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 2018-02-05 00:00:00Z and 2018-02-06 00:00:00Z

[ddnet/ddnet] Pull request closed: #1017 Organize .gitignore and also ignore build dir stuff

07:03

[ddnet:master] 2 new commits

22bacc1 Organize .gitignore and also ignore build dir s... - heinrich5991 ee117df Merge pull request #1017 from heinrich5991/pr_g... - def-

[ddnet/ddnet] Issue opened: #1018 Window size doubles on every launch [macOS]

Every time I launch DDNet, the window size doubles.

[freenode] <nemo> hm. regex might be clearer

13:15

[freenode] <nemo> oh... huh. doesn't seem gitignore supports syntax: regexp ?

13:15

[freenode] <nemo> shrugs

[quakenet] <deen> nemo: no, this is what's matchable: https://git-scm.com/docs/gitignore#_pattern_format

13:27

[freenode] <nemo> yeah. was reading that

13:27

[freenode] <nemo> guess hg is spoiling me again

13:33

[quakenet] <heinrich5991> nemo: using regexes also complicates the tools that have to deal with .hgignores

[ddnet:master] 1 new commit

d234c55 Try macosx highdpi fix - def-

[freenode] <nemo> shrugs

13:44

[freenode] <nemo> like regexes are such a complicated thing â˜º

13:44

[freenode] <nemo> anyway. if a tool supports .hgignore it obv has regex builtin

[ddnet/ddnet] Issue opened: #1019 In-game Demo

Watching demo in game is good idea! Do you want to see the demo with cheater without leave from server?

@heinrich5991 did you have info about the tw protocol?

15:46

https://github.com/heinrich5991/libtw2/tree/master/doc

heinrich5991/libtw2

libtw2 - Some Teeworlds stuff in Rust

15:46

nvm :)

15:46

this is rly handy

https://github.com/heinrich5991/libtw2/blob/master/doc/connection.md

heinrich5991/libtw2

libtw2 - Some Teeworlds stuff in Rust

15:47

does the [ mean something special?

not mandatory

with all the info in doc/ is possible to make a headless client?

no

15:49

connection establishment isn't completely documented

:/

feel free to complete the docs

where should i look at (src) to fulfill that info?

you could take a look into downloader

[freenode] <nemo> oh.

15:57

[freenode] <nemo> mildly on topic to the above

15:57

[freenode] <nemo> I've been poking around at options for finally getting the workplace onto a documentation wiki

15:57

[freenode] <nemo> I ran into ikiwiki

15:57

[freenode] <nemo> http://ikiwiki.info/

15:58

[freenode] <nemo> it seemed not-bad-at-all - kind of a better variant of what hedgewars is currently doing

15:58

[freenode] <nemo> I was wondering if you guys had anything along those lines.

15:58

[freenode] <nemo> a wiki backed by vcs

16:01

[quakenet] <heinrich5991> github wiki works ;)

[freenode] <nemo> heh. ah right

16:19

[freenode] <nemo> not the place to ask for suggestions then ðŸ˜�

16:23

[quakenet] <heinrich5991> mh?

16:23

[quakenet] <Learath2> heinrich5991: I'm not learning rust as fast as I like :D

16:23

[quakenet] <heinrich5991> Learath2: rust is a programming language with new concepts

16:23

[quakenet] <heinrich5991> these tend to take some time learning

16:24

[quakenet] <Learath2> it was more a timeconstraint issue

16:24

[quakenet] <Learath2> feel free to write a parser for the teehistorian rcon things

16:26

[quakenet] <heinrich5991> Learath2: what is the desired output format?

16:30

[quakenet] <Learath2> [YYYY-MM-DD HH:MM:SS] KeyIdent='%s' rcon='%s'

16:31

[quakenet] <Learath2> [YYYY-MM-DD HH:MM:SS] KeyIdent='%s' NickName='%s' logged in

16:31

[quakenet] <Learath2> s/$//out/

16:31

[quakenet] <Learath2> [YYYY-MM-DD HH:MM:SS] KeyIdent='%s' changed nick %s->%s

16:31

[quakenet] <heinrich5991> s#$#/out#

16:32

[quakenet] <Learath2> would also be useful

16:32

[quakenet] <Learath2> keep forgetting you can use something other then /

16:34

[quakenet] <Learath2> heinrich5991: those messages are grep friendly enough, right?

16:35

[quakenet] <heinrich5991> I guess?

16:35

[quakenet] <heinrich5991> dunno

16:35

[quakenet] <heinrich5991> we can change them on a whim ;)

16:35

[quakenet] <Learath2> oh yeah true, we already have the data

16:36

[quakenet] <Learath2> heinrich5991: do you have KeyIdent available in teehistorian files btw?

16:36

[quakenet] <heinrich5991> yes

16:36

[quakenet] <heinrich5991> it's what I recently added, don't remember?

16:36

[quakenet] <Learath2> wasn't sure if you added keyid which is only valid for that session or keyident

keyident IIRC

[quakenet] <Learath2> btw @heinrich5991 bridge is uncloaked if you don't know

16:38

[quakenet] <heinrich5991> yea, I know

16:38

[quakenet] <heinrich5991> haven't added an issue to matterbridge yet

\o/

Image attachment

17:21

what is the use of extra_token and token anyway?

17:21

oh

17:21

nvm i know now

oh great Savander isn't here anymore

[quakenet] <deen> why is that great?

</sarcasm>

because you bullied him away

any of you people familiar with symfony?

savander

feelsbadman

1

[quakenet] <Learath2> well if any of you come accross Savander tell him I'm sorry for talking

ddnet

ddnet2

ddnet

ddnet2

ddnet

[ddnet/ddnet] Issue opened: #1020 fix dummy_hook

`[ 54%] Building CXX object CMakeFiles/DDNet.dir/src/engine/client/client.cpp.o /home/meer/Downloads/Restricted-master/src/engine/client/client.cpp:797:1: Fehler: expected unqualified-id before »if« if(g_Config.m_ClDummyHookFly) ^~ /home/meer/Downloads/Restricted-master/src/engine/client/client.cpp:826:4: Fehler: expected unqualified-id before »if« if(!Size && (!m_DummyInput.m_Direction && !m_DummyInput.m_Jump && !m_DummyInput.m_Hook)) ^~ /home/meer/Downloads/Restricted-maste...

18:22

[ddnet/ddnet] Issue closed: #1020 fix dummy_hook

18:24

[ddnet/ddnet] Pull request opened: #1021 CMakeLists.txt: enable using system libwebsockets, pnglite, wavpack

[ddnet/ddnet] Issue opened: #1022 Update libwebsockets

As of libwebsockets 2.4.1, many APIs have been changed. For example, libwebsocket_callback_on_writable is gone. This stops ddnet from linking against system libraries, which is required by Debian packaging guideline. Please consider updating bundled libwebsockets codes.

spam of issue

18:33

xD

@heinrich5991 consider https://github.com/heinrich5991/libtw2/pull/6

add the unpacker source from ddnet as example by Ryozuki · Pull Request #6 · heinrich5991/libtw2

@Ryozuki I don't want to add code

its rly handy

if you want code, look at the teeworlds/DDNet code

or atleast add this

18:53

https://github.com/ddnet/ddnet/blob/43f246f17495607b80c4c0d9092a506535a4619a/src/engine/shared/compression.cpp#L33

ddnet - DDraceNetwork, a mod of Teeworlds

18:53

you make life easier for others

18:54

(link)

the documentation is not for finding the code

ok wathever

the docs are for describing the protocol

18:54

look – if I want to see how it is implemented in teeworlds, I can go to the code and look

you know that some docs contain code as example?

dunno

they do

18:55

https://nodejs.org/api/dgram.html

18:55

scroll

this is supposed to be some kind of document that should allow clean-room engineering

u will see code examples

i.e. if you never look at the original code, you should be able to produce an equivalent implementation

18:55

if I add original code to the document, it defeats the purpose IMO

its not about original code but an example of implementation

but then people will take the implementation and see it as the only way to do things

18:56

I'm trying to say what part of the implementation matters and which does not

if you can read peoples mind then i guess yes

18:57

well i see you don't want it, i will just clone it and add it

I won't stop you from doing that ^^

[quakenet] <Learath2> s/won't/can't/ :D

?

18:57

what even means s/

I won't and can't stop you, @Learath2, does that make you happy?

18:58

s/search/replacement

[quakenet] <Learath2> positively cheerful :)

18:59

[quakenet] <deen> Learath2: Rico said you forgot to start ictf servers yesterday btw :D

18:59

[quakenet] <Learath2> deen: oh yeah, that was an insane day

18:59

[quakenet] <deen> luckily I was online in the morning and could do it

https://github.com/Ryozuki/libtw2/blob/master/doc/int.md ❤

libtw2 - Some Teeworlds stuff in Rust

[quakenet] <Learath2> heinrich5991: up for some ddnet later?

19:00

[quakenet] <heinrich5991> very much later maybe

19:00

[quakenet] <Learath2> I somehow can't stay up till very-much later the last week, I just pass out at around 2400 for some reason :/

19:01

[quakenet] <heinrich5991> currently trying to do the cmakelists.txt thing

19:01

[quakenet] <deen> It's called "needing a reasonable amount of sleep to function"

19:01

[quakenet] <deen> heinrich5991: wait

19:01

[quakenet] <deen> the debian guy did it already

19:01

[quakenet] <heinrich5991> yea, seen that

19:01

[quakenet] <deen> https://github.com/ddnet/ddnet/pull/1021

CMakeLists.txt: enable using system libwebsockets, pnglite, wavpack by yangfl · Pull Request #1021 · ddnet/ddnet

19:02

[quakenet] <Learath2> well yeah but I'd rather wake up at 0800 and not 0600

19:03

[quakenet] <heinrich5991> deen: didn't follow the rest of the Find modules and (now googled) seem to be a copy-paste from code from the net

19:03

[quakenet] <deen> heh, ok

19:03

[quakenet] <deen> I didn't look at the changed files

wew

19:38

does someone knows what can cause that?

0x00007ffff6e631ea in _int_free (av=0x7ffff718f620 <main_arena>, p=<optimized out>, have_lock=0) at malloc.c:3987
#0  0x00007ffff6e631ea in _int_free (av=0x7ffff718f620 <main_arena>, p=<optimized out>, have_lock=0) at malloc.c:3987
#1  0x000000000043784a in mem_free (p=0x1731640) at src/base/system.c:491
#2  0x00000000004302c4 in CSnapshotStorage::PurgeUntil (this=0x7ffff5b1d390, Tick=2418360) at src/engine/shared/snapshot.cpp:413
#3  0x000000000043eb70 in CServer::DoSnapshot (this=0x7ffff5857010) at src/engine/server/server.cpp:751
#4  0x0000000000443910 in CServer::Run (this=0x7ffff5857010) at src/engine/server/server.cpp:1848
#5  0x0000000000446459 in main (argc=3, argv=0x7fffffffe138) at src/engine/server/server.cpp:2525

not really, you could try running it through valgrind or compiling with -fsanitize=address

hmh ok

19:41

mem_free(pHolder);

19:41

this shit crash

is this ddnet btw?

i guess no

it is

some botty g** shit i think

19:55

o.O

but not the newest

baumalein != ddnet

19:56

wait... rei is it you? xD

no

19:56

i'm Rei

xD

[quakenet] <deen> wrong pointer? freeing twice?

see you later alligator

19:57

in a while crocodile

idk :/

19:59

thats the problem

lol is this in the server code

20:12

for other reasons i ran my server thru valgrind earlier and there were a few uninitialized arrays in server.cpp dosnapshot function

sounds like that should be fixed

i added = { 0 }; to all of them in this function and it made valgrind stop spamming

Image attachment

well, are they actually ever accessed beyond their written position?

20:15

is it a valgrind false positive?

not sure tbh

[quakenet] <eeeee> valgrind might be complaining about reading uninitialized data (but still in bounds)

20:20

[quakenet] <eeeee> reading such data won't segfault, but can lead to non-deterministic behavior

if the functions the arrays r passed to are relying on them to be null terminated i can see why it might cause a problem

20:21

idk much about how snapshots work

[quakenet] <eeeee> i doubt they are null terminated

20:23

[quakenet] <eeeee> it sucks to append to null terminated structs

hm true

[quakenet] <Learath2> phpstorm might be the most overreaching ide that i've ever used, and it's done amazingly well

23:13

[quakenet] <Learath2> apparently you can do bloated right

Exported 174 message(s)