Config
Log for #openttd on 4th October 2022:
Times are UTC Toggle Colours
00:16:58  *** Wormnest has quit IRC
00:28:34  *** Elysianthekitsunesheher[m] has quit IRC
00:40:26  *** Wormnest has joined #openttd
01:20:46  *** BrainkillerXL has quit IRC
01:33:07  *** Wormnest has quit IRC
02:11:38  *** Wormnest has joined #openttd
02:24:15  *** D-HUND has joined #openttd
02:27:37  *** debdog has quit IRC
02:42:26  *** Wormnest has quit IRC
03:08:51  *** tokai|noir has joined #openttd
03:08:51  *** ChanServ sets mode: +v tokai|noir
03:15:28  *** tokai has quit IRC
04:26:09  *** sla_ro|master has joined #openttd
05:04:22  *** Flygon has joined #openttd
05:25:05  <pickpacket> I wish I could see *why* a town hates me so much
05:25:48  <pickpacket> tokai|noir: don't be afraid of missing out :)
06:39:06  *** D-HUND is now known as debdog
07:17:35  <petern> Usually trees.
07:49:52  <pickpacket> not this time apparently
07:54:23  <debdog> a station with a bad rating perhaps?
07:54:34  <pickpacket> huh. That could be it
09:21:49  *** felix_ has joined #openttd
09:21:49  *** felix has quit IRC
09:59:03  <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #10064: Fix: link overlay line width on HiDPI displays https://github.com/OpenTTD/OpenTTD/pull/10064
10:08:57  <andythenorth> is it lunch?
10:09:18  <petern> No.
10:09:23  <petern> Is it refactor?
10:11:44  <andythenorth> usually
10:12:24  <LordAro> yup
10:15:56  <reldred> no it’s dinner
10:17:15  <petern> https://cdn.discordapp.com/attachments/1008473233844097104/1026800161051131944/unknown.png
10:17:15  <petern> This window needs refactor
10:17:57  <glx[d]> It needs an horizontal scrollbar
10:20:12  <pickpacket> andythenorth: I just had lunch
10:21:18  <petern> Which idiot did 64 cargo types? Stupid decisions...
10:21:20  <LordAro> it needs fewer industries
10:21:34  <petern> Yeah, it is... intense.
10:21:48  <pickpacket> glx[d]: do you mean vertical scrollbar?
10:22:09  <LordAro> at least needs to work out how to divide by 2 properly
10:22:34  <LordAro> couple more pixels padding on the bottom too
10:22:43  <pickpacket> pfft. It's too many industries for my taste, but I can't see how that would be universally true. It's a pretty cool NewGRF
10:22:55  <petern> It works out the number of rows needed for cargo types and industry types, and then fills it by columns.
10:23:17  <petern> When you have multiple columns, horizontal scrollbar works better than vertical.
10:23:47  <petern> LordAro, pretty sure I have the pixel problems solved πŸ˜„
10:23:48  <pickpacket> I would prefer a simple list with a horizontal scroll bar though
10:35:05  <petern> in the original game there are no resizable windows, but the smallmap has a small and large mode. the buttons are down the side in the small mode.
10:48:06  <pickpacket> Some windows are resizable in 1.10.3 at least. Did you mean the original TTD?
10:49:26  <petern> What else would "original" mean?
10:57:32  <Rubidium> TTO?
11:04:26  <pickpacket> 🀷
11:04:45  <pickpacket> I have no relation to TTD... OpenTTD is "default" for me
11:07:30  <LordAro> don't worry, petern is just really old
11:09:05  <petern> That would just be TT πŸ˜„
11:13:59  <petern> And I'm not even the longest nor oldest OTTD dev πŸ˜„
11:15:59  <petern> andythenorth: salad time
11:26:13  <pickpacket> time salad!
11:30:27  *** sla_ro|master2 has joined #openttd
11:33:18  *** sla_ro|master has quit IRC
11:40:22  *** sla_ro|master has joined #openttd
11:45:18  *** sla_ro|master2 has quit IRC
12:22:17  *** felix has joined #openttd
12:22:17  *** felix_ has quit IRC
12:35:51  <andythenorth> petern: I had chicken salad
12:35:54  <andythenorth> shall we do pictures?
12:36:12  <andythenorth> the best social content
12:36:17  <andythenorth> https://cdn.discordapp.com/attachments/1008473233844097104/1026835150341951559/9EB1DDD6-8599-4D62-8C2D-74C51BCC245B.heic
12:37:20  <andythenorth> https://cdn.discordapp.com/attachments/1008473233844097104/1026835410984378438/unknown.png
12:51:59  <petern> I think pictures of mine would be frowned upon.
13:20:48  *** Guest2213 has quit IRC
13:20:48  *** JamesRoss[m] has quit IRC
13:20:48  *** karoline[m] has quit IRC
13:20:48  *** udo[m] has quit IRC
13:20:48  *** thomas[m]123 has quit IRC
13:20:48  *** paulus[m] has quit IRC
13:20:48  *** osvaldo[m] has quit IRC
13:20:48  *** jeeg[m] has quit IRC
13:20:49  *** dag[m] has quit IRC
13:20:49  *** phil[m] has quit IRC
13:20:49  *** shedidthedog[m] has quit IRC
13:20:49  *** philip[m]12 has quit IRC
13:20:49  *** joey[m]1 has quit IRC
13:20:49  *** jeremy[m] has quit IRC
13:20:49  *** CornsMcGowan[m] has quit IRC
13:20:49  *** linda[m] has quit IRC
13:20:49  *** fiddeldibu[m] has quit IRC
13:20:49  *** menelaos[m] has quit IRC
13:20:49  *** aperezdc__ has quit IRC
13:20:49  *** rudolfs[m] has quit IRC
13:20:49  *** leward[m] has quit IRC
13:20:49  *** giords[m] has quit IRC
13:20:49  *** elliot[m] has quit IRC
13:20:49  *** nolep[m] has quit IRC
13:20:49  *** ookfof[m] has quit IRC
13:20:49  *** olmvnec[m] has quit IRC
13:20:49  *** natmac[m] has quit IRC
13:20:49  *** glothit7ok[m] has quit IRC
13:20:49  *** dude[m] has quit IRC
13:20:49  *** jact[m] has quit IRC
13:20:49  *** NekomimiGunner18[m] has quit IRC
13:20:49  *** blikjeham[m] has quit IRC
13:20:49  *** EmeraldSnorlax[m] has quit IRC
13:20:49  *** natalie[m] has quit IRC
13:20:49  *** yur3shmukcik[m] has quit IRC
13:20:49  *** christoph[m] has quit IRC
13:20:52  *** twom[m] has quit IRC
13:20:52  *** pothyurf[m] has quit IRC
13:20:52  *** ist5shreawf[m] has quit IRC
13:20:52  *** hylshols7qui[m] has quit IRC
13:20:53  *** hamstonkid[m] has quit IRC
13:20:53  *** karl[m]123 has quit IRC
13:20:53  *** josef[m] has quit IRC
13:20:53  *** Heiki[m] has quit IRC
13:20:53  *** einar[m] has quit IRC
13:20:53  *** andythenorth[m] has quit IRC
13:20:53  *** grag[m] has quit IRC
13:20:54  *** igor[m]12 has quit IRC
13:20:54  *** freu[m] has quit IRC
13:20:54  *** robert[m] has quit IRC
13:20:54  *** nartir[m] has quit IRC
13:20:54  *** labs[m] has quit IRC
13:20:54  *** cawal[m] has quit IRC
13:20:54  *** johanna[m] has quit IRC
13:20:54  *** magdalena[m] has quit IRC
13:20:54  *** patrick[m] has quit IRC
13:20:54  *** lezzbehonest[m] has quit IRC
13:20:54  *** kstar892[m] has quit IRC
13:20:54  *** albert[m] has quit IRC
13:20:54  *** zzy2357[m] has quit IRC
13:20:54  *** vista_narvas[m] has quit IRC
13:20:54  *** thelonelyellipsis[m] has quit IRC
13:20:54  *** luk3Z[m] has quit IRC
13:20:54  *** khavik[m] has quit IRC
13:20:54  *** gretel[m] has quit IRC
13:20:54  *** gdown has quit IRC
13:20:54  *** cjmonagle[m] has quit IRC
13:20:54  *** patricia[m] has quit IRC
13:20:54  *** amal[m] has quit IRC
13:20:54  *** citronbleuv[m] has quit IRC
13:31:01  *** albert[m] has joined #openttd
13:35:19  *** sla_ro|master has quit IRC
13:58:49  <DorpsGek> [OpenTTD/team] kgecmen opened issue #349: [tr_TR] Translator access request https://github.com/OpenTTD/team/issues/349
14:08:59  *** nielsm has joined #openttd
14:24:41  <supermop_toil> YO
15:12:21  *** gelignite has joined #openttd
15:27:58  <supermop_toil> i love the tto ui
15:30:01  <pickpacket> what's the difference between Industries and IndustryTiles in NML?
15:31:59  <andythenorth> Tiles are per tile, and belong to an industry
15:32:10  <andythenorth> they are primarily concerned with displaying the sprites
15:32:21  <andythenorth> so most of their properties, vars and callbacks relate to that
15:32:31  <andythenorth> Industry has the production logic, and the tile layouts
15:32:50  <andythenorth> both Industry and IndustryTile can have location restrictions when constructing industry
15:33:00  <pickpacket> okay. Does that mean I need IndustryTiles to show sprites?
15:33:04  <andythenorth> yes
15:33:09  <andythenorth> they're composed into a layout
15:33:35  <andythenorth> there is a sidebar about the default baseset tiles
15:33:39  <andythenorth> but best ignore that
15:35:32  <pickpacket> Do I have to declare accepted cargoes in both? It's done very differently
15:39:59  <supermop_toil> i think you have to set which tiles accept the cargo - its possible to have tiles of a power plant not accept coal, for example. I believe in the base industries the substation/transformer tile does not
15:40:09  <andythenorth> iirc, there's a flag for making the tiles follow the industry acceptance
15:40:16  <andythenorth> that's relatively modern, so might not be in the examples
15:40:38  <andythenorth> INDTILE_FLAG_ACCEPT_ALL
15:40:56  <andythenorth> use that, unless you're trying to do fancy stuff
15:41:04  <supermop_toil> might be realistic to force players to drop coal off at the furnace part, but mostly it just serves to confuse players
15:41:08  <andythenorth> +
15:41:19  <petern> supermop_toil, that's why https://fuzzle.org/~petern/ottd/tto.png
15:41:35  <supermop_toil> ohhh hell yes
15:41:55  <supermop_toil> pillow-y toy buttons
15:42:01  <petern> (That was 9 years ago, haha)
15:42:06  <LordAro> oh, that's what you've really been working towards
15:42:15  <petern> It's not πŸ˜„
15:42:24  <LordAro> but it could be!
15:42:48  <andythenorth> also did we integrate Doom! yet?
15:43:08  <supermop_toil> has anyone ported openttd to TI-86 yet?
15:44:25  <supermop_toil> i wonder how non-volatile that memory is... will mine that i haven't used in 20 years still have tetris on it?
15:58:39  <petern> Not just battery backed?
16:30:45  <petern> Well that's weird, `#include "safeguards.h"` works fine when it should really be `#include "../safeguards.h"`
16:30:51  *** HerzogDeXtEr has joined #openttd
16:35:39  <LordAro> i imagine cmake is adding -I. or similar to the compiler?
16:36:10  <glx[d]> supermop_toil: the TI-85/TI-86 tetris ? with link play ?
16:37:03  <glx[d]> or .. full path is in include path
16:42:11  <DorpsGek> [OpenTTD/team] glx22 commented on issue #349: [tr_TR] Translator access request https://github.com/OpenTTD/team/issues/349
16:51:23  *** Wolf01 has joined #openttd
16:52:40  <Wolf01> Meh, google dismisses iot core on august 2023... I didn't receive any communication yet, I need to find an alternative service :(
16:53:23  *** Wormnest has joined #openttd
16:56:29  *** wallabra_ has joined #openttd
17:00:01  <supermop_toil> glx[d]: i didn't have my own cable, always had to borrow one to get games
17:01:17  *** wallabra has quit IRC
17:01:17  *** wallabra_ is now known as wallabra
17:27:19  *** Idkwrh has joined #openttd
17:27:20  <Idkwrh> hello
17:27:45  <supermop_toil> hi
17:42:30  *** sla_ro|master has joined #openttd
17:47:16  <WoozyDragon4018> supermop_toil: bruhh πŸ’€πŸ’€
17:48:26  *** amal[m] has joined #openttd
17:48:27  *** andythenorth[m] has joined #openttd
17:48:27  *** aperezdc__ has joined #openttd
17:48:28  *** blikjeham[m] has joined #openttd
17:48:28  *** cawal[m] has joined #openttd
17:48:28  *** christoph[m] has joined #openttd
17:48:28  *** citronbleuv[m] has joined #openttd
17:48:29  *** cjmonagle[m] has joined #openttd
17:48:29  *** lezzbehonest[m] has joined #openttd
17:48:30  *** CornsMcGowan[m] has joined #openttd
17:48:30  *** dag[m] has joined #openttd
17:48:30  *** dude[m] has joined #openttd
17:48:31  *** einar[m] has joined #openttd
17:48:31  *** elliot[m] has joined #openttd
17:48:32  *** EmeraldSnorlax[m] has joined #openttd
17:48:32  *** fiddeldibu[m] has joined #openttd
17:48:33  *** gdown has joined #openttd
17:48:33  *** freu[m] has joined #openttd
17:48:34  *** giords[m] has joined #openttd
17:48:34  *** glothit7ok[m] has joined #openttd
17:48:34  *** grag[m] has joined #openttd
17:48:35  *** gretel[m] has joined #openttd
17:48:37  *** hamstonkid[m] has joined #openttd
17:48:38  *** Heiki[m] has joined #openttd
17:48:38  *** hylshols7qui[m] has joined #openttd
17:48:38  *** igor[m]12 has joined #openttd
17:48:38  *** ist5shreawf[m] has joined #openttd
17:48:38  *** jact[m] has joined #openttd
17:48:38  *** jeeg[m] has joined #openttd
17:48:39  *** jeremy[m] has joined #openttd
17:48:40  *** joey[m]1 has joined #openttd
17:48:40  *** johanna[m] has joined #openttd
17:48:40  *** josef[m] has joined #openttd
17:48:40  *** karl[m]1234 has joined #openttd
17:48:41  *** karoline[m] has joined #openttd
17:48:42  *** khavik[m] has joined #openttd
17:48:42  *** kstar892[m] has joined #openttd
17:48:42  *** labs[m] has joined #openttd
17:48:44  *** leward[m] has joined #openttd
17:48:44  *** linda[m] has joined #openttd
17:48:44  *** luk3Z[m] has joined #openttd
17:48:45  *** magdalena[m] has joined #openttd
17:48:45  *** menelaos[m] has joined #openttd
17:48:45  *** nartir[m] has joined #openttd
17:48:45  *** natalie[m] has joined #openttd
17:48:45  *** natmac[m] has joined #openttd
17:48:45  *** NekomimiGunner18[m] has joined #openttd
17:48:45  *** nolep[m] has joined #openttd
17:48:45  *** olmvnec[m] has joined #openttd
17:48:45  *** ookfof[m] has joined #openttd
17:48:45  *** osvaldo[m] has joined #openttd
17:48:45  *** patricia[m] has joined #openttd
17:48:45  *** patrick[m] has joined #openttd
17:48:46  *** paulus[m] has joined #openttd
17:48:46  *** phil[m] has joined #openttd
17:48:47  *** philip[m]12 has joined #openttd
17:48:47  *** pothyurf[m] has joined #openttd
17:48:47  *** robert[m] has joined #openttd
17:48:48  *** rudolfs[m] has joined #openttd
17:48:48  *** shedidthedog[m] has joined #openttd
17:48:49  *** thelonelyellipsis[m] has joined #openttd
17:48:50  *** thomas[m]1234 has joined #openttd
17:48:50  *** twom[m] has joined #openttd
17:48:51  *** JamesRoss[m] has joined #openttd
17:48:51  *** udo[m] has joined #openttd
17:48:52  *** vista_narvas[m] has joined #openttd
17:48:52  *** Elysianthekitsunesheher[m] has joined #openttd
17:48:53  *** YourOnlyOne has joined #openttd
17:48:53  *** yur3shmukcik[m] has joined #openttd
17:48:54  *** zzy2357[m] has joined #openttd
17:49:26  *** YourOnlyOne is now known as Guest2317
17:50:20  <andythenorth> yo
17:50:23  <andythenorth> was it lunch?
17:50:27  * andythenorth has been busy working
17:59:35  <pickpacket> I'm slowly getting somewhere, I think. I'm struggling because the tutorial covers making a road vehicle, but making a new industry seems to be very different
18:04:22  <pickpacket> "The cargos that are accepted by this tile. CargoID is a cargo type from your cargo translation table, amount is the acceptance of the given cargo in 1/8 units." --- Whaaaa...? "cargo translation table"? Eighths of units?
18:04:39  <pickpacket> INDTILE_FLAG_ACCEPT_ALL makes sense though
18:04:51  <andythenorth> if you're in NML
18:05:02  <Bouke> https://github.com/OpenTTD/OpenTTD/issues/10069 are there debug symbols available for the macOS builds?
18:05:02  <andythenorth> and you use the accept all flag
18:05:17  <andythenorth> you can ignore all of that part of the docs
18:05:53  <andythenorth> however I can explain if you want πŸ™‚
18:07:28  <pickpacket> andythenorth: I'll probably love an explanation at some later point but right now I just want the simplest solution :D
18:07:47  <pickpacket> When an ID for anything is given as XXh what does that mean?
18:07:50  <pickpacket> hex?
18:10:21  *** jfs- has joined #openttd
18:10:21  <jfs-> Bouke: you already have them, that backtrace you posted contains symbol names for all the stack frames including those in openttd code
18:11:08  <Bouke> jfs-: I kinda was expecting line numbers to go with those offsets?
18:11:48  <jfs-> oh... I'm not sure they would make much sense with an optimized build tho
18:13:10  <JGR> Line numbers won't be included in vanilla release builds, if that is turned on they're still mostly correct and useful though
18:14:16  <FLHerne> pickpacket: hex, yes
18:14:20  <Bouke> I thought that's what debug symbols are for? To map offsets to line numbers?
18:14:40  <Bouke> (and also offsets to function names of course)
18:15:02  <FLHerne> pickpacket: cargotable https://newgrf-specs.tt-wiki.net/wiki/NML:Cargotable
18:15:11  <pickpacket> FLHerne: πŸ‘
18:15:31  <jfs-> but like, that backtrace seems like total nonsense to me... it's in some macos input system and then making a savegame load/save stream filter and then suddenly handling a socket event and cleaning out the companies pool?
18:15:31  <jfs-> were you in a network game when it happened? loading a game? saving?
18:15:37  <FLHerne> tile acceptance is in n/8, where 8/8 in a station's catchment area is required for the cargo to be accepted
18:15:48  <FLHerne> for industries and town buildings
18:16:20  <FLHerne> so if you want any one tile to represent the whole industry, just make them all 8/8
18:16:30  <Bouke> I *think* I was about to do a save of a network game.
18:16:45  <pickpacket> FLHerne: but can a tile be 2/8 within a catchment area? Either the tile is in or out, no?
18:16:50  <FLHerne> if they were all 3/8 a station would have to include at least three tiles in its catchment area
18:16:58  <pickpacket> oh! I see
18:17:00  <JGR> The stack trace is not necessarily wrong, the issue is that the symbol offsets are very large
18:17:48  <FLHerne> like a lot of industry features, it's not really obvious what the point is :p
18:17:49  <JGR> Implying that it's using the wrong symbols
18:18:19  <FLHerne> you can add a lot of complexity, but all it does is confuse players for the sake of very marginal 'realism' improvements
18:18:54  <FLHerne> it makes more sense with houses
18:19:43  <Bouke> The `CleanPool` that's included with the stacktrace is something that I've seen before: e.g. https://github.com/OpenTTD/OpenTTD/issues/9876#issuecomment-1138585636 and https://github.com/OpenTTD/OpenTTD/issues/8501#issue-779384225. Is that maybe part of some cleanup that isn't related to the actual crash?
18:20:22  <JGR> `CleanPool() + 226451` indicates that it's not actually in that function
18:21:09  <petern> I know templates can expand a bit but that's a bit excessive πŸ˜„
18:22:04  <JGR> You might have better luck using lldb or addr2line
18:34:41  <Bouke> I cannot reproduce the crash at will. I only have this crash log and the 12.2 binary. Is there any way to get a usable crash location / stack trace?
18:35:23  <JGR> Is the binary form the website/github, or one you compiled yourself?
18:37:37  <Bouke> From the website.
18:43:59  <Bouke> (note: it's the macOS build)
18:46:57  <petern> https://cdn.discordapp.com/attachments/1008473233844097104/1026928424192835695/unknown.png
18:46:57  <petern> Excessive refactor...
18:49:17  *** EmeraldSnorlax[m] has quit IRC
18:49:17  *** einar[m] has quit IRC
18:49:17  *** olmvnec[m] has quit IRC
18:49:17  *** dag[m] has quit IRC
18:49:17  *** jeeg[m] has quit IRC
18:49:17  *** grag[m] has quit IRC
18:49:18  *** elliot[m] has quit IRC
18:49:18  *** linda[m] has quit IRC
18:49:18  *** amal[m] has quit IRC
18:49:18  *** johanna[m] has quit IRC
18:49:18  *** kstar892[m] has quit IRC
18:49:18  *** patrick[m] has quit IRC
18:49:18  *** paulus[m] has quit IRC
18:49:18  *** Guest2317 has quit IRC
18:49:18  *** Heiki[m] has quit IRC
18:49:18  *** NekomimiGunner18[m] has quit IRC
18:49:18  *** thelonelyellipsis[m] has quit IRC
18:49:18  *** thomas[m]1234 has quit IRC
18:49:18  *** karoline[m] has quit IRC
18:49:18  *** twom[m] has quit IRC
18:49:18  *** ist5shreawf[m] has quit IRC
18:49:18  *** fiddeldibu[m] has quit IRC
18:49:18  *** jact[m] has quit IRC
18:49:18  *** udo[m] has quit IRC
18:49:18  *** aperezdc__ has quit IRC
18:49:18  *** phil[m] has quit IRC
18:49:18  *** osvaldo[m] has quit IRC
18:49:19  *** zzy2357[m] has quit IRC
18:49:19  *** philip[m]12 has quit IRC
18:49:19  *** leward[m] has quit IRC
18:49:19  *** gretel[m] has quit IRC
18:49:19  *** giords[m] has quit IRC
18:49:19  *** jeremy[m] has quit IRC
18:49:19  *** vista_narvas[m] has quit IRC
18:49:19  *** hamstonkid[m] has quit IRC
18:49:19  *** cjmonagle[m] has quit IRC
18:49:19  *** khavik[m] has quit IRC
18:49:19  *** nolep[m] has quit IRC
18:49:19  *** freu[m] has quit IRC
18:49:19  *** labs[m] has quit IRC
18:49:19  *** cawal[m] has quit IRC
18:49:19  *** dude[m] has quit IRC
18:49:19  *** magdalena[m] has quit IRC
18:49:20  *** patricia[m] has quit IRC
18:49:20  *** igor[m]12 has quit IRC
18:49:20  *** karl[m]1234 has quit IRC
18:49:20  *** natalie[m] has quit IRC
18:49:20  *** josef[m] has quit IRC
18:49:20  *** christoph[m] has quit IRC
18:49:20  *** shedidthedog[m] has quit IRC
18:49:20  *** JamesRoss[m] has quit IRC
18:49:20  *** rudolfs[m] has quit IRC
18:49:20  *** luk3Z[m] has quit IRC
18:49:20  *** CornsMcGowan[m] has quit IRC
18:49:20  *** gdown has quit IRC
18:49:20  *** joey[m]1 has quit IRC
18:49:20  *** blikjeham[m] has quit IRC
18:49:20  *** andythenorth[m] has quit IRC
18:49:21  *** albert[m] has quit IRC
18:49:21  *** pothyurf[m] has quit IRC
18:49:21  *** natmac[m] has quit IRC
18:49:21  *** glothit7ok[m] has quit IRC
18:49:21  *** ookfof[m] has quit IRC
18:49:21  *** nartir[m] has quit IRC
18:49:21  *** hylshols7qui[m] has quit IRC
18:49:21  *** yur3shmukcik[m] has quit IRC
18:49:21  *** robert[m] has quit IRC
18:49:21  *** lezzbehonest[m] has quit IRC
18:49:21  *** menelaos[m] has quit IRC
18:49:22  *** citronbleuv[m] has quit IRC
18:49:22  *** Elysianthekitsunesheher[m] has quit IRC
18:57:39  <JGR> I've dumped the symbol table, and the address does fall into a large gap
18:58:03  <JGR> The previous symbol is indeed the CleanPool function
18:59:05  *** albert[m] has joined #openttd
18:59:22  <Bouke> So would've "proper" debug symbols have helped here? I mean the compiler must've known what contributed code contributed to that offset?
19:03:13  <JGR> If you do a local build with `-DCMAKE_BUILD_TYPE=RelWithDebInfo` you should get better crash logs
19:03:55  <JGR> The compiler will have known but it won't have written that information anywhere for a release build
19:13:14  <pickpacket> ugh... does anyone have a GIMP pallette for the DOS colours?
19:14:17  <JGR> <https://github.com/OpenTTD/OpenTTD/tree/master/docs/palettes>
19:15:00  <Bouke> Should we include that build flag with the release builds on the website?
19:15:26  <nielsm> it makes the output file significantly larger
19:18:19  <JGR> Macosx seems to produce a disproportionate number of crash bug reports so it may still be worthwhile?
19:18:56  <TrueBrain> Bouke: https://github.com/OpenTTD/OpenTTD/blob/c7f55a058be78841ffe63ba525f3d0879ffe480c/.github/workflows/release.yml#L334
19:18:56  <TrueBrain> Like this? πŸ˜„
19:19:45  <JGR> The macos binaries from my branch are about twice the size of the 12.2 ones from the website because of debug tables
19:20:04  <JGR> It's basically impossible to do anything with bug reports I get from mac users otherwise
19:21:23  <TrueBrain> (Oops, linked Linux, bit lower same code exists for MacOS)
19:21:57  <Bouke> TrueBrain: Apparently, but that's been there for 2 years -- so shouldn't 12.2 already have shipped with them?
19:22:04  <TrueBrain> It should
19:22:14  <TrueBrain> If they aren't, means something stripped the binary
19:22:29  <TrueBrain> Which was not by design
19:23:30  <JGR> If it was stripped the stack trace would not have any symbols at all
19:24:49  <andythenorth> might be time I taught nmlc to say *which* pixels are animated
19:24:55  <andythenorth> not just their bounding box location
19:25:08  <andythenorth> exact pixel positions would be much more useful
19:25:18  <TrueBrain> JGR: Β―\_(ツ)_/Β―
19:26:21  <frosch> c++23 is said to have a stacktrace library πŸ™‚
19:26:46  <frosch> but who knows when/if that is available for osx
19:32:52  <andythenorth> meh nml pixel_stats doesn't track x,y locs
19:32:57  <andythenorth> maybe I should just write my own thing
19:33:00  <nielsm> but it's not like that would help having debug symbols in the binary
19:33:19  <andythenorth> it's really common that other people's sprites have spurious animated pixels
19:33:23  <andythenorth> and I have to manually fix
19:33:38  <andythenorth> particularly their are browns and yellows where the RGB value is almost identical
19:33:58  <andythenorth> total pain in the arse
19:34:06  <frosch> andythenorth: feed them through rgbaeater
19:34:22  <frosch> it's one of the many thing to integrate into nml
19:35:25  <andythenorth> oh V wrote docs and stuff πŸ™‚
19:35:26  <andythenorth> --disallowed_colour_indexes
19:35:52  <andythenorth> all I want is the actual x,y position
19:36:04  <frosch> jfs-: https://www.sandordargo.com/blog/2022/09/21/cpp23-stacktrace-library it has line numbers and stuff
19:36:07  <andythenorth> quite often there are *supposed* to be some animated pixels in the spritesheet
19:36:13  <andythenorth> but not everywhere
19:36:23  <Bouke> So we've established, based on the stacktrace, that `-DCMAKE_BUILD_TYPE=RelWithDebInfo` is enabled for the build, but that we expected a more detailed stacktrace?
19:36:49  <frosch> Bouke: that stacktrace in the issue is broken
19:36:54  <frosch> stack corruption or something
19:37:00  <frosch> more debug symbols would not help there
19:37:13  <TrueBrain> I guess the only way to get "more" is to disable optimizations, honestly πŸ˜„
19:38:07  <frosch> there is no point in "more" if already the first frame is wrong
19:38:08  <andythenorth> "moar"
19:38:21  <TrueBrain> frosch: there is a reason it is quoted πŸ˜›
19:38:23  <frosch> there is no "ServerNetworkGameSocketHandler" in a singleplayer game
19:39:22  <TrueBrain> frosch: so basically this means you can write non-OS specific crashlog code? πŸ™‚
19:39:25  <frosch> and then alternating between savegame compression and unicode input
19:39:37  <frosch> TrueBrain: exactly
19:39:49  <TrueBrain> would be nice ... I always dread having to add that to any application of mine
19:39:59  <Bouke> (it was a network game)
19:40:02  <TrueBrain> especially as OpenTTD's is .. hard to understand πŸ™‚
19:40:08  <Bouke> (see also the greyed out fast forward button)
19:41:02  <TrueBrain> it is almost a reason to upgrade to C++23 πŸ˜›
19:41:43  <JGR> A replacement stack trace library won't produce better results if the symbol table is still very short for whatever reason
19:41:54  <frosch> TrueBrain: it's only available for gcc currently, and possibly only linux
19:42:09  <TrueBrain> yeah .. well .. new good things start somewhere
19:42:17  <nielsm> does it also have a platform-agnostic way to assign helpful names to threads? that does not involve (explicitly) raising SEH exceptions on windows
19:45:32  <Bouke> So based on the stacktrace in that issue, there's no way to further fix this? I would have to attach a debugger and try to reproduce the crash? And (apart from C++ 23) we have no solution to improve the crash reporting (on macOS)?
19:45:53  <nielsm> if you don't have a memory dump file there probably isn't anything to do
19:46:06  <nielsm> but if there is a dump it might be able to reconstruct the true stack somehow
19:46:14  *** Flygon has quit IRC
19:46:18  <nielsm> s/able/possible/
19:46:30  <frosch> did you try loading the crash.sav?
19:46:40  <frosch> if it is game-state related, that would trigger it again
19:46:44  <frosch> but it probably isn't
19:46:49  <Bouke> Yep the save game loaded fine and we've been playing it for a few hours afterwards.
19:47:25  <Bouke> Does OTTD write a memory dump?
19:47:40  <frosch> on linux i have "ulimit -c unlimited" enabled by default
19:47:49  <frosch> no idea whether anything similar exists for osx
19:48:42  <pickpacket> I need to be off to bed... So far I've just been trying to get a "soylent green" one-tile factory to work. I'm not there yet, and I'm struggling with the whole sprite/spriteset/spritelayout shenanigans. Will continue tomorrow evening. If anyone is curious about my progress so far: https://notabug.org/tinyrabbit/tea-tea-deluxe
19:50:17  <nielsm> pickpacket: I think you need to place the FEAT_INDUSTRYTILES block before the FEAT_INDUSTRIES block
19:50:59  <nielsm> you need to define things before you use them
19:51:32  <nielsm> so first you define the raw graphics sprites, then you define the sprite layouts, then you define the tile layouts, and then you define the industry
19:52:04  <pickpacket> oh! I'll try that first thing tomorrow :D Thanks!
19:54:03  <andythenorth> oof, I suppose I should set all the decay periods and stuff for cargos
19:54:07  * andythenorth so bored by that idea
19:54:32  <nielsm> and you need a tilelayout block too, which defines a named layout for the tiles on the map, and then you refer that layout in the layouts property of the industry itself
19:57:18  <petern> andythenorth: making NewGRFs **is** the gameplay...
19:57:39  <andythenorth> only the fun parts though
19:57:59  <andythenorth> https://cdn.discordapp.com/attachments/1008473233844097104/1026946307354529802/unknown.png
19:57:59  <andythenorth> trying to improve this is blah blah
19:58:14  <andythenorth> and dP made another chart showing how all the cargo payment stuff is bollocks for gameplay
19:58:29  <andythenorth> was it lunch? πŸ™‚
19:58:47  <petern> It's Gu dessert
19:58:55  <andythenorth> I ran out of Gu
19:59:01  <andythenorth> our local Tesco resolutely doesn't sell them
19:59:05  <andythenorth> would have to go to Waitrose
19:59:13  <andythenorth> and the problem with Waitrose is the other customers
19:59:28  <JGR> <https://github.com/OpenTTD/OpenTTD/blob/c7f55a058be78841ffe63ba525f3d0879ffe480c/cmake/InstallAndPackage.cmake#L121>
20:00:05  <andythenorth> https://cdn.discordapp.com/attachments/1008473233844097104/1026946835115421796/unknown.png
20:00:06  <andythenorth> but launch the rocket
20:00:11  <andythenorth> I am winning the game
20:00:23  <petern> I can do that in Excel
20:00:23  <JGR> Diffing my branch with master and that is off in mine
20:01:19  <JGR> Bourke: might be worth doing a local build with that changed and see if you get a better result next time
20:01:26  <TrueBrain> TrueBrain: called it! πŸ˜› πŸ˜› πŸ˜›
20:01:32  <nielsm> I wonder if the industry example in NML should be extended to show making a new tile layout and include custom tile graphics
20:02:33  *** gelignite has quit IRC
20:02:35  <TrueBrain> in my humble opinion, we shouldn't strip. Those few extra MiB aren't an issue anymore in this day and age
20:04:29  <Bouke> JGR: This is great! I'll try this. Now I need to get the crash reproduced...
20:05:46  <Bouke> TrueBrain: Agreed; being able to understand a crash will be worth it.
20:09:30  <frosch> is there no option to put the debug symbols into a separate file?
20:09:40  <TrueBrain> Debian does it, so there is; but is that worth the effort?
20:09:55  <TrueBrain> we are TINY, compared to any other game that people actually play πŸ˜„
20:13:39  *** nielsm has quit IRC
20:13:45  <frosch> 7MB vs. 270MB is a difference
20:14:38  <frosch> no idea how often people download osx binaries compared to zbase
20:15:02  <Rubidium> is that compressed?
20:15:05  <frosch> though i guess binaries are deducted from valve's bandwidth, not ours
20:15:15  <frosch> no, uncompressed
20:15:54  <frosch> about 80MB compressed
20:16:19  <frosch> oh, and osx is universal, so even more
20:17:30  <Bouke> https://github.com/JGRennison/OpenTTD-patches/releases/tag/jgrpp-0.48.4 lists the macOS binaries (unstripped) as 15,4 MiB (compressed). On-disk it's 32 vs 54 MiB.
20:37:02  <Bouke> Interestingly macOS reports a different stacktrace for the same crash:
20:37:02  <Bouke> Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
20:37:02  <Bouke> 0   libsystem_kernel.dylib            0x00007fff727ff32a __pthread_kill + 10
20:37:02  <Bouke> 1   libsystem_pthread.dylib           0x00007fff728bfe60 pthread_kill + 430
20:37:02  <Bouke> 2   libsystem_c.dylib                 0x00007fff72786808 abort + 120
20:37:03  <Bouke> 3   org.openttd.openttd               0x0000000101d0dc27 0x101aca000 + 2374695
20:37:03  <Bouke> 4   libsystem_platform.dylib          0x00007fff728b45fd _sigtramp + 29
20:37:05  <Bouke> 5   ???                               0x00007fff8fc2f4c0 OBJC_CLASS_$_NSInputClientWrapper + 40
20:37:05  <Bouke> 6   org.openttd.openttd               0x0000000101dda036 0x101aca000 + 3211318
20:43:18  <TrueBrain> frosch: Where does that number come from? It is a bit without context πŸ™‚
20:43:55  <frosch> i am a bit confused about binary sizes
20:44:12  <frosch> 7MB is the current size of the offical linux-generic binary, uncompressed
20:44:38  <frosch> my local debug build makes 270MB, relwithdebug is 200MB, release is 12MB
20:44:59  <frosch> i did not try stripping my local builds
20:45:11  <frosch> but either way, my local debug builds are huge πŸ™‚
20:45:28  <TrueBrain> Debug builds always are πŸ˜„
20:45:43  <TrueBrain> I do not remember relwithdebinfo being that big, but I might be mistaken
20:45:55  <TrueBrain> I remember a 30% size increase
20:46:10  <TrueBrain> Not a 3000%
20:48:03  <JGR> Bouke: This is the same stack trace, the first few entries in the OpenTTD one are within the crash log writer
20:48:58  <frosch> TrueBrain: maybe the dwarfs need to compensate for something
20:49:13  <TrueBrain> Sounds fair
20:49:23  <Bouke> Oh hmm these don't appear to include Pool<...> as the first frames?
20:50:02  <JGR> Compare the addresses, not the symbol names
20:54:53  <Bouke> Up until `0x00007fff728b45fd` they appear to be the same, but from there on up they are different no? (0-3 vs [00]-[02])
20:56:26  <JGR> [00] to [02
20:56:29  *** Borg has joined #openttd
20:56:40  <JGR> [00] to [02] are within OpenTTD's crash log handler
20:57:21  <JGR> 0 - 3 are from after the log has been written and abort has been called from within the signal handler
21:07:45  <JGR> <https://gist.github.com/JGRennison/94e5bdb9f71def950820cd94a53265a8>
21:08:35  <JGR> [05] 101dda036 is the instruction after the call  highlighted here
21:11:35  <JGR> The symbols referenced in the next function down suggest that it's from one of the macos-specific files
21:13:04  <Bouke> Looks like some sort of callback from the AppKit framework.
21:13:34  <Bouke> JGR: Thanks, although I'm not that well-versed in reading semi-assembly :?
21:13:50  <glx[d]> btw all macos reports we have in issues start with Pool<..> and that's of course incorrect (even the touchbar crash reports)
21:15:39  <JGR> The function below looks like `- (NSAttributedString *)attributedSubstringForProposedRange:(NSRange)theRange actualRange:(NSRangePointer)actualRange`
21:16:00  <JGR> So possibly this is `- (BOOL)hasMarkedText`
21:16:43  <JGR> If `_focused_window` was null it would be consistent with the macos crash report
21:18:44  <Bouke> I do recall that while I clicked "save" I also clicked elsewhere in the game. So that might be it?
21:19:31  <glx[d]> Bouke: to test the crash report with cmake change, you can use master and the touchbar with 32bpp-sse4-anim
21:19:38  <Bouke> (That particular mouse is a little flacky, triggering additional mouse clicks instead of a single click)
21:35:11  <petern> Oh yeah, I was going to do more OpenTTD, but I've instead ended up replacing jquery with native javascript on my... OBS browser views... Christ.
21:35:38  <petern> (The javascript animate() works much better than the jquery animate())
21:45:33  *** Borg has quit IRC
21:46:44  <andythenorth> such!
21:46:51  <andythenorth> I set industry map colours
21:46:54  <andythenorth> and watched youtube
21:49:18  *** Wolf01 has quit IRC
21:55:02  <petern> I split all the NWidget code and then stuffed it into a branch I'll never look at.
22:04:22  *** ChoHag_ has joined #openttd
22:04:22  *** ChoHag has quit IRC
22:05:56  *** sla_ro|master has quit IRC
22:18:08  <JGR> Slightly nicer dump of where the crash is: <https://gist.github.com/JGRennison/43c47733d44e6b8a584a5ea0a6eb7518>
22:20:06  <JGR> Which is here: <https://github.com/OpenTTD/OpenTTD/blob/master/src/video/cocoa/cocoa_wnd.mm#L1020>
22:20:33  <JGR> Looks suspiciously like _focused_window is non-null but dangling/invalid
22:23:41  <petern> Window::Close() clears _focused_window if it matches `this`, but ~Window() itself doesn't do so.
22:24:44  <petern> Not sure if windows are deleted anywhere though.
22:24:51  <petern> (Other than DeleteClosedWindows())
22:36:55  *** amal[m] has joined #openttd
22:36:55  *** andythenorth[m] has joined #openttd
22:36:55  *** aperezdc__ has joined #openttd
22:36:57  *** blikjeham[m] has joined #openttd
22:36:57  *** cawal[m] has joined #openttd
22:36:57  *** christoph[m] has joined #openttd
22:36:57  *** citronbleuv[m] has joined #openttd
22:36:58  *** cjmonagle[m] has joined #openttd
22:36:58  *** lezzbehonest[m] has joined #openttd
22:36:59  *** CornsMcGowan[m] has joined #openttd
22:36:59  *** dag[m] has joined #openttd
22:36:59  *** dude[m] has joined #openttd
22:37:00  *** einar[m] has joined #openttd
22:37:01  *** elliot[m] has joined #openttd
22:37:02  *** EmeraldSnorlax[m] has joined #openttd
22:37:03  *** fiddeldibu[m] has joined #openttd
22:37:04  *** gdown has joined #openttd
22:37:04  *** freu[m] has joined #openttd
22:37:05  *** giords[m] has joined #openttd
22:37:05  *** glothit7ok[m] has joined #openttd
22:37:05  *** grag[m] has joined #openttd
22:37:05  *** gretel[m] has joined #openttd
22:37:06  *** hamstonkid[m] has joined #openttd
22:37:07  *** Heiki[m] has joined #openttd
22:37:07  *** hylshols7qui[m] has joined #openttd
22:37:07  *** igor[m]12 has joined #openttd
22:37:07  *** ist5shreawf[m] has joined #openttd
22:37:08  *** jact[m] has joined #openttd
22:37:08  *** jeeg[m] has joined #openttd
22:37:09  *** jeremy[m] has joined #openttd
22:37:11  *** joey[m]1 has joined #openttd
22:37:11  *** johanna[m] has joined #openttd
22:37:11  *** josef[m] has joined #openttd
22:37:11  *** karl[m]1234 has joined #openttd
22:37:12  *** karoline[m] has joined #openttd
22:37:13  *** khavik[m] has joined #openttd
22:37:13  *** kstar892[m] has joined #openttd
22:37:13  *** labs[m] has joined #openttd
22:37:14  *** leward[m] has joined #openttd
22:37:15  *** linda[m] has joined #openttd
22:37:15  *** luk3Z[m] has joined #openttd
22:37:15  *** magdalena[m] has joined #openttd
22:37:15  *** menelaos[m] has joined #openttd
22:37:15  *** nartir[m] has joined #openttd
22:37:15  *** natalie[m] has joined #openttd
22:37:15  *** natmac[m] has joined #openttd
22:37:17  *** NekomimiGunner18[m] has joined #openttd
22:37:17  *** nolep[m] has joined #openttd
22:37:17  *** olmvnec[m] has joined #openttd
22:37:17  *** ookfof[m] has joined #openttd
22:37:17  *** osvaldo[m] has joined #openttd
22:37:18  *** patricia[m] has joined #openttd
22:37:18  *** patrick[m] has joined #openttd
22:37:18  *** paulus[m] has joined #openttd
22:37:18  *** phil[m] has joined #openttd
22:37:19  *** philip[m]12 has joined #openttd
22:37:19  *** pothyurf[m] has joined #openttd
22:37:19  *** robert[m] has joined #openttd
22:37:20  *** rudolfs[m] has joined #openttd
22:37:22  *** shedidthedog[m] has joined #openttd
22:37:24  *** thelonelyellipsis[m] has joined #openttd
22:37:24  *** thomas[m]123 has joined #openttd
22:37:24  *** twom[m] has joined #openttd
22:37:25  *** JamesRoss[m] has joined #openttd
22:37:25  *** udo[m] has joined #openttd
22:37:26  *** vista_narvas[m] has joined #openttd
22:37:26  *** Elysianthekitsunesheher[m] has joined #openttd
22:37:27  *** YourOnlyOne has joined #openttd
22:37:27  *** yur3shmukcik[m] has joined #openttd
22:37:28  *** zzy2357[m] has joined #openttd
22:38:04  *** YourOnlyOne is now known as Guest2387
22:46:19  <glx[d]> deletion is delayed and should not happen outside DeleteClosedWindows() (the code theorically ensure that)
22:55:01  <glx[d]> and indeed _focused_window is probably not null as EditBoxInGlobalFocus() checks that
22:58:17  *** HerzogDeXtEr has quit IRC
23:54:16  <petern> Yeha

Powered by YARRSTE version: svn-trunk