01:15 < bridge> @furo321 we are such late night coder's its insane 01:18 < bridge> 🚀 01:18 < bridge> :gigachad: 01:22 < bridge> pog 03:56 < bridge> Oh, sorry 03:59 < bridge> I found that there are many robots of "(connecting)" in registered servers, did it 03:59 < bridge> I found that there are many robots of "(connecting)" in registered servers, was it DDOS? 04:03 < bridge> And I found that there are many robots of "(connecting)" in registered servers, was it DDOS? 04:03 < bridge> And I found that there are many robots of "(connecting)" in registered servers at 0:00, was it DDOS? 04:03 < bridge> And I found that there are many robots of "(connecting)" in registered servers at 0:00 (UTC+8), was it DDOS? 04:58 < bridge> Fake player you can ban without problems 05:04 < bridge> im alive 06:07 < ws-client> daily bash quiz! ``x=1;ls | x=2;echo "$x"`` 06:07 < ws-client> a) error 06:07 < ws-client> b) 1 06:07 < ws-client> c) 2 06:12 < ws-client> @meloƞ i got you bro https://zillyhuhn.com/crack/bam4 07:01 < ws-client> @milkeeycat lmao nice https://github.com/MilkeeyCat/deez_bridge 08:41 < bridge> ❤️❤️❤️ 08:58 < bridge> chillerdragon: im proud of that name 09:00 < bridge> So far i spent most of the time to find a working irc docker image 09:24 < ws-client> eww why docker @milkeeycat 09:25 < ws-client> you mean to run a server for testing? or to run your client lib in a container? 09:25 < bridge> i aint installing real irc server on my machine 09:25 < ws-client> why not? 09:25 < bridge> coz that shit is gonna install smth in /etc 09:25 < bridge> i can just run docker compose up and it will work 09:25 < ws-client> watfak no? 09:26 < ws-client> docker and docker compose will put stuff in /etc 09:26 < ws-client> docker is the worst 09:26 < ws-client> just run https://github.com/jrosdahl/miniircd 09:26 < ws-client> at least use podman :p 09:26 < ws-client> you can also test stuff on quakenet 09:26 < bridge> :banhammer: 09:27 < bridge> it already can send messages 😄 09:27 < ws-client> thats poggies 09:27 < bridge> so probably i can bridge my discord sever and #ddnet-off-topic 09:27 < ws-client> soon new bridge for #developer <:poggers2:1008007455936094328> 09:28 < bridge> btw 09:28 < bridge> i was confused about how irc works 09:28 < bridge> so basically u join a server 09:28 < bridge> quakenet for example 09:28 < bridge> and u can make own channel 09:28 < ws-client> yes 09:28 < bridge> like `/join #foo` 09:28 < ws-client> yes 09:29 < bridge> does it get removed when u leave? 09:29 < ws-client> define removed xd 09:29 < ws-client> idk 09:29 < ws-client> whats the different between removed and empty? 09:29 < ws-client> you can also create your self a channel on chat.zillyhuhn.com here https://github.com/discord-irc/irc-websockets/blob/91135aaa4e2062dbfd1e85cb0113b1f3558feb27/src/bridge_connections.ts 09:29 < bridge> ok, another question does #ddnet-off-topic always have anyone in it? 09:30 < ws-client> the client of chat.zillyhuhn.com is mostly in there and my bouncer 09:30 < ws-client> but its same vps so on reboot its empty 09:31 < ws-client> @milkeeycat idk much about irc tbh. But i think on quakenet the first joiner of a channel can take some ownership. And when he rejoins he can request that from Q bot again 09:32 < ws-client> might be that if the channel gets fully empty for a while that someone else can claim ownership again idk 09:32 < ws-client> not sure if the irc spec even has such things defined. I could image that depends on your network 09:44 < bridge> is irc cool again 09:45 < bridge> i just wanted to make something in go 09:47 < ws-client> yea im workin on the backend rn 09:47 < bridge> i connect my bot in 10 tries ;-; 09:48 < bridge> why go 09:48 < bridge> isn't this a rust household 09:48 < bridge> i like go 09:48 < ws-client> go daddy 09:48 < bridge> fukc that shit :justatest: 09:49 < ws-client> erhm wtf is chat.zillyhuhn.com missing half the messages?! 09:49 < bridge> this 09:49 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1202173722430021652/3KVIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIAiCIP8DzVydYHOQvG4AAAAAElFTkSuQmCC.png?ex=65cc7e86&is=65ba0986&hm=0d16b57edef7aadb40b8d17ca6f53c14547f9f9b8e38cad29fd3414a15d31473& 09:49 < ws-client> yes 09:49 < bridge> i opened their site once and i was getting their ads in british for weeks 09:49 < bridge> chillerdragon: send token 09:49 < ws-client> https://zillyhuhn.com/cs/.1706690967.png 09:50 < ws-client> @TsFreddie token is: nimrocks 09:50 < ws-client> Very cool 09:50 < ws-client> @TsFreddie its bring your own security approach. I store passwords in plaintext <:justatest:572499997178986510> 09:50 < ws-client> so make sure to hash them on your side :p 09:50 < bridge> @tsfreddie checkout that coll ass emoji picker :santatrollet: 09:50 < bridge> oops, too late 09:50 < ws-client> rip 09:50 < bridge> @tsfreddie checkout that cool ass emoji picker :santatrollet: 09:51 < bridge> have fun with my bank account 09:51 < ws-client> ye @TsFreddie if u do : and tab that thing is built by milkeey its really pog 09:51 < ws-client> eh i mean : (and a letter) 09:52 < ws-client> <:justatest:572499997178986510> 09:52 < ws-client> i don't understand 09:52 < ws-client> nob 09:52 < bridge> isnt it just like in discord? 09:53 < bridge> idk, tab does nothing for me 09:53 < ws-client> ye i sent wrong instructions it used to be like that before 09:53 < ws-client> its : and a letter 09:53 < ws-client> do like :j 09:53 < ws-client> :j 09:53 < ws-client> xd 09:53 < ws-client> wtf 09:53 < ws-client> ah its off by default omg @MilkeeyCat 09:53 < ws-client> why u do dat btw xd 09:53 < ws-client> @TsFreddie bottom left settings 09:53 < ws-client> turn it on 09:53 < ws-client> :j 09:53 < ws-client> reload the page 09:54 < ws-client> xd 09:54 < bridge> bruh, too many instructions 09:54 < ws-client> ikr 09:54 < ws-client> bajillion instructions to send an emoji xDDD 09:54 < ws-client> @MilkeeyCat when fix xd 09:54 < ws-client> ye it works now 09:54 < ws-client> anyway back to work now 09:54 < ws-client> @ChillerDragon doesnt it work as needed? 09:54 < ws-client> its off for newcomers 09:54 < ws-client> do we want that? 09:54 < ws-client> oh 09:55 < ws-client> i dont even know where to turn it on xD 09:55 < bridge> i mean do you intend to have more newcomers 09:55 < ws-client> no so its not too dramatic 09:55 < ws-client> lmao @MilkeeyCat 09:55 < ws-client> ok gtg bye 09:55 < bridge> bye 10:09 < bridge> Why client function "Localize" doesn't get warning -Wformat-nonliteral? 10:26 < bridge> @furo321 chill Out man, leave some for the Rest of us :justatest: 10:34 < bridge> Localize function is not using printf-like functions, but it's simulating their behavior. 10:34 < bridge> 10:34 < bridge> ``` 10:34 < bridge> GNUC_ATTRIBUTE((format_arg(1))); 10:34 < bridge> Localize function is not using printf-like functions, but it's simulating their behavior. 10:34 < bridge> 10:35 < bridge> ```extern const char *Localize(const char *pStr, const char *pContext = "") 10:35 < bridge> 10:35 < bridge> GNUC_ATTRIBUTE((format_arg(1))); 10:35 < bridge> ``` 10:35 < bridge> Localize function is not using printf-like functions, but it's simulating their behavior. 10:35 < bridge> 10:35 < bridge> ```cpp extern const char *Localize(const char *pStr, const char *pContext = "") 10:35 < bridge> 10:35 < bridge> GNUC_ATTRIBUTE((format_arg(1))); 10:35 < bridge> ``` 10:35 < bridge> Localize function is not using printf-like functions, but it's simulating their behavior. 10:35 < bridge> 10:35 < bridge> ```cpp 10:35 < bridge> extern const char *Localize(const char *pStr, const char *pContext = "") 10:35 < bridge> 10:35 < bridge> GNUC_ATTRIBUTE((format_arg(1))); 10:35 < bridge> ``` 10:39 < bridge> OK now 10:40 < bridge> thanks 11:22 < bridge> Why not we also show the game mode on steam friend list? 11:22 < bridge> is it really needed 11:24 < bridge> I think so 11:25 < afterfx> test 12:21 < bridge> hi from eu 12:35 < bridge> time to upgrade 1 week old gentoo 12:41 < bridge> hi 12:42 < bridge> the guy with the egpu still didn’t respond 13:15 < bridge> @learath2 TIL if u dont update for long its recommended to pass --backtrack=1000 or smth high to portage 13:15 < bridge> Yep, it's a good idea 13:16 < bridge> ah there is also --keep-going 13:16 < bridge> @learath2 is it related? 13:16 < bridge> Not exactly, but it can help when you have some dependency update loop 13:17 < bridge> should i add backtrack=10 to my daily upgrade script? xcd 13:17 < bridge> the one i run when i wake up 13:18 < bridge> No point to it if you are running it daily 13:18 < bridge> ok 13:18 < bridge> morning routine: 13:18 < bridge> - coffee 13:18 < bridge> - boot up gentoo 13:18 < bridge> - upgrade 13:18 < bridge> - gym 13:18 < bridge> :owo: 13:19 < bridge> went to sleep at 7 and woke up 15 mins ago today 13:19 < bridge> feels bad 13:19 < bridge> You only really need to bump up backtrack if something breaks 13:19 < bridge> ok 13:20 < bridge> Wher breakfast? 13:28 < bridge> after gym 13:28 < bridge> well i take some protein with my coffee 13:28 < bridge> but its better to not eat 13:28 < bridge> before gym 13:28 < bridge> so u dont puke or smth 13:29 < bridge> also its odd but i never feel like having breakfast before xd 13:33 < bridge> i prefer to eat more at dinner 13:43 < bridge> Understandable. Working out with a full stomach is very meh 15:01 < bridge> from kernel 6.1 to 6.6 ez 15:01 < bridge> new allocator :poggers2: 15:01 < bridge> SLUB 15:23 < bridge> chiller thanks for recommendation: 15:23 < bridge> 15:23 < bridge> deflate: 415078 - 415013, gz: 415096 - 415031, zlib: 415084 - 415031, lz4: 1124876, brotli: 165840, zstd: 228700 - 191730 15:23 < bridge> 15:23 < bridge> brotli wins against everything in my use case, even when i set compression to best 15:23 < bridge> how does it manage to be so good? 15:23 < bridge> i should add lzma too 😄 15:24 < ws-client> wowo 15:24 < ws-client> someone mentioned it in a tech talk i went to o.O 15:24 < ws-client> didnt expect it to be actually good xd 15:36 < bridge> deflate: 415078 - 415013, gz: 415096 - 415031, zlib: 415084 - 415031, lz4: 1124876, brotli: 165840, zstd: 228700 - 191730, lzma: 179541 - 178466 15:36 < bridge> 15:36 < bridge> it even wins against lzma 15:36 < bridge> 15:36 < bridge> lmao, i didnt expect that 15:41 < bridge> Which function adds quotes before and after nickname if there are spaces in nickname? (When search nickname in chat by clicking ``tab``) 15:43 < bridge> lmao I was looking at this yesterday. May you tell me speed wise if it's even slower than lzma? Or is it actually decently fast (compared to gz or zlib) 15:44 < bridge> I believe brotli uses a pretty big dictionary of a lot of common words/terms, which I believe can help quite a lot if you're benchmarking with small files. I think you can create custom dictionaries with zstd as well, but it doesn't have one by default. 15:45 < bridge> 15:45 < bridge> ty <3 15:46 < bridge> it's as fast as deflate algorithms, so basically the fastest around 15:46 < bridge> wtf. actually insane 15:49 < bridge> interesting that this improves it so much, bcs my data is fully binary, no text ^^ 15:49 < bridge> 15:49 < bridge> but maybe repeating patterns match very well tho 15:50 < bridge> Oh, binary data? In that case, I don't think it helps at all. 15:51 < bridge> well i could probably test a custom dictionary at some point, but dunno if i really want to have one 15:51 < bridge> defeats general purpose 15:51 < bridge> We use it for cold-storage backups 15:51 < bridge> Original Size 15:51 < bridge> 788.10MiB 15:51 < bridge> Compressed size 15:51 < bridge> 113.71MiB 15:52 < bridge> We use it (brotli) for cold-storage backups 15:53 < bridge> for tw demos i used lzma2 at max for now, i should try and see if brotli is better for that too 😄 15:55 < bridge> the uncompressed size is 206437961 btw. so the data is mostly empty, else the compression ratio wouldnt be so insanely high 😄 16:19 < bridge> Q: Have any of you used vcpkg in any actual project? 17:18 < bridge> i only remember my attempt to add dpp to vcpkg registry 17:18 < bridge> is that like some m$ apt 17:34 < bridge> @milkeeycat any particular reason why you did not use an existing project instead of building your own? https://github.com/MilkeeyCat/deez_bridge 17:35 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1202291115659624458/image.png?ex=65ccebdb&is=65ba76db&hm=85d7446de1f9e82066d56dfb8254fa73dac26348323056feb2c3b4563dba4179& 17:46 < bridge> it's m$ cargo 17:46 < bridge> yuzu uses it 17:52 < bridge> https://tenor.com/view/ni%C3%B1o-asustado-huh-what-meme-stare-gif-16752413 18:08 < bridge> https://www.reddit.com/r/europe/comments/1afghmc/pigss_comeback_spain_and_the_southern_countries/ 18:08 < bridge> lmao 18:08 < bridge> > PIGS's comeback: Spain and the southern countries are driving the Eurozone's economic growth against a stagnant Germany and France (in Spanish) 18:09 < bridge> > In the case of Spain, GDP increased by 0.6% in the last quarter of the year, while Italy grew by 0.2% and Portugal by 0.8%. These figures contrast with those of Germany (-0.3%), France (0%), Sweden (0.1%) and Austria (0.2%). What was observed in the final quarter of the year is part of a trend that has been observed since Russia invaded Ukraine in February 2022. While the German economy has remained frozen at its pre-war size, the southern countr 18:09 < bridge> > 18:09 < bridge> > Proof of this is that, compared to pre-Pandemic records, the GDP level of Spain, Portugal, Italy or Greece exceeds that of northern economies such as Germany, the Netherlands or Austria and also that of France. The Spanish economy is now 9.8% larger than in the fourth quarter of 2019 (the last one before the pandemic); Portugal's GDP is 10.8% higher, Italy's is 7.7% above and Greece's (no data yet for the fourth quarter) is 8.6% higher than then. 18:14 < bridge> Is it possible to make this show the difficulty of the map instead of the currently selected one? Kinda misleading when the loaded map is hard difficulty and easy is selected 18:14 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1202300785694871672/image.png?ex=65ccf4dc&is=65ba7fdc&hm=168b483ddcb1d5c9f2e3a3463f0d81fa9143a45c935cce3bf8e2c18875da34fb& 18:15 < bridge> this looks like a kog server 18:26 < bridge> any fix about the update failing? 18:26 < bridge> (instead of downloading the latest version & manually updatingà 18:26 < bridge> (instead of downloading the latest version & manually updating) 18:26 < bridge> no 18:26 < bridge> that's your only option 18:58 < bridge> rip 19:18 < bridge> https://github.blog/changelog/2024-01-30-github-actions-introducing-the-new-m1-macos-runner-available-to-open-source/ 19:18 < bridge> @heinrich5991 @deen we can have ci for macos m1 19:21 < bridge> is that interesting to us? 19:21 < bridge> not sure tbh 19:21 < bridge> it uses macos 14 19:22 < bridge> we can deprecate old macos <14 19:22 < bridge> which is the non m1 19:22 < bridge> and then use c++17 features (yes, macos <14 claims c++17 support but doesnt implement everything) 19:22 < bridge> that seems unrelated to the runner and more related to the minimum mac version we want to support, no 19:22 < bridge> that seems unrelated to the runner and more related to the minimum mac version we want to support, no? 19:23 < bridge> https://github.com/ddnet/ddnet/pull/7692 19:23 < bridge> i guess, but modern macos use m1 xd 19:23 < bridge> or arm 19:24 < bridge> macos is an insult to your precious life time 19:24 < bridge> it is 19:24 < bridge> i vote for removing macos support too 19:24 < bridge> and windows 19:24 < bridge> xDD 19:24 < bridge> but we live in a society 19:24 < bridge> :PES_Clown: 19:26 < bridge> according to https://store.steampowered.com/hwsurvey?platform=mac, afaict >25% of macs are still intel 19:26 < bridge> I'm surprised by how much non-intel macs there are already though 19:30 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1202320009938862100/GFHU2G3aUAAgMEA.png?ex=65cd06c4&is=65ba91c4&hm=f1c1fcae56a92c8da62173a539699dded3b4f68ccd1c09faae71423abedd674f& 19:32 < bridge> @ryozuki u should never be a fan of anyone or anything 19:32 < bridge> 19:32 < bridge> secret life tip 19:33 < bridge> Justar ftw 19:33 < bridge> @ryozuki let's start: 19:33 < bridge> 19:33 < bridge> Rust sucks 19:33 < bridge> xd 19:33 < bridge> Jupstar ftw 19:33 < bridge> who r you, random 19:33 < bridge> I'm a ghost! 19:34 < bridge> 👻 19:34 < bridge> well the unicode for it sucks xD 19:34 < bridge> 🥼🎩 19:34 < bridge> xd 19:34 < bridge> but ur hat is important 19:34 < bridge> yeah plus my cloth is from a labcoat ;) 19:45 < bridge> @robyt3 u here 19:46 < bridge> yes 19:46 < bridge> do you also have the feeling that strtoints 19:46 < bridge> https://github.com/ddnet/ddnet/blob/984457a48e112d27e51623a40c2971fa2fe5053a/src/game/gamecore.h#L83 19:46 < bridge> 19:46 < bridge> is buggy? 19:47 < bridge> it's missing a & 0xff after the +128 calculation 19:47 < bridge> bcs the other way around it does that 19:47 < bridge> https://github.com/ddnet/ddnet/blob/984457a48e112d27e51623a40c2971fa2fe5053a/src/game/gamecore.h#L96-L99 19:47 < bridge> for me it fails on this input: 19:47 < bridge> 19:47 < bridge> "Teleport fx�" 19:47 < bridge> is community a new word for hoster? 19:48 < bridge> Is that supposed to be valid? It contains a replacement character for me 19:48 < bridge> according to rust it is 19:48 < bridge> 😂 19:48 < bridge> but still, isnt it weird that we don't "clamp" the int? 19:48 < bridge> I'm not very sure about those functions, we don't have any tests for them right now 19:49 < bridge> it can "write" into the other bytes 19:49 < bridge> It uses `aBuf` which is `char` so I guess it doesn't need to clamp or AND? 19:49 < bridge> but it gets upcasted to int 19:50 < bridge> and then +128 can go out of range of unsigned char 19:50 < bridge> so in a worst case it "writes" into the other bytes 19:51 < bridge> or i don't understand smth important here 19:51 < bridge> mhh signed char vs unsigned char probably? 19:51 < bridge> well then it's probably correct 19:52 < bridge> Definitely confusing enough that we needed to silence a related compiler warning 19:52 < bridge> is there one? 19:52 < bridge> ah yeah 19:52 < bridge> the ifdef block xd 19:52 < bridge> `#pragma GCC diagnostic ignored "-Warray-bounds"` 19:52 < bridge> but that's above what i meant 19:52 < bridge> `char` is signed on most platforms 19:52 < bridge> but yeah it's simply c not defining char as signed or unsigned 19:53 < bridge> yup 19:53 < bridge> who in this world even decided that it's good that it's not defined 19:54 < bridge> it's just the C way to do things, I guess 19:55 < bridge> well i already hate that i have to deal with strtoints xddd 19:55 < bridge> that function is indeed weird 😦 19:55 < bridge> we inherited it, I think 19:55 < bridge> defs, it's used inside the map format 19:56 < bridge> @robyt3 do you know why there are maps with invalid utf8? 19:56 < bridge> 19:56 < bridge> like how did this even happen in first place? 19:56 < bridge> did they develop on ARM? XD 19:57 < bridge> I don't think the functions checks for valid UTF-8, so it might be one of the remaining way to get invalid UTF-8 into the client 19:57 < bridge> mh ok 19:57 < bridge> Initially UTF-8 wasn't used consistently when the filename comes from the OS, so ANSI filenames containing umlauts are considered as invalid UTF-8 19:58 < bridge> At least from the maps with invalid UTF-8 that Patiga found it looks like that 19:58 < bridge> yea, because we didn't have correct handling for non-ascii filenames on windows at first 19:59 < bridge> can patigas map tool print in json? 19:59 < bridge> @patiga can i serialize to json? 20:01 < bridge> most things yea 20:01 < bridge> not the entire thing to one json object though 20:01 < bridge> with which tool? 20:01 < bridge> there are so many 20:01 < bridge> `twmap-edit --mapdir INPUT_PATH OUTPUT_PATH` 20:02 < bridge> thats the "mapdir" format 20:02 < bridge> most things are json files 20:04 < bridge> ah interesting, it makes it structured 20:10 < bridge> its especially useful to diff maps 20:20 < bridge> @patiga next time you sleep, dream of zlib license. 20:20 < bridge> 20:20 < bridge> That would have saved me some headache xd 20:25 < bridge> :heartw: 20:27 < bridge> im a fan of the best current available language 20:27 < bridge> that sounds pragmatic 20:40 < bridge> has anyone here worked with postgres' inet/cidr datatypes before? 20:40 < bridge> haven't 20:41 < bridge> if anyone has, my question: is it possible to merge overlapping ip ranges to always have a minimal set of ranges? 20:41 < bridge> maybe postgres is the wrong tool 20:50 < bridge> I find the question underspecified 20:50 < bridge> you can test whether one cidr is a subset of another in postgres using `>>=` 20:50 < bridge> if so, you can drop the one that is a subset 21:01 < bridge> well if i'm not wrong a \0 char is encoded as 128 (bcs of the +128) 21:01 < bridge> 21:01 < bridge> but: 21:01 < bridge> https://github.com/ddnet/ddnet/blob/984457a48e112d27e51623a40c2971fa2fe5053a/src/game/gamecore.h#L89 21:01 < bridge> 21:01 < bridge> just strips it as zero 21:01 < bridge> 21:01 < bridge> but zero is not the same as \0 here, because of the +- 128 21:01 < bridge> 21:01 < bridge> 21:01 < bridge> i think that's why i get so many invalid utf8s in my case 21:01 < bridge> e.g. patigas tool strips the zeros before applying the +-128 offset 21:01 < bridge> ah yea, that bit is inconsistent 21:01 < bridge> you have to ignore the last byte 21:02 < bridge> Is there anyone here that knows their way around C++ templates? 21:02 < bridge> you 21:02 < bridge> what is a way around c++ templates? 21:02 < bridge> what's your question? 21:02 < bridge> I kinda ran out of tricks 😄 21:02 < bridge> i used them heavily in the past, but i got extremly bad in cpp :lol: 21:03 < bridge> what you trying to do ? 21:04 < bridge> https://github.com/heinrich5991/libtw2/blob/406f45ad4df6038b91f3f4f666b3adfe0e19906f/map/src/format.rs#L612 21:04 < bridge> I also simply strip the last byte 21:05 < bridge> ```c++ 21:05 < bridge> template || std::is_invocable>> 21:05 < bridge> struct invocable { 21:05 < bridge> template>> 21:05 < bridge> using result_t = std::invoke_result_t; 21:05 < bridge> 21:05 < bridge> template>> 21:05 < bridge> using result_t = std::invoke_result_t; 21:05 < bridge> }; 21:05 < bridge> ``` 21:05 < bridge> This is invalid C++, but it should kinda illustrate what I'm trying to do 21:06 < bridge> || with = is defined ? 21:06 < bridge> I'm trying to extract the return type from either an `std::invocable` or an `std::invocable