IRC logs for #openttd on OFTC at 2024-05-18
⏴ go to previous day
00:12:36 <_glx_> it's a parameter to add in the config
00:18:14 <silent_tempest> Yeah I did thart
02:21:04 *** Wormnest has quit IRC (Quit: Leaving)
02:43:35 *** APTX has quit IRC (Quit: No Ping reply in 180 seconds.)
02:46:10 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:52:49 *** APTX has quit IRC (Remote host closed the connection)
03:31:51 *** Flygon has quit IRC (Remote host closed the connection)
04:40:33 <DorpsGek> - Update: Translations from eints (by translators)
05:19:45 *** HerzogDeXtEr has joined #openttd
06:17:02 <andythenorth> peter1138: I’ll be afk this morning but can later
06:43:27 *** casemate_crusader has joined #openttd
06:43:27 <casemate_crusader> peter1138: i mean i gave a mac
08:30:47 <truebrain> peter1138: you didn't receive it? 😛
08:33:29 <peter1138> I did once, but somehow it managed to die after about a week.
08:46:32 *** mindlesstux has joined #openttd
09:22:23 <kuhnovic> Hmm, I think I've found a way to use std::priority_queue for A* pathfinding
09:23:15 <kuhnovic> ... and still have the ability to update nodes that are not at the top. Need to try it out and benchmark it.
09:56:24 <peter1138> MacBook air M1, £615, oof
09:58:49 <peter1138> Stinks of BO in there
10:10:30 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
10:15:28 *** SigHunter has joined #openttd
10:30:17 *** mindlesstux has quit IRC (Ping timeout: 480 seconds)
10:34:53 *** mindlesstux has joined #openttd
11:04:27 <peter1138> I did not buy the second hand M1 mac from the CEX that smells of B.O.
11:35:31 <andythenorth> can buy a ropey used one on Back Market for £510
11:35:48 <peter1138> That's basically CEX yeah.
11:36:01 <andythenorth> and brand new warrantied is £799 on Amazon, so £615 no
11:36:33 <peter1138> 409 for a M1 Mac Mini on Amazon? Hmm.
11:38:32 <andythenorth> allegedly Asahi Linux is usable on them now
11:50:40 <peter1138> That defeats the point 🙂
11:51:08 <andythenorth> will I have to be your macOS IT support? 😛
12:00:44 <andythenorth> "try turning it off and on again"
12:04:55 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
12:56:41 <klote> andythenorth: Indian accent
13:11:27 *** SigHunter has joined #openttd
13:58:38 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
14:01:03 <peter1138> Yup, noticed that too 🙂
14:02:56 <_glx_> it's funny because ResolverObject is a reference for more than 10 years
14:14:33 <andythenorth> when do we make grfs player-editable?
14:14:44 <andythenorth> endless requests for turning vehicles on and off
14:14:50 <andythenorth> changing costs, capacity etc
14:15:08 <_glx_> get the source and rebuild 😉
14:40:15 <peter1138> They are actually blank when placed on the map, too.
15:28:46 <peter1138> Hmm, maybe they don't appear in game.
15:29:18 *** Wormnest has joined #openttd
15:47:15 *** yeah_its_gloria has joined #openttd
15:47:15 <yeah_its_gloria> hey uh I've been trying to find existing reports about the macOS version of OpenTTD failing to load social plugins properly but so far I've found nothing and I'm not entirely sure who to talk to where to put this soooooo
15:47:15 <yeah_its_gloria> the app fails to boot like this
15:47:15 <yeah_its_gloria> [2024-05-18 11:40:44] dbg: [driver:0] Could not change to foreground application. Error -50
15:47:15 <yeah_its_gloria> dlopen failed trying to load: [...]/steamclient.dylib
15:47:17 <yeah_its_gloria> dlopen([...]/steamclient.dylib, 0x0002): tried: '[...]/steamclient.dylib' (code signature in [...] '[...]/steamclient.dylib' not valid for use in process: mapping process and mapped file (non-platform) have different Team IDs), [more of the same but for different paths])
15:47:17 <yeah_its_gloria> [S_API] SteamAPI_Init(): Failed to load module '[...]/steamclient.dylib'
15:47:19 <yeah_its_gloria> [2024-05-18 11:40:44] dbg: [misc:0] [Social Integration: steam/libsteam-social.dylib] Failed to initialize
16:18:57 *** gelignite has joined #openttd
16:39:00 <yeah_its_gloria> guess that problem is now fixed
16:43:11 <_glx_> we have a template for pull requests
16:47:24 <_glx_> and your previous message could definitely go in "motivation/problem" section of the template 🙂
17:09:02 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
17:11:21 <andythenorth> 15 minutes trying to write a script to modify toml
17:11:29 <andythenorth> GPT got it right first time in 10s
17:13:10 <dwfreed> that says more about you than GPT :P
17:13:36 <andythenorth> it says I needed to use GPT straight away
17:15:53 <yeah_its_gloria> the oopses are intensifying
17:33:24 <_glx_> indeed we don't package it in steam because it fails to load 🙂
17:36:41 *** softinterlingua_9093 has joined #openttd
17:36:41 <softinterlingua_9093> Hello. How i can request to add a language in OpenTTD Translator? I'm interested to translate OpenTTD in Interlingua (langcode: ia)
17:36:45 <_glx_> triggered a build for the PR so we can have a signed binary for testing
17:49:55 <yeah_its_gloria> _glx_: right should've figured
19:02:19 *** salut3585 has joined #openttd
19:02:55 <salut3585> can you approve this?
19:03:36 <peter1138> Did you mean review?
19:04:25 <salut3585> yes, the issues was fixed and need toapprove again
19:13:29 <truebrain> we had to learn this the hard way btw 😄
19:22:33 <yeah_its_gloria> truebrain: with all due respect, my sibling in christ, it removes a printf
19:23:06 <yeah_its_gloria> I'm willing to wager we'll watch the sun explode before this causes problems, and I did consider whether this would cause problems
19:23:10 <truebrain> what a weird reply ....
19:23:18 <yeah_its_gloria> I included it in the commit for that reason
19:23:42 <yeah_its_gloria> since OpenTTD also doesn't use restorable state (yet? if it did it'd need to use NSSecureCoder anyway)
19:23:59 <truebrain> I understand that to you it might seem like a small thing, but we have been around for a lot longer maintaining this code base. So I just simply request you to put this in a small PR. Small effort, and it makes us happy
19:24:41 <truebrain> feels a bit needless to reply in this way to such a simple request ...
19:24:59 <yeah_its_gloria> because it's needless? you're just kinda flexing your experience at me
19:25:05 <yeah_its_gloria> but sure have it your way I'll just need a moment to add the rest
19:25:58 <truebrain> This attitude really isn't helping anyone, and I am not sure what I did to deserve it. Either way, tnx for fixing it.
19:27:26 <peter1138> Standard practice to put unrelated things in different PRs, it doesn't really matter how small it is.
19:27:53 <peter1138> In fact, the smaller the better
19:29:39 <yeah_its_gloria> if I called this "fixing small bugs on macOS" instead of focussing on social plugins primarily would this be any better
19:30:20 <truebrain> No. We have a very simple policy: unrelated changes should go in different PRs. PRs are cheap, PRs are free.
19:30:28 <yeah_its_gloria> can I keep the change in this PR but put the state related thing in a second commit
19:30:52 <truebrain> given the motivations of the lines of code to exist are totally different, I would advise you not to
19:31:43 <truebrain> Gives you also a bit of room to indicate why Apple is asking for these lines to exist
19:31:53 <truebrain> as the documentation around that function is like really poor 😦
19:32:01 <truebrain> (on Apple's side, that is)
19:32:27 <yeah_its_gloria> I mean you have to use AppKit directly to have this behavior so I guess that's why you don't have much documentation for it
19:32:39 <yeah_its_gloria> and they don't like when you try using AppKit directly particularly now I guess
19:32:48 <yeah_its_gloria> still sucks but meh
19:32:56 <truebrain> that alone would be valuable information to have in the description 🙂
19:32:58 <peter1138> Better to document it in the code than in a PR comment.
19:33:25 <peter1138> Or at least a brief summary.
19:33:42 <truebrain> Just to paint you a picture, none of the current developers have owned a Mac in ... 5 years? 10? So we build this all by hopes and prayers ....
19:34:18 <peter1138> I already tried cleaning up some defines that said they were related to MinGW and Windows 98... that did not work well 😄
19:35:51 <yeah_its_gloria> truebrain: is anything aside from the issue tracker to see what other problems there might be
19:35:59 <yeah_its_gloria> might as well make myself more useful than fixing the two bugs that bothered me
19:36:10 <truebrain> we tend to tag MacOS related bugs with that label
19:36:28 <truebrain> but mostly I am pretty sure we use very old APIs in some places, that nobody dares to fix 😛
19:36:53 <truebrain> and yeah, and help with MacOS is most welcome
19:37:16 <truebrain> and as you might have understood by now, a clear motivation/description on why this fixes that, would be much appreciated, as .. you can tell us anything, basically.
19:39:02 <peter1138> #12662 is a mac os fix that could do with testing 🙂
19:39:41 <yeah_its_gloria> truebrain: I could tell you anything? what do you mean?
19:39:55 <_glx_> yeah_its_gloria: oh fixing what bothers you is always a good motivation 🙂
19:40:21 <yeah_its_gloria> it's fun to play as is but I could make it more Fun or whatever
19:40:32 <truebrain> yeah_its_gloria: it is a figure of speech that it is very likely you know more about any of this than most of us, so if you say: this is the right solution, we will be like: okay. That is why documentation and motivation is important
19:41:10 <_glx_> IIRC one of my first patch (way before github) was to stop the annoying disk drive access
19:42:02 <peter1138> > Fix: Remove floppy disk drive
19:44:51 <truebrain> at least your first commit is recorded 😛
19:46:10 <_glx_> oh at that time we were still using sourceforge "tracker"
19:47:51 <_glx_> lol it half fixed an if/else style
19:49:34 <truebrain> nice (and good) description yeah_its_gloria , tnx!
19:52:44 <_glx_> pff I need to fix some workflows
20:02:06 <yeah_its_gloria> truebrain: also if you don't mind me asking, could you re-review the social plugins thingy? I made all the changes™️
20:02:40 <truebrain> did you test glx' binary?
20:05:26 <yeah_its_gloria> it's taking a moment
20:05:43 <yeah_its_gloria> Finder isn't responding
20:05:58 <yeah_its_gloria> oh oops I was compiling other code
20:06:29 <yeah_its_gloria> time to see if Steam works
20:06:31 <yeah_its_gloria> pun unintended
20:07:12 <truebrain> I hate that Apple's / Steam's only solution is to disable security
20:07:19 <truebrain> as weirdly enough, we do cross-sign another of their libraries
20:07:34 <truebrain> not sure why they don't allow us to cross-sign steamclient.dylib .. I guess it depends on the Steam version
20:08:04 <_glx_> well steam is basically mitm 🙂
20:08:07 <yeah_its_gloria> truebrain: it doesn't even need to be cross signing, it's awkward that it rejects *any* dylib
20:08:19 <yeah_its_gloria> you'd think checking for a signed release binary would, you know, be fine
20:08:25 <yeah_its_gloria> because it does this for all Apple signed system frameworks
20:09:03 <_glx_> to test steam plugin you need to start openttd from steam
20:10:10 <yeah_its_gloria> can't wait for the OpenTTD unapproved social integration to do all sorts of code injection shenangians
20:10:31 <yeah_its_gloria> evil integration
20:10:32 <_glx_> we don't allow unapproved plugins
20:10:47 <yeah_its_gloria> aw we already solved this problem?
20:10:56 <peter1138> So did this work initially? I'm sure someone was testing this... 😒
20:11:13 <yeah_its_gloria> it'd always say failed to initialize for me at least
20:11:23 <_glx_> it worked on older macs
20:11:24 <truebrain> peter1138: it did, but for older MacOS 🙂
20:11:36 <yeah_its_gloria> I'm the weirdo running it on an M2 Air with Sonoma
20:11:42 <truebrain> it was all tested and working when I delivered it 😉
20:12:07 <_glx_> testing was hard enough to do
20:12:20 <_glx_> requires finding someone with a mac
20:12:24 <yeah_its_gloria> *the* macOS thing to do now would be like a Metal renderer right
20:12:54 <_glx_> yeah and once you implement Metal support they deprecate it 😉
20:13:18 <yeah_its_gloria> don't jinx it already
20:13:26 <yeah_its_gloria> what would they even call the follow up
20:13:39 <peter1138> Do Macs have some kind of remote desktop functionality?
20:13:41 <yeah_its_gloria> hello dear applers. we made the ipad m6. it runs Steel for your epic games
20:13:56 <yeah_its_gloria> it does have built in VNC support
20:14:06 <yeah_its_gloria> I dunno maybe you can like
20:14:07 <andythenorth> it used to just work
20:14:13 <yeah_its_gloria> make a kilometer long thunderbolt cable
20:14:14 <andythenorth> I used to use VNC with Windows and macs
20:14:15 <yeah_its_gloria> and use sidecar with an ipad
20:14:28 <andythenorth> peter1138: from Linux?
20:14:44 <peter1138> Using a Mac without having to actually use it ;D
20:14:59 <andythenorth> can't you just ssh?
20:15:14 <peter1138> Not great for testing UI changes.
20:15:17 <_glx_> doesn't help to see how the game looks like
20:15:47 <peter1138> Like, did I break truetype font rendering in #12662?
20:15:52 <yeah_its_gloria> what if we simply have people ready with Macs to test it
20:16:21 <yeah_its_gloria> last time I caused a big commotion with social plugins I think I owe something
20:16:50 <peter1138> andy just tends to assume problems are part of the design :p
20:17:09 <_glx_> andy is very good to break things 🙂
20:17:11 <peter1138> "It's always been that way"
20:17:23 <yeah_its_gloria> something about omelettes and eggs I'm sure
20:18:20 <_glx_> newgrf authors mindset, just work around issues instead of reporting them
20:19:12 <andythenorth> it's hard to report OpenTTD bugs
20:19:16 <andythenorth> the rules are 'no patches'
20:19:21 <andythenorth> 'no unreleased grfs'
20:19:26 <andythenorth> 'no changing grfs in the save'
20:19:40 <andythenorth> so then I have to go get a repro on a clean build
20:21:59 <_glx_> unreleased grf is not an issue, if you provide it in the report
20:22:06 <audigex> I legit don't think I've ever found something broken when making GRFs with NML, at least in game
20:22:06 <audigex> Although I'm having a fun one now... my code builds fine locally, but the GitHub automated build throws this error
20:23:11 <audigex> And when I build locally, that line isn't even 1090...
20:23:35 <_glx_> custom nml version locally ?
20:23:53 <audigex> _glx_: Nope my git repo is 100% up to date
20:23:53 <audigex> NML locally is standard I think? I'd have to check
20:24:47 <audigex> The only thing I can think of is that my local build is Windows and Github builds on Linux and there's something different about Python's folder traversal
20:24:47 <audigex> I once had a similar issue with case sensitivitity IIRC
20:25:10 <_glx_> linux is case sensitive
20:26:14 <audigex> Yeah, that was an issue once where I'd renamed a folder but Git still considered some to be in /emu and others to be in /EMU
20:26:14 <audigex> Shouldn't be the problem here though
20:26:39 <audigex> In my local version, the switch in question isn't even called until line 6552, so I'm really not sure why it's causing a problem at like 1090
20:26:43 <_glx_> and all relevant local files are actually added to git
20:27:17 <audigex> Yeah, it's been building absolutely fine for 2 years, except when I added this specific vehicle. All files relating to that vehicle are in Git and pushed to Github
20:28:56 <_glx_> well you might need to modify your action to put the merged nml in artifacts
20:29:04 <_glx_> so you can compare them
20:29:09 <audigex> I assume it's something to do with whatever the fuck is happening here...
20:30:13 <peter1138> That's sub-directory with nothing else in it.
20:30:31 <_glx_> oh Departemental comes first in directory list I think
20:30:39 <audigex> Wrong location actually
20:30:56 <audigex> Oh, so Departmental/ is empty and MPV isn't, so it shows them together?
20:31:25 <peter1138> No, Departmental contains MPV, but nothing else.
20:31:52 <audigex> Departmental is empty other than MPV, so it groups them into one line rather than making me expand both, I see
20:32:04 <peter1138> But I don't know what tool that is on the left.
20:32:17 <audigex> _glx_: How do I do that?
20:32:20 <peter1138> I know VS code do that in its file listing.
20:32:30 <_glx_> anyway if your nml does the includes using wildcards you'll get different order on linux and windows
20:33:33 <audigex> That's possible but sprites and switches are in _spritesets and that's never been a problem for any other unit? So seems unlikely to be the problem here
20:34:00 <audigex> Unless capitalisation matters for ordering maybe? Seems odd but I can try that
20:34:18 <_glx_> it does, D and d are not in the same place
20:34:44 <audigex> So it could be producing:
20:34:47 <peter1138> Life it easier when all filenames are lowercase 😄
20:35:07 <peter1138> Why can I not type any more? 😦
20:35:07 <_glx_> yes that's very possible
20:35:45 <peter1138> Yeah, _ comes between D and d.
20:35:57 <audigex> It seems very plausible then, I hadn't realised I'd done that folder differently... one minute, let me push to Github
20:36:12 <audigex> It's not often I say this but: Linux is dumb. Why would _ come between A-Z and a-z?
20:36:28 <audigex> Aight, ASCII is now my mortal enemy
20:37:01 <_glx_> that's why filenames are usually lowercase
20:37:37 <peter1138> Usual way of forcing order is, well, list explicitly instead of wildcard, or prefix with with two numeric digits.
20:37:39 <audigex> yeah I generally stick with one case system per project, dunno why I used D this time
20:37:49 <peter1138> 00_spritesets, 10_diesel
20:38:04 <audigex> Why two numeric digits not 1?
20:38:44 <peter1138> And so you can stuff things beween 1 and 2 without having to renumber. 10 -> 15 -> 20
20:39:25 <_glx_> and sorting is not reading the number, it's just ascii 😉
20:40:54 <audigex> peter1138: BASIC PTSD intensifies
21:02:57 <andythenorth> should curve_speed_multiplier default to 0 or 1?
21:03:37 <andythenorth> nfo docs suggest 0
21:03:41 <andythenorth> nml docs aren't clear
21:14:08 *** HerzogDeXtEr has joined #openttd
21:40:57 <peter1138> I could just assume it works.
21:46:43 <andythenorth> what will happen if it doesn't?
22:33:31 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:30:45 <yeah_its_gloria> I've been looking into #10044 and like
23:30:51 <yeah_its_gloria> this bug is related to cursor speed somehow???
23:31:03 <yeah_its_gloria> it only happens if the cursor moves between the Dock or menu bar fast
23:38:46 <yeah_its_gloria> also sometimes the window is stuck being unactivated
23:38:51 <yeah_its_gloria> why is Cocoa so weird sometimes
23:40:11 <yeah_its_gloria> and sometimes it changes later
23:40:12 <yeah_its_gloria> what on earth
23:42:10 <yeah_its_gloria> ok I fixed the bug
23:45:47 <yeah_its_gloria> git blame proves useless
23:53:02 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
continue to next day ⏵
