DDraceNetwork - developer

DDraceNetwork

Development / developer

Development discussion. Logged to https://ddnet.tw/irclogs/ Connected with DDNet's IRC channel, Matrix room and GitHub repositories — IRC: #ddnet on Quakenet | Matrix: #ddnet-developer:matrix.org GitHub: https://github.com/ddnet

Between 2023-03-21 00:00:00Z and 2023-03-22 00:00:00Z

HELLO! I am a bot created by @dzlandis:mozilla.org that detects phishing/malicious links sent in your chat rooms and notifies users that they are malicious. Feel free to join #phishbot:matrix.org for questions or concerns. Please kick me if you would not like me here. If you would like me to automatically delete phishing messages or kick users who send phishing links, please give me those permissions. For more information, please visit: https://github.com/dzlandis/phish-matrix-bot

GitHub - dzlandis/Phish-Matrix-Bot: A bot for detecting phishing li...

A bot for detecting phishing links across Matrix. Contribute to dzlandis/Phish-Matrix-Bot development by creating an account on GitHub.

Matrix - Decentralised and secure communication

You're invited to talk on Matrix. If you don't already have a client this link will help you pick one, and join the conversation. If you already have one, this link will help you join the conversation

hi Phish Bot

It sounds like no one invited it

probably Zwelf did it on the matrix side

Learath2

https://tc39.es/ecma262/#sec-html-like-comments

oh wow, take a close look at that again

(edited)

02:55

 are for single-line comments in JS

02:56

so they can actually be used to hide JS code from browsers that do not support JS

02:56

I remembered it correctly ^^

02:57

<!-- console.log("abc");
     console.log("def");
-->  console.log("ghi");

prints just def

02:58

@Ryozuki too ^^

you can also just check if the browser supports javascript & not use a hack (edited)

WTF (@heinrich5991)

heinrich5991#9827

<!-- console.log("abc");
     console.log("def");
-->  console.log("ghi");

prints just def

https://motioncanvas.io/ really cool project lol

Visualize complex ideas programmatically

Phish Bot

HELLO! I am a bot created by @dzlandis:mozilla.org that detects phishing/malicious links sent in your chat rooms and notifies users that they are malicious. Feel free to join #phishbot:matrix.org for questions or concerns. Please kick me if you would not like me here. If you would like me to automatically delete phishing messages or kick users who send phishing links, please give me those permissions. For more information, please visit: https://github.com/dzlandis/phish-matrix-bot

am I malicious justatest

justatest

[ddnet:master] 2 new commits

8d4f851 Fix console chain of inp_controller_guid - Robyt3 323d8bb Merge #6448 - bors[bot]

08:20

[ddnet/ddnet] Pull request closed: #6448 Fix console chain of `inp_controller_guid`

heinrich5991

probably Zwelf did it on the matrix side

hm, no. Not my doing. Is it normal that anti-spam software advertises itself via spam

?

09:23

doesnt make it really trustworthy

https://releases.llvm.org/16.0.0/docs/ReleaseNotes.html

https://www.youtube.com/watch?v=XP7Qx1FF1hA

Holding corporate criminals ACCOUNTABLE; it's about time we start

10:55

related to gpl

https://petals.ml/

and whats this

Run large language models like BLOOM-176B collaboratively — you load a small part of the model, then team up with people serving the other parts to run inference or fine-tuning.

12:41

@Voxel do u know how torrent works?

12:41

its basically a peer to peer compute power sharing

12:41

to run a big model

ahh

All MySQL collations are of type PAD SPACE. This means that all CHAR, VARCHAR, and TEXT values are compared without regard to any trailing spaces. “Comparison” in this context does not include the LIKE pattern-matching operator, for which trailing spaces are significant.

14:04

TIL

14:04

was part of a ctf challenge

is the quote related to mysql being part of a ctf?

the quote was part of what you had to know/find out/guess in a ctf challenge (edited)

timakro is here, pog

Hi, I wasn't here for a while :D I'm looking for C debugging help as I'm kinda rusty. Any idea why this segfaults? I have a coredump

(gdb) frame
#0  0x000055b195b811a1 in on_change_systray (obj=0x55b195bcc940 <systray>) at /home/tim/Downloads/tint2/src/systray/systraybar.c:298
298         GSList *list = g_slist_copy(systray.list_icons);
(gdb) print systray.list_icons
$5 = 0x55b19749d2a0 = {0x55b19736f250, 0x55b19730ab50, 0x55b197441ff0, 0x55b197370a40}
(gdb)

sad that rusty is not a pun

could as well be a pun (wouldn't have issues like these in rust) haha

haha

14:12

idk much context

14:12

i just made a quick google https://stackoverflow.com/questions/5873780/glib-type-testing-macros-cause-segfault-when-passed-a-gslist

Glib type testing macros cause segfault when passed a GSList

Passing a GSList to a Glib macro (Like G_OBJECT_TYPE_NAME(), GTK_IS_WIDGET() etc) causes a segmentation fault. This is a problem as my program has to process a list of GObjects and can't test for

But if thats stackframe #0 it means the bug is in that line, not g_slist_copy, right?

Note that this is a “shallow” copy. If the list elements consist of pointers to data, the pointers are copied but the actual data isn’t. See g_slist_copy_deep() if you need to copy the data as well.

14:14

hmm idk tbh xD

14:17

https://github.com/o9000/tint2/blob/master/src/systray/systraybar.c

14:17

tbh i never used glib

14:17

maybe @Learath2 knows

14:17

he loves c

Yeah I modified this, it crashes on a line I wrote. I tried to patch another segfault but not very successfully ^^

well the only thing i can point out is that ur making a shallow copy, idk much else :p

did you try asan and ubsan? it usually cuts most debugging sessions down to a couple minutes

x/i $rip pls

(gdb) x/i $rip
=> 0x55b195b811a1 <on_change_systray+658>:      mov    (%rax),%rax

Learath2

did you try asan and ubsan? it usually cuts most debugging sessions down to a couple minutes

Hm, may check those out

what does x/i

14:28

oh so $rip is where it crashed

14:28

til

Had to google it: x is "inspect memory", /i interprets it as instruction, $rip is instruction pointer (edited)

14:29

So basically prints the current instruction at time of segfalt

FAILED: release/ddnet_engine_shared.lib C:/Users/source/repos/ddnet/out/build/x64-Release/release/ddnet_engine_shared.lib

(edited)

15:03

how to fix this while freshly building ddnet server?

15:03

on win with vs

15:04

[12/133] Generating release/ddnet_engine_shared.lib seems to have worked, so i dont know

15:10

i think its a Rust issue, but since when do we need Rust to compile the server?

15:14

Rust is not in the ddnet libs, how to get it to work?

Rust is part of the dependencies, you need to install it to build: https://github.com/ddnet/ddnet#dependencies-on-linux--macos

GitHub - ddnet/ddnet: DDraceNetwork, a free cooperative platformer ...

DDraceNetwork, a free cooperative platformer game. Contribute to ddnet/ddnet development by creating an account on GitHub.

15:18

You can google for how to install Rust on Windows

we got it

15:24

thanks

what did you do to fix it?

magick

guyz

16:22

how can i build the new version?

just as the old version

16:23

from what version did u upgrade

there are something different

whats ur OS whats the old ddnet version u built

windows

16:24

idk

16:24

16

is rust installed?

little check

did u do git submodule update i dont even know if we had updates tho xd

the command cargo give me a response so i think that rust is installed

yeah

16:25

else share what error you exactly get

16:26

that is probably easier

Image attachment

16:26

maybe i choose the wrong kit

Jupstar ✪

did u do git submodule update i dont even know if we had updates tho xd

git clone --recursive https://github.com/ddnet/ddnet

use amd64 for example, or mingw x86_64 if u have it installed

amd of VS?

16:30

wait maybe

yes the other is 32bit

16:30

u probably dont want it

yes

16:30

something is happending xd

16:31

d] [ 13%] Built target rust_engine_shared_target [build] mingw32-make: *** [Makefile:155: all] Error 2 [proc] The command: "C:\Program Files\CMake\bin\cmake.EXE" --build f:/KeClient/build --config Debug --target all -j 14 -- exited with code: 2 [build] Build finished with exit code 2

16:31

finished here

16:32

Consolidate compiler generated dependencies of target steam_api
[  0%] Built target steam_api
Consolidate compiler generated dependencies of target md5
[  1%] Built target md5
[  2%] Building CXX object CMakeFiles/engine-gfx.dir/src/engine/gfx/image_loader.cpp.obj
In file included from F:\KeClient\src\engine\gfx\image_loader.cpp:3:0:
F:/KeClient/src/base/system.h:2747:13: error: specialization of 'template<class _Tp> struct std::hash' in different namespace [-fpermissive]
 struct std::hash<NETADDR>
             ^~~~~~~~~~~~~
In file included from c:\mingw\lib\gcc\mingw32\6.3.0\include\c++\bits\stl_bvector.h:1262:0,
                 from c:\mingw\lib\gcc\mingw32\6.3.0\include\c++\vector:65,
                 from F:\KeClient\src\engine\gfx\image_loader.h:6,
                 from F:\KeClient\src\engine\gfx\image_loader.cpp:1:
c:\mingw\lib\gcc\mingw32\6.3.0\include\c++\bits\functional_hash.h:58:12: error:   from definition of 'template<class _Tp> struct std::hash' [-fpermissive]
     struct hash;
            ^~~~
cc1plus.exe: warning: unrecognized command line option '-Wno-nullability-completeness'
mingw32-make[2]: *** [CMakeFiles\engine-gfx.dir\build.make:76: CMakeFiles/engine-gfx.dir/src/engine/gfx/image_loader.cpp.obj] Error 1
mingw32-make[1]: *** [CMakeFiles\Makefile2:383: CMakeFiles/engine-gfx.dir/all] Error 2
mingw32-make: *** [Makefile:155: all] Error 2

 *  The terminal process "C:\Windows\System32\WindowsPowerShell\v1.0\powershell.exe -Command mingw32-make" terminated with exit code: 1. 
 *  Terminal will be reused by tasks, press any key to close it.

better dont use mingw32

16:34

its a pretty broken version

Image attachment

mh then it seems like it detects it incorrectly

maybe for that wait

16:34

408 bytes

cmake-tools-kits.json

328 bytes

3.92 KB

1.94 KB

16:34

this is my vscode configuration

@murpi what is KeClient btw?

?

16:35

i gave a random name to it

mh ok

16:35

always have to be careful in this chat xd

about what

about cheaters

lel

kio

Click to see attachment 🖼️

where did u even get them from

16:37

other/vscode/ddnet.code-workspace contains a ddnet workspace i dunno your tasks.json looks like someone else created it

Jupstar ✪

about cheaters

I had started to create this client that was supposed to have sexy graphics and some features, but then I didn't have time to do it, now I have time and I wanted to continue but I need a decent version, everything worked before, but downloading this new version doesn't work anymore ddnet normal lel

@kio I hope that's not a bot client? You previously posted code which is used for bots

Jupstar ✪

where did u even get them from

this chat and after i edited that and made it works

murpi

@kio I hope that's not a bot client? You previously posted code which is used for bots

like?

problem is you using mingw32 make binary, mix it with Visual studio c++

16:40

maybe copy paste your whole workspace and delete all these .vscode files and try again

You deleted your old repo "KVClient" which had e.g: spinbot

murpi

You deleted your old repo "KVClient" which had e.g: spinbot

was a local texture spin lel

Hm Hm

murpi

You deleted your old repo "KVClient" which had e.g: spinbot

https://github.com/k-i-o/KRV-Client

GitHub - k-i-o/KRV-Client: DDraceNetwork, a cooperative racing mod ...

DDraceNetwork, a cooperative racing mod of Teeworlds - GitHub - k-i-o/KRV-Client: DDraceNetwork, a cooperative racing mod of Teeworlds

16:42

I program for fun, if I had to make cheats inside the client I would do them for sport but it would only be a personal goal, I would not use them and I would not spread them, however my goal now is to make a decent client with nice graphics, so don't worry

Jupstar ✪

maybe copy paste your whole workspace and delete all these .vscode files and try again

without these files which command i had to do

i'd say u dont really have to do a lot simply select a kit and start F7

16:45

then it should compile

16:45

thats all i ever did in vscode be it on windows or linux

Do you recommend uninstalling something and reinstalling it to fix the kits or do I just select I don't know what when I put the kit?

16:51

@murpi do you understand and approve? otherwise tell me and I stop doing anything

16:51

to practice finding memory addresses etc I did some little useless external hacks, but I don't think it's a problem, it's just fun

(edited)

Jupstar ✪

i'd say u dont really have to do a lot simply select a kit and start F7

Image attachment

16:53

it says it can't find the dll, and every time I press ok it tells me a different dll

kio

Click to see attachment 🖼️

then copy them manually

16:54

looks like vsc++ create sub dirs called debug

yes idk why

Is this just trying to compile ddnet master branch?

never happened

Learath2

Is this just trying to compile ddnet master branch?

yes

16:54

Did you post the build log somewhere?

14.03 KB

Hm, rust seems to fail to create something too. Could there be a permissions issue with the folder?

16:56

Though seems the exe files are created normally

I try to reopen vscode with admin permissions

16:58

5.22 KB

so the build worked?

same error

17:00

build worked but no dll generated

can you try copying the dll manually?

I put all the dlls by hand and the error disappeared, but when opening the game it crashes in the loading screen..

isn't libpng not generated but an external lib ?

all lib

it should be copied automatically

17:02

but if it's not, you could try copying it manually

yes i think (edited)

it should be in ddnet-libs, somewhere

kio

I put all the dlls by hand and the error disappeared, but when opening the game it crashes in the loading screen..

crash

Maybe it also failed to copy the data dir? That could appear like a crash too

do we crash without a datadir nowadays?

No, but it would immediately quit out no?

loading screen without texts or images, only a progess bar and after a crash

No text no images does smell like no datadir. Can you check if there is a folder called data next to the exe?

inside the same folder Debug?

Yes

17:06

If not copy the data folder from root aswell and see if it works

kio

@murpi do you understand and approve? otherwise tell me and I stop doing anything

We cannot control what people do with the code, and we have no authority to give approval for any modifications made by third parties. However, I encourage you to consider the impact of your actions on the community and the game as a whole. That includes modifications that could give players an unfair advantage in any gamemode.

Learath2

No, but it would immediately quit out no?

we used to not do that

Learath2

Yes

i will show u a weird thing

heinrich5991

we used to not do that

Might have regressed after setting vulkan as the default. The shaders are in data

murpi

We cannot control what people do with the code, and we have no authority to give approval for any modifications made by third parties. However, I encourage you to consider the impact of your actions on the community and the game as a whole. That includes modifications that could give players an unfair advantage in any gamemode.

okk, I will be more reserved with my things and will not show them

kio

i will show u a weird thing

@Learath2

17:08

I compared two projects, one is my old one that was working and the second is the current one that is not working

Yeah this is where my knowledge of visual studio comes to an end. I honestly haveny compiled on windows in ages

17:09

It is most definitely a mess created by the folder structure

kio

Click to see attachment 🖼️

i used this configurations for the old project

17:09

but i give problems now

Notice your dlls are also there

17:10

Maybe it has something to do with how you used cmake? Did you use the cmake gui or directly open the folder in vs?

seems like there's a bug lingering as well

17:10

data being copied to the wrong dir

yep

17:11

heinrich5991

seems like there's a bug lingering as well

Probably due to how cmake gui generates a subproject for each target

17:13

Or rather the visual studio target of cmake that is

17:13

A similar issue happens in the xcodeproject target

F7 do this command [main] Building folder: KeClient [build] Starting build [proc] Executing command: "C:\Program Files\CMake\bin\cmake.EXE" --build f:/KeClient/build --config Debug --target ALL_BUILD

Learath2

A similar issue happens in the xcodeproject target

xcode

cammostripes

That's not the interesting part

17:14

How you generated this visual studio project is what matters

17:15

If you just open the cloned folder with VS, if I recall correctly it's handled properly

should be

17:17

you can even clone using VS itself

you can look at the initial few seconds and the final few, but the problem could be the visual studio kit, because the one in the build looks like a VS project to me :/ https://www.veed.io/view/83fbb47d-4ad7-40e4-8d66-78cd7294b564?sharingWidget=true&panel=share (edited)

VEED - 2023-03-21 18-16-31.mp4

Make stunning videos with a single click. Cut, trim, crop, add subtitles and more. Online, no account needed. Try it now, free. VEED

Oh you are using vscode hmm

17:27

Mhhhh. Ok, let me try it myself :/

i love u

17:28

ehm

17:41

i did something

The objective & aim of chatting with me should be sex and relationship

bruh (edited)

You are right Laura but this isn't the place

LAURA

The objective & aim of chatting with me should be sex and relationship

we are developers, only cocks or computers (edited)

Yeah, vscode just generates the project in this weird way sadly

wait wait i look some green texts im horny

stop

:/

17:43

looks good wait

Image attachment

That looks like it's make or ninja not visual studio/nmake so what did you change?

i called the folder "gay" to punish the project (actually to change as to workspace so vscode doesn't recognize it as the same project), then i chose unidentified kit and re-added my old config files which weren't working before

17:46

this fules

1.94 KB

411 bytes

cmake-tools-kits.json

319 bytes

3.93 KB

17:46

files

It works with the msys generator btw

my goal by adding cmake-tools-kit was to find a setting with the path I put but nothing came out

Yeah you switched to mingw too

ok

17:46

Image attachment

@heinrich5991 the vs project generator just doesn't work, I'm not sure what we do about that, it's just the way targets are treated

kio

Click to see attachment 🖼️

So close, I guess the rust part failed above

20.6 KB

17:48

i wanna cry

17:49

3 hours only for compile the orginal game

Well if you have visual studio it will just work(tm) in there

17:53

Hm, it wasnt the rust part that failed it seems

you could try doing it in visual studio, that has at least been tried by project members before

Learath2

@heinrich5991 the vs project generator just doesn't work, I'm not sure what we do about that, it's just the way targets are treated

we can probably fix that

I should do this in a clean vm and make a tutorial sometime

first step would be creating a bug report though ^^

17:55

I have a tutorial for msvs

I meant for people that don't want vs

17:55

With msys2 and mingw

ah

17:55

yea, do that

Honestly iirc msvs just works when you clone within it

i don't rly like VS but im trying with that

17:57

it works

17:57

i don't want VS sdfijndfsjahinlk

I'll figure it out with mingw after dinner

1

bruh, where the fuck it take x, y, w, h

Image attachment

Learath2

I'll figure it out with mingw after dinner

oke

CUIRect is stateful

oi

18:13

hi

kio

bruh, where the fuck it take x, y, w, h

Those should technically be m_X, m_Y, m_W and m_H

hello

hi

Learath2

CUIRect is stateful

I don't know how to do anything anymore, they changed all lel methods

18:14

hi

kio

I don't know how to do anything anymore, they changed all lel methods

Hm, I think CUIRect is pretty much the same as it has always been

theozin

hello

Do you need something?

no the class render

I'm Brazilian and I'm using a translator cammo

cammo

Learath2

Hm, I think CUIRect is pretty much the same as it has always been

I think CUIRect wasn't able to draw on its own before

Voxel

Do you need something?

no

18:17

someone who plays

18:18

pls

This is where we talk about the develoupment of the game. If you're looking for someone to play with, feel free to request in #looking-for-group

Voxel

This is where we talk about the develoupment of the game. If you're looking for someone to play with, feel free to request in #looking-for-group

thanks

heinrich5991

I think CUIRect wasn't able to draw on its own before

Oh true, so there is some change there

timakro

Had to google it: x is "inspect memory", /i interprets it as instruction, $rip is instruction pointer (edited)

you can also show which accessed address caused the segfault

19:23

p $_siginfo._sifields._sigfault.si_addr

ah cool

most often will show (void *) 0x0 but you never know

@Chairn Uh nice, a null pointer. But I still don't know where it comes from

Could we get a new Statistic on the website and co that shows the oldest standing records?

timakro

@Chairn Uh nice, a null pointer. But I still don't know where it comes from

could it be a this pointer that is null when you try to call the function?

You mean the function argument?

(gdb) frame
#0  0x000055b195b811a1 in on_change_systray (obj=0x55b195bcc940 <systray>) at /home/tim/Downloads/tint2/src/systray/systraybar.c:298
298         GSList *list = g_slist_copy(systray.list_icons);
(gdb) print systray.list_icons
$3 = 0x55b19749d2a0 = {0x55b19736f250, 0x55b19730ab50, 0x55b197441ff0, 0x55b197370a40}

It's not null as you can see

vscode provides a nice gui for debugging with gdb

And we're not dereferencing anything anyways so...

can you give us the whole function?

20:32

perhaps it's not in that line

ohh this pointer, this is C not C++

this is C++, not C :p

yeye cuz Chairn said this that can't be the problem because I'm working with C

20:36

I'm pretty sure now that somehow the wrong debug symbols got used because I copied my modified binary to /bin so that probably somehow confused gdb. Then the line with the bug would be traywin = (TrayWindow *)l->data; (which makes sense, 0 deref) and maybe my fix worked after all but somehow the old binary is back

[ddnet/ddnet] Issue opened: #6450 Use Steam-API to determine preferred user language before checking user locale

As described in https://github.com/ddnet/ddnet/pull/6449#pullrequestreview-1349549157, also use https://partner.steamgames.com/doc/api/ISteamApps#GetCurrentGameLanguage to determine the configured game language in Steam. Steam uses their own language names (https://partner.steamgames.com/doc/store/localization/languages), so we need to add another entry to the language index for this purpose. Right now Steam does not support Serbian (which is available in ddnet both in Latin and Cyrillic s...

i'd like to quit vsode but the visual debugger things makes me stick around. pretty easy to spot segfaults and fix em with this

[ddnet:master] 5 new commits

413227a Add os_locale_str to get user locale - Robyt3 481698c Move CLanguage and LoadLanguageIndexfile to localization - Robyt3 cddbc78 Add RFC 3066 language tags to language index - Robyt3 a91b51c Select language on first start based on user locale - Robyt3 8057d59 Merge #6449 - bors[bot]

21:13

[ddnet/ddnet] Issue closed: #2459 Use system locale to set language on first execution

21:13

[ddnet/ddnet] Pull request closed: #6449 Select language on first start based on user locale

[ddnet/ddnet-data-svg] Pull request opened: #39 Jungle deathtiles remake

Not pixel perfect, but visually indistinguishable in game i think. !jd_compare_full !jd_compare_smoll The original mapres has a bug on the vertical 1-spike spikes, where they are 0.5 pixels off, making them not line up with the cross tiles. I don't know your policy on these situa...

I'm having a CORS issue accessing the player json data. Is this something that needs to be changed on my side or is it on the ddnet server-side? The following gives a CORS error: fetch("https://ddnet.org/players/?json2=UnlucksMcGee") Although this one does not: fetch("https://ddnet.org/releases/maps.json") (edited)

Image attachment

heinrich5991

that's an error that has to be fixed server-side

It previously seems to be something that was fixed on server-side

yes, needs to be fixed serverside

UnlucksMcGee

I'm having a CORS issue accessing the player json data. Is this something that needs to be changed on my side or is it on the ddnet server-side? The following gives a CORS error: fetch("https://ddnet.org/players/?json2=UnlucksMcGee") Although this one does not: fetch("https://ddnet.org/releases/maps.json") (edited)

open chrome with special conf (edited)

23:38

what is this empty class supposed to do? :/

Image attachment

it's supposed to do nothing

23:38

you can check the #ifdef above to see what it's supposed to do when discord support is compiled in

UnlucksMcGee

I'm having a CORS issue accessing the player json data. Is this something that needs to be changed on my side or is it on the ddnet server-side? The following gives a CORS error: fetch("https://ddnet.org/players/?json2=UnlucksMcGee") Although this one does not: fetch("https://ddnet.org/releases/maps.json") (edited)

works now?

Exported 293 message(s)