IRC logs for #openttd on OFTC at 2025-06-13
⏴ go to previous day
01:14:55 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
01:14:57 *** WormnestAndroid has joined #openttd
01:14:59 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
01:15:15 *** WormnestAndroid has joined #openttd
01:49:42 *** Wormnest has joined #openttd
02:09:18 *** Wormnest has quit IRC (Quit: Leaving)
02:36:04 *** gnu_jj_ has joined #openttd
02:39:41 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:11:24 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:30:41 *** WormnestAndroid has quit IRC (Remote host closed the connection)
03:30:46 *** WormnestAndroid has joined #openttd
04:42:28 <DorpsGek> - Update: Translations from eints (by translators)
06:16:16 <_zephyris> peter1138[d]: Yeah... I'm tempted to 'flatten' the OGFX2 ones, and add a shadow, I think it'd work better overall. But for original graphics meh, like you say, too 3D
06:54:30 <peter1138[d]> Remember they have text drawn over it the station view
07:03:41 <peter1138[d]> (And that used to be the only place they were used)
07:21:25 <peter1138[d]> _glx_: That's what I would have thought, but in master it doesn't get killed.
07:42:22 *** gelignite has joined #openttd
07:48:00 *** D-HUND is now known as debdog
08:40:38 <_zephyris> peter1138[d]: Yeah, but white text... Just an idea for now anyway
08:40:45 <_zephyris> More useful is getting CI working
08:56:01 <peter1138[d]> Also we have the bit that creates an outline.
08:56:10 <peter1138[d]> Outline will include your baked in shadow.
08:58:13 <peter1138[d]> Hmm, I think ITI manages to reuse cargo icons from different climates 🙂
08:59:00 <peter1138[d]> As does IOTC, with some dupes.
09:01:51 <peter1138[d]> NewCargoes has a mix of new and original, but that set doesn't work anyway.
09:03:55 <peter1138[d]> That Limestone 😮
09:05:34 <peter1138[d]> (Also Tourists is too tall)
09:13:34 *** notluke2578 has joined #openttd
09:13:40 <notluke2578> Tourists are generally tall irl
09:31:50 <peter1138[d]> No one has reverted the wiki yet 😉
09:46:28 *** twpol has quit IRC (resistance.oftc.net synthon.oftc.net)
09:46:28 *** greeter has quit IRC (resistance.oftc.net synthon.oftc.net)
09:46:28 *** felix has quit IRC (resistance.oftc.net synthon.oftc.net)
09:48:14 *** greeter has joined #openttd
09:55:56 <peter1138[d]> Is there a Standardised Cargotype Scheme?
10:12:44 <peter1138[d]> So the U+200C characters I removed from Lithuanian got added back.
10:13:01 *** ialokin has joined #openttd
10:14:54 <_glx_> peter1138[d]: In master each Valuate is allowed 1M ops, in my branch it uses the remaining ops so first run might be very short but next one will get around max_ops from setting
10:16:26 <peter1138[d]> Yeah, so I suppose the problem is it still can't suspend in the middle of the valuator.
10:25:54 <_glx_> Yup, the Valuator call itself can be suspended, but if a Valuator call happens in another Valuator call that will fail
10:26:57 <_glx_> Most likely because of the "infinite" loop detection
10:27:56 <_glx_> Hmm but infinite loop in Valuator should be less problematic if suspendable
10:29:20 <_glx_> Maybe I can remove the check
10:43:46 *** ialokin has joined #openttd
11:08:47 <andythenorth> peter1138[d]: Yes
11:12:24 <peter1138[d]> That description for Setting Type is... very long.
11:33:44 <peter1138[d]> Pretty sure it can be shorter than that, but... translations, eh?
11:35:26 *** gelignite has joined #openttd
11:54:55 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:55:08 *** WormnestAndroid has joined #openttd
12:22:07 <_zephyris> I guess hard to spot the context to know it needs to be short from the string name
12:22:49 <_zephyris> (not having checked what the string name is)
12:42:58 <peter1138[d]> Oof, a website with a password reset field that tells you if the email is not known... naughty.
12:55:05 *** reldred has quit IRC (Quit: User went offline on Discord a while ago)
13:29:35 <cmcaine> Did another mockup for the hotkey highlighting thing
14:05:30 <peter1138[d]> Please don't use threads here.
14:05:48 <peter1138[d]> Discord is way to ephemeral, unsearchable, and not everyone is using it.
14:06:14 <peter1138[d]> Github would be a better place to do things.
14:06:35 <cmcaine> I think it looks okay now. Possible extras:
14:06:35 <cmcaine> - Include the hotkeys in the tooltip, maybe specifying whether they are global or not so that people can learn the colour association
14:06:35 <cmcaine> - maybe show global hotkeys that don't currently have a button visible (could possibly be used instead of using colour to highlight the global hotkeys)
14:06:35 <cmcaine> - where a single button has multiple actions associated with it (e.g. polyline has CTRL-A and CTRL-SHIFT-A), maybe show them both somehow?
14:06:35 <cmcaine> I think the default set of hotkeys is pretty bad, so I've been mocking this up with my own config. I think it would be good to offer a "classic" and "modern" set of hotkeys.
14:06:37 <cmcaine> Also, I've spent long enough thinking about hotkeys now that I've changed a bunch of mine and memorised them, so my energy for implementing this is waning 😅
14:09:17 <cmcaine> I'm posting to share this idea for quick feedback and motivation.
14:09:17 <cmcaine> If I actually write a patch then I'll put the mockups and stuff on github in the PR.
14:10:35 <cmcaine> peter1138[d]: Was it Blabla car? I think they do this.
14:14:53 <talltyler> I assume you've read the previous attempt at hotkeys? I don't recall what killed the project, maybe kuhnovic remembers. 🙂
14:14:53 <talltyler> Another approach you might consider, which I think would resolve the "players don't know hotkeys" and maybe scratch your itch of changing hotkey settings, is to instead create a hotkey configuration window. That one gets requested somewhat frequently.
14:15:16 <peter1138[d]> They probably also spent enough time on it to just memorise it all 😉
14:16:13 <peter1138[d]> Does Uniscribe in Wine work the same as on Windows... I'd hope so, but you never know...
14:18:21 <kuhnovic> talltyler: I closed that PR myself, I got stuck and gave up on it
14:18:58 <kuhnovic> That doesn't mean it's impossible, I just didn't feel like working on it any longer 🙂
14:19:41 *** Flygon has quit IRC (Read error: Connection reset by peer)
14:20:40 <talltyler> Right, I just forget what you discovered was harder than it appeared. 🙂
14:27:53 <peter1138[d]> Might be easier with the new string parameter system.
14:30:26 <kuhnovic> IIRC the main difficulty was trying to figure out which hotkey belonged to which button. I tried to automatically resolve it and it just didn't work. I probably just went down the wrong rabbit hole.
14:52:57 <peter1138[d]> So something in our layouter prefers splitting between digits and letters instead of at a ideographic colon or a ZWSP.
14:56:25 <peter1138[d]> (Or even a space, actually.)
15:00:12 *** Wormnest has joined #openttd
15:16:14 <peter1138[d]> Text file is `MD5 Sum码：1234567890THISISTHEAGEOFTHETRAINTHISISTHEAGEOFTHETRAINTHISISTHEA`
15:17:19 <peter1138[d]> The numbers are in a separate run to the letters, so the line break is at the pointer where the run changes.
15:18:17 <_glx_> oh I see the issue with AAAHog, it's actually bad luck (and the way I detect infinite loops), it got suspended while valuating the first item of the list
15:32:27 <michi_cc> peter1138[d]: The usual run algorithm prefers runs as long as possible I think, and numbers are mostly script-agnostic, so will be collapsed into the CJK run.
15:34:28 <peter1138[d]> Right, basically there's nothing to say we want a new run now at specific points.
15:35:12 <michi_cc> A zero-width non-joiner after the : might do something, but this is just a wild guess.
15:39:58 <peter1138[d]> Funnily enough...
15:54:54 <peter1138[d]> Game gotta be entirely deterministic...
16:17:13 <cmcaine> talltyler: Making a hotkey config window seems like a lot more work 🙂
16:17:13 <cmcaine> Yes, I found that PR before I started with any mockups or looking at the code.
16:37:32 <peter1138[d]> Same thing but on "Windows"... a.k.a Wine.
16:38:41 <peter1138[d]> So ZWSP does actually seem do the correct thing with (this) uniscribe, despite what was mentioned in #11766
18:39:09 *** youngking236048 has joined #openttd
18:42:16 <debdog> and people think I am a troll mentioning xlibre
19:50:52 <peter1138[d]> debdog: Who what?
20:12:56 <dwfreed> peter1138[d]: he's just mad he got quieted in #debian for being offtopic
20:15:13 <debdog> peter1138[d]: uhm, that was meant for another channel
20:15:36 <debdog> dwfreed: haha, good catch. except I am not mad
20:16:29 <debdog> dwfreed: just wanted to see who are the fascists in #debian
20:16:39 <dwfreed> debdog: it's offtopic here too, drop it
20:17:08 <debdog> right, as I have said, this was accidently
20:17:29 <debdog> also, drop waht? dwfreed
20:17:35 <dwfreed> then why are you continuing it? shut up, or I apply the network banhammer
20:17:54 <debdog> I just saiod, it was an accident
20:18:19 *** debdog has quit IRC (autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2025-06-13 20:18:19))
20:19:00 <LordAro> i love when someone unknowingly argues with an admin
20:20:00 <_glx_> never a good idea to not listen to ircops
20:27:08 <andythenorth> weird interlude 🙂
20:30:44 <dwfreed> debdog: disconnect now, or I ban you permanently for ban evasion; only warning, don't reply, don't argue; /quit, or get banned
20:32:55 <debdog> dwfreed: what have I done wrong?
20:32:57 *** debdog has quit IRC (autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2025-06-13 20:32:57))
21:02:07 <peter1138[d]> Just use some random unicode characters if you want a ban....
21:30:56 <michi_cc> andythenorth: ⏫ Something for you maybe 🙂
21:36:10 *** debdog has quit IRC (autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2025-06-13 21:36:10))
21:38:37 <andythenorth> michi_cc: I can't test tonight, but that might solve a lot of the compatibility angst in the grf dev channel 😄
21:40:46 <michi_cc> Yes, that is the reason for this. Needing a million (potentially hidden) railtypes to make something like a 4-system electrification locomotive is just stupid. Some mixed types do make sense to build, but most do not.
21:55:23 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:01:30 *** ChanServ sets mode: +v tokai
22:07:24 <talltyler> It’s good to see some out-of-the-box thinking on this 🙂
22:08:24 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
22:09:31 <michi_cc> I don't consider it that much out-of-the-box, more like something NewGRF authors should have been feature requesting since 10 years ago.
22:09:59 <andythenorth> that nml fallback feature obscured the spec too much for them to understand it
22:10:09 <andythenorth> best of intentions
22:10:15 <michi_cc> And speaking of features, the multi-hold ship thingy would be something nice to steal from JGRpp 🙂
22:15:56 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:18:59 <peter1138[d]> If it's done in a way that allows the same for all transport types, sure.
22:37:08 <michi_cc> Road and rail have some kind of multi-cargo transport way already (articulation and extra wagons), and aircraft have a little bit of multi-cargo support.
22:37:45 <michi_cc> I haven't looked at the actual JGRpp implementation though. I like the idea, but no idea if the code is sound or not.
22:42:14 *** WormnestAndroid has quit IRC (Remote host closed the connection)
22:42:15 *** WormnestAndroid has joined #openttd
23:03:53 <_glx_> michi_cc: didn't check, but I think it's some kind of articulated
23:09:11 <_jgr_> Yes, I'm using callback 16, it didn't seem worth inventing something radically different
23:12:19 <peter1138[d]> michi_cc: Yes, but it causes other effects.
23:14:10 <peter1138[d]> A multi-cargo road vehicle can't use non-drive-through bays, for instance.
23:21:27 *** reldred has joined #openttd
23:21:27 <reldred> I honestly forget non drive through stops even exist
23:21:46 <reldred> Been that long since I’ve used them in anger
23:39:38 <michi_cc> peter1138[d]: Well, maybe `Vehicle::cargo_type` (and related vars) need to become a vector and be filled by either a static prop or a new callback.
23:40:32 <michi_cc> Or alternative articulated parts get a hidden flag that completely removes them from any drawing or movement related calculations.
23:52:00 <peter1138[d]> I don't know really, other than being locked into doing things a certain due to a 20 year old spec...
23:56:01 <cmcaine> I wanna draw some text on the screen. What functions should I be looking for?
23:56:01 <cmcaine> - easy way to outline the characters
23:56:01 <cmcaine> - easy way to change the colour of the text
23:57:31 <peter1138[d]> DrawString would be a place to start.
23:58:03 <peter1138[d]> Colour, mostly, there's a few text colours available.
23:58:25 <peter1138[d]> Size, there is only the normal medium font and the small font.
continue to next day ⏵
