IRC logs for #openttd on OFTC at 2021-05-09
⏴ go to previous day
00:15:46 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
00:16:14 *** WormnestAndroid has joined #openttd
00:28:14 <Speeder_> newest openttd can run ridiculously fast?
00:28:41 <Speeder_> also since june 2020 there was any major changes to... the thing I forgot name that you use to specify newgrfs?
00:30:06 <FLHerne> Speeder_: (1) yes (2) you might need to clarify a bit more
00:31:08 <Speeder_> the language to make industries and wtuff
00:31:17 <Speeder_> last major update added the multiple inputs and outputs thingy
00:33:15 <FLHerne> There are a few new vehicle variables about railtypes
00:34:10 <FLHerne> and IND_FLAG_DO_NOT_CLAMP_PASSENGER_PRODUCTION which overrides a strange default behaviour of water industries
00:34:49 <FLHerne> Also procedure calls, but you don't need new OpenTTD for that, it's purely an NML addition
00:35:11 <FLHerne> I think none of the above are in a released version of NML yet, only master branch
00:36:41 <FLHerne> so you can decide how ridiculously fast fast-forward should be
00:37:02 <glx> highly depends on hardware power :)
00:45:12 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
01:02:29 <Speeder_> wasn't possible to build windows version without visual studio?
01:04:37 <Speeder_> build system changed?
01:04:50 <Speeder_> I am almost sure I built my custom OpenTTD last time using MingW
01:04:56 <Speeder_> now I can't figure out how to even fire "make"
01:28:13 <glx> from source root you need to 'mkdir build' then 'cd build' then 'cmake .. -G"MSYS Makefiles" -DCMAKE_BUILD_TYPE=RelWithDebInfo' then 'make'
01:38:22 <Speeder_> that should be in the docs...
01:38:34 <Speeder_> also stupdi cmake >.< I never liked it, and it is being a pain to make it work
01:38:41 <Speeder_> seemly you need to install a specific version of it
01:39:13 <Speeder_> (the default version that comes with mingw doesn't have the option -G"MSYS Makefiles")
01:58:36 *** tokai|noir has joined #openttd
01:58:36 *** ChanServ sets mode: +v tokai|noir
02:02:25 *** Wormnest has quit IRC (Quit: Leaving)
02:05:19 *** tokai has quit IRC (Ping timeout: 480 seconds)
02:05:58 <Speeder_> so my release versio nof openttd had 5mb
02:06:02 <Speeder_> my compiled version has 230mb
02:06:28 <Speeder_> ok, and it doesn't launch
02:51:25 *** Gustavo6046_ has joined #openttd
02:55:18 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
02:55:18 *** Gustavo6046_ is now known as Gustavo6046
02:56:06 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
05:28:39 <LordAro> Speeder_: debug versions are big, this is known
05:28:56 <LordAro> can't help you with the other issues though, works on mingw fine here
06:04:23 <dwfreed> the mozilla bug notes that the issue only occurs pre Win8
06:58:34 *** Progman has joined #openttd
07:25:18 *** jottyfan has joined #openttd
07:25:27 *** sla_ro|master has joined #openttd
08:37:56 *** andythenorth has joined #openttd
09:04:43 <TrueBrain> be happy you can already get one :P
09:28:57 <peter1138> In RelocateAllWindows() it uses Window::IterateFromBack()
09:30:35 <peter1138> It seems this causes the toolbar/statusbar to be repositioned last, after all the other windows. Therefore the caption-visibility test doesn't take account of the current toolbar size/position. Hmm.
09:31:40 *** gelignite has joined #openttd
09:32:21 <peter1138> Yeah, changing it to IterateFromFront() makes the caption test work as expect.
09:32:52 <peter1138> As the toolbars are always on top, seems like it should be Front to Back in this case.
09:34:26 <TrueBrain> owh, look, a change in english.txt
09:34:34 * TrueBrain takes a walk outsides
09:35:33 *** Progman has quit IRC (Remote host closed the connection)
09:44:04 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9223: Fix: lobby window doesn't close if no connection could be established https://git.io/J3FES
09:46:15 <andythenorth> ok the shakes and sweats have stopped
09:46:25 <andythenorth> time to code newgrf?
09:50:20 *** HerzogDeXtEr has joined #openttd
09:54:51 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: WIP: allow multiplayer without port forwarding https://git.io/JOk7C
09:55:04 <TrueBrain> okay .. fixed many many things this morning, all kind of small stuff
09:55:06 <TrueBrain> showing more errors
09:55:11 <TrueBrain> handling connection failures
09:55:26 <TrueBrain> all that is left, as far as I know, is to find a clean way to help clients understand which NewGRFs they are missing
09:58:46 <TrueBrain> but I think I first focus a bit on the game coordinator code .. as I am not sure yet how to approach NewGRFs :)
09:59:24 <peter1138> Feature: Improved multiplayer compatibility (no newgrfs are allowed)
10:00:21 <TrueBrain> the main issue is the name of a NewGRF
10:00:32 <TrueBrain> ID + MD5 is exchanged
10:00:44 <TrueBrain> but the name used to be a separate exchange
10:00:59 <TrueBrain> client told server which he was missing, server told the names
10:01:22 <peter1138> IIRC that was only because of the UDP packet size limit.
10:02:13 <TrueBrain> so I was planning on adding the name for every NewGRF in the info packet
10:02:15 <peter1138> You could fetch the names from bananas, but if they're on there... that doesn't help.
10:02:17 <TrueBrain> but .. that is a lot of data :P
10:02:43 <TrueBrain> well, the idea that I did try, was to only send NewGRF names of ones not on BaNaNaS
10:03:29 <peter1138> Could you just keep the separate exchange?
10:03:39 <TrueBrain> I guess part of my problem is that the server-listing is already huge .. an average server is 1KB of data .. there are 500 of them online
10:03:46 <TrueBrain> so that is 500KB of data :P
10:04:21 <TrueBrain> hmm ... orginally, a client got all the info from the server, so the server knew the NewGRF name too
10:04:30 <TrueBrain> now the client gets the info from the game coordinator, which doesn't know this
10:04:43 <TrueBrain> buttttt .. we can ofc do that if you refresh a server, he does send the names to you
10:05:46 <_dp_> or just compress the server list :p
10:05:46 <TrueBrain> it just means that if you press the NewGRF button, we first need to do a quick refresh
10:06:02 <peter1138> I don't think that's unreasonable.
10:06:27 <TrueBrain> _dp_: that is not a bad idea, except that it should be on the transport layer
10:06:31 <TrueBrain> not on the per-packet layer
10:08:24 <TrueBrain> hmm, I like the idea of: Game Coordinator sends NewGRFs without name .. but the server itself will send it with name, to clients
10:08:31 <TrueBrain> we can show the red/yellow dot just fine
10:08:52 <TrueBrain> and as soon as you see the names in the GUI, the info is requested
10:09:04 <TrueBrain> (as the game doesn't give a shit about names .. it is purely a human thing :P)
10:12:07 <_dp_> btw, can you simplify newgrf downloading while at it?
10:12:20 <_dp_> ideally it should be just one extra "OK" click if everything is on bananas
10:13:30 <TrueBrain> have you seen how massive this PR already is?
10:13:47 <_dp_> you can do it in a separate PR :p
10:13:48 <TrueBrain> but we are open for contributions ;)
10:15:46 *** WormnestAndroid has quit IRC (Remote host closed the connection)
10:16:00 *** WormnestAndroid has joined #openttd
10:23:23 <peter1138> Should I make FR_CHECKER scaled?
10:24:00 <TrueBrain> not sure that makes it look any better :P
10:27:23 <TrueBrain> tnx peter1138 , this solution of NewGRF names seems very trivial to implement :)
10:30:48 *** Progman has joined #openttd
10:41:37 <TrueBrain> took a while to test, as our searchpaths are just a tiny bit too strong .. it kept finding the newgrf for the client too :D
10:42:02 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: WIP: allow multiplayer without port forwarding https://git.io/JOk7C
10:42:09 <TrueBrain> okay, did not expect to solve this today, but the solution is simple and elegant
10:42:57 <TrueBrain> that means my TODO list for the PR is empty :D
10:43:11 <TrueBrain> most likely I forgot 100 things and there are 200 bugs hiding .. but .. at a certain point you go blind on your own code :P
10:50:53 * andythenorth starts there I think :P
10:55:51 *** frosch123 has joined #openttd
10:56:16 <frosch123> TrueBrain: probably meh, but in theory newgrf names are translateable
11:11:15 <TrueBrain> frosch123: yeah, I remember that being a current issue to in some cases
11:11:32 <TrueBrain> so for sure, not my problem for this PR :D :D
11:34:50 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:36:24 *** WormnestAndroid has joined #openttd
11:38:11 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:39:22 *** WormnestAndroid has joined #openttd
11:46:41 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:48:41 *** WormnestAndroid has joined #openttd
12:06:06 <andythenorth> oof brain won't code :)
12:12:54 <DorpsGek> [OpenTTD/OpenTTD] kiwitreekor updated pull request #9161: Feature: NewGRF Bridges without overriding https://git.io/J3cPO
12:15:11 <DorpsGek> [OpenTTD/OpenTTD] kiwitreekor updated pull request #9161: Feature: NewGRF Bridges without overriding https://git.io/J3cPO
12:32:35 <TrueBrain> "so why isn't this working?" .. "owh, I tell the server to connect to the server, instead of the client ..."
12:32:38 <TrueBrain> took me long enough
12:39:03 *** Progman has quit IRC (Remote host closed the connection)
12:44:09 <peter1138> Relays of connected servers, like IRC...
12:56:28 <glx> seems the cmake move broke it
13:26:28 <peter1138> Is increasing station catchment to play "sparse" games cheating? :D
13:27:48 <glx> like rising a small island near oilrig with road stop to fill trains on the coast ?
13:28:21 <peter1138> No I mean doubling the CA_* values to make catchment area increase.
13:31:02 <TrueBrain> hmm .. my PR crashes when I restart the coordinator from time to time .. something in the GUI .. some string that renders wrong .. will be fun to figure out :P
13:32:32 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: WIP: allow multiplayer without port forwarding https://git.io/JOk7C
13:32:47 <peter1138> IMO this is what larger maps are for :)
13:33:07 <peter1138> Lower density towns and longer distances.
13:36:00 <FLHerne> peter1138: It is, that's what buses are for
13:38:38 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: WIP: allow multiplayer without port forwarding https://git.io/J3bsN
13:38:44 <TrueBrain> I think that PR is ready for review :)
13:39:05 <TrueBrain> rewrote the whole Game Coordinator; not production-ready yet, but it does everything what it should
13:39:15 <TrueBrain> it just isn't presistent
13:39:18 <TrueBrain> so .. yeah .. what-ever for now
13:39:42 <TrueBrain> GC detects what kind of connection you have as server
13:39:49 <TrueBrain> and clients are matched as best as we can
13:40:39 <TrueBrain> building binaries, but I think Ubuntu18.04 will prevent that from working ..
13:41:50 *** jottyfan has quit IRC (Remote host closed the connection)
13:42:58 <TrueBrain> ah, no, those are skipped for those kind of releases
13:47:25 *** jottyfan has joined #openttd
13:48:40 <TrueBrain> Python logic is ~400 lines of code
13:48:55 <TrueBrain> total 1000 lines of code, including protocol handling etc
13:49:30 <TrueBrain> owh, 250 lines of code, the logic itself, ignoring whitespaces
13:49:52 <TrueBrain> one giant state-machine :D
13:50:30 <TrueBrain> abcdefghjkmnpqrstuvwxyzABCDEFGHJKMNQRSTUVWXYZ23456789 <- alphabet for invite-codes
13:50:42 <TrueBrain> removes all the annoying "is this an I l i l or 1" shit :P
13:52:30 <glx> and of course my win7 64bit VM is broken, need to redo it
13:52:54 <FLHerne> TrueBrain: You could have L, maybe
13:53:24 <TrueBrain> I didn't take the risk :)
13:55:08 <TrueBrain> hmm, my STUN detection isn't complete yet, lol, oops .. I am cheating :D
13:57:50 *** jottyfan has quit IRC (Quit: jottyfan)
14:00:32 <TrueBrain> hmm .. for the TURN servers I can simply sign the tokens on the Game Coordinator .. means those two don't have to talk to eachother
14:00:39 <TrueBrain> TURN server can simply validate it is a valid token
14:00:45 <TrueBrain> include a timestamp in it
14:02:01 <TrueBrain> LordAro: it is possible I do need Python3.8 on the bananas-boxes .. I think they are a great place to run TURN servers on for now .. cheap, and allows us to see how much they are used .. not something for today or next week, but if you can think about how we would manage to do that, that would be great :)
14:05:17 <DorpsGek> [OpenTTD/website] TrueBrain merged pull request #212: Fix: [Actions] queue publish jobs so they are executed in order https://git.io/J3ghG
14:07:31 *** jottyfan has joined #openttd
14:10:49 <TrueBrain> okay, new binaries published
14:10:59 <TrueBrain> so if anyone wants to test .. would appreciate any and all feedback :)
14:11:10 <TrueBrain> maybe we should make it a thing soon, invite who-ever to join, see how it goes
14:12:25 <TrueBrain> but first I need a few other people to test to make sure I didn't make big boo-boos :P
14:18:46 <TrueBrain> invalid reads on APID chunk
14:18:46 <TrueBrain> ==31684== by 0x8B00F4: Save_NewGRFMapping(OverrideManagerBase const&) (newgrf_sl.cpp:33)
14:19:16 <TrueBrain> als on the IIDS chunk
14:19:24 <TrueBrain> ==31684== at 0x8B06F8: OverrideManagerBase::IsValidID(unsigned short) const (newgrf_commons.h:221) is the first mention of invalid read
14:20:03 <TrueBrain> ah, yes, recent .. peter1138 , you might have introduced invalid reads :D
14:21:16 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9202: Codechange: Don't save unused NewGRF override mappings. https://git.io/J3b8x
14:23:31 <glx> yeah wrong iso used, I installed 32bit, let's retry
14:24:12 <TrueBrain> and thread sanitizer doesn't appreciate my PR .. well, something to figure out another time :)
14:27:10 <TrueBrain> owh, it is not my PR, it is the same issue, just .. more verbose :P
14:28:03 *** lobster has quit IRC (Read error: Connection reset by peer)
14:28:20 *** lobster has joined #openttd
14:31:05 *** lobstarooo has joined #openttd
14:34:32 <glx> oh wrong end value for the loop I think
14:34:59 <glx> it should be GetMaxOffset() (based on assert in OverrideManagerBase::Add()
14:36:21 *** lobster has quit IRC (Ping timeout: 480 seconds)
14:37:01 <glx> 1.11.2, crash inside ATI driver
14:37:07 *** lobstarooo is now known as lobster
14:45:13 <glx> and I don't have the right dlls
14:53:06 <TrueBrain> you got to love OpenGL :D
14:59:43 <glx> ok found the driver, downloading to extract the dlls
14:59:55 <glx> (else I can't get a callstack)
15:00:36 <glx> AMD releasing broken drivers is so usual
15:01:59 <glx> but crash report flow seems way lower than before 1.11.2 ;)
15:03:17 <TrueBrain> for sure it solved many issues :) Just a few pesky ones left :P
15:12:48 <glx> 00007FFE96B9C6BD mov rax,qword ptr [7FFE98EDFEF8h]
15:12:49 <glx> 00007FFE96B9C6C4 call qword ptr [rax+9D8h]
15:12:49 <glx> seems they are playing with function pointers and forget basic safety checks ;)
15:17:21 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
15:20:52 <glx> hmm would it be possible to include openttd.pdb in steam install ?
15:21:38 <glx> could help filling decoded stack trace in crash.log I think
15:22:01 <dwfreed> how big is openttd.pdb ?
15:24:53 <glx> not that big compared to install size of most games ;)
15:30:16 <michi_cc> Hmm, not sure we can do anything about crashing inside opengl32.dll!LoadAvailableDrivers.
15:30:57 <glx> yeah I can see many reports of this driver not working very well
15:31:04 <dwfreed> maybe offer another launch option for windows to pass -v win32
15:31:24 <glx> it's the march version, but april version crashes too
15:34:10 <peter1138> 15:34 <+glx> oh wrong end value for the loop I think
15:34:20 <peter1138> Nah, I think it's testing completely the wrong thing.
15:35:20 <glx> <+glx> it should be GetMaxOffset() (based on assert in OverrideManagerBase::Add()
15:35:37 <glx> and constructor seems to confirm it too
15:36:33 <peter1138> Yeah, but it's meant to save the mapping_ID array, not the entity_overrides array. mapping_ID is GetMaxMapping() long.
15:37:55 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3bXW
15:38:32 <peter1138> glx, yup. That was part of the PR.
15:39:38 <peter1138> glx, I guess really it just needs to check for grfid != 0
15:40:16 <peter1138> Although that assumes there's no default overrides. Hmm.
15:40:57 <peter1138> Otoh, if they are all sequentially allocated, there should be a count somewhere.
15:42:32 <DorpsGek> [OpenTTD/OpenTTD] glx22 opened pull request #9225: Fix: [MinGW] Reapply 48fd7b27 to fix launch on Windows 7 x64 https://git.io/J3b1B
15:45:21 <glx> seems valid slot is found by brute force
15:46:51 <glx> in OverrideManagerBase::AddEntityID()
15:48:14 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #9226: Fix #9202: Invalid test for unset NewGRF override mapping. https://git.io/J3bMy
15:48:57 <peter1138> Forgot to make it a draft
15:49:44 <peter1138> Thanks, I couldn't see it :(
15:50:08 <peter1138> Anyway still testing it, but that's what I have in mind.
15:50:10 <glx> it's hidden near reviewers' list
15:50:35 <peter1138> Oh yes, not a button and doesn't look like a link.
16:11:12 <Rubidium> glx: have you been able to test #9225? Primarily because the fix you reapply mentions that it must be linked before gdi32, however I cannot find out whether it will actually link it before gdi32. I presume that it still fails if gdi32 gets linked before usp10
16:15:14 *** jottyfan1 has joined #openttd
16:16:10 <glx> I tested the exe in a VM, and checked verbose compilation
16:17:15 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9225: Fix: [MinGW] Reapply 48fd7b27 to fix launch on Windows 7 x64 https://git.io/J3b7m
16:17:33 <glx> the line is /H/msys64/mingw64/bin/g++.exe -g -mwindows -Wl,--whole-archive CMakeFiles/openttd.dir/objects.a -Wl,--no-whole-archive -o openttd.exe -Wl,--major-image-version,0,--minor-image-version,0 -static -static-libgcc -static-libstdc++ /H/msys64/mingw64/lib/libpng.a /H/msys64/mingw64/lib/libz.a /H/msys64/mingw64/lib/liblzma.a /H/msys64/mingw64/lib/liblzo2.a /H/msys64/mingw64/lib/liblzo2.a -lopengl32 -lws2_32 -lwinmm -limm32 -lusp10 -lkernel32 -luser32
16:17:33 <glx> -lgdi32 -lwinspool -lshell32 -lole32 -loleaut32 -luuid -lcomdlg32 -ladvapi32
16:17:56 <glx> and without the fix there's no -lusp10 at all
16:18:43 <glx> so cmake adds default libs after our own selection
16:21:16 *** jottyfan has quit IRC (Ping timeout: 480 seconds)
16:21:16 *** jottyfan1 is now known as jottyfan
16:26:39 *** Gustavo6046 has joined #openttd
16:26:48 *** Wormnest has joined #openttd
16:28:51 <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #9225: Fix: [MinGW] Reapply 48fd7b27 to fix launch on Windows 7 x64 https://git.io/J3b1B
16:30:16 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9223: Fix: lobby window doesn't close if no connection could be established https://git.io/J3bFV
16:48:25 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9223: Fix: lobby window doesn't close if no connection could be established https://git.io/J3FES
16:51:42 *** ChanServ sets mode: +v tokai
16:53:38 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3bpX
16:55:21 <glx> oh no conflict in 9017 ;)
16:56:19 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3bhs
16:58:28 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
17:02:37 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3bju
17:03:32 <TrueBrain> got the first report that with many NewGRFs it crashes .. which reminds me .. I should add a patch: "Feature: remove NewGRF limit" :P
17:03:46 <TrueBrain> hmm, no, there is still a limit .. lets calculate
17:05:01 <TrueBrain> @calc (32767 - 142) / 100
17:05:10 <TrueBrain> so .. 255? Sounds like a nice number
17:06:05 <peter1138> Next you'll be off-loading bananas traffic by distributing NewGRFs from the servers ;)
17:06:15 <TrueBrain> I did consider it :P
17:09:48 <andythenorth> that would be really useful for the very niche case of playing multiplayer with grfs that are in development (not on bananas)
17:09:55 <andythenorth> very niche though
17:10:27 <TrueBrain> the issue is: are we allowed to redistribute via servers? :D
17:10:30 <glx> but if not on bananas it won't be possible to distribute from servers
17:10:32 <FLHerne> Won't that cause mb to yell at you?
17:10:43 <peter1138> TrueBrain, it's not us distributing ;)
17:11:49 <peter1138> Hmm, I'm going to assume that chunky bevels and non-integer scaling will not really work.
17:11:57 <glx> I think distributing NewGRF available on bananas from servers is similar as distributing them from bananas
17:11:57 <peter1138> But I should try it anyway :D
17:15:14 <DorpsGek> [OpenTTD/OpenTTD] frosch123 opened pull request #9227: Make Window destruction not rely on undefined behavior https://git.io/J3NvF
17:15:50 <DorpsGek> [OpenTTD/OpenTTD] frosch123 updated pull request #9227: Make Window destruction not rely on undefined behavior https://git.io/J3NvF
17:23:25 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #9227: Make Window destruction not rely on undefined behavior https://git.io/J3NU5
17:23:57 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: multiplayer without port forwarding https://git.io/JOk7C
17:24:01 <TrueBrain> New NewGRF limit ^^ :)
17:24:40 <andythenorth> not us distributing
17:26:11 <TrueBrain> I had the first person asking: why 255? :P
17:26:18 <TrueBrain> but yeah .. with 113 it works :D
17:26:34 <frosch123> one for each client
17:27:10 <TrueBrain> even the NewGRF-check-packet no longer fits in 1460 bytes :P
17:27:41 <frosch123> anyway, does the fios limit still exist?
17:28:46 <TrueBrain> when are we going to rename "master" to "main"? :D
17:29:39 <Rubidium> TrueBrain: after 9017 has been merged I'd guess
17:29:51 <frosch123> when people are done replacing trunk->master in the wiki
17:30:31 <frosch123> TrueBrain: anyway, when you do, don't forget eints
17:34:15 <glx> pff github shows commits in the wrong order
17:34:51 <frosch123> sorry, i should have force-redated them
17:37:33 <FLHerne> Rubidium: Is "ScriptWaypointList_Vehicle" literally just waypoints?
17:37:53 <FLHerne> Or all ordered destinations?
17:38:15 <FLHerne> The description is a bit unintelligible, and then I wasn't sure what a list of only waypoints could be used for :p
17:38:54 <peter1138> Is there an equivalent for Stations?
17:39:13 <peter1138> A Waypoint function would imply it is for Waypoint, I suppose.
17:39:23 <frosch123> FLHerne: there are certainly usecases for lists of stations without waypoints. so the waypoint list probably only exists for completeness
17:40:20 <frosch123> and yes, ScriptStationList_Vehicle exists
17:42:34 <FLHerne> Is it ordered by the order sequence? What if one is visited multiple times?
17:44:48 <frosch123> FLHerne: scriptlist is a weird combination of std::vector, std::map<index, value>, std::set<index> and std::set<value>
17:45:34 <frosch123> probably also std::multimap<value, index>
17:46:10 <frosch123> someone did not want to explain containers to script authors, so they made a container to rule them all
17:46:43 <TrueBrain> the other way is also a possibility: that the author knew nothing about containers and just bashed the keyboard long enough that it compiled :P
17:46:48 <frosch123> we call it "eierlegende Wollmilchsau" :)
17:47:37 <TrueBrain> ugh, MacOS builds are slow ...
17:47:45 <TrueBrain> 20 minutes .. where the rest is long done ..
17:48:20 <FLHerne> Reminds me of PHP's "array"
17:49:48 <TrueBrain> PHP empty() function remains the best one out there
17:49:51 <Rubidium> though it doesn't add duplicates and the sorting depends on the sorter the user sets
17:50:11 <Rubidium> empty("false") -> true too?!?
17:50:36 <TrueBrain> no, that is false :P
17:53:08 <TrueBrain> (in reality, empty() does a bool-cast and checks if the result is false, which is a really weird thing to do for an empty() function)
17:54:58 <Rubidium> TrueBrain: which makes me wonder why "false" is not considered 'false' ;)
17:55:20 <TrueBrain> in security-coding trainings I gave, these kind of things were important to highlight
17:55:31 <TrueBrain> most people I trained were unaware, and assumed both or neither
17:55:37 <TrueBrain> PHP is full of those .... "fun" facts
17:55:54 <andythenorth> I did a PHP once
17:56:04 <andythenorth> and I tried to did a Perl once
17:56:34 <Rubidium> I have only use PHP for some simple local things that definitely never leave the local network, as in... not even connected to the internet at that moment
17:57:06 <frosch123> i used php only once, and that's because tb wanted me to use php
17:59:55 <TrueBrain> I think it was PHP because Rubidium wrote some PHP to start with :P
18:00:03 <TrueBrain> I don't want to underdo his claim .. but ... :D
18:00:19 <TrueBrain> just happy we no longer run PHP on any of our servers :)
18:01:56 <Rubidium> what ran as PHP for OpenTTD?
18:02:10 <TrueBrain> and index.php redirects
18:02:43 <Rubidium> can't remember making any of those
18:03:30 <TrueBrain> owh, and the mirror stuff
18:03:31 <TrueBrain> also had some PHP :P
18:03:37 <TrueBrain> someone rewrote it in bash
18:03:45 <TrueBrain> but the sync script was still PHP
18:03:49 <TrueBrain> which was .. horrible to deal with :D
18:03:57 <peter1138> You know PHP is bad when you end up rewriting in bash...
18:04:09 <frosch123> rewriting php code in bash? :) sounds interesting
18:04:34 <Rubidium> I know I have been fighting with Django and wrote some javascript
18:04:56 <TrueBrain> and you did C++ servers
18:05:04 <TrueBrain> which is also rather unusual this day and age :)
18:05:46 <Rubidium> it reused most of the OpenTTD code, so it got that for it and it was quite light weight
18:06:06 <TrueBrain> it wasn't bad; just the dependency on the OpenTTD code was horrible to maintain :P
18:06:10 <TrueBrain> as with any dependecy
18:08:26 <TrueBrain> (which basically meant we never updated it in 10 years btw :P)
18:08:36 <Rubidium> I must say download.js is also quite a-typical javascript for a website. Usually it's minified and garbled, and therefor not quite readable
18:08:54 <Rubidium> but needed it to be updated?
18:09:20 <TrueBrain> well, yeah, but it was part of a bigger problem
18:09:24 <TrueBrain> it was running on Debian Etch
18:10:03 <TrueBrain> it was fun following the flow when converting to Python :) It had some (positive) surprises up its sleeve
18:10:10 <frosch123> i once ported the bananas server and parts of the master server to postgres
18:10:15 <frosch123> but gave up on the sql functions
18:10:32 <Rubidium> yeah, the different dialects are a nuisance
18:11:52 <frosch123> i knew companies who had 100kloc in oracle sql functions, and then wanted to get away from oracle :p
18:12:07 <TrueBrain> well, staying at Oracle is too
18:12:51 <peter1138> We have some Oracle Application Express stuff. It's... horrible.
18:13:16 <frosch123> TrueBrain: the trigger were not the costs, but the appearance of export restrictions
18:16:13 <Rubidium> FLHerne: I hope it's clearer now
18:21:17 <FLHerne> Rubidium: It is, thanks :-)
18:24:18 *** jottyfan has quit IRC (Quit: jottyfan)
18:25:30 *** Progman has joined #openttd
18:44:40 *** tokai|noir has joined #openttd
18:44:40 *** ChanServ sets mode: +v tokai|noir
18:50:13 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
18:51:29 *** tokai has quit IRC (Ping timeout: 480 seconds)
18:57:14 <DorpsGek> - Update: Translations from eints (by translators)
19:05:09 *** bkilm[m] has quit IRC (Quit: Client limit exceeded: 8192)
19:13:01 *** Gustavo6046 has joined #openttd
20:04:34 *** gelignite has quit IRC (Quit: Stay safe!)
20:07:20 <TrueBrain> right, IPv4 and IPv6 still work fine, w00p :)
20:07:35 <TrueBrain> its funny .. most of the testing I do via WSL2, which in my setup has shitty video performance
20:07:40 <TrueBrain> then you start the Windows build
20:07:52 <TrueBrain> 144fps is even better :D
20:11:48 <glx> oh CI doesn't report warnings for lang files
20:12:50 <Rubidium> glx: that might be a good thing, otherwise you can't remove strings anymore
20:26:36 *** ChanServ sets mode: +v tokai
20:33:28 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
20:36:52 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
20:39:59 *** tokai has quit IRC (Ping timeout: 480 seconds)
21:06:09 <glx> hmm does GC test direct connection yet ?
21:09:44 <glx> because it says "behind NAT", while master can advertise fine
21:11:13 <glx> I am behind NAT, but it's open for openttd
21:11:15 <TrueBrain> It only checks IPv6 if you use that to connect to GC
21:12:03 <TrueBrain> Guess I could check IPv4 in those cases too
21:20:57 <glx> LAN search is funny, first click shows ip:port, second click adds invite code, then both disapear
21:22:19 <TrueBrain> Please write those findings in the PR :D
21:22:52 <TrueBrain> Lovely edge cases to fix :)
21:29:46 <glx> oh and I triggered an assert
21:31:05 <andythenorth> maybe it's time I accepted that nml will optimise switch results that are a constant
21:31:16 <andythenorth> I've spent nearly half a day trying to write code to prevent it
21:31:51 <glx> why would you want to prevent optimisations ?
21:32:28 <andythenorth> I'm trying to prevent them being needed
21:32:43 <andythenorth> but it makes for very fragmented python templating
21:32:51 <andythenorth> have to do a lot of if len(foo) == 1
21:33:06 <glx> it's almost impossible to do with generated nml
21:33:14 <andythenorth> I hate the yellow warnings basically, it feels like I did something wrong
21:33:32 <andythenorth> ammler strongly taught me that compile should never print any warnings
21:33:53 <TrueBrain> If you can reproduce that assert, that would be great :D
21:34:13 <glx> I think it's because last used server is ":3979"
21:34:31 <TrueBrain> Ah, yes .. empty hosts are a problem
21:34:39 <TrueBrain> Need to filter those
21:35:41 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:36:13 * andythenorth makes peace with the madness
21:36:29 <andythenorth> wasting life on irrational beliefs :D
21:37:11 <glx> let's fix openttd.cfg and retry :)
21:39:08 *** sla_ro|master has quit IRC ()
21:45:32 *** Progman has quit IRC (Remote host closed the connection)
21:50:21 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:51:42 *** ChanServ sets mode: +v tokai
21:52:09 *** andythenorth has quit IRC (Quit: andythenorth)
22:07:32 *** tokai|noir has joined #openttd
22:07:32 *** ChanServ sets mode: +v tokai|noir
22:14:26 *** tokai has quit IRC (Ping timeout: 480 seconds)
22:17:51 <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on pull request #9017: Feature: multiplayer without port forwarding https://git.io/J3N7G
22:37:29 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:47:13 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
22:50:56 *** Gustavo6046 has joined #openttd
22:59:25 *** ChanServ sets mode: +v tokai
23:06:16 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
