IRC logs for #openttd on OFTC at 2025-11-13
⏴ go to previous day
02:31:26 *** lobstarooo has joined #openttd
02:31:26 *** lobster has quit IRC (Read error: Connection reset by peer)
02:42:50 *** Wormnest has quit IRC (Quit: Leaving)
03:55:55 *** gnu_jj_ has joined #openttd
03:59:14 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
04:24:03 *** Zathras_11 has joined #openttd
04:24:25 *** Zathras_1 has joined #openttd
04:27:36 *** Zathras has quit IRC (Ping timeout: 480 seconds)
04:27:46 *** Zathras_4 has quit IRC (Ping timeout: 480 seconds)
04:40:31 <DorpsGek> - Update: Translations from eints (by translators)
05:25:18 *** tokai|noir has joined #openttd
05:25:18 *** ChanServ sets mode: +v tokai|noir
05:32:21 *** tokai has quit IRC (Ping timeout: 480 seconds)
06:40:27 *** Flygon has quit IRC (Read error: Connection reset by peer)
07:03:54 *** WormnestAndroid has quit IRC (Remote host closed the connection)
07:03:57 *** WormnestAndroid has joined #openttd
08:08:37 <LordAro> i don't think maintaining bug-compatibility for Next() is necessary. Otherwise, why are you changing anything?
09:21:02 <peter1138> Yeah, compatibility is for API changes.
11:20:22 <xarick> visual studio 2026 is out
11:25:25 <xarick> well im not sure i wanna install it
11:59:49 *** reldred has quit IRC (Quit: User went offline on Discord a while ago)
12:07:33 <xarick> well let's see if clang works
12:15:08 <xarick> so it's just visual studio but with a futuristic look
12:45:07 <kuhnovic> Yes, it's all looks. I'm sure they did absolutely zero improvements at all. And they definitely did not list that in the what's new tab.
12:45:39 <peter1138> Remember when they decided the menu headings should be all-uppercase?
12:46:47 <peter1138> Though I don't remember what version that was.
12:47:47 <kuhnovic> I don't recall that, I think I skipped over that version
13:03:53 <kuhnovic> Ah yes, I went from 2012 to 2015
13:44:33 *** lobster has joined #openttd
13:44:33 *** lobstarooo has quit IRC (Read error: Connection reset by peer)
14:01:02 <xarick> i need to test Begin() on empty lists
14:01:10 <xarick> because I suspect.... stuff
14:40:50 <Borg> peter1138: okey.. all works :) if you want minimap 2x zoom-in w/o too much changes :) I can hand a patch..
14:41:11 <Borg> smallmap_gui.cpp didnt changed much really.. 1.8.0 vs 15.0 :) (master)
15:30:56 *** lobstarooo has joined #openttd
15:30:56 *** lobster has quit IRC (Read error: Connection reset by peer)
16:28:09 *** Wormnest has joined #openttd
16:55:44 <audigex> Is anyone able to tell me why the game locks up (spins) when generating a map with this loaded?
16:55:44 <audigex> I assume it's something to do with the "sheffield_supertram" items, but I can't see anything obviously wrong and I'm getting no errors from either nmlc or OpenTTD
16:56:03 <audigex> I'd offer a crash log but... it doesn't crash, it just spins
17:11:01 <xarick> regression is failing why
17:14:37 <audigex> audigex: Seems to be something to do with variant_group
17:14:37 <audigex> Possibly (due to the way the files are processed in my build) it’s falling down when the variant header refers to a vehicle that hasn’t yet been defined?
17:16:08 <peter1138> There was an issue detecting circular loops of variants at some point, but it's been fixed since.
17:16:49 <peter1138> Variants are always linked at the end once everything is loaded, so order of definition doesn't matter.
17:17:19 <peter1138> It loads in master for me, but there are no rail types and no rail vehicles available.
17:18:15 <peter1138> Not sure why a tram set would disable rail vehicles.
17:18:18 <audigex> If I remove the variant_group it works fine, with the variant_groups set it just spins
17:18:18 <audigex> MacOS nightly from maybe a week ago, I’ll try it on my Windows rig later in case
17:18:26 <audigex> peter1138: It’s also a train set
17:18:36 <xarick> did i just trigger a platform specific regression issue? or
17:19:02 <xarick> windows passes regression, linux and mac fail
17:20:56 <audigex> It’s a tram and rail metro set with metro-as-trams and tram-as-train options (via param, at some point)
17:21:00 <xarick> i dont have linux to see what's happening 🙁
17:23:56 <xarick> 955 seems to be about std::swap
17:25:34 <xarick> the test result of IsEnd() after a list swap differs between windows and linux/mac
17:25:49 <xarick> other than that I dunno what else
17:26:06 <xarick> i dont have a way to verify anything outside windows
17:26:37 <peter1138> Well, you do have variant loops, which should probably be avoided.
17:30:06 <audigex> peter1138: As in two vehicles trying to use the other as a variant group?
17:30:40 <peter1138> Just some kind of loop. Could be A->B->A, or could be A->B->C->D->A.
17:31:33 <peter1138> These IDs are have a loop somewhere: 427, 426, 425, 428, 413, 402, 401, 41c, 3fd, 3fc, 3ff, 3fe
17:31:42 <peter1138> Of course, coming from NML that's probably not very useful.
17:32:47 <audigex> I’ll have a look shortly, baby’s demanded my attention for now
17:35:28 <peter1138> Simple loops were always detected but another fix for loops came 2 months ago. When a loop is detected it just removes the variant stuff, so it shouldn't hang.
17:40:25 *** lobster has joined #openttd
17:40:25 *** lobstarooo has quit IRC (Read error: Connection reset by peer)
17:54:14 *** lobstarooo has joined #openttd
17:54:14 *** lobster has quit IRC (Read error: Connection reset by peer)
18:03:58 <peter1138> I assume SwapList assumes you're not actually iterating before and after without restarting interation...
18:04:34 <peter1138> Retarget() doesn't fix up the bucket iterators.
18:07:00 <peter1138> (Maybe it shouldn't need to)
18:10:15 <andythenorth> lunch was....underwhelming
18:13:11 <_glx_> my guess would be some UB somewhere
18:41:25 <xarick> I don't have the means to test mac
18:43:58 <LordAro> suspicious discussion is suspicious
18:46:30 <xarick> I uncovered a can of worms "bug"?
18:47:23 <xarick> but yeah, i'd also like to get a crash log
18:54:12 <xarick> let me check the sorter tipes of the 2 lists
18:59:08 <xarick> they're both sort by item, ascending apparently
19:19:36 *** ckb has quit IRC (Ping timeout: 480 seconds)
19:59:16 <xarick> there's no bucket_iter in ScriptListSorterItemAscending
20:01:26 <xarick> so is it the suspicious part?
20:14:20 <_glx_> first step is to understand the code 🙂
20:14:59 <_glx_> XXXSorterItemXXX don't need to iterate buckets for obvious reason
20:15:44 <xarick> would this change anything?
20:16:15 <_glx_> no, because that's the list, not the sorter
20:16:37 <_glx_> list has both items and bucket, also for obvious reason
20:26:57 *** Flygon has quit IRC (Read error: Connection reset by peer)
20:35:10 <LordAro> worth throwing ubsan at it?
20:56:55 <andythenorth> hmm what do we call ground?
20:57:24 <andythenorth> `foo_allowed = ['land', 'water']`
20:59:15 <andythenorth> maybe it's tile_type
21:08:50 <andythenorth> pff I guess I just split it into 2 bools
21:09:13 <andythenorth> `allow_on_land = True`, `allow_on_water = False` etc
21:09:22 <andythenorth> some things can't be named
21:10:47 *** Borg has quit IRC (Quit: leaving)
21:27:36 <xarick> who's the mac guy? andythenorth?
21:29:39 <andythenorth> and the compile farm
21:31:13 <xarick> can you reproduce the 14772 regression mac crash and post the log?
21:42:18 *** kuka_lie has joined #openttd
21:44:04 <audigex> Didn't quite work out what was causing it, but I've added "Don't use your own ID as a variant_group" and a file to track A>B, B>C etc and scan for loops, and between the two that seems to have solved it
21:50:26 <_glx_> hmm and macos builds are full of -Wdeprecated-declarations and -Wundef (guess the SDK has been updated)
21:54:13 <xarick> like a headless chicken
21:54:39 <_glx_> VS can debug using WSL2
21:54:50 <xarick> it's visual studio code
22:05:43 <_glx_> hmm for testing I added an assert(false) in Valuate()
22:06:08 <_glx_> when running ctest with a debug build I get ```CMake Error at D:/developpement/GitHub/glx22/OpenTTD/cmake/scripts/Regression.cmake:44 (message):
22:06:08 <_glx_> Unexpected output: Crash encountered, generating crash log...```
22:07:41 <_glx_> but only a regression diff with release build
22:11:09 <xarick> ah, i was missing a compiler
22:13:58 <xarick> nop i still dont know what im doing
22:17:00 *** kuka_lie has quit IRC (Quit: Lost terminal)
22:20:35 <xarick> Bad CMake executable: "". Check to make sure it is installed or the value of the "cmake.cmakePath" setting contains the correct path
22:24:48 <_glx_> you installed all the required packages ?
22:26:26 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:29:02 <xarick> no idea, i have zero idea what i need to do
22:30:33 <_glx_> oh it's a pain to do it in vscode
22:30:47 <_glx_> easier to use a terminal for that
22:33:41 <xarick> I'm spreading openttd repositories everywhere
22:37:19 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:37:31 <kaji_kaede> xarick: Check the settings of the extension you're using.
22:38:22 <kaji_kaede> I assume "" is meant to be the file path it's using to search for CMake, which... Given it's empty... Probably not gonna find CMake there.
22:54:26 <xarick> im under the impression i dont have cmake installed
22:55:42 <xarick> they tell me to adjust cmake.cmakePath
22:55:52 <xarick> but i have no idea where this setting is
22:56:30 <xarick> and have no idea whether i have a cmake executable 🙂 im am that clueless on vs code
23:00:25 <xarick> how to install cmake on WSL thingy
23:05:41 <xarick> i have no idea if i have cmake, because linux likes to not use file extensions...
continue to next day ⏵
