00:09 < bridge> github searching is so fucking broken 00:09 < bridge> i am so mad 00:09 < bridge> like i am searching in a commit, getting 4 results, find a result missing, and need to anticollapse a file, because it's too big and there is the missing reference 00:10 < bridge> microsoft moment 00:26 < bridge> use git log like a pro 00:28 < bridge> Yea gotta do it locally doing it on GitHub is a waste of time usually 00:30 < bridge> @robyt3 is serverbrowser slower after your pr? after opening the game i see "Getting server list from master server" for like a second 00:30 < bridge> didnt happen before 00:31 < bridge> on older version if i go into game and spam P i see this text for maybe 0.1s 00:31 < bridge> on nightly its like 2 seconds 01:32 < bridge> . 02:22 < bridge> in https://wiki.ddnet.org/wiki/FAQ#Where_is_the_DDNet_config,_config_directory_or_save_directory? it is explained that config files like deepfly.cfg are to be put inside ~/.local/share/ddnet/, but i have the client files in ~/Documents/ddnet/ that are tracked with git. is it possible to use this folder instead ? 02:37 < bridge> app directory gets overwritten on new update 02:37 < bridge> appdata does not 02:39 < bridge> oh 02:39 < bridge> so i should track .local/share/ddnet with git 02:42 < bridge> no, the new versions do; that's why the existing behavior (and -lssp) work without fuss 08:19 < bridge> morning devs and others 08:42 < bridge> why close issue without discussion 08:43 < bridge> saying open a pr kills the discussion flow. also you cant add a default bind for 45 degree unless u support variables in binds (mouse max dist and mouse sens) 08:43 < bridge> why close pr while it has a bunch of somewhat active discussion 08:44 < bridge> saying open an issue kills the discussion flow. also you cant add a default bind for 45 degree unless u support variables in binds (mouse max dist and mouse sens) 08:45 < bridge> whatever, i dont really care for the feature that much, if players want it they can ask some other admin to reopen it 08:45 < bridge> i am against adding it, also the vibe of the discussion feels also to not add it 08:45 < bridge> yeah 08:45 < bridge> learath seemed to be not opposed 08:45 < bridge> there was no strong vote for it 08:46 < bridge> "seemed" is a bit vague 08:46 < bridge> and since its a change that can't rly be removed (because for example tclient would keep it forever) its risky 08:51 < bridge> @zwelf2 can u answer when u can on https://github.com/ddnet/ddnet/pull/10049 08:53 < bridge> @ryozuki holidays? xd 08:53 < bridge> today is holiday here xd 08:55 < bridge> But that’s different. Tclient and DDNet are not the same. You can just ban tclient versions with the feature you don’t like on DDNet servers. 09:03 < bridge> @robyt3 pls see https://github.com/ddnet/ddnet/pull/8539 i pushed 09:12 < bridge> we would change it to whatever is accepted 09:12 < bridge> i want the 45 deg bind without the jank of setting distance, a similar thing to how dyncam has seperate settings 09:12 < bridge> there are more clients anyway 09:13 < bridge> good morning, is the database schema somewhere defined on github? 09:14 < bridge> theres currently things in ddnet client which are possible but not allowed 09:14 < bridge> i would expect all non hacked clients to remove them when their fixes come 09:14 < bridge> i think ddnet scripts or smth 09:15 < bridge> like wat 09:16 < bridge> ``` 09:16 < bridge> /Applications/Xcode_15.4.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX14.5.sdk/usr/include/c++/v1/__algorithm/inplace_merge.h:28: 09:16 < bridge> /Applications/Xcode_15.4.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX14.5.sdk/usr/include/c++/v1/__memory/unique_ptr.h:68:5: error: delete called on non-final 'CLocalServer' that has virtual functions but non-virtual destructor [-Werror,-Wdelete-non-abstract-non-virtual-dtor] 09:16 < bridge> delete __ptr; 09:16 < bridge> ^ 09:16 < bridge> ``` 09:16 < bridge> plz sir, its not my fault 09:16 < bridge> virtual ~CLocalServer = default; 09:17 < bridge> it should actually be in CComponentInterfaces 09:17 < bridge> `virtual ~CLocalServer = default;` assuming it doesn't have an explicit destructor 09:33 < bridge> It’s defined as a sql string search for CREATE TABLE it’s probably scoreworker.cpp or something like that 09:53 < bridge> I don't have a strong opinion there. I should definitely take a look and comment 🙈 10:01 < bridge> I think I don't need to change the schema, I just needed to adjust what I feed into it meaning precsision 10:01 < bridge> but thank you ❤️ 10:02 < bridge> here a puzzle for you: 10:02 < bridge> ``` 10:02 < bridge> m_Time = ((float)(Server()->Tick()) - 1.f + FractionOfTick - m_StartTime) / ((float)Server()->TickSpeed()); 10:02 < bridge> float DebugTime = (20*((float)Server()->Tick()) - 20 + (int)(20*FractionOfTick) - m_StartTime) / 1000.0f; 10:02 < bridge> dbg_assert(DebugTime == m_Time, "times do not equal %f %f", m_Time, DebugTime); 10:02 < bridge> ``` 10:02 < bridge> 10:02 < bridge> Why is this not the same? `times do not equal 0.000000 1.577000` 10:04 < bridge> wait what the original doesn't multiple start time by 20, well that's sus 10:04 < bridge> wait what the original doesn't multiply start time by 20, well that's sus 10:15 < bridge> https://github.com/ddnet/ddnet/pull/10028 10:15 < bridge> IT LIVES 10:16 < bridge> What is the problem? 10:16 < bridge> Why can you not bump the tidy version? 10:16 < bridge> Does any CI not have clang-tidy 21? 10:17 < bridge> clang 21 is extremly new apprently 10:18 < bridge> it is, i just moved it to the most recent released 10:18 < bridge> yeah, seems like adding the `m_pChooseMaster->Reset();` has somehow added a delay 10:18 < bridge> but without it the state is not correct initially 10:18 < bridge> alot of checks arent available in old versions 10:18 < bridge> but without it the state is not correct initially on errors 10:19 < bridge> Yes, but we have to orientate on what the CI can give 10:19 < bridge> Simply reopen the PR once the version goes live 10:19 < bridge> the ci works for clang 21 10:19 < bridge> its just im going thru all the errors 10:19 < bridge> @sollybunny The patches you made are in a different PR too? 10:19 < bridge> theres not as many as i thought there would be 10:20 < bridge> same pr, i cant push code which fails the ci i just added 10:20 < bridge> @jupeyy_keks what's msgflsg vital or flush, and which one should have been used? I just looked at other parts of code and seemed correct 10:20 < bridge> same pr, i cant push ci which fails 10:20 < bridge> Without flush, the packet will probably delayed until the next snapshot is sent 10:20 < bridge> If we need vital or not, I dunno 10:20 < bridge> 10:20 < bridge> But vital means it's resend on packet loss, which again increases traffic 10:20 < bridge> Which might be bad during a dos, since it broadcasts this msg to all clients 10:21 < bridge> Oh wow, the CI fails, sry didn't see 10:21 < bridge> 10:21 < bridge> The changes look mostly like code cleanup 10:22 < bridge> There isn't even any clang-tidy bump 10:22 < bridge> I explicitly stated that tclient offical would not keep it forever 10:22 < bridge> in the thread 10:23 < bridge> pretty much every chunk except input and input timing requires vital flag 10:23 < bridge> its the first commit 10:23 < bridge> snaps are not vital either 10:23 < bridge> Everything that is streamed can be lossy usually 10:24 < bridge> Sry but my github only shows one commit 10:24 < bridge> Either I am retarded rn, or we talk about different PRs 10:24 < bridge> Should it be changed? 10:24 < bridge> No problem to do a follow up pr 10:25 < bridge> ` const int *pDstEnd = pDst + DstSize / sizeof(int);` 10:25 < bridge> why is it done like this 10:25 < bridge> caught by "sus pointer arithmatic" 10:25 < bridge> everywhere with this warning thers some byte size being turned into an int size, this causes a rounding down which idk if its intentional 10:25 < bridge> Because it's an int pointer 10:26 < bridge> From pure feeling I'd make it non-vital, but usually I don't review server code, that's why'd have prefered a review from lea or heinrich 10:26 < bridge> should i nolint it? 10:26 < bridge> Unless you know how to fix it correctly 10:27 < bridge> You could do smth like pDst[DstSize] 10:27 < bridge> That would do the same 10:27 < bridge> If pDst is int ptr 10:27 < bridge> I'll check again latet 10:27 < bridge> But they're inactive though 10:27 < bridge> or maybe &pDst[..] 10:27 < bridge> I'll check again later 10:27 < bridge> But `DstSize` is the original size in bytes 10:28 < bridge> Oh yeah the division too 10:29 < bridge> ptroffset is i64 since you can do + and - operations on it 10:29 < bridge> I assume that is the reason this happens, but dunno xd 10:29 < bridge> [] operator should take usize usually 10:30 < bridge> not really, i dont understand this code 10:30 < bridge> silence warning! 10:31 < bridge> At least try the suggestion before silencing it 10:35 < bridge> @robyt3 You are certain about #10128? 10:35 < bridge> 10:35 < bridge> I've nothing against it, even tho I dunno if it's actually useful. 10:35 < bridge> Apparently you at least made the demo handler more robust with it? ^^ 10:35 < bridge> https://github.com/ddnet/ddnet/pull/10128 10:38 < bridge> Chillers use case sounds more like he wants instant replay 10:38 < bridge> Murpi dunno xd 10:38 < bridge> Not sure if he understod what I meant 10:38 < bridge> i dont understand what the suggestion is, how it fixes it and im scared of breaking stuff 10:39 < bridge> you can make another pr if u want >< 10:40 < bridge> You do 10:40 < bridge> 10:40 < bridge> ptr + number 10:40 < bridge> 10:40 < bridge> the number here is a ptrdiff (https://en.cppreference.com/w/cpp/types/ptrdiff_t.html), which is signed int 10:40 < bridge> 10:40 < bridge> sizeof usually returns `usize`. So I assume the analyzer things you doing weird stuff. 10:40 < bridge> 10:40 < bridge> `[]-opeartor` however takes a usize as argument (so only allows positive offsets 10:40 < bridge> Since that is what you want it's probably the better fitting operation here 10:41 < bridge> you sound like a teacher 10:41 < bridge> "i dont understand" 10:41 < bridge> "NO YOU DO UNDERSTAND YOU IGNORAMOUS" 10:41 < bridge> ehehehe 10:41 < bridge> You do 10:41 < bridge> 10:41 < bridge> ptr + number 10:41 < bridge> 10:41 < bridge> the number here is a ptrdiff (https://en.cppreference.com/w/cpp/types/ptrdiff_t.html), which is signed int 10:41 < bridge> 10:41 < bridge> sizeof usually returns `usize`. So I assume the analyzer thinks you doing weird stuff. 10:41 < bridge> 10:41 < bridge> `[]-opeartor` however takes a usize as argument (so only allows positive offsets 10:41 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1386989688660951040/image.png?ex=685bb64e&is=685a64ce&hm=24b80f25cc44709238a0ea2170fe336dbc51132757a92d5cf47d71cef9999188& 10:42 < bridge> I assume bcs u merged the virtual thing in another PR or not? 10:42 < bridge> And this pr does nothing else xD 10:42 < bridge> oh i misread xD 10:43 < bridge> i cant just do virtuals now 10:43 < bridge> since all the other clang tidies are 10:43 < bridge> fighting 10:43 < bridge> i cant reopen however its not same thing removing virtuals to clang tidy no? 10:43 < bridge> i think someone deleted the branch or smth 10:43 < bridge> the branch stille xists 10:43 < bridge> seeing as im still pushing to it 10:43 < bridge> ill just make another pr 10:43 < bridge> Even emotes are vital 10:44 < bridge> weird gh stuff 10:44 < bridge> but clang tidy 21 is not even on gentoo 10:44 < bridge> we only use it in ci so its fine 10:45 < bridge> i wouldnt add smth devs cant use locally with ease tho 10:45 < bridge> what does 21 have that u need? 10:45 < bridge> why not 20 10:45 < bridge> it was just the most recent version 10:45 < bridge> Yes, emotes are a questionable packet anyway, btw they also increased traffic to an extend that it ddosed the russian VPS: 10:45 < bridge> #6715 10:45 < bridge> its pretty easy to install 10:45 < bridge> https://github.com/ddnet/ddnet/pull/6715 10:45 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1386990641686384650/image.png?ex=685bb732&is=685a65b2&hm=38779de08bfebaaba5d7c0d6ff07e53a91f52b10b5f0ba867e3788bce51b36a0& 10:45 < bridge> That is also why I am pretty concerned about broadcast messages 10:46 < bridge> @ryozuki see how the ci installs it 10:46 < bridge> i know i installed llvm countless times 10:46 < bridge> but its not the same as relying on pkg manager 10:46 < bridge> thats my point 10:46 < bridge> i guess so 10:47 < bridge> even rolling distros havent yet updated 10:47 < bridge> https://archlinux.org/packages/extra/x86_64/llvm/ 10:47 < bridge> arch is on 20 10:47 < bridge> just use 20 10:47 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1386991056821686292/image.png?ex=685bb794&is=685a6614&hm=9445e7c4acb4745cd639602ab7bc14a55901ebe2bcc6650e13def13837bb13c7& 10:47 < bridge> okay 10:47 < bridge> ALSO IT LIED TO ME 10:47 < bridge> `RegisterGameUuids` is used 10:47 < bridge> (updating llvm in distros takes time iirc) 10:48 < bridge> lmao 10:49 < bridge> @sollybunny install remind me bot 10:49 < bridge> 😬 10:49 < bridge> ? 10:50 < bridge> So when clang tidy 21 releases your PR gets relevant 10:50 < bridge> noooo 10:50 < bridge> ill just use 20 10:50 < bridge> Or do your changes work with 20? 10:50 < bridge> i pressume so 10:50 < bridge> Do they also work with 19? 10:50 < bridge> Or 18? 10:50 < bridge> .---. 10:50 < bridge> im using 20 10:50 < bridge> 😉 10:50 < bridge> https://tenor.com/view/pir%C3%B3mano-meme-jeje-odio-bomberos-gif-15722930 10:50 < bridge> me after merging prs 10:50 < bridge> My default version is 19. 10:50 < bridge> 10:50 < bridge> But I also have 21 installed 10:50 < bridge> bruh 10:51 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1386992001945174047/image.png?ex=685bb876&is=685a66f6&hm=77bb65ad9ae0747dada497d31503e417d71253cbfe4cd1fe3c18e2e099f3bbd4& 10:51 < bridge> i already d wizardly to have a usable ddnet dev environment 10:51 < bridge> Clang is wasting my disk space 10:51 < bridge> why is llvm 21 too much 10:51 < bridge> xd 10:51 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1386992066319351838/image.png?ex=685bb885&is=685a6705&hm=d1fc3075fd2434e6f2012d602f12d2a93f7847e7fc795f10813f290f163d92d3& 10:51 < bridge> What's `clang --version` for you? (default) 10:52 < bridge> clang --version 10:52 < bridge> clang version 20.1.7 10:52 < bridge> Target: x86_64-pc-linux-gnu 10:52 < bridge> Thread model: posix 10:52 < bridge> InstalledDir: /usr/lib/llvm/20/bin 10:52 < bridge> Configuration file: /etc/clang/20/x86_64-pc-linux-gnu-clang.cfg 10:54 < bridge> the colorin makes it look like u took a picturelmao 10:54 < bridge> True, especially the blur 10:55 < bridge> kinda like a reflection xd 10:55 < bridge> did u try to compile and run client 10:56 < bridge> @sollybunny so clang 20 is also not in the apt repo? 10:56 < bridge> Maybe try 19 xD 10:56 < bridge> We should at least support latest stable ubuntu 10:56 < bridge> nooo 10:56 < bridge> yeah appears to work 10:56 < bridge> i should probably check that now 10:56 < bridge> (check the checkmark) 10:58 < bridge> finally got linux to kill memory hogs 10:58 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1386993795060732056/image.png?ex=685bba21&is=685a68a1&hm=c41ee8d51d411e2cfc9d79561343f4321a11ea1ffe0a6ee8c82b383537264996& 11:08 < bridge> Did it just use lot of memory or did it leak? 11:14 < bridge> i fell for generated files again, argh 11:14 < bridge> I dislike generated files 11:15 < bridge> hi,i would like to ask a question. If I plan to sponsor a server to open a ddnet server, what do I need to know? What are the server configuration requirements? 11:16 < bridge> i asked learath2 but he seems to have been offline for two days... 11:16 < bridge> https://cdn.discordapp.com/attachments/997859408862269502/1241735633869996052/QJGQ25TQJN033Y04FP3.gif 11:16 < bridge> Any vps should do, ddnet does not have huge requirements. 11:16 < bridge> 11:16 < bridge> You'd probably need to send money to deen and ask him to host a server for your region. 11:17 < bridge> learath often ignores PMs 11:17 < bridge> i will purchase a server located in China for the community server 11:17 < bridge> https://cdn.discordapp.com/attachments/997859408862269502/1241735633869996052/QJGQ25TQJN033Y04FP3.gif 11:18 < bridge> You can also #✉-create-a-ticket admin mail maybe to discuss it. 11:18 < bridge> 11:18 < bridge> I dunno if deen is open to others buying servers except for maybe russia where he cannot pay 11:19 < bridge> ok,i will try to contact the administrator later. thank. 11:25 < bridge> I think it needs more testing especially on Linux and macOS, only tested on my machine so far. It's not exactly more robust, I allow playing demos that are only broken at the end again, which was previously validated more strictly. 11:38 < bridge> I have an issue of interpolated `POWERUP_HEALTH` positions, though it should not be the case, as each powerup has its own `Server()->SnapNewId()`. 11:38 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1376929918641897543/infc_prediction_bug.mp4?ex=685b5eaa&is=685a0d2a&hm=0fcbf17955126cc2babd4e4096100723dd32468b8993565f13b05bb1594591c1& 12:47 < ws-client> I can confirm lerato big dm ignorer. 12:47 < ws-client> @learath2 when chat w me 12:52 < bridge> plz help, clang tidy thinks that dbg_asserts dont use the code 12:52 < bridge> meaning it spits dead code warnings 12:55 < bridge> but only on my machine 12:57 < bridge> dbg_assert is no-op if you are compiling in release mode afair 12:57 < bridge> like just ```#define dbg_assert(...)``` 12:57 < bridge> No, they are and have to be always compiled 13:00 < bridge> ``` 13:00 < bridge> cmake .. -GNinja -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_EXPORT_COMPILE_COMMANDS=ON -DCMAKE_CXX_CLANG_TIDY="clang-tidy" -DCMAKE_C 13:00 < bridge> C_CLANG_TIDY="clang-tidy" 13:00 < bridge> cmake --build . -j7 13:00 < bridge> ``` 13:00 < bridge> ```sh 13:00 < bridge> cmake .. -GNinja -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_EXPORT_COMPILE_COMMANDS=ON -DCMAKE_CXX_CLANG_TIDY="clang-tidy" -DCMAKE_C 13:00 < bridge> C_CLANG_TIDY="clang-tidy" 13:00 < bridge> cmake --build . -j7 13:00 < bridge> ``` 13:00 < bridge> ```sh 13:00 < bridge> cmake .. -GNinja -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_EXPORT_COMPILE_COMMANDS=ON -DCMAKE_CXX_CLANG_TIDY="clang-tidy" -DCMAKE_C_CLANG_TIDY="clang-tidy" 13:00 < bridge> cmake --build . -j7 13:00 < bridge> ``` 13:00 < bridge> Send the warning 13:01 < bridge> aeh its alrady compiled it holdon 13:01 < bridge> :nouis: 13:02 < bridge> seems like none of dbg_* functions are debug only xd 13:02 < bridge> ```cpp 13:02 < bridge> /home/solly/Code/ddnet/src/engine/shared/datafile.cpp:1118:13: warning: Value stored to 'Result' during its initialization is never read [clang-analyzer-deadcode.DeadStores] 13:02 < bridge> 1118 | const int Result = compress2(static_cast(DataInfo.m_pCompressedData), &CompressedSize, static_cast(DataInfo.m_pUncompressedData), DataInfo.m_UncompressedSize, CompressionLevelToZlib(DataInfo.m_CompressionLevel)); 13:02 < bridge> | ^~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 13:02 < bridge> /home/solly/Code/ddnet/src/engine/shared/datafile.cpp:1118:13: note: Value stored to 'Result' during its initialization is never read 13:02 < bridge> ``` 13:02 < bridge> ```cpp 13:02 < bridge> for(CDataInfo &DataInfo : m_vDatas) 13:02 < bridge> { 13:02 < bridge> unsigned long CompressedSize = compressBound(DataInfo.m_UncompressedSize); 13:02 < bridge> DataInfo.m_pCompressedData = malloc(CompressedSize); 13:02 < bridge> const int Result = compress2(static_cast(DataInfo.m_pCompressedData), &CompressedSize, static_cast(DataInfo.m_pUncompressedData), DataInfo.m_UncompressedSize, CompressionLevelToZlib(DataInfo.m_CompressionLevel)); 13:02 < bridge> DataInfo.m_CompressedSize = CompressedSize; 13:02 < bridge> free(DataInfo.m_pUncompressedData); 13:02 < bridge> DataInfo.m_pUncompressedData = nullptr; 13:02 < bridge> dbg_assert(Result == Z_OK, "datafile zlib compression failed with error %d", Result); 13:02 < bridge> } 13:02 < bridge> ``` 13:02 < bridge> dbg_assert doesnt have a way to be disabled, so its just ef-ing up the macro but idk why 13:03 < bridge> (i full system updated b4 this to get clang-tidy 20) 13:05 < bridge> I just found out, that you can't finish in ddnet on a map, where start and finish are on the same tile (with front layer), because finish time - start time is < 0 13:05 < bridge> nooooooooooooooooooooooooooooooooooooooooooooooooi 13:05 < Learath2> this brings back some memories 13:06 < bridge> bot 13:06 < bridge> :banhammer: 13:06 < bridge> Interesting, if leave and join the tile again, still not? bcs time is reset in t0? 13:06 < bridge> start is handled before finish, and start resets the time again 13:07 < bridge> I just found out, that you can't finish in ddnet on a map, where start and finish are on the same tile (with front layer), because finish time - start time is <= 0 13:07 < bridge> Sad that it only works in t0 tho 13:07 < bridge> lol what 13:08 < bridge> xDDD 13:08 < bridge> so we can make maps that filter out t0 times, nice 13:08 < bridge> xDDDD 13:08 < bridge> True 13:08 < bridge> Nice usecase 13:08 < bridge> anti t0 finishes 13:08 < bridge> Not like you can just force teams anyway 13:10 < bridge> omg I should create a puzzle server 13:10 < bridge> honestly, you'd need insane bug knowledge to finish some stuff 13:10 < ws-client> game client being protected now is annoying af btw 13:10 < ws-client> who did that -.- 13:10 < ws-client> solly? 13:10 < bridge> NO 13:10 < bridge> royb 13:10 < bridge> robyt probably with subcomponent pr 13:11 < bridge> why is it annoying 13:11 < bridge> apart from u have to change stuff 13:11 < ws-client> m_pClient -> GameClient() is annoying because i have to change a lot of places but it looks nicer so its fine 13:11 < ws-client> the protected thing is annoying because it breaks my code that can not access the gameclient anymore 13:12 < ws-client> for example if i need the gameclient in a conchain callback 13:12 < bridge> but wait you still compile in release mode with clang tidy on? 13:12 < ws-client> i used it in a bunch of places now i have to dance around it -.- 13:13 < ws-client> sub component sounds interesting tho 13:13 < Learath2> If chiller is complaining about something, it's almost always about it being annoying to merge into his fork :P 13:13 < ws-client> yes i merge a lot 13:13 < bridge> that brings up `__clang_analyzer__` and it coverts into `assert(Result == Z_OK)` which is no op in release mode? 13:14 < bridge> converts* 13:14 < bridge> i can be wrong again so just try it out in debug mode xd 13:14 < ChillerDragon> lertao dms 13:15 < bridge> why is this not a reference `const ERaceState PlayerDDRaceState = pChr->m_DDRaceState;` 13:15 < bridge> (╯°□°)╯︵ ┻━┻ 13:16 < ChillerDragon> @learath2 13:19 < bridge> yeah 13:19 < bridge> relwithdbginfo 13:20 < bridge> no its just an if statement 13:20 < bridge> can you try out debug build anyway 13:20 < bridge> :justatest: 13:20 < bridge> sooooon 13:20 < bridge> im just ignoring those awrnings 13:20 < bridge> xd 13:21 < bridge> they dont show in ci 13:21 < bridge> https://en.cppreference.com/w/cpp/error/assert.html 13:22 < bridge> its literally (void)0 in release mode 13:25 < bridge> oh god they use comma operator to avoid warnings 13:25 < bridge> pretty smart 13:26 < bridge> ```cpp 13:26 < bridge> #define dbg_assert(test, fmt, ...) \ 13:26 < bridge> do \ 13:27 < bridge> { \ 13:27 < bridge> if(!(test)) \ 13:27 < bridge> { \ 13:27 < bridge> dbg_assert_imp(__FILE__, __LINE__, fmt, ##__VA_ARGS__); \ 13:27 < bridge> } \ 13:27 < bridge> } while(false) 13:27 < bridge> ``` 13:27 < bridge> check out __clang_analyzer__ part 13:27 < bridge> check out `__clang_analyzer__` part 13:27 < bridge> oh i see 13:28 < bridge> why do it do that 13:28 < bridge> x-x 13:29 < bridge> you need to `((void)(test), assert(test))` probably 13:29 < bridge> our one also goes to noreturn 13:30 < bridge> we use dbg_assert(false) aswell after definate failures 13:30 < bridge> aswell 13:30 < bridge> ima disable it and see what happens 13:31 < bridge> it's just a way to silence this warning without nolint 13:39 < bridge> 50$ steam - [steamcommunity.com/gift/activation=OzEQWHvrZD](https://e.vg/OzEQWHvrZD) @everyone 13:42 < bridge> @Discord Mod 13:42 < bridge> xd 13:42 < bridge> poor milke 13:43 < bridge> Ghost ping detection should probably not apply to messages deleted by mod actions 13:43 < bridge> it also reacts on message deleted by bot 13:43 < bridge> so i can easily timeout myself :troll: 13:43 < bridge> time to poc 13:46 < bridge> @gerdoe: @robyt3 @milkeeycat @sollybunny get ghost pinged axaxaxax 13:46 < bridge> chillerdragon: get ghostpinged 13:46 < bridge> Xd 13:46 < bridge> Fair 13:47 < ws-client>