IRC logs for #openttd on OFTC at 2023-06-27
β΄ go to previous day
02:00:17 *** Wormnest has quit IRC (Quit: Leaving)
02:37:02 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:16:05 *** ChanServ sets mode: +v tokai
04:22:47 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
05:38:47 *** nielsm has quit IRC (Remote host closed the connection)
07:07:29 *** D-HUND is now known as debdog
07:27:03 *** peter1138 has joined #openttd
07:27:03 <peter1138> Hmm, maybe I should replace the apache server frontending by misskey instance.
07:31:24 <pickpacket> misskey... isn't that an activitypub server?
07:33:49 <peter1138> Yes. I think if I make a post I'm hitting some limit with concurrent connections, as everything goes a bit slow, but CPU usage isn't particulary high.
07:37:33 <pickpacket> peter1138: which AP server are you currently using?
07:38:24 <peter1138> Oh. I meant "my" instance, not "by".
07:39:23 *** m3henry has quit IRC (Quit: m3henry)
08:15:32 <peter1138> Phew, realised that scss doesn't allow changing variables, it ignores the definition. So I can override bootstrap colours completely externally.
08:40:34 *** virtualrandomnumber has joined #openttd
08:41:18 *** virtualrandomnumber has quit IRC ()
08:45:25 <dwfreed> truebrain: irc usernames are the same as linux usernames
08:47:12 <dwfreed> which is also odd, because there are *definitely* linux usernames that start with _
08:50:39 <truebrain> I find it mostly odd I couldn't find anything defining what restrictions they hold .. so I just hope it is nickname-but-not-starting-with-_ π
08:51:19 <dwfreed> must start with alphanumeric
08:52:05 <truebrain> bit of whac-a-mole now π
08:52:36 <dwfreed> I'm reviewing the list of character attributes in the important ircds for this and will have a regex for you in a sec
08:55:05 <truebrain> a nickname can start with a `[` .. let's see what Discord usernames allow ..
08:55:33 <truebrain> `Theyβll be limited to lowercase characters (a-z), numbers (0-9) and two special characters (period and underscore). `
08:55:42 <truebrain> okay, so period is already filtered out
08:55:45 <truebrain> and underscore we now strip
08:55:59 <dwfreed> period is fine in ident
08:56:06 <truebrain> but not in nickname π
08:57:23 <dwfreed> [a-zA-Z0-9][-.0-9A-Z\[\\\]^_`a-z{|}~]*
08:57:35 <truebrain> and they can start with a number? Lol, too funny π
08:57:40 <truebrain> tnx, that is very useful
09:04:24 *** _zephyris has joined #openttd
09:04:24 <_zephyris> _zephyris: I was wondering (now the IRC fun has resolved!) if anyone can help with this ^^, getting debug info on missing baseset sprites?
09:06:53 *** brickblock19280 has joined #openttd
09:06:53 <brickblock19280> I don't think that works over the bridge but you could try to use the sprite aligner
09:09:57 <_zephyris> What do you mean? Just scroll through the extended sprite index range looking for question marks?
09:10:55 <peter1138> It would be useful if the game just listed the IDs, at least with a debug level set.
09:11:04 <peter1138> (Maybe it can, I've not checked)
09:14:13 <_zephyris> Yeah, even just listing action5/replace_new features with missing sprites would be handy...
09:15:23 <truebrain> right, then this should be the proper fix dwfreed; tnx again π
09:15:44 <truebrain> push before pressing "create PR" ... lol
09:19:05 <brickblock19280> _zephyris: Have you replaced the tunnel portal less tunnel sprites?
09:21:45 <_zephyris> Yup, done those ones...
09:22:18 <_zephyris> It's a weird number to be missing. Tbh I don't quite believe it, can't find a blue question mark in game.
09:23:19 <_zephyris> Hmm, and I guess if sprites were missing from the end of an action5/replace_new range they won't appear in the sprite aligner anyway...
09:23:33 <_zephyris> I'll let you know in 8000+ clicks!
09:33:45 <_zephyris> Well, 3500 to get through the extra sprites... And no question marks...
09:37:40 <brickblock19280> Some might be openttd fallback sprites
10:05:51 <_zephyris> brickblock19280: Ah, yeah, true...
10:09:48 <Eddi|zuHause> _zephyris: debug level should be "-d sprites=1"
10:12:43 <Eddi|zuHause> _zephyris: there doesn't seem to be a way to tell which exact sprites are missing
10:33:04 <Eddi|zuHause> i think i found something
10:45:46 <truebrain> let's bring this live, to avoid people not being able to talk on IRC ever again (famous last words) π
10:48:14 <Eddi|zuHause> UserError("Uhm, would you be so kind not to load a NewGRF that changes the type of the map generator sprites?"); <-- haha :p
10:51:59 <peter1138> Which barely anyone uses now π
10:53:36 *** brickblock19280 has quit IRC (Remote host closed the connection)
10:53:36 *** truebrain has quit IRC (Remote host closed the connection)
10:53:36 *** _glx_ has quit IRC (Remote host closed the connection)
10:53:36 *** peter1138 has quit IRC (Remote host closed the connection)
10:53:36 *** _zephyris has quit IRC (Remote host closed the connection)
10:53:36 *** DorpsGek_v has quit IRC (Remote host closed the connection)
10:53:49 *** DorpsGek_v has joined #openttd
10:54:02 *** truebrain has joined #openttd
10:54:02 <truebrain> right, IRC bridge reloaded π
10:54:40 <orudge> [11:51:40] <peter1138> Which barely anyone uses now <-- given only 6 people disappeared now, maybe more than it seems still use it? :D
10:54:57 <Eddi|zuHause> am _I_ still visible?
10:55:27 *** _zephyris has joined #openttd
10:55:27 <_zephyris> Eddi|zuHause: Yeah, still useful though. 19 fallback sprites... Maybe I'll modify the fallback grf to all bright pink...
10:55:48 <Eddi|zuHause> _zephyris: this change could help you?
10:55:53 <_glx_> All action5 have fallbacks
10:56:32 <truebrain> orudge: What does the map generator have to do with people disappearing? π
10:56:39 <truebrain> I am scared now π
10:57:00 <_zephyris> Eddi|zuHause: Oh, yes, perfect! Thank you.
10:57:00 <orudge> Heh, I thought that reply was related to "not being able to talk on IRC ever again"
10:57:08 <truebrain> I know; that is why it is funny π
10:57:24 <_zephyris> Eddi-zviaGitHub: Strong support for this ^^^!
10:57:29 <truebrain> and for your context orudge, 6 people talked on Discord between 0300 CEST and 1300 CEST. Now count how many talked on IRC π
10:58:08 <orudge> Hey, I never said all the IRC users were *active* ;)
10:58:11 <truebrain> well, talked in this channel .. I muted most other channels π
10:58:28 <truebrain> and I am saying we only connect Discord users to IRC when they actually talk π
10:58:33 <truebrain> otherwise we would flood 7500+ people in π
10:58:46 <truebrain> (I am not joking, there are 7500 users on Discord)
10:58:53 * orudge can't say he has ever used Discord
10:59:00 <truebrain> you are missing out
10:59:11 <orudge> I do have Matrix these days, mainly for work-related things
10:59:24 <orudge> but I have enough bloated Chromium-embedding clients on the go
10:59:29 <orudge> and I presume Discord is the same :D
10:59:42 <truebrain> One can assume all they want π
11:26:36 <pickpacket> What kind of stats can we get from steam? Number of current installations of the game? Total hours played last month? Number of active players..?
11:27:44 <Eddi|zuHause> steam makes a survey about machine stats, but i don't know if they propagate the results to the game devs
11:44:03 *** peter1138 has joined #openttd
11:50:38 <pickpacket> for the game! on steam!
11:50:49 <pickpacket> 93% is pretty awesome :D
11:51:02 <Eddi|zuHause> we should offer each negative reviewer their money back
11:52:54 <pickpacket> and we should offer a free trial period
11:57:09 <Rubidium_> we should return them 10 times the MSRP :D
11:58:37 <Eddi|zuHause> "manufacturer suggested retail price"?
12:08:07 *** locosage has joined #openttd
12:08:07 <locosage> is it my internet or coordinator having issues? can't get server list in the game
12:09:57 *** brickblock19280 has joined #openttd
12:10:19 <locosage> weird, everything works but not server list
12:11:15 <Eddi|zuHause> probably your router/NAT/ISSP eating UDP packages
12:13:31 <pickpacket> brickblock19280: Swedish. Interesting.
12:14:44 <Eddi|zuHause> you know something weird i noticed, when i was in belgium, my phone didn't find a roaming network, but i was near the czech border recently (even though i didn't cross it), it did find czech roaming.
12:15:37 <_glx_> roaming used to be a pain when being near borders
12:16:15 <_glx_> (unexpected costs on the invoice)
12:17:41 <_glx_> Eddi|zuHause: incompatible frequencies ?
12:17:46 <Eddi|zuHause> i did get a text message about roaming conditions, but i haven't thoroughly read it
12:27:23 <Eddi|zuHause> _glx_: dunno anything about frequencies. did belgium shut off G2?
12:29:51 <truebrain> Eddi|zuHause: FYI, OpenTTD only uses UDP on local networks these days .. since 12.0 .. like a while now π
12:34:38 <peter1138> Hmm, of course, SetDParamStr() introduces scope issues.
13:10:33 <Rubidium_> peter1138: hopefully not for long... I'm almost at the point of allowing temporaries for SetDParamStr
13:16:54 *** bigyihsuan has joined #openttd
13:16:54 <bigyihsuan> what's the latest firs/xis/derivative that exists? i want to peek at the source so i can think about making a fork
13:17:24 <Eddi|zuHause> because there aren't enough forks yet?
13:18:16 <bigyihsuan> but i want to make an aerospace industry-themed firs/xis fork (think 1940s to near future rocketry and all of its industry chains)
13:18:26 <peter1138> Don't be like RedHat, eddi π
13:18:35 <brickblock19280> Start from firs
13:18:41 <bigyihsuan> brickblock19280: roger
13:19:30 *** emperorjake has joined #openttd
13:19:30 <emperorjake> AXIS is based on FIRS 4.4.0 but there are newer versions of FIRS
13:19:34 <Eddi|zuHause> that would have been my suggestion as well
13:20:39 <bigyihsuan> brickblock19280: docks for delivering rocket parts would be a cool thing
13:21:27 <brickblock19280> It separates the water industry ground from the rest of the ground sprites compared to the original code
13:28:47 <bigyihsuan> popped the repo into vscode and immediately my linter's going crazy lmao
13:29:13 <bigyihsuan> no type annotations smh :suspensive_railway:
13:29:26 <brickblock19280> Py nml is annoying
13:32:37 <bigyihsuan> just with 1 file open
13:32:52 <bigyihsuan> weh, might have to turn off type linting for this one
13:41:38 <locosage> dbg: [net] getaddrinfo() for address "coordinator.openttd.org:3976" took 5 seconds
13:41:38 <locosage> dbg: [net] This is likely an issue in the DNS name resolver's configuration causing it to time out
13:43:33 *** andythenorth has joined #openttd
13:43:33 <andythenorth> Lol what, since when is python expected to have type annotations?
13:44:02 <andythenorth> Is there a new PEP?
13:56:17 <andythenorth> Iβm not going to add them, youβll have to ignore the linter
14:07:11 <bigyihsuan> i've turned off type linting
14:15:40 <LordAro> pretty sure andy's python predates type hinting
14:32:13 <_glx_> aren't anything "Any" in python ?
14:33:03 <_glx_> that's so unhelpful with intellisense
14:35:40 <andythenorth> LordAro: Iβm on 3.11, seems itβs a 3.10 thing
14:50:34 <peter1138> `:Polish ZΕoty` ... can we just change that to `:Polish Zloty`?
14:55:48 *** Wormnest has joined #openttd
14:56:53 <merni> but probably not uncommon in software
15:02:12 <Eddi|zuHause> we should probably make it a translator rule to not use letters that aren't regularly used in the language
15:02:49 <peter1138> Is it wrong when translating a Polish word to English to replace letters that are used in Polish but not English?
15:03:58 <Eddi|zuHause> there was a similar topic for (i think) japanese recently
15:04:42 <peter1138> There are a lot of fonts that don't include a `Ε` which tehrefore cannot be used in OpenTTD, just because of one currency that most people won't use.
15:04:50 <_glx_> it was readme in japanese and english language
15:05:30 <peter1138> Most of the issue was the currency code translations, but for zloty it's in the name too π
15:05:56 <Rubidium_> bear in mind that the hard coded currency also uses that character but that won't be found by the standard font finding stuff
15:07:02 <peter1138> Yes, if you select Zloty and the font doesn't support it, you get everything in your font, but the currency is "z?"
15:07:46 <peter1138> Whereas if you select ZΕoty and the font doesn't support `Ε`, then you get everything in some random fallback font.
15:08:25 *** ahyangyi has joined #openttd
15:08:25 <ahyangyi> Please, just use `Zloty` in the English UI.
15:09:20 <ahyangyi> I mean, `Chinese Renminbi` isn't terribly right as well. I could argue for `Chinese RΓ©nmΓnbΓ¬` with some pedantic reasoning such as "Chinese is a tonal language!"
15:09:29 <ahyangyi> But nobody wants that
15:11:59 <_glx_> ideally we could use the user select font, and fallback only for missing chars, but that's probably not easy
15:12:48 <Eddi|zuHause> haven't they made fallback rules for Ε->l and ΓΆ->o and stuff for unicode letters?
15:13:54 <Eddi|zuHause> "THEY". "UP THERE".
15:14:31 <Eddi|zuHause> can't think of a funny conspiracy theory right now
15:14:44 <peter1138> And the hardcoded currency symbol does use `Ε`, but there are already translations that don't use `Ε`, so the fact it'll show up as `z?` is nothing new.
15:15:19 <merni> Wait, why are currency *symbols * translated?
15:15:46 <_glx_> isn't `Ε` in sprite font? (somewhere in extra)
15:16:03 <merni> Eddi|zuHause: Those are symbols, not words
15:16:34 <Eddi|zuHause> _glx_: yes. but the sprite font is ignored when using truetype-fonts
15:16:56 <_glx_> could be a nice fallback
15:17:08 <peter1138> Currency symbols are not translated.
15:17:18 <Eddi|zuHause> probably makes string handling more complicated than necessary
15:17:56 <brickblock19280> Eddi|zuHause: Γ to oe would be better
15:17:56 <peter1138> Currency *names* are translated.
15:28:04 <peter1138> I can use more fonts in most EU languages than English, because of that single Ε.
15:41:15 <LordAro> andythenorth: type hinting has been a thing since 3.8, iirc
16:08:04 *** HerzogDeXtEr has joined #openttd
16:23:28 <Eddi|zuHause> i still have programs on 2.7
16:27:30 <jfs> alternatively, ship the game with sprite font covering all currency symbols
16:28:21 <jfs> maybe even a fallback truetype font that can be used for those single symbols if that's possible
16:31:38 *** gebik4544 has joined #openttd
16:31:38 <gebik4544> what about those pesky Czech symbols like Ε?
16:41:49 *** nerryoob has joined #openttd
16:41:49 <nerryoob> Hey I originally asked this on add-on-development but apparently this is a better place for it. I've been working on a patch to let trains run on tram tracks but when I was trying to get them to visually drive on one side of the road (like road vehicles do) I got a little confused by how sprite drawing works. The values of x, x_offs, x_extent etc don't seem to affect where the sprite for the
16:41:49 <nerryoob> train is drawn, as changes to them in Train::UpdateDeltaXY and DoDrawVehicle didn't seem to have any impact. Any idea how I can change the visual position of the train sprite?
16:46:24 <Eddi|zuHause> nerryoob: the sprite offsets are relative to the bounding box. you need to move the bounding box, not the sprite
16:47:54 <Eddi|zuHause> nerryoob: there are screen coordinates (x and y), and world coordinates. you need to deal with world coordinates (x, y and z)
16:59:54 *** gelignite has joined #openttd
17:21:51 <peter1138> jfs: Those symbols ARE in the sprite font.
17:22:07 <peter1138> The problem is, CJK and Cyrillic are not.
17:25:00 <jfs> hence why I'm suggesting letting the game have a per-character fallback to some specific, included font data (not to any arbitrary system font)
17:25:47 <jfs> actually no different from "falling back" to the included font/sprite data for the special symbols for transportation type and a few more
17:30:52 *** _aD has quit IRC (Ping timeout: 480 seconds)
17:35:18 <andythenorth> why can't we multi-thread by sharding the map tiles?
17:35:49 <andythenorth> I don't mean "we should", I mean "there are reasons, what are they?"
17:41:47 <truebrain> andythenorth: because many actions require information from surrounding tiles .. so a big mutex nightmare, if I would guess π
17:42:43 *** frosch123 has joined #openttd
17:42:43 <frosch123> c++26 adds hazard pointers. you can add a huge mess then π
17:57:21 <frosch123> LordAro: don't worry, hazard pointers do not meet openttd's determinism requirements
17:58:13 <andythenorth> shame we can't apply eventual consistency to MP π
17:58:53 <andythenorth> "your train crashed, several years ago"
17:59:33 <frosch123> if the train is moving away from you very fast, it's actually realistic :p
17:59:50 <andythenorth> do we allow near-light-speed trains?
18:00:23 <frosch123> not sure whether daylength patches add a setting for lightspeed
18:00:54 <andythenorth> lightspeed might be variable anyway
18:04:20 <truebrain> I just saw Einstein turning around in his grave
18:05:55 <Eddi|zuHause> i don't think the currently finitely many possible speed values include a range of relativistic speeds anyway :p
18:06:05 <jfs> const_cast<float>(c) = 200000000;
18:15:29 <LordAro> frosch123: i am scared
18:35:59 <truebrain> is frosch doing it again? Tssssk .. stop scaring the young people frosch123 ! π
18:41:19 <frosch123> i dealt with a library using something hazard-pointer-ish only once. it was so buggy, we decided to rm-rf it
18:43:20 <DorpsGek> - Update: Translations from eints (by translators)
18:58:00 *** virtualrandomnumber has joined #openttd
18:58:31 *** virtualrandomnumber has quit IRC ()
19:15:58 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
19:45:39 *** gelignite has quit IRC (Quit: Stay safe!)
20:58:36 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:06:38 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
21:32:24 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:49:47 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:57:37 *** _aD has quit IRC (Ping timeout: 480 seconds)
continue to next day β΅
