DDraceNetwork - developer

louis

a solid block is 60x60 relative to a tee

I also don't get this ratio

02:13

Character's size is 28x28 in a 32x32 subpixel environment (edited)

02:14

I think it's 28, having 2 pixels free space inside a 1 tile hole on both sides

i think character's size is 29x29

02:26

in regards to gametiles

02:26

in regards to other tees, it's probably different

02:27

02:27

that's a drawing of the 60x60 hitbox (a visualization of the wiki explanation) (edited)

Is there any confirmation of the tee's size? There was on the wiki I believe

ddnet-maps

[ddnet-maps:master] 1 new commit

abc23fd A Electric - ddnet-maps

Cellegen

Is there any confirmation of the tee's size? There was on the wiki I believe

look up my links from github

louis

Click to see attachment 🖼️

not 30 but 32 btw

louis

i think character's size is 29x29

28x28

gerdoe

not 30 but 32 btw

since a full tile is 32, and a tee is smaller than a tile, it can't really be 32

gerdoe

28x28

(without looking at code) iirc within a tile, a tee has 3 positions: left, middle and right. that would suggest 29. however, that would also mean that the distance is 30.5. with your 28, the 30 would sound correct

10:43

hm okay too much speculation. left, middle right would actually suggest 30 so nvm happy

Patiga

since a full tile is 32, and a tee is smaller than a tile, it can't really be 32

bruh i have eye cancer it seems

11:01

28/2 + 32/2 = 30

11:01

that's it

Robyt3

[ddnet/ddnet] Issue opened: #6346 Discrepancy of text line height depending on mapped screen size and font size

With the current font size 16 used for the ingame server info MOTD, the line height is calculated inaccurately: !textrender-lineheight-16 There is an empty space at the bottom due to the text not being rendered at exact positions aligned with the text lines. The line height is always identical to the font size. If you change the font size to 10, then this space disappears, as the ...

Robyt3

[ddnet/ddnet] Pull request opened: #6347 Minor refactoring of base `mem_*` functions

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/ddnet/ddnet/#using-addres...

huh, is there any use of mem_has_null

12:21

usecase*

could yall pls send me some screenshots of how this web page looks for you? https://zillyhuhn.com/tmp/

@deen How to open your own server

12:26

thats how it looks like for me ._.

12:26

is my font broken watafak is this "fl" bs

ima_2fac89e.png

12:27

On my phone it looks good

same

same?

phone thing

so it looks good for you? idk who is responsible for that maybe my system font or browser idk

12:30

whats ur os/browser?

12:31

Ah nice i think i got the fix font-feature-settings:"liga" 0;

oh yeah I didn’t realize you were trying to fix the ligatures

12:43

I wouldn’t have expected them to exist in a monospace font

12:45

where are you talking from right now

12:45

an IRC or matrix bridge?

both :D

1

gerdoe

huh, is there any use of mem_has_null

It's used in io_read_all_str to ensure that a string does not contain any null bytes except the null termination

i quickly switched to matrix to send a screen from ma phone

12:47

yea im surprised too that they exist there @Ewan it took me quite some time to even realize what kept messing up my indentation/spacing

12:48

i kept fixing it on my phone which broke it on my desktop xd

yeah and why on earth would it happen in the first place if just f and l next to eachother

12:48

maybe if it was entered as its own character, sure

12:49

but nothing just automatically ligatures like that

12:49

that’s why we have kerning

12:49

seems like a bug or localization thing

ChillerDragon: your terminal map editor is looking good :p

?xd

Patiga

(without looking at code) iirc within a tile, a tee has 3 positions: left, middle and right. that would suggest 29. however, that would also mean that the distance is 30.5. with your 28, the 30 would sound correct

isn't it 4 positions

13:51

4 position suggests 29

@Jupstar ✪ https://github.com/Overv/vramfs

GitHub - Overv/vramfs: VRAM based file system for Linux

VRAM based file system for Linux. Contribute to Overv/vramfs development by creating an account on GitHub.

14:01

we need to put ddnet in vram

louis

isn't it 4 positions

you are right, there are 4 positions: .44 .47 .50 and .53

Ryozuki

we need to put ddnet in vram

ez. does it already support that the GPU loads from disk directly? tho wouldnt really help us, since we have to decode pretty much all files anyway

xd

bors[bot]

[ddnet:master] 5 new commits

cb6de6a Group and order mem_* function implementations - Robyt3 d4809fa Change mem_has_null return type to bool - Robyt3 18cfbb5 Use size_t for mem_* function size parameters - Robyt3 ec7f556 Fix possible integer overflows using mem_* functions - Robyt3 9532e23 Merge #6347 - bors[bot]

15:27

bors[bot]

[ddnet/ddnet] Pull request closed: #6347 Minor refactoring of base `mem_*` functions

is there a player named Asko_YT ? i got a spam email lol, i wonder if there is a player with a similar name who was abused for that

Hello friends! It’s me again… I have a question. I made a bad implementation of a startup flag for the ddnet client a month ago or so. Now I had the idea to make all kinds of flags like (better to say command line argument) like for joining a specific server, thus joining the server. That could start ddnet kind of “headless”. Like ./ddnet —connect <server_ip>. And finally I could achieve my dream of dreams to make a “ddnet launcher” that is an electron app with a server list and it would use the client with the —connect flag to join the server. So you could make a really fancy ddnet ui with css and js and after all you would have +1 electron app on your PC

1

cauldron

Hello friends! It’s me again… I have a question. I made a bad implementation of a startup flag for the ddnet client a month ago or so. Now I had the idea to make all kinds of flags like (better to say command line argument) like for joining a specific server, thus joining the server. That could start ddnet kind of “headless”. Like ./ddnet —connect <server_ip>. And finally I could achieve my dream of dreams to make a “ddnet launcher” that is an electron app with a server list and it would use the client with the —connect flag to join the server. So you could make a really fancy ddnet ui with css and js and after all you would have +1 electron app on your PC

You can already pass any console command on the command line. The client also support URLs (ddnet ddnet://localhost:8303), map files and demo files as command line arguments.

Robyt3

Seems a bit hacky how it uses different flags to trigger the UI code. You could add a separate command render_demo <filename> instead of adding a hard-coded command line parameter. The command would start the demo player with the given demo, render it and quit. Then you'd use an external script to launch the client with this argument multiple times to render an entire folder.

This is from last month… so I once made a “app shortcut” for a mobile app with a big code base. It used javafx I think for asynchronous stuff and I just could “emit” an event or something like that when my “shortcut” or “3d touch” or whatever was activated. And then there were subscribers. But how is it in C++. Is there a framework for asynchronous programming? And how does the init process and lifecycle of the game work? I mean especially the UI. Not the game tick and everything ingame.

Robyt3

You can already pass any console command on the command line. The client also support URLs (ddnet ddnet://localhost:8303), map files and demo files as command line arguments.

I saw that in the code (that you can pass f1 commands). But I didn’t think of a f1 command for joining a server. I always try to reinvent the wheel

cauldron

I saw that in the code (that you can pass f1 commands). But I didn’t think of a f1 command for joining a server. I always try to reinvent the wheel

ddnet "connect localhost:8303" also works

cauldron

This is from last month… so I once made a “app shortcut” for a mobile app with a big code base. It used javafx I think for asynchronous stuff and I just could “emit” an event or something like that when my “shortcut” or “3d touch” or whatever was activated. And then there were subscribers. But how is it in C++. Is there a framework for asynchronous programming? And how does the init process and lifecycle of the game work? I mean especially the UI. Not the game tick and everything ingame.

There is no way to get information about events from the running client except the console output

Is there a f1 command to render a demo?

18:53

I found demo_play

There are start_video/stop_video but they don't work on client launch. play demos/demoname.demo can be used to play a demo, but there is no command to render a demo from the command line. Not sure what demo_play does and why we have demo_play and play.

18:55

Looking at the code, demo_play pauses/unpauses the currently playing demo

Robyt3

There is no way to get information about events from the running client except the console output

What is an event here?

Whereas play starts the demo playback from a file

cauldron

What is an event here?

Basically whatever console output there is

18:56

For example chat: *** 'Robyt3' entered and joined the game when you join a server

18:57

It's not really designed with general purpose events that could be used to control the client externally

I was once working on a mobile app and you could just emit something. It was a kind of global thing. And there were subscribers for the event you’d emit. And then they would execute their code. Is there something like this in the ddnet client code? (edited)

You can send command to the server/client with a FIFO file on unix or a Named Pipe on Windows

18:59

But the only way to get information from the server/client is to parse the console output (or read the memory, which would be very painful)

I mean when I was working with the mobile app you would emit the event inside the code and somewhere else in the code subscribers would trigger.

You mean entirely inside the ddnet code? That would be a callback.

I mixed it up, it’s not javafx, thats for UI. I ment RxJava

19:02

https://github.com/ReactiveX/RxJava

GitHub - ReactiveX/RxJava: RxJava – Reactive Extensions for the JVM...

RxJava – Reactive Extensions for the JVM – a library for composing asynchronous and event-based programs using observable sequences for the Java VM. - GitHub - ReactiveX/RxJava: RxJava – Reactive E...

There is no general purpose event system in ddnet where you can send arbitrary event objects to arbitrary subscribers

19:03

There are only various callbacks for specific things

19:04

For example a callback that gets invoked when the config file gets saved

19:05

You can chain callbacks for all config variables, so you can do additional things when a specific variable changes

So starting ddnet with the coonect flag where in the app startup does it gets triggered?

Robyt3

ddnet "connect localhost:8303" also works

This.

The command line arguments are parsed by calling ParseArguments in the main function in client.cpp

Right

But some commands are stored and only executed a bit later when StoreCommands(false) is called, because those commands that have CFGFLAG_STORE don't work immediately

Ok thank you

19:10

I appreciate all the help.

roby when add RXJava to ddnet?

2

There is that https://github.com/ReactiveX/RxCpp

GitHub - ReactiveX/RxCpp: Reactive Extensions for C++

Reactive Extensions for C++. Contribute to ReactiveX/RxCpp development by creating an account on GitHub.