IRC logs for #openttd on OFTC at 2026-10-19
⏴ go to previous day
00:04:11 <brickblock19280> I don't see it being more complicated honestly
00:05:50 <brickblock19280> nearby_tile_is_water should be enough to cover that usecase
01:59:49 *** MinchinWeb[m] has quit IRC (Remote host closed the connection)
02:00:06 *** MinchinWeb[m] has joined #openttd
03:57:58 *** Philemon has joined #openttd
04:01:20 *** Phileman has quit IRC (Ping timeout: 480 seconds)
04:20:39 *** MinchinWeb[m] has quit IRC (Remote host closed the connection)
04:21:03 *** MinchinWeb[m] has joined #openttd
05:00:48 *** tokai|noir has joined #openttd
05:00:48 *** ChanServ sets mode: +v tokai|noir
05:08:03 *** tokai has quit IRC (Ping timeout: 480 seconds)
07:03:14 <kuhnovic> IMO sea planes should be able to move like a ship (and use regular ship docks), but can also take off into the sky given enough room to do so. This would require a straight line of X water tiles, where X could be made newgrf-configurable. After takeoff it would move like any other airplane would, and could even land on regular airports. And implementing this would probably be a total nightmare.
07:07:34 *** skee has quit IRC (Read error: Connection reset by peer)
07:25:22 <peter1138> Just flip v->type ;-)
07:46:03 * peter1138 changes variations around a bit.
08:08:15 <kuhnovic> Sprite variations? 👀
08:09:46 <reldred> rock variations would be nice, iirc zeph had a few candidate shapes for ogfx2 that would work well as tile variations to break up the sameyness of it
08:13:00 <peter1138> The PR is already there. I'm just changing the layout a bit to make it more consistent.
08:13:53 <kuhnovic> Are you going to un-draft it 😄 ?
08:14:26 <peter1138> Is it un-hacky enough? :D
08:16:50 <peter1138> "Change: Fis" still irks ;)
08:17:27 <peter1138> "Change: Improve UI when starting scenario editor"
08:18:00 <kuhnovic> The action code could use some deduplication. The actual variation mechanism looks quite clean IMO.
08:20:00 <kuhnovic> You could argue that it's specific to tileindex-based sprites. Why can't we have vehicle sprite variations for example? But that can open up a whole new can of worms. Definitely a separate PR.
08:32:37 <peter1138> Well yes, I've been alluding to that. What other variations could we do.
08:33:05 <peter1138> That opens up the question of how do we specify them, and what if they're combinatorial?
08:33:16 <peter1138> Plus of course anything NewGRF can usually just do it by itself.
08:33:44 <peter1138> It's worth thinking about up front, as it may affect the NewGRF layout.
08:35:31 <peter1138> There isn't a lot of code duplication though, it's similar but different enough to not be worth deduplicating imho.
08:40:18 <peter1138> Although probably the variation code should be in spritecache.cpp rather than viewport.cpp
08:45:27 *** toktik has quit IRC (Remote host closed the connection)
08:50:42 <kuhnovic> Yeah that makes sense
08:51:17 <kuhnovic> I'm mostly thinking about the baseset. As you say, newgrfs have a lot of control already
09:58:59 <peter1138> It can't be something that changes without a refresh. For a tile, the tile hash never changes.
10:00:00 <peter1138> Altitude could be considered but I don't think that's hugely useful. NewGRFs can already do that.
10:05:06 *** toktik has quit IRC (Remote host closed the connection)
10:09:59 <DorpsGek> - Update: Translations from eints (by translators)
10:18:45 <kuhnovic> What do you mean by "without a refresh"?
10:48:49 <LordAro> D:/a/***/***/src/lang/danish.txt (1368): warning: Unwanted UTF-8 character U+200B in sequence 'Kun til slutningen af linjen' (FATAL)
10:49:04 <LordAro> eints/strgen need tweaking?
11:44:26 <peter1138[d]> Why do they (translators) keep (very rarely) adding unnecessary unicode :/
11:46:09 <_jgr_> How would they know that those characters are somehow bad/special?
11:57:56 <_zephyris> "Change: Improved UI workflow for entering the scenario editor"... _fixed_ it.
11:59:03 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
12:02:45 <_zephyris> Does it need a check/warning for unusual whitespace?
12:05:59 <_zephyris> A zero width space wouldn't be appropriate for that string
12:06:08 <_zephyris> Is is invisible AI injection for watermarking?
12:10:01 <LordAro> i don't think LLM text watermarking works quite so obviously
12:14:10 <peter1138[d]> Right, it's not that I expect translators to know that U+200B is special. It's that I *don't expect* translators to be typing in such codes. It's not generally something that is typed accidentally and does perhaps suggest some tool was used to generate the text.
12:14:56 <LordAro> Certainly for a typographically simple language such as Danish
12:15:05 <_zephyris> Some googling suggests it's an unintended quirk of some translation tools, and a deliberate watermarking strategy for some genAIs
12:16:42 *** toktik has quit IRC (Remote host closed the connection)
12:17:33 <LordAro> how feasible is it to work out which translator modified that line?
12:18:59 <peter1138[d]> > danish: 86 changes by bscargo
12:32:52 *** zyliety has quit IRC (Quit: User went offline on Discord a while ago)
12:34:41 <rito12_13> Possibly they don't visit IRC/Discord channel
12:38:48 <_zephyris> Which whitespace should be used? Only space? Do we use nbsp for value-unit pairs?
12:39:30 <peter1138[d]> We have {NBSP} for that.
12:39:53 <peter1138[d]> Anyway, since OpenTTD ignores these characters now, we could just remove this check.
12:46:56 <_zephyris> Ah, of course, forgot that one.
12:47:55 <_glx_> some invisible whitespaces have a real meaning for some scripts but clearly not in this case
12:48:40 <brickblock19280> _zephyris: I wouldn't consider it unreasonable, you would potentially want to split slutningen into slut
12:48:40 <brickblock19280> -ningen if it was right at the end. It's the most appropriate part of the word to split.
12:49:11 <_glx_> but openttd doesn't split words
12:50:31 <peter1138[d]> That's not where it is :-)
12:50:57 <peter1138[d]> > Kun til slutningen af <U+200B><U+200B>linjen
12:51:28 <brickblock19280> That's definitely not normal
12:51:31 <_glx_> definitely AI watermarking
12:51:44 <peter1138[d]> Of course, if you had "slut<U+200B>ningen" then it would be able to split the word, if it was wrapping.
12:52:39 <peter1138[d]> We wouldn't add a hyphen though.
12:53:44 <peter1138[d]> LordAro, we're having those dreaded thoughts again.
12:53:59 <peter1138[d]> "Is it time for mudguards?"
12:54:12 <LordAro> we just require it when the clocks change
12:54:26 <LordAro> simplifies things as people can't be trusted to make proper judgements
12:54:28 <peter1138[d]> That's our rule. It looks like it's going to be wet the day before.
12:54:29 <_glx_> that will be quite soon
12:55:39 <LordAro> it's a week earlier than i thought it was too
12:56:00 <LordAro> and my winter bike is still broken :(
12:58:28 <peter1138> Mine is fixed, but it weighs about 7-8 kg more than my normal bike, so...
12:58:42 <peter1138> I'll go with the shitty flappy things instead.
12:58:55 <LordAro> mm, I think I'll have to
12:59:00 <peter1138> "Made an effort" "Not much of one" "It's still too short"
12:59:41 <LordAro> I took two wheels to the wheel repair shop this morning. That was quite exciting - one strapped to my back, the other in my hand
12:59:56 <LordAro> The headwind and rain did not help matters
13:00:41 <peter1138> I need to find a new mechanic :(
13:42:41 <_zephyris> It's feeling chilly, are we feature frozen for RC?
13:42:49 <_zephyris> Thinking about a good time to do font updates.
13:43:38 <peter1138> Any time you update really :)
14:07:45 <LordAro> certainly no need to strictly be in step
14:20:34 <DorpsGek> - Add: summary for week 42 of 2026 (by OpenTTD Survey)
14:22:00 <_zephyris> Cool. There's been a fair few fixes, so I'll do a release and a vendor at some point
16:09:41 <tony_pixel> Could it be possible to add badge filtering in vehicle replacement list also?
16:17:21 <peter1138> So funnily enough, it was Danish that caused us to add the U+200B test in the first place.
16:22:03 *** toktik is now known as Guest3140
16:24:06 *** Guest3140 has quit IRC (Ping timeout: 480 seconds)
16:31:44 <peter1138> Also, yeah, it was the same translator last time.
16:32:10 <peter1138> And the time before.
16:33:42 <LordAro> worth trying to engage with them first, or just unauth and wait?
16:35:08 <peter1138> Make strgen output the string "U+200B" instead of complaining ;-)
16:35:30 <peter1138> Or make strgen just ignore it.
17:11:27 <peter1138> Bah, Doom levels that don't let you backtrack.
17:19:15 *** gelignite has joined #openttd
18:17:03 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
18:19:46 *** WormnestAndroid has joined #openttd
19:52:41 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
19:53:04 *** WormnestAndroid has joined #openttd
20:09:46 *** jfs_ has quit IRC (Ping timeout: 480 seconds)
21:11:55 *** WormnestAndroid has quit IRC (Remote host closed the connection)
21:12:03 *** WormnestAndroid has joined #openttd
22:29:46 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:30:27 *** WormnestAndroid has quit IRC (Remote host closed the connection)
23:30:49 *** WormnestAndroid has joined #openttd
23:30:51 *** WormnestAndroid has quit IRC (Remote host closed the connection)
23:31:05 *** WormnestAndroid has joined #openttd
continue to next day ⏵