IRC logs for #openttd on OFTC at 2023-02-09
⏴ go to previous day
00:50:23 *** Xaroth51 has quit IRC (Ping timeout: 480 seconds)
01:27:48 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
01:28:33 *** bryjen has quit IRC (Quit: Leaving)
01:31:13 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
02:01:11 *** WormnestAndroid has joined #openttd
02:04:22 *** Wormnest has joined #openttd
02:56:41 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
03:34:48 *** Wormnest has joined #openttd
03:35:33 *** Tirili has quit IRC (Quit: Leaving)
03:47:59 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:50:40 *** D-HUND is now known as debdog
04:24:59 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
04:27:07 *** Wormnest has quit IRC (Quit: Leaving)
04:30:30 *** Tirili has quit IRC (Quit: Leaving)
04:49:23 *** smoke_fumus has joined #openttd
04:57:30 *** WormnestAndroid has joined #openttd
04:57:41 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
04:57:43 *** WormnestAndroid has joined #openttd
05:05:50 *** birdjj has quit IRC (Read error: Connection reset by peer)
05:30:00 *** Flygon has quit IRC (Ping timeout: 480 seconds)
05:58:37 *** birdjj has quit IRC (Ping timeout: 480 seconds)
06:04:08 *** keoz has quit IRC (Ping timeout: 480 seconds)
06:42:36 <pickpacket> is it possible to see income per cargo type?
07:16:47 *** sla_ro|master has joined #openttd
07:57:36 *** HerzogDeXtEr has joined #openttd
09:41:25 *** sla_ro|master has quit IRC ()
10:20:43 *** JohnFranklin has joined #openttd
10:21:06 <Rus> Just a quick question - for the 13.0 release will the .deb files for Linux be released any time soon or are these no longer being put together?
10:21:40 *** Rus is now known as Guest4278
10:21:41 *** Guest4278 is now known as Rus[d]
10:23:44 *** JohnFranklin has left #openttd
10:29:31 <Rus[d]> Thanks for confirming - better re-engineer my docker 😄
11:10:35 <Samu> I've been having 2nd thoughts about the ai settings fiasco, and came to realise that if I'm actually going to have the gsconfig window look more like the ai settings window without the config part, why not just use the ai settings window instead, and thus making my whole PR point invalid
11:11:42 <Samu> but then I wake up and see ppl embracing my PR
11:21:18 <TrueBrain> Rus[d]: Or just use the generic Linux package 🙂
11:22:07 *** sla_ro|master has joined #openttd
11:31:07 <Rus[d]> I'm using the Bataeu docker image that pulls the .deb files from the CDN as a part of the CI/CD build
11:31:07 <Rus[d]> With 13.0 that obviously fails so will him to fix or I may fork it and do it myself with the generic Linux files
11:36:49 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:44:50 <JohnFranklin[d]> What is "km-ish/h"
11:45:05 *** NGC3982_ has quit IRC (Ping timeout: 480 seconds)
11:48:14 *** NGC3982 has joined #openttd
11:49:49 <dP> so it's just about the same as km but not quite
11:50:06 <JohnFranklin[d]> Okay that's why 200 km/h in the source becomes 201 km/h in the game
11:51:20 <dP> yep, unless you use km/h units in the source
12:03:37 *** WizzStar has joined #openttd
12:03:37 <WizzStar> any guides for making own content (newgrf's) ?
12:05:26 <kamnet> Lots of patience and practice. What kind of NewGRF do you want to make?
12:13:44 <WizzStar> i have a lot of patience and alot of time to practice xD an NewGRF for newobjects
12:14:37 <WizzStar> i can pixel art 😉 but dont know how to get it into OTTD
12:24:30 <dP> if you don't mind it being written in python I can probably do the most coding for you
12:24:47 <dP> as I'm developing a framework for making grfs and looking for some real projects to use it on
13:23:38 <Samu> To do: debug window invalidate data showing / hiding break
13:24:45 <Samu> To do 2: hide config buttons from gs config if accessed via debug window
13:27:20 <Samu> To do 3: the propagated thing glx mentioned
13:27:39 <Samu> though i don't really consider it to be a bug
13:28:47 <Samu> I don't know how to show/hide widgets
13:32:18 <DorpsGek> Rubidium_: Error: You don't have permissions to perform this action.
13:33:10 *** Rubidium_ is now known as Rubidium
13:33:51 <DorpsGek> Rubidium: Error: You don't have permissions to perform this action.
14:50:22 <Samu> I tried to understand why some places use EnforcePrecondition and others don't
14:50:39 <Samu> and I thought it was probably to do with DoCommands being run
14:51:10 <Samu> in the majority of cases, if it runs a DoCommand, then it uses EnforcePrecondition
14:51:32 <Samu> if it doesn,'t then it's done without EnforcePrecondition
14:59:33 *** WormnestAndroid has joined #openttd
15:29:25 <Samu> question: if I do (int)value_in_byte with value_in_byte being 255, will i get a -1 or a 255 after the cast?
15:35:20 *** greeter has quit IRC (Remote host closed the connection)
15:41:01 *** greeter has joined #openttd
15:57:35 <Samu> ScriptGroup::GetNumEngines is difficult to fix...
15:57:52 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
15:58:11 <Samu> IsValidEngine returns different values for deity
15:59:44 <Samu> IsValidEngine in the scope of company x
16:15:45 *** Wormnest has joined #openttd
16:31:05 <glx[d]> you'll get 255 samu, it's like (int64)uint32_value
17:09:10 <Samu> same link, can you check description is fine?
17:15:23 <LordAro> glx[d]: depending on the shortcircuiting behaviour ;)
17:16:48 <Samu> i am deciding against fixing the vehicle type issue for now
17:16:55 <Samu> will do it on a separate pr
17:18:00 <Samu> otherwise I repeat the mistake of doing everything in the same pr
17:19:46 <Samu> all and default, yeah my bad
17:27:55 <glx[d]> yeah it's very easy to fall in rabbit hole and try to fix everything
17:29:41 *** CplBurdenR3294 has joined #openttd
17:29:41 <CplBurdenR3294> I fell in to a rabbit hole to try and fix the issues with a game of mine. That was 20 years ago
17:29:45 <CplBurdenR3294> Still modding it 😄
18:07:54 <Samu> hmm, i dont know how to use template
18:09:24 <LordAro> Samu: that's the truest thing you've said all week
18:10:14 <frosch> TrueBrain: do we use wrangler? cloudflare considered it necessary to tell use they are dropping wrangler v1
18:10:31 <TrueBrain> we do, and I got the same email. Good for them 😛
18:11:20 <Samu> what is the criteria for using EnforcePrecondition
18:11:38 <Samu> cus i'm still confused, some places use it, some don't, there's no real consistency about it
18:12:17 *** tokai|noir has joined #openttd
18:12:17 *** ChanServ sets mode: +v tokai|noir
18:14:06 <dP> been fighting them the whole week, trying to make cmserver compile with 13.0
18:14:21 <Samu> from what I see, it's usually used for DoCommands which sets LastError messages
18:14:50 *** herms123 has joined #openttd
18:15:18 <dP> somewhat got code like this working though
18:15:18 <dP> bool Game::pre_command_accepted(const Container &cc, NetworkClientInfo *ci) {
18:15:18 <dP> byte cmd = cc.get_command();
18:15:18 <dP> Company *c = Company::GetIfValid(ci->client_playas);
18:15:18 <dP> //sell check, disable sell check after x years
18:15:20 <dP> if ((cmd == CMD_SELL_VEHICLE || cmd == CMD_DEPOT_SELL_ALL_VEHICLES)
18:15:20 <dP> && _settings_game.citymania.same_depot_sell_years > 0
18:15:22 <dP> && (_cur_year - c->inaugurated_year <= _settings_game.citymania.same_depot_sell_years))
18:15:24 <dP> for (auto &e : cost.cm.events.get<event::VehicleSold>()) {
18:15:26 <dP> if (!CanSellVehicle(cc.tile, e.vehicle_id)) {
18:15:28 <dP> "Recently bought vehicles (<{} years) can only be sold at the same location.",
18:15:30 <dP> _settings_game.citymania.same_depot_sell_years));
18:15:59 <LordAro> mm, lots of very poorly indented lines
18:16:08 <LordAro> TrueBrain: couldn't it at least preserve the indentation? :p
18:16:23 <TrueBrain> sadly, no, as the IRC protocol doesn't allow for that 🙂
18:16:32 <TrueBrain> well, some unicode chars might
18:17:06 <frosch> be careful with fancy unicode chars, oftc likes to ban people using them, assuming only bots use them
18:17:11 <frosch> we learned that the hard way :p
18:17:20 <TrueBrain> well, we are bots on IRC
18:18:32 <TrueBrain> so a patch that checks a version which is unrelated to the problem is needed to fix a problem on the backend of our backend
18:19:05 *** tokai has quit IRC (Ping timeout: 480 seconds)
18:19:20 <LordAro> i'm still pretty confident that that PR is the wrong solution
18:19:33 <TrueBrain> that is why I made a comment to say the same thing, yes
18:20:10 <TrueBrain> which is confirmed by the fact that changing backend on the same codebase works fine 😛
18:20:23 <TrueBrain> so if anything, it should check the backend, not the SDL version
18:21:50 <dP> dP: has two interesting parts:
18:21:50 <dP> - checking command type with automatic typecasting into lambda arg
18:21:50 <dP> - checking command effects with the test run (before executing)
18:22:25 *** herms has quit IRC (Ping timeout: 480 seconds)
18:23:02 <TrueBrain> LordAro: PR in general feels weird .. setting `SDL_SetRelativeMouseMode` constantly also seems like the wrong approach
18:24:10 <TrueBrain> but yeah, it requires someone delving into SDL, Wayland, and mouse wrapping 🙂
18:25:58 <TrueBrain> I am surprised that libcurl + winhttp works on so many system so far .. I expected a tiny bit of CA issues honestly 😛
18:31:11 * glx[d] is looking at using SQInteger in more places in the API and (int32) feels partly wrong here
18:32:14 <frosch> i think using fmt is a good solution to deal with "unknown" integer types like SQInteger
18:33:04 <frosch> OTTD_PRINTF_SIZEX macros are meh 🙂
18:44:04 <TrueBrain> I think that is all we need 😄
18:45:02 <TrueBrain> who wants to do a quick review? 😄
18:45:58 <DorpsGek> - Update: Translations from eints (by translators)
18:47:25 <glx[d]> nothing should change for current openttd
18:47:37 <TrueBrain> should-would-could, but for that we have staging 🙂
18:48:32 <glx[d]> yeah and it will be easy to test (just need to remember what to change/set)
18:49:17 <TrueBrain> first I need to remember how I change a parameter in AWS for a container ... it has been a while 😛
18:54:04 <TrueBrain> I wonder if we can type it, so any attempts at WASM as faster/easier? 😛 😛
19:14:13 *** gelignite has joined #openttd
19:28:48 *** jellyknight has joined #openttd
19:32:12 *** gelignite has quit IRC (Ping timeout: 480 seconds)
19:32:30 *** gelignite has joined #openttd
19:35:10 *** geli has quit IRC (Ping timeout: 480 seconds)
19:37:40 *** jellyknight has quit IRC (Ping timeout: 480 seconds)
19:37:45 <TrueBrain> staging says: all good
19:38:33 <TrueBrain> `* old SSL session ID is stale, removing`
19:38:44 <TrueBrain> pretty sure it could be improved, as it seems it is not reusing the session currently
19:38:47 <TrueBrain> but those are further details 🙂
19:40:48 *** gelignite has quit IRC (Ping timeout: 480 seconds)
19:57:58 <Samu> I think Rubidium wants me to create a template somewhere, and that place seems to be a non-existant script_core.cpp
19:59:04 <pickpacket> I'm playing 13.0 now and it is *seriously* laggy compared to 12.2
19:59:07 <Samu> and i have no idea how to make a template :p
20:01:23 <pickpacket> 12.2 runs smoothly on a 1024x1024 map with 560 vehicles. Sure, top shows it at 102% CPU but it's smooth
20:02:19 <pickpacket> The 13.0 is probably a debug build!
20:04:10 <Samu> wow, seems simple, at a glance
20:09:11 <TrueBrain> okay, https is now enabled for CDN downloads .. frosch : mind checking if it also works for you with my PR?
20:09:18 <TrueBrain> and that it stays on https 😄
20:12:10 <glx[d]> I think we forgot to disable asserts for 13.0
20:13:21 <TrueBrain> did someone not follow the release procedure? 😛
20:15:37 <frosch> downloaded 300MB via https at correct speed
20:16:10 <frosch> the reconnect between each file is noticeable, so downloading many small files would suck, but i guess we don'T have that requirement
20:16:57 <TrueBrain> I think I can improve on that, but Ihave to fiddle with it
20:17:01 <TrueBrain> it should reuse connections
20:17:39 <glx[d]> I confirm https working here too
20:21:22 <frosch> well, it reuses http/2 the connection, not sure why it appear to lag between files
20:21:44 <Samu> what is the encoding used for the source files cpp hpp?
20:21:56 <Samu> i'm creating a new file, never did such thing before!
20:21:57 <TrueBrain> owh, for me it setup a new secure connection on every download
20:21:59 <TrueBrain> which is a bit expensive
20:26:11 <TrueBrain> frosch: happen to know if that also happened before this patch?
20:27:09 <TrueBrain> glx[d]: so 13.1 when?
20:27:45 <glx[d]> I guess no end user will notice 😉
20:28:07 <frosch> hmm, i think the logs are lieing to me, it does reconnect
20:28:52 <TrueBrain> I assumed it would keep the session open on its own, but possibly I need to create a single curl object or something .. have to fiddle with it a bit 🙂
20:43:17 *** Xaroth51 has joined #openttd
20:43:32 *** Xaroth51 is now known as Xaroth
20:43:46 <frosch> the PR is slower overall compared to master. downloading nuts takes 5s in master, 7s in PR
20:44:39 <TrueBrain> yeah, a thread would solve that; but that is something for a follow-up PR, to refactor this whole mechanism 😛
20:44:46 <TrueBrain> but the stutters between downloads? Are they more/less?
20:44:54 <frosch> my complete testset takes 56s in master, 62s in PR
20:45:04 <TrueBrain> okay; that is fully acceptable to me
20:45:04 <frosch> so, may also be random
20:45:22 <TrueBrain> will see if I can fix reusing of sessions; that should improve things slightly
20:45:27 <TrueBrain> and is nicer on the server 😛
20:51:12 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:27:25 <TrueBrain> okay, connection reuse is a lot quicker 😛
21:28:23 <TrueBrain> mind trying this frosch , see if that improves things for you too?
21:33:04 <frosch> 54s, faster than master 🙂
21:33:12 <TrueBrain> also less laggy for me
21:33:16 <TrueBrain> feels a lot smoother at least
21:33:31 <TrueBrain> I still should move this to a thread, but .. yeah .. one problem at the time 🙂
21:33:49 <TrueBrain> right, review, approval, let's ship it! 😄
21:34:19 <frosch> 90ms to download 2 small gs
21:34:41 <TrueBrain> next step would be to move the CDN to Cloudflare R2
21:34:48 <TrueBrain> that should improve latency numbers a lot 😄
21:35:51 <TrueBrain> hmm .. I would need to change the bananas-server so it sends https traffic to one CDN, and http to another .. as Cloudflare doesn't support http
21:36:05 <TrueBrain> but that shouldn't be too hard, and can be done server-side
21:36:39 <TrueBrain> and after 14.0 release, when most people start using https, we can remove the http CDN; older clients will just fall back to custom TCP
21:37:06 <TrueBrain> sounds like a plan! 🙂
22:05:11 <TallTyler> 14.0 whenever NoDL gets merged 😛
22:05:37 <TallTyler> Although I plan to tackle timetable auto separate next
22:08:44 *** sla_ro|master has quit IRC ()
22:10:45 <Samu> it's my first attempt at a template + new file
22:21:20 *** keoz has quit IRC (Ping timeout: 480 seconds)
22:28:36 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:28:46 <Samu> ah, i solved it, i think
22:30:09 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:45:23 <glx[d]> this docker rely on us providing .deb IIRC
22:48:14 <Samu> how do i pass a default value to a template? :(
22:54:45 <Samu> Also tried static void BroadcastNewEvent(CompanyID skip_company = MAX_COMPANIES, Args ... args) {
23:07:13 *** Samu has quit IRC (Quit: Leaving)
continue to next day ⏵