00:01 <+sctt_> heinrich5991: it's ok for you to use this style? 00:01 <+sctt_> https://ibb.co/bQ36U5 00:02 <@heinrich5991> mhmh. dbg_msgs are usually one line 00:02 <@heinrich5991> if you do it this way, the first line will have [] stuff and the others won't 00:02 <@heinrich5991> is that intended? 00:02 <+sctt_> heinright591: maybe i will just issue more dbg_msgs? 00:03 <@heinrich5991> yea 00:03 <+sctt_> ok :) 00:03 <@heinrich5991> but if(!pStorage) doesn't really fit there ^^ 00:04 <+ddnet-discord> are you programming with nano? :o 00:05 <+sctt_> Ryozuki: well to be honest when i started this i didn't think i would take it that seriously xD 00:05 <+ddnet-discord> why not vim? 00:06 <+ddnet-discord> "I've been using vim for about 2 years now, mostly because I can't figure out how to exit it." 00:06 <+ddnet-discord> funny :) 00:06 <@heinrich5991> :) 00:07 <+sctt_> hahahha 00:07 <+sctt_> well the reality is that i've always used nano and never found a reason to switch to vi, even if i've used it a couple of times 00:08 <+ddnet-discord> well its all personal preference, my eyes wouldnt stand with that green :P 00:08 <+sctt_> i know vi is more powerful, but in a way i've never found indispensable 00:08 <+sctt_> Ryozuki: well the colors are configurable 00:08 <+ddnet-discord> ik :P 00:08 <+sctt_> Ryozuki: i like retro green xD 00:28 <+sctt_> heinrich5991: i've pushed :P 00:32 <@heinrich5991> great :) 00:32 <@heinrich5991> couple of things: 00:32 <@heinrich5991> I still see some of the code indented with 8 spaces instead of tabs 00:32 <@heinrich5991> DataWriter and DataReader should get a g_ prefix 00:33 <@heinrich5991> it really doesn't hurt to declare the variables in the main function a bit later, at their use site 00:33 <@heinrich5991> i.e. in the loops 00:33 <@heinrich5991> but that's it, I think :) 00:33 <@heinrich5991> sctt_: ^ 00:34 <+sctt_> i noticed the spaces, i'm fixing now 00:34 <+sctt_> xD 00:34 <+sctt_> ok for data 00:40 <+sctt_> heinrich5991: ok done^^ 00:42 <@heinrich5991> ok, looks good now. you really like nonlocal variables, aha :) 00:42 <@heinrich5991> sctt_: the newest commits are made by root@term.station 00:42 <@heinrich5991> if that's ok, I'll go ahead and merge 00:42 <+sctt_> "you really like nonlocal variables" what do you mean? xD 00:42 <+sctt_> oooh 00:43 <+sctt_> sorry i've read non-canonical, i don't know why :/ 00:43 <+sctt_> i hate globals, i never use them 00:43 <+sctt_> i've not for a piece of code that does only one thing and it's under 100 lines xD 00:43 <@heinrich5991> not what I mean, I ment that you declare all your variables at the start of the main function instead of in the loops where they're used 00:45 <+sctt_> i did that only because it was like this on the code i started and you didn't tell me to move them 00:45 <+sctt_> i'm trying to do things as you are used to :/ 00:46 <+sctt_> there are people who likes to have data list in the top part xD 00:46 <@heinrich5991> anyway, [00:42:15] <@heinrich5991> sctt_: the newest commits are made by root@term.station 00:46 <@heinrich5991> is that ok or do you want to change that? 00:47 <+sctt_> i don't really care about the name, i would change only if you prefer that 00:47 <@heinrich5991> nah, d/c 00:47 <+sctt_> ok^^ 00:48 <@heinrich5991> I'll wait for the travis ci build to finish before merging 00:48 <@heinrich5991> should hopefully be done in a couple of minutes ... 00:48 <+sctt_> ok :) 00:48 <+sctt_> to be honest i'm surprised you are interested in map_image_replace tool 00:49 <+sctt_> i hadn't think about a merge when i wrote it 01:02 <@heinrich5991> w/e build still not started, will go to sleep and merge tomorrow 01:02 <@heinrich5991> good night sctt_ 01:02 <+sctt_> yeah no pressure 01:02 <+sctt_> goodnight and thanks! 01:02 <@heinrich5991> thanks for the pull request! :) 01:02 <+sctt_> :) 01:40 <+ddnet-discord> @everyone Countdown for DDNet!!! 21days 10hours 20 minutes 07:48 < ddnet-commits> [ddnet] def- pushed 4 new commits to master: https://git.io/v5cpk 07:48 < ddnet-commits> ddnet/master 0849fa6 sctt: fixing map_resave and adding map_replace_image... 07:48 < ddnet-commits> ddnet/master 2c748eb root: Fixing PR over heinrich revision 07:48 < ddnet-commits> ddnet/master 7cd95f6 root: Fixing spaces and globals 14:02 < ddnet-commits> [ddnet] heinrich5991 opened pull request #847: Rename `GetUncompressedDataSize` to `GetDataSize` (master...pr_ddnet_uncompressed_datasize) https://git.io/v5CgC 15:57 <+redix> Some days ago someone wrote that the precision on the ghost (and i guess the auto recorded demos) is dropped. 15:57 <+ddnet-discord> Learath2 15:57 <+redix> Thsi is probably due to this line: https://github.com/ddnet/ddnet/blob/0a9f2913935feccd93c66d739b9b8a3279a7443d/src/engine/client/client.cpp#L3661 15:58 <+ddnet-discord> 348533194355900416 15:58 <+ddnet-discord> Learath2 - 19.08.2017 Can someone take a look at how we store ghosts? We seem to drop the precision, we shouldn't 16:00 <+redix> The servers use dots as decimal-point character but some languages use commas (at least german) 16:01 <+redix> the scanf call which extracts the time drops the precision because of this 16:02 <+redix> Solution would be to remove the line (at least in the latest sdl version it seems to be fixed) or do the time parsing ourselves 16:03 <+ddnet-discord> so maybe try to remove that line and pull request with information about fix ? https://hg.libsdl.org/SDL/rev/2f18ea79bc03 16:05 <+redix> Question is whether the systems where the bug occured have to fixed version now 16:06 <+ddnet-discord> I think ddnet provides dlls with a client? Is he not? 16:07 <+ddnet-discord> or libs, don't know what is used on linux 16:09 <+redix> The bug occured on linux. I don't know whether the bundled or system libs are used for releases 16:09 <+ddnet-discord> then i don't know 16:09 <+ddnet-discord> On windows, there are bundled dlls 16:10 <+ddnet-discord> deen: are you here? 16:11 <+redix> ddnet-libs does not contain a bundled version for sdl... 16:12 <+redix> only for windows... not for linux 16:12 <+ddnet-discord> :/ only for windows, I see 16:13 <+ddnet-discord> sorry, I'm not familiar with C/C++. I'm web dev 😛 16:13 <+redix> In this case doing the parsing ourselves might be the better solution 16:13 <+redix> okay :D 16:13 <+redix> will look at this later again 16:32 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5C7U 16:32 < ddnet-commits> ddnet/master 43af1fd oy: fixed sprite rendering. Closes #834 16:34 < ddnet-commits> [ddnet] def- pushed 2 new commits to master: https://git.io/v5C7Z 16:34 < ddnet-commits> ddnet/master 2682480 heinrich5991: Rename `GetUncompressedDataSize` to `GetDataSize`... 16:34 < ddnet-commits> ddnet/master 25a5eff Dennis Felsing: Merge pull request #847 from heinrich5991/pr_ddnet_uncompressed_datasize... 17:22 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5Cho 17:22 < ddnet-commits> ddnet/master 7f0eb53 def: Initialize in order to ensure destruction in order (fixes #848) 17:29 <+ddnet-discord> @heinrich5991 43af1fd what doesitdo? 17:31 <+ddnet-discord> https://github.com/teeworlds/teeworlds/blob/master/scripts/tw_api.py#L213 17:31 <+ddnet-discord> What does this line do? 17:31 <+ddnet-discord> don't know Python, trying to rewrite to PHP 17:41 <+ddnet-discord> Nevermind 😃 Python automatically split string to array of characters 17:41 <+ddnet-discord> 😃 18:17 <+ddnet-discord> Great, I've done collecting data from MasterServers by PHP, yey. :D https://packagist.org/packages/savander/twservers 20:23 <@deen> Learath2: I have trouble with the fetcher 20:23 <@deen> It fetches the first 16 KB, and writes them and then I get [2017-08-30 20:21:47][fetcher]: task failed. libcurl error: Failed writing body (0 != 16200) 20:36 <@deen> wtf, failes with Debug, but works with Release 20:36 <@deen> (in cmake) 20:37 <@deen> ok, same in bam at least 20:37 <@deen> why are our map downloads broken in debug? 20:45 <@heinrich5991> redix: I think the precision is dropped on the server side 20:46 <@heinrich5991> there's a tw vanilla issue about this I think 20:55 <+redix> heinrich5991: what exactly do you mean? The issue I encountered is that a message like "xxx finished in: 0 minute(s) 21.50 second(s)" is recognized as 0:21.00 if your system language expects 21,50 20:56 <@heinrich5991> ok, no, that's something different 20:57 <@deen> redix: we already set LC_ALL 20:57 <+redix> not sure if this is what Learath2 meant though 20:57 <@deen> maybe we need to set some other thingy 20:57 <+eeeee> changing https://github.com/ddnet/ddnet/blob/5a437b110f8efdb1c27e38f26b7681ff7d1c33cb/src/engine/client/fetcher.cpp#L170 to "return io_write(..." might fix the downloads. https://curl.haxx.se/libcurl/c/CURLOPT_WRITEFUNCTION.html 20:57 <@deen> but not using scanf is probably a better idea 20:59 <+redix> LC_ALL is the problem... without LC_NUMERIC all should be fine. But without sscanf might be a better solution 21:00 <@deen> eeeee: thanks, that works! 21:00 <@deen> redix: make a PR please 21:00 <+redix> okay 21:01 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5Wgf 21:01 < ddnet-commits> ddnet/master e480b65 def: Fix fetcher write function to return size 21:32 <@Learath2> deen: sry totally missed that 21:32 <@deen> np :) 21:32 <@deen> I have a cool new feature 21:32 <@deen> That was written by timakro 2 years ago 21:32 <@Learath2> hype? 21:32 <@deen> and then abandoned, just made it work 21:32 <@deen> yeah, best feature ever, +1000% player growth tomorrow 21:33 <+eeeee> Learath2: not your fault curl api has no type safety 21:33 <@deen> (but i'd like to get the crashes fixed before releasing) 21:33 <@Learath2> what be the feature? 21:33 <@deen> eeeee: can you even have type safety in a c api like that? 21:34 <+eeeee> not with a generic "setopt" function which does everything. but if they had a specialized function for setting each callback then it could be safe i think. 21:34 <@deen> Learath2: This: 21:34 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5WwM 21:34 < ddnet-commits> ddnet/master 16f3ddc def: Show if user has rank on map in browser (+filter)... 21:35 <@heinrich5991> how? :) 21:35 <@heinrich5991> *clicks link* 21:35 <@Learath2> that doesnt sound too safe :D 21:35 <@deen> compile it yourself and see! 21:35 <@deen> works already with default servers 21:35 <@deen> client-only change 21:36 <+eeeee> kek 21:37 <@Learath2> eh that sounds safe :) 21:37 <@Learath2> nice idea 21:37 <@deen> maybe need to make sure it only includes the real "ddnet type" servers 21:37 <@deen> not block, ictf etc 21:38 <@deen> and we have the /players/ web thingy already 21:38 <@deen> So https://ddnet.tw/players/?json=deen is super fast 21:38 <+ddnet-discord> sounds good 😮 21:39 <@deen> I wanted to add ranks to that too, but I feel like the server browser looks overloaded if we display another number 21:39 <@Learath2> i really should have made fetchtasks just jobs :/ 21:39 <@deen> and then people would all change their name to milk just to have a nice looking server browser 21:40 <+ddnet-discord> hahahah 21:40 <@deen> Oh, and we expose the player name to the web server every time you press refresh, not sure what to think about that 21:41 <+eeeee> but it's our webserver so it's safe, no? 21:41 <@deen> yeah 21:41 <@deen> and it's only on ddnet tab 21:41 <@deen> so these people will go on our servers anyway 21:42 <@Learath2> deen: could the traffic be an issue? 21:42 <@deen> nope 21:42 <@heinrich5991> not even mysql-webserver traffic? 21:42 <@Learath2> well we could always throttle client side if it does 21:42 <@deen> I ran time.gif 21:42 <@deen> heinrich5991: no mysql required 21:42 <@heinrich5991> kk 21:42 <@heinrich5991> cached? 21:42 <@Learath2> heinrich5991: the player pages are prebuilt iirc 21:42 <@deen> heinrich5991: we have all that info in ram for /players/ 21:42 <@heinrich5991> generated? 21:42 <@deen> at all times 21:42 <@heinrich5991> cool 21:42 <@deen> including all ranks, times etc 21:42 <@heinrich5991> sounds safe 21:42 <@deen> would be way too slow to request that dynamically 21:42 <@heinrich5991> ok 21:42 <@heinrich5991> didn't know that 21:43 <@deen> that's the main memory user on our server actually :D 21:43 <@Learath2> deen: how do we have it on ram? 21:43 <@deen> [pid: 11463|app: 0|req: 544/544] 84.191.82.137 () {38 vars in 453 bytes} [Wed Aug 30 21:42:10 2017] GET /players/?json=milk => generated 20415 bytes in 0 msecs (HTTP/1.1 200) 1 headers in 51 bytes (0 switches on core 0) 21:43 <@deen> Learath2: it's a uwsgi python thingy, scripts/players.py 21:44 <@deen> so should we show the flag or the rank of player instead? 21:45 <@heinrich5991> flag 21:45 <@heinrich5991> (my vote) 21:46 <+ddnet-discord> flag 21:46 <@Learath2> just a checkered finish flag rite? 21:46 <@deen> yes 21:46 <@Learath2> +1 21:46 <@deen> and empty flag when unfinished 21:46 <@Learath2> maybe nothing when unfinished to keep it cleaner? 21:46 <@deen> and no flag at all for other tabs (and non-ddnet server kinds) 21:47 <@deen> hm 21:47 <@deen> then it doesn't look aligned 21:47 <@deen> or you mean just leaving it empty? 21:47 <@Learath2> i mean empty yeah 21:47 <+ddnet-discord> yea, could be cleaner 21:47 <+ddnet-discord> like with password servers 21:47 <@deen> yeah, doesn't look bad 21:47 <@deen> and I have to write less code 21:47 <@deen> great 21:48 <+ddnet-discord> you know Immediately if you finished 21:49 <@deen> anyone want to figure out the fixes for the new issues I made? have to fix the crashes before release 21:50 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5Wop 21:50 < ddnet-commits> ddnet/master a1d2ecb def: Empty when no rank 21:56 <@Learath2> deen: #849 is #825 21:57 <@Learath2> need to terminate it before quitting the main process 21:57 <@deen> ok, can you do that? 21:58 <@Learath2> it would be quite antisocial of me to code on a boat :P 21:58 <@Learath2> can try to sneak it in after everyone else goes to sleep if you really need me to :) 21:58 <@deen> nah, that's fine 21:58 <+ddnet-discord> lol 21:58 <+ddnet-discord> coding on boat 21:58 <+ddnet-discord> new achievement 21:58 <@Learath2> @Savander yean i'd never hear the end of that one 21:59 <+ddnet-discord> you should show them DDNet, maybe they will start to code with you 21:59 <+ddnet-discord> as wel 21:59 <+ddnet-discord> well 😮 22:03 < ddnet-commits> [ddnet] heinrich5991 opened pull request #852: Add testing; fix, rename and test the `CEditor::ExtractName` function (master...pr_ddnet_fix_extractname) https://git.io/v5W6B 22:03 <@heinrich5991> does someone have ubuntu here? 22:04 <@deen> gtest? 22:04 <@heinrich5991> google test 22:04 <@heinrich5991> used by llvm, chromium 22:04 <@deen> I know gtest 22:04 <@deen> just wondering, you're adding that 22:05 <@heinrich5991> I have no experience with testing frameworks, so I just picked the seemingly most popular one 22:05 <@heinrich5991> if you know something better, I can change that 22:05 <@deen> nono, i have lots of experience with gtest from work 22:05 <@heinrich5991> cool 22:05 <@heinrich5991> so did I do it correctly? ^^ 22:06 <@deen> dunno, there are no tests yet :D 22:06 <@heinrich5991> there are five 22:06 <@deen> you forgot to add them? 22:06 <@heinrich5991> https://github.com/ddnet/ddnet/pull/852/files#diff-b46255edcb8ce021a21c6b190484bc23 22:06 <@heinrich5991> no 22:06 <@deen> ah 22:07 <@deen> not used to tests not starting with Unit* :D 22:07 <@heinrich5991> :) 22:07 <@deen> you could use parameterized teests 22:08 <@heinrich5991> that sounds like something reasonable 22:08 <@deen> but in this case just use a fixture class with a helper method and call that 22:08 <@heinrich5991> a what? 22:08 <+ddnet-discord> new age of ddnet? unit tests ? 22:08 <@heinrich5991> what is a fixture class? 22:08 <@deen> and each test uses the fixture class and calls the method 22:08 <@deen> I'll just merge it and show you? 22:08 <@heinrich5991> sure 22:09 <@heinrich5991> make run_tests to run the tests 22:09 <@deen> make: *** No rule to make target 'run_tests'. Stop. 22:10 <@heinrich5991> in the build folder 22:10 <@deen> ah, have to install gtest 22:10 <@heinrich5991> oh yea, that too 22:10 <@heinrich5991> do you happen to have ubuntu? 22:10 <@deen> arch 22:10 <@heinrich5991> (that's the CI environment) 22:10 <@heinrich5991> meh, already tested there :D 22:11 <+ddnet-discord> I want to learn how to create webhooks for ddnet :C 22:11 <@deen> webhooks for discord? 22:11 <+ddnet-discord> webhooks? 22:12 <@heinrich5991> . 22:12 <@heinrich5991> duh, irc window broken :( 22:12 <@heinrich5991> better 22:12 <+ddnet-discord> yes 22:12 <+ddnet-discord> like 22:12 <@deen> it's just a POST to https://discordapp.com/api/webhooks/SECRETCODEHERE 22:12 <+ddnet-discord> imagine i finish a map on ddnet 22:12 <@deen> with a json {'content': msg} 22:12 <+ddnet-discord> https://discordapp.com/developers/docs/resources/webhook 22:13 <+ddnet-discord> i wanted it to send a message on my dioscord 22:13 <+ddnet-discord> saying that i did it 22:13 <+ddnet-discord> discord webooks, if you are interested 22:13 <@deen> Well, you can't do that because you don't have ssh on the ddnet server 22:13 <+ddnet-discord> lemme see savander 22:13 <+ddnet-discord> whats ssh 22:13 <@deen> i mean, it's a script we run on our sever every few minutes 22:13 <+ddnet-discord> RockuS. DDnet send post data to discord, to the specific url 22:13 <+ddnet-discord> oh 22:13 <@deen> and it just checks for new ranks and puts them on discord 22:13 <+ddnet-discord> that's why, you can see notifications :P. 22:13 <+ddnet-discord> i see 22:14 <+ddnet-discord> aw rip then 22:16 <+ddnet-discord> what is that irclink thing 22:16 <+ddnet-discord> its a webhook 22:17 <+ddnet-discord> for you to send messages 22:17 <+ddnet-discord> You still can parse ddnet site 22:17 <+ddnet-discord> but i don't know anything related to programming and shit 22:17 <+ddnet-discord> to gather information by your script 22:17 <@heinrich5991> or get the records from the ddnet discord bot 22:17 <@Learath2> i give permission to the bot to post here 22:17 <+ddnet-discord> or you can do that, right 😛 22:17 <+ddnet-discord> but idk how 22:17 <+ddnet-discord> Actually, its something more 22:18 <+ddnet-discord> webhook's are a bit different to pure bots 22:18 <+ddnet-discord> and i don't understand either of them 22:18 <@Learath2> webhooks get to send messages 22:18 <+ddnet-discord> bots do actions 22:18 <@Learath2> bot's can parse and act on messages on discord 22:18 <@deen> heinrich5991: like this: 22:18 <+ddnet-discord> i see 22:19 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5WPK 22:19 < ddnet-commits> ddnet/master 58c5303 def: Change unit test layout 22:19 <+ddnet-discord> Omg, DDNET WILL HAVE UNIT TEST S 😮 22:19 <+ddnet-discord> NO MORE CRASHES 😮 22:19 <+ddnet-discord> what's that XD 22:19 <+ddnet-discord> ehm, in PHP it test specific function 22:19 <+ddnet-discord> if it still works :v 22:19 <+ddnet-discord> or whole feature 22:20 <+ddnet-discord> it's too complicated 22:20 <+ddnet-discord> i'm not C programmer, unfortunately :? 22:20 <@heinrich5991> deen: this doesn't look like much of an improvement 22:21 <@heinrich5991> tbh 22:21 <@deen> not using macros is the improvement 22:21 <@heinrich5991> ok 22:21 <@deen> otherwise it does the same, yes 22:21 <@heinrich5991> why do you need a whole class though? 22:21 <@heinrich5991> function would work too, right? 22:22 <@heinrich5991> what does TEST_F do different from TEST? 22:22 <@deen> oh, probably yes. we always use classes so that we can use constructor/destructor for additional setup/cleanup for each test 22:22 <@deen> and some magic stuff 22:22 <@deen> not sure if we'll need that in ddnet 22:22 <@heinrich5991> not in that case, probably 22:26 < ddnet-commits> [ddnet] Redix opened pull request #853: Ghost and Race demo time extraction fix (master...pr_ghost_time) https://git.io/v5WXB 22:26 <@Learath2> redix: could you fix the actual timing while at it? 22:26 <@Learath2> we should store more then t 22:27 <@Learath2> the seconds 22:27 <+redix> the code extracts up to 3 decimal places 22:28 <@Learath2> redix: unless you changed sth it doesn't 22:28 <@heinrich5991> I guess you could also replace decimal commas with dots or so? 22:28 <@heinrich5991> before trying to parse 22:28 <@Learath2> it looked to me as it did aswell, after trying apparently it doesn't 22:30 <+redix> Learath2 what exactly happens? for me everything is fine now :O 22:35 <@Learath2> get the time 7.25, then get the time 7.20 you get new record but your old ghost is what exists 22:35 <@Learath2> Tezcan can show it to you if you find him ingame 22:36 <+redix> this should be fixed now... the client used 7.00 for both. So the second wasn't better 22:37 <+ddnet-discord> Oh so you did fix it up 22:37 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5W15 22:37 < ddnet-commits> ddnet/master 040fc9c def: Implement right click and middle click on country and type filters... 22:38 <+redix> i think so... at least if Tezcans system uses , for decimal-points 22:44 <+ddnet-discord> Tezcans server is just a normal ddnet server modified to store 1 more digit. It s record messages are same 22:44 <+ddnet-discord> it has more features tho 22:45 <+redix> sure... but the system he ran the client on when he noticed the bug 22:52 <+ddnet-discord> We could send a NETMSG_RACE_STATE aswell to indicate finishes and stuff 22:52 <+ddnet-discord> So might not need that weird parsing 22:55 <@heinrich5991> what is NETMSG_RACE_STATE? 22:56 <+redix> advantage of this parsing is that it works with any version of ddnet or whatever race mod you are running 22:56 <@heinrich5991> yep 22:56 <+ddnet-discord> it will be only for ddnet tab i guess ? 22:57 <+redix> some even leave the minute part out if your time is below 60 seconds. works with this pr :D 22:58 <+ddnet-discord> Wait a second, I am dev? o.O 22:58 <+ddnet-discord> btw, how reliably is the race time counter? 22:59 <+ddnet-discord> thousands ? 22:59 <+ddnet-discord> usually speedrun games have a high precision timer 22:59 <@heinrich5991> 0.02 22:59 <@heinrich5991> seconds 22:59 <+ddnet-discord> :/ 22:59 <@heinrich5991> IIRC 22:59 <@heinrich5991> unless we applied the race patch 22:59 <+ddnet-discord> for example, teesites race severs 22:59 <+ddnet-discord> has thousandths 22:59 <@heinrich5991> yes 23:00 <+ddnet-discord> It would be too late, to change that thing ? 23:00 <+ddnet-discord> i saw 0.0001 on some games 23:00 <+ddnet-discord> I mean, to make bigger precision 23:00 <+ddnet-discord> or maybe not(?) i dont remember 23:00 <+ddnet-discord> 0.0001 seems too high 23:00 <+ddnet-discord> but who knows. 23:00 <+ddnet-discord> yeah xD 23:00 <+ddnet-discord> If i remember, in Trackmania 23:01 <+ddnet-discord> yes 23:01 <+ddnet-discord> i saw precision like that 23:01 <+ddnet-discord> hmm 23:01 <+ddnet-discord> trackmania, every ms counts 23:01 <+ddnet-discord> even half of a ms 😄 23:01 <+ddnet-discord> yes 23:01 <@heinrich5991> nah, trackmania was 0.001 I think 23:01 <+ddnet-discord> I think, more like 0.002 23:02 <+ddnet-discord> i never saw 0.003/5/7 23:02 <+ddnet-discord> it was always even number 23:02 <+ddnet-discord> http://dedimania.net/tm2stats/?do=stat&Envir=Stadium&UId=7yN2DsBnbG8swVNQ81oAcS4pGe5&Show=RECORDS 23:02 <+ddnet-discord> 0.001? 23:03 <+ddnet-discord> nevermind 23:03 <+ddnet-discord> 😄 23:03 <+ddnet-discord> http://dedimania.net/tm2stats/?do=stat&Envir=Stadium&RecOrder3=REC-ASC&UId=oFCBcoNpAKYW7fSu9w5DmdtzKAg&Show=RECORDS 23:03 <+ddnet-discord> ah right 23:03 <+ddnet-discord> xD 23:04 <+ddnet-discord> but what with performance? Could it cause more usage? 23:05 <+ddnet-discord> Btw, i had idea to add "Exclude Gamemode" in filter's 23:05 <@heinrich5991> no 23:06 <@heinrich5991> re performance 23:06 <+redix> well 0.001 precision is mostly useful for fastcap and short runs 23:06 <+ddnet-discord> no, to exlucde gamemode? or to performance? 23:06 <@heinrich5991> [23:06:06] <@heinrich5991> re performance 23:06 <+ddnet-discord> https://ddnet.tw/ranks/solo/#map-Gorou e.e 23:06 <+ddnet-discord> what does it mean 😄 23:07 <+ddnet-discord> i think he means it doesnt affect perfomence ? 23:07 <+ddnet-discord> I meant "re" word. But i guess i understand, it's like "it was about that topic" 23:07 <@heinrich5991> yes 23:08 <@heinrich5991> why exclude gamemodes? 23:08 <+ddnet-discord> sometimes i got brain lags. I am apologize 23:08 <@heinrich5991> doesn't exclude suffice? 23:08 <+ddnet-discord> Not really 23:08 <+ddnet-discord> because, i see o lot of ddrace servers 23:09 <+ddnet-discord> and if I want to see everything else, but not ddrace 23:09 <@heinrich5991> you type ddrace into exclude? 23:09 <+ddnet-discord> it's not really possible. Or maybe i don't know how to use exclude in filter at bottom properly 23:10 <+ddnet-discord> Like that / 23:10 <+ddnet-discord> http://www.wasz.ovh/img/P8nY9yVC80230 23:11 <+ddnet-discord> It looks like it doesn't work 23:12 <@heinrich5991> maybe it doesn't capture game mode, like quick search 23:12 <+ddnet-discord> But, actually its weird 23:12 <+ddnet-discord> it exclude every ddnet server 23:12 <@heinrich5991> because of ddracenetwork in the title 23:12 <@heinrich5991> probably 23:12 <+ddnet-discord> Ah right, because ddnet has "DDrace" word 23:12 <+ddnet-discord> in server name 23:12 <+ddnet-discord> it doesn't work for gamemodes 23:14 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5W9Z 23:14 < ddnet-commits> ddnet/master 7d71dae def: Fetcher: sephamore instead of spinlock, exit thread gracefully 23:15 <+ddnet-discord> what does the fetcher fetch? 23:15 <+ddnet-discord> Maybe i could try to add that thing ? 23:15 <+ddnet-discord> I would learn something 😄 In C 23:15 <@deen> @Ryozuki http downloader 23:15 <+ddnet-discord> ah 23:15 <+ddnet-discord> not always that Web things 23:19 < ddnet-commits> [ddnet] def- pushed 1 new commit to master: https://git.io/v5W9j 23:19 < ddnet-commits> ddnet/master a20db6e def: Don't shutdown sound twice (fixes #851) 23:20 <+ddnet-discord> ha! Cool ~! 23:20 <+ddnet-discord> http://www.wasz.ovh/img/LEMjhhX980301 23:20 <+ddnet-discord> i can use CLion to compile/debug ddnet 23:20 <+ddnet-discord> D: 23:20 <@heinrich5991> ? 23:20 <@deen> What's the status with #853? redix? Should I wait before releasing new client? 23:21 * heinrich5991 is currently reviewing #853 23:23 <+redix> The code fixes at least the bug i encountered. Looks a bit weird maybe but it works and is quite flaxible :D 23:23 <+redix> flexible* 23:33 <+ddnet-discord> NETMSG_SNAPSINGLE, // ? 23:33 <+ddnet-discord> best comment ever 23:33 <@heinrich5991> a snapshot diff contained in only a single message 23:34 <+ddnet-discord> you know the whole tw protocol? 23:34 <+ddnet-discord> nice :o 23:34 <+ddnet-discord> (well most the enums there are obvius) 23:34 <+ddnet-discord> what about 23:34 <+ddnet-discord> NETMSG_SNAPSMALL, // 23:34 <@heinrich5991> not used 23:35 <+ddnet-discord> why its there then? 23:35 <@deen> @Ryozuki You know that heinrich5991 reimplemented the tw protocol in Rust? 23:35 <+ddnet-discord> no, can see source? 23:35 <@heinrich5991> yes, thanks to deen 23:35 <@heinrich5991> heinrich5991/libtw2 23:35 <@deen> thanks to me? 23:36 <@heinrich5991> yes 23:36 <@deen> how so? 23:36 <@heinrich5991> you bugged me long enough to put it on github ^^ 23:36 <@deen> ah, that's what you mean 23:48 <+ddnet-discord> static const unsigned char SECURITY_TOKEN_MAGIC[] = {'T', 'K', 'E', 'N'}; 23:48 <+ddnet-discord> idk why i find lot of the tw code to be funny 23:49 <+ddnet-discord> What do you think, about add to exclude gametypes ? https://github.com/ddnet/ddnet/blob/master/src/engine/client/serverbrowser.cpp#L248-L266 23:52 <@heinrich5991> Savander: better add gamemode support to exclude filter 23:52 <+ddnet-discord> yea, i mentioned that code 23:53 <@heinrich5991> deen: you broke the build 23:53 <@heinrich5991> /Users/travis/build/ddnet/ddnet/src/engine/client/fetcher.h:18:2: error: unknown type name 'SEMAPHORE' 23:53 <+ddnet-discord> i added it now, 😛 But i can't really make pull request 23:53 <@heinrich5991> why not? 23:53 <+ddnet-discord> i mean, i don't know how 😮 23:54 <@heinrich5991> mh 23:54 <@heinrich5991> do you know how to use git? 23:55 <@heinrich5991> deen: (on mac: https://travis-ci.org/ddnet/ddnet/jobs/270160021) 23:55 <+ddnet-discord> nevermind 23:55 <+ddnet-discord> i did it from github.com 23:55 < ddnet-commits> [ddnet] savander opened pull request #854: Add gametype support to exclude field (master...patch-1) https://git.io/v5Wd2 23:55 <+ddnet-discord> im using Git, but for my own project's 23:55 <+ddnet-discord> withour pr's 23:55 <+ddnet-discord> without* 23:55 <@deen> oh grea,t we have no semaphores on mac?! 23:57 <+eeeee> they must be called iSemaphores on mac 23:58 <@deen> cool, file size never worked in teeworlds 23:59 <@deen> makes me wonder why that doesn't cause more problems 23:59 <@heinrich5991> file size? 23:59 <@deen> io_length