector removal merged first :p
18:30:27 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
18:32:08 <Alberth> niels: not much use, it never changes after initial construction
18:32:56 <Alberth> TB just deleted all source code :p
18:33:34 *** Progman has joined #openttd
18:36:43 *** supermop_work has joined #openttd
18:37:58 *** supermop_work_ has joined #openttd
18:40:58 <glx> pff stupid cmake regex failures
18:41:21 <glx> string sub-command REGEX, mode MATCH regex "" matched an empty string.
18:41:34 <planetmaker> rm -rf / path/to/somewhere ?
18:41:37 <glx> stuff like that is allowed in any other language
18:41:47 *** HerzogDeXtEr has joined #openttd
18:42:33 <LordAro> planetmaker: rm -rf /home/$USRE/.myfiles more like
18:43:07 <planetmaker> I left the space in the path deliberately :P
18:43:30 <LordAro> i think Steam's was rm -rf ~/$SOMEVARIBEL
18:43:41 <planetmaker> yeah, I think so, too
18:45:07 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhjWb
18:45:17 <Alberth> he just updated the PR ".. (and removing all other project-related code)"
18:48:21 <LordAro> pfft, who needed that anyway?
19:06:49 *** J0anJosep has joined #openttd
19:06:50 <peter1138> Yeah, the std::vector PR needs to be updated and then reviewed.
19:08:32 <J0anJosep> How is the multidock PR going?
19:09:10 <J0anJosep> I had done some work on it (multidock, ship separation and others). Maybe I can help at some point.
19:09:30 <peter1138> I looked at that in the past but got kinda lost.
19:10:15 <J0anJosep> I can create a separate branch to check it.
19:11:10 <J0anJosep> But I still have to rebase all my previous work to your catchment improvement.
19:11:13 <peter1138> Basically in all my previous attempts I was stumbling on how to efficiently get the docking tiles, without making them "physical" station tiles.
19:11:33 <peter1138> Doing that would be awkward for savegame compatibility.
19:11:38 <J0anJosep> Why not making them physical station tiles?
19:12:00 <peter1138> Because there are plenty of cases where there's no space in an existing game to put a station tile there.
19:12:58 *** frosch123 has joined #openttd
19:13:00 <J0anJosep> If you make the tile on the water the tile where ships dock, that is solved. You can then tell specifically which tracks can be used for docking, which prevents weird 45º "docking".
19:14:54 <peter1138> Okay that would work, but I'm not sure. It effectively force-breaks every save.
19:15:56 <peter1138> Because existing docks get changed.
19:17:03 <J0anJosep> You can update them. The only thing is new docks will have a visible tile and and invisible tile, insetad of two visible tiles (the one in a sloped tile and the one in the water.)
19:18:12 <peter1138> Yeah, that "only thing" is what I mean, it's quite a change.
19:22:59 <J0anJosep> I also know this is quite a change, but is doable. Savegame compatibility isn't affected by it.
19:23:35 <J0anJosep> And it also allows for docks in 1-tile wide rivers, without having to terraform the river.
19:24:01 <peter1138> I intend to allow all that, but just keep the existing docks as they are.
19:24:11 <peter1138> Well, appearance-wise, anyway.
19:26:39 <J0anJosep> Ok. Other types of docks? How will they be?
19:27:04 <peter1138> However a NewGRF defines them.
19:29:11 <peter1138> I'm torn, because the method I've used works.
19:29:36 <peter1138> If it was guaranteed to always have the space I'd just plonk a docking tile down on the end :/
19:30:56 <J0anJosep> What if there are three different docks with the same docking tile?
19:31:40 <peter1138> Yeah that's also a problem.
19:31:49 <peter1138> Not a problem with your method or my current method.
19:33:28 *** m3henry has joined #openttd
19:33:49 <J0anJosep> Yes. I chose to do it that way because I thought about separating ships as well.
19:34:05 *** andythenorth has joined #openttd
19:34:28 <peter1138> Not show how separating ships is relevant.
19:35:17 <J0anJosep> I also thought about doing it adding a new docking tile, but then three different competitors could share the same docking tile.
19:35:52 <andythenorth> ^^ does the OP there just not know about pressing 'escape'
19:36:36 <m3henry> tram > bendy bus > plane > bigger plane > EMU > HST > DMU > 3rd rail EMU. I just need to take a boat and helicopter today and I'll be a transport tycoon
19:36:49 <peter1138> My docking tile is 'virtual'
19:37:36 <peter1138> It's 1 bit in the map array that says signals ships and the PF to do additional stuff.
19:38:54 <peter1138> The bit available on any water tile, and also station tiles (for buoys) & bridge heads (aqueducts).
19:39:03 <J0anJosep> Ok. I was thinking if only 1 ship could be in a tile, sharing that tile could be troublesome.
19:41:39 <peter1138> Effectively we'd break backwards compatiblity by making the existing water part the docking tile.
19:41:44 *** m3henry has joined #openttd
19:41:59 <peter1138> But it solves lots of issues in the long run.
19:42:13 <J0anJosep> By the way, your idea looks ok.
19:42:16 <peter1138> J0anJosep, backwards compatibility means it stays the same.
19:42:46 <J0anJosep> I'm lost know. My way makes it breaks backwards compatibility?
19:43:04 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhjlF
19:43:05 <peter1138> Yes, because it changes the visual representation.
19:43:16 <peter1138> I'm not talking about technical compatibility.
19:43:41 <peter1138> Obviously old saves still load, but the result is differen.t
19:43:53 <m3henry> Are there any save versions which can't be loaded by 1.9.0?
19:44:04 <peter1138> On the other hand, it's appealling because it solves more issues
19:44:15 <peter1138> m3henry, anything from a patchpack :p
19:44:44 <peter1138> Any save from an official version should be loadable.
19:44:45 <m3henry> well yeah, but you know that's not what I meant
19:44:51 <peter1138> If not, there's a bug.
19:45:00 *** supermop_work_ has quit IRC
19:45:04 <peter1138> And it's not that improbable, savegame conversion is ... fun.
19:46:15 <m3henry> :3 At what point will maintaining complete backwards compat become too large a burden?
19:47:13 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
19:47:39 <peter1138> Usually you only need to take care about converting from the previous to the current version.
19:47:49 <peter1138> Everything before that is handled by the code that's already there.
19:48:27 <peter1138> But it can get awkward when things are shuffled around. We basically do intermediate conversions.
19:50:42 <peter1138> J0anJosep, with my method, when I encounter a tile marked as a docking tile, I have to check all 4 adjacent tiles.
19:51:13 <peter1138> J0anJosep, it's not slow but it's clearly more complex than just checking an actual station tile.
19:51:30 <peter1138> J0anJosep, there's also the extra effort needed to maintain the docking tile flag.
19:52:49 <J0anJosep> It's not that hard. I discarded that way because it was too complex at the moment I tried it (years ago, few coding experience in OpenTTD).
19:53:27 <nielsm> compare bits 16..23 to cargotype("COAL") etc
19:53:32 <peter1138> J0anJosep, simply, making it an actual station tile is superior, apart from having to wipe out the existing water part of the dock.
19:53:48 *** m3henry1 has joined #openttd
19:54:00 <J0anJosep> peter1138, if it is more lines of coding but the result is better, go with it.
19:54:08 <andythenorth> think I'll need to wrap that in an economy switch
19:54:20 <peter1138> J0anJosep, i'm leaning to deferring to the community :p
19:54:49 <peter1138> J0anJosep, did you ever do anything with custom graphics or not get that far?
19:57:56 <andythenorth> wonder if I can just set() all the accepted cargos across all economies
19:58:14 <J0anJosep> peter1138, I didn't go that far. But there are 13+16 free bits (even after adding dock tracks, reserved tracks, the edges and the corners of river banks, and trackdir preferences on the map array). I think it should be no problem.
19:58:33 <peter1138> Free bits wasn't an issue I am worried about :)
19:59:07 <J0anJosep> Would be any added problem then? I don't think so.
19:59:27 <peter1138> No problem, just wondering if you'd done anything on it.
19:59:42 <peter1138> If not, that's good because there's no thing to throw away :)
20:00:15 <J0anJosep> Did you do it already?
20:00:24 <peter1138> No, not got that far yet, but I have plans.
20:02:45 <andythenorth> hmm this is going to be slow to compile
20:03:54 <peter1138> Right, I'm leaving it for now.
20:04:01 <peter1138> #7364 needs attention.
20:04:16 <peter1138> I'm concerned about adding flags to say a callback result is invalid.
20:04:22 * andythenorth wonders about an nml linter
20:04:38 <andythenorth> it's tedious waiting 40s for a syntax error
20:08:36 <peter1138> Hmm, if I disable the group, will it still load, or throw a different error.
20:10:34 <peter1138> dbg: [grf] [d/firs.grf:3341] GetGroupFromGroupID(0xF6:0x89): Groupid 0x00F6 does not exist, leaving empty
20:10:39 <peter1138> Hmm, I guess that still works.
20:11:13 <peter1138> Means you can't have diagnostics.
20:11:33 <andythenorth> do we have some problems with windows not being marked dirty?
20:11:46 <peter1138> andythenorth, be more specific please.
20:11:47 <andythenorth> industry window notably, and newgrf industry debug
20:11:59 <andythenorth> I've also got station signs missing parts
20:12:12 <andythenorth> I'll see if it's reproducible
20:12:25 <peter1138> Station signs missing parts would be something for nielsm to look into.
20:12:39 <peter1138> I never use the newgrf debug windows so ... no idea.
20:14:41 <andythenorth> also industry text failing to update / misdrawn, but harder to repro
20:14:42 *** Thedarkb-T60 has joined #openttd
20:15:50 <andythenorth> just checking my branch to see if it's something weird locally
20:16:29 <andythenorth> I was messing around with SDL PR, and blitters
20:17:09 <andythenorth> classic "I can't just post a savegame"
20:17:19 <andythenorth> due to unreleased + broken newgrfs
20:20:29 <andythenorth> is there a TL;DR spec in bullet points? o_O
20:20:47 <andythenorth> in principle offering GS control of zones is potentially interesting
20:20:48 <nielsm> but read zone_type.h it should show most ideas
20:20:59 <nielsm> yeah GS control is one goal
20:21:35 <andythenorth> I have long wanted GS based on opening up a map
20:21:43 <andythenorth> (removing restrictions as reward for completing goals)
20:21:49 <nielsm> railroad tycoon 2 allowed some potentially interesting gameplay with territories granting you permissions after achieving goals/hitting marks
20:21:59 <andythenorth> the problem I forsee
20:22:07 <andythenorth> is that realistically I will never ever write a GS
20:22:14 <andythenorth> and not many other people do
20:22:22 <nielsm> ow, late, need to get food before everything closes
20:22:23 <andythenorth> otherwise it's great :P
20:22:43 <nielsm> there should also be tools for setting it up in scenario editor without GS :)
20:23:07 <nielsm> so you could do dumb things like make a map where each company has an island to themselves and can't interfere
20:23:24 <andythenorth> I can't repro this dirty redraw bug (signs screenshot)
20:23:43 <andythenorth> I tried things like SDL recently, maybe I had a bad configure or something
20:24:30 * andythenorth deletes the screenshot
20:29:46 <nielsm> hmm, I guess "bedrock mountain" might not be a great zone after all? since it's not quite exclusive with being a territory
20:30:46 <nielsm> unless maybe restricting towns from building on it at all?
20:31:49 *** gelignite has joined #openttd
21:02:49 *** octernion has joined #openttd
21:13:09 <peter1138> Should I "break" existing savegames by turning the water-part of docks into the docking tile?>
21:15:11 <peter1138> Well I don't think we've ever done a change like that before
21:15:43 <andythenorth> major version API change
21:15:50 <andythenorth> save up a few other things to go with it
21:15:59 <nielsm> if some saves may fail to load that's def. version 2.0
21:16:03 <andythenorth> or make a savegame migration tool
21:16:17 <peter1138> No, savegames won't fail to load.
21:16:27 <andythenorth> it just changes behaviour, yes?
21:16:27 <peter1138> And no, fuck waiting around for a mythical 2.0.0
21:16:34 *** m3henry has joined #openttd
21:16:58 <peter1138> Hmm, I might simplify this PR to make it backportable.
21:16:59 <andythenorth> 5th November 2019
21:17:01 <nielsm> so if the docking tile is blocked by something else, does the dock then disappear or what happens?
21:17:11 <andythenorth> 2.0.0, bonfire party!
21:18:06 <peter1138> nielsm, the part of a dock that is on water will turn into effectively a water tile (but it's a station tile)
21:18:13 <peter1138> It can't be blocked.
21:18:34 <andythenorth> TrueBrain: not yours?
21:18:48 <TrueBrain> a bonfire in nov? Are you crazy?
21:18:54 <TrueBrain> I was assuming it was indoors at your place :)
21:19:06 <andythenorth> we are celebrating an English hero that day
21:19:21 <andythenorth> a man who was legally killed for trying to blow up our democracy
21:19:23 <TrueBrain> nobody cares about that :P
21:19:51 <andythenorth> we do, much fireworks
21:19:57 <peter1138> Okay, I had my fish.
21:19:59 <TrueBrain> 2 months later again fireworks
21:20:04 <peter1138> What's for dessert?
21:20:05 <TrueBrain> you love your fireworks
21:20:12 <TrueBrain> peter1138: a bonfire?
21:24:18 *** supermop_work has joined #openttd
21:24:32 <LordAro> smoked fish is very nice
21:24:33 <andythenorth> flames are so entrancing
21:24:38 <andythenorth> as is deleting stuff
21:24:40 <nielsm> I remembered I have ice cream in the freezer
21:25:08 *** supermop_work_ has joined #openttd
21:25:16 <andythenorth> so is that nml 16-cargo PR done?
21:25:58 <andythenorth> FIRS exercises some of the spec there
21:26:22 <andythenorth> accept / produce props, produce block, cb 37, new flag, and new tile flag
21:26:32 <andythenorth> everything I tried so far works now
21:26:44 <andythenorth> 'works' = nml compiles it
21:27:04 <andythenorth> and with #7364 it works in OpenTTD as expected
21:27:23 <andythenorth> not saying it's 'done done' and bug free, but we have .z versions for bug fixes eh?
21:27:37 <andythenorth> nml needs to get out there, it will only find bugs by authors using it
21:28:21 <m3henry> These merge conflicts are rather confusing
21:28:24 <andythenorth> I can't keep up with who has rebased what
21:29:12 <nielsm> I think I may have had a dream about some kind of nml2 syntax recently, and I'm not sure that is healthy
21:29:26 <nielsm> but I don't remember any specifics
21:31:24 <DorpsGek_II> [OpenTTD/nml] andythenorth commented on pull request #15: Industries: support 16 cargos in / 16 cargos out https://git.io/fhjBi
21:31:47 <andythenorth> planetmaker: ^ I'm proposing that we approve that, ship it, and wait for bug reports
21:32:03 <andythenorth> the error handling of the old prod. cb formats is poor
21:32:14 <andythenorth> that could be fixed in a future release
21:32:23 <andythenorth> nml needs to scorch the earth a bit, and just get into use
21:35:39 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7364: Fix e66cec8f86: Permit loading of industry production callback with invalid cargo type. https://git.io/fhhbB
21:38:25 <peter1138> So... ditch half of my multiple docks patch and go with butchering the existing docks?
21:38:49 <peter1138> Is that too much change, aesthetically?
21:39:39 <andythenorth> what gets the most interesting result?
21:40:05 <m3henry> Oh I get it, st->industries_near has been changed to std::vector by someone else right?
21:40:50 <peter1138> m3henry, it's a std::set now
21:41:17 <peter1138> StationList as well.
21:42:15 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhjRs
21:43:22 <andythenorth> it's my PR, so I shouldn't
21:43:40 <nielsm> should it be squashed or used as-is?
21:43:43 <peter1138> I needs to be squashed.
21:43:52 <peter1138> Well, you could do a squash commit, I suppose.
21:44:06 <andythenorth> there's a button right there in GH :P
21:44:52 <peter1138> We definitely don't want that kind of history in repos :p
21:44:59 <andythenorth> certainly, I would place a small bet that nobody else is going to test that PR :)
21:45:18 <andythenorth> we have a handful of nml industry sets, and not many of those authors can install nml from a PR
21:45:21 <peter1138> 1 commit may be too little.
21:45:30 <peter1138> Should I manually squash bits?
21:46:02 <peter1138> Hmm, no, I'm busy :p
21:48:59 <andythenorth> just squash it to one, move on?
21:57:54 <LordAro> so what's wrong with 2 tile docks?
21:58:01 <LordAro> i've not been paying attention
22:01:05 <peter1138> Nothing as such, but you can't reliably convert them to be 3 tile docks (3rd tile as an actual station tile but just water)
22:01:18 <peter1138> This is a 2 tile dock still, just the second part is water.
22:01:36 <LordAro> why do we want 3 tile docks?
22:02:06 *** supermop_work_ has quit IRC
22:02:07 <peter1138> By making the 3rd tile an actual station tile, it simplifies a lot.
22:02:54 <LordAro> how aren't they already effectively 3 tiles long then?
22:02:55 <peter1138> That image handily illustrates why the docking tile can't be converted.
22:03:09 <peter1138> the tile at the end is not part of the station
22:03:19 <peter1138> It is where the ship loads, though.
22:03:32 <peter1138> Basically all vehicles stop ON a station tile to do loading
22:04:10 <peter1138> I'm of the opinion that changing the way docks work like this is unacceptable.
22:04:52 <peter1138> Er, "like this" ... i specifically mean replacing the 2nd part with something that looks like water.
22:05:36 <peter1138> This is what I have working in my branch.
22:05:59 <peter1138> the red square is just debugging
22:06:14 <LordAro> but that image is post-samu patch, isn't it? did that get merged?
22:06:18 <peter1138> Each red square is a regular water tile with a special flag in the map array that says it is a docking tile.
22:06:54 <peter1138> LordAro, it illustrates the current possibilities of docks.
22:07:21 <peter1138> samu's patch was about the ability to place docks in those locations, which is not possible, but you can terraform or place buoys afterwards.
22:07:38 <peter1138> But samu's patch is irrelevant here, just a handy picture :)
22:07:55 <peter1138> So with the "red square method"...
22:08:28 <peter1138> There's a bit in the map array, and if a ship/pf encounters it, it has to look at the 4 adjacent tiles to see what to do.
22:08:48 <peter1138> Kinda ugly, but does work. And doesn't "break" docks like in docks3.png
22:09:29 <andythenorth> replace everything with docking tiles, build objects for the quays and stuff?
22:09:35 <andythenorth> or rail station tiles?
22:09:46 <andythenorth> marico and friends?
22:09:50 <peter1138> LordAro, and andythenorth misses the point :)
22:10:41 <peter1138> andythenorth, and no, a dock tile has to be a dock tile, not a "newobject"
22:11:00 <andythenorth> the tile would be a dock tile
22:14:34 <peter1138> Yeah, well, part of this plan is use-newobjects-style graphics chain, but with station variables.
22:14:40 <peter1138> Might need some duplication.
22:14:56 <peter1138> Things like platforms etc are irrelevant, cos it's a rail station.
22:20:07 <andythenorth> oof, so is anyone ever going to write a GS? o_O
22:20:38 <andythenorth> it's been about 8 years :)
22:22:29 <peter1138> There's lots of them?
22:23:57 <peter1138> LordAro, so yeah. "removing" the part on water is a no go, right?
22:25:17 <LordAro> peter1138: i wouldn't like it, anyway :p
22:25:38 <peter1138> I better edit landscape_grid.html then :)
22:28:15 <peter1138> Oh, and of course, I think it will be possible to make dock tiles which a docking tiles in the future.
22:28:22 <peter1138> Just not for the existing docks.
22:29:03 <LordAro> just not necessarily for existing docks
22:29:36 <peter1138> Well, awkwardly, going that route means this map bit only needed for old maps. Hmmmmm...
22:29:44 <peter1138> Maybe I should come up with a different way.
22:30:01 <LordAro> how awful would it be for old style docks (with no position in front of them) to use the 2nd dock tile instead?
22:30:34 <peter1138> Didn't we agree that it's not good?
22:30:48 <LordAro> i mean just clip through it
22:31:18 <LordAro> but the number of docks like that must be very small
22:33:19 <LordAro> the number of docks that either have had land terraformed in front of them or had a buoy placed in front of them must be vanishingly small
22:33:45 <peter1138> So convert to 2 tile dock if it can't be converted to a 3 tile dock?
22:34:03 <LordAro> i'm wildly speculating, but yeah
22:35:52 <andythenorth> there's always something :P
22:36:06 <LordAro> well station catchment has already been changed
22:37:40 <peter1138> They "just work" right now.
22:38:14 <LordAro> maybe a good time to properly remove oil rig's specialness? :p
22:47:24 <nielsm> and make industries with own road stops too?
22:47:41 <nielsm> somehow industries with own rail stations?!
22:50:35 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhSz0
22:51:43 <andythenorth> where's my industry airport screenshot :P
22:52:24 <nielsm> airport owned by a third party and you rent terminals at it
22:53:41 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhSz0
22:53:57 <andythenorth> there was a patch for arbitrary airports at industries
22:55:13 <peter1138> airports at industries?
22:56:12 <andythenorth> it was a yexo patch
22:59:58 <andythenorth> found some nfo for it :P
23:00:13 <andythenorth> I probably have a test grf in forum pms somewhere
23:00:25 <andythenorth> I think it was a wind turbine that a helicopter could land on
23:02:19 <m3henry> hopefully that was the last rebase
23:07:04 <peter1138> andythenorth, hmm, source no longer available.
23:07:49 <andythenorth> hmm, bedtime also
23:11:57 <peter1138> TrueBrain, is it possible to access the old hg.openttd.org repos somehow?
23:12:21 <peter1138> TrueBrain, or at least long enough to make a clone of it all...
23:12:52 <TrueBrain> sure. what patch do you have that is only there? :p
23:12:54 <peter1138> TrueBrain, I mean the stuff under hg.openttd.org/developers/
23:13:16 <peter1138> Dunno what it is, never seen it :p
23:13:23 <peter1138> But of course the link is no more.
23:13:40 <TrueBrain> give me a moment .. need to boot my PC and stuff
23:13:47 <peter1138> It's not that important!
23:14:12 <TrueBrain> nah. small effort :)
23:17:48 <peter1138> I need to allow bridges above station tiles :p
23:20:04 <TrueBrain> right; hg .. where are they hiding ..
23:21:47 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhj0p
23:23:19 <TrueBrain> wow .. that was 3 server-iterations ago :D
23:25:07 <TrueBrain> send you the tarball in PM
23:25:14 <TrueBrain> that is a tarball of the server-side hg repo
23:25:17 <TrueBrain> no clue what is in there, etc
23:25:40 <TrueBrain> possibly the answer to world peace
23:25:55 <TrueBrain> I assume you can figure out how to get information out of it? :)
23:26:15 <peter1138> $ hg clone airports.hg airports
23:26:29 <peter1138> Gives me a regular tree.
23:26:37 <peter1138> Of course, it's hg so makes no sense, but other than that... ;)
23:26:48 <TrueBrain> convert to git and put it on GitHub? :P
23:26:51 <peter1138> Heh, hg log doesn't page :D
23:26:57 <TrueBrain> anyway, enjoy :D If you need any other, lemme know!
23:27:11 <DorpsGek_II> [OpenTTD/OpenTTD] nikolas updated pull request #7086: Change #6173: Update SDL driver to use SDL 2.0 https://git.io/fhamZ
23:27:38 <TrueBrain> we have backups of backups of backups of old servers, so yeah :D Happy they are useful :P
23:28:14 <TrueBrain> "Subscriptions 683"; tnx GitHub .... nice to know .. I follow ... 600+ things ...
23:29:00 <TrueBrain> there was some dust involved :P
23:29:04 <LordAro> TrueBrain: mine said "1000+" and then i decided to close the tab
23:29:23 <TrueBrain> yeah ... I unwatched OpenTTD a while ago :D (well .. I watch releases)
23:29:38 <TrueBrain> I don't understand how watching is useful for these kind of repositories
23:29:47 <LordAro> mm, i should probably unwatch
23:29:53 <TrueBrain> just a notification when something is created
23:29:54 <TrueBrain> that is what I want
23:29:59 <TrueBrain> not all things happening with it from there
23:31:02 <peter1138> Hmm, I think this is just the trunk branch. Odd.
23:31:09 <TrueBrain> no branches in there?
23:32:51 <TrueBrain> no, there are merges in there
23:33:03 <TrueBrain> so I guess somewhere deep down is something of interest
23:34:28 <Eddi|zuHause> rebasing a branch with existing sync commits is somewhat of a pain
23:34:32 <TrueBrain> so yeah, without an upstream attached to this .. this is difficult to see
23:34:52 <peter1138> Yeah, it's like a "standalone" branch :/
23:34:59 <peter1138> 1 clone per feature? o_O
23:35:02 <peter1138> LordAro, is it terrible?
23:35:14 <LordAro> peter1138: better than just clipping, i guess
23:35:39 <LordAro> and as long as its only a possibility for the "legacy" docks
23:36:13 <TrueBrain> the diff is pretty big
23:38:00 <TrueBrain> I did a git clone, changed to the latest sync version, removed all the files, copied the hg clone content in there
23:38:03 <TrueBrain> and there you have it :P
23:38:09 <TrueBrain> lost the history, got the content :D
23:38:16 <LordAro> that looks suspiciously like newgrf airports :o
23:38:26 <TrueBrain> possibly because it is? :P
23:39:18 <peter1138> Wait what, we already have newgrf_airport and newgrf_airporttiles?
23:39:21 <TrueBrain> sorry for the patch extension; it was refusing to show .patch inline :P
23:39:52 <Eddi|zuHause> yeah, the only thing missing from newgrf airports is state machines
23:40:05 <peter1138> What does the current stuff do? o_O
23:40:09 *** Smedles has joined #openttd
23:40:12 <TrueBrain> right, I am off to bed :) Night all!
23:40:16 <Eddi|zuHause> OpenGFX+Airports
23:40:17 <LordAro> opengfx+ airports is a thing
23:40:27 <Eddi|zuHause> mostly, providing alternate graphics and rotating
23:40:30 <LordAro> reskins and rotations
23:41:04 <peter1138> Small Seaplane Port
23:41:11 <Eddi|zuHause> (sorry, that was from a silly scene from a comedy movie like 30 years ago)
23:41:52 <Eddi|zuHause> peter1138: yeah, airport tiles can be on water, like industry tiles
23:42:05 <Eddi|zuHause> peter1138: there's just no check to only allow seaplanes
23:42:06 <peter1138> And this has been there for 9+ years...
23:42:37 <peter1138> Can't place a seaplane port on water...
23:42:47 <Eddi|zuHause> must be on the shore, afaik
23:42:56 <peter1138> No, it must be on land.
23:43:07 <Eddi|zuHause> well, i never tried
23:43:25 <peter1138> Oh, now I'm seeing glitching station signs :/
23:44:56 <peter1138> Hmm, well, time for a debug build.
23:50:36 <peter1138> Urgh, blocked oilrig...
23:51:44 <LordAro> peter1138: can't really just remove a tile from the oilrig :p
23:52:27 <peter1138> Just remembered also that multiple docks can currently 'share' the 3rd tile.
23:52:41 <peter1138> Different stations, different owners even.
23:53:04 <peter1138> And this method fails for oilrigs yes.
23:55:41 <peter1138> With my 'first' method none of that matters.
23:56:02 <peter1138> And actually I can easily make all tiles around an oil-rig be docking tiles.
23:56:49 *** Beerbelott has left #openttd
23:57:26 <peter1138> Although that requires a bit extra effort in my "flood" routine.
continue to next day ⏵
