DDraceNetwork - developer

I just had an idea of how to sneak past the account system :O

9:12AM

We shall make commits where new clients have features/functionality that weren't before to complete new maps unbeatable from old clients

9:13AM

So this will be the start of ddnet2.0

9:13AM

If you wanna stick to old client you can but you will not be able to complete 2.0 maps

9:14AM

and then we will sneak the account system in.

9:14AM

If you wanna play on old maps you can but the new season is here and you can choose what you want

to play ddnet 2.0 maps you would need ddnet battle pass as well?

nah

9:24AM

you just would be able to complete them ig

9:24AM

like some maps could require 3 dummies instead of 1

what was the latest version with stars when tee in freeze?

which is not possible currently

Mr.Gh0s7

like some maps could require 3 dummies instead of 1

just launch 2 clients santatrollet

9:25AM

thats how ive finished quantum

9:25AM

or whatever the map's name

MilkeeyCat

just launch 2 clients santatrollet

You don't have dummy copy though or toggle

will ddnet 2.0 have more stuff for dummy players pepeW

i aint updating

This was just an example

if they remove dummies

9:27AM

then it will be fastest git pull in my life

I guess what I wanted to say is break compatibility to make bug fixes. Old ppl can still play on ddnet1.0 but the new ddnet will have new networking, new compatibility for mods, new options etc. that weren't possible to add before.

MilkeeyCat

what was the latest version with stars when tee in freeze?

I would say before 16.5

9:30AM

(for the server)

Mr.Gh0s7

I guess what I wanted to say is break compatibility to make bug fixes. Old ppl can still play on ddnet1.0 but the new ddnet will have new networking, new compatibility for mods, new options etc. that weren't possible to add before.

Also I am not saying since we could add accounts we could add this and that because those features would require accounts. I am saying that one of the bugfixes would be accounts whatever that entails :) -- sincerely ghosty

4

:O

Powered by https://github.com/TsFreddie/Oolong

GitHub - TsFreddie/Oolong: A TypeScript framework based on PuerTS a...

A TypeScript framework based on PuerTS and MithrilJS - GitHub - TsFreddie/Oolong: A TypeScript framework based on PuerTS and MithrilJS

2

10:19AM

haven't really written any documentations for it tho

TsFreddie

Powered by https://github.com/TsFreddie/Oolong

it looks so pogg

how does it work though does it have a master window that's transparent or does the window actually change size?

you can write ts in unity? pepeW

i can now

Mr.Gh0s7

how does it work though does it have a master window that's transparent or does the window actually change size?

the bottom bar just changes size

10:24AM

basically a react-like framework with hacky DOM implementation in Unity

seems like the saving in my editor broke a little, and was saving it to some .tmp file

bug? when i set the value to 0.2, it starts dropping down to 0.190ish

floating point shenanigans

cyberFighter

bug? when i set the value to 0.2, it starts dropping down to 0.190ish

yes that is def a bug. it should be at most 0.001 off, depending on the implementation

ddnet often does not round to nearest integer, but simply converts to int that could be a problem maybe

The lineinput uses float, whereas the value in the map file is fixed-point

Robyt3

The lineinput uses float, whereas the value in the map file is fixed-point

but even for fixed point you can round up and down

Also, envelope point value is fixed point with scale 1024 whereas time uses scale 1000

someone experience with being banned because of port scan attacks?

2:29PM

my VPS was banned. so either some software has a vulnerable, my SSH key was hacked or it's false positive

2:29PM

apparently my server sent it from port 80

2:30PM

so probably apache

how come shift + left click (filling selection with blocks) doesnt work with wasd (moving around in editor)

e.g. ger2.ddnet.org https://www.abuseipdb.com/check/176.9.114.238 also has 34% report rate, and many of these reports also claim port scanning

2:33PM

so i kinda assume it's just a dos

Robyt3

Also, envelope point value is fixed point with scale 1024 whereas time uses scale 1000

mh int(0.2 * 1024) / 1024 = 0.19921875 vs nearest_int(0.2 * 1024) / 1024 = 0.200195312 5 could explain it xdd

@heinrich5991 did u know about this https://doc.rust-lang.org/nightly/core/panic/struct.Location.html#method.caller

Location in core::panic - Rust

A struct containing information about the location of a panic.

3:38PM

i discovered this can be nice if u have asserts for tests in another function u call inside the test

3:39PM

 #[track_caller]
    pub fn run_program_assert_output(
        program: &(String, Program),
        entry_point: &str,
        args: &[JITValue],
        outputs: &[JITValue],
    ) {
        let result = run_program(program, entry_point, args);
        assert_eq!(result.return_values.as_slice(), outputs, "assert mismatch on: {}", std::panic::Location::caller());
    }

3:39PM

xd

3:40PM

my FFR tip

3:40PM

my business is done

isnt that default behavior for panics anyway?

oh and also the crate this friday is https://docs.rs/educe/0.4.23/educe/

educe - Rust

Educe

3:40PM

its nice for partialeq impls

Jupstar ✪

isnt that default behavior for panics anyway?

i think not if the panic is inside a function but u care about the one who called that one

3:41PM

the assert here would point me to this function, not where this function was called

ah u mean like a backtrace?

3:41PM

panic also has a backtrace function afair

Ryozuki

@heinrich5991 did u know about this https://doc.rust-lang.org/nightly/core/panic/struct.Location.html#method.caller

didn't know it

Approved in RFC 2091, this feature enables the accurate reporting of caller location during panics initiated from functions like Option::unwrap, Result::expect, and Index::index. This feature adds the #[track_caller] attribute for functions, the caller_location intrinsic, and the stabilization-friendly core: panic

:Location::caller wrapper.

3:42PM

@Jupstar ✪ oh this very feature allows that on unwraps

3:42PM

https://rustc-dev-guide.rust-lang.org/backend/implicit-caller-location.html

Implicit Caller Location - Rust Compiler Development Guide

A guide to developing the Rust compiler (rustc)

for me panics already report that

its already in stable btw

3:42PM

yeah

3:42PM

it was implemented some time ago

3:42PM

3:43PM

@Jupstar ✪ this was how it wAS BEFORE

3:43PM

sry caps

Ryozuki

@Jupstar ✪ this was how it wAS BEFORE

i never used dark ages rust

3:43PM

i think a panic on a fn with track caller will show the caller location

3:44PM

instead of the panic inside the fn

3:44PM

time to check rustp layground

i'd be useful if errors could do the same

3:45PM

i hate to do .map_err(my text that explains everything)

https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=02da3e737b0d37e8157f2e8fe32a739f

Rust Playground

A browser interface to the Rust compiler to experiment with the language

3:45PM

it does that!

3:45PM

ok

3:45PM

nvm

3:45PM

im blind

3:45PM

it doesnt

3:46PM

kek

rustup update version-2050

https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=087d6bae5e916cc9bf70997e474f579e

Rust Playground

A browser interface to the Rust compiler to experiment with the language

3:46PM

omg

3:46PM

i had it on the wrong fn

3:46PM

it works xDDD

3:46PM

it reports the panic at line 10

Ryozuki

oh and also the crate this friday is https://docs.rs/educe/0.4.23/educe/

why is it called educe?

thats my point

3:47PM

@Jupstar ✪

heinrich5991

why is it called educe?

the author is chinese

3:47PM

idk

3:47PM

https://magiclen.org/educe

3:48PM

oh so theorically

3:48PM

i dont even need the format string on the assert

3:48PM

cuz the assert itself will show up where the fn is called

3:48PM

marvelous

cool, now do the same for anyhow! calls

3:50PM

or tell me how to do it if it already exists

@Jupstar ✪ what does anyhow!

it doesn't really help me if a higher level component panics xd

just put #[track_caller]

anyhow! creates a anyhow Error

Jupstar ✪

mh int(0.2 * 1024) / 1024 = 0.19921875 vs nearest_int(0.2 * 1024) / 1024 = 0.200195312 5 could explain it xdd

yep, that fixes it

Ryozuki

just put #[track_caller]

it's really cool, but it would be nice for errors too, not only for panic u know

Jupstar ✪

it's really cool, but it would be nice for errors too, not only for panic u know

for errors hmm

3:51PM

i think there is backtrace support

how often do u call panic in your programs xD u probably do anyhow::Result

3:51PM

xdd

yeah

3:52PM

its all here https://doc.rust-lang.org/std/backtrace/index.html

std::backtrace - Rust

Support for capturing a stack backtrace of an OS thread

mh nice, didnt know this works for errors directly

available since 1.65.0

but i assume only with debug symbols?

@Jupstar ✪ u use nightly right

also doesnt work on WASM

do u use thiserror

:/

Ryozuki

do u use thiserror

sometimes

3:54PM

if i need to have an error as type

#[derive(Error)]
pub struct Error {
    pub backtrace: Backtrace,
    pub source: ErrorImpl,
}

3:54PM

and ErrorImpl can be ur enum

3:55PM

#[derive(Error, Debug)]
pub enum MyError {
    Io {
        #[from]
        source: io::Error,
        backtrace: Backtrace,
    },
}

3:55PM

sadly thiserror needs nightly for this

well and is the question if it works on WASM for example

3:55PM

and without debug symbols

https://docs.rs/anyhow/latest/anyhow/struct.Error.html#method.backtrace

Error in anyhow - Rust

The Error type, a wrapper around a dynamic error type.

maybe i should write my own macro

3:55PM

i am macro pro now

xd

cyberFighter

bug? when i set the value to 0.2, it starts dropping down to 0.190ish

why does that value even use 0-1.000 instead of 0-255 (edited)

i think it's simply because we use it for position etc too

4:10PM

it's not type aware

@Jupstar ✪ Would the color pipette from upstream work with all graphics backends?

does it get the color form the backend?

not sure how it does it, but it must get the hovered color from the engine I suppose

i mean if u take a screenshot, u get the colors u see

4:15PM

i'd maybe remove the alpha value tho

4:15PM

(the screenshots should already do that anyway)

ah, looks like upstream literally takes a screenshot of only the hovered area

4:15PM

https://github.com/teeworlds/teeworlds/pull/1434/files

Color picking for editor by msiglreith · Pull Request #1434 · teewo...

Added some color picking features for the editor: eyedropper/pipette and HSV color picker (requested by @Sonix-). Pipette is implemented by reading pixels from the current backbuffer, which involve...

I mean taking a screenshot should work. I don't think u be to reduce it to this area only in the backend already

4:20PM

Does it do it per frame btw

4:20PM

Or (edited)

4:20PM

Cache

Looks like it does it every frame justatest

4:21PM

Weirdly hooks into the screenshot command to read 1x1 pixel

I guess we could return screenshot data from graphics class. I dunno if it's nice that graphics class writes files anyway xd

Would be cleaner to add a separate command that reads only one desired pixel

4:24PM

And only call when the mouse position changed at least

4:24PM

Although that can also lead to issues if menus are opened with keys

Robyt3

[ddnet/ddnet] Issue opened: #7430 Add color pipette to editor

See https://github.com/teeworlds/teeworlds/pull/1434. Should be a separate graphics command to read one pixel instead of hooking into the screenshot command though. The picked color could be copied to clipboard so it can be applied to other color properties with shift+left click. Or we should consider adding a small palette of recently used colors for quick access.

4:31PM

Robyt3

[ddnet/ddnet] Pull request opened: #7431 Fix inaccurate envelope point value rounding

Round to nearest integer instead of truncating in f2fx to ensure correct round-trip with fx2f.

Checklist

[X] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test (especially base/) or added coverage to integration test
[ ] Considered possible null pointers and out of bounds array indexing
[ ] Changed no physics that affect existing maps
[ ] Teste...

GitHub

Click to see attachment 🖼️

Is that function not also used for physics?

4:35PM

Should be careful if so xd

It's only used in editor and tools, it converts float to fixed point. Loading maps should only use the opposite function

ah then fine

4:41PM

right physics use an ever lower value xD

4:41PM

256 or smth

is there a bind to start the local server?

5:33PM

*command

./DDNet-Server

5:35PM

xd

5:35PM

i don't think there is a console command

uff ok xd

github-merge-queue[bot]

[ddnet:master] 2 new commits

859727c Fix inaccurate envelope point value rounding - Robyt3 de52ded Merge pull request #7431 from Robyt3/Editor-Envelope-Accuracy - heinrich5991

6:55PM

heinrich5991

[ddnet/ddnet] Pull request closed: #7431 Fix inaccurate envelope point value rounding

thx

7:27PM

are those changes immediately put into the steam nightly tho? i dont really understand how it works

nightly should be build every night, hence the name

7:32PM

and nightly contains all changes from the master branch

@cyberFighter

hi

who go playing

i need someone who knows about the game

8:31PM

someone i can ask questions to

u could start with asking the question

Jupstar ✪

u could start with asking the question

how can i ban other clients expect steam

mh there is some checksum sent to the server, with which u could identify these clients i think

8:35PM

better ask @heinrich5991

8:35PM

but he will probably ask you "why do you want to ban all other clients"

It's something you'll have to figure out on your own

Jupstar ✪

but he will probably ask you "why do you want to ban all other clients"

because i want the ban bot clients someone spoofing

8:37PM

versions

or are you the bot client and want to join other servers?

8:38PM

what is your server? the spoofing might not even stop by that

8:38PM

and it's probably no spoofing

8:38PM

it's simply many connections

Jupstar ✪

what is your server? the spoofing might not even stop by that

ziprnet

banning based on client IDs is mostly futile

8:57PM

try to ban them based on their behavior

https://starcel.net/

Starcel

Unbounded Thinking

Link without context is a censorable offense

Replying to @TsFreddie https://starcel.net/

10:53PM

:ban:

no context really

oh damn the demo menu is changed

11:22PM

i like it

Robyt3

[ddnet/ddnet] Pull request opened: #7432 Add CLock wrapper for std::mutex to replace lock_* functions, use CLock instead of std::mutex, add thread-safety annotations

Replace usages of platform specific lock_* functions with std::mutex through the wrapper class CLock. Move lock classes to base/lock.h. The CLock wrapper class is only necessary because the clang thread-safety attributes are not available for std::mutex except when explicitly using libc++. Use CLock and CLockScope instead of std::mutex and add clang thread-safety analysis annotations everywhere except for usages in engine graphics and video, as those usages also involv...

TIL

clock