IRC logs for #openttd on OFTC at 2020-12-05
⏴ go to previous day
00:03:18 *** Smedles has joined #openttd
00:12:12 *** Delzur has quit IRC (Remote host closed the connection)
00:52:49 *** Progman has quit IRC (Remote host closed the connection)
00:57:17 *** ChanServ sets mode: +v tokai
01:04:14 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
01:48:42 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
02:34:44 *** Wormnest has quit IRC (Quit: Leaving)
03:56:23 *** debdog has quit IRC (Ping timeout: 480 seconds)
05:00:19 *** Smedles has joined #openttd
05:45:21 *** Smedles has joined #openttd
05:50:24 *** Smedles has quit IRC (Remote host closed the connection)
05:54:23 *** Smedles has joined #openttd
05:54:45 *** Smedles has quit IRC (Remote host closed the connection)
05:55:50 *** Smedles has joined #openttd
06:00:02 *** m1cr0man has quit IRC (Quit: G'luck)
06:03:39 *** m1cr0man has joined #openttd
06:52:23 *** JohnnyB has joined #openttd
07:20:10 *** tokai|noir has joined #openttd
07:20:10 *** ChanServ sets mode: +v tokai|noir
07:27:03 *** tokai has quit IRC (Ping timeout: 480 seconds)
07:52:32 *** Smedles has joined #openttd
09:57:23 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8351: Fix: do not add an offset to a nullptr https://git.io/JIGMV
10:00:00 <TrueBrain> lol .. that immediately asserts ... oops ..
10:04:22 *** HerzogDeXtEr has joined #openttd
10:05:08 <TrueBrain> this code turns out to be more special than I expected ... lol
10:06:48 *** Smedles has joined #openttd
10:10:03 <TrueBrain> oops, didn't mean to push :D
10:15:15 <TrueBrain> GetVariableAddress can return a nullptr, awesome :D
10:15:19 <TrueBrain> this is getting better and better :P
10:18:51 <TrueBrain> what a terrible piece of code :D
10:20:25 <Wolf01> Why do games have that size? My new 1TB SSD is already full... and only 10 games are installed there
10:20:25 *** JohnnyB has quit IRC (Read error: Connection reset by peer)
10:21:50 <LordAro> Wolf01: gotta have them 4k graphics
10:21:54 <Wolf01> Even the OpenTTD folder passed 1GB, it used to be really small... that should be 'cause of andy &co. :P
10:22:19 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8351: Fix: do not add an offset to a nullptr https://git.io/JIG92
10:23:57 <Wolf01> I should start to use raid for games, so I can continue to add SSDs, but performance?
10:25:38 *** andythenorth has joined #openttd
10:34:45 <TrueBrain> this small inline function got a bit bigger :(
10:35:27 <TrueBrain> but fuck, this function was unpredictable :P
10:35:48 <LordAro> looks a bit like the sort of thing that was translated from the ASM
10:36:26 <LordAro> it is still quite small with the asserts removed :)
10:36:39 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8351: Fix: do not add an offset to a nullptr https://git.io/JIGQE
10:37:28 <TrueBrain> but so yeah, this could still be UB in some different cases :P
10:38:20 <TrueBrain> and I cannot believe some other code was doing the thing the function was doing .... that just .. :P
10:39:42 <TrueBrain> introduced in a patch that had nothing to do with saveload :P
10:40:09 <andythenorth> how anything ever works ever
10:40:24 <TrueBrain> anyway, I am sure now there is no longer an addition to a nullptr in that function :D
10:40:36 <andythenorth> never watch things being made
10:43:44 <andythenorth> very close to my day job
10:44:09 * andythenorth does *not* sell voting software
10:44:14 <andythenorth> for all the reasons and more
10:50:00 <TrueBrain> owh, I love CMake .. it makes things easier :)
11:16:12 *** Progman has joined #openttd
11:29:07 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:29:20 *** WormnestAndroid has joined #openttd
11:42:50 <TrueBrain> compressed, without OpenGFX, OpenTTD in your browser is now ~2.5MB to download .. that is not bad :D
11:47:54 <andythenorth> I have web pages bigger than that
12:10:00 *** sla_ro|master has joined #openttd
12:22:40 <LordAro> TrueBrain: am ready to try a reset
12:23:49 <LordAro> TrueBrain: rerunning ansible resulted in removing `proxy_hide_header X-Cache-Status;` from the nginx configs, was that something you added (and didn't commit)?
12:24:53 <LordAro> hmm, nope, wasn't in the original PR
12:25:00 <LordAro> must have been something i added at some point
12:25:12 <LordAro> no, i'm being backwards, ignore me
12:29:14 <milek7> TrueBrain: >and on a non-global, with the address of 0. This on its turn returns a nullptr .. and for some reason that works :P
12:29:32 <milek7> I think adding nullptr to nullptr is legal :P
12:47:51 *** virtualrandomnumber has joined #openttd
12:48:07 *** virtualrandomnumber has quit IRC ()
12:51:17 *** arikover has joined #openttd
12:53:45 *** jottyfan has joined #openttd
12:57:06 *** arikover` has joined #openttd
13:02:24 *** arikover has quit IRC (Ping timeout: 480 seconds)
13:02:31 *** jottyfan has quit IRC (Quit: jottyfan)
13:04:10 <TrueBrain> milek7: I was more referring to the fact that function returned a nullptr
13:04:16 <TrueBrain> LordAro: will reset 1 of them now
13:05:10 <TrueBrain> well, I want to test the health-check anyway, so we have to wait like 10 minutes before reprovisioning :D
13:06:29 <TrueBrain> wow, that went .. smooth
13:06:41 *** jottyfan has joined #openttd
13:07:31 <TrueBrain> but but .. you convinced me that it wasn't a good idea! If we provision other VPSes, the IPv6s might have changed :P
13:07:35 <TrueBrain> it is difficult to keep up with that
13:08:07 <TrueBrain> so we convinced each other of the others opinion .. well, great
13:08:49 <LordAro> we could just discard whatever configuration the VPS starts with and set it up with systemd-networkd
13:09:33 <TrueBrain> hmm .. come to think of it, my argument against your PR is a bit thin
13:09:37 <TrueBrain> if we change VPSes, we need to change DNS too
13:09:42 <TrueBrain> so there is already work involved no matter what
13:11:03 <TrueBrain> both VPSes are reinstalled, your SSH key is loaded
13:11:31 <DorpsGek> - Add: IPv6 configuration to ansible (by LordAro)
13:11:49 <LordAro> TrueBrain: any word on that X-Cache-Status header?
13:11:55 <TrueBrain> I never added it, so I dunno :P
13:12:10 <LordAro> i suppose i must have added/removed it at some point
13:12:11 <TrueBrain> it is a local header
13:12:13 <TrueBrain> so you cannot remove it :P
13:12:20 <TrueBrain> it is one added by nginx :)
13:12:45 <TrueBrain> so having that header is correct even
13:12:53 <LordAro> WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!
13:13:55 <TrueBrain> it correctly emailed me twice (sentry)
13:14:03 <TrueBrain> and it will repeat that mail every hour it is still down
13:14:20 <TrueBrain> you know what, every 3 hours
13:14:48 <TrueBrain> funny, OVH deprecated Debian 10 for some reason
13:14:55 <TrueBrain> but a reinstall just brings it back to that state
13:14:57 <TrueBrain> no clue what they mean :P
13:16:46 <LordAro> not sure yet, looks like acme-challenge failed
13:16:52 <TrueBrain> did you provision IPv6 before certbot?
13:17:04 <TrueBrain> as I guess HTTP challenge will fail otherwise :)
13:17:39 <LordAro> possibly? pretty sure it didn't have ipv6 setup the first time i did this
13:17:53 <TrueBrain> but possibly I also didn't have the AAAA records set up :)
13:18:06 <TrueBrain> well, it should fall back to IPv4 of course, but it was the one thing I could think of :)
13:18:32 <LordAro> i'll move ipv6 earlier
13:19:10 <TrueBrain> otherwise it has little reason to fail, I would think :)
13:21:49 <TrueBrain> any detail on the error?
13:22:03 <TrueBrain> IPv6 at least did came alive correctly, so that is good :D
13:24:20 <TrueBrain> firewall issue? If you start a webserver yourself? (python3 -m http.server)
13:24:39 <LordAro> 2020-12-05 13:21:57,739:DEBUG:acme.standalone:Successfully bound to :80 using IPv6
13:24:42 <LordAro> 2020-12-05 13:21:57,740:DEBUG:acme.standalone:Certbot wasn't able to bind to :80 using IPv4, this is often expected due to the dual stack nature of IPv6 socket implementations.
13:25:15 <TrueBrain> but okay, the IPv6 socket should accept IPv4 in that case
13:25:19 <TrueBrain> so it should be fine
13:26:51 <LordAro> http.server only supports ipv6 with python3.8, and the server has 3.7 :p
13:27:05 <TrueBrain> at least you can check if the firewall isn't blocking anything
13:27:13 <LordAro> firewall isn't set up yet
13:27:23 <TrueBrain> just ruling out the defaults
13:27:33 <TrueBrain> who knows what a new version of debian does, etc etc
13:27:45 <TrueBrain> so if you can reach http over IPv4 with the python server, we know that isn't the issue :)
13:28:52 <TrueBrain> you did install firewalld
13:28:58 <TrueBrain> default config possible?
13:29:18 <TrueBrain> (pre-task does apt install firewalld)
13:31:09 <TrueBrain> possibly you want to install everything before the roles, except for the 2 tests
13:31:14 <TrueBrain> not sure if that is possible in ansible :D
13:31:27 <LordAro> it's perfectly possible, just needs some rearchitecting
13:31:33 <TrueBrain> that way, you deploy a hardened system first, install certbot / nginx next, and finish with 2 tests
13:31:44 <LordAro> it's something i had in mind to do anyway
13:32:00 <TrueBrain> great minds think alike :P :P
13:33:14 *** Smedles_ has joined #openttd
13:34:19 *** Smedles has quit IRC (Ping timeout: 480 seconds)
13:37:48 <TrueBrain> hmm, the bootstrap crashes on Linux for me ... "throw logic error"
13:38:18 <LordAro> i suspect it's been quite some time since anyone actually tested it
13:38:45 <TrueBrain> euuhhh ... this code cannot work, I would guess
13:38:51 <TrueBrain> if (!BaseGraphics::SetSet(nullptr)) goto failure;
13:38:57 <TrueBrain> if (name.empty()) {
13:39:01 <TrueBrain> that ...... sounds bad?
13:39:26 <LordAro> where name is the parameter?
13:39:32 * andythenorth doing work at weekends
13:39:46 <LordAro> but .empty would imply that it's not a pointer...
13:40:00 <TrueBrain> bool BaseMedia<Tbase_set>::SetSet(const std::string &name)
13:40:03 <LordAro> is it being coerced to '0' or something?
13:40:22 <LordAro> that should not even compile
13:40:49 <TrueBrain> I don't understand C++ :P
13:40:54 <TrueBrain> but to me that looks very weird
13:41:58 <TrueBrain> it recently changed from "const char *" to "const std::string &"
13:42:14 <TrueBrain> and from StrEmpty(name) to name.empty()
13:42:36 <TrueBrain> but indeed, how does this compile :D
13:42:41 <glx> I think std::string(nullptr) construct an empyt string, but I mau need to check
13:43:05 <LordAro> yeah, coerces to the std::string(const char*) constructor
13:43:16 <TrueBrain> well, that is the error indeed :P But how is it runtime ...
13:43:33 <TrueBrain> okay, so at least I found who to blame :P
13:44:02 *** arikover` has quit IRC (Quit: ERC (IRC client for Emacs 26.3))
13:44:23 <TrueBrain> if (!BaseGraphics::SetSet("")) goto failure; fixed the issue
13:44:26 <TrueBrain> is that a legit fix?
13:45:59 <LordAro> "The behavior is undefined if [s, s + Traits::length(s)) is not a valid range (for example, if s is a null pointer)."
13:46:24 <TrueBrain> sorry LordAro , didn't mean to distract you :P
13:46:34 <LordAro> is alright, i'm just waiting for ansible to run :p
13:46:58 <LordAro> i am surprised there's no compiler warning for that
13:47:27 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8353: Fix a49fdb7ebb: bootstrap crash when trying to load new baseset https://git.io/JIZln
13:47:40 <TrueBrain> bad michi_cc for not testing bootstrap :P :P :P :P (just kidding :D)
13:47:59 <TrueBrain> finding bugs before any beta, w00p :P
13:48:27 <LordAro> bet that's not the only one of those
13:48:53 <TrueBrain> for SetSet() it is, it seems
13:48:59 <michi_cc> Hey, frosch approved it :p
13:49:00 <TrueBrain> but indirect calls, I did not check
13:49:18 <TrueBrain> we do have these: BaseGraphics::SetSet({})
13:49:20 <TrueBrain> no clue what {} is?
13:49:22 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8353: Fix a49fdb7ebb: bootstrap crash when trying to load new baseset https://git.io/JIZl1
13:49:28 *** Progman has quit IRC (Remote host closed the connection)
13:49:32 <LordAro> TrueBrain: basically the same thing
13:49:40 <TrueBrain> so that is the better solution, k
13:49:52 <michi_cc> Probably actually the better way to express an empty string.
13:50:06 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain dismissed a review for pull request #8353: Fix a49fdb7ebb: bootstrap crash when trying to load new baseset https://git.io/JIZl1
13:50:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8353: Fix a49fdb7ebb: bootstrap crash when trying to load new baseset https://git.io/JIZln
13:50:12 <TrueBrain> I always prefer more the same, so .. yeah :)
13:50:17 <michi_cc> It's simply calling the default constructor, just in an lexically unambigous way.
13:50:48 <LordAro> mm, std::string a = ""; is more code than std::string b = {};
13:50:51 <TrueBrain> owh, and in C++ it knows the type .. okay, that would somewhat make sense
13:52:50 *** sla_ro|master has quit IRC ()
13:54:06 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8353: Fix a49fdb7ebb: bootstrap crash when trying to load new baseset https://git.io/JIZ8g
13:54:36 <TrueBrain> now lets see if I can get emscripten to work with a bootstrap :)
13:55:29 <LordAro> TrueBrain: can you reset the servers again? think i've got it working, but they're now in a weird half-state
13:59:55 <LordAro> [4~TrueBrain: also, fun fact - due to new certification requirements, i now have to add a firewall to all our servers at work (even though they're all internal only)
14:00:04 <LordAro> just as well i already know how to do it :)
14:00:23 <TrueBrain> and yes, even internal servers need firewalling
14:00:33 <TrueBrain> I like that certifications demand that :P
14:00:52 <TrueBrain> reducing attack surface \o/
14:03:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8353: Fix a49fdb7ebb: bootstrap crash when trying to load new baseset https://git.io/JIZln
14:03:52 <Timberwolf> andythenorth: that reminds me of videos from wealthy Russians where they buy supercars but live in the middle of nowhere, so upload videos of them driving round muddy fields and dirt tracks.
14:03:53 <TrueBrain> �[1;31mError: Uhm, would you be so kind not to load a NewGRF that makes the 'query' sprite a non-normal sprite?�[0;39m
14:07:55 <andythenorth> Timberwolf also so muddy yes :)
14:08:03 <andythenorth> hmm is flex a thing in css yet?
14:14:32 <TrueBrain> LordAro: not really working? :)
14:15:06 <LordAro> firewall isn't doing what i want it to
14:15:15 <TrueBrain> well, port 80 is open at least :P
14:16:33 <TrueBrain> why didn't we just make a GRF with a-z in it, for bootstrapping? :P
14:18:07 <FLHerne> Hm, ambiguous question?
14:18:29 <FLHerne> It's a thing you can use in CSS unless you have specific customers who insist on *really* stupid browsers
14:18:57 <LordAro> FLHerne: guess who has that?
14:19:10 <andythenorth> guess who has that
14:19:19 <andythenorth> but at some point, things have to give
14:19:39 <andythenorth> IE 11 is EOL August 2021
14:19:41 <LordAro> TrueBrain: so... i need to make "reboot" part of the process...
14:19:54 <FLHerne> I mean, even IE10 has flexbox
14:20:08 <TrueBrain> and that cannot be done in place?
14:20:15 <TrueBrain> rmmod && insmod? :D
14:20:18 <TrueBrain> a reboot is fine, really
14:20:21 <andythenorth> "Partial support is due to large amount of bugs present (see known issues)"
14:20:22 <LordAro> i don't think so? it's kernel-level stuff
14:20:25 <FLHerne> I try to design things to at least look decent in IE11, because a few people still use that
14:20:29 <LordAro> i guess mod stuff might work?
14:20:33 <FLHerne> And it's not too much trouble with flexbox
14:20:43 <TrueBrain> a reboot is fine too; this doesn't have to be gold-plated
14:20:50 <TrueBrain> it just has to be easy to reprovision a VPS
14:20:52 <FLHerne> Certainly easier than trying to do without it at all
14:20:52 <andythenorth> also we have customers, it turns out, who run Edge and friends using the IE 7 compatibility mode
14:21:00 <andythenorth> this increasingly causes lolz issues
14:21:07 <LordAro> i can easily make it conditional on "if iptables has actually been installed"
14:21:23 <FLHerne> "Error: Your browser was configured by a moron. Please fire your IT department."
14:21:29 <TrueBrain> do what is needed LordAro :) It is fine really :)
14:22:03 <andythenorth> FLHerne legacy apps :P
14:22:07 <andythenorth> that can't be updated
14:22:43 <FLHerne> I reiterate my demand to fire the IT dept. :p
14:23:45 <FLHerne> If you can't update even the frontend of your software for over a decade, then no-one knows how it works and you'd be nuts to rely on it for anything
14:26:34 <Timberwolf> It's got better in the last few years as teams have started getting "the CEO wants to know why it doesn't work on their new iPad" tickets.
14:30:26 <LordAro> TrueBrain: one more reset?
14:31:26 <TrueBrain> reset given; 1 is already up, 2 is doing its thing
14:32:27 <TrueBrain> I do not like the host-key-checking ignore, honestly
14:32:52 <LordAro> makes it easy when you're resetting the server so often :p
14:32:58 <LordAro> i'll take it out of the PR though
14:34:06 <TrueBrain> if you make roles, and feel free to say no, possibly add the IPv6 stuff in its own "ovh" role?
14:34:10 <TrueBrain> making it more clear it is ovh-specific?
14:38:02 <TrueBrain> hmm .. do I simply add a ttf to emscripten data, so the bootstrap works .. or do I find another way to load OpenGFX in the local storage ...
14:42:21 <andythenorth> Timberwolf you ever bothered transporting vehicles in your FIRS 4 test games?
14:46:30 <Timberwolf> andythenorth: Lemme check, I definitely got the vehicles chain running in Nentwood transport but not sure if that's 4 or 3.
14:46:52 <andythenorth> I am mostly curious about wagons, and maybe splitting the chain a bit
14:47:06 <andythenorth> I never ever bother transporting the vehicles, not in 10 test games or so
14:47:24 <TrueBrain> owh boy ... found a nasty bug in Firefox .. if emscripten requests a mouse lock, it does so even if you are in another tab
14:48:39 <Timberwolf> I think I mostly did it to test the vehicle sprites, admittedly :)
14:51:01 <Timberwolf> Some of those cars should use the secondary company colour, really.
14:53:55 <Timberwolf> Ah, Cargopositor is repeating a single object, that's why.
14:54:11 <Timberwolf> I should make it able to do recolour ramps based on an array when it's repeating objects.
14:58:39 <LordAro> TrueBrain: so i still had to do a few things manually, but i have updated the playbook to reflect that
14:58:51 <LordAro> you're welcome to reset one more time if you want :)
14:59:16 <LordAro> oh, yes, separate out the ovh stuff
14:59:42 <TrueBrain> there is a TODO in there too :P
14:59:46 <TrueBrain> I would not do that
15:00:03 <TrueBrain> "# Do this super early" <- not a useful comment :)
15:00:19 <TrueBrain> those TODOs are completely pointless
15:00:23 <TrueBrain> as either you will never do it
15:00:27 <TrueBrain> or by the time you do it, you changed your mind
15:00:32 <TrueBrain> it only confuses people ;)
15:01:29 <TrueBrain> same as with "Unlikely to be necessary after upgrading to bullseye" ... so you aren't sure? So why mention it? Only confuses people :P
15:02:11 <TrueBrain> let me know when you are done splitting roles, I will reset one more time :D
15:04:18 <TrueBrain> reinstall done on both
15:04:27 <LordAro> i was running ansible when you did that :p
15:04:37 <LordAro> luckily, because i got rid of the known-hosts hack, it failed :p
15:05:26 <TrueBrain> bananas-server is picking up the mirrors as they come and go, that works pretty well honestly
15:05:51 *** frosch123 has joined #openttd
15:09:05 <TrueBrain> so the emscripten result becomes 600KB bigger because of the bootstrap (requires freetype and a ttf)
15:09:13 <TrueBrain> not sure that is worth it for a one-off thing :D
15:11:07 <LordAro> reset iptables -> reboot doesn't work so well, because the reboot puts all the old rules back again
15:12:29 <TrueBrain> okay, I am going to see if I can do this in Javascript .. we just don't have OpenGFX somewhere not compressed .. which is a bit annoying :D
15:13:29 <supermop_Home> andythenorth I have only managed to get around to making vehicles in 1-2 games
15:14:13 <supermop_Home> once in a desert game where I needed to deliver them to make towns grow at all
15:16:35 *** Progman has joined #openttd
15:18:42 <TrueBrain> LordAro: right, I am off for a bit; before I do, do you want me to reset them once more, or are you good now? :D
15:19:09 <LordAro> TrueBrain: i think we're done :)
15:19:27 <LordAro> just 2 minor tweaks to push to the PR
15:19:41 <TrueBrain> cool; I will review tonight, and bring this to production tomorrow :)
15:20:45 <LordAro> that firewall is not doing anything...
15:34:17 <LordAro> TrueBrain: think i've fixed - reset pls whenever you're around, no rush
15:34:25 * LordAro goes to do house work or something
15:49:40 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
16:07:55 <FLHerne> Any idea why Black insists on adding a blank line after one specific docstring, but not the others?
16:16:31 *** Progman has quit IRC (Remote host closed the connection)
16:20:19 <andythenorth> Timberwolf are you recolouring cargos with recolour sprites, or in your compositor?
16:22:20 <DorpsGek> [OpenTTD/nml] frosch123 opened pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
16:23:44 <Timberwolf> Road Vehicles was all in the compositor, because I hadn't figured out how to do sprite stacking. Means it's quite wasteful of file size :/
16:24:10 <andythenorth> I considered doing it recolour sprites to save compile time
16:24:19 <andythenorth> but testing it would be so painful
16:24:25 <FLHerne> frosch123: Should probably be part of `make test`, and include flake8?
16:24:49 <FLHerne> And then we can drop the special-case github action, because we're running `make test` anyway...
16:25:06 <FLHerne> Hm, maybe not if we want it to mutate the source
16:26:14 <FLHerne> The --check version should probably be in there
16:28:20 * andythenorth considers splitting vehicles into 'cars' and 'trucks'
16:28:24 <andythenorth> and having 2 auto plants
16:30:55 <Timberwolf> (The pink is so the renderer knows what should be occluded by the wagon)
16:33:41 <DorpsGek> [OpenTTD/nml] FLHerne commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZFO
16:36:27 <andythenorth> classic use of pink
16:36:31 <andythenorth> I have it for masks
16:41:12 <frosch123> hmm, flake8 prints a lot of issues for nml
16:41:40 <frosch123> somehow the .flake8 does not work
16:44:15 *** sla_ro|master has joined #openttd
16:47:45 <frosch123> ok, got a newer flake, that understands that
16:48:12 <frosch123> still have 54 reports though
16:48:56 <FLHerne> .flake8 disables a bunch of things
16:49:44 <FLHerne> frosch123: That looks like you've got some non-standard flake8 plugin(s) installed
16:50:21 <FLHerne> frosch123: Specifically, google says flake8-bugbear
16:50:30 <frosch123> ah, yes, i have that
16:50:33 <FLHerne> Also, some of these might be worth looking at :D
16:50:37 <frosch123> it found important issues in eints
16:51:19 <DorpsGek> [OpenTTD/nml] frosch123 updated pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
16:52:12 <DorpsGek> [OpenTTD/nml] frosch123 commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZAs
16:59:54 *** michi_cc has joined #openttd
16:59:54 *** ChanServ sets mode: +v michi_cc
17:02:33 <DorpsGek> [OpenTTD/nml] FLHerne commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZx5
17:08:53 <DorpsGek> [OpenTTD/nml] FLHerne commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZpo
17:10:40 <DorpsGek> [OpenTTD/nml] frosch123 updated pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
17:10:53 <frosch123> didn't see the black error between all the flake errors :)
17:12:22 <DorpsGek> [OpenTTD/nml] FLHerne commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZhv
17:26:13 <DorpsGek> [OpenTTD/nml] frosch123 updated pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
17:26:14 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
17:26:40 *** WormnestAndroid has joined #openttd
17:27:46 <DorpsGek> [OpenTTD/nml] frosch123 updated pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
17:37:56 <DorpsGek> [OpenTTD/nml] FLHerne commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIneF
17:39:14 <frosch123> he, i had "formatters" first, but my editor squiggled it
17:40:28 <DorpsGek> [OpenTTD/nml] LordAro commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JInvt
17:58:44 <FLHerne> LordAro: I like it using `make test`, because then I can run that and be sure the CI won't complain :p
17:59:02 <FLHerne> You can always run `make regression` if you only want the compiler tests
18:00:12 <LordAro> it's not a test though :p
18:02:56 <FLHerne> Testing whether the code meets the project standards is a valid thing to test
18:08:12 *** gelignite has joined #openttd
18:14:36 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
18:26:51 *** Progman has joined #openttd
18:29:33 *** m1cr0m4n has joined #openttd
18:30:38 *** m1cr0man has quit IRC (Ping timeout: 480 seconds)
18:33:43 <DorpsGek> [OpenTTD/nml] frosch123 updated pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
18:43:39 *** WormnestAndroid has joined #openttd
18:45:40 *** otetede2 has joined #openttd
18:50:51 *** jottyfan has quit IRC (Quit: jottyfan)
18:51:42 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
18:52:09 *** WormnestAndroid has joined #openttd
18:57:17 *** otetede2 has quit IRC (Quit: Nettalk6 - www.ntalk.de)
19:00:14 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
19:09:09 *** WormnestAndroid has joined #openttd
19:38:17 <frosch123> FLHerne: did you look at the bugbear reports? otherwise i'll look at them
19:42:08 <FLHerne> frosch123: I'm going through them
19:42:23 <FLHerne> (was eating for a bit)
19:45:08 <FLHerne> I don't see the point of `B011 Do not call assert False...`
19:45:40 <andythenorth> what does that patch do? :O
19:45:54 <andythenorth> oh it's look ahead
19:45:58 <FLHerne> They shouldn't be hit, so we don't care if -O removes them
19:46:16 <frosch123> FLHerne: i think it's about "assert" being ignored. so code behind a "assert False" becomes reachable
19:46:21 <frosch123> a "throw" would stay
19:47:05 <frosch123> a regular assert is passed in the normal case, so the code behind it is regulary executed
19:47:20 <frosch123> behind a "assert False" follows whatever
19:47:54 <frosch123> so you get an entirely new code path
19:48:36 <TrueBrain> LordAro: both resetted
19:49:01 <frosch123> yay, our tie breaker arrived
19:50:14 <frosch123> TrueBrain: is running "flake8" and "black --check" in "make test" a convenient solution for development, or is it an incredible stupid thing to mess with packagers?
19:50:40 <TrueBrain> both, ish. Haha, not the answer you wanted :D
19:50:43 <LordAro> TrueBrain: no merge yet pls
19:51:00 <TrueBrain> personally, I make my "make test"s always run flake/black too, simply because I want to make sure what ever goes out, is in working order
19:51:11 <TrueBrain> so the CI does it, but also on binary build, it runs flake/black again
19:51:30 <TrueBrain> but it does require that also packagers have that installed
19:52:03 <FLHerne> It might be annoying if you do have extra flake8 plugins installed
19:52:45 <DorpsGek> [OpenTTD/nml] TrueBrain commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIn36
19:52:58 <TrueBrain> FLHerne: absolutely
19:53:04 <TrueBrain> this is why I personally use the GitHub workflow
19:53:13 <TrueBrain> where the test is done by another part than the release
19:53:24 <TrueBrain> the release simply skips the tests, as it was executed after the job that does the tests
19:53:29 <TrueBrain> but .. that is bad if anyone ever does it manually
19:53:53 <TrueBrain> basically, there is no right way here, just preferences :) I would say: run it too often rather than not often enough
19:54:07 <TrueBrain> "make flake" runs black, well, that is a lie :D
19:54:07 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
19:54:38 <frosch123> "pyproejct.toml" wtf
19:54:54 <TrueBrain> yeah ... still not sure if it is a failed project or not
19:55:06 <TrueBrain> some people say: it is a PEP, we will use it
19:55:09 <LordAro> TrueBrain: it works! minor update to PR incoming
19:55:12 <TrueBrain> others say: it is bullshit, I don't use it :P
19:55:31 <TrueBrain> and Sentry triggered the correct emails, nice
19:56:34 <TrueBrain> FLHerne: btw, nobody uses flake8 plugins anymore really :P That makes it all a bit easier :)
19:56:47 <frosch123> so toml is a mixture of ini an yaml
19:56:52 <frosch123> not sure the world needed that
19:57:06 <LordAro> TrueBrain: bugbear does seem quite useful to have though
19:57:14 <LordAro> frosch123: it's basically ini
19:57:21 <TrueBrain> frosch123: NOBODY knows ... it is a fucked-up format
19:57:22 <LordAro> almost no one uses any of the advanced stuff
19:57:29 <TrueBrain> it was one of the "new things" .. total disaster :P
19:58:02 <TrueBrain> LordAro: bugbear is useful, but CodeQL covers that too :)
19:58:47 <DorpsGek> [OpenTTD/nml] TrueBrain approved pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JInsG
19:58:53 <TrueBrain> faster, that is sure
20:00:19 <DorpsGek> [OpenTTD/nml] TrueBrain commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JInsa
20:00:24 <TrueBrain> I couldn't resist :
20:01:00 <TrueBrain> in general, using Makefile is just for old people :P
20:01:05 <TrueBrain> enough trolling for one night :D
20:02:06 <TrueBrain> LordAro: so I can roll the VPSes out to production? (well, tomorrow, not tonight)
20:03:15 <frosch123> luckily we are not so old that we would use perl
20:03:18 <TrueBrain> my workflow is: push to GitHub, create PR, run black locally, fix the problems, get an email telling me I fucked up, pushing again :)
20:03:41 <TrueBrain> I am honestly REALLY happy that I am just not old enough to used Perl
20:03:52 <frosch123> haha, the first version of that pr only contained the modfiying black :p
20:03:56 <TrueBrain> it sounds horrible to be stuck with the thought Perl is a good language
20:04:26 <TrueBrain> and pyproject.toml .. I am still not sure if it is a good idea to use or not
20:04:33 <TrueBrain> the reason I haven't done it yet on the other repositories
20:04:44 <frosch123> it claims to replace setup.py
20:04:44 <TrueBrain> it is just .. ugh ...
20:04:49 <TrueBrain> "net niet", as we say in Dutch
20:04:52 <frosch123> so i do not fancy having both
20:04:57 <TrueBrain> that was the whole idea, yes
20:05:05 <TrueBrain> pip supports both, I believe
20:05:21 <TrueBrain> but pyproject.toml also configures tooling
20:05:21 <frosch123> yeah, but what would it do if there are both :p
20:05:27 <TrueBrain> so it is not uncommon to have both
20:05:47 <TrueBrain> if you only put black in pyproject.toml
20:06:16 <frosch123> the vpses are sooo spammmy
20:06:16 <TrueBrain> but again, I am still on the fence about it myself, so meh
20:06:29 <TrueBrain> frosch123: you wanted alerting! :D
20:06:49 <TrueBrain> you wanted Sentry to tell us when they were offline
20:07:04 <TrueBrain> was your suggestion :P I just executed it :D
20:07:18 <frosch123> very well, my peasant :)
20:07:31 <TrueBrain> and it honestly was a good idea :P
20:07:48 <frosch123> i am sure the lord is also pleased
20:08:24 <frosch123> did sinterklaas bring lots of cookies?
20:08:27 <TrueBrain> ugh, when I just look at pyproject.toml my stomach says: FUCK NO
20:08:44 <TrueBrain> I did not celebrate Sinterklaas ... in my age that no longer happens, really :P
20:08:49 <LordAro> TrueBrain: yes, go for it :)
20:08:58 <TrueBrain> LordAro: sweet! Tnx :D
20:09:41 <DorpsGek> [OpenTTD/nml] frosch123 commented on pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JInGV
20:09:54 <DorpsGek> [OpenTTD/nml] frosch123 merged pull request #174: Add: make target to run black-formatter as it is used in PR checks. https://git.io/JIZ5m
20:11:04 <frosch123> no cooies for old people? being old sucks
20:11:17 <TrueBrain> in so many ways ......
20:11:59 <TrueBrain> I have this mask for covid-bla, with a reindeer on it
20:12:09 <TrueBrain> so I put it on, and put 2 christmas balls on each end
20:12:17 <TrueBrain> I was "too old" for that stuff
20:12:25 <TrueBrain> and I should "act move like someone of my age when going outside"
20:13:25 <andythenorth> and you don't even have a job :P
20:13:50 <TrueBrain> and I haven't shaved in a while
20:14:00 <TrueBrain> perfect for xmas .. I just sit outside, and make some money
20:14:51 *** daspork has joined #openttd
20:16:10 <frosch123> not shaving also works in home office
20:17:03 <frosch123> but beards annoy me, so i never last longer than 5 days or so
20:17:26 <TrueBrain> just sit it out after 5 days for a day or 2
20:17:32 <TrueBrain> and you are good again for a few weeks :P
20:17:50 *** Smedles has joined #openttd
20:18:45 *** WormnestAndroid has joined #openttd
20:30:28 <Wolf01> I haven't shaved since the first days of june :S
20:31:14 <TrueBrain> do bird live in it already?
20:31:35 <Wolf01> Yes, I found an ostrich some days ago
20:31:43 <frosch123> do wolfs switch between summer and winter pelt?
20:33:02 <TrueBrain> glx: I am still happy with your cross-compile solution :) It works very well :D
20:33:13 <TrueBrain> lot more flexible than what we used to do to cross-compile :)
20:33:32 * LordAro made it from ~Feb to August without shaving
20:34:10 <TrueBrain> for some reason I don't think the not shaving was the issue .. but the lack of showering .. like, dude!
20:34:21 <glx> the only issue is you need the same configuration for native and cross build (you can use debug tools with release cross build)
20:35:18 <TrueBrain> glx: I don't follow; I can use a debug host tools to build a release build for a target, not?
20:35:42 <glx> cmake prevents it for some weird reason
20:36:10 <glx> it's hardcoded in the import files
20:36:30 <TrueBrain> well, haven't run into that issue, but I did switch from debug to release a few times already
20:36:39 <TrueBrain> they are even completely different cmake versions
20:37:09 <TrueBrain> but anyway, it is a smooth experience for me .. never had this much fun cross compiling :D
20:37:42 <glx> yeah it's very easy to use
20:38:47 <glx> but I still can't build arm windows version (because there's a #warning and that's not supported by MSVC)
20:39:12 <glx> but commented out it builds fine
20:39:34 <glx> of course I can't test the result
20:40:33 *** natalie[m] has quit IRC (Ping timeout: 480 seconds)
20:40:38 *** grag[m] has quit IRC (Ping timeout: 480 seconds)
20:40:38 *** dekeract[m] has quit IRC (Ping timeout: 480 seconds)
20:40:38 *** jeeg[m] has quit IRC (Ping timeout: 480 seconds)
20:40:38 *** menelaos[m] has quit IRC (Ping timeout: 480 seconds)
20:40:39 *** josef[m] has quit IRC (Ping timeout: 480 seconds)
20:40:43 *** christoph[m] has quit IRC (Ping timeout: 480 seconds)
20:40:43 *** julie[m] has quit IRC (Ping timeout: 480 seconds)
20:40:48 *** hamstonkid[m] has quit IRC (Ping timeout: 480 seconds)
20:40:48 *** shedidthedog[m] has quit IRC (Ping timeout: 480 seconds)
20:40:48 *** paulus[m] has quit IRC (Ping timeout: 480 seconds)
20:40:49 *** udo[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** ciet[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** jeremy[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** joey[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** rudolfs[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** patricia[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** yoltid[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** hylshols7qui[m] has quit IRC (Ping timeout: 480 seconds)
20:40:53 *** dude[m] has quit IRC (Ping timeout: 480 seconds)
20:40:58 *** amal[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** magdalena[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** khavik[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** ircer[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** cawal[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** nartir[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** karoline[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** fjodor[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** phil[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** Aileen[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** ist5shreawf[m] has quit IRC (Ping timeout: 480 seconds)
20:41:08 *** tonyfinn[m] has quit IRC (Ping timeout: 480 seconds)
20:41:09 *** jact[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** patrick[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** philip[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** linda[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** iarp[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** freu[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** einar[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** yoyo[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** blikjeham[m] has quit IRC (Ping timeout: 480 seconds)
20:41:13 *** twom[m] has quit IRC (Ping timeout: 480 seconds)
20:41:14 *** nolep[m] has quit IRC (Ping timeout: 480 seconds)
20:41:18 *** osvaldo[m] has quit IRC (Ping timeout: 480 seconds)
20:41:18 *** olmvnec[m] has quit IRC (Ping timeout: 480 seconds)
20:41:18 *** robert[m] has quit IRC (Ping timeout: 480 seconds)
20:41:18 *** CornsMcGowan[m] has quit IRC (Ping timeout: 480 seconds)
20:41:19 *** natmac[m] has quit IRC (Ping timeout: 480 seconds)
20:41:23 *** cyberjunkie[m] has quit IRC (Ping timeout: 480 seconds)
20:41:23 *** christoph[m]1 has quit IRC (Ping timeout: 480 seconds)
20:41:23 *** emeraldsnorlax[m] has quit IRC (Ping timeout: 480 seconds)
20:41:28 *** Heiki[m] has quit IRC (Ping timeout: 480 seconds)
20:41:29 *** cjmonagle[m] has quit IRC (Ping timeout: 480 seconds)
20:41:32 <glx> worse than openttdcoop ;)
20:41:33 *** yur3shmukcik[m] has quit IRC (Ping timeout: 480 seconds)
20:41:33 *** fiddeldibu[m] has quit IRC (Ping timeout: 480 seconds)
20:41:39 *** igor[m] has quit IRC (Ping timeout: 480 seconds)
20:41:43 *** ad5twoknebor[m] has quit IRC (Ping timeout: 480 seconds)
20:41:58 *** ookfof[m] has quit IRC (Ping timeout: 480 seconds)
20:42:12 <TrueBrain> they really should have more than one exit gateway .. that would make this so much less worse
20:42:18 *** labs[m] has quit IRC (Ping timeout: 480 seconds)
20:42:23 *** gretel[m] has quit IRC (Ping timeout: 480 seconds)
20:42:28 *** karl[m]1 has quit IRC (Ping timeout: 480 seconds)
20:42:28 *** dave[m] has quit IRC (Ping timeout: 480 seconds)
20:42:28 *** elliot[m] has quit IRC (Ping timeout: 480 seconds)
20:42:29 <frosch123> and still there are 100+ idlers here
20:42:33 *** albert[m] has quit IRC (Ping timeout: 480 seconds)
20:42:38 *** pina[m] has quit IRC (Ping timeout: 480 seconds)
20:42:43 *** johanna[m] has quit IRC (Ping timeout: 480 seconds)
20:42:48 *** leward[m] has quit IRC (Ping timeout: 480 seconds)
20:43:00 <glx> frosch123: same in many discord servers :)
20:43:00 <TrueBrain> amazing how much smoother Chrome is with OpenTTD than FireFox
20:43:04 <TrueBrain> it is day and night ...
20:43:18 *** pothyurf[m] has quit IRC (Ping timeout: 480 seconds)
20:43:18 *** glothit7ok[m] has quit IRC (Ping timeout: 480 seconds)
20:43:36 <frosch123> isn't running such things the main reason why chrome was created?
20:44:05 <TrueBrain> no clue about the history, it is just surprising that FireFox hasn't caught up yet
20:44:46 <frosch123> well, at this point using firefox is a political statement
20:45:33 *** dag[m] has quit IRC (Ping timeout: 480 seconds)
20:46:02 <TrueBrain> but the way of building is rather clean, and works with the emscripten supplied docker image
20:46:12 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
20:46:48 *** WormnestAndroid has joined #openttd
20:47:39 *** otetede has joined #openttd
20:48:08 <frosch123> hmm, when you start downloading stuff from bananas, you will hit the local storage quota pretty fast :p
20:48:25 <TrueBrain> few things need fixing, like the error when you load it
20:48:32 <TrueBrain> like .. I disabled sound, why bug me with this
20:49:22 <frosch123> oh, you found a way to capture the mouse :o
20:49:31 <frosch123> i didn't expect such a thing to exist
20:49:52 <TrueBrain> Stadia runs on Chrome, so it had to exist :P
20:49:55 <TrueBrain> but took a bit to find
20:50:07 <TrueBrain> it is still not exactly what I would like; re-entering the window is not correct
20:50:22 <TrueBrain> which sometimes makes it a bit confusing where your mouse is
20:51:23 <TrueBrain> but here CloudSaves come up ... files in local storage can be evicted :)
20:51:31 <TrueBrain> I can already see the mad emails :P
20:51:42 <michi_cc> Are OTTD desktop releases going to switch to Electron now? :p
20:52:17 <TrueBrain> Fastforward will be disabled on these builds .. I was thinking of a message like: "Unlock this feature for 1 euro"
20:52:44 <frosch123> TrueBrain: my content_download folder is 6GB. i think the quote for local stroage is 100MB
20:52:45 <TrueBrain> "Fast-forward is technically not possible in this version; download the Desktop version to use this feature"
20:52:57 <TrueBrain> frosch123: it is 50% of your diskspace, actually
20:53:01 <TrueBrain> with a max of 2GB per group
20:53:12 <TrueBrain> but it is an LRU, so it will sort itself out just fine :P
20:53:46 <frosch123> hmm, my savegame folder is 2GB, 1.6 of those are downloads from flyspray
20:53:54 <TrueBrain> and with 6GB, it means you have 50% of BaNaNaS on your disk :P
20:54:04 <frosch123> 273MB are desync replays
20:54:05 <TrueBrain> (BaNaNAS is 12GB :P)
20:54:17 <michi_cc> Cloud BaNaNaS and cloud savegame storage.
20:54:25 <TrueBrain> BaNaNaS is already on the cloud :P
20:54:33 <frosch123> even more bandwidth
20:54:45 <TrueBrain> render server-side, less bandwidth :P
20:54:59 <TrueBrain> I did consider to look into what it would take to bring OpenTTD to Stadia
20:55:03 <michi_cc> Direct loading via HTTP, not indirect via local storage.
20:55:11 <frosch123> ah, nvm, another 1GB of desync replays
20:55:22 <frosch123> next time my disk is full, i know what to delete
20:55:26 <TrueBrain> michi_cc: you mean, streaming? :D
20:55:39 <TrueBrain> only download the sprites used
20:55:52 <TrueBrain> but seriously, it would be cool to run OpenTTD in Stadia :D
20:55:59 <TrueBrain> just it would require adding controller support
20:56:34 <Xaroth> Playing OpenTTD with a controller sounds.. painful.
20:56:50 <TrueBrain> people play it on Android too
20:56:55 <TrueBrain> how much difference can there be
21:06:26 <andythenorth> oof maybe I should release FIRS 4
21:06:40 <andythenorth> it has open bugs, and the gameplay makes no sense
21:07:42 <frosch123> is there a risk of your maintainers quitting?
21:08:56 <TrueBrain> that answer is always yes. But I guess you want to know how big the risk would be? :P
21:09:01 <andythenorth> doesn't seem to be
21:10:23 *** grag[m] has joined #openttd
21:11:30 <andythenorth> can GS read the town registers?
21:13:49 * andythenorth has half an idea, needs a help
21:14:02 <andythenorth> so I just learnt about the statue rating bonus
21:14:21 <andythenorth> could I generate rating bonus in a town if certain cargos are delivered?
21:15:24 <frosch123> yes, i think gs can do that
21:15:53 <frosch123> hmm, no wait. gs can influence town rating, you meant station rating, right?
21:16:46 <frosch123> looks like gs cannot change station rating, but that could be added
21:16:57 <frosch123> gs-triggered ad campaigns
21:17:34 <andythenorth> I don't know if the idea is good
21:17:56 <andythenorth> but my alternative idea is to boost industry production, but that just gets even more vehicles
21:18:22 <frosch123> well, considering you cannot max-out station ratings unless you transport cargo at 250km/h, and people's ocd requires to reach 100%, i think there is potential for gs
21:18:27 <andythenorth> the station ratings are so silly currently
21:18:53 <andythenorth> increasing ratings for delivering town cargos (goods, alcohol, vehicles) might be valid
21:19:18 <andythenorth> it's kind of nonsensical, but might make a good mechanic
21:20:14 <andythenorth> I guess we love the company most that delivers the beer?
21:21:08 *** christoph[m] has joined #openttd
21:21:54 <frosch123> TrueBrain: is the "attack" on the old server due to you posting about it's old os?
21:22:20 <TrueBrain> frosch123: the machine is off
21:22:47 <frosch123> well, the ip is still assigned to us for a few days
21:22:56 <TrueBrain> but how do you attack an IP that is offline :D
21:23:00 <TrueBrain> anyway, this is why I love OVH
21:23:03 <frosch123> but ok, people would have to know it
21:23:20 <TrueBrain> we don't have to ask for anything, they moved us to their mitigation infra, and they are dealing with the traffic
21:25:21 *** paulus[m] has joined #openttd
21:25:36 *** hamstonkid[m] has joined #openttd
21:25:49 <TrueBrain> now I am curious what kind of attack .. but I don't think they give that information :)
21:25:59 <TrueBrain> I wonder if it is an old service trying to reach the old server and go bananas :P
21:26:03 <TrueBrain> like someone's mail client, or what-ever :D
21:26:05 <FLHerne> But I don't know what it should say :p
21:27:00 *** natalie[m] has joined #openttd
21:28:19 <frosch123> FLHerne: i think the interesting info is constants vs register: version 0 uses contants, both version 1 and 2 use registers
21:29:19 <FLHerne> frosch123: There's a lot of `if version in (0, 1): <stuff> elif version == 2: <blah>` going on
21:29:22 <LordAro> TrueBrain: frosch123: attack? huh?
21:29:23 *** shedidthedog[m] has joined #openttd
21:29:26 <FLHerne> So clearly there are important differences
21:29:44 <frosch123> unless you want to add that version 0 and 1 use industry slots, while version 2 uses cargoids
21:30:07 <frosch123> FLHerne: but not sure, whether duplicating that info from the specs makes any sense
21:30:10 <TrueBrain> LordAro: OVH emailed us there is a DDoS attack going on on one of our IPs
21:30:21 <TrueBrain> and that they are mitigating it
21:30:44 <TrueBrain> well, the IP is offline, as in: the server is powered down
21:30:52 <TrueBrain> so a very odd moment to attack it :P
21:30:57 <TrueBrain> last time this happened was back in 2017 btw
21:31:00 <TrueBrain> but not the first time :)
21:31:09 <FLHerne> frosch123: I think it makes sense
21:31:10 <LordAro> by your message above, against the old IPs rather than the new one?
21:31:25 <FLHerne> Also, why are versions 1 and particularly 0 supported?
21:31:26 <frosch123> FLHerne: sub_in/add_out are "list of values" for v0/1, but "list of pairs of values" for v2
21:32:07 <FLHerne> In what cases are those used instead of 2?
21:32:23 <frosch123> v0/1 are probably used if you use the nml 0.4 syntax
21:32:27 <frosch123> supprot for that was readded
21:32:43 <frosch123> v2 is for the nml 0.5 syntax
21:32:58 <frosch123> v0/1 is mostly the same, it's more of a optimisation thingie
21:33:22 <FLHerne> But 2 functionality is a superset of 1?
21:33:26 <FLHerne> So we should just use that
21:33:49 *** ookfof[m] has joined #openttd
21:33:51 <frosch123> generating v2 for nml 0.4 source is a "hard problem" :)
21:33:53 <FLHerne> I guess that would be a lot of rewriting of the deprecated codepath
21:35:12 *** otetede2 has joined #openttd
21:35:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8355: Add: support for emscripten (play-OpenTTD-in-the-browser) https://git.io/JIn4l
21:35:23 *** khavik[m] has joined #openttd
21:35:38 <frosch123> FLHerne: it would require nml to "understand" quite a bit of the authors intention
21:35:55 <TrueBrain> LordAro: we don't have "new IPs" .. as in, they are AWSes, and change on a regular interval :D
21:36:18 <TrueBrain> well, they could have targetted the new VPSes, I guess, but, no
21:39:07 *** joey[m] has joined #openttd
21:39:28 <frosch123> FLHerne: i think it's easier to split Action2Production into two classes Action2ProductionOld and Action2Production, just like ProduceOld and Produce classes
21:40:10 <frosch123> there is almost no shared code path in that class for old vs new
21:41:34 *** patricia[m] has joined #openttd
21:41:38 *** otetede has quit IRC (Ping timeout: 480 seconds)
21:41:42 <TrueBrain> awh, you cannot upload random extensions on GitHub .. too bad :P
21:43:25 <frosch123> on gist? if you use "git" instead of the webinterface, you can
21:43:45 <TrueBrain> I was trying to upload files in a comment of the Pull Request :P
21:43:49 <TrueBrain> didn't want to add it to git itself :)
21:44:12 <frosch123> oh, images in comments, yeah no idea
21:44:26 <TrueBrain> well .. wasm files :P
21:44:35 <TrueBrain> it does support images :D
21:44:45 <TrueBrain> cannot upload them in a gist either :(
21:44:58 *** rudolfs[m] has joined #openttd
21:44:58 <TrueBrain> guess I have to make a workflow and publish them as artifacts :)
21:45:04 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #8355: Add: support for emscripten (play-OpenTTD-in-the-browser) https://git.io/JInBa
21:47:43 *** otetede2 has quit IRC (Quit: Nettalk6 - www.ntalk.de)
21:48:03 *** ciet[m] has joined #openttd
21:48:10 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8355: Add: support for emscripten (play-OpenTTD-in-the-browser) https://git.io/JInB7
21:48:44 <frosch123> wow, when came that "build prs as wasm for testing" into play? that's and insane/amazing idea
21:48:58 <TrueBrain> I have been saying that for ... months now? :)
21:49:00 *** jeremy[m] has joined #openttd
21:49:09 <TrueBrain> I keep telling you guys, you think too narrow minded about what WASM can change :P
21:49:41 <frosch123> well, i knew about "building prs" and i knew about "playing in browser". but never about putting those together
21:50:09 <TrueBrain> michi_cc jokes about using Electron to bring OpenTTD to the desktop, but ... it does make supporting Mac so much easier :P
21:51:58 <TrueBrain> I need a place to upload the current progress, without it costing me tons of money :P
21:52:14 <frosch123> TrueBrain: no idea why you meantion lzo. lzma is our current savegame format, so if there are bug reports with savegame examples, they would use lzma. but lto is pointless
21:52:17 <TrueBrain> I guess I first need to build the workflow to produce the artifacts, see if that works
21:52:21 <andythenorth> AWS is $0.61 / month
21:52:26 <FLHerne> TrueBrain: Selecting 'Exit' from the menu does not have a nice UX ;-)
21:52:35 <TrueBrain> frosch123: I will strike "try LZO support" from my list :)
21:52:47 <TrueBrain> FLHerne: good point, let me add it to the list
21:53:17 <FLHerne> frosch123: Being able to load old savegames is nice
21:53:40 <FLHerne> And "loads this old savegame wrong, or not at all" is a bug that gets reported every so often
21:53:52 <TrueBrain> frosch123: I guess it requires a change in CMake too, as it now tells me: LZO not found; compiling OpenTTD without LZO is strongly disencouraged
21:54:00 <FLHerne> Probably not a priority though
21:54:29 <TrueBrain> (our CMake files, not CMake itself :P)
21:54:43 *** amal[m] has joined #openttd
21:55:40 <frosch123> FLHerne: old savegames use zlib. lto is only used for ancient savegames
21:55:46 <TrueBrain> adding LZMA is easy btw, milek7 already did that. It is just something external, so I have to see how to do that in such way it is easily reproducable
21:56:14 <FLHerne> s/old/ancient/ in what I said, then :p
21:56:17 <frosch123> FLHerne: we also omitted lzo from bananas scenarios :)
21:56:39 <Timberwolf> andythenorth: I did think about those at one point :)
21:56:50 <andythenorth> they were in Horse 1
21:56:59 <andythenorth> I think I'm going to do it as dual-headed
21:57:25 <andythenorth> I don't like the look of the single unit, dual cab version so much
21:57:35 <TrueBrain> I like how clean the PR is :D
21:57:36 <FLHerne> TrueBrain: fwiw, fast-forward has a noticeable effect in my browser
21:57:39 <TrueBrain> just a few CMake files ... hmmmmmm
21:57:46 <FLHerne> (I guess it's about 2x, which would make sense)
21:57:48 <TrueBrain> FLHerne: open the FPS meter
21:58:00 <TrueBrain> you can see your Hz of your screens there
21:58:09 <andythenorth> supermop_Home looks like my loft right now
21:58:14 <TrueBrain> (mine is 144, so FF goes pretty quick, in comparison)
21:58:55 <andythenorth> supermop_Home really actually spooky close to what I can see, but in OO gauge manifestation
21:59:01 <TrueBrain> so your screens are 60Hz :)
21:59:14 <TrueBrain> and having a fast-forward that is only twice as fast ... that is asking for bug-reports
21:59:18 <FLHerne> I mean, 2x seems useful
21:59:20 <TrueBrain> so I rather disable it, explaining what is going on
21:59:26 <TrueBrain> or keep it enabled, but show an error nevertheless
21:59:32 <FLHerne> Perhaps my perspective is biased by too many slow laptops
21:59:48 <TrueBrain> normally it is a lot faster than 2x :P
21:59:55 *** dekeract[m] has joined #openttd
22:00:06 <TrueBrain> but okay, I can see your point. So I keep it enabled, but I show a warning that FF is limited by the browser
22:00:12 <FLHerne> For most of my games, my old laptop struggled to maintain 1x
22:00:13 <TrueBrain> and that a desktop would most likely have a faster FF
22:00:18 <frosch123> FLHerne: the first revision in git already has zlib
22:00:19 <FLHerne> 2x would have been great
22:00:40 <frosch123> so lzo savegames only existed in the first N months of ottd in 2004
22:00:56 <FLHerne> frosch123: Well, I remember git being an amazing new innovation, oh you mean including the SVN import
22:01:10 <supermop_Home> andy I was going to buy a Kenneth grange product this week but missed the sale
22:01:23 <FLHerne> supermop_Home: This BA stuff?
22:01:59 <TrueBrain> FLHerne: we imported SVN in git :)
22:02:09 <TrueBrain> we only lost 975 revisions of subversion because of stupidity :P
22:02:15 <TrueBrain> (and from BEFORE my time, I want to add)
22:02:25 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #8355: Add: support for emscripten (play-OpenTTD-in-the-browser) https://git.io/JIn0g
22:02:54 <frosch123> TrueBrain: who linked truelight to truebrain on github? :p
22:03:59 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8355: Add: support for emscripten (play-OpenTTD-in-the-browser) https://git.io/JIn0D
22:04:04 <TrueBrain> frosch123: I did, of course :)
22:04:52 <TrueBrain> who else would go through that trouble :P
22:05:27 <milek7> now that you got mouse capture working, you could restore original scroll_mode
22:05:44 <TrueBrain> it doesn't change the mouse position on the host
22:05:49 <TrueBrain> (which scroll_mode 0 tries)
22:06:11 <TrueBrain> you can try it yourself by changing it in the settings
22:06:17 <TrueBrain> it feels terrible :P
22:07:34 <milek7> I thought relative mode would take care of that
22:07:57 <TrueBrain> I am sure there is a way to do this correctly, as Stadia is doing it for many games
22:07:59 *** nielsm has quit IRC (Ping timeout: 480 seconds)
22:08:02 <TrueBrain> I just couldn't figure that part out
22:08:12 <TrueBrain> this was the best I could create which feels okay :)
22:08:54 <TrueBrain> it might be SDL2 port, it might be a setting we do .. dunno :)
22:08:58 <milek7> ffwd could be done with ugly kludge
22:09:03 <milek7> measure time between LoopOnce calls, and call gameloop enough to fill available time
22:09:16 <TrueBrain> we are not going to add hacks :P
22:10:21 <TrueBrain> the main issue you introduce with your suggestion milek7 , is that you are not vsync'd again
22:10:35 <TrueBrain> so the mouse etc will feel weird and strange again
22:10:47 *** philip[m] has joined #openttd
22:10:49 <TrueBrain> (sync'd with Animation Frame)
22:11:25 <TrueBrain> we could use threading, as SDL2 is said to work from a thread too, and use the draw-thread solution
22:11:29 <TrueBrain> but that requires some enginering
22:12:01 <TrueBrain> currently the draw-thread is eager; it would need to be changed to be in lockstep of the Animation Frame
22:12:06 <TrueBrain> not impossible; just a lot of effort :)
22:12:11 *** Aileen[m] has joined #openttd
22:12:32 <TrueBrain> I do not mind some limitations to a web-version at this moment in time, honestly
22:12:43 <milek7> why it wouldn't be vsynced? I wanted to call GameLoop() multiple times during animation frame (but not enough to overflow available frame time)
22:15:24 *** karoline[m] has joined #openttd
22:16:37 <TrueBrain> I don't see that happening; but I love to be proven wrong of course :)
22:16:56 <milek7> (kludge would be in guessing how many times GameLoop could be called based on past performance)
22:17:14 <TrueBrain> you would also have to find out how much time you have; not sure that info is available
22:17:22 <TrueBrain> and I don't know what happens if you overshoot
22:17:28 *** magdalena[m] has joined #openttd
22:17:35 <milek7> yes, that's the kludge :P
22:17:41 <TrueBrain> but ... future-work :)
22:17:42 <FLHerne> If you overshoot you just miss the next frame?
22:17:48 <TrueBrain> I think it becomes sluggy again
22:17:53 <TrueBrain> like you noticed on the version yesterday
22:18:01 <TrueBrain> one could argue that is okay if you are in FF
22:18:37 <FLHerne> It would if you did that consistently
22:18:58 <TrueBrain> shrug needs experimenting
22:19:01 <FLHerne> Missing it once shouldn't be the end of the world
22:19:04 <TrueBrain> my focus is first on getting it complete ;)
22:19:17 <TrueBrain> and I can fully live without FF working :P
22:19:35 *** menelaos[m] has joined #openttd
22:19:43 <FLHerne> You need to get the content-download working, so that people can download aBase and zBase into a temporary storage
22:19:56 <TrueBrain> well, they can store it in memory
22:20:22 <TrueBrain> what works for you?
22:20:28 <TrueBrain> owh, FF = Fast Forward
22:20:43 <TrueBrain> that was a thing before FireFox existed :D
22:24:29 *** linda[m] has joined #openttd
22:26:46 *** iarp[m] has joined #openttd
22:26:57 *** labs[m] has joined #openttd
22:28:31 *** gretel[m] has joined #openttd
22:32:52 *** ist5shreawf[m] has joined #openttd
22:33:39 *** nartir[m] has joined #openttd
22:34:21 *** patrick[m] has joined #openttd
22:36:25 *** fjodor[m] has joined #openttd
22:37:39 *** phil[m] has joined #openttd
22:39:08 <milek7> make[2]: *** No rule to make target '/home/milek7/openttd-ems/src/source/build-host/src/settingsgen/settingsgen', needed by 'src/table/settings.h.timestamp'. Stop.
22:41:04 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
22:41:54 *** WormnestAndroid has joined #openttd
22:42:27 *** blikjeham[m] has joined #openttd
22:44:26 <milek7> I should call these docker commands in root dir?
22:44:32 <TrueBrain> milek7: build the host files via the docker, that might fix it. I will update the info tomorrow
22:45:33 <TrueBrain> There are absolute paths in the files it seems, confusing the build via docker :D
22:45:52 <TrueBrain> It is cute, if you think about it :D
22:50:12 *** Smedles_ has joined #openttd
22:52:47 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
22:53:04 *** WormnestAndroid has joined #openttd
22:54:45 *** tonyfinn[m] has joined #openttd
22:55:04 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:55:08 *** Smedles has quit IRC (Ping timeout: 480 seconds)
22:57:47 *** osvaldo[m] has joined #openttd
22:57:51 *** gelignite has quit IRC (Quit: Stay safe! Stay at home! Stop the chain reaction!)
23:01:26 * andythenorth thinks about bed
23:01:31 *** andythenorth has quit IRC (Quit: andythenorth)
23:04:54 *** m1cr0m4n has quit IRC (Ping timeout: 480 seconds)
23:06:29 *** sla_ro|master has quit IRC ()
23:09:08 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
23:18:25 *** olmvnec[m] has joined #openttd
23:18:30 *** christoph[m]1 has joined #openttd
23:20:24 <DorpsGek> [OpenTTD/OpenTTD] michicc opened pull request #8356: Fix 63ccb36ef3: Crash trying to load TTO/TTD savegames. https://git.io/JInrA
23:21:25 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8356: Fix 63ccb36ef3: Crash trying to load TTO/TTD savegames. https://git.io/JInrj
23:21:36 *** jeeg[m] has joined #openttd
23:25:36 *** dave[m] has joined #openttd
23:28:30 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
23:28:43 *** WormnestAndroid has joined #openttd
23:29:31 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8356: Fix 63ccb36ef3: Crash trying to load TTO/TTD savegames. https://git.io/JInrA
23:31:28 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8347: Change: [NewGRF] Use aircraft property 12 also for helicopters. https://git.io/JInK3
23:36:44 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:37:24 *** freu[m] has joined #openttd
23:37:39 *** cyberjunkie[m] has joined #openttd
23:38:11 *** WormnestAndroid has joined #openttd
23:41:57 <milek7> TrueBrain: using all that warping logic in _cursor.UpdateCursorPosition makes no sense when we want to use relative mode
23:53:33 *** ad5twoknebor[m] has joined #openttd
23:54:31 *** yoltid[m] has joined #openttd
23:56:34 *** m1cr0man has joined #openttd
continue to next day ⏵
