IRC logs for #openttd on OFTC at 2021-01-02
⏴ go to previous day
00:10:18 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on issue #8295: Game crashes trying to sell huge wagon chain https://git.io/JJ9zk
00:13:52 <DorpsGek> [OpenTTD/OpenTTD] michicc opened pull request #8476: Codechange: [OSX] Prevent the compiler from using SSE4 instructions unless we want to. https://git.io/JL5q4
00:16:56 <TrueBrain> NO_SSE4_FOUND ... that name is just a mindfuck
00:21:50 <supermop_Home> can houses use 2cc?
00:24:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8476: Codechange: [OSX] Prevent the compiler from using SSE4 instructions unless we want to. https://git.io/JL5qp
00:25:04 <michi_cc> TrueBrain: I'm just follow the scheme like the Intel section above (LIFETIME_DSE_FOUND) :p
00:25:14 <TrueBrain> I know :) It still is a shitty name :P
00:25:16 <TrueBrain> but what can you do :D
00:26:30 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8476: Codechange: [OSX] Prevent the compiler from using SSE4 instructions unless we want to. https://git.io/JL5q4
00:26:48 <TrueBrain> there is another MacOS bug report that has at least 3 different crashes reported in it :P
00:27:10 <TrueBrain> after next nightly, I am going to close that and ask them to create new tickets if they still have crashes .. nearly impossible to figure this out now :(
00:27:19 <TrueBrain> just because it also crashes for you, doesn't make it the same crash :D
00:33:17 *** Gustavo6046 has joined #openttd
00:44:43 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8477: Fix: prevent clients from making emergency saves twice if server disconnects https://git.io/JL5mn
00:45:16 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8477: Fix: prevent clients making emergency saves twice if server disconnects https://git.io/JL5mn
00:49:59 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8477: Fix: prevent clients making emergency saves twice if server disconnects https://git.io/JL5mn
00:50:00 <TrueBrain> ugh, English is hard ... :P
00:52:34 *** Gustavo6046_ has joined #openttd
00:52:51 *** Gustavo6046 has quit IRC (Read error: Connection reset by peer)
00:52:51 *** Gustavo6046_ is now known as Gustavo6046
00:58:08 *** Gustavo6046 has quit IRC (Read error: Connection reset by peer)
00:58:18 *** Gustavo6046 has joined #openttd
01:26:04 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
01:26:15 *** Progman has joined #openttd
01:27:07 *** WormnestAndroid has joined #openttd
01:36:47 *** WormnestAndroid has quit IRC (Remote host closed the connection)
01:39:23 *** WormnestAndroid has joined #openttd
01:40:15 *** DasPoseidon has quit IRC (Quit: DasPoseidon)
01:49:11 *** Progman has quit IRC (Remote host closed the connection)
01:51:31 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
01:51:35 *** WormnestAndroid has joined #openttd
02:07:01 *** WormnestAndroid has quit IRC (Remote host closed the connection)
02:08:40 *** WormnestAndroid has joined #openttd
02:24:54 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
02:25:15 *** WormnestAndroid has joined #openttd
02:25:18 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:25:44 *** WormnestAndroid has joined #openttd
02:38:19 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
02:38:31 *** WormnestAndroid has joined #openttd
03:09:52 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
03:17:12 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:18:18 *** WormnestAndroid has joined #openttd
03:25:28 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
03:38:38 *** ekangmonyet has quit IRC (Ping timeout: 480 seconds)
03:38:44 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
03:39:14 *** WormnestAndroid has joined #openttd
04:07:51 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
04:08:35 *** WormnestAndroid has joined #openttd
05:34:30 *** HerzogDeXtEr has joined #openttd
06:02:26 *** snail_UES_ has quit IRC (Quit: snail_UES_)
06:43:45 <Flygon> Playing OpenTTD on a 1440p monitor feels weird as heck.
06:43:58 <Flygon> Having to play on 2x zoom for the graphics to look the correct 'size'.
06:50:50 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8477: Fix: prevent clients making emergency saves twice if server disconnects https://git.io/JL5WS
07:25:17 *** sla_ro|master has joined #openttd
07:27:23 *** WormnestAndroid has quit IRC (Remote host closed the connection)
07:27:36 *** WormnestAndroid has joined #openttd
07:27:58 *** andythenorth has joined #openttd
07:46:23 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8477: Fix: prevent clients making emergency saves twice if server disconnects https://git.io/JL5ld
10:29:09 *** Progman has joined #openttd
10:53:32 *** jottyfan has joined #openttd
10:58:56 *** jottyfan has quit IRC (Quit: jottyfan)
11:11:50 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
11:48:37 *** azulcosta has joined #openttd
11:48:48 *** DasPoseidon has joined #openttd
11:49:27 <azulcosta> morning to all. good year
11:50:04 <azulcosta> i don't know if someone answered my question 12h ago...
11:51:47 *** jottyfan has joined #openttd
11:52:43 <azulcosta> my problem, is 4 out of 11 buses, in sprites 1 and 5 (vertical sprites), only show the masked colors ok.... the original roof and windows and all original 32bpp sprite bus are invisible only in these 2 vertical sprites... i'm using windows ottd palette...
11:53:05 <azulcosta> look bus at right corner
11:55:40 <azulcosta> befor the masked colors, the buses showed up ok....
12:03:13 *** jottyfan has quit IRC (Quit: jottyfan)
12:19:08 <andythenorth> Timberwolf the 32bpp Foden is on the left?
12:20:12 <azulcosta> are you talking to me ?
12:21:45 <Timberwolf> andythenorth: yes :)
12:21:56 <andythenorth> yeah the 8bpp version is a better game sprite
12:22:21 <andythenorth> lot of 32bpp stuff looks muddy to me
12:22:49 <DorpsGek> [OpenTTD/OpenGFX] Mailaender commented on issue #47: nmlc ERROR: Image file "sprites/png/infrastructure/locks_normal.gimp.png": cannot identify image file https://git.io/JL7wG
12:25:52 <Timberwolf> azulcosta: could you gist your code (particularly the templates and alternative sprites) and upload the 32bpp and mask files somewhere?
12:27:16 <azulcosta> the masks aren't uploaded yet on github...
12:27:47 <azulcosta> the buses pack are the already availabe for download in game... i have a video presentation (32bpp with 4x zoom), buses from 1915-2010
12:28:15 <azulcosta> i'll make a zip right away, of the 4 problem buses
12:28:33 <azulcosta> can i send it here ?
12:29:28 <Timberwolf> make a branch on github with the source, I'll see if there's anything obvious I can see from my 32bpp days.
12:33:36 <azulcosta> how can i add a branch ?... github is a bit complicated yet to me...
12:38:49 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
12:39:03 *** WormnestAndroid has joined #openttd
12:56:03 <azulcosta> woops, i think i've discovered the problem
12:56:49 <azulcosta> original vertical sprites were converted to 24bpp... a not 32
12:57:12 <Timberwolf> Oh, there we go... so yes, they'd have a weird alpha channel.
12:57:30 <azulcosta> i must make more tests now
12:59:16 <azulcosta> it's weird, because i can't save png to 32bpp ? only 16/24 ?
13:00:49 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JIWz8
13:02:54 <azulcosta> timberwolf: i'll re-save all problem graphics in 32bpp now... paintshop pro doesnt have 32, but photoshop does... i'll re-export and re-save and test again.... thanks for the time helping me... cheers to all
13:03:34 <Timberwolf> np... yeah, I've only ever used 32-bit ARGB graphics files.
13:03:54 <Timberwolf> I didn't really help, I just downloaded the files while you fixed it yourself :)
13:07:42 <LordAro> Timberwolf: TimberRubberDuck
13:10:58 *** azulcosta has quit IRC (Ping timeout: 480 seconds)
13:38:15 <michi_cc> Timberwolf: If just seen you nice Road vehicle preview image, and I noticed a possible improvement for the roads in it.
13:38:57 <michi_cc> The roads, especially the pure asphalt one, look a lot like pasted or floating above the grass and not really embedded in the landscape.
13:39:42 <michi_cc> If one carefully looks at the original road and rail graphics, one notices that they all have a tiny border (like 1-2 pixels) of slightly varied greens and yellows along.
13:40:25 <michi_cc> Having just a single pixel border of darker greenish colours might make your roads much better at meshing with the grass below.
13:57:16 <Timberwolf> That's an interesting idea, I should have a try of that at some point.
13:58:03 <Timberwolf> Finding a green that GoRender won't mess with in at least one orientation may be a challenge, I could end up adding some new lighting/shading options.
13:58:33 <Timberwolf> Also needs snow/town awareness so you don't get random green next to town pavements.
14:04:59 <michi_cc> Yeah, you'd need to provide different version for that indeed.
14:09:10 <andythenorth> post process composition
14:09:24 <andythenorth> PIL is really easy, probably ImageMagick also the same
14:42:52 <Samu> i just discovered towns can terraform land to sea level
14:43:00 <Samu> have ocean water fill in
14:43:50 <Samu> unsure if the water destroys roads or houses, have to check it out
14:46:33 <supermop_Home> speaking of grass borders.. do you have to make separate groundsprites for houses for each climate to get the correct grass to show through?
14:48:37 <Eddi|zuHause> i think you can make the default grass show through, which will automatically change to snow, etc.
14:49:13 <supermop_Home> just by leaving parts of the groundsprite transparent?
14:57:03 <Samu> i have a vague that in the ideal scenario, towns terraforming to sea level may end up flooding industries build on height 0
14:57:28 <Samu> this is interesting, let's see if i can detect that
14:57:31 <supermop_Home> are industries regularly built at height 0?
14:57:47 <Samu> i don't mean oil rigs in this case
14:58:19 <Eddi|zuHause> other industries can as well, if you terraform the right way
15:00:20 <Samu> town expand command in the scenario editor builds multiple times
15:00:53 <Samu> and the result may become flooded roads and houses
15:01:03 <Samu> i wonder if this happens in a real game
15:02:36 <supermop_Home> not really enough wood colors... out of a range of 8, using 2-6 for SE and 3-7 for SW
15:02:41 <Samu> just detected a house being flooded during world generation
15:03:03 <supermop_Home> sounds very biblical
15:06:48 <Samu> a road was flooded during the "run tile loop"
15:31:01 <supermop_Home> andythenorth is there an easy way to recolor from the 58-5f greens to the 68-6f browns?
15:31:14 <supermop_Home> or should i just make 2 sets or sprites?
15:31:42 <Samu> a road and a house were flooded
15:32:32 <andythenorth> supermop_Home there's no recolouring set up in FIRS, I'd just make 2 sets and paint bucket I think :)
15:32:54 <Eddi|zuHause> supermop_Home: theoretically that is super easy
15:33:31 <Eddi|zuHause> you define a recolour map for 58->68, 59->69 etc.
15:33:48 <Eddi|zuHause> and in the recolour callback, you return that new recolour map
15:34:03 <Eddi|zuHause> sprinkle in a bit of old fashioned GRM
15:35:06 <Eddi|zuHause> if you use sprite stacking, each sprite can have its own recolour map
15:38:27 <andythenorth> not so much in ogfx
15:40:17 <Samu> a terraform command was executed at that tile
15:46:13 <Samu> so far the terraform hasn't exposed a road or house
15:47:03 <Samu> well, it did during world gen, but after that it hasn't yet
15:56:41 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JL561
15:58:16 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JL56A
15:59:06 <TrueBrain> pfft, LordAro points to a PR which fails CI .. not sure :D :P :D
15:59:49 *** Speedy` has quit IRC (Remote host closed the connection)
16:00:00 <TrueBrain> always this stupid MacOS :D
16:10:18 *** Speedy` has joined #openttd
16:19:10 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
16:41:12 <LordAro> TrueBrain: it hadn't failed when i pointed to it :p
16:45:00 *** jottyfan has joined #openttd
16:48:35 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JL5Xp
16:50:29 *** snail_UES_ has joined #openttd
16:52:53 *** DasPoseidon has quit IRC (Remote host closed the connection)
16:58:13 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JL51a
16:58:57 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JIWz8
17:01:17 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JIWz8
17:16:26 *** J0anJosep has joined #openttd
17:22:01 <J0anJosep> Hi! I have a branch in my repo which I would like to have binaries to distribute.
17:23:16 <J0anJosep> Is it possible to create them from the current infrastructure of the OpenTTD organization?
17:23:52 <FLHerne> TrueBrain would be the person to ask
17:24:11 *** frosch123 has joined #openttd
17:26:56 <J0anJosep> Thanks! I will read it. I'll wait in IRC to see what TrueBrain says about it too.
17:29:25 <FLHerne> J0anJosep: Basically, if you have a GitHub fork of OTTD and you do <something> to trigger that 'release' workflow, GitHub should compile a set of binaries for you
17:31:32 <J0anJosep> FLHerne: I did it some time ago, but not sure about how to do it now.
17:41:23 *** DasPoseidon has joined #openttd
17:41:34 *** DasPoseidon has quit IRC ()
17:48:03 <DorpsGek> [OpenTTD/OpenTTD] LordAro opened pull request #8478: Add: [CMake] Explicitly support txz for Arch Linux https://git.io/JL5DX
17:54:25 <frosch123> hmm, are discussions supposed to stay there forever? it worked fine when there were only 3, but now :p i guess "beta"
17:55:00 <LordAro> frosch123: good yuletide period?
17:55:56 *** Wormnest has joined #openttd
17:56:34 <frosch123> i am too far south for that
18:04:16 <michi_cc> ^^^ might incidentally fix the recently reported SetPortAlphaOpaque problem.
18:05:07 <michi_cc> frosch123: I think you are supposed to give some mustard to PR#7000 whenever convenient :)
18:17:15 *** DasPoseidon has joined #openttd
18:20:47 *** Progman has quit IRC (Remote host closed the connection)
18:22:42 <michi_cc> LordAro: #ifdef for what?
18:23:00 <LordAro> michi_cc: for 10.10 or newer
18:23:32 <LordAro> yes, runtime requirement, not build time
18:24:08 <michi_cc> "#if (MAC_OS_X_VERSION_MIN_REQUIRED < MAC_OS_X_VERSION_10_10)" is built-time. We don't target 10.10 on the CI though.
18:24:18 <DorpsGek> - Update: Translations from eints (by translators)
18:25:23 <michi_cc> We target 10.9 by default as that is the first version with the proper c++ runtime lib.
18:25:46 *** DasPoseidon has quit IRC (Remote host closed the connection)
18:26:03 <michi_cc> Incidentally, Gestalt was deprecated in 10.9 but the replacement only introduced in 10.10, making this an awkward thing.
18:26:42 *** DasPoseidon has joined #openttd
18:27:35 <Samu> towns can't autoslope on their own roads
18:28:05 <LordAro> michi_cc: sounds like apple
18:28:11 <Samu> i mean, the terraform command where the resulting road is not autosloped
18:28:54 *** azulcosta has joined #openttd
18:36:43 <michi_cc> LordAro: Yeah, that's why I just said screw the warnings.
18:42:04 *** virtualrandomnumber has joined #openttd
18:42:21 *** virtualrandomnumber has quit IRC ()
18:44:22 <Samu> making town terraforming around town roads a bit more permissive
18:44:22 *** DasPoseidon has quit IRC (Quit: DasPoseidon)
18:48:01 <andythenorth> Samu ever considered making content? :)
18:50:06 <Samu> looks like it's a bad idea
18:50:16 <Samu> 194 cases of flooded houses
18:50:40 <Samu> ok let's not have it happen for houses then
18:53:42 <glx> michi_cc: there's still 1 warning :)
18:55:15 <TrueBrain> J0anJosep: currently, we can create binaries from PRs created .. if you want to build binaries in your own fork, that is also possible, but would require setting up some infrastructure yourself (mostly, an S3 account and some other AWS configuration)
18:55:43 <TrueBrain> I do plan on supporting building other repositories which are published inside our infrastructure, but I have not yet figured out how to do this in a clean way with regards to secrets
18:56:02 <TrueBrain> basically, GitHub Actions needs some secrets to know where to publish binaries and with what credentials ..
18:56:30 <TrueBrain> so it depends a bit on what that branch is and what you intend to do with it
18:59:01 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8478: Add: [CMake] Explicitly support txz for Arch Linux https://git.io/JL595
19:00:22 <TrueBrain> LordAro: Arch has no package manager or something?
19:00:25 <snail_UES_> can I use the “git” command from the Mac OS Terminal?
19:00:41 <snail_UES_> I’m trying to run it, but the command doesn’t work
19:00:46 <LordAro> TrueBrain: of course, it just uses tar.xz :p
19:01:44 <J0anJosep> TrueBrain: It is a branch in my repository: MultitileDepots. I don't have time to play with it and check it. So I would like to create binaries and ask people to try the PR and get some feedback.
19:02:01 <TrueBrain> J0anJosep: open up a draft PR and we can do that for you :)
19:02:17 <snail_UES_> when I try to revert to a previous version, I get the following:
19:02:21 <TrueBrain> otherwise I don't currently have a solution
19:02:27 <snail_UES_> error: could not revert bd3a5876b0... Fix #7644: [Cocoa] Manually set colorspace to sRGB
19:02:27 <snail_UES_> hint: after resolving the conflicts, mark the corrected paths
19:02:28 <snail_UES_> hint: with 'git add <paths>' or 'git rm <paths>'
19:02:36 <J0anJosep> TrueBrain: Thank you
19:02:56 <TrueBrain> LordAro: what is that URL suppose to show me?
19:03:00 <snail_UES_> the above happens if I run “ git git revert -n bd3a5876 “
19:03:01 <LordAro> TrueBrain: actually, i've misread things
19:03:03 <TrueBrain> besides the URL being wrong :P
19:03:16 <J0anJosep> TrueBrain: By the way, today I pushed master to my repository and in triggered the CI.
19:03:31 <TrueBrain> J0anJosep: yup; that is GitHub Actions for you :)
19:03:54 <TrueBrain> snail_UES_: that happens because the revert is too old and cannot be done cleanly; means you have to manually do so
19:04:35 <TrueBrain> J0anJosep: basically, your fork is identical to our repository; everything we do, yours will do too. So pushes to master trigger a CI .. same as creating a PR to your own fork :)
19:04:54 <TrueBrain> LordAro: what kind of things did you misread? :D
19:05:17 <TrueBrain> (having 3 conversations at once, LIKE A BOSS)
19:06:14 <J0anJosep> TrueBrain: Ok, but I triggered the CI for a useless case... :P
19:06:29 <TrueBrain> you can disable the CI workflow in your fork manually, if it annoys you enough :)
19:07:27 <J0anJosep> TrueBrain: It doesn't anoy me at all, but it does something useless. Don't worry.
19:07:57 <LordAro> TrueBrain: so i'm not entirely wrong - arch uses tar.xz for packages, but "installing" is just putting things into a particular folder, which is then bundled into a tar.xz by the package creator (`makepkg`)
19:08:20 *** DasPoseidon has joined #openttd
19:08:21 <LordAro> so there's no meaningful package format for arch
19:08:38 <snail_UES_> thanks, posted it on JGR’s thread
19:09:20 <TrueBrain> LordAro: cool; so yeah, leave the comment out, or add something like "Arch uses .tar.xz for packaging" or what-ever :)
19:09:55 <LordAro> TrueBrain: thinking about it, i think it might have largely switched to .tar.zst now
19:10:04 <LordAro> but regardless, nothing worth worrying about for us
19:10:15 <LordAro> no metadata or anything
19:11:02 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JL5Hl
19:11:07 <TrueBrain> and some minor stuff there :)
19:11:16 *** Eddi|zuHause has quit IRC (Remote host closed the connection)
19:11:23 <TrueBrain> so many code in `fileio.cpp` is confusing as fuck ... :P
19:11:53 <TrueBrain> but really, calling a variable "tmp" is just stupid
19:11:58 *** Eddi|zuHause has joined #openttd
19:12:54 <TrueBrain> reading it more closely, it is just lazyness
19:13:09 <TrueBrain> "lets make a variable we reuse for many things similar but not really because I want to type 3 lettes"
19:13:46 <TrueBrain> guess it was true when it was tmp[1024]
19:13:50 <TrueBrain> now it just looks weird :D
19:14:19 <TrueBrain> so it is now twice as fast! :P
19:15:11 <TrueBrain> it will still be shitty slow, but at least slightly less slow :D
19:15:31 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8478: Add: [CMake] Explicitly support txz for Arch Linux https://git.io/JL5DX
19:16:48 <michi_cc> glx: Should be gone now :)
19:16:56 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8478: Add: [CMake] Explicitly support txz for Arch Linux https://git.io/JL5H6
19:19:14 <LordAro> TrueBrain: don't quite follow what you want me to do regarding .clear()
19:19:20 <TrueBrain> without adding remotes :D
19:20:43 <TrueBrain> it makes it more explicit, that is all
19:21:00 <TrueBrain> owh, I failed in being consistent, one sec
19:21:26 <TrueBrain> you see the same in the block below, with SP_PERSONAL_DIR
19:21:39 <TrueBrain> and lower, with SP_BINARY_DIR
19:22:13 <TrueBrain> because "tmp" is being reused like crazy like that, if someone moves blocks around, your code all of a sudden does something different
19:22:23 <TrueBrain> so I guess you could also add: else { tmp.clear(); }
19:22:28 <TrueBrain> which does the same, I guess :P
19:22:34 <LordAro> marginally redundant, but not in a meaningful way
19:23:01 <TrueBrain> that sentence .. wtf are you saying? :D :P
19:23:11 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JIWz8
19:23:35 <TrueBrain> it is just the "tmp" variable really
19:23:41 <TrueBrain> if that would be scoped better, clear() is not needed
19:24:00 <TrueBrain> you see that a bit later in that function too, where there is an else{} clearing tmp, just .. because you don't know the content :P
19:25:10 <TrueBrain> I am still not used to "return {}" statements :D
19:25:13 <TrueBrain> I love that it is valid :)
19:25:26 <LordAro> cheaper than `return "";` :)
19:25:46 <TrueBrain> but when you write too much Python ... it looks even funnier :D
19:27:02 <TrueBrain> lol @ hs.dat being written in .config/openttd/
19:27:05 <TrueBrain> that is not a config file :P
19:27:35 <TrueBrain> okay ... both the old folders and the new work for me
19:27:40 <TrueBrain> I guess that is good enough, right? :D
19:27:42 <LordAro> iirc this came up when you were on your journey of discovery last time
19:28:45 <TrueBrain> I am just happy everyone will now use XDG :)
19:28:49 <TrueBrain> instead of having 2 formats on Linux :)
19:29:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JL5Qf
19:29:47 <TrueBrain> J0anJosep: wow,that is a lot of work :o nice :)
19:34:14 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8477: Fix: prevent clients making emergency saves twice if server disconnects https://git.io/JL5Qs
19:34:59 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8477: Fix: prevent clients making emergency saves twice if server disconnects https://git.io/JL5mn
19:39:24 <J0anJosep> TrueBrain: I'm sorry I divided it into so many commits. Do you think it is necessary next time I push to a draft PR to squash it into one single commit?
19:40:27 <TrueBrain> I don't care one way or the other :)
19:40:44 <TrueBrain> I would suspect that the PR is currently too big for any review, if that is what you are after
19:40:56 <TrueBrain> but creating releases works just fine no matter how you structure your PR :)
19:41:20 <TrueBrain> one note: previews (the HTML version) are done every time you push something new. Releases (which I will give you the URL when the run is done) are only created on demand
19:41:27 <TrueBrain> so you have to poke me (or any other dev) if you want a new version to be build
19:44:11 <TrueBrain> J0anJosep: building fails because Windows fails to compile; let me know when you looked into it / fixed it, so I can build you a new release :)
19:46:12 <DorpsGek> [OpenTTD/OpenTTD] LordAro merged pull request #8357: Codechange: Drop libxdg-basedir dependency in favour of finding the directories ourselves https://git.io/JIWz8
19:47:16 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8073: Missing Savefiles - please warn if started from ~/.config/openttd/ https://git.io/Jvphb
19:47:19 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #8073: Missing Savefiles - please warn if started from ~/.config/openttd/ https://git.io/Jvphb
19:47:22 <DorpsGek> [OpenTTD/OpenTTD] LordAro merged pull request #8478: Add: [CMake] Explicitly support txz for Arch Linux https://git.io/JL5DX
19:48:16 <TrueBrain> I have the feeling the search-paths are getting there ... now for the "content_download" folder issue ...
19:48:24 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8475: Docs: indicate where to find an excellent tutorial on using CMake https://git.io/JL5QH
19:50:20 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8463: Change: Default settings improved for new players https://git.io/JL5QN
19:50:26 <DorpsGek> [OpenTTD/OpenTTD] LordAro merged pull request #8463: Change: Default settings improved for new players https://git.io/JL96K
19:51:57 *** J0anJosep has quit IRC (Quit: Konversation terminated!)
19:55:02 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #8465: Documentation: COMPILING.md does not describe how to make a non-debug build on non-Windows https://git.io/JLHTh
19:55:05 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8475: Docs: indicate where to find an excellent tutorial on using CMake https://git.io/JL79j
19:56:18 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8465: Documentation: COMPILING.md does not describe how to make a non-debug build on non-Windows https://git.io/JLHTh
19:56:34 <TrueBrain> and indeed, not being able to close discussions is annoyingggggg
19:57:04 <TrueBrain> I can lock it .. guess we should lock the settings discussion, as it came to an end?
19:57:44 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8471: Fix #8462: Don't check whether to grow a town road on water https://git.io/JL57n
19:57:54 <frosch123> yes, nothing worse than someone continueing a discussions that is done
19:58:07 *** J0anJosep has joined #openttd
19:59:02 <frosch123> "done" category is probably weird :p
19:59:48 <FLHerne> TrueBrain: Well, now it uses .tar.zst, but whatever
20:00:11 <TrueBrain> clearly someone is still processing his backlog :)
20:00:22 <TrueBrain> frosch123: yeah .. people do ask for that, a "locked" category
20:00:23 <FLHerne> LordAro: That's not quite true, pkgs also have metadata files
20:00:51 <LordAro> FLHerne: mm true, but it's not something that anything that's not makepkg is expected to create
20:01:31 <FLHerne> Now I have to read more backlog to find out what the question was :p
20:02:59 <FLHerne> Then we should probably add a CMake module for calling makepkg?
20:03:57 *** _2TallTyler has joined #openttd
20:07:44 <_2TallTyler> Timberwolf: Your post about parameters in your RV set should be required reading for new grf authors. Good stuff. I'll be referring back to it whenever I need the courage to change something big in my own sets.
20:07:56 <TrueBrain> "1,686 contributions in the last year " <- oops ... (my GitHub profile)
20:09:10 <TrueBrain> locked the settings discussion
20:09:49 <TrueBrain> so .. "content_download" .. it is a weird one .. it is always based on where-ever your "config" file is (or ".local" stuff for XDG), and it has only (and exactly) 1 place
20:09:53 <TrueBrain> it does not use normal search-paths ..
20:10:07 <TrueBrain> 2 ways to solve it: add more search-paths, or change the code that it uses the normal search-paths to find content_download content
20:10:13 <TrueBrain> the first is easy, the second is annoyingly difficult :P
20:12:24 <TrueBrain> mainly it is annoying because of how MidiFile is implemented, ironicly
20:13:36 <TrueBrain> welcome back btw frosch123 ; I am a bit slow, I know :D
20:15:27 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8457: Change: Ability to convert between road types for town roads and bridges now indi… https://git.io/JL55t
20:16:34 <TrueBrain> if I am not reading the code wrong, midi playback only works from "content_download/baseset" folder, not "baseset" .. odd
20:17:02 <TrueBrain> ah, no, it is written there
20:20:52 <TrueBrain> I got lost in that discussion :P
20:21:18 <TrueBrain> but discussions are stateless yes ... which is an annoying fact
20:21:40 <Samu> LordAro, CMD_BUILD_ROAD also ends up running CMD_LANDSCAPE_CLEAR and inherits the DC flags from the source
20:26:43 <michi_cc> Discussions really is beta for now. While it's nice that the discussions list shows how long ago the discussion started, the time of the last comment would make much more helpful.
20:26:59 <TrueBrain> michi_cc: leave it as feedback :) (if nobody has already)
20:27:14 <TrueBrain> GitHub has the tendency to listen to feedback .. shocker, I know :)
20:28:13 <frosch123> aren't they sorted by last comment added?
20:28:48 <Samu> without DC_NO_WATER, CMD_BUILD_ROAD calls CMD_LANDSCAPE_CLEAR and reaches ClearTile_Water which say it is possible to clear the water. If the command was run in DC_EXEC mode, you would see rivers being demolished and replaced with roads
20:28:49 <TrueBrain> they are sorted by last activity, it seems
20:29:02 <TrueBrain> but it doesn't tell you when this activity was :)
20:30:07 <frosch123> the tooltip shows the latest comment, so you can test your memory :)
20:30:08 <LordAro> Samu: you know my feelings on PR comments in irc
20:32:38 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8471: Fix #8462: Don't check whether to grow a town road on water https://git.io/JL55o
20:32:56 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8481: Add: always set PERSONAL_DIR "/content_download" in search path https://git.io/JL556
20:35:06 <LordAro> TrueBrain: i did not realise there was an entirely separate DetermineBasePaths implementation
20:35:35 <TrueBrain> LordAro: the question is .... did you want to know? :)
20:36:07 <DorpsGek> [OpenTTD/OpenTTD] LordAro requested changes for pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JL555
20:37:01 <TrueBrain> how do I load custom fonts in OpenTTD .. hmm ... lets find out!
20:37:59 <TrueBrain> they ... require absolute paths?
20:38:01 <TrueBrain> what is this madness?
20:39:13 <michi_cc> Not necessarily, unless you want to use something that is not known by the OS (respectivly Fontconfig).
20:39:24 <TrueBrain> I was using my own ttf, yes
20:40:00 <michi_cc> If it's not an absolute path, OTTD probably tries it as an OS font.
20:40:30 <TrueBrain> let's find out what this "probably" is, and where this is done :)
20:40:35 <TrueBrain> hard to find in the code ...
20:41:28 <michi_cc> For Linux, something something fontconfig
20:41:45 <michi_cc> Maybe fontdetection.cpp
20:41:59 <TrueBrain> fontcache.cpp, it seems :)
20:42:05 <TrueBrain> always a puzzle, if you cannot grep the setting itself :P
20:42:57 <michi_cc> Ah, fontdetection is for when just give the font name.
20:43:19 <TrueBrain> * First type to load the fontname as if it were a path. If that fails,
20:43:19 <TrueBrain> * try to resolve the filename of the font using fontconfig, where the
20:43:19 <TrueBrain> * format is 'font family name' or 'font family name, font style'.
20:43:25 <TrueBrain> it suggests it should work, a relative path
20:43:37 <frosch123> both works, path and "font name, including font modifiers"
20:44:13 <TrueBrain> absolute paths work, yes
20:44:19 <TrueBrain> relative paths seem to be from the working dir
20:44:23 <TrueBrain> not using the search path :)
20:44:45 <TrueBrain> without using our own FIOS system :)
20:44:52 <frosch123> maybe the android port bundles a font for special ottd chars
20:45:14 <frosch123> but pretty sure noone plans to put fonts on bananas :)
20:45:34 <TrueBrain> I wonder if we should use our search-paths
20:46:21 <TrueBrain> we could, I guess, by first checking if we can find the filename
20:46:37 <TrueBrain> (and in what search-path)
20:46:48 <TrueBrain> after that, hand it over to FT_New_Face
20:47:02 <TrueBrain> would be more consistent, I guess?
20:47:26 <frosch123> do modifiers like ", bold" work for paths?
20:47:38 <TrueBrain> I doubt that; a ttf mostly is either bold or not
20:47:47 <TrueBrain> the ones I have on disk are, anyway :)
20:47:58 <frosch123> oh, i thought they were collections of fonts
20:48:16 <TrueBrain> the only times I use it for webfonts, they have different files for different weights
20:58:44 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8482: Add: use our search-paths to find fonts based on relative filenames too https://git.io/JL5Fv
20:58:55 <TrueBrain> that was easier than I expected honestly
21:03:13 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep updated pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLyT1
21:03:15 <TrueBrain> frosch123: we always load "face" 0 when you use a filename
21:03:33 <TrueBrain> I think that means it won't do any ", bold" :)
21:03:49 <TrueBrain> (this is what font-config resolves, from what I gather from the code)
21:07:07 <frosch123> oh, i didn't know ottd had special code for font-by-path and font-by-name. i assumed fontconfig handled both
21:07:29 <TrueBrain> I am surprising people left and right today with my PRs fixing weird bugs :P
21:08:45 <Samu> a road and a house were flooded runing world generation "run tile loop" phase, because during town generation, terraform command is used and towns are grown multiple times, ending up building in tiles that are posteriorly flooded
21:09:14 <TrueBrain> the Windows font-loader first checks if there is a "." in the name, before it checks for file-based approach
21:09:33 <Samu> it happens very very rarely
21:09:46 <TrueBrain> and does some custom "Bold" implementation too .. marked as "poor man's way"
21:10:01 <Samu> in 12k towns, about 1 occurrence of flooding during tile loop
21:10:09 <Samu> not sure if it's worth a pr
21:10:33 <TrueBrain> okay, Windows does this font-stuff pretty different ... that is annoying :D
21:15:26 *** J0anJosep has quit IRC (Quit: Konversation terminated!)
21:15:49 <michi_cc> OSX does a third thing :)
21:16:09 <TrueBrain> I am so not liking this bug now :P
21:16:26 <TrueBrain> let me see if I can theo-craft this for Windows and OSX :D
21:16:34 <TrueBrain> (as in, I cannot test it, so .. that will be fun :P)
21:17:30 <TrueBrain> wait, no, I only have to fix _freetype
21:17:34 <TrueBrain> as that is what small_font controls :)
21:18:01 <michi_cc> Ah, no, OSX still uses FreeType. It's not using Fontconfig, though, in case you are doing something there.
21:18:46 <TrueBrain> this happens before fontconfig, or what-ever OS-based version is there, is called
21:18:54 <TrueBrain> so OSX should be good
21:18:58 <TrueBrain> it is Windows that is just ... weird
21:19:20 <michi_cc> GetFontByFaceName is what is different with OSX
21:20:16 <michi_cc> Windows is not using FreeType as the Unicode layout stuff only works with fonts loaded through the OS and not FreeType.
21:20:42 <TrueBrain> yeah ... which makes the _freetype variable a bit evil :D
21:22:24 <michi_cc> Hmm, I supposed OSX text layout will break horribly if you try a font that FreeType can read but OSX (for whatever reason) not.
21:22:46 <TrueBrain> GetFontByFaceName is never called if you use a .ttf
21:23:52 <michi_cc> But deep in CoreTextParagraphLayoutFactory, we try to load the font through the OS.
21:24:13 <TrueBrain> huh? Lol .. so FT_New_Face result is not used?
21:25:02 <TrueBrain> that would be very odd, as why would it load it at all in that case?
21:26:34 <michi_cc> No, rendering is done by FreeType, but the layout needs various tables that are contained in the font file.
21:27:14 <TrueBrain> I gave up on fonts long ago :)
21:27:20 <michi_cc> Basically, I think it is somewhat broken if you try to use a font that FreeType and OSX interpret differently.
21:28:05 <TrueBrain> guess we should ask our in-house OSX user to test it :D
21:28:07 <michi_cc> Ideally, we would let OSX render the font, too, but when I looked at it getting the pixels was heaps more complicated than on Windows.
21:30:25 <TrueBrain> hmm .. okay, I can compile under Windows
21:30:31 <TrueBrain> now ... how to set the current working dir ... hmm
21:31:11 <TrueBrain> hmm, guess I do not have to really
21:32:04 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick opened pull request #8483: Fix: Don't terraform during town generation https://git.io/JL5bM
21:32:57 <TrueBrain> how to add parameters to running in MSVC ... eeuuhhhh
21:34:19 <michi_cc> It's easy if you go the project file route and somewhere totally buried on some json file if you do "open directory".
21:34:23 <TrueBrain> the box that it should be in, is empty ...
21:35:40 <TrueBrain> yippie, found it :D
21:40:57 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8482: Add: use our search-paths to find fonts based on relative filenames too https://git.io/JL5Fv
21:40:58 <TrueBrain> now with Windows support :D
21:42:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7830: Load font from openttd config file directory, not from working directory. https://git.io/JewMG
21:42:42 <glx> in MSVC cwd is set do bin
21:44:12 <TrueBrain> in Windows I now too could load ./myfont.ttf :D
21:44:16 <Eddi|zuHause> unless you do weird stuff, windows almost always sets the cwd to the binary location
21:44:38 <TrueBrain> (where myfont.ttf is only on the search-path, not in cwd)
21:51:33 *** gelignite has joined #openttd
21:53:01 <supermop_Home> where do you want these pngs uploaded andythenorth?
21:53:55 <andythenorth> supermop_Home forums maybe?
21:54:16 *** _2TallTyler has quit IRC (Quit: Page closed)
21:58:14 <supermop_Home> needs some more timbers on outside, plus a chimney and awning
22:00:46 <andythenorth> ok this needs an ogfx PR :)
22:01:04 <andythenorth> "not tonight matthew" :P
22:01:08 *** DasPoseidon has quit IRC (Quit: DasPoseidon)
22:01:24 <andythenorth> supermop_Home nice regularity of form btw :)
22:01:38 <andythenorth> good building sprites rely a lot on rhythm, usually of windows
22:03:48 <supermop_Home> its all modular in photoshop
22:04:01 <supermop_Home> so i can stack more floors etc
22:04:42 <supermop_Home> other than balconies poking out, you cant see but there is a floor slab for each floor too
22:05:05 <supermop_Home> to help get everything lined up and to make construction stages easily
22:06:19 <supermop_Home> ground tiles with greenery will have to wait a bit though.. don't feel like doing those yet, so these will just have grey concrete for now
22:06:51 <supermop_Home> wood siding ill have done tonight, then idk... how about / angled versions?
22:08:38 *** jottyfan has quit IRC (Quit: jottyfan)
22:13:38 <andythenorth> I like the regular grid placement personally
22:13:56 <andythenorth> we need a portico and a pool :P
22:14:31 <supermop_Home> by / i mean facing SE instead of SW
22:15:10 <supermop_Home> portico can go on front or back of lobby, or be a separate module?
22:15:44 <andythenorth> have you tried a layout with them stacked?
22:15:56 <andythenorth> portico might want a tile with a drive and some garden
22:16:00 <andythenorth> not sure how big it would be though
22:16:05 <andythenorth> sometimes trial and error is only way
22:16:22 <supermop_Home> of the 4 guest room floorplates... is a particular one best for ogfx? thinking the ogfx sprite should be lobby on north tile and guest room on south
22:17:32 <supermop_Home> stacked like tall or just a 1x2 layout?
22:18:38 <andythenorth> I mean just laid out yes :)
22:26:59 <andythenorth> Sam 2 ships done, such progress
22:27:09 <andythenorth> drawing ships more like drawing factories than trains
22:28:29 *** Samu has quit IRC (Quit: Leaving)
22:32:30 <DorpsGek> [OpenTTD/OpenTTD] michicc opened pull request #8484: Fix: [OSX] Register fonts loaded directly from disk for text layout https://git.io/JL5xV
22:32:57 <michi_cc> TrueBrain: This interacts with #8482, whichever is merged first causes the other to need updating.
22:38:43 <TrueBrain> michi_cc: easy solution: approve and merge mine, and fix yours :D :D :P
22:42:47 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8482: Add: use our search-paths to find fonts based on relative filenames too https://git.io/JL5pf
22:42:51 <michi_cc> Your fault if it breaks :P
22:44:12 <TrueBrain> I promise I will fix it if it does :) can't merge atm, on mobile and not logged in :(
22:44:26 <TrueBrain> So feel free to merge for me :)
22:44:42 <DorpsGek> [OpenTTD/OpenTTD] michicc closed issue #7830: Load font from openttd config file directory, not from working directory. https://git.io/JewMG
22:44:45 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8482: Add: use our search-paths to find fonts based on relative filenames too https://git.io/JL5Fv
22:54:57 <DorpsGek> [OpenTTD/OpenTTD] michicc updated pull request #8484: Fix: [OSX] Register fonts loaded directly from disk for text layout https://git.io/JL5xV
22:56:45 <michi_cc> It would still be better if OSX would also render the glyphs, but that is a while different thing.
23:02:45 <TrueBrain> One issue at the time :D
23:03:31 <DorpsGek> [OpenTTD/OpenTTD] mattkimber opened pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
23:07:00 <andythenorth> supermop_Home top left one best?
23:15:06 <supermop_Home> lobby plus column 'A' guest rooms?
23:18:05 <supermop_Home> also doubled up snow on chalet style roofs to make it 1px thicker
23:18:35 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5hB
23:20:41 * andythenorth must to sleep soon
23:21:16 <supermop_Home> ill send sprite sheet and psd via forums, unless via GitHub somehow is better
23:22:15 <DorpsGek> [OpenTTD/OpenTTD] michicc updated pull request #8484: Fix: [OSX] Register fonts loaded directly from disk for text layout https://git.io/JL5xV
23:22:59 <andythenorth> forums is fine :)
23:23:08 *** andythenorth has quit IRC (Quit: andythenorth)
23:25:16 <Timberwolf> michi_cc: Thanks, that feels like a nice approach of collecting everything together as a MutableSpriteCache.
23:29:11 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5hb
23:33:24 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5hj
23:35:34 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8481: Add: always set PERSONAL_DIR "/content_download" in search path https://git.io/JL5jJ
23:37:11 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:37:40 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8484: Fix: [OSX] Register fonts loaded directly from disk for text layout https://git.io/JL5jY
23:39:43 <LordAro> APPROVE ALL THE THINGS
23:40:42 <Eddi|zuHause> quick! approve #7000 before frosch looks at it :p
23:43:23 <frosch123> pff, i am looking at it just now
23:47:13 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5j5
23:47:15 <Eddi|zuHause> i think people are going crazy. now they're shooting at trains
23:49:35 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdee
23:49:53 <DorpsGek> [OpenTTD/OpenTTD] frosch123 requested changes for pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdev
23:51:05 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdek
23:52:29 <frosch123> Eddi|zuHause: guis resolve the vehicle sprites every time, and do not use the cache. those guis can also show different sprites than the viewports due to var10
23:52:53 *** Progman has joined #openttd
23:53:40 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdeZ
23:55:18 *** sla_ro|master has quit IRC ()
23:56:42 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLde2
23:57:07 <LordAro> michi_cc: what do you think about #8452? seems broadly positive to me, but you've probably done more work with the map array than most
23:57:34 <DorpsGek> [OpenTTD/OpenTTD] stormcone commented on pull request #8483: Fix: Don't terraform during town generation https://git.io/JLdeV
23:59:31 <michi_cc> LordAro: I like the idea. I've not checked if all the descriptions actually match.
continue to next day ⏵
