Config
Log for #openttd on 10th May 2021:
Times are UTC Toggle Colours
00:02:42  *** tokai|noir has joined #openttd
00:02:42  *** ChanServ sets mode: +v tokai|noir
00:09:48  *** tokai has quit IRC
00:20:31  *** tokai|noir has quit IRC
02:45:32  *** glx has quit IRC
02:51:23  *** D-HUND has joined #openttd
02:54:46  *** debdog has quit IRC
03:29:25  *** Wormnest has quit IRC
04:05:54  *** smoke_fumus has joined #openttd
04:14:12  *** Flygon has joined #openttd
05:14:48  *** Wuzzy has quit IRC
06:03:06  *** andythenorth has joined #openttd
06:36:05  *** sla_ro|master has joined #openttd
07:10:22  *** HerzogDeXtEr has joined #openttd
07:36:58  <TrueBrain> fixing up a PR with N commits is getting to me .. boy, that are so many recompiles again and again and again ..
07:39:24  * andythenorth sends cookies
07:39:32  <TrueBrain> hmmm ... COOKIE!
07:51:45  *** andythenorth has quit IRC
08:01:22  *** iSoSyS has joined #openttd
08:04:02  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3xB8
08:06:15  *** iSoSyS has quit IRC
08:20:49  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9230: Change: reworded many of the network errors during connect/listen https://git.io/J3x0B
08:22:00  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3x02
08:23:09  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #9229: Fix: some memory/resource leaks https://git.io/J3xEi
08:24:19  <Rubidium> TrueBrain: time for ccache?
08:24:41  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9230: Change: reworded many of the network errors during connect/listen https://git.io/J3x0B
08:25:29  <TrueBrain> Rubidium: I considered it 20 times already, yes
08:25:30  <TrueBrain> but .. lazy
08:26:01  <TrueBrain> gives me time to get a drunk etc .. I guess ..
08:26:21  <Rubidium> have fun with your sword fight on the swiveling chair
08:26:39  <TrueBrain> exactly :P
08:28:35  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3xuR
08:30:04  *** Samu has joined #openttd
08:30:16  *** KenjiE20 has quit IRC
08:43:07  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
08:45:27  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
08:47:15  <TrueBrain> hmm .. strictly seen I should split "connection type" into IPv4 and IPv6
08:47:22  <TrueBrain> as they can have completely different types :)
08:47:32  <TrueBrain> but lets not confuse users with too much info
08:59:00  *** leward[m] has quit IRC
10:29:17  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #9222: Fix to Polish town names generation https://git.io/J3x7a
10:41:43  *** KenjiE20 has joined #openttd
11:21:38  *** urdh has quit IRC
11:39:10  *** andythenorth has joined #openttd
11:39:15  <andythenorth> cheese on toast then?
11:39:35  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #9230: Change: reworded many of the network errors during connect/listen https://git.io/J3pJm
11:39:42  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
11:39:56  <TrueBrain> right, fixed some pesky use-after-free errors ..
11:40:12  <TrueBrain> unique_ptr are  nice, but .. sometimes a bit tricky to make sure you destroy an object AFTER it leaves it :D
11:40:31  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9230: Change: reworded many of the network errors during connect/listen https://git.io/J3x0B
11:40:49  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
11:41:14  <TrueBrain> valgrind and threadanalyzer are now happy \o/
11:41:20  <LordAro> \o/
11:41:29  <Rubidium> next stop coverity?
11:41:45  <TrueBrain> CodeQL .. you can run that more often over the code :P
11:42:25  <Rubidium> well... building with coverity's tools requires quite a long sword fight
11:42:46  <LordAro> how about clang's scan-build ?
11:44:40  <Rubidium> would be interesting if that finds the resource leaks in master
11:45:22  <TrueBrain> right, hopefully I fixed the GC issues glx ran into .. hard to test for me, so I will wait for him :)
11:47:07  *** andythenorth has quit IRC
11:49:25  <TrueBrain> scan-build: No bugs found.
11:49:28  <TrueBrain> I guess that is a lie :P
11:51:26  <Rubidium> well, it's rather saying it's not that good
11:51:47  <LordAro> i think it's got some options that need to be turned on
11:51:51  <TrueBrain> nahhhhh
11:51:55  <TrueBrain> you have to use it correctly
11:51:59  <TrueBrain> that is much more important
11:52:00  <LordAro> or that
11:52:04  <TrueBrain> manual says: scan-build make
11:52:14  <TrueBrain> but .. no .. you have to first make CMake know about it too :P
11:52:17  <LordAro> ah yes
11:52:36  <TrueBrain> hmm .. even still, same result
11:54:45  <Rubidium> LGTM is actually quite good
11:54:56  <TrueBrain> LGTM == CodeQL, is getting there
11:55:01  <TrueBrain> for Python it is about 80% correct
11:55:03  <TrueBrain> 20% bullshit :P
11:55:49  <TrueBrain> cool thing is you can create/use your own set of checks
11:56:38  <TrueBrain> LGTM got bought up by GitHub, and GitHub integrated CodeQL in their interface .. just .. it takes for ever to run .. LGTM still runs the code too on its own .. it is a bit of a hybrid world atm :P
11:57:01  <TrueBrain> ideal we would run CodeQL on every PR, as that reports things that got added
11:57:09  <Rubidium> https://lgtm.com/projects/g/OpenTTD/OpenTTD/snapshot/7912427a57710450047acbede9a753511222ecf6/files/src/order_gui.cpp?sort=name&dir=ASC&mode=heatmap#V557 On first glance it looks like it's BS, but it actually has a point here
11:57:39  <TrueBrain> that is one thing I love about CodeQL in GitHub .. the existing issues, sure, have to be fixed .. but I don't want PRs to introduce new ones
11:57:56  <TrueBrain> Rubidium: lol, yes, we have more of thse; those are fun :)
11:58:02  <TrueBrain> okay, finally have scan-build working with CMake
11:58:44  <TrueBrain> finding potential memory leaks
12:00:17  <Rubidium> though the nasty thing about "fixing" that one is that when somebody makes the scrollbar's capacity/position 32 bits, it's technically broken
12:00:54  <TrueBrain> the function is just a bit weird .. casting to uint .. like .. there checking for < 0 was too much effort?
12:00:57  <TrueBrain> but a bit later you do?
12:01:09  <Rubidium> similarly coverity is complaining quite a bit about memory being uninitialized within ZeroedMemorysomething
12:01:26  <Rubidium> the casting to uint it probably to stop the compiler complaining about signed vs unsigned comparisons
12:01:37  <TrueBrain> well, it does a bit more :P
12:02:05  <TrueBrain> scan-build finds 136 things
12:02:06  <Rubidium> yeah, it does... but that's probably the side effect of the cast and not the intended purpose of the cast
12:02:24  <TrueBrain> but in the end it is the reason LGTM complaints, not?
12:02:35  <Rubidium> yup
12:02:47  <TrueBrain> scan-build mostly finds dead-code
12:02:54  <TrueBrain> 21 memleaks
12:02:59  <TrueBrain> 35 dereference of null pointer
12:04:06  <TrueBrain> it too has issues with our ZeroedMemoryAllocator
12:05:15  <TrueBrain> https://user-images.githubusercontent.com/1663690/117656512-bb45ef00-b198-11eb-864f-6741cb00dc05.png
12:05:18  <TrueBrain> reporting is pretty well done
12:05:47  <Rubidium> coverity has that one too ;)
12:05:54  <TrueBrain> they all do I hope
12:06:01  <TrueBrain> :D
12:07:02  <Rubidium> CodeQL doesn't
12:07:30  <TrueBrain> :o
12:08:08  <Rubidium> CodeQL seems to be looking more at the "math" side of thing and less at the memory things
12:11:00  <TrueBrain> scan-build reports a few that neither of the other two report
12:11:12  <TrueBrain> haven't evaluated if it is correct :P
12:13:32  <TrueBrain> LordAro: you should let CodeQL run on master too for FreeRCT :P
12:13:33  <TrueBrain> now it cannot diff
12:14:41  <LordAro> ha
12:17:34  <TrueBrain> https://github.com/TrueBrain/OpenTTD/actions/runs/827983141 <- lets see if my copy/paste works :P
12:22:46  <TrueBrain> FreeRCT compiles a lot quicker :P
12:27:28  *** supermop_Home has joined #openttd
12:29:31  *** glx has joined #openttd
12:29:31  *** ChanServ sets mode: +v glx
12:38:34  <TrueBrain> https://github.com/OpenTTD/OpenTTD/pull/9017 <- glx, new version; should now correctly detect your direct-IP, even over IPv4 :P
12:38:50  <TrueBrain> no promises, as I cannot really test that scenario :P
12:39:05  <glx> I'll redo my different tests :)
12:40:08  <TrueBrain> 10 minutes for CodeQL to run on our code .. well, still faster than building MacOS
12:40:20  <LordAro> ha
12:40:24  <TrueBrain> 81 alerts
12:40:31  <glx> but I was surprised by the IPv6 firewall
12:40:39  <LordAro> TrueBrain: https://github.com/FreeRCT/FreeRCT/pull/201 this is your fault
12:40:47  <TrueBrain> LordAro: yeah, I got the ping, tnx for pinging again :P
12:40:48  <TrueBrain> :D
12:40:56  <LordAro> TrueBrain: what ping?
12:40:57  <TrueBrain> glx: yeah ... shitty consumer hardware being shitty
12:40:58  <LordAro> :p
12:41:14  <TrueBrain> CodeQL reports a lot of "Multiplication result converted to larger type "
12:41:25  <TrueBrain> also "Comparison of narrow type with wide type in loop condition "
12:41:28  <TrueBrain> and that is about it
12:41:52  <TrueBrain> so the default CodeQL rules are -not- useful :P
12:42:06  <TrueBrain> can't remember how to load the extended set ..
12:42:12  <glx> at least it's a progress, my previous router didn't have the option at all, every IPv6 aware device was freely accessible from the outside
12:42:38  <TrueBrain> yeah .. I talked about the shittyness of IPv6 deployment a few days ago .. I am not going to repeat that rant :D :D
12:43:11  <glx> but I'd like to have some fine tuning to allow stuff to pass through
12:43:32  <TrueBrain> there is a reason I put a Unity Security Gateway after my ISP router
12:43:35  <TrueBrain> they can go fuck themselves
12:44:26  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9217: Add: [Actions] Check CI annotations to detect compile warnings https://git.io/J3pG5
12:44:29  <TrueBrain> Unify
12:44:29  <Rubidium> TrueBrain: the "Comparison of narrow type with wide type in loop condition" is actually useful; it triggers on the following: for (byte i = 0; i < 256; i++) {}
12:44:30  <TrueBrain> typing is hard
12:44:57  <TrueBrain> Rubidium: I am sure they are all valid complaints .. I was more complaining that what CodeQL does by default is a very narrow set of validation
12:45:05  <TrueBrain> bit too narrow :P
12:45:34  <glx> TrueBrain: the exit step ?
12:45:49  <TrueBrain> glx: that whole step, it does .. thing ...
12:45:50  <TrueBrain> :P
12:45:55  <glx> ah
12:48:08  <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #9217: Add: [Actions] Check CI annotations to detect compile warnings https://git.io/J3Q7w
12:48:42  <glx> it just does stuff on arrays extracted from json :)
12:48:54  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9214: Cleanup: [Fluidsynth] Remove fluid_player_join https://git.io/J3pZH
12:48:55  <TrueBrain> still unreadable :P
12:49:14  <glx> first one merges 2 arrays in one, line by line
12:49:28  <glx> second one adds all values
12:49:52  <TrueBrain> yeah, you can stop trying to explain :P Pretty sure it will never become readable to me :D :D
12:49:53  <TrueBrain> its fine
12:49:55  <glx> but I aggree, jq is unreadable
12:53:58  <TrueBrain> too bad it is hard to comment on commit messages in GitHub .. as frosch has been rather creative with them :P Lets see, how to do this ..
12:58:20  <glx> oh something is different with #9017, trying IPv6 only, without firewall it says "Public", with it it's "Remote connections blocked"
12:58:40  <glx> was "Behind NAT" before
13:02:19  <glx> IPv4 only (and dual stack) is now "Public" in both cases
13:10:36  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9227: Make Window destruction not rely on undefined behavior https://git.io/J3p8e
13:11:10  <glx> need to check something with previous version
13:11:18  <TrueBrain> hmm .. so if you don't have IPv4 and your firewall is closed, it says remove connection blocked
13:11:28  <TrueBrain> can you run the game with -dnet=5 and show me the output?
13:11:57  <glx> I'm building previous version to test something with IPv4 and windows firewall
13:12:17  <TrueBrain> you cannot revert the GC by going back a version :P
13:12:24  <TrueBrain> so not sure what you expect to see different there :D
13:13:15  <glx> I don't remember if I tried IPv4 without firewall rule (or blocked connections) yesterday
13:13:25  <TrueBrain> yesterday doesn't matter :P
13:13:28  <TrueBrain> the binary hardly changed
13:13:36  <TrueBrain> binary -> code, what-ever :)
13:13:46  <glx> because now it says "remote blocked"
13:22:57  <TrueBrain> there you go
13:23:03  <TrueBrain> my timing was a bit too strict
13:24:02  <glx> ah yes it's back to "Behind NAT"
13:27:15  <glx> ok IPv4 only and IPv6 only work as expected, no trying dual with different firewall status
13:31:51  <glx> both open -> public, IPv4 open -> public, IPv6 open -> Behind NAT, both closed -> Behind NAT
13:32:07  <TrueBrain> can you setup the IPv6 one?
13:32:11  <glx> seems correct as when IPv6 is open it depends on windows firewall
13:32:14  <TrueBrain> so I can see what the GC think
13:32:24  <TrueBrain> owh, so it is not really open :P
13:32:41  <glx> I mean open on the router yeah
13:33:27  <glx> and for IPv4 I'm just talking about windows firewall ;)
13:33:39  <glx> of course it touches IPv6 too
13:33:48  *** nielsm has joined #openttd
13:34:43  <TrueBrain> I am making small changes to the GC btw, so a lot of restarts; but things should auto-recover :)
13:36:39  <glx> yeah I can see connection lost from coordinator :)
13:36:51  <glx> but it comes back quickly
13:37:00  <TrueBrain> made the invite codes a bit smaller
13:37:05  <TrueBrain> looks a bit more sensible now
13:37:51  <Timberwolf> There's a feature for the mobile client, "scan this QR code to join a game"
13:38:23  <TrueBrain> it is too bad we don't have anything to copy stuff to the clipboard .. that I would like in-game :)
13:38:58  <TrueBrain> (we only read from the clipboard from what I can tell)
13:39:16  <glx> oh invite code is not easily accessed from dedicated console
13:39:24  <TrueBrain> huh?
13:39:40  <glx> it's printed when registered
13:39:41  <TrueBrain> you don't know how to copy/paste stuff from a terminal in Windows? :P
13:40:04  <glx> but what if there's a lot of logs after it ?
13:40:44  <TrueBrain> most common usecase is, I assume: you start the dedicated server, you copy the invite code, and give it to people :)
13:40:52  <TrueBrain> only after that, there will be log-lines
13:41:03  <TrueBrain> when nobody is on the server, there won't be anything to log, I would guess
13:42:35  <glx> maybe it could be added to server_info command
13:42:38  <TrueBrain> anyway, I will also add it to the console command "server_info"
13:42:44  <TrueBrain> but I doubt it will see much use :)
13:44:53  <glx> anyway it seems to work fine with the weird configurations I can setup :)
13:45:17  <TrueBrain> good! Now try connecting too :P
13:46:07  <TrueBrain> we really should do a play-test tbh .. just invite random people .. just to see how often STUN really works ..
13:49:20  <TrueBrain> https://user-images.githubusercontent.com/1663690/117669526-4332f580-b1a7-11eb-9285-dc0c75204b33.png <- some emails are cute
13:49:34  <TrueBrain> sadly, even our changed contact page doesn't prevent people from emailing us ..
13:49:40  <TrueBrain> worst part, he is not even using 1.11 ..
13:50:25  <glx> oh snowline issue I guess
13:50:32  <TrueBrain> ofc
13:54:19  <glx> ok with everything blocked, connecting to the server works
13:58:29  <glx> and with windows firewall open, and IPv6 blocked on router it uses direct IPv4 connection
14:00:23  <TrueBrain> hmm, this setup is odd .. it should use IPv6 now, and it is using IPv4 ..
14:00:43  <TrueBrain> well, possibly IPv6 simply doesn't work via the public IPv6 :D
14:01:13  <glx> hmm let's see if I can close IPv4 easily on the router
14:01:50  <TrueBrain> let me test if I can join via IPv6
14:02:14  <glx> ok I closed IPv4 I think
14:02:54  <TrueBrain> ypi dod
14:02:56  <TrueBrain> you did
14:03:00  <glx> I like this new box, no need to reboot to apply changes
14:03:02  <TrueBrain> okay, IPv6 does work for me at least
14:03:13  <TrueBrain> owh, via STUN
14:03:15  <TrueBrain> that is odd
14:03:20  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #9214: Cleanup: [Fluidsynth] Remove fluid_player_join https://git.io/J3QE0
14:03:35  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #9229: Fix: some memory/resource leaks https://git.io/J3NM8
14:03:57  <glx> yeah I can see it using STUN for my client too
14:04:15  <TrueBrain> one thing I do not do yet .. I just test if I can TCP connect to IPv6:port
14:04:19  <TrueBrain> not if it can reach the actual server
14:04:41  <TrueBrain> hmm, no
14:04:44  <TrueBrain> owh, I think I know why
14:04:46  <TrueBrain> give me a sec
14:05:40  <TrueBrain> there we go!
14:05:53  <TrueBrain> and if you open IPv4 gain, it should still prefer IPv6
14:05:54  <glx> yup direct connection
14:06:12  <TrueBrain> yeah, it was the weird :: error you had yesterday ..
14:06:17  <TrueBrain> I was sending the IPv6 like any normal system
14:06:21  <TrueBrain> but OpenTTD wants [] around it
14:06:24  <TrueBrain> otherwise it gets confused :P
14:08:38  <glx> reopened IPv4, should I restart the server to test ?
14:08:49  <TrueBrain> or change to local back to public, but yeah
14:08:53  <TrueBrain> it doesn't retest a server that is online
14:09:37  <TrueBrain> yup, IPv6 first :D
14:09:38  <TrueBrain> sweet
14:09:55  <TrueBrain> and IPv4 if it has to
14:10:07  <glx> ok blocked IPv6 now
14:10:43  <TrueBrain> took a bit longer, as timeouts .. but that falls back correctly
14:11:54  <glx> oh of course it still used direct connection for me, I'm behind the router
14:30:19  *** WormnestAndroid has quit IRC
14:30:32  *** WormnestAndroid has joined #openttd
14:40:48  <Rubidium> glx: is https://github.com/OpenTTD/catcodec/pull/1 still valid/up-to-date?
14:41:39  <DorpsGek> [OpenTTD/OpenTTD] telk5093 opened issue #9231: ``screenshot minimap`` does not generates files in dedicated server https://git.io/J3pD4
14:41:39  <glx> it should
14:41:54  <glx> it's missing many targets though
14:42:24  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9232: Fix 91b8ce07: dedicated servers could no longer create screenshots https://git.io/J3pDg
14:42:32  <LordAro> lol
14:42:51  <LordAro> oh i see, you were ready for it :p
14:43:00  <TrueBrain> honestly, I was just too late :P
14:43:02  <LordAro> i thought that was fast, even for you
14:43:11  <TrueBrain> I wanted to have the PR before the issue :D
14:44:18  <TrueBrain> I love how the diff concluded the wrong thing
14:44:51  <glx> lol and dedicated doesn't build :)
14:45:16  <glx> oh not your code
14:45:23  <TrueBrain> that is a nice show of irony :)
14:47:02  <TrueBrain> wow, cancel workflow is slow as fuck
14:50:47  <TrueBrain> took me a long time before I understood he meant minimap screenshots btw :P Was funny :D
14:50:55  <TrueBrain> I was like: no, you cannot make screenshots on a dedicated server, there is no blitter
14:50:59  <TrueBrain> he: but it was working before!
14:51:04  <TrueBrain> yeah ... minimap screenshots :D
14:51:07  <TrueBrain> language is hard :P
14:51:18  <TrueBrain> I should have asked what screenshot, instead of assume :)
14:52:19  <glx> minimap, the most unusable feature on modern monitors ;)
14:53:01  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9220: 1.11.2: test suite is failing https://git.io/J35st
14:53:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #9220: 1.11.2: test suite is failing https://git.io/J35st
14:53:14  <TrueBrain> it looks good on a website :)
14:53:38  <glx> in game it's just too small, and unzoomable
14:55:54  <glx> check annotations works fine ;)
14:56:08  <TrueBrain> yeah ... other parts don't :P
14:56:11  <TrueBrain> stupid apt-update
14:57:04  <DorpsGek> [OpenTTD/OpenTTD] kloczek commented on issue #9220: 1.11.2: test suite is failing https://git.io/J35st
14:58:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9220: 1.11.2: test suite is failing https://git.io/J35st
15:04:17  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9188: `autosave/netsave.sav` is not numbered (unlike regular `autosave/autosave#.sav`) https://git.io/J3guT
15:06:03  <LordAro> catcodec is missing a dorpsgek file ^^
15:06:17  <TrueBrain> has nothing to do with 9188, so that is an odd ^^ :P
15:06:19  <TrueBrain> :D :D
15:06:29  <LordAro> ^^ being eyes, not a '^'
15:06:58  <TrueBrain> ^-^
15:08:46  <peter1138> o_O
15:09:24  <DorpsGek> [OpenTTD/OpenTTD] embeddedt opened issue #9233: Online content not working cross-domain in Emscripten https://git.io/J3p5u
15:09:33  <glx> should we add annotations as required ?
15:11:49  <glx> hmm actually we could require annotations and remove others from the list, as annotations fails if any of the build fails
15:12:48  <glx> ah and dedicated is not required ;)
15:13:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9089: New setting: Grow cities around stations https://git.io/JOyyW
15:13:07  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #9089: New setting: Grow cities around stations https://git.io/JOyyW
15:13:12  <TrueBrain> glx: I wouldn't for now
15:13:14  <TrueBrain> lets first see how it goes
15:20:31  *** sla_ro|master has quit IRC
15:21:59  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9233: Online content not working cross-domain in Emscripten https://git.io/J3p5u
15:22:06  <TrueBrain> "cross-domain" has nothing to do with this, but what-ever :P
15:23:13  <TrueBrain> 80% chance his IPv6 is broken :P
15:23:25  <TrueBrain> but the version he tried didn't have Happy Eyeballs yet
15:32:20  <Timberwolf> You can already sorta do #9089 in newgrf, Tyler does it to change the types of buildings around stations. I guess it's not a big leap from there for the grf to go, "don't provide a building if not near a station"
15:32:52  <Timberwolf> Not sure what the limits are as I've not tried.
15:33:42  <Timberwolf> Very A-Train. (Which had that weird mechanism where stations would sprout a road, then stuff would grow along the road)
15:38:16  <peter1138> Yeah, NewGRF houses can probably do some... funky things.
15:38:18  <milek7> TrueBrain: connection is handled by browser, happy eyeballs on openttd side have nothing to do with this
15:40:29  *** gelignite has joined #openttd
15:43:53  *** Wormnest has joined #openttd
15:58:13  *** Progman has joined #openttd
15:59:43  <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on issue #9233: Online content not working cross-domain in Emscripten https://git.io/J3p5u
15:59:46  <DorpsGek> [OpenTTD/OpenTTD] embeddedt closed issue #9233: Online content not working cross-domain in Emscripten https://git.io/J3p5u
16:29:02  *** frosch123 has joined #openttd
16:34:16  *** didac has joined #openttd
16:34:42  <didac> hallo
16:50:02  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9233: Online content not working cross-domain in Emscripten https://git.io/J3p5u
16:54:55  *** sla_ro|master has joined #openttd
16:57:51  *** Wolf01 has joined #openttd
17:09:44  <DorpsGek> [OpenTTD/OpenTTD] TheDude-gh opened pull request #9234: UDP query of game script https://git.io/J3hW0
17:15:10  <peter1138> Oof, sqrt/atan/cos/sin.
17:15:35  <peter1138> UDP query eh?
17:16:17  <glx> while we are removing UDP
17:17:00  <TrueBrain> I love how people ignore templates
17:17:03  <TrueBrain> can we just close PRs with "no"
17:17:06  <TrueBrain> that are opened like that?
17:17:19  <peter1138> Why as much as "no"? :D
17:20:51  <peter1138> Also... Azure Pipelines?
17:21:12  *** Flygon has quit IRC
17:21:49  <peter1138> Such an old commit, and... merged instead of rebased.
17:32:11  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9234: UDP query of game script https://git.io/J3h0G
17:32:17  <TrueBrain> just random code without context
17:32:18  <TrueBrain> what-ever
17:32:41  <TrueBrain> I tried my mind-read device .. but it was out of order :(
17:38:21  <FLHerne> TrueBrain: If the bot closes them, people can't be offended
17:38:43  *** andythenorth has joined #openttd
17:38:51  <TrueBrain> personally I agree, but OpenTTD has shown to think differently about that :P
17:39:34  <FLHerne> Stalebot was annoying, but that was because the underlying reason for closing them was daft
17:45:54  <peter1138> Wasn't stalebot just andythenorth?
17:50:22  <FLHerne> lol
17:50:24  <DorpsGek> [OpenTTD/OpenTTD] TheDude-gh commented on pull request #9234: UDP query of game script https://git.io/J3hzA
17:50:41  <TrueBrain> how do you miss a template? Lol ..
17:52:00  <glx> the other option seems better
17:52:33  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9234: UDP query of game script https://git.io/J3hg8
17:52:39  <TrueBrain> lets try again ...
17:52:55  <TrueBrain> people seem to have a hard time understanding what a motivation is :) A description of the PR is not a motivation :P
17:53:38  <FLHerne> TrueBrain: "When servers has a game scripts, it and could be shown in ingame lobby window and could be added also to web server listing ." sounds reasonable
17:53:50  <FLHerne> for part of the same reason that it shows grfs now
17:54:00  <TrueBrain> but then a string alone is enough
17:54:03  <TrueBrain> so there is more to this
17:54:04  <FLHerne> Players like to know what kind of game they're joining
17:54:29  <andythenorth> was it me? :o
17:54:41  <andythenorth> oh dear
17:54:46  <FLHerne> TrueBrain: Isn't that what this does?
17:54:49  <TrueBrain> no
17:54:55  <FLHerne> PACKET_UDP_CLIENT_GET_GAMESCRIPT,///< Requests the name of a game script
17:55:01  <TrueBrain> well, kinda
17:55:03  <TrueBrain> the PR is weird
17:55:08  <TrueBrain> it is really hard to understand what it is trying to do
17:55:22  <TrueBrain> it also sends the shortname, for example
17:55:33  <FLHerne> Ok, it sends the version too, but that's useful in case the version matters for gameplay
17:55:37  <peter1138> Where does HandleIncomingNetworkGameInfoGRFConfig come into it?
17:55:41  <glx> name and version should be enough
17:55:48  <glx> short name has no use
17:56:09  <TrueBrain> we are filling in so many blanks :P
17:56:33  <glx> and it's easy to merge name and version into a single name.version string
17:56:34  <DorpsGek> [OpenTTD/OpenTTD] TheDude-gh commented on pull request #9234: UDP query of game script https://git.io/J3h2G
17:58:36  <andythenorth> should I apologise to someone somewhere?
17:59:32  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9234: UDP query of game script https://git.io/J3h29
17:59:49  <peter1138> I'm probably being dumb but I can't see where the request for this information is made.
18:00:20  <TrueBrain> peter1138: this PR is totally broken; that is not you
18:00:59  <TrueBrain> but it is also not a PR that is anywhere near ready .. he basically has an idea and wrote some code to showcase it :)
18:01:09  <peter1138> 2 years ago.
18:01:23  <peter1138> Well, nearly.
18:01:37  <TrueBrain> not sure why he is faking a GRF, but possibly to allow autodownloading
18:01:40  <TrueBrain> which is not needed for GSes :P
18:02:20  <_dp_> are ais even possible in mp officially?
18:02:24  <TrueBrain> sure
18:02:37  <peter1138> The GameInfoGRFConfig bit is not in the original commit, it's from the merge commit.
18:03:23  *** urdh has joined #openttd
18:04:14  <TrueBrain> lol, then I have no clue :P
18:05:03  <peter1138> Also, StationListv1 still gets autoselected as an AI. Hmm.
18:05:18  <TrueBrain> did you update? :)
18:05:23  <TrueBrain> I believe it was fixed recently, not?
18:05:42  <peter1138> 	function UseAsRandomAI()  { return false; }
18:06:24  <peter1138> Maybe I have an old version of it lying around :)
18:06:35  <TrueBrain> otherwise it is still broken :'(
18:07:31  <peter1138> Yup.
18:07:55  <peter1138> It's an old version, because the test AIs are no longer installed.
18:08:02  <TrueBrain> haha, lol
18:08:15  <TrueBrain> so one solution prevented the other solution to kick in
18:08:16  <TrueBrain> nice
18:08:23  <peter1138> yes :D
18:09:02  <peter1138> Hmm, and resolving that found another "issue"
18:09:16  <peter1138> If you remove a gamescript that is running, openttd crashes.
18:09:22  <TrueBrain> :o
18:10:13  <peter1138> During the cleanup routines. Hmm.
18:11:01  <TrueBrain> right, #9017 .. I need more testing ..
18:11:10  <TrueBrain> I am tempted to just ping Discord for people to join a server
18:11:15  <TrueBrain> but I am a bit afraid how that would end up :P
18:12:33  <TrueBrain> weird that we don't show author ingame of NewGRFs
18:17:29  <peter1138> I think you can't search content download by author either.
18:17:40  <TrueBrain> I think we don't even send it
18:17:42  <TrueBrain> very bad
18:17:47  <peter1138> So there are authors as tags instead.
18:17:54  <TrueBrain> anyway, PR9017, join my game!! Lets see if you guys can find a bug :)
18:21:24  <peter1138> Scaling viewport scrolling by framerate is... not going well.
18:21:37  <peter1138> It overshoots if there's a framerate spike :p
18:22:02  <peter1138> Let's find this PR
18:26:12  <TrueBrain> the chat improvement is very noticeable peter1138
18:26:19  <TrueBrain> in fact, I don't notice it, which is what I notice :D
18:38:53  <peter1138> Hmm
18:39:00  <peter1138> That is quite useful
18:39:14  <peter1138> No longer do you need to faff about with the console
18:39:20  <TrueBrain> indeed
18:41:29  <TrueBrain> aaahhhh ... I collided two trains and now lost a lot of money
18:41:29  <TrueBrain> nooooo
18:45:34  <andythenorth> are you playing the game?
18:45:36  <andythenorth> wtf?
18:45:40  <TrueBrain> I even pinged you!
18:45:48  <andythenorth> oh you did
18:45:51  * andythenorth was playing tanks
18:46:00  <andythenorth> ok
18:46:10  * andythenorth gets the binary
18:46:51  <TrueBrain> and get others on Discord to join too :P
18:47:55  <andythenorth> do I need an invite or something?
18:48:15  <TrueBrain> nah, just search internet, and you will see my lonely server being up there
18:48:55  <TrueBrain> weird, my server just got a new invite code assigned .. I need to trace that tomorrow :P
18:58:20  <FLHerne> Hm, I can't connect
18:58:30  <FLHerne> It's hung up on 1/6 Connecting
18:58:43  <TrueBrain> it even switch to TURN for you, hmm
18:58:54  <TrueBrain> I wonder if I opened the port for it, let me check
18:59:10  <TrueBrain> yeah, I did
18:59:21  <TrueBrain> also odd that STUN didn't work
18:59:57  <FLHerne> sorry, afk
18:59:59  <FLHerne> bad timing
19:00:05  <TrueBrain> no worries
19:04:37  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/J3hy3
19:04:38  <DorpsGek>   - Update: Translations from eints (by translators)
19:05:22  <TrueBrain> funny how for people with dualstack, if it picks IPv4 or IPv6 is a bit RNG
19:05:30  <TrueBrain> both are equally fine ofc
19:06:23  <supermop_Home> hi
19:07:01  <andythenorth> https://steamcommunity.com/app/1536610/discussions/0/3130541756147167200/
19:13:09  <peter1138> Unplayable eh?
19:14:51  <DorpsGek> [OpenTTD/OpenTTD] TheDude-gh commented on pull request #9234: UDP query of game script https://git.io/J3hSi
19:15:02  <glx> seems I'm connected via IPv4
19:15:41  <TrueBrain> 50/50 odds, basically :P
19:15:48  <TrueBrain> I can delay IPv4 over IPv6 a bit I guess
19:15:52  <TrueBrain> to make it prefer IPv6
19:15:54  <TrueBrain> might be wise
19:28:06  <TrueBrain> I don't like playing with frosch123 ... he is way too good at this ..
19:29:57  <frosch123> pff, you mean i play serious?
19:30:16  <TrueBrain> no clue if this is your serious, but you are doing a lot better than any of us :P
19:30:59  <frosch123> i thought everyone else was fighting on github :p
19:31:31  <frosch123> anyway, i seriously dislike the new depot-clone behavior
19:31:36  <frosch123> who changed that :p
19:31:58  <TrueBrain> that it doesn't open the new vehicle window?
19:32:10  <peter1138> It really is 3 years old :p
19:32:10  <frosch123> yes
19:32:17  <TrueBrain> I really like that it doesn't :)
19:32:21  <TrueBrain> when I clone, I don't care :P
19:32:43  <TrueBrain> I am more annoyed that "delete" doesn't remove the order I selected :D
19:32:47  <frosch123> but when i open a new route, i non-share clone a train, and then just want to change some orders
19:32:49  <_dp_> frosch123, you can use citymania client, it's configurable there ;)
19:37:48  <TrueBrain> okay, so far 1 IPv6 STUN that fails, and FLHerne cannot join .. but we need more people to test :D
19:45:21  *** smoke_fumus has quit IRC
20:06:51  *** gelignite has quit IRC
20:09:37  *** lobstarooo has joined #openttd
20:16:01  *** lobster has quit IRC
20:16:08  *** lobstarooo is now known as lobster
20:41:41  <FLHerne> TrueBrain: back now
20:42:00  <TrueBrain> FLHerne: seems you are not the only one with this issue .. and seems to be the old issue you had
20:42:05  <TrueBrain> a bit RNG if you can connect or not
20:42:11  <TrueBrain> will look into it tomorrow what is going wrong exactly :D
20:42:18  <TrueBrain> I think I have some timings off
20:42:36  <FLHerne> TrueBrain: It seems I can connect now
20:42:44  <TrueBrain> yeah, it is a bit playing the odds
20:43:03  <frosch123> it took longer than other players though
20:43:14  <TrueBrain> that is just his shitty connection :P
20:46:25  *** arikover has joined #openttd
20:49:13  *** didac has quit IRC
20:54:08  *** Samu has quit IRC
20:58:29  *** iSoSyS has joined #openttd
21:03:05  *** Elouin has quit IRC
21:03:14  *** Elouin has joined #openttd
21:04:48  <TrueBrain> so we found a nice bug .. there is a flow where a client can be disconnected, through a crash, where the server does see him go
21:04:52  <TrueBrain> but none of the clients is informed
21:04:57  <TrueBrain> I will have fun tomorrow tracing bugs :P
21:05:13  <LordAro> dunno about the original crash though :(
21:05:19  <frosch123> ah, i thought it was double for the server as well
21:05:20  <TrueBrain> not your first
21:05:25  <TrueBrain> frosch123: nope
21:05:26  <LordAro> naturally, reran in gdb, did not reproduce
21:05:32  <TrueBrain> LordAro: same as last time :P
21:05:34  <LordAro> maybe i should always run in gdb
21:05:46  *** urdh has quit IRC
21:05:52  <frosch123> LordAro: get an os where you can enable core dumps by default
21:06:11  *** nielsm has quit IRC
21:06:21  <LordAro> well it's a self-compiled debug build of TB's branch anyway
21:06:27  <LordAro> on mingw
21:07:20  *** Progman has quit IRC
21:09:38  *** WormnestAndroid has quit IRC
21:11:37  <TrueBrain> but you keep having those random crashes
21:11:40  <TrueBrain> might be mingw ofc
21:11:45  <TrueBrain> but would be interesting to find out what it is :D
21:11:49  <LordAro> indeed
21:12:02  *** WormnestAndroid has joined #openttd
21:26:41  <glx> oh I was waiting to build my second 5.5M tunnel
21:26:51  <LordAro> netsave!
21:27:09  <TrueBrain> :D
21:27:19  <frosch123> so the tunnel was no accident? :p
21:27:42  <glx> no it was part of a 2 tracks line
21:28:00  <glx> the cost was a surprise though
21:28:13  <glx> I didn't check before building it
21:28:52  <frosch123> ah, found it
21:29:02  <glx> a very busy line of 2 trains ;)
21:30:46  <frosch123> you all built road vehicles :)
21:31:10  <glx> well for gold mines a train was not efficient
21:31:50  <TrueBrain> grrr, cannot reproduce the TURN bug .. so annoying if you locally cannot reproduce bugs :P
21:32:25  *** andythenorth has quit IRC
21:33:15  <DorpsGek> [OpenTTD/team] pelya opened issue #212: Please transfer openttd-android-translate repo to the new translator https://git.io/J3jLO
21:44:48  *** sla_ro|master has quit IRC
21:47:37  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #9235: Fix warnings from LGTM https://git.io/J3jqC
21:48:57  *** arikover has quit IRC
21:49:46  *** Wolf01 has quit IRC
21:53:02  *** frosch123 has quit IRC
21:59:16  <glx> MSVC doesn't like
22:36:37  <DorpsGek> [OpenTTD/OpenTTD] TheDude-gh updated pull request #9234: UDP query of game script https://git.io/J3hW0
23:05:15  *** HerzogDeXtEr has quit IRC

Powered by YARRSTE version: svn-trunk