Pathos

Also I want to report that this isn't just me

Restarting the client fixes it only for some people

ChillerDragon

[ddnet/ddnet] Pull request opened: #9001 Change color of non standard vanilla gametypes

Brings DM+ closer to DM instead of using a dark green.

before

!before

after

!after

Checklist

[x] Tested the change ingame
[x] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test (especially base/) or added c...

9000!

<ChillerDragon> yes its over 9000!

Vappy

9000!

8,09958998668719085829131208009794E31681

where i can find firedelay

<ChillerDragon> i tunes

04:12

<ChillerDragon> rcon tune grenade_fire_delay 0

04:12

<ChillerDragon> see also tunes

04:12

<ChillerDragon> @iuwywuiywuiywiuwyuwyui

ws-client

<ChillerDragon> rcon tune grenade_fire_delay 0

tunes ok ,where i can find lines of this?

<ChillerDragon> wat

nvm

<ChillerDragon> @Patiga can there be no gamelayer?

04:30

<ChillerDragon> https://zillyhuhn.com/cs/.1726806627.png

Yo, anyone knows why there's no gmock.h header file :\ (I did steps described in Running tests (Debian/Ubuntu) part in readme)

MilkeeyCat

Yo, anyone knows why there's no gmock.h header file :\ (I did steps described in Running tests (Debian/Ubuntu) part in readme)

probably because you didn't install it?

06:45

Isn't this some lib? xd

i have google-mock package installed (edited)

Try the download gtest cmake flag. Check what the CI does

06:58

Also deleting the build directory might help

deleted it already 3 times xd

Which cmake command did you run?

cmake -DCMAKE_EXPORT_COMPILE_COMMANDS=1 -DDEV=ON .. && make run_tests

Try the GTEST one

with -DDOWNLOAD_GTEST=ON it works

07:05

thanks xd

github-merge-queue[bot]

[ddnet:master] 2 new commits

b443b40 Change color of non standard vanilla gametypes - ChillerDragon 5558a69 Merge pull request #9001 from ChillerDragon/pr_dm_plus_colors - def-

07:18

def-

[ddnet/ddnet] Pull request closed: #9001 Change color of non standard vanilla gametypes

@Ryozuki https://github.com/edg-l/ddstats/blob/86940cd45074ea440bee048f8551252bd08eb557/src/routes/events/%2Bpage.svelte#L42 The close code must be either 1000, or between 3000 and 4999 otherwise switching between pages breaks (i.e. going from the /Events page to any other page raises an InvalidAccessError)

ddstats/src/routes/events/+page.svelte at 86940cd45074ea440bee048f8...

A DDNet / teeworlds fancy information frontend. Contribute to edg-l/ddstats development by creating an account on GitHub.

:O

github-merge-queue[bot]

[ddnet:master] 2 new commits

582e9fa add support for vertical arrow key navigation in multi-column listboxes - dobrykafe b6f80e5 Merge pull request #8999 from dobrykafe/pr-listbox-qol - def-

07:44

def-

[ddnet/ddnet] Pull request closed: #8999 Add support for vertical arrow key navigation in multi-column listboxes

ws-client

<ChillerDragon> @Patiga can there be no gamelayer?

only when you create a map from scratch, you cant save a map without a gamelayer

<ChillerDragon> hm okay then the optional makes sense. Bit annoying for type checking if a map is loaded hmm..

08:12

<ChillerDragon> maybe it would be nicer if making a map from scratch creates a game layer for you? like the editor does.

ws-client

<ChillerDragon> maybe it would be nicer if making a map from scratch creates a game layer for you? like the editor does.

you should then pass everything for gamelayer to appear

08:24

current approach is not sane, but making default gamelayer for ::new(...) is kinda meh

ChillerDragon

[ddnet/ddnet] Issue opened: #9002 Editor: click drag navigation can catch quads and alter their position

https://github.com/user-attachments/assets/f41ae00f-3df8-474d-a7aa-d8a45036c9e8 Mouse based navigation is impossible without breaking your map if it looks like this: !image The group the quad is in needs para to be able to get moved under the cursor on move.

08:50

feels good

<ChillerDragon> where did the input field for envelope values go in the editor?

08:58

<ChillerDragon> and how to resize the envelopes field in general?

ws-client

<ChillerDragon> where did the input field for envelope values go in the editor?

Right click the envelope point

ws-client

<ChillerDragon> and how to resize the envelopes field in general?

Mouse wheel and Shift+mouse wheel

<ChillerDragon> woah segfault xd

09:01

<ChillerDragon> @Robyt3 that seems to zoom not make the view port bigger

09:02

<ChillerDragon> last time i used the editor clicking on "envelopes" multiple times would open it in 3 different sizes

ws-client

<ChillerDragon> @Robyt3 that seems to zoom not make the view port bigger

You can drag the envelope menu up to make it larger

<ChillerDragon> a

You can also drag the setting editor and layers

<ChillerDragon> i see nice

09:03

<ChillerDragon> thanks

Robyt3

[ddnet/ddnet] Pull request opened: #9003 Fix some editor quick actions not being clickable, small refactoring

Add separate UI element IDs for the quick action buttons in the prompt dialog, to ensure that the UI element IDs are different from the IDs used for the regular menu buttons. Otherwise, some buttons could not be activated with clicks because the menu button with the same ID was rendered first. Closes #8998.

Checklist

[X] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Wr...

09:08

ChillerDragon

[ddnet/ddnet] Issue opened: #9004 Editor: add shift+r to rotate backwards

"r" rotates but "shift+r" does not rotate backwards

n flips the tile, why shift+n doesnt

zhn

n flips the tile, why shift+n doesnt

In general I'd say it's better if hotkeys are only activated when no additional unnecessary modifiers are pressed, so we can more easily add new hotkeys in the future without breaking users' existing workflow

Robyt3

In general I'd say it's better if hotkeys are only activated when no additional unnecessary modifiers are pressed, so we can more easily add new hotkeys in the future without breaking users' existing workflow

yeah, and thats cool actually

09:20

i got used to these hotkeys tbh

2

2

10:03

not me but funny

github-merge-queue[bot]

[ddnet:master] 4 new commits

3002183 Fix some editor quick actions not being clickable - Robyt3 9597241 Add margin between quick action label and description - Robyt3 7b64152 Remove unnecessary Skip and DescColor variables - Robyt3 034e497 Merge pull request #9003 from Robyt3/Editor-Prompt-Fix - def-

10:16

def-

[ddnet/ddnet] Pull request closed: #9003 Fix some editor quick actions not being clickable, small refactoring

10:16

def-

[ddnet/ddnet] Issue closed: #8998 Some quick actions are not double clickable

Ryozuki

Click to see attachment 🖼️

This might not be as bad as it seems with context. Hetzner offers dedicated system configurations where a tech has to setup and make changes for you. So it's probably not a vps, maybe someone asked for extra hard drives and they forgot to plug it back in.

10:26

Basically a non event if that's the case since your downtime was planned and an unspecified duration so who cares. (edited)

go ddnet.

12:04

what the fxxkin'

12:05

nightly builds

棉裤

Click to see attachment 🖼️

Yes ? You're toggling it from 0 to 1 or 1 to 0 (NVM I didn't see the 100) (edited)

12:12

That's how a toggle works ^^

jxsl13

curse the person sending weird master server data: cannot unmarshal number into Go struct field Skin.servers.info.clients.skin.name of type string

the info data should be treated as free-form

Matodor

someone know what is MR prefix? map route?

move restriction

1

Ewan

I think it would be cool if ddnet servers could just fork themselves when voting (up until a defined maximum). add a new type of map vote that will move anyone who presses f3 to the new server, and a command to join the last detached vote if u missed it

https://github.com/ddnet/ddnet/issues/6754

Move players to new server on map vote · Issue #6754 · ddnet/ddnet

As discussed on irc @heinrich5991 suggested to move all players that voted yes on a map vote to a new server. This fixes the issue of players in the race being interrupted by map votes. And also so...

how can I find hammer's hitbox implementation in source?

KemeL

how can I find hammer's hitbox implementation in source?

meloƞ

what

Learath2

https://www.ietf.org/archive/id/draft-ietf-tls-esni-08.html ECH seems rather shit to me. It seems they only really considered the case where you are behind some protection cartel

ech is great. change my mind. you just put anything into the client outer, it doesn't matter

Faulty

And when approximately do we plan on increasing this size? I would be extremely grateful. I can do a separate PR on this if needed to make it work now, not when the 128 server support PR is accepted since it might not be anytime soon

maybe we could try reducing the size of the json a little. 128 bytes per player doesn't sound too little, but I guess json is verbose…

ws-client

<ChillerDragon> @Teero nah i got no powers nowhere :c

ChillerDragon: if you want to maintain 0.7, it might be possible. the problem is that we have literally no one who wants to do that currently

heinrich5991

maybe we could try reducing the size of the json a little. 128 bytes per player doesn't sound too little, but I guess json is verbose…

Are you suggesting to add another endpoint to the master server and shorten the names of the json parameters?

heinrich5991

ech is great. change my mind. you just put anything into the client outer, it doesn't matter

How does it not matter? How is the "client-facing server" supposed to handle the case of ech rejection otherwise?

meloƞ

Yes ? You're toggling it from 0 to 1 or 1 to 0 (NVM I didn't see the 100) (edited)

im doing 0 <-> 100, but it's 0 <-> 1

It would of course be nice to use grpc serialisation, but it's not really appropriate in our case

Robyt3

[ddnet/ddnet] Pull request opened: #9005 Fix toggle command always setting value to 1

Regression from #8971.

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
[ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddn...

GitHub

Click to see attachment 🖼️

nice

heinrich5991

[ddnet/ddnet] Issue closed: #8945 Extend antibot to be a generic plugin system

MilkeeyCat

@heinrich5991 is it possible to make https://github.com/heinrich5991/libtw2/tree/master/wireshark-dissector work with wireshark 4.0?

check out the older version that worked with wireshark 4.0

heinrich5991

check out the older version that worked with wireshark 4.0

i already compiled 4.2 xd

ChillerDragon: why would you be against quic?

jxsl13

doesn't like arm64

report a bug

github-merge-queue[bot]

[ddnet:master] 2 new commits

e1a21f5 Fix toggle command always setting value to 1 - Robyt3 b34edb6 Merge pull request #9005 from Robyt3/Client-Toggle-Fix - def-

13:30

def-

[ddnet/ddnet] Pull request closed: #9005 Fix toggle command always setting value to 1

heinrich5991

ChillerDragon: why would you be against quic?

is quic quick enough? did u see the reports from cloudflare?

13:35

ok not cloudflare

13:35

https://dl.acm.org/doi/10.1145/3589334.3645323

QUIC is not Quick Enough over Fast Internet | Proceedings of the AC...

that was about file downloads, probably doesn't matter for ddnet

13:35

IIRC

13:36

the diagrams show that current QUIC impls don't manage to saturate a data link if it has more than 500 Mb/s

13:37

I hope we don't have server-client connections with more than 1 Mb/s

Learath2

How does it not matter? How is the "client-facing server" supposed to handle the case of ech rejection otherwise?

I don't understand what you said. if the server gets a request from a client that doesn't do ECH, they'll include the normal SNI. if they don't, the server unpacks the actual SNI from the inner handshake, the outer one doesn't matter

heinrich5991

I don't understand what you said. if the server gets a request from a client that doesn't do ECH, they'll include the normal SNI. if they don't, the server unpacks the actual SNI from the inner handshake, the outer one doesn't matter

What if the server can not decrypt the inner hello? Then a connection is supposed to be established with the outer hello, through which the "client-facing server" needs to give out an ECHConfig to be used

13:50

6.3.2.1: When the server rejects ECH or ..., it continues with the handshake using the plaintext "server_name" extension instead. Clients that offer ECH then authenticate the connection with the public name, as follows:

The client MUST verify that the certificate is valid for ECHConfigContents.public_name. If invalid, it MUST abort the connection with the appropriate alert.

...

13:52

So I can't just put "anything" in there. I need to put a valid SNI for which I can produce a certificate as far as I understand. Now if you are behind cloudflare or akamai or whatever, that's great. You put cloudflare.com in there, cloudflare.com provides the ECHConfig great. What am I supposed to do with my topology? I'm forced to expose the one domain that I own in plaintext

Ryozuki

Click to see attachment 🖼️

lmaooo

Unless I grossly misunderstand something. ESNI definitely had the better idea when they used DNS for the keys. ECH's approach seems to be designed around funneling people to using some "client-facing server" provider

13:56

Or making no assumptions about intent, at the very least completely designed around that usecase

@heinrich5991

Issue 1. When downloading the same file, the in-kernel UDP stack issues much more packet reads (netif_receive_skb) than TCP, leading to a significantly higher CPU usage. This is because none of the QUIC implementations we examine uses UDP generic receive offload (GRO) where the link layer module combines multiple received UDP datagrams into a mega datagram before passing it to the transport layer. This is in sharp contrast to the wide deployment of TCP segmentation offload, and recent advocacy of UDP send-side offload (GSO). • Issue 2. In the user space, QUIC incurs a higher overhead when processing received packets and generating responses. This overhead can be attributed to multiple factors: the excessive packets passed from the kernel (Issue 1), the user-space nature of QUIC ACKs, and the lack of certain optimizations such as delayed ACK in QUIC.

13:59

it looks like some issues are also bad implementations

13:59

or not optimized

heinrich5991

report a bug

It's more of a wireshark from brew does seem to be amd64 and my problem was cross compilation from arm64 to amd64

https://loco.rs/

Loco.rs - Rust Web Framework

Loco.rs is like Ruby on Rails for Rust. Use it to quickly build and deploy Rust based apps from zero to production.