⏴ go to previous day
00:19:21 *** wallabra has quit IRC (Ping timeout: 480 seconds)
01:19:53 *** Wormnest has quit IRC (Quit: Leaving)
01:35:06 *** wallabra has joined #openttd
02:50:04 *** gnu_jj_ has joined #openttd
02:52:59 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:53:00 *** WormnestAndroid has joined #openttd
02:53:09 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:53:09 *** WormnestAndroid has joined #openttd
02:53:13 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:53:21 *** geizeskrank has quit IRC (Ping timeout: 480 seconds)
02:53:21 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
02:53:24 *** WormnestAndroid has joined #openttd
02:56:54 *** geizeskrank has joined #openttd
03:50:21 *** D-HUND has joined #openttd
03:53:53 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:07:16 *** keikoz has joined #openttd
04:13:55 *** Flygon has quit IRC (Read error: Connection reset by peer)
04:59:12 *** toktik has quit IRC (Remote host closed the connection)
04:59:26 *** toktik has joined #openttd
05:12:20 *** keikoz has quit IRC (Ping timeout: 480 seconds)
06:15:07 *** Gukky has joined #openttd
06:15:32 *** Gukky has quit IRC ()
06:21:15 <andythenorth> hmm
06:21:20 <andythenorth> if nml has #include
06:21:44 <andythenorth> can we cache the results of the parse for each included block, and only update if its changed?
06:24:16 *** D-HUND is now known as debdog
06:27:30 *** izhirahider has quit IRC (Read error: Connection reset by peer)
06:27:43 *** izhirahider has joined #openttd
07:10:19 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #13940: Codechange: Use unique_ptr to manage grffiles. https://github.com/OpenTTD/OpenTTD/pull/13940
07:16:36 <andythenorth> dumping `// comment` from the properties block saves 6 MB of nml
07:16:45 <andythenorth> might have reduced nml parse time a tiny bit
07:18:43 <andythenorth> probably a rounding error 😛
07:43:16 *** keikoz has joined #openttd
07:56:29 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #13941: Codechange: CargoSpec::Iterate() already tests cargo validity. https://github.com/OpenTTD/OpenTTD/pull/13941
08:17:38 *** mindlesstux has quit IRC (Quit: The Lounge - https://thelounge.chat)
08:18:32 *** mindlesstux has joined #openttd
08:38:55 <DorpsGek> [OpenTTD/survey-web] survey-summary[bot] pushed 1 commits to main https://github.com/OpenTTD/survey-web/commit/191c71974aea5a477009b8c00fec1e03ca08844d
08:38:56 <DorpsGek> - Add: summary for Q1 of 2025 (by OpenTTD Survey)
08:46:22 *** SigHunter has quit IRC ()
08:49:21 *** SigHunter has joined #openttd
10:15:19 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356572628017746013/image.png?ex=67ed0e36&is=67ebbcb6&hm=294d5106c57ee7d62538e0d3989b5269a2f08d62ffc6bf53929f49a4b9c46d02&
10:15:19 <xarick> :/
10:31:41 <xarick> oh, right, today is also a stupid day, everyone's going jokes
10:36:58 <peter1138> I guess OpenTTD is now dead. No April 1st release.
10:40:33 <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #13941: Codechange: CargoSpec::Iterate() already tests cargo validity. https://github.com/OpenTTD/OpenTTD/pull/13941#pullrequestreview-2732259173
10:43:34 <xarick> strange, when supersimpleai crashes, it's always on company 0
10:43:49 <xarick> very suspicious
10:45:58 <_glx_> Company 0 is usually the human one
10:46:37 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356580506308710401/image.png?ex=67ed158c&is=67ebc40c&hm=d22db0967ed6cb57fb2745930e36fadf0036f907019ed270a8a24e28f7d3b5d2&
10:47:46 <xarick> i made a spectator slot for myself so i could use company 0 as AI
10:49:05 <_glx_> I know, but maybe the AI does something preventing it to work in slot 0
10:50:30 <_glx_> Of course the screenshot doesn't show the crash reason, only the end of stack
10:51:54 <xarick> this is the other crash: https://www.tt-forums.net/download/file.php?id=232247
10:57:28 <Rubidium> lets remove dedicated builds and big endian support. Survey says nobody uses them :D
10:58:38 <xarick> ah rats, it's using libraries
10:59:12 <xarick> i hate it when AIs aren't standalone
10:59:29 <_glx_> I'd check around railbuilder.nut:50
10:59:32 <xarick> need to open multiple folders as workspaces
11:00:11 <_glx_> Then the callers to see why it's passing null
11:01:15 <xarick> oh, my library folder isn't all that clean 😦
11:04:16 <xarick> gonna check if i made changes to Graph.AyStar.4
11:04:24 <xarick> i think i didn't, but since it's a folder...
11:04:29 <xarick> and not a .tar
11:07:57 <xarick> confirmed, it's unchanged, gonna delete folder
11:16:37 <xarick> wow his code is quite complex, akin to mine 🙂
11:22:54 <xarick> MyAITileList_Passinglane wow
11:23:06 <xarick> class MyAITileList_Passinglane extends AITileList
11:26:52 <xarick> wow!
11:26:52 <xarick> `function MyAITileList_Passinglane::AddPassinglane(ps_id, ps_entry, ps_frontentry, ps_blockentry, ps_frontblockentry, ps_exit, ps_frontexit, ps_blockexit, ps_frontblockexit)`
11:45:40 <_glx_> Hmm that will be fun on saving
11:46:18 <_glx_> And more after loading
11:47:34 <_glx_> Maybe the API should mention the instance framework
11:52:42 <_glx_> Though based on the name my guess it just uses these extra params to fill a normal AITileList, like other derivatives
12:39:11 <_glx_> and looking at the code my guess is wrong
12:40:45 <_glx_> so somehow `MyAITileList_Passinglane::GetTileByValue()` returns null
12:43:51 <_glx_> haha so the author created `MyAITileList_Passinglane` to actually use it like a table
12:43:58 <_glx_> or an array
12:44:33 <_glx_> silly
12:46:00 <xarick> counter is -1
12:47:26 <xarick> GetExitPair(counter - 1), I dunno, but that looks weird
12:47:37 <xarick> get exitpair of (-2)?
12:48:34 <xarick> oh my bad
12:48:51 <xarick> he has a `count` and a `counter` variables
12:48:55 <xarick> count is 1
12:52:23 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356612156879016067/image.png?ex=67ed3306&is=67ebe186&hm=1b943ff3bdeb1f00286ce168987907f8e512cc577d367cfc34d3a2109804f6a1&
12:52:23 <xarick> this looks odd
12:52:35 <xarick> wrong offset?
12:53:00 <xarick> pink wasn't here at the time of the crash btw
12:54:02 <xarick> there should be 5 passing lanes around this area
12:54:07 <xarick> i can only find 2
12:58:53 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356613795337998386/screenshot77.png?ex=67ed348d&is=67ebe30d&hm=94b6d9a62480616136b09d18d77669d4245c50d1dad06068e74c0920c63912c1&
12:58:53 <xarick> okay, the 3rd passing lane seems to be not built, and if it were built, it would be on water.
12:59:20 <xarick> i wonder if depots indicate a "passing lane"
13:00:35 <_glx_> hmm could it be trying to build stuff near map border ?
13:00:56 <xarick> doubtful, this place isn't nowhere close to the border
13:15:01 *** Flygon has joined #openttd
13:16:55 <xarick> i can't see 5 passing lanes
13:17:15 <xarick> BuildPassingLaneSection must have not returned null on failure - me thinks
13:17:44 <xarick> i see in the log it successfully built the 5 passing lanes...
13:17:50 <xarick> i don't see where
13:26:12 <_glx_> code is not super readable (arrays of arrays)
13:26:29 <__abigail> Making unreadable code is my specialty 🫡
13:36:03 <xarick> BuildOrCheckPassingLaneSection - wowo... what a dreaded function to look at
13:36:14 <xarick> I can't look at it 😦
13:37:14 <xarick> a bajillion variables
13:38:20 <xarick> 7 "vectors" which are actually... arrays of arrays
13:38:56 <xarick> oh, they're not
13:45:51 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356625614769094801/image.png?ex=67ed3f8f&is=67ebee0f&hm=386a0df48d0460e960231f2b0459219efdcfa1922f8a0f47eb3ecae05a8b36d0&
13:45:51 <xarick> how can he keep up with these many
13:48:08 <xarick> well, I've done worse 😐
13:48:14 <xarick> so i shouldn't be complaining
13:49:15 <xarick> i have no idea what type or kind of passing lane type failed
13:49:21 <xarick> failed to fail
13:49:40 <xarick> it's most likely around here where it fails, but I'm not gonna dig in
13:49:52 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13934: Add #12939: [NewGRF] Add road-/tram-/rail-type variable 0x45 to get mutual road-/tram-/rail-type on same tile https://github.com/OpenTTD/OpenTTD/pull/13934#pullrequestreview-2732769485
13:53:05 <xarick> this AI name should have been SuperComplexAi
14:01:04 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on discussion #12939: Level crossing extensions in NewGRF https://github.com/OpenTTD/OpenTTD/discussions/12939
14:02:04 <Rubidium> it's hard to make something look simple, so if it looks complex it's probably simple
14:07:42 <DorpsGek> [OpenTTD/OpenTTD] frosch123 merged pull request #13926: Fix 089c7a2: SDL2 application name hint was not effective. https://github.com/OpenTTD/OpenTTD/pull/13926
14:07:50 <DorpsGek> [OpenTTD/OpenGFX2] zephyris opened issue #154: Clean builds fail due to interpretation of `%` by make https://github.com/OpenTTD/OpenGFX2/issues/154
14:09:56 <DorpsGek> [OpenTTD/OpenTTD] frosch123 approved pull request #13939: Fix: NewGRF details view does not show the "No information available"… https://github.com/OpenTTD/OpenTTD/pull/13939#pullrequestreview-2732853579
14:11:02 <DorpsGek> [OpenTTD/OpenTTD] frosch123 merged pull request #13939: Fix: NewGRF details view does not show the "No information available"… https://github.com/OpenTTD/OpenTTD/pull/13939
14:11:06 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356631969563410513/image.png?ex=67ed457a&is=67ebf3fa&hm=1ab4cb545a27b8b726c14cb547c3b02899170a9b583afcee1613977bb1735d37&
14:11:06 <xarick> wow, this is a "passing lane"
14:11:29 <xarick> so it takes many forms and shapes and can even be single line
14:11:31 <xarick> and depotless
14:12:14 <xarick> need to study this AI more indepth
14:14:29 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #13934: Add #12939: [NewGRF] Add road-/tram-/rail-type variable 0x45 to get mutual road-/tram-/rail-type on same tile https://github.com/OpenTTD/OpenTTD/pull/13934#issuecomment-2769517706
14:57:03 *** mastertrams has joined #openttd
14:57:03 <mastertrams> I know it wasn't historically possible, but have I just imagined that there's now a way to read OpenTTD savefiles outside the game?
15:06:29 <peter1138> Depends what you mean by read. It's always been possible to read the raw data.
15:06:49 <_jgr_> mastertrams: <https://github.com/TrueBrain/OpenTTD-savegame-reader>
15:28:02 *** nielsm has joined #openttd
15:31:47 *** michi_cc_ has quit IRC (Ping timeout: 480 seconds)
15:42:52 *** michi_cc_ has joined #openttd
15:57:51 <DorpsGek> [OpenTTD/OpenGFX2] zephyris opened pull request #155: Codechange: Makefile rewrite to fix multiple issues https://github.com/OpenTTD/OpenGFX2/pull/155
16:04:36 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13940: Codechange: Use unique_ptr to manage grffiles. https://github.com/OpenTTD/OpenTTD/pull/13940#issuecomment-2769869994
16:07:13 <DorpsGek> [OpenTTD/OpenGFX2] rubidium42 commented on pull request #155: Codechange: Makefile rewrite to fix multiple issues https://github.com/OpenTTD/OpenGFX2/pull/155#pullrequestreview-2733247527
16:22:03 *** Wolf01 has joined #openttd
16:22:23 <xarick> vcpkg install failedhttps://github.com/SamuXarick/OpenTTD/actions/runs/14200603444/job/39786375885
16:23:12 <Wolf01> No release today?
16:31:10 <Rubidium> there seems to be little interest in a release
16:35:28 <xarick> "savegames uploaded here never leave your computer"
16:36:16 <xarick> i can't delete them?
16:37:06 <peter1138> It means the savegame is not actually uploaded, it's processed only on your computer.
16:38:57 <xarick> how long does it take to analyze
16:41:13 <peter1138> Depends on the savegame. If it's a normal savegame, you won't notice it.
16:41:22 <peter1138> If it's one of your silly saves, it might never finish.
16:42:28 <xarick> RESULT_CODE_HUNG
16:42:31 <xarick> well... ok
16:44:37 <_glx_> xarick: rebase
16:48:01 <xarick> i can't rebase master
16:49:54 <_glx_> oh I see, seems bad luck, your action picked a too recent image
16:50:10 <_glx_> anyway it's something to fix
16:55:54 *** WormnestAndroid has quit IRC (Remote host closed the connection)
16:55:59 *** WormnestAndroid has joined #openttd
16:58:08 <DorpsGek> [OpenTTD/OpenTTD] glx22 opened pull request #13942: Codefix f719fa678f: [CI] Also apply CMake fix to normal CI https://github.com/OpenTTD/OpenTTD/pull/13942
17:00:09 <_glx_> this time it was lzo
17:00:26 *** Wormnest has joined #openttd
17:01:40 <Rubidium> whack-a-mole-o'clock :(
17:01:46 <Rubidium> no wonder we can't do releases anymore
17:02:23 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #13942: Codefix f719fa678f: [CI] Also apply CMake fix to normal CI https://github.com/OpenTTD/OpenTTD/pull/13942#pullrequestreview-2733395091
17:03:14 <Rubidium> I guess it needs some more work
17:03:50 <_glx_> and it doesn't work, but I did exactly like nightly workflows
17:04:50 <Rubidium> maybe it doesn't work for nightlies either, but we were lucky with the worker we got
17:05:16 <_glx_> possible yes
17:06:11 <_glx_> yup nightly used last week image
17:07:17 <_glx_> let's try another way
17:08:22 <peter1138> Hmm, std::vector but pre-reserved? Or std::deque? (But with unknown allocation size...)
17:09:01 <DorpsGek> [OpenTTD/OpenTTD] glx22 dismissed a review for pull request #13942: Codefix f719fa678f: [CI] Also apply CMake fix to normal CI https://github.com/OpenTTD/OpenTTD/pull/13942#pullrequestreview-2733395091
17:09:04 <DorpsGek> [OpenTTD/OpenTTD] glx22 updated pull request #13942: Codefix f719fa678f: [CI] Also apply CMake fix to normal CI https://github.com/OpenTTD/OpenTTD/pull/13942
17:09:24 <_glx_> hoping it will pick sunday's image
17:09:57 <_glx_> and it did
17:18:25 <_glx_> and still fails
17:23:29 <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #13941: Codechange: CargoSpec::Iterate() already tests cargo validity. https://github.com/OpenTTD/OpenTTD/pull/13941
17:25:34 <DorpsGek> [OpenTTD/OpenGFX2] zephyris updated pull request #155: Codechange: Makefile rewrite to fix multiple issues https://github.com/OpenTTD/OpenGFX2/pull/155
17:28:59 <Rubidium> _glx_: might you need to pass that -DCMAKE_... to the cmake --build line as well?
17:29:29 <_glx_> I'll try using action's `env`
17:30:12 <DorpsGek> [OpenTTD/OpenTTD] glx22 updated pull request #13942: Codefix f719fa678f: [CI] Also apply CMake fix to normal CI https://github.com/OpenTTD/OpenTTD/pull/13942
17:31:20 <_glx_> for the step now, but might better to put it globally like `CTEST_OUTPUT_ON_FAILURE`
17:33:13 <_glx_> oups I failed at copy/paste
17:35:46 <DorpsGek> [OpenTTD/OpenTTD] glx22 updated pull request #13942: Codefix f719fa678f: [CI] Also apply CMake fix to normal CI https://github.com/OpenTTD/OpenTTD/pull/13942
17:45:59 <_glx_> well, I don't have more ideas
17:46:41 <_glx_> https://cdn.discordapp.com/attachments/1008473233844097104/1356686222398853391/image.png?ex=67ed7801&is=67ec2681&hm=36a4a03ee636c454e97c95f921980b5ca0771f95e0e6c4c19ed3e708906f85e5&
17:46:41 <_glx_> env var is set, but has no effect
17:48:26 <DorpsGek> [OpenTTD/OpenTTD] zephyris commented on issue #13915: [Bug]: Excessive (spam of) vehicle testing offers https://github.com/OpenTTD/OpenTTD/issues/13915
17:50:09 <_glx_> there's also <https://github.com/microsoft/vcpkg/issues/44726> so I guess the only option is to wait
17:53:14 <xarick> is there an evil newgrf which could make a new type of rail with a max speed lower than the engines that can run on it?
17:53:44 <_glx_> oh seems like `vcpkg_cmake_configure` in portfile.cmake skips anything we could set
17:54:23 <_glx_> xarick: I think that already exists
17:54:29 <_glx_> same for road types
17:54:44 <xarick> 🙂
17:55:33 <michi_cc> If you read in the other channels here, you could even see requests to add GRF parameters to individually change the top speed for each rail type.
17:56:47 <xarick> getting a matching engines, wagons and railtypes is so complicated
17:57:13 <xarick> some railtypes can be electrified
17:57:27 <xarick> the compatibilities confuse me
17:57:38 <_glx_> oh have fun with different electrification too 🙂
17:58:12 <_glx_> 25kV AC, 3kV DC, 1.5kV DC
17:58:24 <_glx_> overhead, third rail
17:58:51 <belajalilija> c:
17:58:52 <_glx_> plus the different gauges
18:00:28 <_glx_> it might be easier to check grfid and use hardcoded decisions
18:19:00 <peter1138> Either way you "just" need to choose a train and railtype that are compatible.
18:23:51 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13940: Codechange: Use unique_ptr to manage grffiles. https://github.com/OpenTTD/OpenTTD/pull/13940#issuecomment-2770333351
18:29:44 <_zephyris> So, I can copy and adapt opengfx workflows (https://github.com/OpenTTD/OpenGFX/tree/master/.github/workflows)
18:29:50 <_zephyris> Where else should I look?
18:33:48 <_zephyris> I've found where it points to in OpenTTD/actions, but no big insights to my untrained eye...
18:42:36 *** tokai has joined #openttd
18:42:36 *** ChanServ sets mode: +v tokai
18:43:41 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #13940: Codechange: Use unique_ptr to manage grffiles. https://github.com/OpenTTD/OpenTTD/pull/13940
18:47:52 <xarick> okay, this is where my AI stumbles, especially with a NewGRF such as Iron Horse... <https://github.com/SamuXarick/LuDiAI-AfterFix/blob/master/TrainMain.nut#L111-L129>
18:48:28 <xarick> Utils.ArrayHasItem is ... bad, just bad
18:48:38 <xarick> main source of slowdown
18:49:36 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
18:52:19 <xarick> ArrayGetValue is also there
18:52:29 <xarick> triple slow
18:54:44 <xarick> i barely remember why I did this
18:55:15 <Rubidium> luckily the commit message making that change explains why you did it
18:56:09 <xarick> i dont think so
19:04:48 <DorpsGek> [OpenTTD/OpenTTD] frosch123 opened pull request #13943: Remove C-style string parsing and composing https://github.com/OpenTTD/OpenTTD/pull/13943
19:15:55 <DorpsGek> [OpenTTD/OpenTTD] frosch123 closed pull request #13862: Codechange: Use std::string_view in FormatString, and validate string bounds while parsing https://github.com/OpenTTD/OpenTTD/pull/13862
19:15:58 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #13862: Codechange: Use std::string_view in FormatString, and validate string bounds while parsing https://github.com/OpenTTD/OpenTTD/pull/13862#issuecomment-2770452787
19:18:19 <xarick> i need a different method to do this
19:33:56 <DorpsGek> [OpenTTD/website] rubidium42 opened pull request #346: Add: 'Release or not' news item https://github.com/OpenTTD/website/pull/346
19:43:31 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356715622335123629/image.png?ex=67ed9362&is=67ec41e2&hm=d625a61f1974e963fb597a465f06267ab49caaf0f920c9831915f15894c5302e&
19:43:31 <xarick> so i manually expanded all branches of iron horse, took me 25 minutes
19:45:55 <DorpsGek> [OpenTTD/OpenTTD] github-advanced-security[bot] commented on pull request #13943: Remove C-style string parsing and composing https://github.com/OpenTTD/OpenTTD/pull/13943#pullrequestreview-2733890649
19:47:38 <Rubidium> frosch123: funny to see Marathi being used :D
19:50:15 <frosch123> I did not debug today's commits yet, left for tomorrow :)
19:50:42 <Rubidium> it looks like a great leap forward
19:51:44 <frosch123> For now I am annoyed that codeql does not understand maybe_unused. Nor the c++23/26? Placeholder
19:52:00 <_glx_> Rubidium: only mentioned releases in the post, but even dev starts to become problematic since PR fail to build too
19:52:08 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #13944: Codechange: Use default operator<=> to compare Linkgraph Hop. https://github.com/OpenTTD/OpenTTD/pull/13944
19:54:23 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356718358162051454/image.png?ex=67ed95ef&is=67ec446f&hm=b5f21c475a3ff581ee8069493b04d98d7185a7b3ad6c04e422af13cce9abf6a6&
19:54:23 <xarick> Iron Horse, according to an AI
19:54:52 <_glx_> many are just variants
19:55:17 <peter1138> Andy saw variants and thought it was a challenge.
19:55:20 <xarick> my ai only selects passengers and mail
19:55:44 <_glx_> just the API doesn't easily filter that out (unlike GUI)
20:00:15 <xarick> some of his variants are double the size
20:00:30 <xarick> of the base
20:02:37 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356720429523275857/image.png?ex=67ed97dd&is=67ec465d&hm=6d300260c33e016f099e20cbc99c9253de2dc25fd1f9a03157dadffdd6333d4c&
20:03:57 *** gelignite has joined #openttd
20:11:52 <_glx_> hmm since API has AINewGRF and AINewGRFList, maybe we should add some AIEngine::GetGRFID and AIVehicle::GetGRFID
20:12:16 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356722859371921468/image.png?ex=67ed9a20&is=67ec48a0&hm=ed8ad75d757987fadf07cbe0778394aa83a367a003ee96f5b696cd902403ca00&
20:12:34 <xarick> 14071 ticks
20:12:37 <xarick> for mail
20:13:00 <xarick> and that's not even going into the Utils.Array thing, I disabled that part
20:13:21 <_glx_> yes because your AI tries every combination
20:13:44 <_glx_> suboptimal with most used grfs
20:15:20 <_glx_> and it probably fails when newgrf has strict rules about how a consist must be to be allowed to start
20:15:20 <xarick> 1196 wagons can transport mail... that's just insane for 1 newgrf
20:15:42 <_glx_> almost all wagons can refit to mail
20:16:01 <_glx_> I think
20:20:32 <xarick> 506 * 1196 * 5 = total combinations
20:21:09 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356725092452663346/image.png?ex=67ed9c34&is=67ec4ab4&hm=45e62dad3c4dc8f9e77ce85376aa88d73b572ba4102602e3f9cfc21caabdd1af&
20:21:09 <xarick> hmm yeah...
20:21:41 <xarick> I can't do this method
20:32:10 <xarick> clearly I misunderstand arrays
20:36:48 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356729031600640050/image.png?ex=67ed9fdf&is=67ec4e5f&hm=4cfa6851cd5affc816da5b210a6f14b874837a610a6a6151c9910b5d34985937&
20:37:19 <xarick> ArraySetValue did nothing
20:37:36 *** SigHunter_ has joined #openttd
20:40:23 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
20:43:00 <xarick> `local railtypes = Utils.ArrayGetValue(engineWagonPairs, pair);`
20:43:33 <xarick> it returns an array... or a reference/pointer to the array, i dunno something fishy going on here
20:43:50 <xarick> but there are no such thing in squirrel
20:44:04 <xarick> or is it?
20:46:03 <_glx_> no pointer in squirrel
20:48:01 <_glx_> so if you return the array it's actually a copy
20:48:15 <_glx_> IIRC
20:54:08 *** nielsm has quit IRC (Ping timeout: 480 seconds)
20:58:06 <xarick> seems to actually return the real one
20:59:48 <xarick> let me try again
21:04:50 <_glx_> indeed after testing it seems to not be a copy
21:05:38 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356736287268278474/image.png?ex=67eda6a1&is=67ec5521&hm=15b961370b9a9f36d1b88593984269d4dbf5204824990035e90ac8a6e89ec227&
21:06:00 <xarick> ArraySetValue is useless
21:06:17 <xarick> unless the value i set is actually not an array
21:06:46 <xarick> the value here is "railtypes"
21:07:24 <xarick> same memory address?
21:09:12 <_glx_> and after checking source I can confirm every type is by reference except integer, float, bool and user pointer
21:10:12 <_glx_> so yes by reference they have the same address
21:12:00 <xarick> thx
21:12:40 <_glx_> .clone() exists to make copy
21:13:44 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356738327738581113/image.png?ex=67eda888&is=67ec5708&hm=b69f768d1a9ef3d4c3bd1188b4469e6c787b4fa53ae6bc7bd7f66957bdb50b9d&
21:13:44 <xarick> I'm trying to interpret this magic I did here...
21:14:01 <xarick> what is actually happening
21:14:11 <xarick> it works, but I don't remember why I did this
21:14:56 <xarick> it should be named ArrayHasArray I guess
21:16:32 <_glx_> seems very complex
21:18:00 <_glx_> haha my testing was wrong anyway since I was actually testing with list (ie instance), not array
21:28:26 <DorpsGek> [OpenTTD/OpenGFX2] zephyris opened pull request #156: Fix: 0.25x and 0.5x zoom selectors weren't converted to 8bpp https://github.com/OpenTTD/OpenGFX2/pull/156
21:29:25 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1356742273370427473/image.png?ex=67edac35&is=67ec5ab5&hm=8faefb934856c87a0a1201eb2adb45698db3aab1d8fe13718db35ccb1743e59e&
21:29:31 <_glx_> and it seems we don't support `clone` for API instances
21:29:50 <xarick> item[0] is the engine
21:29:56 <xarick> item[1] is the wagon
21:30:28 <xarick> Array Has item: [engine, wagon]
21:30:33 <xarick> okay, that's starting to make sense
21:31:20 <DorpsGek> [OpenTTD/OpenGFX2] zephyris commented on issue #149: [Bug]: Tunnel icons are 1x zoom only. https://github.com/OpenTTD/OpenGFX2/issues/149
21:33:21 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:33:44 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:42:01 <DorpsGek> [OpenTTD/OpenGFX2] zephyris opened pull request #157: Fix: 2x zoom tunnel icons were excluded from Classic baseset https://github.com/OpenTTD/OpenGFX2/pull/157
21:48:20 <xarick> oh, foreach then do while
21:48:36 <xarick> interestingly slow
21:54:37 <xarick> foreach engine { foreach wagon { foreach railtype { foreach [engine,wagon] item in array { do { while } } } } }
21:54:48 <DorpsGek> [OpenTTD/OpenGFX2] zephyris updated pull request #157: Fix: 2x zoom tunnel icons were excluded from Classic baseset https://github.com/OpenTTD/OpenGFX2/pull/157
21:54:59 <xarick> + iron horse
21:55:26 <xarick> need ideas to make this more efficient
21:55:29 <xarick> and saveable
21:56:16 <DorpsGek> [OpenTTD/OpenGFX2] PeterN approved pull request #157: Fix: 2x zoom tunnel icons were excluded from Classic baseset https://github.com/OpenTTD/OpenGFX2/pull/157#pullrequestreview-2734130863
21:59:34 <xarick> engineWagonPairs = [ [[engine, wagon], [railtype1, railtype2]] ], this seems to be the structure
22:01:08 <xarick> railtypes have a minimum of 1 entry
22:01:30 <xarick> a maximum of how many compatible types the engine wagon combo can run on
22:01:42 <xarick> variable array size
22:02:42 <DorpsGek> [OpenTTD/OpenGFX2] zephyris merged pull request #157: Fix: 2x zoom tunnel icons were excluded from Classic baseset https://github.com/OpenTTD/OpenGFX2/pull/157
22:02:47 <DorpsGek> [OpenTTD/OpenGFX2] zephyris closed issue #149: [Bug]: Tunnel icons are 1x zoom only. https://github.com/OpenTTD/OpenGFX2/issues/149
22:03:35 <xarick> not sure if it's NewGRF specific
22:04:09 <xarick> what is the maximum number of different railtypes an engine can run on?
22:05:43 <peter1138[d]> "Yes"
22:05:49 <peter1138[d]> Technically all of them.
22:05:57 <xarick> wow, great
22:29:44 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
22:30:51 *** zanooda2000 has quit IRC (Quit: User went offline on Discord a while ago)
22:31:30 <_glx_> annoying handling of metamethods in squirrel, thrown errors are ignored
22:33:22 *** WormnestAndroid has joined #openttd
22:37:13 <_glx_> because metamethod failing is the same as metamethod not being implemented
22:38:38 <peter1138> Hmm.
22:41:14 *** gelignite has quit IRC (Quit: Stay safe!)
22:42:26 <_glx_> hmm let's try to not handle nicely in try-catch blocks in squirrel_helper.hpp
22:45:07 <_glx_> yeah not a good idea, it ends up in thread NOT_REACHED()
22:54:04 <DorpsGek> [OpenTTD/OpenTTD] dependabot[bot] opened pull request #13945: Bump Apple-Actions/import-codesign-certs from 3 to 5 in the actions group https://github.com/OpenTTD/OpenTTD/pull/13945
23:20:57 <DorpsGek> [OpenTTD/OpenTTD] stormcone opened pull request #13946: Fix: Reference to the correct section of the README if a graphics or … https://github.com/OpenTTD/OpenTTD/pull/13946
23:37:29 <DorpsGek> [OpenTTD/OpenTTD] glx22 opened pull request #13947: Fix: [Script] Prevent cloning of API instances https://github.com/OpenTTD/OpenTTD/pull/13947
23:53:24 <_glx_> seems they are rolling back runner images
23:57:10 <_glx_> my PR ran on quite old 20250309.1.0
23:57:26 <_glx_> (or it's just luck)
continue to next day ⏵
