00:24 < bridge_> 2 ideas : 00:24 < bridge_> 1 - Add a search bar in the Players Tab 00:26 < bridge_> what's your use case for this? 00:26 < bridge_> 2 - Add a queue system for full servers : When server full and someone attempt to connect, he gets a number, that number gets +1. When someone disconnects server sends a ping back to all ppl that tried to connect. If they still are trying to connect, n°1 gets connected, the others get their number -1. If someone is no longer trying to connect, -1 to all the next ones 00:26 < bridge_> something like this idk 00:26 < bridge_> this has an open issue: https://github.com/ddnet/ddnet/issues/3453#issuecomment-1638880126 00:27 < bridge_> something like this idk maybe the method i described make servers burn 00:27 < bridge_> theres 64 players in the Players tab i'm searching for a specific one i know his name 00:27 < bridge_> like... it's a Search bar 00:27 < bridge_> why do we have a search bar for servers 00:28 < bridge_> why wouldn't we?? 00:28 < bridge_> "let's go fng" -> types fng -> wins 00:28 < bridge_> we could even have the same system as in server browser, where u could sort players by muted, emote muted, or friends 00:28 < bridge_> exactly 00:28 < bridge_> so lets get it for Players tab too :D 00:29 < bridge_> because it takes time to implement and might clutter the UI if we don't need it 00:29 < bridge_> why? you can search a player name in the server searchbar 00:30 < bridge_> won't clutter 00:30 < bridge_> https://cdn.discordapp.com/attachments/293493549758939136/1133164363797233795/image.png 00:30 < bridge_> to mute rapidly 00:31 < bridge_> do we have a /mute cmd btw ? 00:31 < bridge_> f1 mute 00:31 < bridge_> ah ok ok 00:31 < bridge_> Idea 3 : Add a /mute cmd. Unmute if the player is already muted 00:32 < bridge_> why 00:32 < bridge_> f1 mute 00:32 < bridge_> since it's client-side, it has to be in the console 00:32 < bridge_> understandable 00:32 < bridge_> ~~Idea 3 : Add a /mute cmd. Unmute if the player is already muted~~ 00:34 < bridge_> Or we can make the ugliest hack known to mankind. Intercept slash commands on the client :BigBrain: 00:34 < bridge_> seems pointless 00:34 < bridge_> I did not dare say this out loud 00:35 < bridge_> It is very very pointless 00:35 < bridge_> not entirely 00:35 < bridge_> Slightly easier to use? 00:35 < bridge_> how does the mute command work in the console? we don't complete player names in the local console yet, do we? 00:36 < bridge_> Nope, we don't complete them anywhere 00:36 < bridge_> we complete them in chat 00:36 < bridge_> Ah, yep that would help 00:36 < bridge_> ႇႀႂႅႀႂ 00:37 < bridge_> anyone with support for this font can screenshot how this looks like? 00:37 < bridge_> I only see boxes 00:37 < bridge_> anyone with support for this font can screenshot what this looks like? 00:38 < bridge_> https://cdn.discordapp.com/attachments/293493549758939136/1133166439478591528/Screenshot_20230725_013848_Discord.jpg 00:39 < bridge_> Our text renderer iirc can't handle this 00:40 < bridge_> I got a file with this filename 00:40 < bridge_> I wanted to see whether it's some fancy emoji 00:40 < bridge_> hm 00:40 < bridge_> but I don't see it 00:40 < bridge_> I wonder if it's a different encoding 00:42 < bridge_> it was xml-escaped in the file name 00:42 < bridge_> I guess it's not 00:42 < bridge_> the original filename was `пикник-0003` 00:42 < bridge_> https://github.com/ddnet/ddnet/issues/1087 00:43 < bridge_> https://github.com/ddnet/ddnet/pull/1080 00:43 < bridge_> https://github.com/ddnet/ddnet/pull/1082 00:43 < bridge_> https://github.com/ddnet/ddnet/issues/1085 00:43 < bridge_> https://github.com/ddnet/ddnet/pull/1080 00:43 < bridge_> https://github.com/ddnet/ddnet/pull/1082 00:43 < bridge_> the original filename was `пикник-0003.jpg` 00:43 < bridge_> Lol 00:43 < bridge_> thanks @DDNet 00:45 < bridge_> ``` 00:45 < bridge_> Content-Disposition: inline; filename="=?iso-8859-1?Q?пикник-0003.jpg?=" 00:45 < bridge_> Content-Transfer-Encoding: base64 00:45 < bridge_> Content-Type: application/octet-stream; name="=?iso-8859-1?Q?пикник-0003.jpg?=" 00:45 < bridge_> ``` 01:27 < bridge_> xd 02:47 < bridge_> oh sweet 03:06 < bridge_> ```cls 03:06 < bridge_> #666 03:06 < bridge_> ``` 03:06 < bridge_> https://github.com/ddnet/ddnet/pull/666 06:11 < bridge_> murpi only has 1 follower 06:11 < bridge_> we should fix that 09:38 < bridge_> And get him to 0? 09:38 < bridge_> (@tsfreddie) 09:38 < bridge_> if you can chiller 09:38 < bridge_> i dare you 09:51 < bridge_> ok now he has 4 09:51 < bridge_> now that should be good enough 09:51 < bridge_> follow 4 follow anyone :D 09:58 < bridge_> i follow if u have interesting stuff or do interesting stuff 09:58 < bridge_> same 09:58 < bridge_> most of my followers are bots or accounts from the web3 boom 09:58 < bridge_> https://cdn.discordapp.com/attachments/293493549758939136/1133307391992152114/image.png 09:58 < bridge_> :gigachad: 09:59 < bridge_> someone from microsoft follows me xd 09:59 < bridge_> cool 09:59 < bridge_> mixed feelings kek 10:01 < bridge_> i know a few ms people but they don't follow on github 10:01 < bridge_> most ppl don't use this feature 10:51 < bridge_> does the tele pred pr have pred for kill tiles too? 11:25 < bridge_> no 11:27 < bridge_> I think respawn tiles put you the furthest away from the nearest tee, but if you don't have /Showall then I don't think the client can know which respawn tile the server will pick 11:35 < bridge_> https://cdn.discordapp.com/attachments/293493549758939136/1133331649254404217/RDT_20230725_1135098494746713858065622.jpg 11:35 < bridge_> xorg wins over wayland 11:35 < bridge_> the x.org maintainers went to wayland IIRC 11:37 < bridge_> is vsync still forced? 11:40 < bridge_> ddg'ing "vsync forced wayland" leads me to https://old.reddit.com/r/linux_gaming/comments/lscc3v/ (RIP reddit) which leads me to the following merged pull request: https://gitlab.freedesktop.org/wayland/wayland-protocols/-/merge_requests/65 11:41 < bridge_> Yes 11:42 < bridge_> The prs for the drivers are still open 11:42 < bridge_> Nori will notify us when they in 😂 11:43 < bridge_> ☝️🤓 since its gitlab its a MR not a PR 11:45 < bridge_> xd 11:47 < bridge_> just call the concept whatever you want 11:47 < bridge_> ye its a joke 11:48 < bridge_> how well does ddnet work in wayland 11:49 < bridge_> does the xwayland bridge have forced vsync too? 11:51 < bridge_> my quick ddg search yields no results 11:55 < bridge_> @heinrich5991 Do you mean I should remove the entire commit adding `IO_NEWLINE` or just always use `\n` for the JSON writer? Should I add another constant `#define JSON_NEWLINE "\n"` then or use the literal `"\n"` everywhere in the tests? https://github.com/ddnet/ddnet/pull/6874#pullrequestreview-1544214616 11:57 < bridge_> I guess removing the entire commit with `IO_NEWLINE` would be good. don't add another constant for `JSON_NEWLINE`, but directly use `\n` everywhere 11:58 < bridge_> what do you think about that? @robyt3 11:59 < bridge_> Seems okay, I just added `IO_NEWLINE` to avoid the duplicate testing code, but only using `\n` also works 12:21 < bridge_> guys its possible to change the background of a map? no entities background 13:03 < bridge_> Yes 13:04 < bridge_> hi jopstat 13:04 < bridge_> jopstar 13:04 < bridge_> jup 13:04 < bridge_> juice man 13:05 < bridge_> https://discord.com/channels/252358080522747904/293493549758939136/1101897900289704047 13:05 < bridge_> @ryozuki ^ 13:05 < bridge_> in the editor? while playing? 13:05 < bridge_> @mpft hi Ewan 13:05 < bridge_> Iwan 13:05 < bridge_> Ip man 13:05 < bridge_> it’s me 13:06 < bridge_> Ip man 13:06 < bridge_> Legendary 13:06 < bridge_> :happy: 13:07 < bridge_> :win10: 13:08 < bridge_> I use windows in my room. But that's about all euphoria you'll get from me about windows 13:11 < bridge_> english confused me for a second 13:12 < bridge_> why is wireshark also on discord? 13:13 < bridge_> is there no good open-source communication medium left? 13:13 < bridge_> (sent to discord) 13:14 < bridge_> I recommend ChillerDragons IRC chat . It's debatable if it's good. But it's defs unique xd 13:27 < bridge_> lol 14:22 < bridge_> i feel validated 15:29 < bridge_> I just realized I started too many personal projects and none of them can be done because i'm just exhausted after work everyday 15:29 < bridge_> I'm not even doing overtime recentlyr 15:29 < bridge_> I'm not even doing overtime recently 15:29 < bridge_> very good 🙂 15:30 < bridge_> :brownbear: the last project shipped 15:30 < bridge_> so we can breath for a while 15:30 < bridge_> so we can breathe for a while 15:32 < bridge_> I keep falling asleep at 10pm and wake up at 4am lately 16:22 < bridge_> @heinrich5991 @learath2 @jupeyy_keks https://lock.cmpxchg8b.com/zenbleed.html 16:23 < bridge_> https://news.ycombinator.com/item?id=36848680 16:23 < bridge_> > It turns out that with precise scheduling, you can cause some processors to recover from a mispredicted vzeroupper incorrectly! 16:23 < bridge_> > 16:23 < bridge_> > This technique is CVE-2023-20593 and it works on all Zen 2 class processors, which includes at least the following products: 16:23 < bridge_> > 16:23 < bridge_> > AMD Ryzen 3000 Series Processors 16:23 < bridge_> > AMD Ryzen PRO 3000 Series Processors 16:23 < bridge_> > AMD Ryzen Threadripper 3000 Series Processors 16:23 < bridge_> > AMD Ryzen 4000 Series Processors with Radeon Graphics 16:23 < bridge_> > AMD Ryzen PRO 4000 Series Processors 16:23 < bridge_> > AMD Ryzen 5000 Series Processors with Radeon Graphics 16:23 < bridge_> > AMD Ryzen 7020 Series Processors with Radeon Graphics 16:23 < bridge_> > AMD EPYC “Rome” Processors 16:23 < bridge_> > 16:23 < bridge_> > The bug works like this, first of all you need to trigger something called the XMM Register Merge Optimization2, followed by a register rename and a mispredicted vzeroupper. This all has to happen within a precise window to work. 16:24 < bridge_> > 16:24 < bridge_> > We now know that basic operations like strlen, memcpy and strcmp will use the vector registers - so we can effectively spy on those operations happening anywhere on the system! It doesn’t matter if they’re happening in other virtual machines, sandboxes, containers, processes, whatever! 16:24 < bridge_> > 16:24 < bridge_> > This works because the register file is shared by everything on the same physical core. In fact, two hyperthreads even share the same physical register file. 16:24 < bridge_> old: https://discord.com/channels/252358080522747904/293493549758939136/1133110893643837440 16:24 < bridge_> noooo 16:24 < bridge_> u beat me to it this time 16:24 < bridge_> xd 16:24 < bridge_> somehow i missed it lol 16:25 < bridge_> But u didn't tag me to force me read the news xdd 16:25 < bridge_> xdd 16:26 < bridge_> cvtsi2sd 16:26 < bridge_> So if it's fixed in newer ryzens. Did they know about it? 16:26 < bridge_> man assembly instructions rly got some names 16:26 < bridge_> xdd 16:26 < bridge_> I payed 250€ for my 16 cores lmao 16:26 < bridge_> is 7020 new? 16:27 < bridge_> it's only fixed if you updated your microcode since essentially yesterday 16:27 < bridge_> That's graphics 16:27 < bridge_> I mean newer hardware 16:27 < bridge_> It didn't mention Ryzen 7000 series 16:28 < bridge_> @chairn this probs also interests u 16:28 < bridge_> @chairn indeed. That's why u have to buy newest hardware 16:28 < bridge_> Don't do the same mistake as me 16:28 < bridge_> is it even fixed in the latest hardware? 16:28 < bridge_> i've seen it already 16:29 < bridge_> saw a poster abut that kind of optimization with intel chips on november, leading to the same data leak 16:29 < bridge_> last hardware is probably even worse 16:29 < bridge_> they keep introducing new vulnerabilities 😄 16:29 < bridge_> @chairn if I buy u a modern gaming PC. Would u use it? 16:30 < bridge_> no 16:30 < bridge_> So u intentionally use slow hardware? 16:30 < bridge_> Gaming laptop 16:30 < bridge_> As alt 16:30 < bridge_> rly? 16:30 < bridge_> compiling ddnet in 20 secs is nice 16:30 < bridge_> In 6 even nicer 16:31 < bridge_> on release 16:31 < bridge_> can u compile it in 6? xd 16:31 < bridge_> i should try compiling ddnet on a tmpfs too 16:31 < bridge_> maybe it speeds up 16:31 < bridge_> Haven't tried lately. But the 16 core upgrade boosted it again xd 16:31 < bridge_> xd 16:32 < bridge_> @jupeyy_keks @heinrich5991 btw im interested in this, do you introduce extensive testing on ur ddnet rust projects? property testing? 16:32 < bridge_> i rly wonder if game dev big houses do some kind of automated testing on gameplay 16:33 < bridge_> specially for games with precise physics 16:33 < bridge_> I'm leaning a bit on libtw2 for reimplementing the teeworlds 0.6 protocol. that has lots of testing and also quickcheck-like testing 16:33 < bridge_> probably not 16:33 < bridge_> we could do that with @zwelf2's teehistorian replayer at some point 🙂 16:33 < bridge_> that would be kind of like differential testing? 16:34 < bridge_> I introduced testing for network mostly. Else only smaller tests for either how I want to use libs or libs I introduced 16:34 < bridge_> https://en.wikipedia.org/wiki/Differential_testing 16:35 < bridge_> ``` 16:35 < bridge_> CPU vulnerabilities found in the past few years: 16:35 < bridge_> 16:35 < bridge_> https://en.wikipedia.org/wiki/Meltdown_(security_vulnerability) 16:35 < bridge_> https://en.wikipedia.org/wiki/Spectre_(security_vulnerability) 16:35 < bridge_> https://aepicleak.com/ 16:35 < bridge_> https://en.wikipedia.org/wiki/Software_Guard_Extensions#SGAxe 16:35 < bridge_> https://en.wikipedia.org/wiki/Software_Guard_Extensions#LVI 16:35 < bridge_> https://en.wikipedia.org/wiki/Software_Guard_Extensions#Plundervolt 16:35 < bridge_> https://en.wikipedia.org/wiki/Software_Guard_Extensions#MicroScope_replay_attack 16:35 < bridge_> https://en.wikipedia.org/wiki/Software_Guard_Extensions#Enclave_attack 16:35 < bridge_> https://en.wikipedia.org/wiki/Software_Guard_Extensions#Prime+Probe_attack 16:35 < bridge_> https://www.vusec.net/projects/crosstalk/ 16:35 < bridge_> https://en.wikipedia.org/wiki/Hertzbleed 16:35 < bridge_> https://www.securityweek.com/amd-processors-expose-sensitive-data-new-squip-attack/ 16:35 < bridge_> ``` 16:35 < bridge_> xd 16:36 < bridge_> Testing in async is really hard xd 16:37 < bridge_> @ryozuki where do u want to introduce tests? 16:37 < ChillerDragon> heinrich be like 16:38 < ChillerDragon> "I use duckduckgo BTW" 16:38 < bridge_> personally everywhere, but specially physics 16:38 < bridge_> aka tests where the code affects the users the most directly 16:39 < bridge_> https://observablehq.com/@ayhanfuat/the-fall-of-stack-overflow 16:40 < bridge_> Well gonna be interesting how hard it is to pass zwelfs tests unintentionally ^^ he said he added like 400 tests 16:40 < bridge_> zwelf is a true ruster 16:40 < bridge_> i actually never checked the code from his repo 16:40 < bridge_> gitlab rly makes it invisible a lot for me sadly 16:40 < bridge_> He's the most active silent dev xd 16:40 < bridge_> xd 16:41 < bridge_> im also active dev, just at work 16:41 < bridge_> kek 16:41 < bridge_> well i got a small side project now, ddstats 16:42 < bridge_> Total players: 3338 16:42 < bridge_> nice 16:42 < bridge_> I'll do the 1 quadrillion moddablest tw version ever 16:42 < bridge_> It's finished before the universe dies 16:42 < bridge_> I'll do the 1 quadrillion fps moddablest tw version ever 16:42 < ChillerDragon> > Total players: 3338 16:42 < ChillerDragon> all bots 16:43 < bridge_> urs 16:43 < ChillerDragon> i hooman 16:43 < bridge_> with 8 clients 16:43 < ChillerDragon> no proof 16:43 < bridge_> 200% Tru tho 16:47 < bridge_> what should I say to be neutral on that? 16:48 < bridge_> i guess just say ur search result said that 16:48 < bridge_> but i dont mind u saying ddg xd 16:49 < ChillerDragon> I am vegan BTW 16:49 < bridge_> Chillig 16:50 < bridge_> nice if you are, not nice if this is supposed to be the "how do you know someone is vegan" joke 16:50 < ChillerDragon> its the latter 16:50 < ChillerDragon> if in doubt its me being offensive 16:50 < bridge_> ChillerDragon I'm vegan bte 16:51 < ChillerDragon> you are? 16:51 < bridge_> Btw 16:51 < bridge_> Yes 16:51 < ChillerDragon> pog 16:51 < bridge_> 😬😬 16:51 < ChillerDragon> ye im not ._. 16:51 < bridge_> Shame 16:51 < bridge_> Boomer 16:51 < ChillerDragon> indeed 16:54 < bridge_> But I bet I ate a fly lately. They fcking everywhere xd 16:54 < ChillerDragon> eww 16:55 < bridge_> Free protein 16:55 < ChillerDragon> https://c.tenor.com/CNKi8EAo15MAAAAM/frog-eating.gif 16:55 < ChillerDragon> jopsti face reveal 16:55 < bridge_> Froggers 16:55 < bridge_> jfroggers 16:55 < ChillerDragon> xd 17:05 < bridge_> @jupeyy_keks do u eat eggs 17:06 < ChillerDragon> thats not vegan... 17:06 < ChillerDragon> ryo trol 17:07 < bridge_> I'm not sure in what context you mean it. But always no xd 17:07 < bridge_> wanted to know if u vegetarian instead 17:07 < bridge_> xd 17:07 < bridge_> I didn't like eggs ever anyway xD 17:08 < ChillerDragon> eggs s good what can i say 17:08 < bridge_> Nowadays it's not even hard to be vegan xd. One trillion alternatives. 17:08 < bridge_> 17:08 < bridge_> 10 years ago it was much harder xd 17:10 < bridge_> Thanks hipster kids xd 17:12 < ChillerDragon> omagawd @heinrich5991 can we discuss redirects? 17:12 < bridge_> sure 17:12 < ChillerDragon> pog 17:12 < bridge_> @kaffeine0 17:13 < ChillerDragon> why u insisting on dropping the connection? 17:13 < ChillerDragon> ok whatever i dont care my client will be up to date i will be redirected 17:13 < ChillerDragon> if it bothers me on my server i can change the code 17:13 < bridge_> correct 17:13 < ChillerDragon> ok then if it has to happen how? 17:14 < ChillerDragon> how do i ensure the client is not dropped before it got the redirect message? 17:14 < bridge_> check if the client is too old 17:14 < ChillerDragon> no i mean 17:14 < ChillerDragon> if its recent enough 17:14 < bridge_> if so, drop the client right after sending the (flushed) redirect message 17:14 < bridge_> if it's recent enogh 17:14 < ChillerDragon> so flushing it and then dropping 17:14 < ChillerDragon> thats all the gurantee there is? 17:14 < bridge_> send the redirect message and wait for the client to drop the connection 17:15 < bridge_> if the client sends further messages, ignore them completely 17:15 < ChillerDragon> so drop the connection after a timeout? 17:15 < bridge_> I'd be fine without the timeout 17:16 < ChillerDragon> will it? 17:16 < ChillerDragon> its udp after all 17:16 < bridge_> that's the same as normal disconnects then 17:16 < ChillerDragon> you mean kicks? 17:17 < ws-client> chillerdragon: it sounds like what you want to ask is: are disconnects vital 17:17 < ChillerDragon> https://paste.zillyhuhn.com/28 17:17 < ChillerDragon> this doesnt even work on local server 17:17 < ChillerDragon> it just kicks me 17:17 < ChillerDragon> i cant imagine what happens on a laggy unstable connection 17:18 < ChillerDragon> i just do not want to drop clients that did not get the info where to redirect to 17:18 < bridge_> make em vital? 17:18 < ChillerDragon> doesnt help 17:18 < ChillerDragon> if the client is dropped 17:18 < ChillerDragon> vital wont be resend 17:18 < ChillerDragon> will it? 17:19 < bridge_> that's why I say the server doesn't drop the client if it's recent enough(?) 17:19 < bridge_> read above 17:19 < ChillerDragon> thats not what i talk about 17:19 < ChillerDragon> i talk about packet loss 17:19 < ChillerDragon> for recent clients 17:19 < bridge_> yes 17:19 < ws-client> lmao 17:19 < ws-client> i think u both missunderstand each other 17:19 < ChillerDragon> me too? 17:19 < ws-client> chiller: dont disconnect at all 17:19 < ChillerDragon> omagwd 17:19 < bridge_> I'm saying with the above solution, there's no problem 17:19 < ws-client> if client is new 17:20 < bridge_> the client disconnects 17:20 < ChillerDragon> o 17:20 < ChillerDragon> only disconnect old`? 17:20 < bridge_> yes 17:20 < ChillerDragon> why do you even want to drop it in the first place then? 17:21 < bridge_> thanks Jupstar 17:21 < ChillerDragon> yes jupsti pro 17:21 < bridge_> because old clients won't disconnect automatically 17:21 < bridge_> I want the protocol to work similarly for old and new clients 17:22 < ChillerDragon> okay 17:22 < ws-client> chillerdragon: sneak a Kick("Update client") in ;) 17:22 < ChillerDragon> i see so kick is backwards compability i like it 17:22 < ChillerDragon> gonna use the same in my 0.7 pr 17:22 < bridge_> yes, the kick is backcompat 17:22 < ChillerDragon> kick those 0.6 clients 17:22 < ChillerDragon> axaxaxa 17:23 < ChillerDragon> ok nvm the code snippet i sent would even redirect on localhost i was just using an old client :D 17:24 < ChillerDragon> so and waddup the cookies now? 17:24 < ChillerDragon> opaque id? 17:24 < ChillerDragon> DHL tracking number 17:24 < ChillerDragon> and password 17:24 < bridge_> are you interested in implementing it? 17:24 < ChillerDragon> i want cookies! 17:24 < bridge_> I don't feel like it's absolutely necessary tbh 17:25 < ChillerDragon> but cookies should be unrelated to redirect 17:25 < ChillerDragon> they should be sent on any connect regardless of redirect 17:25 < ChillerDragon> what i want is redirecting to other ips ._. 17:25 < ChillerDragon> to send jopsti to russia 17:25 < bridge_> why do you want to redirect to other IPs? 17:26 < ChillerDragon> if i switch hoster for example 17:27 < ChillerDragon> or load balance tees around on a multi server cluster 17:27 < bridge_> that seems out of reach tbh 17:27 < ChillerDragon> heck redirecting to url would even better 17:27 < ChillerDragon> to enforce 0.7 connection for all hehe 17:27 < ChillerDragon> @heinrich5991 out of wat 17:27 < ChillerDragon> how? 17:28 < ChillerDragon> i dont get the security risk 17:28 < bridge_> load-balancing tees 17:28 < bridge_> multiserver cluster 17:28 < ChillerDragon> seems like a useless limitation 17:28 < ChillerDragon> on servers that require less state or where state can be sent across network that is totally reachable 17:29 < bridge_> one of the things that just came to my mind regarding 'security', I think the target server should probably communicate that it wants to receive the new clients 17:29 < ChillerDragon> i mean using CSaveTee you can basically send player state over the network and sync across servers 17:29 < bridge_> one of the things that just came to my mind regarding 'security', I think the target server should probably communicate that it wants to receive redirects from that address 17:29 < ChillerDragon> cross origin policy 17:29 < ws-client> xD 17:30 < ChillerDragon> do you want to load tees from that url? 17:30 < ws-client> i dont want to be redirected to teep*rn.com 17:30 < ChillerDragon> you have no rights 17:30 < ChillerDragon> pst 17:30 < ChillerDragon> then do not connect to chillertrap.com in the first place 17:30 < ChillerDragon> its known associate 17:31 < ChillerDragon> @heinrich5991 i see to avoid some kind of unwanted flood to servers that dont want any visitors 17:31 < ws-client> what i dislike is that i might be redicrected to a different country 17:31 < ChillerDragon> such as area51.nsa.gov 17:31 < ChillerDragon> you can possibly get users in legal trouble 17:31 < ChillerDragon> if you make them send udp packets 17:31 < ChillerDragon> to blacklisted ips 17:32 < ChillerDragon> phew idk how to solve that 17:33 < ChillerDragon> thats why i gotta just slap my whole network behind a proxy that is one unfied ip and routes all traffic under the hood to wherever i want @heinrich5991 rip your proxy support 17:33 < bridge_> do you run more than one server atm? 17:33 < ChillerDragon> ok no redirect to custom ips for now 17:34 < ChillerDragon> i dream big and act little 17:34 < bridge_> cookies? 17:34 < bridge_> chiller implementing a tee browser 17:34 < ChillerDragon> but i had two 2vps connected to one teeworlds kubernetes cluster for 1 day xd 17:34 < ChillerDragon> cookies pog 17:34 < ChillerDragon> for tee tracking 17:34 < ChillerDragon> and auto login 17:34 < ChillerDragon> and facebook ads 17:35 < bridge_> "cookies" will come with accounts anyway 17:35 < ChillerDragon> generically usable for 3rd parties? 17:35 < bridge_> ye 17:35 < ChillerDragon> pls no GDPR popups on connecting to servers 17:35 < ChillerDragon> THIS SERVER USES COOKIES! 17:36 < bridge_> you do know that this is only related to cookies not necessary for the service's operation 17:36 < ChillerDragon> ok amazing so jopsti will code cookie feature for me 17:36 < ChillerDragon> i know nothing 17:36 < ChillerDragon> thats what i have lawyer jopsti for 17:37 < ChillerDragon> he covers all my legal concerns 17:37 < bridge_> chiller legit has a company of unpaid employees, all from this chat 17:37 < bridge_> im his google 17:37 < ChillerDragon> xd 17:37 < bridge_> i want my teecoins 17:37 < ChillerDragon> *sends some teecoins* 17:45 < ChillerDragon> jopsti do u know how to hac opengl? 17:45 < ChillerDragon> https://paste.zillyhuhn.com/Es 17:46 < ChillerDragon> i did arch update after decades of procrastination 17:46 < ChillerDragon> and am too lazy to reboot 17:47 < ChillerDragon> ok watever i reboot xd 17:47 < bridge_> just reboot 17:48 < bridge_> kill xserver xd 17:50 < bridge_> doesn't work 17:50 < bridge_> driver/library mismatch 17:50 < bridge_> chillerdragon: speaking of laziness. Would you rather update a cert, or learn the public key fingerprint by heart xd 17:50 < ChillerDragon> fair point 17:50 < bridge_> context? ^^ 17:51 < ChillerDragon> context is the hotfix is harder than the fix 17:52 < bridge_> a cert today expired, apparently certbot was down for some reason. but i wasnt at my computer, so i just copied the fingerprint and checked it later, in good faith that i'm currently not attacked xdd 17:52 < bridge_> 17:52 < bridge_> now that i learned it, i dont need to even start certbot anymore 17:53 < ChillerDragon> xd 17:53 < bridge_> Yet another bizarro chiller use 😄 17:54 < bridge_> When request ip over avian carrier support for ddnet? 17:55 < ChillerDragon> omagawd lerato 17:55 < bridge_> ur biggest fan 17:56 < bridge_> waste 64*8 bytes for each big snap? 17:56 < bridge_> probably a bit much 17:57 < ChillerDragon> is this a reply? 17:57 < ChillerDragon> who wastes bytes? 17:57 < bridge_> I'm writing the dissector 17:58 < ChillerDragon> ? 17:58 < bridge_> there is no reply 17:58 < ChillerDragon> ou you do snap dissecting? o.O 17:58 < ChillerDragon> very pog 17:58 < bridge_> I was considering to waste 64*8 bytes for each snap spanning at least two messages 17:58 < bridge_> currently only snap reassembly 17:59 < bridge_> not dissecting the actual snaps 17:59 < ChillerDragon> so the snap it self isnt wasting 17:59 < ChillerDragon> your implementation is? 17:59 < bridge_> I was considering an implementation that's a bit more wasteful 17:59 < bridge_> but I decided to do it another way 17:59 < ChillerDragon> just tell your users to buy more ram 18:00 < bridge_> I'm a user 18:00 < bridge_> can't buy more ram for this machine 18:00 < ChillerDragon> why not? 18:00 < bridge_> soldered 18:00 < ChillerDragon> your laptop? 18:00 < bridge_> yes 18:00 < ChillerDragon> hm 18:01 < ChillerDragon> just add a new section on ddnet.org/donations 18:01 < ChillerDragon> development hardware 18:01 < ChillerDragon> and buy new laptop 18:01 < ChillerDragon> im half kidding 18:02 < bridge_> should be full kidding 18:02 < ChillerDragon> ok i kicked clients wiht current or older version in #6914 idk if thats the way to be back compat 18:02 < chillerbot> https://github.com/ddnet/ddnet/issues/6914 18:03 < ChillerDragon> all other version checks seem to predict the new release number already 18:04 < bridge_> please don't start a new standard ^^ 18:04 < ChillerDragon> ok then explain the old 18:04 < bridge_> predict the next version number :p 18:04 < ChillerDragon> rly? 18:05 < ChillerDragon> is that how its done? 18:05 < bridge_> 17.2 18:05 < ChillerDragon> lmao 18:05 < bridge_> According to the law of jupsti you should just get an entire new machine in that case 18:05 < ChillerDragon> agree 18:05 < ChillerDragon> jopsti law good 18:05 < bridge_> sounds bad for the environment 18:06 < ChillerDragon> omg vegan heinrich be like 18:06 < bridge_> just buy a new environment 😉 18:06 < ChillerDragon> yes! 18:06 < bridge_> But it's great for streamlining development. Only target the very bleeding edge and infinite ram/storage/network bitrate 18:06 < ChillerDragon> just turn your old laptop into a server and keep using it 18:06 < bridge_> ChillerDragon: the impl of the thing should probbaly go into the engine 18:06 < ChillerDragon> the server side? 18:06 < bridge_> yes 18:06 < ChillerDragon> instead of gamecontext? 18:06 < ChillerDragon> u say that now? 18:06 < ChillerDragon> after hours of engineering 18:06 < ChillerDragon> omahgawd 18:06 < bridge_> 😄 18:06 < ChillerDragon> oke 18:07 < bridge_> it's a system message after all 18:07 < ChillerDragon> ok mr protocol designer 18:07 < bridge_> Just unengineer it and you'll get your time back 18:07 < bridge_> and you also need the engine to ignore all further messages from a client 18:07 < ChillerDragon> this is exhausting 18:07 < ChillerDragon> whats that supposed to mean 18:09 < bridge_> above `int GameFlags = 0;`, check whether the client was redirected already and ignore the packet in that case 18:09 < bridge_> in src/engine/server/server.cpp 18:10 < ChillerDragon> so also for new clients? 18:10 < ChillerDragon> i thought only kick old clients? 18:10 < bridge_> only needed for new clients 18:10 < bridge_> yes 18:10 < ChillerDragon> omg 18:10 < ChillerDragon> we going circles 18:11 < ChillerDragon> is this really needed? 18:11 < ChillerDragon> and for what? 18:11 < bridge_> old clients: send redirect message, kick. 18:11 < bridge_> new clients: send redirect message, ignore further messages from the client. 18:11 < ChillerDragon> why ignore from new clients 18:11 < ChillerDragon> and for how long? 18:11 < bridge_> forever 18:11 < ChillerDragon> so lifeban all redirected clients? 18:11 < bridge_> lifeban? 18:11 < bridge_> only ignore messages on that connection 18:12 < bridge_> they can disconnect and reconnect immediately if they want to 18:12 < bridge_> but they can't continue using that connection 18:12 < ChillerDragon> how does one even identify a connection 18:12 < bridge_> client ID? 18:12 < bridge_> you have a client struct 18:12 < ChillerDragon> isnt that reused 18:12 < ChillerDragon> so i add a bool in the struct? 18:12 < ChillerDragon> hmkay 18:12 < ChillerDragon> bloat imo 18:12 < ChillerDragon> but whatever 18:13 < bridge_> checking for correct utf-8 is probably also "bloat" in your opinion? ^^ 18:13 < bridge_> that has led to some wonderful exploits, even some in TW ^^ 18:14 < bridge_> the idea is that you specify your protocol very precisely 18:14 < ChillerDragon> you should push spec driven development 18:14 < ChillerDragon> writing a teeworlds protocol standard first and discussing that 18:14 < ChillerDragon> and then the implementation 18:15 < ChillerDragon> for a 2d kirby shooter 18:15 < ChillerDragon> axaxaxax 18:16 < bridge_> the 2d kirby shooter that's casually DoSed 18:16 < bridge_> probably needs some consideration for protocols 18:23 < ChillerDragon> what is the difference between kick and dropping traffic? 18:24 < ChillerDragon> im thinking about calling ``DelClientCallback`` instead of bloating in a new ignore message concept 18:24 < bridge_> arent u the one that wrote the doc for whole protocol xd 18:24 < ChillerDragon> yes! 18:24 < bridge_> the thing you also wanted to have. re-delivery of the redirect message 18:24 < ChillerDragon> yes! 18:24 < ChillerDragon> dont you think that makes sense? 18:25 < bridge_> yes 18:25 < bridge_> chillerdragon will be the first to add a client that can ignore redirects xdd 18:25 < bridge_> that's the answer to "what's the difference between kick and dropping traffic" 18:25 < ChillerDragon> axaxax jopsti tru 18:25 < ChillerDragon> i see 18:36 < ChillerDragon> interesting dropping there just gets my client stuck 18:36 < ChillerDragon> no connection problems 18:36 < ChillerDragon> i guess the server still sends keepalive lemme check 18:37 < ChillerDragon> https://paste.zillyhuhn.com/Ep 18:37 < ChillerDragon> hmm 18:37 < ChillerDragon> is this dropped enough for you @heinrich5991 ? 18:37 < ChillerDragon> it seems super bugged 18:38 < bridge_> do you guys think ddnet is a good start to a novice programmer? 18:38 < ChillerDragon> i really dont wanna see some clients end up in that state seems like a debugging nightmare 18:38 < ChillerDragon> i started with ddnet and i turned out great 18:38 < ChillerDragon> or did i 18:39 < ChillerDragon> @Iza c++ has more pitfalls and annoyances for beginners than python or other high level scripting languages 18:39 < bridge_> if u code for yourself its probably ez to get some stuff working without lot of effort 18:39 < bridge_> but honestly the codebase isnt really built for intuitivity 18:39 < ChillerDragon> getting it to compile is 70% already 18:40 < bridge_> Yeah i heard python is great for beginners, but c++ is just more useful for more stuff, isnt it? 18:40 < ChillerDragon> depends 18:40 < ChillerDragon> why do you want to code anyways 18:40 < bridge_> i mean generally: if u start coding with small games.. u directly see (visually) what you did.. that's quite nice 18:40 < bridge_> It'll make you a decent programmer though, if you manage to understand ddnet you are pretty decent 😄 18:40 < bridge_> python is never great lmao 18:40 < bridge_> it's like purposely going into hell 18:40 < ChillerDragon> pygame pog 18:40 < bridge_> no, I don't think c++ is useful for more stuff than python 18:40 < ChillerDragon> idk actually never tried it 18:41 < bridge_> I thought I may can learn something that’s going to be useful for my future 18:41 < bridge_> Or rather the future of jobs 18:41 < ChillerDragon> what do you want to do in your future? 18:41 < bridge_>