IRC logs for #openttd on OFTC at 2025-05-19
⏴ go to previous day
01:39:49 *** tokai|noir has joined #openttd
01:39:49 *** ChanServ sets mode: +v tokai|noir
01:46:45 *** tokai has quit IRC (Ping timeout: 480 seconds)
02:14:54 *** Wormnest has quit IRC (Quit: Leaving)
02:40:30 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:40:33 *** WormnestAndroid has joined #openttd
02:40:35 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:40:36 *** WormnestAndroid has joined #openttd
02:40:48 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:40:49 *** WormnestAndroid has joined #openttd
03:00:35 *** gnu_jj_ has quit IRC (Ping timeout: 480 seconds)
07:05:14 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
07:05:43 *** WormnestAndroid has joined #openttd
07:10:47 <pickpacket> andythenorth: moin
07:11:12 <pickpacket> github tells me that nightly workflows for my openttd fork will be disabled soon. I didn't even know I had them
07:18:31 <DorpsGek> - Add: summary for week 20 of 2025 (by OpenTTD Survey)
08:54:43 *** Flygon has quit IRC (Read error: Connection reset by peer)
09:51:29 <xarick> weird issues with vehicle ids in regression results
09:51:44 <xarick> some effect vehicle probably took an id
09:52:55 <xarick> I did not touch this part
09:53:50 <xarick> but now I have to touch it
09:53:58 <xarick> change those 20's to 21
09:54:29 <xarick> totally unrelated to my tests
09:58:46 *** michi_cc_ has joined #openttd
10:01:59 <peter1138[d]> So timing changes means the ID changes.
10:04:01 <peter1138[d]> Could you place your own changes further down, after this stuff?
10:05:43 <xarick> erm... I mean, I could but...
10:06:33 <xarick> I placed my changes in the appropriate class method tests
10:06:44 <xarick> they're neatly organized
10:10:07 <peter1138[d]> Yeah, so probably not.
10:11:09 <peter1138[d]> Are you able to verify that it's an effect vehicle and not some other change that shouldn't be occuring?
10:11:54 <peter1138[d]> (Probably the easiest way to do that is to modify OpenTTD to temporarily add a message outputting the ID when a new effect vehicle is created)
10:12:45 <xarick> gonna verify what it is
10:13:35 <peter1138[d]> If it is the case, then you can update the unit tests for it, and explain in the PR itself why the unit test changed 🙂
10:14:49 <xarick> these are the tests I added
10:14:59 <xarick> I didn't buy any vehicle
10:25:56 <peter1138[d]> Doesn't matter, any build command will suspend the script until the next tick. And that will allow effect vehicles to be created by the game.
10:26:24 <peter1138[d]> So it's not an issue with your changes, it's just an artifact of how the game works.
10:27:25 <xarick> i'm getting a vehicle_id 20 being a ship, testing refit capacities
10:27:43 <xarick> this refit part happens before
10:27:56 <xarick> ship is sold at the end of refit tests i guess
10:29:14 <xarick> yeah, there's a SellVehicle
10:29:22 <_glx_> You added 11 ticks delay
10:29:42 <peter1138[d]> The effect vehicle could be before that 🙂
10:30:52 <_glx_> All tests depending on fixed ID have be updated
10:33:20 <_glx_> xarick: If the built vehicle is now 21, then you need to update the order commands to work with 21
10:33:42 <xarick> vehicle 20 is an animation of some kind
10:35:49 <peter1138[d]> Yup, now you have successfully confirmed that's what it is, not some other unintended change. So just update the IDs.
10:42:23 <xarick> i wanted to know what animation was it
10:43:09 <peter1138[d]> Basically irrelevant.
10:43:21 <peter1138[d]> But they do have x_pos and y_pos set.
10:47:57 <peter1138[d]> Some of these things misuse tile as a target identifier.
10:48:21 <peter1138[d]> Which is funny, because they are still in the vehicle tile hash, but with a tile that isn't really a tile.
11:05:12 <xarick> oh gosh, what a joke, now later on, on the next commit, vehicle is 20 again
11:06:02 <peter1138[d]> So the effect vehicle is deleted 🙂
11:06:49 <xarick> multiple road waypoints tests per commit, is hard!
11:15:17 <peter1138[d]> If you re-order the commits, maybe the change is nullified.
11:29:35 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:29:42 *** WormnestAndroid has joined #openttd
12:23:59 <xarick> which station functions are you referring to
12:33:06 <xarick> interesting, I wonder what it returns on a drivethrough station (not waypoint)
12:42:39 *** reldred has quit IRC (Quit: User went offline on Discord a while ago)
12:47:08 <xarick> too bad 33411 is a normal station 😦
12:55:32 <xarick> should I add a test for IsRoadStationTile on a Drivethrough station?
12:57:14 <xarick> it returns true, not sure if I agree with it
12:59:13 <xarick> I made no changes to IsRoadStationTile at all, but then again, it was never tested against a drivethrough in regression before
12:59:52 <xarick> not sure if it warrants a fix...
13:00:26 <peter1138[d]> Are you suggesting that a drive through station is not a road station tile?
13:00:51 <xarick> yes, since there's IsDriveThroughRoadStationTile
13:01:17 <LordAro> what if i told you things can have more than one property?
13:01:35 <peter1138[d]> A drive through road station tile is still a road station tile.
13:02:27 <xarick> I see, might be too late to make changes now
13:02:39 <LordAro> also those changes would be wrong
13:02:46 <peter1138[d]> Bah, this CD won't rip with whipper as it's got a read error near the end. Sometimes it'd be nice to have "a rip" even if it's not a 100% accurate rip, right.
13:10:01 <xarick> I'm creating a PR to add the test, just for the sake of completeness, not going to change behaviour
13:28:54 <xarick> probably fine to return true
13:54:22 <xarick> Bribe has a ~93% chance to succeed
13:54:50 <xarick> 93% chance to deny exclusive transport rights of someone else
13:55:32 <xarick> in AI wars, that's kind of cheap
14:16:55 <xarick> meh... frosch closed some stuff
14:19:26 <peter1138[d]> Yes, and reasons were given.
14:20:34 <xarick> not sure I rebase those that are closed
14:21:43 <peter1138[d]> Correct, there is no point in doing so.
14:57:13 <peter1138[d]> Ok, raw cd-paranoia managed to grab it, inaccurately, but better than nowt.
15:09:01 <LordAro> "Through the magic of knowing what I'm doing, I have determined what you're talking about"
15:09:08 <LordAro> is this professional enough to say to a colleague?
15:09:19 <peter1138[d]> Probably the same issues as the other rejected PRs.
15:09:27 <peter1138[d]> LordAro, well...
15:09:48 <LordAro> How about, "Through the magic of being good at my job" ?
15:12:08 <peter1138[d]> I would suggest something invoking your experience...
15:12:35 <kuhnovic> "I've been it this for 30 years so..." 😛
15:12:52 <andythenorth> "GPT suggested that..."
15:12:56 <LordAro> with some prompting they sent a picture that gave the necessary additional context
15:13:24 <peter1138[d]> It wasn't one of Xarick's PRs then...
15:29:46 <LordAro> me: "you should probably do something about this too <picture>" colleague: "non-matching response" me/colleague: <follow up comments> colleague: "oh yeah, the image didn't load yet"
15:32:08 <peter1138[d]> Time for another 200km ride.
15:47:26 *** Wormnest has joined #openttd
16:00:40 <xarick> `local x = array(99999999)` let's test this
16:03:50 <xarick> nevermind, i didn't save main.nut
16:08:55 <xarick> ignored the assert, it didn't crash openttd
16:12:11 *** foodliker has quit IRC (Quit: User went offline on Discord a while ago)
16:44:24 *** gelignite has joined #openttd
17:38:06 <xarick> is this too many settings?
17:38:56 *** belajalilija has joined #openttd
17:43:01 <kuhnovic> If you ask yourself "would I ever change this setting" and the answer is "no" or "most likely not", just remove it
17:45:22 <xarick> think i'm gonna get rid of those load orders modes
17:45:39 <xarick> they're a pain to manage
17:46:26 <xarick> and also I have further ideas regarding load orders, I just didn't bother changing them yet
17:47:14 <xarick> cargodist also ruins them
17:51:25 <xarick> I'm also waiting for unbunch for scripts
17:51:40 <xarick> if it ever gets implementefd
17:54:31 *** akimoto has joined #openttd
17:54:41 <_glx_> these load orders looks like self infliged pain 🙂
17:55:34 <_glx_> just let the AI decide by itself what is best depending on actual game settings
17:59:20 <andythenorth> did I break it? 😮
18:00:10 <peter1138[d]> No, just rebased after merging #14280.
18:12:09 *** akimoto has quit IRC (Ping timeout: 480 seconds)
18:33:52 <andythenorth> OpenTTD is finished after 14281? 🙂
19:50:25 *** tokai|noir has quit IRC (Quit: c('~' )o)
20:19:44 <xarick> i keep getting crashes on loading my AI
20:19:56 <xarick> due to the horrible timely manner things are saved
20:21:17 <xarick> it expected a dock at the tile
20:22:05 <xarick> dock isn't there, it was being removed, but the member variable wasn't updated on time to reflect no dock
20:25:44 <xarick> is it possible to set some flag that the Save function can check to prevent it from saving before time?
20:27:01 <xarick> something like... This operation must be done before saving!
20:27:43 *** debdog has quit IRC (Quit: Initiating getting-the-hell-out-of-here maneuver!)
20:35:48 <peter1138[d]> When do you update the local data?
20:36:20 <peter1138[d]> Sounds like your AI will break horrible if the player joins the company 🙂
20:43:09 <andythenorth> oops I've broken company colours
20:43:15 <andythenorth> 'mistakes were made'
20:47:52 <xarick> okay, I managed to salvage this crash
20:48:09 <xarick> if (this.m_dock_to == -1 || (!AIMarine.IsDockTile(this.m_dock_to) || AITile.GetOwner(this.m_dock_to) != ::caches.m_my_company_id)) {
20:48:51 <peter1138[d]> You managed to invoke the spoiler feature.
20:49:03 <peter1138[d]> Wrap code in backticks to avoid.
20:50:00 <peter1138[d]> `if (this.m_dock_to == -1 || (!AIMarine.IsDockTile(this.m_dock_to) || AITile.GetOwner(this.m_dock_to) != ::caches.m_my_company_id)) {`
21:00:36 <andythenorth> maybe I should just finish all the unfinished trains in the other branch, and ignore the brake vans branch
21:03:49 *** treky97 has joined #openttd
21:03:49 <treky97> Hey all sorry to bother but my game just crashed and I figured for once I'd actually be useful and send in the .json.log, .dmp, .sav, and .png. The error message says to send it to the github but I'm not entirely sure where within the github or how. Any advice would be great, thanks in advance
21:09:35 <treky97> Just post it as an issue?
21:09:45 <xarick> got my AIs fighting over transport rights kek
21:13:28 <xarick> i should be more verbose about successful bribes and failed bribes
21:13:47 <xarick> a thing for tomorrow, gotta go sleep, cyas good night
21:14:42 <peter1138[d]> All your logging probably slows your script down :p
21:25:54 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:30:34 *** nielsm has quit IRC (Ping timeout: 480 seconds)
22:15:07 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
continue to next day ⏵
