IRC logs for #openttd on OFTC at 2023-03-05
β΄ go to previous day
00:05:44 *** EmperorJake has joined #openttd
00:05:44 <EmperorJake> AXIS merges Stainless and Alloy steel
00:38:56 *** bryjen has quit IRC (Quit: Leaving)
00:43:59 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
00:44:02 *** WormnestAndroid has joined #openttd
01:11:16 *** Eddi|zuHause has joined #openttd
01:50:45 *** Tirili has quit IRC (Quit: Leaving)
02:04:32 *** WormnestAndroid has quit IRC (Read error: No route to host)
02:05:18 *** WormnestAndroid has joined #openttd
02:24:15 *** lobstarooo__ has quit IRC (Read error: Connection reset by peer)
02:54:29 *** Wormnest has quit IRC (Quit: Leaving)
03:02:31 *** Vadtec has quit IRC (Quit: Vadtec was here....)
03:31:52 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:33:23 *** D-HUND is now known as debdog
04:13:54 *** TROILUS has quit IRC (Read error: Connection reset by peer)
04:14:17 *** TROILUS has joined #openttd
05:19:16 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
07:11:05 *** TROILUS has quit IRC (Quit: Leave)
07:11:46 *** TROILUS has joined #openttd
07:46:10 <petern> glx22viaGitHub: Hidden road types should not be buildable by towns either
07:47:15 <petern> I think I had a patch or pr for rhat
07:57:50 *** bryjen has quit IRC (Quit: Leaving)
08:14:28 <andythenorth> trying to get chatGPT to read FIRS .dot cargoflow
08:19:02 <andythenorth> is chatGPT the best teddy bear coding partner?
08:20:23 <andythenorth> `As an AI language model, I'm not a physical teddy bear and I cannot physically interact with you like a teddy bear could. However, I can help you with a wide range of language-related tasks and answer your questions to the best of my knowledge.`
09:13:40 <TrueBrain> hmm .. according to ISO-3166, Wales doesn't exist ..
09:15:16 <TrueBrain> I do wonder how they got their TLD in that case .. π
09:17:25 <TrueBrain> should have `GB-WLS` assigned to it .. hmm
09:17:33 <andythenorth> I can verify that Wales does exist
09:17:36 <andythenorth> I have been there
09:19:47 *** nielsm has quit IRC (Ping timeout: 480 seconds)
09:19:50 <TrueBrain> weird, wikipedia has a code assigned to it, but the ISO from 2022 does not
09:20:18 <TrueBrain> it does to the subparts of Wales
09:20:36 <TrueBrain> similar issues with the other countries in UK, like Scotland
09:21:06 <TrueBrain> yeah, okay, so England, Northern Ireland, Scotland and Wales are missing, in that sense
09:21:14 <TrueBrain> guess it got too complicated π
09:21:17 <TrueBrain> a country in a country π
09:22:05 <TrueBrain> `on 28 September 2021, UK replaced GB as the official country code on car registration plates`
09:22:25 <PopeofMemes> you think thats bad?
09:22:28 <andythenorth> we're quite consistent in our inconsistency
09:22:41 <TrueBrain> if weird is bad? I leave that for the historians
09:22:45 <andythenorth> we've been doing it for thousand years or more π
09:23:09 <TrueBrain> weird landborders we have all over the world; that is nothing weird
09:23:20 <TrueBrain> but not knowing if you want to be called GB or UK .. that is just odd π
09:23:20 <PopeofMemes> there are Dutch parts in Belgium, in the Netherlands, in Belgium, **In the Netherlands**
09:23:52 <PopeofMemes> imagine just going to the bathroom in Belgium and eating breakfast in the Netherlands
09:23:58 <TrueBrain> I think I will just manually add those four entries, as this is just too weird not to have
09:24:11 <andythenorth> it's very simple, United Kingdom of Great Britain and Northern Ireland
09:24:44 <andythenorth> The island of Ireland isn't Ireland
09:25:04 <TrueBrain> it is like they half-way were like: nah, we're good
09:32:05 <TrueBrain> seems GB is the only country with "this issue"
09:46:33 <TrueBrain> I am seriously considering removing the ability to name 1 version of a NewGRF different than the other .. it doesn't actually seem to be used in a way that is helpful, and it looks like it is more likely to confuse the author π
09:52:20 <andythenorth> I was surprised it could be done
09:52:44 <TrueBrain> yeah, and take your GRFs for example .. the package itself is called without `(trains)`, and some versions do add it
09:52:48 <TrueBrain> doesn't seem like you actually intended that π
09:53:36 <andythenorth> I intended the package name to change everywhere for that package
09:53:49 <TrueBrain> yeah, and I see more of those packages π
09:54:14 <andythenorth> If I want to fundamentally change the behaviour of the grf
09:54:14 <andythenorth> Such that it needs a name change for a specific version
09:54:14 <andythenorth> Then it needs a grfid bump
09:55:14 <TrueBrain> okay ... during this initial conversion, shall I also look at titles to see if I can spot regions?
10:00:29 <TrueBrain> the chances of misclassifying increases π
10:20:07 <TrueBrain> okay, I need some extra set of eyes on this ...
10:20:07 <TrueBrain> Is there anything in that list that trivially points to a region?
10:21:16 <andythenorth> did you engineer that 'Furry' would be first? π
10:21:21 <andythenorth> not judging anything or anyone π
10:24:08 *** Brickblock1 has joined #openttd
10:24:08 <Brickblock1> Arabian town names does not have a region
10:24:14 <TrueBrain> because that is not a region π
10:24:41 <TrueBrain> in the description it lists some countries, but I cannot automatically pick up on those π
10:25:12 <Rubidium> Alpine climate? The alps are considered a region, although there are several regions called alps...
10:25:32 <TrueBrain> different kind of "region", not one 3166 sees as a region π
10:26:05 <Brickblock1> some train sets also don't show up
10:26:16 <TrueBrain> "don't show up", what do you mean?
10:26:44 <Brickblock1> don't have a region is what I meant, sorry for the confusion
10:26:51 <Rubidium> you didn't add towns, right?
10:26:55 <Brickblock1> but they might not have tags
10:26:55 <TrueBrain> like? "some" is a bit too vague π Sorry π
10:27:09 <Brickblock1> CP-trainset and PNR trains
10:27:09 <TrueBrain> Rubidium: no, only 3166-2 at lowest resolution
10:27:20 <TrueBrain> Brickblock1: yeah, impossible for me to figure out where they belong π
10:27:37 <TrueBrain> I could peek in the description, but that becomes very tricky
10:27:42 <TrueBrain> will check in a moment how bad that would be π
10:27:52 <Rubidium> TrueBrain: all the Auz ones, but that's just slang interfering there I guess
10:28:16 <TrueBrain> I have no clue what it actually refers to
10:28:22 <TrueBrain> there is nothing in any of the descriptions to give an indication
10:28:33 *** sla_ro|master has joined #openttd
10:28:35 <Rubidium> Auz -> Australia I'd reckon
10:28:46 <TrueBrain> but that is a guess
10:28:58 <TrueBrain> at least, I don't know π It could also be used as trademark name, and be of something different π
10:29:06 <Rubidium> Lombard town names -> town names of the Lombardy region, but that's like having Frisian town names
10:29:31 <TrueBrain> okay, the tt-forums post does mention Australian sometimes, but also American West ..
10:30:21 <TrueBrain> I rather not classify something that misclassify .. but looking at the tt-forums post, I feel a bit more empowered to classify it as Australian π
10:31:00 <Brickblock1> some of them are mostly generic tho
10:31:15 <TrueBrain> okay, then I will not set a region to Auz, and leave it to the author π
10:31:21 <Rubidium> "Auustralian Wide Ballast Tracks" <- someone made a typo
10:31:41 <Rubidium> though, he description mentions Australian
10:32:20 <dP> a lot of town names actually
10:32:31 <dP> Tibetan, Lombard, Winnipeg...
10:33:07 <Rubidium> TrueBrain: you ditched the UN MU49?
10:34:05 <TrueBrain> these kind of remarks always confuse me, as people don't state what they see, but they make a conclusion based on something they see .. doesn't help me Rubidium π
10:34:37 <TrueBrain> UK is annoying .. as it is too short to do generic matching on π
10:34:58 <TrueBrain> tnx Brickblock1 , added the bunch
10:40:17 <Rubidium> TrueBrain: which one confused you? The question about UN M49, if so you previously stated "only 3166-2" so I wondered whether state names I have seen in UN M49 would be considered regions by you(r code) or not. Or was it the "Names of Texas", then that is actually the name of a NewGRF
10:40:26 <TrueBrain> Brickblock1: how that happened, lol. Let me fix π
10:40:46 <Brickblock1> has Chicago in description I guess
10:41:08 <TrueBrain> Rubidium: no, I said 3166-2 was the lowest resolution π I didn't say it was the only set used π But lowest -> highest I guess ..
10:41:58 <TrueBrain> Brickblock1: it has `chicago` in the tags
10:42:24 <TrueBrain> but that is still a misclasfficication π Let me fix π
10:42:52 <Brickblock1> I wouldn't call those Canadian
10:43:30 <TrueBrain> `canada` in the tags
10:43:33 <TrueBrain> but .. yeah .. let's not
10:44:46 <Brickblock1> Greenlandic town names show up with Europe but I don't know how that would be changed
10:45:06 <TrueBrain> Rubidium: btw, for a complete picture, Texas is not in UN M49, as that just defines cities and regions upwards. Texas however is in 3166-2, so I mapped it now
10:46:04 <TrueBrain> Brickblock1: tags also included denmark, but let me add greenland too π
10:46:23 <Brickblock1> yeah Denmark makes some sense
10:47:50 <Brickblock1> city objects being France feels wrong to me
10:48:01 <Brickblock1> but I assume it was in tags
10:48:58 <TrueBrain> yeah, had "new york" and "paris" in the tags
10:49:05 <TrueBrain> new york didn't make it π
10:49:28 <TrueBrain> reloaded the import to fix all the mentioned things
10:50:04 <FLHerne> TrueBrain: many of those have implicit regions that would be pretty hard to autodetect
10:52:57 <FLHerne> TrueBrain: like, NOHAB is a Swedish locomotive, all the Long Vehicles vehicles are European, the Transrapid track set is either German or Chinese
10:53:47 <FLHerne> Ikarus trucks are Hungarian
10:53:55 <TrueBrain> FLHerne: yeah, I have no intention of getting this 100% right
10:54:05 <TrueBrain> but what we can deduce, I have no problem adding .. but it has to be automated π
10:54:06 <Brickblock1> that loco wasn't built for use in Sweden originally so marking Sweden is in ways weird
10:54:36 <TrueBrain> I mainly do this as many of these NewGRFs have no active authors anymore
10:54:46 <FLHerne> Brickblock1: Nordic generally then :p
10:55:01 <FLHerne> oh, I didn't realise they ran in Hungary
10:55:02 <Brickblock1> it was also used in Hungary
10:55:55 <Brickblock1> it was also used in Belgium and Luxembourg
10:56:05 <TrueBrain> I think this will have to do .. seems like it is a fine balance between effort and result
10:56:19 <TrueBrain> as if we go any deeper, we get discussions like you guys just had π
10:56:57 <FLHerne> TrueBrain: I think lowering resolution beyond country codes would make sense
10:57:28 <TrueBrain> FLHerne: I honestly don't know what you try to say with that sentence, sorry π¦
10:57:39 <FLHerne> if there was a 'Europe' tag that (implicitly?) included all the European countries, and could also be put on "stuff that's clearly from /somewhere/ in Europe"
10:57:54 <TrueBrain> did you check any of the work I did? π
10:58:32 <FLHerne> not yet, I only read > [10:41] <TrueBrain> Rubidium: no, I said 3166-2 was the lowest resolution < 20min ago
10:58:51 <TrueBrain> yeah, and later I correct lowest -> highest, as I always confuse resolution direction π
10:59:45 <TrueBrain> euh, that URL doesn't work for you ..
11:01:05 <TrueBrain> tnx all for the bunch of suggestions/corrections; we can always fix things later, but for now this seems to work just fine π
11:05:52 <TrueBrain> in-game search is also rather content with this information, good π
11:22:58 <Xarick> the original NoCAB performs much better than NoNoCAB
11:26:42 *** sla_ro|master has quit IRC ()
11:28:41 <Xarick> let's put them face to face
11:28:50 <Xarick> with infinite money, let's see who does better
11:38:37 <TrueBrain> lol, some sets are funny .. a tag "durango" for a townname set of colorado
11:38:45 <TrueBrain> but durango is also a region in Mexico
11:52:47 <Xarick> NoCAB crashes to excessive CPU usage π¦
12:03:41 <gebik> TrueBrain: its town in colorado too as I looked it up
12:06:04 *** TROILUS has quit IRC (Read error: Connection reset by peer)
12:06:05 *** TROILUS1 has joined #openttd
12:06:09 *** TROILUS1 is now known as TROILUS
12:10:30 <TrueBrain> Even .. mobile phones are the worst
12:16:04 <Xarick> omg why did i install visual studio 17.5.1
12:16:10 <Xarick> i can't build yet again
12:24:43 *** HerzogDeXtEr has joined #openttd
12:25:36 <Xarick> which version of visual studio do you have installed?
12:29:36 <Xarick> i wonder if i can revert to 17.4.5
12:30:38 <klote[d]> Hi, i managed to get opettd server running but its using version 12.1
12:30:48 <klote[d]> how do i update it to the newest version?
12:39:57 <TallTyler> TrueBrain: Hmm that might be my NewGRF. Durango is the location of one of Americaβs best-known tourist railroads. No idea why I added that as a tag though.
12:43:52 <poglet> Then extract it using: tar -xf openttd-13.0-linux-generic-amd64.tar.xz
12:46:38 <klote[d]> downloaded it to the wrong folder i think
12:46:50 <TrueBrain> TallTyler: It is indeed yours π
12:55:34 <klote[d]> poglet: Hi thanks, this is kind of confusing.
12:55:51 <klote[d]> am i supposed to just swap the openttd program?
12:56:21 *** Kuhnovic has joined #openttd
12:56:21 <Kuhnovic> KuhnovicviaGitHub: Can one of the lead devs slap a Preview label on my PR? This really needs to be play tested π
12:56:47 <klote[d]> because i dont want to lose config info
12:58:04 <klote[d]> i used apt install openttd and its running fine
12:58:10 <klote[d]> i just want to upgrade
13:00:12 <poglet> Yeah sorry when i checked about 2 weeks ago apt didn't have the latest version so i had to go though these steps to get it working. Maybe someone here knows how long it takes for apt to get updated, not sure what the process is for that.
13:02:45 <FLHerne> poglet / klote[d]: Ubuntu doesn't update the major version of its OpenTTD packages except with new Ubuntu releases
13:02:55 <FLHerne> they only backport bugfixes
13:03:33 <FLHerne> (which is silly in my opinion but typical policy for non-rolling-release distros)
13:03:45 <klote[d]> So i tried to install the depencys
13:04:06 <klote[d]> but they are also not all availible :x
13:05:45 <FLHerne> This guide you're trying to follow is outdated and a bad idea generally
13:06:04 <klote[d]> then what am i supposed to do...
13:06:28 <klote[d]> how do i upgrade the 12.1 to 13...
13:06:46 <FLHerne> unpack it anywhere you like
13:06:58 <FLHerne> execute the `openttd` binary in the resulting folder
13:08:01 <FLHerne> it can coexist with the distro package, or you could uninstall that to avoid confusion
13:09:01 <klote[d]> what am i doing wrong here?
13:10:59 <FLHerne> ^ enter ./openttd (explicitly "the one in this directory") rather than relying on $PATH
13:11:17 <FLHerne> it might be interesting to see what `which openttd` returns if you run that
13:11:32 <FLHerne> now that isn't supposed to happen
13:11:47 <klote[d]> is probably because i have missing depencys?
13:12:19 <klote[d]> or does it need command -D behind it?
13:12:32 <klote[d]> to run server terminal?
13:12:33 <FLHerne> the generic binary is supposed to be built in a way that minimises the need for such things
13:12:46 <FLHerne> please upload the crash.log it printed the path for
13:13:34 <FLHerne> you could try with -D, but I expect it will still crash
13:13:45 <klote[d]> it gave specific error now
13:20:08 <FLHerne> (the zip, but don't unpack the tar inside that)
13:25:08 <klote[d]> its not responding to any commands.
13:28:41 <klote[d]> none of the rcon commands are working
13:40:09 <TrueBrain> lol, one minor issue with 3166 .. something like Aruba is part of NL, so it is part of Europe, it says
13:40:14 <TrueBrain> that is not exactly true π
13:42:16 <andythenorth> administrative regions vs. geographical regions π
13:42:22 <andythenorth> welcome to my day job π
13:42:47 <TrueBrain> always makes me giggle how bad those datasets actually are π
13:47:50 <klote[d]> ok so ingame rcon commands dont work π
13:51:38 <Rubidium> TrueBrain: what's wrong with the dataset? Does the data set declare Aruba as part of Europe, or did you assume transitivity?
13:52:17 <TrueBrain> that is not what I said; I never claimed anything wrong with the dataset. I just said there was an issue π
13:53:44 <TrueBrain> and andy summarized the issue very nicely π
13:55:05 <TrueBrain> some heightmaps tagged 20 regions .. lol
13:55:14 <TrueBrain> I will be using the region that is above that instead π
13:55:19 <TrueBrain> it won't fit in network packets otherwise π
13:58:40 <andythenorth> so that sea-regions-pathfinder patch π
13:58:52 <andythenorth> a tile will know which region it is in?
14:02:27 <andythenorth> kinds of want to stop this happening on map gen π
14:02:28 <TrueBrain> wauw, Albert still plays OpenTTD? π
14:02:49 <andythenorth> come back Alberth!
14:04:07 <andythenorth> hmm I already adjust some industry placement with GS after map gen
14:04:16 <andythenorth> maybe I can do more of that
14:04:47 <andythenorth> I'm not sure it's fair to the channel though, I will have a lot of stupid questions / rants
14:04:59 <TrueBrain> Similar question as earlier, but this time for heightmaps and scenarios ..
14:04:59 <TrueBrain> Did I miss anything obvious?
14:05:31 <andythenorth> what was the thing people asked? We ping some content authors when we do this stuff?
14:05:37 <TallTyler> I can make you a Stage channel, "Andy complains about GS" π
14:05:41 <andythenorth> not sure how to ping, who is mod?
14:05:58 <andythenorth> "who is in our consultee database / contact list?"
14:06:11 <TrueBrain> I don't really need anyone specifically to look over this, just humans π
14:06:13 <andythenorth> TallTyler: I can just complain to chatGPT now
14:06:31 <TallTyler> I think the consensus was, make a thread in Discord channel #development-forum and ping the `@NewGRF Developer` role (which OpenTTD Developers now have permission to do)
14:06:58 <andythenorth> thanks...well next time we can do that π
14:07:36 <TallTyler> Firefox logo heightmap?
14:07:39 <petern> I think my legs are done for
14:08:48 <TallTyler> Where is the region "Lakes"?
14:08:56 <TrueBrain> lol, let me put it on the ignore list π
14:09:36 <TrueBrain> TallTyler: There are regions called "West", and "Rivers", which tend to match on tags in the wrong way π
14:09:47 <TrueBrain> I already added many on the ignore list .. Lakes is a new one π
14:10:38 <TrueBrain> seems that was the last one π
14:10:44 <TallTyler> A lot of scenarios in None are real places that didn't get categorized
14:11:29 <andythenorth> are there also regions in a fictional world?
14:11:30 <TrueBrain> "a lot" sadly isn't matching my filters π
14:11:42 <andythenorth> do we need a separate namespace for 'Italo Calvino type regions'?
14:12:03 <TallTyler> La Pampa 512x512 is Argentina according to Google
14:12:21 <TallTyler> Same for Santa Cruz 512x512
14:12:38 <TallTyler> Fallout OpenTTD is the United States π
14:12:52 <andythenorth> not _the_ United States though
14:12:59 <andythenorth> _a_ United States
14:13:05 <TallTyler> Tierra del Fuego is also Argentina
14:13:32 <andythenorth> first one to link Borges wins a prize π
14:13:34 <TrueBrain> those scenarios had no tags at all
14:13:48 <TrueBrain> hmm .. would require some more fuzzy matching on the name ..
14:13:53 <TallTyler> And then RT3 Go West! is the northeast US: New York and most of the New England states
14:14:10 <TallTyler> I made several versions of that though, and the FIRS one isn't in "None"
14:14:18 <TallTyler> Acadia 1750 is Nova Scotia, Canada
14:14:47 <TrueBrain> TallTyler: slow down π We only set the region for content that already had an indicator for region
14:14:56 <TallTyler> Malaysia is presumably the country of the same name π
14:15:05 <TrueBrain> nothing for RT3 Go West had any indications in its tags nor name
14:15:08 <TrueBrain> so they are not tagged
14:15:13 <TallTyler> There are also two Fiji scenarios
14:15:34 <TallTyler> Not complaining, just making a list for you π
14:15:35 <TrueBrain> anyway, it is not useful to list them like this; but yes, all content with a name matching a region and no tags, aren't matched atm .. will see what I can do there
14:15:56 <TallTyler> I have to run an errand now but will be back in about 30 minutes
14:16:16 <TrueBrain> there are just too many darn regions to accidentially match π
14:19:16 <andythenorth> petern: the other one π
14:19:47 <andythenorth> TrueBrain: TrueBrain seems like it might be a human curation shaped problem π
14:19:53 <andythenorth> can the bot do it?
14:19:58 <klote[d]> Why is it when i download and run openttd in linux the list_dirs is pointing towards Windows path structure...
14:19:58 <Xarick> wow, thinking of pathfinding
14:20:39 <TrueBrain> andythenorth: huh? That did not compute
14:21:06 <andythenorth> I'm not sure it computed when I wrote it
14:21:13 <andythenorth> it made sense 2 mins ago
14:21:27 <klote[d]> need to use rcon command first
14:22:24 <TrueBrain> andythenorth: lol; anyway, this is a bot doing it, which is the problem
14:23:34 <petern> Christ, I just flung my phone down the stairs π¦
14:26:46 <TrueBrain> There are 3 `Mayotte` .. eeuuhh .. which to pick π Lol
14:27:36 <pickpacket> is it possible to reduce the speed of the game below 'normal'?
14:31:18 <TrueBrain> wow, also many Santa Cruz .. you near something this way about the world π
14:36:55 <TrueBrain> TallTyler: okay, fixed a bunch of entries based on their name; I hope I didn't introduce false positives π
14:39:58 <andythenorth> shall I try and teach GS about newgrf industry layouts? π
14:40:07 <andythenorth> and how to terraform land for them?
14:40:41 <frosch> TrueBrain: do you use "world" as key word to classify stuff?
14:40:55 <frosch> i am looking at "africa", and 3 out of 6 items look wrong
14:41:17 <TrueBrain> frosch: "world" is on a blacklist, and does not participate in the region selection
14:41:31 <TrueBrain> frosch: people keep giving me these vague remarks .. I need a bit more specific information π
14:41:51 <TrueBrain> I don't even know what content-type you were looking at π
14:42:14 <TrueBrain> did NML CAT made it into the list again? It is such an annoying one ..
14:42:29 <frosch> no idea how "world flags" "world train set" and "nml cat" ended up there
14:42:29 <TrueBrain> owh, this one is spelled differently, great
14:42:36 <andythenorth> hmm how can I find a GS block of connected sea tiles?
14:42:39 <andythenorth> of minimum size π
14:42:52 <TrueBrain> frosch: I do; they had tags like that
14:43:03 <frosch> ah, wolrd train set is fine, it lists all the continents in the description
14:43:14 <TrueBrain> basically, they tagged themselves as "Africa", "America", "Asia", "Australia", "Europa"
14:44:22 <TrueBrain> and I will fix the NML CAT .. it matches on "Lakes", which happens to be a region in Africa π
14:44:39 <TrueBrain> frosch: similar for World Flags btw, so I guess it is fine
14:45:31 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
14:46:32 <TrueBrain> pretty sure they meant to target americas, but they wrote america .. π
14:46:46 <frosch> how odd... i assumed "kiwi..." to be new-zealand stuff, but it is korean...
14:46:56 <TrueBrain> yeah, I had to double check that one too π
14:47:03 <TrueBrain> also if it was south or north ..
14:49:31 <TrueBrain> had the tag `Australian`
14:49:52 <TrueBrain> what is unfortunate about it?
14:50:36 <TrueBrain> next import it will also add JP; original tag was "japan set" .. I did not have that in my list yet π
14:51:18 <TrueBrain> there is a region called `Balkan` in `Turkmenistan`
14:51:50 <TrueBrain> will add it to the ignore list
14:52:05 <TrueBrain> frosch: still no clue what you meant with this π
14:52:26 <andythenorth> it's not really regional, it's just indicating other grfs it is designed to work with?
14:52:49 <frosch> TrueBrain: i see how it got classified as australian, but i think it was not meant to be
14:53:00 <TrueBrain> the author explicitly did
14:53:04 <TrueBrain> there is Auz in the list
14:55:11 <TrueBrain> frosch: that is up to the author ofc .. not much we can do about that, I guess π
14:56:33 <Xarick> omg the triggle A Hog bankrupted
14:56:41 <Xarick> infrastructure maintenance
14:57:27 <klote[d]> Hi, my server is not showing up in the server list... i can only see it lan
14:57:50 <TrueBrain> frosch: many of these regions exist in both countries π But I will make a priority fall here, and assume Luxembourg is meant π
14:57:56 <klote[d]> Is this a firewall thing?
15:00:07 <glx[d]> no it's the `server_game_type` setting
15:00:55 <TrueBrain> frosch: explicit tag by the author
15:01:55 <glx[d]> port forwarding is a nice thing to do, but not required anymore
15:02:44 <TrueBrain> frosch: later on you find a similar one with also "uk" on the list
15:02:47 <frosch> oh, the tags are still part of global.yaml
15:03:00 <frosch> i only checked the diff that removed them from version.yaml
15:03:18 <TrueBrain> I kinda ignored version.yaml tags .. purely based it on the ones in global π
15:03:26 <TrueBrain> I should make a run with the old tags in versions there too
15:03:53 <frosch> yeah, way too many people write stuff into version-specific entries
15:04:04 <frosch> the gui does not hide them well enough
15:04:15 <TrueBrain> I think I will remove `name` for sure
15:04:21 <TrueBrain> `description` I can see a use-case for
15:04:29 <TrueBrain> but honestly, we just need to redo the UI to give a better experience
15:04:34 <TrueBrain> andy keeps saying he is going to do it π
15:04:51 <frosch> currently we have the package text as placeholder text
15:05:07 <frosch> maybe we need a checkbox instead, which defaults to "use package info"
15:05:27 <TrueBrain> something like that, and that unchecking shows the textbox
15:05:32 <TrueBrain> but .. in the current UI this is just .. ugh
15:06:11 <Xarick> I'm scared of posting a PR that was rejected in the past π¦
15:06:42 <TrueBrain> what did Einstein say ... trying the same thing twice expecting a different outcome is the definition of stupidity?
15:07:10 <TrueBrain> frosch: ghehe π Yeah, using the tags as regions isn't perfect π
15:07:22 <Xarick> it was rejected in 2019
15:08:06 <Xarick> OPF was still a thing bakc then
15:09:17 <frosch> creating a random town name set, by using RL names from one specific region, and adding random prefixes and suffixes π
15:11:22 <TrueBrain> there is only one region by that name π To the ignore list!
15:11:49 <klote[d]> what does reload_cfg setting do
15:12:12 <klote[d]> wiki doesnt show explenation
15:12:17 <TrueBrain> frosch: yeah .. they all also do something in Mexico, they claim π
15:12:43 <TrueBrain> was funny at first, as it preferred Mexico City over Mexico ..
15:14:18 <frosch> there are so many sets on bananas, which my geometry skills would never have put into some region. now i can find them π
15:14:56 <frosch> my geometry skills are okay, maybe that's my issue with geography
15:15:28 <TrueBrain> I learnt a lot of new things today π
15:16:32 <michi_cc[d]> Hmm, looks like there's some release constipation again.
15:16:40 <frosch> i really don't understand the geography spam in some tags
15:17:01 <TrueBrain> I hope this new system also reduces it in some places ..
15:17:06 <TrueBrain> but I sided with the authors for now
15:17:25 <TrueBrain> if you disagree with any, do let me know
15:17:43 <andythenorth> michi_cc[d]: ? π
15:18:10 <michi_cc[d]> I thought I read some talk of 13.1 in here, but nothing seems to have happened so far.
15:19:14 <frosch> TrueBrain: can you detect which sets list regions from 3 or more continents?
15:19:23 <TrueBrain> sure, with some effort
15:19:33 <TrueBrain> what you want to do with that list?
15:19:55 <frosch> maybe they are just "generic" and should have no region
15:23:37 <TrueBrain> frosch: 4 hits in newgrfs .. now to find their grfid
15:23:40 <TrueBrain> but not many, at least π
15:24:11 <TrueBrain> these kind of errors happens a lot ... π
15:24:20 <TrueBrain> but I can fix this one π
15:25:20 <frosch> "future reality inspired maglevs" <- some names ...
15:27:16 <TrueBrain> 797a0201, 57545453, 4a448807, 5f5f0200 have 3 or more continents frosch
15:27:37 <Xarick> No common parent found for this merge commit (max-depth of 256 reached) what is this
15:27:55 <TrueBrain> 2 of them you already knew about
15:30:52 <frosch> i think JF Ratt Roads (5f5f0200) is the only valid one of them
15:30:58 <frosch> the other 3 are just region spam
15:31:06 <TrueBrain> okay, now I just need to fix regression tests ..
15:31:21 <glx[d]> looks like more validation is needed
15:31:40 <TrueBrain> glx[d]: nice list π
15:33:35 <glx[d]> updated the gist with the test code
15:38:53 <TrueBrain> explicit requested by author in this case
15:38:58 <TrueBrain> should we take action on it?
15:40:08 <frosch> if it is not too much work, we should fix the french canadian and brazilian portoguese stuff
15:40:11 <TrueBrain> okay, BR / PT is this one set, I can fix that
15:40:17 <TrueBrain> already did the canadian one
15:40:23 <andythenorth> hmm which will be less work?
15:40:36 <andythenorth> teaching the game whether an industry is buildable at a specific location?
15:40:46 <andythenorth> or watching me try to figure it out in GS?
15:41:17 <frosch> it's ridiculous how many regions only exist in the dropdown, because world train set tags them
15:41:40 <TrueBrain> and you can argue two sides of the coin there .. either that is a good thing, or a bad thing π
15:43:14 <TrueBrain> lol .. prussia -> russia π
15:46:34 <TrueBrain> "New England" is nothing π It is a collection of things .. hmmmmm
15:48:30 <frosch> hmm, what info do you use for classification? only the old tags, or also the description?
15:48:45 <TrueBrain> I don't want to touch descriptions, as they contain a lot of bloat π
15:49:54 <frosch> do you feel like adding more tags? or rather "no thanks"? π
15:50:26 <TrueBrain> so far, I only did things automated, with a very big script
15:50:45 <petern> 'nsfw' for george's LV4 (are they even on there)
15:50:45 <TrueBrain> what you suggest is close to curation, I guess; not a bad thing, but I kept away from that currently π
15:51:16 <TrueBrain> frosch: called "kriti", so it doesn't match
15:53:11 <Eddi|zuHause> you can probably safely exclude tags that are used only once
15:53:46 <frosch> plenty of very specific townname sets, which are classified correctly
15:54:51 <Eddi|zuHause> whenever someone says "wrong georgia" i need to think of that tiny island near antarctica :p
15:56:23 <andythenorth> uuf can I be bothered to rewrite the industry placement entirely in GS? π
15:58:13 <frosch> how can a region have such a generic name as "midlands" though
15:58:25 <TrueBrain> similar for West / East / etc
15:58:55 <Eddi|zuHause> how many cities can be called "new city"? or "red castle"? :p
16:01:32 <TrueBrain> okay, locally fixed all the links given so far; keep them coming π
16:02:36 <andythenorth> so currently the GS doesn't init until after map gen AFAIK
16:02:48 <andythenorth> and the grf has no idea whether a GS is present
16:03:13 <andythenorth> it seems pointless to place a lot of industries at map gen, then have GS delete them all
16:08:22 <TrueBrain> restarting the service to load in the new database
16:10:16 <Xarick> hmm i could provide the real pathfinder cost as distance
16:10:33 <Xarick> i guess it's what you wanted
16:10:50 <Xarick> it's just that the distance isn't effectively used
16:10:59 <Xarick> it's just passed around for nothing
16:11:18 <frosch> shall i report all the world map scenarios, which spam all regions?
16:11:27 <TrueBrain> well, I wasn't planning on fixing them
16:11:30 <TrueBrain> so that sounds weird π
16:11:44 <frosch> good, then i can skip reporting them π
16:12:06 <TrueBrain> frosch: hmm .. this one is a bit odd .. Papua New Guinea is part of Melanesia, it seems, but the description suggests it is part of Indonesia ..
16:12:18 <TrueBrain> some political thing?
16:12:30 <TrueBrain> owh, no, I am reading it wrong
16:12:37 <frosch> no, geography/political discrepancy π
16:12:38 <TrueBrain> it contains both PNG and Indonesia
16:12:50 <TrueBrain> okay, so the new info is correct
16:12:59 <frosch> one island, half indonesia, half new guinea
16:13:00 *** Wormnest has joined #openttd
16:13:03 <frosch> but the african guinea is wrong
16:13:13 *** WormnestAndroid has joined #openttd
16:14:13 <TrueBrain> yeah, fixed it in last reload π But was still wondering about the remaining two, but those are fine π
16:14:22 <TrueBrain> as for the "world" maps, I really don't know what a good solution is ..
16:14:39 <TrueBrain> we can introduce "world" or "earth" (as we also have "mars")
16:14:50 <TrueBrain> but .. I am sure if we don't curate it, people will do what is done currently anyway
16:14:59 <TrueBrain> reason I limited the regions you can enter to 10, which I think is still a lot π
16:17:15 <frosch> if in doubt, blame the author?
16:18:08 <frosch> oh, probably same author, just ottd vs gh account
16:23:04 <andythenorth> hmm how do I prospect within a fixed region of the map?
16:23:30 <andythenorth> I guess I draw a rect of tiles or something and monte carlo them
16:24:05 <andythenorth> how do I undo terraforming in GS?
16:24:18 <andythenorth> keep the tile height and contents?
16:24:27 <andythenorth> then rebuild it?
16:24:42 <andythenorth> I need to terraform tile for a specific industry layout
16:24:54 <andythenorth> I can only go tile-by-tile, and the terraforming might fail
16:25:00 <andythenorth> so I need to be able to revert
16:25:18 <andythenorth> GS can place houses and stuff I assume
16:25:33 <TrueBrain> frosch: I don't get what Victoria it should be π
16:26:01 <frosch> both refer to lake victoria in africa
16:26:13 <TrueBrain> ah, not known in 3166, fine, will mute it π
16:28:24 <TrueBrain> all 4 are fixed now
16:34:10 <TrueBrain> will investigate π
16:34:10 *** sla_ro|master has joined #openttd
16:43:14 <TrueBrain> you will find 2 more; fixed
16:43:24 <frosch> that's all, looks awesome π
16:44:02 <TrueBrain> also for checking at this level that you did π
16:44:38 <frosch> there are plenty of scenarios and heightmaps with no tags, which could have some, but author's fault
16:44:47 <TrueBrain> yeah .. so much we can do automated
16:47:12 <frosch> lol, did i just spend 2 hours with this? :p
16:47:22 <frosch> well, time to cook then
16:47:26 <TrueBrain> I spend two days with this ... π
16:48:06 <TrueBrain> ah, found why Merrimack Valley went wrong; the version has other tags than the package
16:49:09 <TrueBrain> so .... I guess we really should remove "regions" from version too
16:49:18 <TrueBrain> but that would be another PR, to collapse name and region
16:55:05 <TrueBrain> okay, I can already do that .. there are no tags worth talking about per version; the globals are always a superset, basically
17:02:58 <TrueBrain> need to rebase that commit on the latest .. but first, dinner π
17:05:00 *** WormnestAndroid has quit IRC (Read error: No route to host)
17:05:09 *** WormnestAndroid has joined #openttd
17:07:42 <frosch> i keep wondering about some form of "fictional" tag, which people can use for their comic-based or alternate-reality stuff
17:07:59 <frosch> nothing you could auto-detect, but may still be useful
17:09:17 <frosch> but "fictional" is so arbitrary... it's may be something a single person created by putting japan and ireland onto the same heightmap next to each other (yes, that exists), or some popular culture dc-universe stuff
17:23:32 *** ChanServ sets mode: +v tokai
17:30:22 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
17:30:39 <andythenorth> I have no idea π
17:31:37 <andythenorth> `foreach (industry_spec in firs["industries"]) {
17:31:37 <andythenorth> local map_curator = industry_spec.economy_variations[firs.get_economy_id()].map_curator;
17:31:37 <andythenorth> firs.map_curators.industry_types.append(map_curator)
17:31:49 <andythenorth> oh maybe it's the fingerprinting
17:32:29 <glx[d]> there's an op limit for loading
17:35:11 <TrueBrain> frosch: I like the idea
17:36:53 <TrueBrain> Just the name is tricky .. maybe something for a later PR π
17:39:16 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
17:43:15 <frosch> i stopped liking the idea :p it's so subjective
17:43:21 *** audigexJon has joined #openttd
17:43:21 <audigexJon> frosch: Possibly because of "Kingston"?
17:45:43 <TrueBrain> Owh, right, I was going to remove the ability to set region per version .. Will be global-only :p
17:46:10 <TrueBrain> Avoids the confusion as with this package π
17:50:36 <Xarick> I don't understand yapf at times
17:55:28 <TrueBrain> why would it be that on one page the CSP does not trigger, and on another it does .. hmm
17:58:42 <TrueBrain> ah ... I set that header for only that one page π Lol
17:58:59 <glx[d]> TallTyler: I marked #10492 so #10545 should be marked too π
18:00:01 <glx[d]> and I think circular reference issue happens in 13.0 too
18:00:03 <TallTyler> Huh, I didn't remove that label
18:00:19 <glx[d]> maybe we set labels at the same time
18:00:27 <TallTyler> I added the Script API label and we must have hit save at the same time
18:01:13 <TallTyler> I'd never remove your labels intentionally π
18:01:42 <glx[d]> unless I use the wrong one π
18:02:18 <TrueBrain> okay, region support is now ready for review π
18:02:19 <TallTyler> I don't know if all the labels are useful, but I try to keep up with them on issues and PRs as one small way I can contribute π
18:02:52 <TrueBrain> and if people could stop making BaNaNaS changes, I can update that PR π (I am not serious; we will find a moment to make the change happen π )
18:08:04 <TrueBrain> Hihi, that now works .. π
18:09:28 <Rubidium> glx[d]: regarding #10545, there are way more throws in that function to consider
18:12:11 <TrueBrain> too bad I cannot demo how the editor looks like, as it is somewhat weird but best I can do atm .. but logging in doesn't work, as ngrok only allows forwarding a single port
18:20:42 <Xarick> this is what autoreplace/autorenew does to my vehicles...
18:26:07 <andythenorth> python list `foo = ["cabbage", "dogs"]`
18:26:39 <andythenorth> chameleon rendering of that in template using `${foo}` is `['cabbage', 'dogs']`
18:27:17 <andythenorth> squirrel wants `["cabbage", "dogs"]`
18:27:58 <Xarick> what is this warning? should I be worried?
18:34:10 <glx[d]> of course only MSVC likes what I wrote
18:38:24 <JGR> In general you'd be better off with values and std::move for most of these string parameters
18:41:29 <Xarick> I find the max_penalty value to be quite short for ships
18:42:52 <Xarick> works fine for roads and rails
18:44:11 <DorpsGek> - Update: Translations from eints (by translators)
18:46:19 <Rubidium> glx[d]: isn't it way easier to just pass the vector as parameter? And potentially statically allocate it in GetEncodedText and clear it there for each call?
18:46:51 *** sla_ro|master has quit IRC ()
18:56:33 *** gelignite has joined #openttd
19:26:43 *** lobstarooo__ has joined #openttd
19:28:00 *** lobstarooo has joined #openttd
19:34:47 *** lobstarooo__ has quit IRC (Ping timeout: 480 seconds)
19:34:51 *** lobstarooo is now known as lobstarooo__
19:39:56 <TrueBrain> so this is happening? π
19:40:02 <TrueBrain> tnx for the reviews frosch π
19:40:35 *** lobstarooo__ has quit IRC (Remote host closed the connection)
19:40:51 *** lobstarooo__ has joined #openttd
19:42:17 <TrueBrain> ugh, exactly when the CI is busy, ofc π
19:43:54 <TrueBrain> 1669 files changed, lol
19:44:02 <TrueBrain> +2000, -10000 lines
19:44:27 <TrueBrain> tags are bloat! π
19:48:22 <LordAro> though i wonder if tags should be kept temporarily, just in case, things need to be reverted
19:48:38 <LordAro> may be easier if the db doesn't need reverting as well
19:49:26 <TrueBrain> so I rather have we remove it in the assumption we no longer needed it, then keep it and completely forget about it π
19:50:03 <andythenorth> oof such GS templating with python π
19:50:09 <andythenorth> keeps me quiet I guess
19:50:13 <andythenorth> writing actual working code
19:53:37 <TrueBrain> okay, staging shows it is all working as expected .. so just need an approval on BaNaNaS to bring this live π \o/ π₯³
19:55:47 <klote[d]> Hi, im tryin to load a specific map during startup of the server
19:57:04 <frosch> some items had an "archived:" at the end, which was reordered alphabetical in that PR
19:57:12 <frosch> i assume we added those manually before
19:57:17 <klote[d]> i added the line load 2 in the on_server.scr document
19:57:17 <TrueBrain> frosch: yup; I assumed the same π
19:57:22 <klote[d]> but it doesnt seem to work
19:57:36 <TrueBrain> frosch: we really should make that a button in the interface π
19:57:43 <klote[d]> but when i directly type load 2 in the rcon it works just fine
19:57:49 <klote[d]> even in the server terminal
20:00:31 <klote[d]> what scr document am i supposed to load command line to start a specific save game when server starts?
20:02:29 <LordAro> klote[d]: not trying to disuade you or anything, but since you're on discord, some of the other help channels may prove more... helpful
20:02:36 <LordAro> this tends to be very dev focused
20:02:57 <LordAro> no idea, i'm not on it :p
20:03:00 <LordAro> but i know they exist
20:03:02 <klote[d]> most help so far was from this openttd.
20:03:15 <TrueBrain> Discord channel #openttd-help: gives the suggestion they want to help π
20:08:26 <TrueBrain> this will disrupt everything for a bit, as the system wasn't designed to process so many mutations at once π
20:15:37 <TallTyler> I didn't realize that this backend improvement to Bananas would be implemented immediately, without having to release a new OpenTTD. Very nice! π
20:19:38 <FLHerne> klote[d]: You can run `openttd -g blah.sav` to load a particular save immediately
20:19:54 <FLHerne> (add -D as usual for dedicated server)
20:19:56 <frosch> TallTyler: a new release would involve in-game dropdowns
20:20:00 <frosch> noone implemented that yet
20:20:02 <TrueBrain> TallTyler: I will give an example (referring to a discussion we had the other day :P)
20:20:26 <klote[d]> FLHerne: Yeah but when i restart server it wont load that specific map
20:20:26 <TrueBrain> TallTyler: I like deploying things without downtime π
20:20:53 <klote[d]> I want it to always load the savegame when game restarts
20:21:58 <FLHerne> then yeah, you need rcon/adminport
20:21:58 <Xarick> what does inline do? explain me like i'm 4 years old
20:22:25 <glx[d]> depending on when the scr is executed, it might happen to do the load command while generating the map
20:22:47 <FLHerne> klote[d]: actually, if you use `restart_game_date` it should restart the same map, I think?
20:23:11 <glx[d]> I know for instance it's not possible do add commands after load in a script
20:23:27 <glx[d]> because they are then ran during the load
20:23:57 <Xarick> Yapf code sometimes have inline, sometimes doesn't
20:24:03 <Xarick> i never know the difference
20:24:28 <glx[d]> compiler knows (inline is a hint for compiler and it may not follow it)
20:25:30 <JGR> inline is more than a hint when it comes to headers, one definition rule, etc
20:26:13 <Xarick> so I never know why or when to use it
20:26:39 <klote[d]> tried adding it to other documents doesnt work lol
20:27:09 <klote[d]> what about autoexec.scr - This Script is executed on gamestart [all - use this for custom aliases per ex.] ?
20:27:43 <klote[d]> allot of servers have ingame commands
20:28:00 <Rubidium> Xarick: I think compilers have become smart enough to not have to add inline everywhere, except when it comes to headers like JGR already said
20:28:03 <klote[d]> is that some sort of bot?
20:29:13 <Rubidium> Xarick: any given the trivialness of those two functions, I reckon both are compiled exactly the same by most compilers
20:30:38 *** sla_ro|master has joined #openttd
20:31:39 <Kuhnovic> I finally got my PR to build the preview but it keeps crashing. I thought it was because of my changes but I tried previews from different PR and I get the same. I build two docks, send a ship to go back and forth between the two, and after some time I get a crash. I tried both Chrome and Edge, same results. Has anybody else had this issue?
20:32:12 <glx[d]> saving crashes for map >= 256x256 tiles
20:33:53 <glx[d]> could still be related to autosave
20:34:52 <glx[d]> oh maybe something with lzma
20:35:57 <Kuhnovic> Turned auto save off, lets give it another go
20:37:09 <glx[d]> as we patch lzma source to build emscripten version
20:37:25 <klote[d]> this wiki is pointing towards a bot called soap but i cant seem to find a download.
20:38:34 <glx[d]> yeah unfortunately it's on openttdcoop
20:42:02 <TrueBrain> owh bah, I don't always filter empty regions .. so now it is committing `regions: []`
20:42:04 <TrueBrain> not that it matters
20:42:06 <TrueBrain> it just annoys me π
20:42:14 <TrueBrain> it did the same with tags, which annoyed me equally
20:43:36 <klote[d]> libottdadmin <- people using this to automate things on servers?
20:44:00 <nielsm> Xarick: in C++ the "inline" keyword does not mean the function should be inlined, it means the function is allowed to be defined multiple times and not cause an error. You need to use "inline" on free functions defined in header files, i.e. functions that have their function body in the .h file, and are not inside a class. For any other cases, the "inline" keyword does nothing.
20:47:29 <glx[d]> it's a lib to connect to server admin port
20:48:12 <Kuhnovic> glx[d]: disabling autosave did the trick. Look mum, no buoys!
20:48:23 <frosch> oh dear.. i expected something like this to happen... you can only have 25 discussion categories :p
20:48:29 <frosch> we have 64 translations
20:49:03 <Xarick> glx[d]: ,it crashes for me for 256x256
20:49:44 <TrueBrain> so, time for a new organization called OpenTTD-translators, and a repository per language? π
20:49:54 <TrueBrain> just for everyone to have a discussion in their own language?
20:50:03 <TrueBrain> such a pro-move of GitHub, removing discussions from teams π
20:50:26 <glx[d]> or tag discussion with a [ISO_CODE]
20:50:58 <glx[d]> and hope translators will be smart
20:51:56 <TallTyler> It works, although I haven't committed any changes that use the new codes yet
20:52:17 <TrueBrain> and sorry, I wanted to give it a spin of what was in my head, but BaNaNaS blabla took prio π
20:53:11 <TallTyler> I think I can continue to run the economy using `_economy_date`, but outside real-time mode, keep that in sync with `_date` in `OnNewDay()` instead of updating them separately
20:53:36 <frosch> ok, back-up plan: check which teams actually discussed anything in the past, and only give them categories :p
20:54:23 <TrueBrain> TallTyler: sounds like you have a plan π
20:55:02 <TallTyler> It's a lot to slog through, but these codes were a big mental hurdle to overcome, so hopefully I can make faster progress
20:55:14 <andythenorth> oof do I want to write a recursive python->squirrel list converter? π
20:55:18 <TrueBrain> welcome to development π
20:55:20 <andythenorth> I do not like recursive things π
20:56:46 <frosch> shall i include those language, where a poor soul posted 1 post, and noone replied? π
20:58:08 <TrueBrain> frosch: nah; do the important ones first π
20:59:47 <nielsm> TallTyler: I don't know the full context of it, but in that text stuff with choice between calendar/realtime units, be careful if those stand-alone STR_TIME_MONTH_SINGULAR/STR_TIME_MONTHS_PLURAL can be used in a context together with concrete numbers since the plurals can be much more complex then
20:59:56 <TallTyler> I'm getting a weird crash when closing OpenTTD. Am I reading this right that `AI::Uninitialize` is causing the crash, as it's at the top of the stack trace?
20:59:56 <TallTyler> `[00] openttd 0x00007FF786EB9BFB AI::Uninitialize + 427 (C:\Users\tyler\Documents\GitHub\OpenTTD\src\ai\ai_core.cpp:195)`
21:00:18 <TallTyler> (if I'm right, maybe it's not something I broke :P)
21:03:03 <glx[d]> I have `delete _settings_newgame.ai_config[c];` for this line
21:03:27 <TallTyler> nielsm: The standalone units are mostly used in setting names and helptexts where either no number exists, or the number is hardcoded into the string. I can't replace the latter (as far as I'm aware) because I can't SetDParam()s for these strings, as they're only referenced in the setting .ini file
21:04:32 <frosch> 6 teams used the discussions more than once, in 6 teams a single person made a post without nay reply, in 3 teams the only post is from me for some admin stuff :p
21:04:45 <klote[d]> Can any one point me towards a openttd server bot that allows ingame commands and automated messages
21:05:03 <klote[d]> or are people making this themselves?
21:05:33 <glx[d]> I think most write admin bots tailored to their server
21:05:40 <TallTyler> glx[d]: Same, and I should note that I can't reliably reproduce the crash
21:06:30 <TallTyler> So far it's only when exiting the game from the intro menu, but not every time
21:06:31 <nielsm> TallTyler: yeah that's what I thought too, it just feels dangerous to put those strings as standalone... I wonder if it would be better to introduce a formatting code that selects between two strings based on the realtime on/off setting. that will allow translators to put whatever form is appropriate within strings (like grammatical case might differ?)
21:07:32 <TallTyler> I don't quite understand your proposed solution. Could you write some psuedocode for what the strings might look like?
21:07:48 <nielsm> also that could solve issues that might occur if the words sometimes need to occur at the start of a sentence and would need to be capitalized, instead of having to need even more variations of those static strings
21:09:41 <nielsm> like a graph legend text: "Payment for one unit of cargo delivered in number of {RTS "days" "seconds"}"
21:09:45 <frosch> on the plus side: the translation that used the discussions are all from stable countries, so I can use flags again
21:10:27 <nielsm> "{RTS "Yearly" "Period-wise"} check enabled"
21:10:54 <nielsm> RTS = Real-Time Setting, in this idea
21:11:11 <TallTyler> Hmm, that does look significantly better
21:11:43 <TallTyler> Do we have any string codes that take parameters like that?
21:12:00 <nielsm> well the plural and case codes?
21:12:41 <glx[d]> better use plural as reference π
21:13:08 <glx[d]> just ignore the index ref
21:13:08 <JGR> Isn't that likely to result in needing to nest plural/case codes inside your {RTS} block?
21:13:14 <JGR> That could get complicated
21:13:35 <TallTyler> Oh duh, plurals π
21:14:54 <nielsm> JGR: I think this form would only be used for situations that are numeral-less, so nested plurals shouldn't be needed, for the cases where you do have numerals you use the inserts for "{NUM} <timeunit>"
21:16:02 <TallTyler> The unit control codes do that already, to handle unit conversion like 1 day = 2 seconds
21:17:16 <TallTyler> Itβs a bit silly to do that with months = minutes and years = periods since they are both 1:1, but I liked the consistency
21:23:30 *** gelignite has quit IRC (Quit: Stay safe!)
21:27:17 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:31:20 *** sla_ro|master has quit IRC ()
21:41:14 <Xarick> I need to come up with a strategy to deal with autoreplace on my AI. Need a way to not care, or not keep track of all vehicles in a route, but still know that the route has vehicles. I don't want to invoke vehicle lists, because they're going to be cpu hogs over time with the amount of vehicles increasing.
21:41:43 <Xarick> I still need to know which vehicles are not profitable, and that's the complicated part
21:42:39 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:43:30 <TrueBrain> frosch: Nope, seems good!
21:45:02 <TallTyler> The βfirst come first serveβ policy makes sense to me
21:45:56 <frosch> newsletter sent, let's see whether at least team notifications work, and people show up :p
21:50:01 <TallTyler> I got the notification
21:50:28 <TallTyler> en_US translators represent, maybe the most useless team! π
21:50:49 <frosch> i think on devzone we had someone sign up for en_GB
21:51:03 <frosch> we did not offer that this time :p
21:51:54 <Xarick> getting a list of vehicles of a route via list of vehicles sharing orders
21:52:41 <Xarick> what I really wanted though was ScriptVehicleList_Group to be similar in design as ScriptVehicleList_SharedOrders
21:54:21 <Xarick> i could use ScriptVehicleList_SharedOrders if ships were smart enough to reach nearest depots
21:57:49 <Xarick> or I could look into openttd code, and make groups keep track of vehicles in it
22:03:35 <Xarick> if I alter an order in a shared order list to send to depot and stop, all vehicles would go stop, and I only wanted to act on 1 vehicle. If I unshare the vehicle, the vehicle becomes "out of the loop", ScriptVehicleList_SharedOrders becomes ineffective to keep track of vehicles
22:08:44 <frosch> hmm, how are those scripts deployed?
22:10:48 <Xarick> The way I have my AI working right now is to never use autoreplace. It currently creates a list of vehicles per group, and manually do replacements on vehicles of that group. If the vehicles I keep track are suddenly crashed, or belong to another route (because I build more vehicles reusing the id I was tracking, via horrible checks), I remove them from the list to keep it as fresh as possible
22:11:11 <frosch> looks like no deployment needed. stuff is directly triggered from main branch
22:12:53 <Xarick> this way I don't lose track of any vehicle I create, but at the same time, it makes vehicle maintenance, route maintenance slow... not automated, meaning my AI performance is not at the top.
22:13:21 <Xarick> it becomes a big problem when i deal with hundreds of routes
22:13:38 <Xarick> vehicles end up waiting in depots for a replacement for years
22:14:37 <Xarick> if I could find a way to shift this burden to autoreplace and autorenew, my AI would be faster, wouldn't have to deal with these tasks
22:15:24 <TrueBrain> Does seem the repo needs some glx[d] love π
22:16:21 <Xarick> it would compete better, have more time for pathfinding and so on
22:16:47 <frosch> is glx master of check-annotations?
22:17:10 <TrueBrain> He became a master, after fixing so many repos
22:17:45 <glx[d]> another repo needs action bumps ?
22:34:41 <Xarick> TallTyler: found a river at sea level
22:45:51 *** lobstarooo__ has quit IRC (Remote host closed the connection)
22:48:35 <frosch> and master is 7 commits behind main
22:49:18 <TrueBrain> Please don't use master/main
22:50:00 <glx[d]> I copied nml workflow for this one
22:50:13 <TrueBrain> Means we should fix nml too π
22:50:53 <TrueBrain> The idea behind it is, that majors always stay compatible
22:51:08 <TrueBrain> Main ofc can break :p
23:19:29 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
23:21:48 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:33:28 *** WormnestAndroid has joined #openttd
23:46:59 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
continue to next day β΅