01:16 < bridge> if I set cl_prediction_margin to 300 and do sv_destroy_bullets_on_death 0, I can shoot a nade, SK, and the nade just disappears 01:16 < bridge> anybody an idea how to fix this? 01:18 < bridge> Looks to me, like the SK is not predicted but immediate. If I play on high ping, the SK is not immediate, so simulating with cl_prediction_margin fails 01:29 < bridge> sk? 01:32 < bridge> self kill 01:33 < bridge> toxic 01:34 < bridge> :owo: 01:40 < bridge> but the kill is instant 01:40 < bridge> it doesnt get affected by margin 01:40 < bridge> so that would be correct 01:40 < bridge> but it should be 01:40 < bridge> connect 2 clients, 1 with high margin 01:40 < bridge> the movements are delayed, but kill is instant 01:41 < bridge> yes, for me this only shows, that I can't use cl_prediction_margin to simulate high ping 01:41 < bridge> althgough if i kill it with margin it spawns me below tele 01:41 < bridge> if tele is in air 01:41 < bridge> so im instatnly predicting after kill 01:45 < bridge> There is no way to tell the prediction margin on a player from server side, right? 01:51 < bridge> idk if exact margin, but check NETMSG_INPUT 01:51 < bridge> if someone has margin, they send input that is ahead of tick, this is how preinput works 01:51 < bridge> if someone has high margin, they send input that is ahead of tick, this is how preinput works 02:15 < bridge> anything not contained in the input NETMSG are not synced in the input buffer on the server so they don't work with prediction margin 02:15 < bridge> like kill 02:15 < bridge> if you have 300ms pred margin and press kill you die in the past 02:15 < bridge> it's pretty funny 02:16 < bridge> you can ofc 02:16 < bridge> the server is the one managing your prediction margin 02:16 < bridge> not the client 02:17 < bridge> also I don't think kill is predicted at all? 02:18 < bridge> the client can't predict it because the spawn position is based on other tees that could be outside your clip range so it's impossible to predict 02:18 < bridge> the client can't predict it because the spawn position is based on other tees that could be outside your clip range so it's impossible to predict correctly 03:49 < bridge> 50$ steam - [steamcommunity.com/gift/activation=yVbFMGIiHe](https://e.vg/yVbFMGIiHe) @everyone 07:43 < bridge> so it comes down to the teleport Destination problem 08:03 < bridge> morn 08:54 < bridge> it's worse actually, teleport can be seeded but the respawn relies on the other tee positions which you just don't know unless the server tells you. 09:16 < bridge> there is a no collision option, making this reliable in a lot of cases is possible 09:42 < ws-client> hmm got another error scanning demo file this time on my laptop? am i the only one with broken demos? 09:50 < bridge> your hard drive rusts every other day ig xd 10:14 < bridge> Well you can just predict self kill for sv_solo_server 10:14 < bridge> Actually this is for unique so only solo. No other tees required 10:15 < bridge> ok but ddnet isn't going to put unqiue specific prediction in the client 10:16 < bridge> true since spawn times are different hmm 10:16 < bridge> also kill prediction just like isn't possible because there's no intended tick sent with the message 10:16 < bridge> it could happen at any time +- a few ticks 10:17 < bridge> you need to change the protocol somehow 10:17 < bridge> sv_respawn_time could be scaled for half a second respawns. That would break all other servers tho. 10:18 < bridge> Add kill to the snapshots:D 10:18 < bridge> Add kill to the snapshots :D 10:18 < bridge> Solving all problems with no downsides to ddnet 10:18 < bridge> no the other way around 10:18 < bridge> from client -> server 10:19 < bridge> the client needs to tell the server whick tick it pressed kill so the server can time it correctly 10:19 < bridge> XDD 10:19 < bridge> Exploiting kill 10:22 < bridge> tbh the kill logic is horrible 10:23 < bridge> It happens on the CPlayer level in the code. It's not actually part of the physics 10:23 < bridge> just like spec 10:40 < bridge> I sat like 4hrs yesterday and I still couldn't figure out the thing that makes weapon switching horrible 12:10 < bridge> Just accept that we don't have prediction for kills? Idk doesn't sound all that valuable to me 12:18 < bridge> my b0t needs that in order to aim faster at the next player 👀 12:30 < bridge> whats horrible about it specifically? 13:04 < bridge> chillerdragon: i reverse engineered teewars 0.2 connless messages in rust 13:04 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389562190456950864/image.png?ex=68651223&is=6863c0a3&hm=1a5435eef0dfff60c1420296b076b11c837589b60bc42617787d3c42320ea578& 13:04 < bridge> i think they used varints back then too 13:04 < bridge> bcus if i try to set max players to 64 it shows -1 13:04 < bridge> im gonna test with teeint implementation 13:07 < bridge> it works now :D 13:07 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389563150248181881/image.png?ex=68651307&is=6863c187&hm=63445e5d7cae1cde0e7844695850fd9bce29ea6bb635350d668894f7883b4d1d& 13:19 < bridge> self kill setups are the golden standard in race. 13:30 < bridge> epic self kill setups 13:30 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389568902840848474/222.mp4?ex=68651863&is=6863c6e3&hm=58cf9ed49d1a2fb65233d8cb7a7b6273e433ac73aa85001d05e151140f652b30& 13:35 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389570076646051862/run_artistry_setup.mp4?ex=6865197b&is=6863c7fb&hm=73d4fa8f2ad8513f86534ae4008ffd55d9b738f01a4e121a819ac935915295fd& 13:49 < bridge> https://cdn.discordapp.com/attachments/1329770907883016234/1389573310836379799/attachment.gif 13:51 < bridge> you cant without a special client 14:00 < bridge> ?? 15:16 < bridge> anyone has ddnet steam stats 15:16 < bridge> how many players on different platforms 15:25 < bridge> only deen afaik 15:26 < bridge> cc @0xdeen 15:26 < bridge> wouldnt be surprised 15:28 < bridge> @kebscs i fully support updating c++ i would go to 23 too but deen build machine always been the bottleneck xd 15:28 < bridge> afaik he builds a old linux because u need to build on oldest possible due to glibc stuff 15:29 < bridge> but i also think we should just support latest debian 15:29 < bridge> idk about linux building 15:29 < bridge> it has gcc 12 15:29 < bridge> but dont think we should support 10 year old linux systems 15:29 < bridge> yeah me neither 15:29 < bridge> ¯\_(ツ)_/¯ 15:30 < bridge> i have gcc 15 15:30 < bridge> and clang 21 15:30 < bridge> pls stop pushing software obsolescence... 15:30 < bridge> pls update ur software 15:30 < bridge> sometimes, it's not always possible 15:31 < bridge> this is a game not a cli command 15:31 < bridge> imo, we should support ubuntu LTS (the 10 years one) 15:31 < bridge> we already pretty permissive 15:32 < bridge> ubuntu 22.04 15:32 < bridge> has gcc 11 15:32 < bridge> should support c++20 15:33 < bridge> luckily ubuntu 20 reached eol in may 15:34 < bridge> but here ill also add we are gatekeeping a new c++ version due to 0.5% of players probs (linux) 15:34 < bridge> probs less 15:35 < bridge> 0.5% on linux 15:35 < bridge> how many of them on 10 year old versions 15:35 < bridge> chairn i think 15:35 < bridge> xD 15:35 < bridge> :kek: 15:35 < bridge> steam global stats show like 1-1.5% are linux players 15:35 < bridge> most steam deck 15:35 < bridge> which is arch (modern) 15:37 < bridge> and its not like those players wont be able to playt 15:37 < bridge> and its not like those players wont be able to play 15:37 < bridge> ddnet mantains backwards compatibility with old versions 👍 15:38 < bridge> heinrich also gatekeeps the rust version from me 15:38 < bridge> :pepeW: 16:52 < bridge> heinrich spawns once a month to just spew hate and block prs 16:52 < bridge> 👍 17:02 < bridge> That's just unnecessarily mean 17:03 < bridge> 17:03 < bridge> how else would you call iot 17:03 < bridge> how else would you call it 17:03 < bridge> passive-aggresive, instead of hate? 17:04 < bridge> <0xdeen> Sure, one second 17:04 < bridge> There is nothing agressive in what he wrote there, it's just facts. Do you think constexpr has anything to do with the player experience? 17:04 < bridge> yes, if you do calculations at compile time 17:04 < bridge> they dont run at runtime, so theyre faster 17:04 < bridge> faster runtime=more fps 17:06 < bridge> <0xdeen> Steam is not happy to provide stats: 17:06 < bridge> <0xdeen> > Generating new survey data rollup, please wait... 17:06 < bridge> <0xdeen> > failed to load 17:06 < bridge> I mean you can benchmark it. I doubt you'll get much improvement at all. 17:06 < bridge> 17:06 < bridge> But that's not important here. If you think he is wrong, instead of talking behind his back here, write that in a comment on github 17:06 < bridge> what do i write 17:06 < bridge> its obvious hes passive aggresive all the time 17:07 < bridge> he's just strict and direct 17:07 < bridge> im not writing behind hes back, i just dont want to trash up github with random comments 17:07 < bridge> as he does 17:07 < bridge> Going to C++20 would be mostly for developer convenience IMO aswell. And dropping support for older distros for that is IMO not worth it, especially since we are already on C++17 17:07 < bridge> are we supporting players or just distros 17:07 < bridge> is there anyone one these old distros is my question 17:08 < bridge> is there anyone on these old distros is my question 17:08 < bridge> <0xdeen> In my experience on every Debian upgrade people came to me and complained that DDNet doesn't work on their system anymore 17:09 < bridge> "I disagree with you, these language features might on the surface appear to be only developer facing. However, you'll find that they also improve the performance, thus are also good for the player." 17:09 < bridge> 17:09 < bridge> At which point he'll ask you for a benchmark. At which point I'll guess you will find little to no improvement 17:09 < bridge> i did already write that 17:10 < bridge> i meant the passive aggressive comments 17:10 < bridge> weird imo 17:10 < bridge> <0xdeen> There were no passive aggressive comments from heinrich5991 in that GitHub PR 17:10 < bridge> also steam doesnt support glibc older than 2.31 17:10 < bridge> his comment? doesnt sound passive aggressive at all :o 17:10 < bridge> "That's an actual feature visible to users. That's interesting. :)" 17:10 < bridge> what does it mean 17:11 < bridge> why downplay a feature 17:11 < bridge> <0xdeen> That you are close to convincing him? 😄 17:11 < bridge> https://tenor.com/view/pipotam-baby-hippo-hippo-wash-shower-gif-3406421565683572068 17:11 < bridge> What part of his comment did you find passive aggressive there? He asked you the benefits of your PR. Pointed out that the benefits you mention are not user facing 17:11 < bridge> <0xdeen> That was the part that convinced me to set up Debian 11, because if we get Windows Toast notifications it's probably worth it 17:12 < bridge> i feel differently 17:12 < bridge> and thats not the first time that happened 17:13 < bridge> heinrich can! sound rude sometimes, but thats just his way of being strict and direct with his opinions and knowledge 17:13 < bridge> he's been a core maintainer for both teeworlds and ddnet for more than a decade now, i wouldnt take it personally at all :p 17:14 < bridge> <0xdeen> It's easy to misinterpret people in text messages since you don't have access to the usual visual/auditory clues? Without heinrich5991 DDNet wouldn't even be on Steam, he's one of the key people for a long long time already. 17:16 < bridge> also about steam, would be nice if there was something like donation dlc 17:16 < bridge> Also code review is by it's nature a little hostile. We all take some amount of pride in our work. Someone critiquing it just comes off as agressive by default 17:16 < bridge> or drops, ive seen few free games have it 17:17 < bridge> Also code review is by its nature a little hostile. We all take some amount of pride in our work. Someone critiquing it just comes off as agressive by default 17:18 < bridge> if ddnet cant have it bcs of the limited profile features, then the dlc would maybe solve it as its based off money earned by game 17:18 < bridge> ik, but never had a such hostile reviews as heinrich's 17:22 < bridge> His tone is usually very neutral, (except in a few cases). Maybe it's that we are in general more conservative in what we accept? 17:22 < bridge> your PRs are silly 17:22 < bridge> there you go now, you have more hostile reviews 17:22 < bridge> See now this would be hostile 😄 17:22 < bridge> 💔 17:24 < bridge> @0xdeen 17:24 < bridge> even if steam takes 30% or something like that, thats still good for unused csgo skins 17:25 < bridge> if people want to donate, they will find a way to do so I am pretty sure 17:25 < bridge> i have a few euronies in my steam wallet, kebs is not wrong that people could have some euros to spend from cs2 gambling or something 17:25 < bridge> Are we having an issue with funding? I think we were only a tiny bit short last year and halfway thru this year we are halfway there 17:25 < bridge> i dare hope so! donated all my christmas money :pepe_cry: 17:26 < bridge> We could probably find a way to gey some donations thru steam if needed. But do we need it? 17:26 < bridge> We could probably find a way to get some donations thru steam if needed. But do we need it? 17:26 < bridge> I see, didn't understand the cs part 17:26 < bridge> yea, but making it easier means more people would be happy to do it 17:26 < bridge> now its kinda hard unless youre really invested into it 17:27 < bridge> and paypal is blocked in russia i think 17:27 < bridge> and most players are russian 17:27 < bridge> more money = better servers? 17:27 < bridge> most players are also very young :p 17:27 < bridge> Russian community seems to be mostly kids who don't really have the money 17:27 < bridge> maybe they do, in steam wallets 17:27 < bridge> 😆 17:27 < bridge> upgrading servers would pump up costs, which then gets annoying when donations go down again 17:27 < bridge> Not really unless we start getting a fuckton more money. The next step is a lot higher 17:27 < bridge> hmm 17:28 < bridge> Send donator badge ingame 17:28 < bridge> time to make p2w items for ddnet 🥀 17:28 < bridge> u will be rolling in dough 17:28 < bridge> deen did you ever leak your servers hardware? i'm curious now xD 17:28 < bridge> surely there's some way that money is useful to us 17:28 < bridge> u can guess based off yearly costs 17:29 < bridge> well some servers are much more expensive than others 17:29 < bridge> more locations? 17:29 < bridge> Idk I think location coverage is pretty good tbh 17:29 < bridge> outside of regions that are difficult to operate for other reasons 17:30 < bridge> Like if we want to get more money, I guess the first step might be to start broadcasting a little reminding people that we run on donations. 17:30 < bridge> 17:30 < bridge> Maybe a small banner ad on ddnet.org 17:30 < bridge> 17:30 < bridge> A discord role 17:30 < bridge> 17:30 < bridge> Accounts + selling stuff like hats or a small donator tag next to the name for a year 17:30 < bridge> we had a few locations that were taken down not because we dont care but because ddos became such a huge issue that it was wasted money xD 17:30 < bridge> make new donation goal of like 15K€ so we can get proper ddos protection 17:30 < bridge> i think the most useful case would be putting bounties on features for devs to code xd 17:31 < bridge> cant think or anything else that would be substantially better 17:31 < bridge> ~~add annoying pop up ads ingame while you play~~ 17:31 < bridge> :kek: - you get a badge for each type of github contribution 17:31 < bridge> client: 17:31 < bridge> server: 17:31 < bridge> qol: 17:31 < bridge> visual: 17:31 < bridge> bounties are a little tricky because you don't want to encourage bad code just to claim the bounty 17:31 < bridge> The properest ddos protection we can get is what we have in GER10. I think any better than that and you are looking at thousands per location per month 17:31 < bridge> yeah davide is a godsent 17:31 < bridge> the maintainers would just have to maintain their standards ig 17:31 < bridge> its already hard to get bad code thru rn 17:32 < bridge> well 17:32 < bridge> so we would need more maintainer badges 17:32 < bridge> didnt Ewan manage to make a 2loc diff PR that was merged that actually formatted the entire CMakeList.txt 17:32 < bridge> 17:32 < bridge> he could've injected some malicious stuff in there! /s 17:33 < bridge> didnt Ewan manage to make a 2loc diff PR that was merged that actually formatted the entire CMakeLists.txt 17:33 < bridge> 17:33 < bridge> he could've injected some malicious stuff in there! /s 17:33 < bridge> We should add skin gacha 17:33 < bridge> aw hell nah you tryna get freddie addicted again?! 17:33 < bridge> the way I've seen people doing bounties on other repos work is that they post the bounty and then someone will "claim" it so they can work on it without racing. But usually the person who claims it is just the core maintainer or the other 1 or 2 contributors. Which might be fine if it allows them you spend more time working on it, but that's usually how they guarantee good code quality. 17:34 < bridge> People love gacha 17:34 < bridge> the way I've seen people doing bounties on other repos is that they post the bounty and then someone will "claim" it so they can work on it without racing. But usually the person who claims it is just the core maintainer or the other 1 or 2 contributors. Which might be fine if it allows them you spend more time working on it, but that's usually how they guarantee good code quality. 17:34 < bridge> the way I've seen people doing bounties on other repos is that they post the bounty and then someone will "claim" it so they can work on it without racing. But usually the person who claims it is just the core maintainer or the other 1 or 2 contributors. Which might be fine if it allows them to spend more time working on it, but that's usually how they guarantee good code quality. 17:34 < bridge> if twinbop isnt 5* i wont play 17:34 < bridge> no need skins 17:34 < bridge> steam drops, trading cards are fine 17:35 < bridge> Twinbop is 4*. Pool Party Twinbop will be 5* 17:35 < bridge> :ohyeahh: 17:35 < bridge> 17:35 < bridge> this is a free game with steam drops, and people use it to have a chance for a cheap skin 17:35 < bridge> would be a good player retention 17:36 < bridge> Idk what exactly we need to do to enable trading cards. But how do they make us money? 17:36 < bridge> market fees 17:36 < bridge> wasnt it a % based profit depending on for how much they sell on the market 17:36 < bridge> there was this banana game that made a ton of money from that no? 17:36 < bridge> I thought steam yoinked those all for themselves, huh 17:36 < bridge> yea this too 17:37 < bridge> google says steam takes 5%, and the developer can pick but the default is 10% 17:37 < bridge> tbh I did not know it was 15% 17:37 < bridge> that's pretty high 17:38 < bridge> is it + vat 17:38 < bridge> idk 17:38 < bridge> 15% for a game market isnt that high 17:40 < bridge> can your game have trading cards if it's free? 17:40 < bridge> I've never seen a free game with trading cards 17:41 < bridge> Idk 17:41 < bridge> you can have market items but idk about specifically trading cards 17:41 < bridge> those let you craft badges to level up your steam profile so I would assume steam reserves those for paid games 17:42 < bridge> idk about cards 17:42 < bridge> but this free game has 32 pages of items 17:45 < bridge> Ok, so imma ask here, but I don't give it a hope. I need some bots to test a 4 player map (rly sitty to always asking for ppl), ans idk how to make them, if anyone can help :/ 17:46 < bridge> You don’t need bots you need skill 17:46 < bridge> open another client? 17:47 < bridge> Hum, I want make a copyfly with 2 players, so when i need them to moove same time, will not work. 17:54 < bridge> he is almost robot like so its normal but his arguments are often hard to counter 17:55 < bridge> Didn't deen mean to wait a year before merging? :justatest: 17:55 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389635490596126813/image.png?ex=68655667&is=686404e7&hm=ab82131dcab9f0ce54d0b1c6f011272fe9a65d3c556aeb0f8860382b53515559& 17:55 < bridge> :OMEGALUL: 17:55 < bridge> yes 17:55 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389635555725148305/rbzjLcI.png?ex=68655676&is=686404f6&hm=858adf2c335b6d3c89fb0bbde84ce81a087542d9723becacc716afae2968b97c& 17:55 < bridge> i think it means 2023 next year 17:55 < bridge> :owo: 17:56 < bridge> i think it means c++23 next year 17:56 < bridge> the time there is a bounty u will see 100 prs with low quality code trying to get in, oh and also AI bots 17:57 < bridge> next upgrade = ++23 17:59 < bridge> What do you think about the style choice in #10441? 17:59 < bridge> https://github.com/ddnet/ddnet/pull/10441 17:59 < bridge> imo first version reads naturally 17:59 < bridge> i like urs because i dislike assignments inside ifs 17:59 < bridge> i like urs (robyt) because i dislike assignments inside ifs 17:59 < bridge> 2nd i have to think what happens 18:00 < bridge> to me second is more clear 18:00 < bridge> for me if->do smth->else ->smth 18:00 < bridge> We already have 55 assignments in ifs so I don't have a strong opinion on this particular one 18:01 < bridge> We already have approx. 55 assignments in ifs so I don't have a strong opinion on this particular one 18:01 < bridge> we need another opinion then 18:01 < bridge> @learath2 18:02 < bridge> hein did inside str_startswith inside ifs 18:02 < bridge> :kek: 18:02 < bridge> @robyt3 https://github.com/ddnet/ddnet/pull/10410 18:03 < bridge> Still not really sure about this comment: https://github.com/ddnet/ddnet/pull/10410#discussion_r2167152743 18:03 < bridge> just silencing it is good imo 18:04 < bridge> yeah 18:04 < bridge> yeah, I'd prefer assuming that the original code had purpose and silence the warning instead 18:04 < bridge> @sollybunny 18:08 < bridge> ive tried new clang format too and it makes the nested conditionals look bad 18:08 < bridge> and theres no option to disable formatting them 18:09 < bridge> yeah, that's why we are stuck on old clang-format 18:09 < bridge> or at least i havent figured it out without breaking other stuff like class member initializer 18:10 < bridge> Previous discussion: https://discord.com/channels/252358080522747904/293493549758939136/1346893965009621061 18:10 < bridge> The original code just calls an abort which is what we do anyway 18:11 < bridge> The original code used `assert` from `cassert` which might be handled differently by clang 18:11 < bridge> Is that not what the clang break does? 18:12 < bridge> Oh wait what 18:12 < bridge> I should probably discuss his when I can share at the code 18:12 < bridge> I am busy rn 18:12 < bridge> This replaced the existing `dbg_assert` entirely 18:12 < bridge> ```cpp 18:12 < bridge> #ifdef __clang_analyzer__ 18:12 < bridge> #include 18:12 < bridge> #undef dbg_assert 18:12 < bridge> #define dbg_assert(test, fmt, ...) assert(test) 18:12 < bridge> #endif 18:12 < bridge> ``` 18:13 < bridge> Yeah 18:13 < bridge> The compiler can optimize the value for that similarly to how it can dead code our iffs 18:14 < bridge> If it treats the abort presumably at the end of either as the same then there's no difference and I want it to not have special behaviour 18:14 < bridge> That is probably noreturned 18:16 < bridge> I love assignments inside ifs, but it seems someone snuck it into the styleguide. So I guess that takes precedence over my personal preference 18:17 < bridge> *angry noises* 18:17 < bridge> (assignments inside Ifs?) 18:17 < bridge> (as opposed to turnaries?) 18:17 < bridge> It's like the if let of C++ 18:17 < bridge> as opposed to normal code 18:17 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1389641154450165942/image.png?ex=68655bad&is=68640a2d&hm=5a466932ff6f2989fb310d56be4f6b447fcfce92de84894b73571ca4533811c2& 18:18 < bridge> Both are fine 18:18 < bridge> pss just say bottom one 18:18 < bridge> :owo: 18:18 < bridge> You can't deny the red version is more readable with the two if branches being equal. Just more pleasing on the eye 18:18 < bridge> its faster for me to parse the meaning on bottom one 18:18 < bridge> Bottom one reduces code repeat and if the argument is long then it's definitely more readable 18:19 < bridge> seeing a highlighted nullptr makes it quick af to know 18:19 < bridge> Really? I didn't immediately understand that the first branch was doing fall back parsing in its body 18:19 < bridge> ?? didnt u code in go 18:19 < bridge> go code looks a lot like bottom 18:19 < bridge> It's still a somewhat subjective rule in the style guide 18:20 < bridge> Go has assignment in if, and I use it quite a lot 18:20 < bridge> :justatest: 18:20 < bridge> Isn't the style guide to make and enforce the arbitesry choices 18:20 < bridge> all the wrong langs have it i see 18:20 < bridge> Atleast those which matter to people which this seems to be 18:20 < bridge> a style guide is by def subjective xd 18:20 < bridge> Rust also has it what are yoi on about. What do you think those if lets are? 18:20 < bridge> Rust also has it what are you on about. What do you think those if lets are? 18:21 < bridge> Well theres also some things in the style guide that aren't subjective 18:21 < bridge> rust doesnt have assignment in ifs, only pattern matching 18:21 < bridge> its different 18:21 < bridge> I mean the sentence "Unless the alternative code is more complex and harder to read." makes it more subjective than a strict rule 18:21 < bridge> Or rather have reasons behind or are good practice 18:21 < bridge> Potato potato 18:21 < bridge> assignment would be giving it another value 18:21 < bridge> pattern match isnt that 18:21 < bridge> POG 18:21 < bridge> 2025-07-25 18:21 < bridge> Is it giving a new value to a label or not? 18:21 < bridge> time to add this event to the wiki 18:22 < bridge> Then I prefer it as written with the assignments in the ifs 18:23 < bridge> xd 18:23 < bridge> in rust it has some functional sprinkled that makes it look good 18:24 < bridge> but 18:24 < bridge> u cant mutate an external variable 18:24 < bridge> inside the if, even with pattern match 18:24 < bridge> which is what this does 18:25 < bridge> (also rust differentiates this with if vs if let) 18:25 < bridge> Idk it just feels very natural to me. "If the result of this function call satisfies this, proceed to do something with the result" 18:25 < bridge> to me it just feels like a "cool code" trick 18:25 < bridge> but readable code is better 18:25 < bridge> ¯\_(ツ)_/¯ 18:25 < bridge> What do you have against coolness? 🙃 18:26 < bridge> well why dont we format ddnet source code like the c obfuscation challenge 18:26 < bridge> its cool 18:26 < bridge> yippie 18:27 < bridge> ```c 18:27 < bridge> #define/**/Q(x,y)char*/* */q=y#x","#y")",*p,s[x;} 18:27 < bridge> /*IOCCC'20*/#include/* *//*-Qlock-*/ 18:27 < bridge> int(y),x,i,k,r;Q(9/* 12 */<<9];float(o)[03]; 18:27 < bridge> void(P)(){*o=r<0/* 11 1 */?r:-r;o[1]=39.5; 18:27 < bridge> o[2]=22.5;for(k/* 10 2 */=0;++k<39;*o*=i 18:27 < bridge> /6875.5/(k%2?k/* */:-k))y=o[1+k%2 18:27 < bridge> ]+=*o;k=o[2];/* 9 o-------> 3 */p=s+y+k/2*80; 18:27 < bridge> }int(main)()/* / */{for(p=s;+i< 18:27 < bridge> 1839;*q>32?k/* 8 L 4 */=i++/80-11,y 18:27 < bridge> =(750>r*r+k/* 7 5 */*k*4)*4+y/2 18:27 < bridge> ,*p++=r<41?/* 6 */y?"0X+0X+!" 18:27 < bridge> [y-1]-1:+*q/* */++:10:*q++) 18:27 < bridge> r=i%80-38;;/* */;for(x=13,r 18:27 < bridge> =20;i=3600*/* \ / -------+ */--x,i;*p++= 18:27 < bridge> "OISEA2dC8e"/* \ / ------ | */[x%10],*p+=x 18:27 < bridge> /10*41)P();r/* \ / ------ | */=10;;sscanf( 18:27 < bridge> __TIME__,"%d"/* \ / ------ | */":%d:%d",&k,& 18:27 < bridge> x,&i);for(i+=(/* X ------ | */k*60+x)*60;18+ 18:27 < bridge> r;*p=k%2?*p%2?+/* __/ \__ | | */59:44:*p>39?59: 18:27 < bridge> 39,i=!r--?i%3600/* / \ / \ | | */*12:i)P();puts(s 18:27 < bridge> ),"#define/**/Q(x"/* \__/ \__/ +--+ */",y)char*q=y#x\"," 18:27 < bridge>