Teeworlds - development

Teeworlds

discord.gg/teeworlds / development

For discussions around the development of the official Teeworlds

Between 2019-01-26 00:00:00Z and 2019-01-27 00:00:00Z

hmm, made a script to crop the gametype icons, now they all look blank in the browser :| guess imageshark compresses too much

@Dune sorry bud I have not worked on this script in a bit. I forgot that I was. Got caught up on some things.

sure :)

[teeworlds/teeworlds] Issue opened: #1998 First click on a server after leaving one acts like a double click

I was reported this bug by someone ingame (@fudgyking maybe?), tested and confirmed it

14:35

[teeworlds/teeworlds] Issue opened: #1999 On joining a server, the vote subject isn't visible

This gotta get fixed somehow: !image

[teeworlds/teeworlds] Issue opened: #2000 [Sliders] Can't click to jump to a specific value nor enter that value manually

It would be great to just click somewhere on the slider to change a value or manually enter a value. It feels sloppy the way it is right now.

Can anyone test if they insta crash when they type "logout" in f2 after logging in ?

20:13

it always does for me on my LAN server

Im fixing it

never happens and still doesn't for me. would be nice if you can reproduce

I figured out why it happened

22:04

I was running a server without the new maplist feature https://github.com/teeworlds/teeworlds/commit/bc38f677506bb0244e4c3f986332f9bb72a8c4b9

merged "List maps" by Learath2 and cinaera. skipped redundant "map...

…command. closes #1086

22:04

but my client had the feature

ooh

22:04

and that crashes on disconnect? nice find :)

so in "CConsole::DeregisterTempMapAll()"

22:04

m_pTempMapListHeap->Reset(); crashes

22:04

cause there's no check

22:04

to see if it exists

make a separate branch/commit for that fix if possible :p

well idk if its needed cause it only can happen if you logout on a up to date client on an old server

well yeah it is, shouldn't be there

22:06

0.7.3 clients should be able to use rcon on 0.7.2 and < servers

so I make a branch for a 1 line fix?

the fix is consequential even though the amount of code isn't

22:06

it's usually best not to mix in fixes and features

22:07

if by any chance the feature is delayed or doesn't make it for some reason, the fix still should, right?

its already merged no

huh? what is?

the maplist feature

by @LordSk 🦋's word, the best commits do not add volumes of code :D

22:08

yes it is

22:08

I meant, don't mix it with other UI features or something

yes I agree that commits need to be granular etc

22:08

nono ofc I wont commit that fix with random other things in it :p

I used to everytime I thought the rest was benign but it always turns out to be a bad idea :p

now whats the best way to put that fix in the official repo

just PR it?

22:11

git checkout master
git branch fix-maplist-crash

remote origin?

git push origin fix-maplist-crash yeah, you can only push to your fork since that's where you have write access

ye

[teeworlds/teeworlds] Pull request opened: #2001 Add check before resetting the local rcon map list

Fixes a crash if the server does not have the MapListEntries logic

seems like it worked

yeah, all good :)

Ah, the good old pointers. Nice catch, Fudgy!

Exported 46 message(s)