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 2021-08-01 00:00:00Z and 2021-08-02 00:00:00Z

have you changed something to the servers? i get ping spikes suddenly

idk if i should ask this question here but is there a way to manually call -fire? how does +fire work anyways

Nope. +commands are special in that they get called on both strokes

as in, +fire is called when m1down and m1up? or there's an internal -fire called when m1up

Deleted User

idk if i should ask this question here but is there a way to manually call -fire? how does +fire work anyways

https://wiki.ddnet.tw/wiki/Binds#Toggles

DDNet client has a multitude of commands and settings that can be bound to nearly any key or mouse button. You can enter commands into the console, which by default is accessed by pressing F1.

07:51

greenthing

Ryozuki

@Patiga maybe we could do something to promote editing the wiki

Maybe we could create a channel for discussions about the wiki again and encourage to ask there if in doubt.

yeah

07:58

@murpi your decision

07:58

or any other admin

we could enable threads for this, if this is something you'd like

i dont think threads help with this, we just need a wiki channel

08:00

it will make more ppl aware it exists and maybe even make ppl contribute, and if they have doubts how to do x thing we can help there

08:01

( how to do x in the wiki*)

how can a language without tenary operator be so popular...

09:16

i mean bash also has none but its not really missing there

if u talk about rust its fine tbh

09:26

and u can make a macro if u rly want it

09:26

macro_rules! either {
    ($test:expr => $true_expr:expr; $false_expr:expr) => {
        if $test {
            $true_expr
        }
        else {
            $false_expr
        }
    }
}

09:28

or https://doc.rust-lang.org/rust-by-example/flow_control/match.html

match - Rust By Example

Rust by Example (RBE) is a collection of runnable examples that illustrate various Rust concepts and standard libraries.

09:28

in fact

09:28

match is way better than a ternary operator specially if u have would chain them

hahaha ofc the rust fanboi jumps in

09:39

rust is pain in the ass to read some tenary aint gonna save it also except the few hardcore fans rus isnt tooo popular.. i was talking about python

lmao

10:10

its not a pain in the ass

10:10

bash is a pain in the ass to read

10:11

but in python u can do if x else y in one line

10:11

a = b if cond else c (edited)

yea thats no tenary

its the same

10:28

ur just obstinated

dafuq is obstinated

10:29

u use some crazy words bro

10:29

are you the new learath?

Image attachment

ty

i didnt know it either

fancy

greenthing

10:39

i know it cuz in spanish its more common

10:39

and found out it exists in english

cheat

10:41

imagine me using words like wunderkind, wanderlust or hamster

hamster?

whats in english again

10:43

the smol mouse

10:43

Embedded image

10:43

thats a hamster

ye

Deleted User

idk if i should ask this question here but is there a way to manually call -fire? how does +fire work anyways

Internally if you register a command starting with + it will get the stroke direction as it's first argument, and it'll be called on both strokes

[quakenet] ChillerDragon

how can a language without tenary operator be so popular...

It's so, soooo, soooo annoying and it's imo just rust people trying to distance themselves from C

10:53

Same with the pre/post-in/de crement operators

are you using the discord reply feature? :(

10:53

when do we get a bridge with reply support

Learath2

Same with the pre/post-in/de crement operators

actually there arearguments that pre post operators are bad

10:54

btw he is talking about python

Ryozuki

actually there arearguments that pre post operators are bad

ofc there are, computer science people can shit out 50 arguments against any feature half useful

Image attachment

10:55

https://github.com/dtolnay/rust-faq#why-doesnt-rust-have-increment-and-decrement-operators

ofc the most common of them, people are stupid

poggers

Image attachment

people are stupid so we have decided to reduce our language to english which will be evaluated through gpt3 to generate the bytecode

poggers

Image attachment

10:58

im in favour of this not being in rust

ofc you are, you are the prime advocate of people are stupid camp

10:59

I would kill for assignments to be expressions too but that is also too complicated for the new breed of developer we all seem to want to encourage

even a master C developer will eventually have a bug due to this too

10:59

im sure

10:59

and rust main selling point is not shooting urself in the foot

Ryozuki

even a master C developer will eventually have a bug due to this too

There is no way a MASTER C developer misses a sequence point after however many years of development

there is

ofc there is because you'd rather treat everyone as babies

11:01

if my grandma can't code with her 0 years of experience it's a bad language

i treat everyone as being able to make mistakes

11:01

and even at the most professional level people do mistakes

11:01

not just in programming

The compiler catches it even, you are being a goof

Learath2

if my grandma can't code with her 0 years of experience it's a bad language

if this is the argument level of this convo then we can stop

Your argument is that everyone can make mistakes thus a MASTER C developer with his 10 years or so of experience will miss the most basic sequence point mistake ever we can indeed stop, you are arguing in bad faith

no im not

s/^/if /

11:06

Besides the operators themselves don't have to require UB, you can if you hate optimization just put sequence points after each use of the operators

11:06

There you go, baby-proof version of the operator

yes rust devs are babies and c are chads

11:06

have fun

or just include one of them, with the post version, even baby-proofer version of them

Ryozuki

yes rust devs are babies and c are chads

you call me out on my shitty argument and you boil mine down to a meme, good job

u keep calling baby tyhis baby that its hard to take it serious

So sorry my words offend you, I'll use alternative terminology if it will make you actuall engage in the argument

can i turn off werror when running cmake/make ?

Just including the post-crement and post-decrement operators, and inserting a sequence point after them makes it so that you can't "shoot urself in the foot"

11:10

@ChillerDragon Why? You can always edit CMakeLists if you have to, but why not fix the warning or suppress it properly if it's a false-positive?

I don't think we should support i++. If we do, it would need to be an expression that returns the pre-incremented value, like it does in C, and that would therefore only be usable with implicitly-copyable types. But x += 1 can work with non-implicitly-copyable types, and I think it's a bad idea to have cases where x += 1 works but x++ doesn't.

fixing the warnings adds unneccesary additional complexity imo

https://github.com/rust-lang/rust/issues/12111

cant i pass something to cmake?

Why is that a bad idea? I don't see anything implicitly wrong with that

[ddnet/ddnet] Pull request opened: #3979 Add script to strip out graphic code

Add a script to strip out the client graphic code. Which creates a client that can be run in a non graphical environment. Can be used by modders who want a headless client for things like console chat or similar. Also useful to have a client that is almost the same as the normal ddnet client but can run in the CI. Should pick up obvious crashes of the client on launch that are unrelated to graphic code. Also can be used for further CI testing where a headless client connects to the server an...

chillerdragon: can you give an example of a warning you don't want to fix?

already solved see the passing pipeline :)

15:28

@heinrich5991:the gfx strip comments out almost all graphic function bodys and thus generates some unused warnings of static methods and global variables see https://github.com/ddnet/ddnet/runs/3213592364#step:10:435 but cmake -E env CXXFLAGS='' works like a charm

Build software better, together

GitHub is where people build software. More than 65 million people use GitHub to discover, fork, and contribute to over 200 million projects.

@chillerdragon maybe just suppress those that are fine so your stripped build is also checked

@Learath2: oky lemme try

16:26

ok looks like CXXFLAGS='-Werror -Wno-error=unused-function -Wno-error=unused-variable' does the job

the path should be ok, when i press wind+r and past the same path that open the good map

Image attachment

Idk if that has been asked already but how come there is no toggle-tile for doors? Seems so intuitive to have. Making slightly complex door parts is a mess...

I think its something like it has always been like that. Who knows who invented those doors back in the middle ages. Was it greyfox while he was drunk? maybe..

16:33

i didnt map something door related in years but i can not remember missing something. Can't u achieve a toggle using 2 tiles?

ofc but in some scenarios a toggle is so much cleaner.

Learath2

Internally if you register a command starting with + it will get the stroke direction as it's first argument, and it'll be called on both strokes

"Internally if you register a command starting with + it will get the stroke direction as it's first argument, and it'll be called on both strokes" Maybe my cs knowledge is insufficient to understand this but, how can you call a +fire twice without calling -fire first? +fire means to "fire, and then keep holding it" right? I previously mentioned a problem which can be replicated like this: Define 2 modes: 1. Half deepfly - "bind a bind mouse1 '+toggle cl_dummy_hammer 1 0'" 2. Deepfly - "bind b bind mouse1 '+fire; +toggle cl_dummy_hammer 1 0'" if you connect your dummy, and press the keys in the following order: b m1down a m1up m1down m1up b m1down (<- supposed to be +fire and +toggle cl_dummy_hammer 1 0, but try it, your main wouldn't actually +fire) The problem is when you switch to Half deepfly mode, the "+fire" has been removed from mouse1, so I assume the second stroke wouldn't be registered, so your main is stuck in the +fire status

17:37

so during the last transition (...b m1down) you need to click twice (m1down m1up m1down) for your main to actually hammer once

17:37

is there any way to solve this problem?

@Deleted User ok so you can pretend the stroke direction is an extra argument. So when you press +fire. There is a sort of internal "fire" command that gets called as "fire 1". And when you let go "fire 0"

ah that makes more sense, thanks!

Deleted User

is there any way to solve this problem?

No idea

What happens when you put fire instead of +fire?

19:11

Never tried it

It won't work. That's why I said "sort of". In reality "fire" is just not exposed to the console

Oh k

i KnOw HoW tO oPtImIzE current CCD

23:45

and you guys too i guess

Exported 122 message(s)