01:31 <+bridge> [ddnet] https://discord.com/channels/252358080522747904/342454474117218334/725130443011260516 01:31 <+bridge> [ddnet] Ghost bug 07:24 <+bridge> [ddnet] @ExP u are the second person this week to tell us😋 08:00 <+bridge> [ddnet] https://www.reddit.com/r/C_Programming/comments/hej4a3/why_uintptr_t_is_superior_than_void/ 08:03 <+bridge> [ddnet] "In C or C++, as mandated by the strict aliasing rule, pointer arguments in a function are assumed not to alias if they point to fundamentally different types, except for char* and void*, which may alias to any other type. Some compilers allow the strict aliasing rule to be turned off, so that any pointer argument may alias any other pointer arguments. In this case, the compiler must assume that any accesses through these pointers can alias 08:03 <+bridge> [ddnet] :poggers: 09:30 <+ChillerDragon> omg they took github down to make new ugly ui :/ 09:30 <+ChillerDragon> time to move to @Ryozukis gogs instance :D 09:48 <+bridge> [ddnet] :greenthing: 10:00 <+bridge> [ddnet] hello I have a problem with mysql when I go to the server to do (/top5) nothing is displayed and its gives me the same error for (/mapinfo) 10:00 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725259087008497734/unknown.png 10:01 <+bridge> [ddnet] I use the ddnet 0.6 / 0.7 mod 10:12 <+bridge> [ddnet] Your mysql server is set to use latin1 encoding 10:12 <+bridge> [ddnet] you have to switch it to the utf8mb4 10:13 <+bridge> [ddnet] i think in /etc/mysql/my.cnf and restart the mysql server 10:14 <+bridge> [ddnet] With the ddnet7 mod, mysql works very well. 10:17 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725263227164033125/unknown.png 10:19 <+bridge> [ddnet] are you using mysql btw or mariadb? 10:19 <+bridge> [ddnet] we only tested with mariadb 10:20 <+bridge> [ddnet] I only use MySQL 10:21 <+bridge> [ddnet] oh, didn't know we broke it :/ 10:22 <+bridge> [ddnet] I have to do like her? 10:22 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725264598017769472/unknown.png 10:23 <+bridge> [ddnet] at the bottom I see latin1 as the default 10:23 <+bridge> [ddnet] would be good to change that to utf8mb4 10:24 <+bridge> [ddnet] I have changed now I will try 10:24 <+bridge> [ddnet] this should fix /mapinfo 10:25 <+bridge> [ddnet] but top5 is probably still broken 10:27 <+bridge> [ddnet] ```[2020-06-24 10:26:05][sql]: MySQL Error: COLLATION 'utf8mb4_general_ci' is not valid for CHARACTER SET 'latin1'``` 10:27 <+bridge> [ddnet] I always have the same error when I change and after I restart mysql 10:29 <+bridge> [ddnet] see https://stackoverflow.com/questions/25852239/collation-utf8-general-ci-is-not-valid-for-character-set-latin1 10:29 <+bridge> [ddnet] @deen couldn't going into /spec send also that char update with info the tee is not chatting? 10:29 <+bridge> [ddnet] (but with utf8mb4 instead of utf8) 10:29 <+bridge> [ddnet] @Soreu yes, that would be the clean solution 10:33 <+bridge> [ddnet] how i can put them in utf8 10:33 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725267226692681768/unknown.png 10:34 <+bridge> [ddnet] in /etc/mysql there should be config files where you can change that 10:37 <+bridge> [ddnet] I opened all the files in mysql and there is nothing written (latin1) 10:38 <+bridge> [ddnet] i'm using ubuntu 18.04 10:40 <+bridge> [ddnet] ok, then latin1 is just the default 10:41 <+bridge> [ddnet] in my.cnf you could at the end put the settings 10:41 <+bridge> [ddnet] ok 10:41 <+bridge> [ddnet] Something like this: 10:41 <+bridge> [ddnet] ``` 10:41 <+bridge> [ddnet] [client] 10:41 <+bridge> [ddnet] default-character-set = utf8mb4 10:41 <+bridge> [ddnet] [mysqld] 10:41 <+bridge> [ddnet] character-set-server = utf8mb4 10:41 <+bridge> [ddnet] [mysqldump] 10:41 <+bridge> [ddnet] default-character-set = utf8mb4 10:42 <+bridge> [ddnet] [mysql] 10:42 <+bridge> [ddnet] default-character-set = utf8mb4 10:42 <+bridge> [ddnet] ``` 10:42 <+bridge> [ddnet] and then restart mysql 10:45 <+bridge> [ddnet] now the (/mapinfo) works well but not the (/rank & /top5) 10:45 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725270284999065670/unknown.png 10:46 <+bridge> [ddnet] yeah, those are a different issue. I asked Zwelf to take a look 10:46 <+bridge> [ddnet] ok thanks 10:50 <+bridge> [ddnet] upgrading mysql could also help, what version are you running? 10:52 <+bridge> [ddnet] oh u merged it 10:53 <+bridge> [ddnet] this is the last update 10:55 <+bridge> [ddnet] however the ddnet7 mod works very well when you execute the top5 command 10:57 <+bridge> [ddnet] yeah, we changed something there recently 10:57 <+bridge> [ddnet] you could revert to DDNet 13.2.2 for now 10:57 <+bridge> [ddnet] until we fix it 10:59 <+bridge> [ddnet] I can wait it doesn't matter I tested the version if it works well 11:09 <+bridge> [ddnet] that worked for me 11:22 <+bridge> [ddnet] @deen don't collect $200 🙂 11:22 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725279690654220299/unknown.png 11:40 <+bridge> [ddnet] @jason54 can you tell me the mysql version for reproducing? 11:41 <+bridge> [ddnet] because mysql 8 seems to support it 11:55 <+bridge> [ddnet] @deen 11:55 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725287896881299456/unknown.png 12:00 <+bridge> [ddnet] ok, that explains it 12:02 <+bridge> [ddnet] it comes from me? 12:02 <+bridge> [ddnet] See the discussion in https://github.com/ddnet/ddnet/issues/2343 12:03 <+bridge> [ddnet] Basically you're using an old MySQL version on your distribution. We are thinking about whether we want to keep supporting them 12:06 <+bridge> [ddnet] so i will try to install mariadb 12:26 <+bridge> [ddnet] you also need at least mariadb 10.2, so it's possible your distribution doesn't have that as well. which distro and version are you using? 13:07 <+bridge> [ddnet] now it works very well thank you 14:10 <+bridge> [ddnet] 0.7 has no color change delay so they can basically be rainbow 14:39 <+bridge> [ddnet] @nori are you sure? I do see the spam protection there in the code 14:41 <+bridge> [ddnet] @Learath2 yes 14:42 <+bridge> [ddnet] @Learath2 i think its only clientsidr 14:42 <+bridge> [ddnet] no some1 in game doing this 14:43 <+bridge> [ddnet] ah no 14:43 <+bridge> [ddnet] yea 14:46 <+bridge> [ddnet] @nori what do you mean? 14:46 <+bridge> [ddnet] No delay 15:03 <+bridge> [ddnet] https://gfycat.com/ultimateearlyflounder 15:05 <+bridge> [ddnet] nothing to see here 🤐 15:05 <+bridge> [ddnet] https://tenor.com/view/fizzer1k-5597-3689-move-along-nothing-to-seehere-gif-12353147 15:22 <+bridge> [ddnet] Yeah instead of letting people know of the bug abuse it 15:22 <+bridge> [ddnet] class act 15:23 <+bridge> [ddnet] @nori were you on 0.6? 15:24 <+bridge> [ddnet] I see what I missed 15:26 <+bridge> [ddnet] atleast it was only visual so I don't have to clean up ranks 15:27 <+bridge> [ddnet] when is the patch going online? 15:27 <+bridge> [ddnet] as soon as it compiles 15:27 <+bridge> [ddnet] $build_status 15:27 <+bridge> [ddnet] 🟢 15:28 <+bridge> [ddnet] @Learath2 yes was on 0.6 15:28 <+bridge> [ddnet] @jao you seem to have broke the check for master somehow, it should still be pending 15:29 <+bridge> [ddnet] @Learath2 is it possible to use 0.7 compatibility layer to connect sixup servers on client side? 15:29 <+bridge> [ddnet] That was the initial plan actually, but it proved a little too hard 15:30 <+bridge> [ddnet] R.I.P fun 😦 15:30 <+bridge> [ddnet] The entire protocol is there if you want to try implementing it, but it's quite a bit of work 15:30 <+bridge> [ddnet] this is so sad 15:31 <+bridge> [ddnet] not like I can restart the servers with people on it, so you get to mess with it a couple more hours atleast 15:39 <+bridge> [ddnet] @Learath2 why is the emote timer so low? seems glitchy 15:49 <+bridge> [ddnet] We used to have it slower, but apparently people enjoyed this more for some reason, idk 16:03 <+bridge> [ddnet] <ᶰ°Konͧsti> $build_status 16:03 <+bridge> [ddnet] 🟢 16:03 <+bridge> [ddnet] <ᶰ°Konͧsti> :danAmazing: 16:05 <+bridge> [ddnet] Am I correct in assuming that GameID in the record_race database is the Client ID? 16:05 <+bridge> [ddnet] If not, what is it? 16:06 <+bridge> [ddnet] storing the client id makes 0 sense 16:06 <+bridge> [ddnet] gameid is probably the ddnet server version 16:06 <+bridge> [ddnet] i dont remember 16:06 <+bridge> [ddnet] Server version wouldn't make sense either as they are heavily differing on the same day 16:07 <+bridge> [ddnet] ? 16:07 <+bridge> [ddnet] just look at the source code 16:07 <+bridge> [ddnet] thats true too 16:07 <+bridge> [ddnet] ty 16:08 <+bridge> [ddnet] pData->m_GameUuid 16:11 <+bridge> [ddnet] i think its random 16:11 <+bridge> [ddnet] RandomUuid 16:11 <+bridge> [ddnet] Can't be that random 16:11 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725352508293120180/unknown.png 16:12 <+bridge> [ddnet] right side is the amount of occurences 16:12 <+bridge> [ddnet] hmm look at what RandomUuid 16:12 <+bridge> [ddnet] does 16:12 <+bridge> [ddnet] github search feature is rly bad 16:13 <+bridge> [ddnet] ikr 16:13 <+bridge> [ddnet] Well either way, I don't think the UUID matters that much to me anyway. Thus I am probably just going to ignore it 16:13 <+bridge> [ddnet] Thank you anyway though 16:18 <+bridge> [ddnet] gameid is the teehistorian id 16:23 <+bridge> [ddnet] it's randomly generated after each map change and server restart 16:25 <+bridge> [ddnet] 816 occurences still sound a bit high. is that run_blue? ^^ 16:46 <+bridge> [ddnet] > 816 occurences still sound a bit high. is that run_blue? ^^ 16:46 <+bridge> [ddnet] it's luminati 16:46 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725361157250220122/unknown.png 16:46 <+bridge> [ddnet] do you understand the gameid now? 16:48 <+bridge> [ddnet] I think about half of it? semi-random teehistorian ID, with some correspondence to the map 16:49 <+bridge> [ddnet] it's completely random, each time a new map is started on the server 16:49 <+bridge> [ddnet] I atleast understand now that it doesn't hold any information I will be using ^^ 16:49 <+bridge> [ddnet] good to know what it is thought 16:57 <+bridge> [ddnet] "222 commits to master since this tag " 16:57 <+bridge> [ddnet] looks like this is gonna be a big update 16:58 <+bridge> [ddnet] massive 16:58 <+bridge> [ddnet] sad my feature got closed 16:58 <+bridge> [ddnet] which one? 16:59 <+bridge> [ddnet] freeze 16:59 <+bridge> [ddnet] most things use client capabilities, like irc 16:59 <+bridge> [ddnet] the excuse that its hard to maintain cuz possiblities, i find it a bit weak, you just need more testing then 17:00 <+bridge> [ddnet] this one https://github.com/ddnet/ddnet/pull/2307 @Learath2 17:00 <+bridge> [ddnet] gl in seeing /spec tees in dark areas 17:00 <+bridge> [ddnet] eh we don't really need the client capabilities though 17:00 <+bridge> [ddnet] @Ravie "also if the outline isn't visible in very dark areas, the nickname definitely will be." 17:00 <+bridge> [ddnet] you can disable name plates 17:00 <+bridge> [ddnet] I'm playing with nameplates disabled 17:00 <+bridge> [ddnet] you can also disable the spec tees 17:00 <+bridge> [ddnet] also it shouldnt matter much if u cant see them 17:00 <+bridge> [ddnet] @Lady Saavik well you couldn't see them at all before, now you can see them a little better 😛 17:00 <+bridge> [ddnet] its not like they are super important 17:01 <+bridge> [ddnet] enable nameplates u can see them before u can see the tee 17:01 <+bridge> [ddnet] =easy advantage 17:01 <+bridge> [ddnet] short names = ez advantage 17:01 <+bridge> [ddnet] we really should do a issue cleanup 17:01 <+bridge> [ddnet] true 17:01 <+bridge> [ddnet] lol 17:01 <+bridge> [ddnet] most old issues 17:01 <+bridge> [ddnet] are invalid 17:01 <+bridge> [ddnet] i made a list a day ago which got addressed tho 17:01 <+bridge> [ddnet] i can see `heinrich5991` from a mile away 17:02 <+bridge> [ddnet] i thing we should do this one asap https://github.com/ddnet/ddnet/issues/198 17:02 <+bridge> [ddnet] we need to have a good idea 17:02 <+bridge> [ddnet] that wont be hated 17:02 <+bridge> [ddnet] there actually aren't that many maps dark enough to have the circle not visible at all 17:02 <+bridge> [ddnet] meh I really liked the 4th option much better then all the others but I guess we have the transparent one now 17:02 <+bridge> [ddnet] @Ryozuki for me spec tees are important and others weapons aren't, what now 17:02 <+bridge> [ddnet] I think /spec is much more useful on kog 17:02 <+bridge> [ddnet] i dont care about kog 17:02 <+bridge> [ddnet] you can see who's alive etc 17:03 <+bridge> [ddnet] well u can now see spec tees 17:03 <+bridge> [ddnet] we only improved 17:03 <+bridge> [ddnet] whats your complain here 17:03 <+bridge> [ddnet] u can get it on kog if their admins work 17:03 <+bridge> [ddnet] we still have a bug with it though https://github.com/ddnet/ddnet/issues/2344 17:03 <+bridge> [ddnet] and implement the ddnet character network 17:03 <+bridge> [ddnet] ye 17:03 <+bridge> [ddnet] maybe u got an idea for a fix 17:03 <+bridge> [ddnet] I'd guess send characters for spec'ed tees as well 17:04 <+bridge> [ddnet] I don't complain 17:04 <+bridge> [ddnet] and add a flag that says they're spec'ed 17:04 <+bridge> [ddnet] (only for ddnet version >= something) 17:04 <+bridge> [ddnet] the ddnet char does that already?? 17:04 <+bridge> [ddnet] I'm just thinking about a solution that will work well for most players 17:04 <+bridge> [ddnet] for some reason its not sent or something 17:04 <+bridge> [ddnet] we don't send a character when someone is spec'ed, I think @Ryozuki 17:04 <+bridge> [ddnet] not only for one specific group or settings 17:05 <+bridge> [ddnet] (this is our way of making them invisible) 17:05 <+bridge> [ddnet] stop being so aggressive towards me 17:05 <+bridge> [ddnet] ah yes sure 17:08 <+bridge> [ddnet] even on relatively dark maps and in freeze it's pretty visible, unless your display is on the surface of the sun 17:08 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725366766460469308/unknown.png 17:08 <+bridge> [ddnet] > i dont care about kog 17:08 <+bridge> [ddnet] @Ryozuki that's sad, as we always cared about old versions, vanilla client, ppl playing with different settings and so on 17:08 <+bridge> [ddnet] @Lady Saavik the feature could also be used by kog 🙂 17:09 <+bridge> [ddnet] yes, they justn eed to implement this 17:09 <+bridge> [ddnet] we're not doing something exclusively for ddnet servers 17:09 <+bridge> [ddnet] its not possible to do it backwards compatible 17:09 <+bridge> [ddnet] @Ravie left is barely visible for me + in game it would be smaller 17:09 <+bridge> [ddnet] I can see it clearly with my brightness on 50% in a relatively bright room 17:10 <+bridge> [ddnet] the point is to keep it as subtle as possible at the same time 17:10 <+bridge> [ddnet] yes I like the version without feet 17:11 <+bridge> [ddnet] @heinrich5991 by flag you mean playerflag? 17:11 <+bridge> [ddnet] I mean a flag in the extended playerobj 17:11 <+bridge> [ddnet] do we have something like that already? 17:11 <+bridge> [ddnet] we have a DDNetPlayer 17:11 <+bridge> [ddnet] i guess its that 17:11 <+bridge> [ddnet] ye, that probably 17:11 <+bridge> [ddnet] ye 17:12 <+bridge> [ddnet] ah no 17:12 <+bridge> [ddnet] DDNetCharacter 17:12 <+bridge> [ddnet] the thing is ddnetcharacter already has that 17:12 <+bridge> [ddnet] i think 17:12 <+bridge> [ddnet] ExPlayerFlags = ["AFK", "PAUSED", "SPEC"] 17:12 <+bridge> [ddnet] ah nice 17:13 <+bridge> [ddnet] so we can just still send the `Character` (which we currently don't) to up-to-date ddnet clients 17:14 <+bridge> [ddnet] that will then know to not display the tee as-is, but as spec'ed 17:16 <+bridge> [ddnet] does that automatically not send stuff to 0.6 clients? @Learath2 17:16 <+bridge> [ddnet] template magic stops that 17:16 <+bridge> [ddnet] I'm scared 17:16 <+bridge> [ddnet] me too 17:17 <+bridge> [ddnet] The time to complain about it was when I was struggling with it a couple weeks ago 17:17 <+bridge> [ddnet] ^^ 17:19 <+bridge> [ddnet] @heinrich5991 messages in the protocol7 namespace use one version of `SendPackMsg` and the rest the original version 17:23 <+bridge> [ddnet] SendChatTarget with target -1 sounds like a bad idea 17:26 <+bridge> [ddnet] @heinrich5991 a speced character is NetworkClipped? 17:26 <+bridge> [ddnet] i guess its that 17:26 <+bridge> [ddnet] @Learath2 `elif any(c['status'] == 'pending' for c in data['check_suites']):` whats wrong? 17:26 <+bridge> [ddnet] idk it was showing green while it was still building 17:27 <+bridge> [ddnet] $build_status 17:27 <+bridge> [ddnet] 🟢 17:27 <+bridge> [ddnet] like right now 17:27 <+bridge> [ddnet] @Ryozuki I think it's 17:27 <+bridge> [ddnet] ``` 17:27 <+bridge> [ddnet] if (m_Paused) 17:27 <+bridge> [ddnet] return; 17:27 <+bridge> [ddnet] ``` 17:27 <+bridge> [ddnet] hmm i guessz 17:28 <+bridge> [ddnet] (pause <-> spec got switched around by deen at some point, might be an artifact of that) 17:28 <+bridge> [ddnet] should i just add an enum somewhere? 17:29 <+bridge> [ddnet] Can someone fix #2337 for me? 17:29 <+bridge> [ddnet] https://github.com/ddnet/ddnet/issues/2337 17:29 <+bridge> [ddnet] @Learath2 i copied it from u 17:29 <+bridge> [ddnet] @Ryozuki enum for what? 17:29 <+bridge> [ddnet] well it's quite possible that I had it wrong, I'll take a look 17:30 <+bridge> [ddnet] the version 17:30 <+bridge> [ddnet] 13022 17:34 <+bridge> [ddnet] doesnt work 17:40 <+bridge> [ddnet] even sending the ddnet char always 17:40 <+bridge> [ddnet] doesnt work 17:40 <+bridge> [ddnet] weird 17:41 <+bridge> [ddnet] maybe its just a client bug 17:42 <+bridge> [ddnet] do you always send the non-ddnet char? 17:42 <+bridge> [ddnet] that's the place where the position is 17:42 <+bridge> [ddnet] ah 17:42 <+bridge> [ddnet] wait 17:45 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/725376000514326619/unknown.png 17:45 <+bridge> [ddnet] doesnt work either 17:46 <+bridge> [ddnet] hmmmm 17:48 <+bridge> [ddnet] you know that the player is spectating 17:48 <+bridge> [ddnet] shouldnt the client know the pos anyway 17:48 <+bridge> [ddnet] idk 17:48 <+bridge> [ddnet] im confused 17:48 <+bridge> [ddnet] \/spec is the thing that makes your tee disappear 17:48 <+bridge> [ddnet] ah 17:48 <+bridge> [ddnet] maybe ::Snap isn't even called 17:48 <+bridge> [ddnet] because it's probably removed from the gameworld 17:49 <+bridge> [ddnet] have you checked if it's getting called? 17:49 <+bridge> [ddnet] oh wait 17:49 <+bridge> [ddnet] in 17:49 <+bridge> [ddnet] CNetObj_DDNetCharacter 17:49 <+bridge> [ddnet] we dont actually get m_Spec 17:49 <+bridge> [ddnet] ? 17:50 <+bridge> [ddnet] ah, it's in the explayerflags 17:50 <+bridge> [ddnet] ye 17:50 <+bridge> [ddnet] but the tee probably isn't snapped 17:50 <+bridge> [ddnet] so the client has no idea where the fuck the tee is supposed to be 😛 17:51 <+bridge> [ddnet] true, i guess already connected clients keep the character 17:52 <+bridge> [ddnet] @heinrich5991 what is FakeSnap? 17:52 <+bridge> [ddnet] for 16player compatibility 17:52 <+bridge> [ddnet] One day I hope we'll actually collaborate on features rather then people writing behemoths 17:52 <+bridge> [ddnet] wdym 17:52 <+bridge> [ddnet] arent we collaborating? :justatest: 17:53 <+bridge> [ddnet] we are collaborating on the project, but none of the actual features, it's a mishmash of one man projects glued together by deen 17:53 <+bridge> [ddnet] whats ur solution 17:54 <+bridge> [ddnet] yes, deen pulls together the project a lot 17:55 <+bridge> [ddnet] how about we all talk about things before going ahead and implementing thousands of lines 17:55 <+bridge> [ddnet] I'm sure we could have come up with a lot more cleaner structure for sixup if people actually wanted to talk about it 17:56 <+bridge> [ddnet] or the http masters, another waste of time, I implemented 3-4 versions of it, heinrich implemented god knows how many, you started doing a couple 17:56 <+bridge> [ddnet] which one do we have right now? none because it's not easy to write a feature of this size without discussion and collaboration 17:57 <+bridge> [ddnet] well i wasnt rly interested in 0.7 support :/ 17:57 <+bridge> [ddnet] @Learath2 maybe its just that we have 4 developers 17:57 <+bridge> [ddnet] not much tbh 17:57 <+bridge> [ddnet] 4-6 17:57 <+bridge> [ddnet] how is that a waste of time 17:57 <+bridge> [ddnet] http masters would be awesome 17:58 <+bridge> [ddnet] you're right, more collaboration would help even features 17:58 <+bridge> [ddnet] @onby the waste of time is that we have like 10 attempts or so, but none is in the actual codebase 17:58 <+bridge> [ddnet] @onby we all implemented a version of it a couple times over, yet we don't have a single release candidate 17:58 <+bridge> [ddnet] whats wrong with all the attempts? 17:58 <+bridge> [ddnet] surely they arent much worse than the udp one 17:58 <+bridge> [ddnet] we also lack player testing 17:58 <+bridge> [ddnet] we find bugs most of the time after releasing 17:58 <+bridge> [ddnet] @Learath2 one of the reasons for that might be that we have different activity cycles 17:59 <+bridge> [ddnet] if theyre implemented redundantly (like, patches for clients/servers to register to these servers as well as the normal ones) then even in the worst case it would be okay 17:59 <+bridge> [ddnet] It'd be nice if we could find a group of people that are willing to help us test release candidates 17:59 <+bridge> [ddnet] servers to register, clients to pull list from** 17:59 <+bridge> [ddnet] if we think how the http master should work, implementing it is a breeze, atleast the code outside ddnet 18:00 <+bridge> [ddnet] @heinrich5991 you're very active with the prs today 🙂 18:00 <+bridge> [ddnet] like the normal master 18:00 <+bridge> [ddnet] except http 18:00 <+bridge> [ddnet] lol 18:00 <+bridge> [ddnet] why should the underlying idea change 18:00 <+bridge> [ddnet] just got back from cycling and got > 50 emails from github alone 😄 18:00 <+bridge> [ddnet] this was in my todo on a master i was making in golang: 18:00 <+bridge> [ddnet] 18:00 <+bridge> [ddnet] Servers should be able to register both ipv4 and ipv6 and be identified as the same server. 18:00 <+bridge> [ddnet] Clients should be able to know a ipv4 and ipv6 belongs to the same server. 18:00 <+bridge> [ddnet] Master server should ping server entries to know that port is forwarded and clients can connect aka "fwcheck". 18:00 <+bridge> [ddnet] Prevent multiple entries across masters 18:00 <+bridge> [ddnet] Add a maximum amount of registered servers per ip 18:00 <+bridge> [ddnet] Admin endpoints or maybe use a websocket server 18:00 <+bridge> [ddnet] Add better logging 18:00 <+bridge> [ddnet] even if the http masters were functionally a clone of the udp ones it would still be an improvement bcs its more resistant to spoofing and other kinds of attack 18:01 <+bridge> [ddnet] these other features would be nice too but not a reason to put it off 18:01 <+bridge> [ddnet] @onby I had it working on #1834 and a shitty PoC server coded in python 18:01 <+bridge> [ddnet] https://github.com/ddnet/ddnet/pull/1834 18:02 <+bridge> [ddnet] heinrich5991 commented 1 hour ago 18:02 <+bridge> [ddnet] Not working on this right now. 18:02 <+bridge> [ddnet] pff 18:02 <+bridge> [ddnet] heinrich no u 18:02 <+bridge> [ddnet] but its true 18:02 <+bridge> [ddnet] He is right though, we are not working on this right now 18:02 <+bridge> [ddnet] it was abandoned 18:02 <+bridge> [ddnet] well i dont say it has to be done now 18:02 <+bridge> [ddnet] but i think it would be good to do eventually 18:02 <+bridge> [ddnet] However I'm quite sad that a request for comment only received one comment, the one closing it 😄 18:02 <+bridge> [ddnet] :justatest: 18:02 <+bridge> [ddnet] maybe we should have a roadmap 18:03 <+bridge> [ddnet] with features to discuss and implement 18:04 <+bridge> [ddnet] @Learath2 I have a feeling this breaks chat in demos, but I don't know whether it worked before 18:04 <+bridge> [ddnet] well if my opinion was important and i wasnt lazy to login to github i wouldve commented in support of this 😅 18:05 <+bridge> [ddnet] I mean pretty much everyone supports the idea, I was more looking for comments on the protocol, the implementation 18:06 <+bridge> [ddnet] @heinrich5991 is there a specific line that gave you the feeling? 18:06 <+bridge> [ddnet] eh yes 18:06 <+bridge> [ddnet] i doont understnad why the char doesnt send a snapshot 18:06 <+bridge> [ddnet] I think previously the server code handled the -1 18:06 <+bridge> [ddnet] @noby translate this to non nerd pls 18:06 <+bridge> [ddnet] @Ryozuki I think OnSnapshot is not getting called, could you verify that? 18:06 <+bridge> [ddnet] http masters are better because http is based on tcp which is a lot harder to do spoofed ddos attacks against 18:07 <+bridge> [ddnet] ah yes 18:07 <+bridge> [ddnet] @Learath2 I'd guess now the server sees each chat message multiple times 18:07 <+bridge> [ddnet] do that 18:07 <+bridge> [ddnet] how does that even work with 0.6/0.7 stuff 18:07 <+bridge> [ddnet] [11:05 AM] Learath2: I mean pretty much everyone supports the idea, 18:07 <+bridge> [ddnet] heh guess so xd 18:07 <+bridge> [ddnet] I bet that broke it already 😄 18:07 <+bridge> [ddnet] i cant find anywhere OnSnapshot in code 18:07 <+bridge> [ddnet] I asked about this a couple weeks ago aswell when I first added the `CHAT_SIX` thing 18:07 <+bridge> [ddnet] ripgrep 18:08 <+bridge> [ddnet] the function you posted 18:08 <+bridge> [ddnet] where you showed that you changed it @Ryozuki 18:08 <+bridge> [ddnet] Snap( 18:08 <+bridge> [ddnet] yes 18:08 <+bridge> [ddnet] theorically its always called when u snap the world 18:08 <+bridge> [ddnet] i think 18:08 <+bridge> [ddnet] not if the character isn't in the world 18:08 <+bridge> [ddnet] which I think it isn't while you're in spec 18:09 <+bridge> [ddnet] Well I don't have a fix for it and it needs to be fixed for a release 18:09 <+bridge> [ddnet] int CPlayer::Pause(int State, bool Force) 18:09 <+bridge> [ddnet] i dont see this method deleting the character 18:09 <+bridge> [ddnet] it's so easy to verify whether Snap is getting called 18:09 <+bridge> [ddnet] why don't you try it? 18:09 <+bridge> [ddnet] ok wait 18:10 <+bridge> [ddnet] @Ryozuki ProcessPause calls `CCharacter::Pause` which is what removes the character 18:10 <+bridge> [ddnet] and the character is removed in `CCharacter::Pause` which is called from your function 18:10 <+bridge> [ddnet] ohh 18:11 <+bridge> [ddnet] apparently heinrich made it 18:11 <+bridge> [ddnet] then i should simply not remove the entity 18:11 <+bridge> [ddnet] I made /spec and remade /pause back in the days 😉 18:11 <+bridge> [ddnet] @Ryozuki but then it suddenly has collision and so on 18:11 <+bridge> [ddnet] you can't not remove the entity 18:12 <+bridge> [ddnet] hjmhhmhmhmh 18:12 <+bridge> [ddnet] it's sensible that it's no longer in the world 18:12 <+bridge> [ddnet] How do you retain the information when you join before someone goes in spec? 18:12 <+bridge> [ddnet] ye, I also wondered about this 😄 18:13 <+bridge> [ddnet] probably something akin to UAF 😛 18:13 <+bridge> [ddnet] wats UAF 18:13 <+bridge> [ddnet] use after free 18:13 <+bridge> [ddnet] (i.e. probably from some data structure that isn't meant to be live at that point) 18:14 <+bridge> [ddnet]