01:57 < ws-client> <ChillerDragon> Hmm ddnet master seems to pass pipelines but my pending pr and fork fail both in unit tests whats up with that <:tee_thinking:478629518358085653>
01:57 < ws-client> <ChillerDragon> https://github.com/ddnet/ddnet/actions/runs/8273728097/job/22637984086?pr=7947
01:57 < ws-client> <ChillerDragon> https://github.com/ZillyInsta/ddnet-insta/actions/runs/8273683092/job/22637865388
01:58 < ws-client> <ChillerDragon> https://zillyhuhn.com/cs/.1710377866.png
01:58 < ws-client> <ChillerDragon> Did I break that?
02:34 < ws-client> <ChillerDragon> jopsti ur also netcup customer right? They shut down my vps once when i was port scanning for a mc server which seems fair. Now they took down my server again saying I did scan the ip 138.201.27.25 with src port 25+- and destination port 25+- no idea what happend there but sounds like bullshit.
02:34 < bridge> <Ewan> lmfao
02:35 < bridge> <Ewan> why port scanning for mc
02:35 < ws-client> <ChillerDragon> Any tips on how to get the vps back? Did that happen to you as well?
02:35 < bridge> <Ewan> are u troll
02:35 < ws-client> <ChillerDragon> @Ewan liveoverflow
02:35 < bridge> <Ewan> we had 3 ppl come in and grief on my last mc server
02:35 < bridge> <Ewan> rly glad i had coreprotect
02:36 < ws-client> <ChillerDragon> I might have to switch hoster if netcup keeps doing this. They are holding all my data hostage and keeping all my services down until i write them a letter that proofs i fixed the problem. Which is tricky in this case.
03:42 < bridge> <murpi> ddnet.org was unreachable a couple hours ago, and I assume the bot didn't attempt to reconnect to Discord once the outage resolved itself
04:37 < bridge> <miguilim> chillerDragon: add the infos about integer overflow:
04:37 < bridge> <miguilim> 
04:37 < bridge> <miguilim> https://protocol-docs.teeframe.com/fundamentals.html#integer-overflow
04:37 < bridge> <miguilim> https://protocol-docs.teeframe.com/packets/snaps-concept.html#calculating-crc
04:37 < bridge> <miguilim> https://protocol-docs.teeframe.com/packets/snaps-concept.html#string-packing-to-snap-items
04:38 < bridge> <miguilim> I also added a new topic talking about snap IDs https://protocol-docs.teeframe.com/packets/snaps-concept.html#snap-ids
05:06 < bridge> <miguilim> I also added a new topic talking about snap item IDs https://protocol-docs.teeframe.com/packets/snaps-concept#snap-item-id-id
05:11 < bridge> <miguilim> chillerDragon: added the infos about integer overflow:
05:11 < bridge> <miguilim> 
05:11 < bridge> <miguilim> https://protocol-docs.teeframe.com/fundamentals.html#integer-overflow
05:11 < bridge> <miguilim> https://protocol-docs.teeframe.com/packets/snaps-concept.html#calculating-crc
05:11 < bridge> <miguilim> https://protocol-docs.teeframe.com/packets/snaps-concept.html#string-packing-to-snap-items
07:02 < ws-client> <ChillerDragon> <:poggers2:1008007455936094328> @miguilim 
07:13 < ws-client> <ChillerDragon> @robyt3 the client not being able to quit is getting really annoying is someone on that? If not can we revert the change that waits for all systems to shutdown? I sometimes reboot multiple clients in a 5sec interval and now i have to wait up to 20seconds with it also trapping my mouse on shutdown that is not okay
07:14 < bridge> <Ewan> same
07:14 < ws-client> <ChillerDragon> I would patch it my self but the maintenance effort of that patch is huuuge
09:50 < bridge> <blazulite> hi
09:50 < bridge> <blazulite> uhh I'm trying to host my own FNG2 server using AWS (EC2)
09:51 < bridge> <blazulite> Everything's been OK but when I try to open the start.sh
09:52 < bridge> <blazulite> it says "Permission denied"
09:52 < bridge> <blazulite> I already did "$ uwf allow 8303" too
09:53 < bridge> <blazulite> any kinf of help would be appreciated 🥺
09:53 < bridge> <blazulite> any kind of help would be appreciated 🥺
09:55 < bridge> <furo321> Make it executable, `chmod +x start.sh`
10:06 < bridge> <blazulite> alright, I'm gonna check if th at works
10:06 < bridge> <blazulite> thanks in advance
10:08 < bridge> <blazulite> It seems like it doesn't work anyway
10:08 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217761248272322601/image.png?ex=6605338a&is=65f2be8a&hm=8a6f6537710f5c3018d8baf6c245e39f68c48647c8e3651b8bc0becd80d4fe51&
10:08 < bridge> <blazulite> It seems like that doesn't work neither
10:08 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217761248272322601/image.png?ex=6605338a&is=65f2be8a&hm=8a6f6537710f5c3018d8baf6c245e39f68c48647c8e3651b8bc0becd80d4fe51&
10:12 < bridge> <Peakies> chmod +x fng2_srv
10:12 < bridge> <robyt3> ChillerDragon: The client not quitting should already be mostly fixed in nightly. All there is left is adding some more specific timeout values for some of the HTTP requests, which have the default timeouts that probably cause them to wait very long if you have bad internet. I'm also working on overhauling the skin loading in general.
10:14 < bridge> <blazulite> OK WOW THANKS
10:14 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217762682577358848/image.png?ex=660534e0&is=65f2bfe0&hm=b42607a084adc6be1e5b7226d36626f63683e72c37e9ae12ed69fd8e83b44dd9&
10:14 < bridge> <Peakies> yw
10:14 < bridge> <blazulite> Why does the "configure your firewall/nat thing keeps happening though
10:14 < bridge> <blazulite> if I already put `uwf allow 8303`
10:15 < bridge> <Peakies> ufw*
10:15 < bridge> <blazulite> OH
10:15 < bridge> <robyt3> ChillerDragon: The ASAN run on #8098 also failed with a random segfault around the same time. Does it happen on master? Maybe CI is broken again.
10:15 < bridge> <DDNet> https://github.com/ddnet/ddnet/pull/8098
10:15 < bridge> <blazulite> then that may be whay I did wrong I guess
10:16 < bridge> <Peakies> maybe
10:17 < bridge> <blazulite> nope it still doesn't work
10:19 < bridge> <Peakies> iptables -I INPUT -p udp --dport 8303 -j ACCEPT
10:21 < bridge> <blazulite> wasn't sure if that was an -L or -I so I tested with both and none of them worked
10:21 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217764605758275584/image.png?ex=660536aa&is=65f2c1aa&hm=e24a68c92382de14a23ece3925c6498ec6915c16b06e45aac77714fd878ba8d6&
10:25 < bridge> <blazulite> wasn't sure if that was an `-l` or `-i` so I tested with both and none of them worked
10:25 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217764605758275584/image.png?ex=660536aa&is=65f2c1aa&hm=e24a68c92382de14a23ece3925c6498ec6915c16b06e45aac77714fd878ba8d6&
10:26 < bridge> <Peakies> ```
10:26 < bridge> <Peakies> ufw enable
10:26 < bridge> <Peakies> ufw allow 8303/udp
10:26 < bridge> <Peakies> ufw status
10:26 < bridge> <Peakies> ```
10:28 < bridge> <blazulite> alright
10:28 < bridge> <blazulite> I did that
10:28 < bridge> <blazulite> time to check again if it works
10:29 < bridge> <blazulite> wtf
10:29 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217766561109114920/image.png?ex=6605387d&is=65f2c37d&hm=9cc4d2c100f0b520c425c1daa4eb6f32d40289aed4588d6e81e7101d85424c44&
10:29 < bridge> <blazulite> important to know that literally everything's on allow
10:29 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217766657632768020/image.png?ex=66053894&is=65f2c394&hm=fef6a7093b2fc783c6b045c58cf22519018fe2925121b9ae2bc811e3953ce96a&
10:30 < bridge> <blazulite> important to know that literally everything's on `ALLOW`
10:30 < bridge> <blazulite> https://cdn.discordapp.com/attachments/293493549758939136/1217766657632768020/image.png?ex=66053894&is=65f2c394&hm=fef6a7093b2fc783c6b045c58cf22519018fe2925121b9ae2bc811e3953ce96a&
10:31 < bridge> <Peakies> then I don't know sorry 😦
10:32 < bridge> <blazulite> nah don't worry man you tried to help and that's all I care abt 😅
10:32 < bridge> <blazulite> thanks and have a nice day :3
10:32 < bridge> <Peakies> tnx you too ❤️
10:35 < bridge> <blazulite> anyway I think it's funny how every time I tried to host a teeworlds server I always finish having networking issues
10:37 < bridge> <furo321> I think you need to open the port through the web console on AWS. Try googling it.
10:37 < bridge> <blazulite> well that makes sense
10:37 < bridge> <blazulite> I'd try that
10:37 < bridge> <blazulite> thanks!
10:46 < bridge> <blazulite> uhm
10:46 < bridge> <blazulite> I added a Custom UDP inbound via the Security Group thing
10:46 < bridge> <blazulite> now I can't connect to the instance
10:46 < bridge> <blazulite> 💀
10:48 < bridge> <furo321> Over ssh?
10:49 < bridge> <blazulite> nope, I'm trying to connect via the "EC2 Instance Connect" thing, not an ssh client
10:49 < bridge> <blazulite> I also tried to connect via the instance connection endpoint and it still does not let me connect
10:53 < bridge> <furo321> Not sure, I think because you have enabled `ufw` now. It might be blocking that connection because you're not allowing ssh any more.
11:50 < bridge> <blazulite> that makes lots of sense
12:06 < ws-client> <ChillerDragon> @robyt3 about the startup times nice to hear I will merge again didn't do that in a few weeks.
12:07 < ws-client> <ChillerDragon> @robyt3 about the CI i have had multiple segfaults and ubsans today in two prs and in one fork. But also created a test branch on master with a readme change that passed so idk what is happening there
12:09 < ws-client> <ChillerDragon> > anyway I think it's funny how every time I tried to host a teeworlds server I always finish having networking issues
12:09 < ws-client> <ChillerDragon> @blazulite yes networking is the hardest part about it. Because setting up the game server it self is so smooth c:
12:09 < ws-client> <ChillerDragon> one other trap is configs and maps but thats about it
12:14 < bridge> <blazulite> ChillerDragon: ikr 😭
12:15 < bridge> <blazulite> I tried to also remove the inbound rules from the security group and I still can't connect via the EC2 Instance Connect
12:15 < bridge> <blazulite> My brain is exploding :D
12:46 < ws-client> <ChillerDragon> @robyt3 i just merged my client and on first launch I had a 20 sec freeze plus gnome popup saying your program froze. Is this my custom http call having wrong timeouts? But the logs showed that one as done already its usually fast and clean.
12:46 < ws-client> <ChillerDragon> I can not reproduce it anymore i still have a 2 seconds delay but its much better than 20
12:49 < bridge> <heinrich5991> 20 sec freeze might be just the way we load
12:49 < bridge> <heinrich5991> unrelated to http and everything
12:49 < bridge> <heinrich5991> we don't process window messages during loading IIRC
12:49 < ws-client> <ChillerDragon> i mean quitting not booting
12:49 < bridge> <heinrich5991> ah
12:49 < ws-client> <ChillerDragon> load is smooth
13:21 < bridge> <heinrich5991> how would I write a small custom HTTP server to test our HTTP client impl?
13:36 < bridge> <learath2> `python3 -m http.server`?
13:47 < bridge> <teero777> yo where was that one library for tw maps in rust?
13:57 < ws-client> <ChillerDragon> @heinrich5991 yea wat lerato said. Or do you need it to be slow? Then I would go with express/sinatra/flask or any other web framework where you can define routes in 5 lines of code and put a sleep in there
13:59 < ws-client> <ChillerDragon> @teero777 on the blaze it git cloud
14:00 < ws-client> <ChillerDragon> high dutch people gitlab.com/patiga/twmap
14:00 < bridge> <teero777> send link!
14:00 < bridge> <teero777> cant find it
14:00 < ws-client> <ChillerDragon> https://gitlab.com/Patiga/twmap
14:00 < bridge> <teero777> <3
14:00 < ws-client> <ChillerDragon> https://zillyhuhn.com/cs/.1710421241.png
14:00 < ws-client> <ChillerDragon> blaze
14:15 < bridge> <ryozuki> ```rust
14:15 < bridge> <ryozuki> if if if a == b {
14:15 < bridge> <ryozuki>     b == c
14:15 < bridge> <ryozuki> } else {
14:15 < bridge> <ryozuki>     a == c
14:15 < bridge> <ryozuki> } {
14:15 < bridge> <ryozuki>     a == d
14:15 < bridge> <ryozuki> } else {
14:15 < bridge> <ryozuki>     c == d
14:15 < bridge> <ryozuki> } {
14:16 < bridge> <ryozuki>     println!("True!");
14:16 < bridge> <ryozuki> } else {
14:16 < bridge> <ryozuki>     println!("False!");
14:16 < bridge> <ryozuki> }
14:16 < bridge> <ryozuki> ```
14:16 < bridge> <ryozuki> https://without.boats/blog/iterators-and-traversables/
14:28 < bridge> <ryozuki> thats lit
14:28 < bridge> <ryozuki> @heinrich5991 @learath2 @jupeyy_keks @chairn 
14:28 < bridge> <ryozuki> 
14:28 < bridge> <ryozuki> # Happy PI day!
14:28 < bridge> <ryozuki> :owo:
14:32 < bridge> <ryozuki> https://cdn.discordapp.com/attachments/293493549758939136/1217827707115540550/450px-Happy_Pi_Day_sign_for_sales_on_pies_at_Delicious_Orchards.png?ex=6605716f&is=65f2fc6f&hm=2fbf266df1017890d2256839bc5fe3a750c1bfc8ec00955cb6f23ed77f80e66c&
15:15 < bridge> <cyberFighter> tf are u doign with those 14 cents saved
16:25 < bridge> <heinrich5991> need custom headers, custom status, custom sending delays, etc.
16:28 < bridge> <learath2> hyper was pleasent enough to use in rust, maybe that works for you?
16:28 < bridge> <learath2> Not very ergonomic though
16:37 < bridge> <learath2> Btw I don't like what we ended up with in #8019. `Gpu`, `Ui`, `Id` all look meh. At the very least acronyms should have been all capital
16:37 < bridge> <DDNet> https://github.com/ddnet/ddnet/pull/8019
16:37 < bridge> <learath2> And all the renames absolutely ruined blames for years to come
16:38 < bridge> <learath2> Lose lose
16:44 < bridge> <learath2> `Rename all variables for strict camel-casing of abbreviations` will be the new `copied refactor to trunk` for the next decade
16:46 < bridge> <heinrich5991> I suggest using ignoreRevs or tig, both very powerful in dealing with that
16:48 < bridge> <learath2> Atleast renames don't add lines like mass styling. Hopefully this will be the last time we do massive 300 file commits just to change styles
16:49 < bridge> <heinrich5991> I still want one for proper prefixing of variable names
16:50 < bridge> <learath2> :PainChamp:
17:01 < bridge> <murpi> https://tenor.com/view/imnota-cat-fubuki-gif-18573529
17:59 < bridge> <teero777> @patiga are there any example on how do place some block with libtw?
17:59 < bridge> <teero777> @patiga are there any example on how do place some blocks with libtw?
17:59 < bridge> <teero777> @patiga are there any examples on how do place some blocks with libtw?
18:01 < bridge> <teero777> @patiga are there any examples on how do place some blocks with twmap?
18:30 < bridge> <peter0x44> https://teewebs.net
18:30 < bridge> <peter0x44> Does this thing still work? Or no?
18:30 < bridge> <peter0x44> And if it does, can I connect to a lan server with it in the same way?
18:32 < bridge> <louis> @murpi /upload still seems to be down
18:36 < bridge> <furo321> The project itself should still work, but seems like the website is broken. The server you will be connecting to needs to support websockets with `-DWEBSOCKETS=ON`.
18:40 < bridge> <furo321> The project itself should still work, but seems like the website is broken. The server you will be connecting to needs be compiled with `-DWEBSOCKETS=ON`.
18:55 < bridge> <learath2> I seem to have trauma from years of propaganda. I know exactly what data structure I should be using but there is this voice in my head telling me it's wrong. Normalize using linked lists, especially ones backed by contiguous storage
19:04 < bridge> <murpi> @learath2 SkinDB Bot seems to be offline or not connecting, maybe you wanna check?
19:05 < bridge> <learath2> I gave it a restart and it just came back up, it doesn't log at all so nothing to diagnose really
19:06 < bridge> <learath2> And YADDB dying yesterday was indeed a connection issue, no idea why it never recovered
19:14 < bridge> <zhn> i have a silly question about `CVoteOptionServer`, why does it have `char m_aCommand[1]`?
19:15 < bridge> <zhn> like, why exactly `[1]`
19:19 < bridge> <learath2> It's an old hack to compensate for C++s lack of flexible array members
19:20 < bridge> <learath2> It's technically UB, but we allocate more than needed for just `CVoteOptionServer` and access the extra allocated space through `m_aCommand`
19:20 < bridge> <zhn> huh
19:21 < bridge> <zhn> so its pointer to last structure byte and first for string that lays in heap's padding memory?
19:22 < bridge> <learath2> It's treated that way yes, but as far as C++ is concerned it is an array of a single char that we are reading beyond the bounds of
19:22 < bridge> <peter0x44> I do this sometimes because you can pass it to functions without having to write &
19:23 < bridge> <peter0x44> Like
19:23 < bridge> <peter0x44> ```
19:23 < bridge> <peter0x44> void example(int*);
19:23 < bridge> <peter0x44> 
19:23 < bridge> <peter0x44> int x[1];
19:23 < bridge> <peter0x44> example(x);
19:23 < bridge> <peter0x44> ```
19:23 < bridge> <zhn> looks useful for string only though
19:23 < bridge> <zhn> strings*
19:23 < bridge> <zhn> like how would you find the actual length without passing it
19:24 < bridge> <learath2> If you don't store a string, you use a size member
19:24 < bridge> <zhn> ah, fair
19:25 < bridge> <learath2> It's a nasty nasty hack that no compiler can break because it's the goto workaround for flexible array members
19:25 < bridge> <zhn> does it need to be refactored in future anyway?
19:25 < bridge> <zhn> in ddnet i mean
19:25 < bridge> <zhn> the way vote options are stored, the struct itself and vote handling in general
19:25 < bridge> <learath2> I wouldn't really mess around with it without good reason
19:26 < bridge> <zhn> /shrug making `CGameContext` less god class
19:26 < bridge> <zhn> nice xd
19:26 < bridge> <learath2> Most people would rework that to `std::string m_Command` and I don't like to introduce random heap allocations because C++ is a deficient language
19:29 < bridge> <learath2> WG14 suggests people use `char m_aCommand[SOME_MAX_SIZE]` but allocate less if needed. Which makes the access not UB at the cost of making `sizeof CVoteOptionServer` insane
19:32 < bridge> <zhn> hm what about just using `const char*` and allocate string outside the structure (it will be placed after the structure anyway i think) in the heap
19:32 < bridge> <zhn> `CHeap::StoreString(...)` seems sane
19:33 < bridge> <zhn> idk, maybe it's just my view, `m_aCommand[1]` looks weird for me xd
19:33 < bridge> <learath2> I don't like it, I want nothing to do with it, but I'm sure you can get heinrich or deen to merge it
19:33 < bridge> <learath2> Adding a pointless layer of indirection
19:34 < bridge> <zhn> anyway thanks for help, wouldn't understand it myself
21:21 < bridge> <teero777> yo
21:21 < bridge> <teero777> whats the point of multiplying by the ramp value and then dividing by it again?
21:21 < bridge> <teero777> https://cdn.discordapp.com/attachments/293493549758939136/1217930507707355147/image.png?ex=6605d12d&is=65f35c2d&hm=218b399857e810efe7751b8f7c70a2a53cbe8a9537fab3acbddf907ae5844cd2&
21:21 < bridge> <teero777> is it affected by MoveBox
21:21 < bridge> <teero777> is it affected by MoveBox?
22:00 < bridge> <borsmatyas02> are you collecting data about player behaviour?
22:01 < bridge> <borsmatyas02> if yes can you share statistics?
23:12 < bridge> <0xdeen> There are some stats and raw data in https://ddnet.org/stats/