Minetest logo

IRC log for #minetest-dev, 2017-04-06

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:17 Lunatrius joined #minetest-dev
00:22 octacian joined #minetest-dev
01:00 Tmanyo joined #minetest-dev
01:25 calculon joined #minetest-dev
02:31 benrob0329-phone joined #minetest-dev
02:45 benrob0329-phone joined #minetest-dev
02:56 benrob0329 joined #minetest-dev
03:02 diemartin joined #minetest-dev
03:04 DI3HARD139 joined #minetest-dev
03:06 diemartin octacian, what I meant is, instead of comparing a version number (which may be different in forks), the client should announce what features it supports and mods should enable/disable stuff based on that instead
03:06 octacian huh, ok
03:06 octacian And how do I get this feature list?
03:07 diemartin and no, the get_client_features thing is hypothetical
03:07 octacian Oh, I see.
03:07 octacian Just realized that lol
03:11 diemartin offtopic: I didn't even see your message, just read that in the logs. ShadowBot notices could be more visible :/
04:05 benrob0329-latop joined #minetest-dev
04:09 SaTa_ joined #minetest-dev
04:17 diemartin joined #minetest-dev
05:01 sofar #5407 two weeks no review, anyone?
05:01 ShadowBot https://github.com/minetest/minetest/issues/5407 -- Hardware coloring for item stacks by juhdanad
05:02 sofar extensively tested by me already
05:05 sofar csm review: 0 for 1. localplayer needs a rebase and it's 30 commits that need squashing
05:09 VanessaE :(
05:09 sofar it's not even that large, lol
05:10 sofar the conflicts should be trivial to fix for the author
05:10 VanessaE I'm waiting for juhdanad's overlay feature to go in, too.  some of my mods need it (I use hacky workarounds that cost FPS)
05:10 sofar same with me with colored itemstacks
05:10 Hunterz joined #minetest-dev
05:10 VanessaE say, did you ever get the chance to play around with UD's auto-coloring feature?
05:11 sofar nope
05:11 VanessaE ok.
05:11 sofar honestly I'm not a creative person, so coloring isn't something I'd ever do
05:12 sofar well, I mean to say I prefer playing survival
05:12 VanessaE right.
05:12 sofar most my minetest worlds are of course just test beds for code
05:12 sofar lol
05:12 VanessaE so, basically it's all air, stone, and water. :)
05:12 sofar that and *cough* secret project
05:14 VanessaE heh.
05:14 benrob0329 ha
05:26 VanessaE meanwhile,
05:27 * VanessaE begs and pleads for someone to do something about mesh gen lag interfering with smooth part of the renderer :P
05:29 VanessaE (active stuff like a mesecons circuit oscillating once a second makes things get choppy because of the repeated mesh regen events)
05:31 benrob0329 perhaps the mesh could simply be optimised every second?
05:31 benrob0329 instead of complealy rebuilt
05:32 VanessaE I doubt that would be possible.  I always figured it could be moved to a lower-priority thread than the renderer.
05:32 VanessaE (or however the two are related)
05:37 nerzhul joined #minetest-dev
05:37 VanessaE there's you a candidate, sofar  :)
05:39 nerzhul hello
05:39 benrob0329 joined #minetest-dev
05:40 nerzhul sofar, to answer, localplayer is available later in the luastack due to the luastack inited before localplayer object was loaded
05:40 nerzhul if i remember we can expose localplayer to Lua directly at stack creation but object is not reliable for mod operations, we choose to expose it after server sent localplayer data
05:41 nerzhul sfan5, are you around ?
05:45 sofar nerzhul: sometimes it never comes up
05:45 sofar https://www.youtube.com/watch?v=sCEYomGL8Lk
05:45 nerzhul hu ?
05:45 nerzhul wtf particles xD
05:46 sofar just a test with too many, seeing how far I can push it
05:46 sofar nerzhul:
05:46 benrob0329 stress test much?
05:46 nerzhul particles for rain i imagine, but i think weather is more a shader than particles :p
05:46 sofar https://gist.github.com/sofar/17acc02aab7c50b53e7103795c465a59
05:46 sofar nerzhul: it's just a test
05:47 sofar torch smoke is actually the first one I'll make I think
05:47 benrob0329 they need to despawn on impact
05:47 nerzhul sofar, interesting usecase
05:47 benrob0329 thats better
05:47 sofar benrob0329: they can, thanks to my patch
05:48 nerzhul sofar, can you look at #5523 it's a tiny server side addition ?
05:48 ShadowBot https://github.com/minetest/minetest/issues/5523 -- API lua get max lag by Dumbeldor
05:48 VanessaE sofar: add smoke to active furnaces also :)
05:48 sofar nerzhul: see that `gist`? It sometimes *never* gets localplayer
05:48 nerzhul sofar, it's very strange
05:48 nerzhul https://github.com/minetest/minetest/pull/5480/files#diff-34f48ad91ac6c202ac60b0348ae90e30R1869
05:49 nerzhul object is exposed here ^
05:49 VanessaE (that, btw, is why HDX's furnaces have a rectangular dark spot on top - to look like it has a flue)
05:49 nerzhul and it's unconditional if you enable client modding
05:51 sofar nerzhul: idk, it worked a few times just fine
05:52 nerzhul for #5525 we just need to move helpers from game to common
05:52 ShadowBot https://github.com/minetest/minetest/issues/5525 -- [CSM] missing vector helpers.
05:54 sofar yeah, it was more a note ;)
05:54 nerzhul where was they declared... hmmm
05:54 sofar same with the yaw/pitch/dir stuff
05:54 sofar does CSM have minetest.after? is it reusing the same code from server side?
05:54 sofar oh, dumbass
05:54 nerzhul vector is in common
05:55 sofar yes it has after()
05:55 nerzhul oh we just forgot to include it :p
05:55 sofar after is in common too
05:56 nerzhul yes and it's already included
05:57 nerzhul i push the trivial change to include vector into game
05:58 sofar :+1:
05:59 nerzhul done
06:04 calculon joined #minetest-dev
06:06 sofar #5516 is fine with me
06:06 ShadowBot https://github.com/minetest/minetest/issues/5516 -- [CSM] Add support for positional audio. by red-001
06:07 nerzhul sofar, we can merge it then ?
06:08 VanessaE *sniff*  do I smell an Ambience Mod update in the near future? :P
06:08 ssieb joined #minetest-dev
06:09 benrob0329 /me gets exited
06:09 diemartin pls
06:14 sofar nerzhul: it's too small of a change to make a fuss about
06:14 nerzhul merging
06:14 sofar nerzhul: besides, if you haven't reviewed it in like 2 minutes... lol
06:14 sofar VanessaE: not yet
06:14 VanessaE aw :(
06:14 sofar still need localplayer api as well
06:14 VanessaE oh right
06:14 sofar did we get get_node already?
06:14 VanessaE cheapie's waiting on that for a chat6 feature also
06:15 nerzhul sofar, ofc, but i needs some later polish & rebase :p
06:15 VanessaE I'll admit I was kinda "meh" on CSM, but I'm starting to see the value in it :)
06:15 nerzhul VanessaE, without API the interest was limited at start, but as i said, we need more and more API :p
06:15 nerzhul sofar, #5499
06:15 ShadowBot https://github.com/minetest/minetest/issues/5499 -- [CSM] Add `on_joinplayer` and `on_leaveplayer` callbacks. by red-001
06:16 VanessaE nerzhul: indeed, and it'll come.
06:16 diemartin I'm sure Minetest felt this way in the first 0.4 :P
06:17 diemartin or whenever it was the Lua API was added
06:18 nerzhul celeron55, can you create client mods sections on forum as more and more users are interested on CSM, this will permit to communicate about this and make ideas pop in the same forum area :)
06:18 VanessaE why not just prefix it [CSM] instead of [Mod] ?
06:20 nerzhul it can be, but as it's not the same usecase i think it's better to have a clear separation, and it also permits to promote CSM directly and prevent end users to mismatch the mod installation between cleint part and server part
06:21 nerzhul i will prepare a forum post about the future release and a teasing changelog
06:22 VanessaE it's quite okay for a client-side mod that doesn't interact much with a server, such as chat6 or perhaps ambience, but I'm still very much against the notion of "client part and server part"; if a mod needs part to run on the client and part on the server, the server should send the pieces the client needs.
06:23 red-001 of course
06:24 red-001 we don't want to end up like minecraft
06:24 VanessaE precisely.
06:24 VanessaE plus, users can...show a distinct lack of intelligence.
06:24 VanessaE asking them to go fetch this or that mod is like asking them to go fetch a gold bar from Fort Knox :P
06:25 nerzhul when we send mods from server to client it will be part of server mod in fact :)
06:25 nerzhul but we will see this use case after release
06:25 VanessaE nerzhul: I assumed as much.
06:25 sofar joined #minetest-dev
06:27 benrob0329-phone joined #minetest-dev
06:40 nerzhul i pushed translations
06:43 VanessaE did they push back? :)
06:44 nerzhul ?
06:44 VanessaE nevermind, just making a joke.
06:45 nerzhul i think i'm always in my bed :p
06:59 celeron55 i wonder how to actually organize that
06:59 VanessaE ?
07:00 celeron55 should the new section be called "Client-only Mods" or something
07:00 VanessaE may as well keep it consistent - client-side mods.
07:01 celeron55 how do we differentiate between those and the ones that have a server- and a client-side part in the same package in the future?
07:01 VanessaE Hybrid mods?
07:02 paramat joined #minetest-dev
07:02 VanessaE or "combination mods" maybe
07:03 VanessaE modders should put both parts of their mods in one post.  seems pretty straightforward to me.
07:04 celeron55 there's no need for that when it's just part of the server-side mod and the servere just sends the stuff to the client
07:04 VanessaE right.
07:04 celeron55 the way they will be used will be exactly the same as current server-only mods
07:04 celeron55 that's the confusing thing
07:06 VanessaE well I got nuthin' then :)
07:07 celeron55 and they will be posted right into wip mods and mod releases just like the current ones
07:07 celeron55 but we'll also have these that don't have anything to do with the server; they need a place
07:09 VanessaE if it were up to me, I'd just put them all in Mod Releases and replace the [Mod] prefix in the topic with something appropriate.
07:09 VanessaE s/topic/title/
07:10 celeron55 i think i could add a subforum under "Mods" called "Client-side modding", which would hold both mod releases and dicussion, and then things can be separated further once we know what is needed, if anything
07:10 VanessaE that could work.
07:11 nerzhul i push a little lint optimization for local use + recent regression with red-001 pr merge
07:12 celeron55 if nerzhul is ok with the suggestion, then i'd like nerzhul to make the first post in there so that the first post isn't just some random "oh i wonder what this is about?"
07:12 celeron55 i can add some guidelines later, i don't think they're needed immediately
07:13 celeron55 or, maybe we want a random post like that
07:13 nerzhul celeron55, whyu not, i'm going to work, see you in 30 mins to discuss about his
07:14 celeron55 hmm, i think i'll add it right now and see what people start talking about
07:14 celeron55 it'll be interesting
07:14 VanessaE careful
07:14 red-001 you might want to move the existing mods
07:14 VanessaE don't tempt the offtopic'ers....
07:14 celeron55 no? well maybe not then 8)
07:15 VanessaE next thing you know azekill_DIABLO and ABJ will be in there trolling :)
07:15 VanessaE and MineYoshi or whatever it was
07:15 red-001 lol
07:16 celeron55 ok i'm adding it; if it gets out of hand, i'll just delete it and re-create it later
07:16 red-001 I wouldn't be surprised if most of the posts on the forums are in offtopic
07:17 celeron55 well it's like 10%; you can see the stats on the index page
07:20 epoch I have problems jumping up one block whem my FPS is around 2.
07:20 epoch :>
07:21 VanessaE epoch: then consider raising your FPS. :P
07:22 epoch can I disable textures?
07:22 epoch or lower viewing distance to 3?
07:23 celeron55 what kind of toaster is it
07:24 VanessaE it's not a toaster, it's a potato :)
07:24 celeron55 maybe one designed for four bread slices will be faster
07:24 VanessaE I thought Minetest wasn't multi-breaded? :)
07:24 epoch just a single core toaster.
07:25 epoch it sure makes good toast though.
07:25 epoch evenly brown on both sides.
07:27 celeron55 do you get 2 fps with shaders and everything disabled and viewing_range=1 and enable_fog=0?w
07:27 VanessaE ok, time for me to stop trolling an go to bed :P
07:27 celeron55 -w
07:27 nrzkt joined #minetest-dev
07:28 nrzkt Errr sorry on last commit i forgot to restore clang-format binary used..' i will push the ci Fox at work
07:28 nrzkt Fix*
07:28 epoch all the shaders are disabled. left only partiticles on.
07:28 epoch had fancy leaves.
07:29 epoch was using 20 viewing distance
07:29 celeron55 20 is obviously too much for it
07:29 celeron55 you can't set it with +/- but if you edit minetest.conf before launching the game, it'll take 1 just fine
07:30 celeron55 that'll cut drawn vertices in half from what you have now and you will probably get 4 fps
07:30 epoch I was using - and it stopped at 20
07:30 paramat use opaque leaves
07:31 epoch opaque is faster than simple?
07:31 paramat yes
07:31 paramat er, maybe
07:31 epoch viewing_range complains about stuff less than 20.
07:31 celeron55 edit the config file directly; your text editor won't complain
07:32 epoch yep yep
07:33 calculon joined #minetest-dev
07:35 epoch with viewing_range at 3 and fog still enabled it is upper 20s not moving.
07:36 celeron55 oh, that's amazing
07:36 epoch I usually just avoid the areas that use a lot of textures and blocks.
07:37 epoch and stuff is playable.
07:37 celeron55 where do you live? i'd imagine in most parts of the world you could get a better computer from any junk pile anywhere
07:37 epoch it is an alright piece of mid 00s junk.
07:38 epoch the laptop my kid is playing it one runs it just fine.
07:38 epoch on*
07:39 epoch I like it better without the fog.
07:39 celeron55 i originally developed MT on a late-00s high-end laptop with an intel gpu and it's still the lowest recommended set-up
07:41 epoch I have 1.5GB RAM, model name: Intel(R) Pentium(R) M processor 1.86GHz, Debian.
07:46 nrzkt joined #minetest-dev
07:49 nrzkt joined #minetest-dev
07:50 kilbith joined #minetest-dev
08:35 nrzkt pushing a last lint fix since recent merges & push
08:54 paramat how do i make a setting serverside? #5519
08:54 ShadowBot https://github.com/minetest/minetest/issues/5519 -- Sneak: Add option for classic move code by paramat
09:06 ssieb joined #minetest-dev
09:26 celeron55 paramat: a generic system for that doesn't exist; only custom fields in the base protocol and eg. privileges and other specific things are currently used in a similar purpose
09:26 nrzkt joined #minetest-dev
09:27 celeron55 you might have to make it if you want it :P
09:29 celeron55 arguably it should be a player-specific setting
09:29 celeron55 if it's global to the server, then there's no option to have it differ between players; if it's player-specific, then it's easy to just make it the same for everyone while leaving other options
09:34 paramat ok. yeah i was thinking it might actually be ok clientside
09:35 paramat differing players on a server will want the option
09:44 celeron55 i meant similar to set_physics_override
09:44 celeron55 probably inside it, actually, in this case
09:44 celeron55 but dunno
09:48 YuGiOhJCJ joined #minetest-dev
10:04 Fixer joined #minetest-dev
10:12 paramat i see. physics override is per-player but dictated by server, hmm
10:13 troller joined #minetest-dev
10:13 paramat currently the 'sneak glitch' physics override controls the option for sneak ladder in new move code
10:14 paramat players would have to request their old/new move code setting to the server admin though
10:19 celeron55 well, if the admin wants to allow both, just like add a chat command or something
10:20 celeron55 there are many ways to go about this
10:20 celeron55 one could think of it as a privilege too, similar to fly or fast or noclip
10:23 celeron55 but the only way that allows everything is the physics override; then regular mods can do anything with it
10:45 YuGiOhJCJ hello, my PR https://github.com/minetest/minetest/pull/4622 has now a conflict that must be resolved, I think I found the reason of this conflict: the "src/CMakeLists.txt" file has been recently modified by this commit: https://github.com/minetest/minetest/commit/86b1542181a92841d174e1caec94d084a34e5158#diff-95e351a3805a1dafa85bf20b81d086e6 where the LuaJIT detection has been removed from here, should I do an other PR
10:45 YuGiOhJCJ from an other branch in a new fork? (remark: I did my PR on the master branch of my fork which is not recommended)
10:55 celeron55 you can make a new PR if you'd like to switch away from your master branch anyway
10:59 paramat hmm chat command that sets a player's physics override, that would give a server the choice to either dictate old/new code or allow players to choose, seems best
11:00 YuGiOhJCJ celeron55, ok I will do that because I will be able to resolve the conflict easily starting from scratch and my code will be in a dedicated branch instead of in the master branch which is recommended
11:03 nrzkt YuGiOhJCJ: just rebase your branch
11:04 nrzkt you can also erase your branch, recreate a proper branch and force push the new change to the remote branch (same) it permits to update pr
11:05 YuGiOhJCJ rebase which branch? in my fork I only have one branch, the master branch
11:05 nrzkt okay
11:05 nrzkt then, do git reset --hard HEAD~1 to revert your commit, next, git remote add upstream https://github.com/minetest/minetest.git && git pull upstream master && git push origin master
11:07 YuGiOhJCJ hehe that was my first PR so that's why I did it wrong, OK I will try to revert some commits like you suggest instead of removing my fork and start from a new one
11:11 celeron55 is it common practice in cmake to add string values to something that was a boolean before?
11:11 celeron55 like adding REQUIRED to something that was TRUE/FALSE before?
11:11 Hunterz joined #minetest-dev
11:11 YuGiOhJCJ I must admit that I am not totally convince by this solution
11:12 celeron55 i guess it'll cope with it fine even if someone used 1/0 in place of TRUE/FALSE before so maybe it works
11:12 YuGiOhJCJ I prefer to keep a TRUE/FALSE boolean
11:12 sfan5 paramat: physics overrides
11:12 celeron55 i would have just added a new setting called REQUIRE_SYSTEM_LUAJIT
11:12 celeron55 or something
11:12 YuGiOhJCJ oh OK I see that was one of the suggestions
11:13 YuGiOhJCJ I think it's better indeed
11:13 Megaf joined #minetest-dev
11:13 celeron55 oh i see, it was already suggested
11:13 YuGiOhJCJ after discussion we have decided to use a string... not sure if it is the best choice
11:15 YuGiOhJCJ nrzkt, after 'git push origin master' I got ' ! [rejected]        master -> master (non-fast-forward)'
11:16 YuGiOhJCJ 'hint: Updates were rejected because the tip of your current branch is behind its remote counterpart. Integrate the remote changes (e.g. 'git pull ...') before pushing again.'
11:21 paramat yes will move setting to a physics override
11:24 paramat i'm also making it clear this is certainly temporary, as in probably until end of next release, which will probably coincide with 0.5, which gives us an excuse to break stuff
11:27 Player_2 joined #minetest-dev
11:43 YuGiOhJCJ ok found, that was just because I need to "force" the push
11:44 celeron55 YuGiOhJCJ: you may want to force-push (but be sure you know what it means in your case before you do)
11:44 YuGiOhJCJ hehe :D
11:44 celeron55 doh 8)
11:46 blaze joined #minetest-dev
11:50 nrzkt yes sorry i forget to tell you force :)
11:50 nrzkt YuGiOhJCJ: seems your PR disappear
11:51 YuGiOhJCJ yes it has been automatically closed after I typed the git commands
11:52 YuGiOhJCJ well it does not matter I will just create a new PR now from a new branch of my fork
11:55 troller joined #minetest-dev
12:49 Hunterz joined #minetest-dev
13:03 YuGiOhJCJ ok the new PR is here: https://github.com/minetest/minetest/pull/5526
13:17 Megaf joined #minetest-dev
13:27 nrzkt i will push a clang-format rule fix (moving braces from Linux to custom because we don't fully use linux style for braces)
13:33 nrzkt our continuation indent is 1 or 2 tabs ?
13:34 Megaf left #minetest-dev
13:34 sfan5 2
13:34 nrzkt nice, i'm fixing this rule in clang-format too
13:35 nrzkt i'm pushing the continuation indent fix for clang format, with src/wieldmesh.h style fix using clang-format. I also include a .gitignore addition for CLion IDE which uses cmake-build-debug and cmake-build-release as default directories for building cmake probjets
14:04 benrob0329 joined #minetest-dev
14:06 octacian joined #minetest-dev
14:06 octacian joined #minetest-dev
14:15 RobbieF joined #minetest-dev
14:17 troller joined #minetest-dev
14:18 YuGiOhJCJ joined #minetest-dev
14:36 XtremeHacker joined #minetest-dev
14:39 twoelk joined #minetest-dev
14:50 DS-minetest joined #minetest-dev
14:51 Lunatrius joined #minetest-dev
15:05 paramat #5527 , also makes simple sneak ladder work, but needs testing for problems
15:05 ShadowBot https://github.com/minetest/minetest/issues/5527 -- Sneak: Detect ledge for 2-node pull-up by paramat
15:07 red-001 could https://forum.minetest.net/viewtopic.php?f=9&t=17092 , https://forum.minetest.net/viewtopic.php?f=9&t=17046 and https://forum.minetest.net/viewtopic.php?f=9&t=17052 be moved to the new csm mod section on the forums?
15:09 DS-minetest oh, a csm section. why isn't it split up into WIP and release?
15:10 benrob0329 because csm itself is wip?
15:10 nrzkt i don't see mod section ? celeron55 did you hide it ?
15:10 Megaf joined #minetest-dev
15:10 red-001 https://forum.minetest.net/viewforum.php?f=53
15:10 nrzkt benrob0329: CSM is WIP but it doesn't mean we block it
15:10 benrob0329 no, not block it
15:10 red-001 it doesn't show up on the board index
15:11 nrzkt if i remember celeron55 can you show it on board index ?
15:11 DS-minetest https://forum.minetest.net/viewforum.php?f=46 it's shown
15:11 red-001 up it does in the user-content subsection
15:11 nrzkt nice, i cannot post to it
15:11 nrzkt xD
15:11 red-001 I assume it's some sort of bug in the forum software
15:11 nrzkt celeron55: also we need to write inside it :p
15:13 celeron55 meh, it needs to work also?!
15:13 celeron55 i thought just adding it would be enough
15:14 benrob0329-phone joined #minetest-dev
15:16 nrzkt xD
15:16 nrzkt nice i got rights, can i add posts as stick ?
15:17 red-001 https://forum.minetest.net/viewtopic.php?f=17&t=16838
15:17 red-001 I wonder for how many more years will this guest killing go on for
15:19 benrob0329-phone Till the end of time itself
15:20 celeron55 that topic stopped to make sense before it was even created
15:21 red-001 that's true generating guest names got broken quite some time ago
15:23 nrzkt celeron55: can you stick this post ? https://forum.minetest.net/viewtopic.php?f=53&t=17176&p=262141#p262141 (i can't remember the word for making it in top page in the section
15:23 nrzkt i didn't finish it, but it's the startup
15:24 celeron55 it's now a sticky
15:24 nrzkt ty
15:24 * celeron55 closes glue tube
15:25 celeron55 altough maybe it should be an announcement
15:25 celeron55 well whatever for now
15:26 nrzkt celeron55: could be nice too
15:26 DS-minetest make it locked
15:26 celeron55 apparently it's possible to post to it now, then? i didn't change anything
15:27 DS-minetest it's locked now
15:28 celeron55 yes i locked it; but i was wondering what nrzkt meant with "celeron55: also we need to write inside it :p"
15:29 red-001 could someone move the posts I mentioned?
15:29 DS-minetest that were topics, not posts
15:29 nrzkt celeron55: if you lock it i cannot edit it xD
15:29 celeron55 lol
15:30 celeron55 i made it show up in the index now
15:30 benrob0329 joined #minetest-dev
15:30 celeron55 and unlocked it
15:32 nrzkt ty
15:36 benrob0329 joined #minetest-dev
15:40 nrzkt red-001: maybe VanessaE can move your posts ?
15:42 troller joined #minetest-dev
15:53 benrob0329 joined #minetest-dev
16:01 twoelk left #minetest-dev
16:13 YuGiOhJCJ joined #minetest-dev
16:36 Megaf joined #minetest-dev
16:48 red-001 why is it not possible to post to the csm section?
16:50 red-001 nvm
16:50 red-001 I wasn't logged in
16:59 paramat joined #minetest-dev
17:00 Megaf joined #minetest-dev
17:05 Darcidride joined #minetest-dev
17:18 Fixer https://forum.minetest.net/viewtopic.php?f=6&t=17165 lighting related bug?
17:27 Megaf joined #minetest-dev
17:32 Krock joined #minetest-dev
17:32 Krock joined #minetest-dev
17:34 Megaf joined #minetest-dev
17:41 ssieb joined #minetest-dev
17:41 benrob0329 joined #minetest-dev
17:43 benrob0329 joined #minetest-dev
18:09 red-001 #5529
18:09 ShadowBot https://github.com/minetest/minetest/issues/5529 -- [CSM] Add function to get the server protocol version. by red-001
18:12 twoelk joined #minetest-dev
18:23 DS-minetest joined #minetest-dev
18:29 nerzhul joined #minetest-dev
18:32 numzero joined #minetest-dev
18:39 rubenwardy joined #minetest-dev
18:40 nerzhul sofar, wow i missed Linuxdirk speach on https://forum.minetest.net/viewtopic.php?f=53&t=17046
18:40 red-001 you didn't miss much
18:41 sofar yup
18:42 nerzhul we don't have a closed source client, you can modify all and do all your cheat shit easily to know where ores are etc, with Lua API it's just simpler, a range limit could be good maybe, but easily removable patching core :)
18:42 nerzhul oh i answered on that post, i forget :p
18:49 Megaf joined #minetest-dev
18:51 rubenwardy FUCK
18:51 Krock eh
18:51 Krock quick, force push
18:52 rubenwardy fixed
18:52 rubenwardy I'm testing out a GUI git client, some reason it pushed to upstream when I said origin
18:52 Krock -- nothing happened -- ;)
18:53 Krock still need to type in password for upstream here. makes it much more fail-safe
18:53 rubenwardy same
18:53 rubenwardy it didn't ask me for the password though
18:53 Krock strange
18:53 rubenwardy as it's linked to my GH account
19:00 rubenwardy yay, I set the "push URL" of upstream to git@null, so it shouldn't be able to push now
19:05 numzero joined #minetest-dev
19:08 rubenwardy turns out that because I pulled from upstream/master, it then set that as the push location
19:27 nerzhul what happended ?
19:28 nerzhul i need to find why a egrep -c exit 1 the CI...
19:30 rubenwardy My GUI clean messed up and pushed to upstream, I then forced-pushed to remove the commit 2 minutes later
19:31 nerzhul which GUI client ?
19:32 rubenwardy GitKraken
19:32 nerzhul it seems egrep is buggy
19:32 nerzhul gitg is a nice GUI, light and does some trivial things nicely
19:32 rubenwardy it decided to change my upstream remote when I told it to git pull upstream master
19:33 rubenwardy I want it for a nice tree view of the history
19:33 rubenwardy gitg doesn't seem to do that
19:33 nerzhul egrep -q should return 0 if no match it returns 1...
19:33 rubenwardy https://blog.axosoft.com/wp-content/uploads/2015/10/1-main-view.png
19:34 nerzhul git lg ?
19:34 nerzhul [alias]
19:34 nerzhul lg = log --oneline --all --graph --decorate --color
19:34 nerzhul add this to your .gitconfig
19:34 nerzhul and type git lg
19:34 nerzhul :D
19:37 Hijiri Was it all items that are limited to ~30000 registrations, or just nodes?
19:37 Krock if there's a limit then it's at 65535 registrations
19:38 Krock items and nodes together AFAIK
19:38 sfan5 nerzhul: i use
19:38 sfan5 lg = log --color --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %s %Cgreen(%cr)%C(bold blue)<%an>%Creset' --abbrev-commit
19:38 sfan5 does yours look different by much?
19:39 rubenwardy different colors, no author
19:39 sfan5 Hijiri: somewhere in the protocol an u16 is used to store the number of items/nodes probably, limiting it to 2^16
19:50 Hijiri ah
19:50 Hijiri so the use of hacks like registering 100 versions of a pick for 100 dig levels would be a bad idea
19:50 sfan5 yes very
19:51 sofar itemstack meta may be a solution?
19:51 Hijiri yes, but only once someone implements dig level overriding
19:53 Hijiri which seems like it would be not too hard to implement
19:53 Hijiri maybe I will try to do it
19:54 sofar +1
19:57 Hijiri ItemStack already has a getToolCapabilities, looks like there's an obvious place to change
20:07 rubenwardy #5506
20:07 ShadowBot https://github.com/minetest/minetest/issues/5506 -- Add Joystick type detection and Xbox controller support by rubenwardy
20:09 Hijiri I'm thinking the modified toolcaps is passed as a serialized JSON table in a "toolcaps" key, does that sound fine?
20:10 sfan5 wouldn't it be nicer to be able to define several toolcaps ahead of time and have "toolcaps" just be an int
20:11 sfan5 would be less flexible but nicer imo
20:11 Hijiri I can think of cases where you would want the extra flexibility
20:11 Hijiri like stacking enchantments
20:12 Hijiri unless I define 100 toolcaps or something
20:15 nerzhul rubenwardy, sfan5 for the release, are you okay for middle may ? do you think it's possible ? will we finish the sneaking problems ?
20:17 nerzhul other question, can we tell 0.4.16 will be the last 0.4.X release permitting us to fix some problems and make 0.5 a really, as CSM is a preview for next release and should be massively enhanced ?
20:18 rubenwardy pretty sure server side modding was still a massive WIP in 0.4.0
20:18 benrob0329 http://semver.org/
20:18 nerzhul without doubt, CSM is a sufficient step for a 0.5 and i think all of us see some things which needs refactor
20:19 nerzhul benrob0329, we don't follow semver as we have an extra useless 0. :p
20:19 red-001 benrob0329, not as if we use ot
20:19 red-001 s/ot/it
20:19 benrob0329 Its just a suggestion..
20:19 rubenwardy benrob0329, we don't use semvar
20:19 rubenwardy we should use semvar, but meh
20:20 RobbieF joined #minetest-dev
20:21 nore well, with the leading 0, we *do* follow semver :)
20:21 nore if we are still doing the "initial development"
20:21 nerzhul nore not exact as we don't have patch release
20:22 nore nerzhul: no, that's what I'm saying
20:22 nerzhul nore rubenwardy is this okay for you if we try to plan release middle may ?
20:22 nerzhul i'm preparing a teasing post on forum
20:22 nore we're following semver because 0.x.y is legal and anything may change at anytime (lol)
20:22 nore mid-may is ok for me I think, need to see what should be merged before then
20:23 rubenwardy it's fine
20:24 nerzhul nore, yes i need to finish my players to db backend (the retro compat mode by default for this release with a warning)
20:24 twoelk lol, and I had hoped the next release might be called 0.5.0, because it allready breaks loads of stuff and have csm become default with 0.6.0 - and thus getting closer to semver
20:24 paramat about release .. if devs want a release ASAP that's ok with me but there needs to be more notice, 6 weeks. 2 weeks will be freeze and we need at least a month to finish up big projects
20:25 nerzhul paramat, it's why i talk about this now
20:25 sfan5 middle may? maybe
20:25 sfan5 but definitely not before that
20:25 nerzhul 1 month to fix the sneak problems, finish the players to db and add last remaning interesting calls on CSM
20:26 sofar june 1st to nail it down?
20:26 paramat i'd prefer 2 months notice
20:26 sofar 8 weeks time
20:26 * twoelk thinks mid-may a wonderfull time as his birthday is thereabouts
20:26 sofar who's doing the release? paramat need a break?
20:26 paramat it wasn't me last time :] we managed it together
20:26 nerzhul what should be done for the release ? for Android i fixed the compat problems it should be okay for me
20:27 nerzhul 28 may is a sunday, or 4th june, as you like
20:27 paramat no need for 1 release manager, it's stressful and unpleasant anyway
20:28 nerzhul paramat, no need, we just need to discuss , maybe it could be nice to use a little bit more the forum with polls inside our section :p
20:28 paramat 4th june is good for me
20:28 nerzhul many users seems to complain on forum we don't communicate
20:28 nerzhul sofar, nore rubenwardy 4th june ?
20:28 twoelk maybe a dev-version run test server might be usefull - with events for some mass related tests ;-P
20:28 sofar fine with me
20:29 nore Okay for me
20:29 nerzhul okay i will communicate on forum about this date
20:29 sofar running a -dev daily server sounds interesting
20:29 Hijiri item metadata refs are newer than stable, right?
20:29 sfan5 yes
20:29 nore +1
20:29 Hijiri sfan5: thanks
20:29 nerzhul sofar, a daily dev server ? but to add what ?
20:29 sofar invite people to crash it, try new features etc.
20:29 twoelk such server could run massiv logs or have datamonitoring tools added
20:30 VanessaE if I weren't afraid of breaking shit, I'd add a cron job to update my servers' build nightly.
20:30 nerzhul i can run it without problem if needed i have many CPU unused on my servers (i will plug postgresql databases)
20:30 sofar better properly isolate it :)
20:30 sofar have it restart every time a commit is added to master
20:30 sofar in either mtg or minetest
20:30 nerzhul sofar, a little bit over engineered, 1 time per 12h can be sufficient :p
20:31 nerzhul imagine i commit translations bim 12 restart :p
20:31 twoelk have irc attached to minetest-dev
20:31 nerzhul i can plug my C++ library on my MT server to talk on IRC and Slack but, lol :p
20:32 benrob0329 Why not have it update every hour if there is a commit?
20:33 nerzhul i'm okay to have this server if needed, but not now, maybe i can prepare work on it next week
20:33 nerzhul i have 8 CPU and 32GB ram on my server, i can use some of them for MT :p
20:33 twoelk during feature freeze might be a good time
20:34 nerzhul ofc
20:35 nerzhul running in debug mode and my FreeBSD has coredump enabled i can easily use lldb or gdb to look at crashed if related to cpp
20:35 paramat and we intend to have 2 weeks feature freeze? i know it sounds a long time but we needed this last time, it can always be cut short if things are good. so freeze 2 weeks beofre 4th june?
20:35 benrob0329 Living on the bleeding edge is the only way to go
20:35 nerzhul paramat, i'm okay for 2 weeks
20:35 VanessaE whoa whoa
20:36 VanessaE before feature freeze there are some things that need to get in
20:36 sofar VanessaE: look at the dates
20:36 VanessaE I saw.
20:36 nerzhul seems reasonable and we have 1 month and 2 weeks to merge what we need to merge
20:36 sofar almost 6 weeks until freeze?
20:36 nerzhul maybe you want me to open a roadmap issue to list features we want to close before feature freeze ?
20:36 VanessaE but you know as well as I do that if someone doesn't jump on it now, it'll be 1-2 days into freeze before someone realizes "feature X" was needed before relse.
20:36 VanessaE release*
20:37 nerzhul sofar, 4th june - 2 weeks ~ 6 weeks
20:37 sofar that's why we should start filling the milestone
20:37 nerzhul VanessaE, we have all features for a release, the 0.4.16 changelog is the biggest release we did in the last 2 years, at least
20:38 VanessaE nerzhul: I know.  just trying to keep feature priorities in mind :)
20:38 nerzhul it's why i propose to open a roadmap PR like we did on CSM, permitting core devs to list all needed features to have before FF, what do you think about it rubenwardy nore sofar sfan5 ?
20:38 VanessaE right now the only item in the 0.4.16 milestones is CSM stuff.
20:39 rubenwardy The blocker label is for this
20:39 rubenwardy or the milestone
20:39 VanessaE rubenwardy: the blocker label doesn't always work
20:39 paramat yes use 'milestone'
20:39 VanessaE I've seen issues go unresolved between releases that were tagged "blocker" and then got put off.
20:40 sofar roadmap PR is ok, but removing items from the milestone should get major approval
20:40 nerzhul oh sofar yes, it's better to use milestone
20:40 nerzhul i forgot this important github/gitlab thing for mt, i'm using it everytime at work :p
20:40 sofar well but it's hard to discuss milestone adding/removal in github
20:41 VanessaE I'm just trying to do what little I can to try to steer things in a sane way :)
20:41 nerzhul sofar, maybe have an issue in the milestone for discussion and PR/issues needed for milestone into it :)
20:43 nerzhul #5361 is very interesting PR for formspec why not add it ?
20:43 ShadowBot https://github.com/minetest/minetest/issues/5361 -- FormSpec : Add background color for textarea/field/pwdfield by adelcoding1
20:44 * benrob0329 should redo the dropped item animation before going on vacation
20:45 nerzhul i started to add some PR to milestone, don't hesitate to add it
20:45 VanessaE another item to add:
20:46 VanessaE wasn't there a PR to improve the mesh tiling?
20:46 VanessaE e.g. fewer polys for the same large flat areas
20:46 nerzhul there is too many PR we should close old PR if nobody wants to integrate it :(
20:46 sofar I suggest we just throw a ton of things in and then as a team work through them
20:47 sofar I mean, I'm partial to itemstack coloring myself :)
20:47 sofar and the PR is done
20:47 VanessaE #5057 is too useful not to merge, people keep screaming for #1118 to go in,
20:47 ShadowBot https://github.com/minetest/minetest/issues/5057 -- Undersampling by numberZero
20:47 ShadowBot https://github.com/minetest/minetest/issues/1118 -- Meta set nodedef by Ekdohibs
20:48 nerzhul #5530 roadmap issue opened
20:48 ShadowBot https://github.com/minetest/minetest/issues/5530 -- Minetest 0.4.16 roadmap
20:48 VanessaE can't find the fewer-polys PR
20:49 GreenDimond joined #minetest-dev
20:50 nerzhul sofar, #5529 approval needed when you get time :)
20:50 ShadowBot https://github.com/minetest/minetest/issues/5529 -- [CSM] Add function to get the server protocol version. by red-001
20:50 sofar one was merged a while back
20:50 sofar nerzhul: that one is trivial, :+1:
20:50 nerzhul merged
20:50 sofar afk espresso run
20:51 VanessaE #5186, soft node overlay
20:51 ShadowBot https://github.com/minetest/minetest/issues/5186 -- Soft node overlay by juhdanad
20:51 VanessaE (that's waiting on some refactor PR)
20:51 nerzhul milestone view is updated dynamicly, fun
20:51 GreenDimond I just read the chat logs for today, and I am slightly confused... at first you were talking about the release date being mid-May... but then begining of June? Which one is it?
20:52 nerzhul GreenDimond, june, i'm writing the post on the forum atm, but i need to finish to write the current interensting changes
20:52 GreenDimond Ah, thankyou :)
21:03 calculon joined #minetest-dev
21:19 VanessaE #3502 needs reconsideration also.
21:19 ShadowBot https://github.com/minetest/minetest/issues/3502 -- Far map and improved map transfer and rendering (WIP) by celeron55
21:20 nerzhul maybe on next release for this, the change is too huge
21:20 VanessaE note the date..
21:20 VanessaE that should have gone in for 0.4.15.
21:21 VanessaE this one's an example of things being put off for too long, tbh.
21:22 nerzhul it's far from being finished
21:22 VanessaE and it never will be.
21:22 VanessaE not at this rate. :)
21:22 nerzhul personnaly i don't have time for this i have 5 PR opened to finish which needs some polishing + i want to do some more performance optimizations on the C++ code client side
21:23 VanessaE just fyi, you advised merging it after 0.4.14 :)
21:23 nerzhul i will push a PR to fix linter in CI because egrep is a shitty command which have wrong documentation :p
21:23 nerzhul i move egrep in linter to awk which works as intended
21:24 nerzhul + code style fix in non whitelisted files since recent brace style fix in clang-format rules (to make them match our rules which are a little bit different than linux baremetal rules)
21:24 nerzhul VanessaE, i remember but i don't tell merge it if it's not finished :p
21:24 VanessaE I only said it needs reconsideration :)
21:25 nerzhul 173 PR needs reconsideration xD
21:25 * VanessaE shrugs
21:26 nerzhul i posted release announce https://forum.minetest.net/viewtopic.php?f=18&amp;t=17179
21:28 nerzhul paramat, if i remember you can change the channel topic here no ?
21:28 nerzhul i don't have rights for this it seems
21:29 sofar can we add all BUGFIX prs to 0.4.16 milestone?
21:29 VanessaE sofar: +1
21:29 sofar that should really be self-explanatory
21:29 sofar I can do it, btw
21:29 red-001 are signed commits still an issue?
21:30 sofar no
21:30 VanessaE #4427 should be closed -- this should be handled with CSM
21:30 sofar canonical is dead
21:30 ShadowBot https://github.com/minetest/minetest/issues/4427 -- Implement client-side auto-pickup by Rogier-5
21:31 sofar VanessaE: tricky, client-side pickup would require allowing the client mods to manipulate entities on the map
21:31 red-001 ^
21:31 sofar that's a security issue
21:31 sofar server-side mod, sure
21:31 VanessaE sofar: cannot CSM trigger a simple "punch" action?
21:31 sofar it can not
21:31 red-001 VanessaE, it could but do we want it to be able to?
21:31 VanessaE indeed I have a mod on my servers that does it.
21:31 paramat nerzhul i can't
21:32 VanessaE red-001: is it any different from the user punching an item manually?
21:32 nerzhul sofar, it's a milestone, it's tasks to do for release
21:32 red-001 VanessaE, it could be used to gain an advantage in pvp
21:32 VanessaE but ok, point taken.
21:32 sofar auto-aim auto-trigger
21:32 red-001 e.g. auto punch
21:32 nerzhul red-001, signed commits are not an issue, as you see i updated weblate and daily ubuntu builds works
21:33 nerzhul VanessaE, CSM doesn't write anything except in the console if i remember
21:33 red-001 maybe it should be removed from the git guidelines then?
21:33 paramat c55 is not happy with adding farmap unfortunately
21:33 nerzhul CSM write on those functions could be nice in a fork for schools, to teach AI using client, you code your player
21:34 nerzhul red-001, where did you see it ?
21:34 red-001 http://dev.minetest.net/Git_Guidelines#Upstream_commit_rules
21:35 red-001 No signed git commits. The Launchpad PPA would break if there was a signed git commit in master.
21:35 sofar nuke it
21:37 nerzhul it's wrong these days
21:37 nerzhul refresh the page red-001 i think you misread :p
21:37 sofar lol
21:38 sofar I was thinking the same thing
21:38 nerzhul in fact sofar you don't care about ubuntu, but moreover neither ubuntu ppa nor weblate has problems with signed commit, this rule is just obsolete: p
21:42 nerzhul sofar, i'm okay with #5498 too
21:42 ShadowBot https://github.com/minetest/minetest/issues/5498 -- Show media download progress in percent on progress screen by orwell96
21:43 sofar it's too short to waste time on debating :)
21:43 * sofar tries to figure out how to benchmark mariadb
21:43 nerzhul sofar, sysbench with mysql options
21:44 nerzhul look at percona blog, they are very good in MySQL database (percona is a mysql fork like mariadb)
21:44 nerzhul for MariaDB you also need to enable performance_schema and some interesting views, pure benchmark without a read DB on it is useless
21:45 sofar how can you recommend a launchpad project to me?
21:45 sofar you know better
21:46 sofar ignore/ OT
21:47 nerzhul :o
21:52 kaeza joined #minetest-dev
21:58 XtremeHacker joined #minetest-dev
22:02 nerzhul nice job guys this evening, see you later :)
22:03 nerzhul just merging #5498 before going to bed
22:03 ShadowBot https://github.com/minetest/minetest/issues/5498 -- Show media download progress in percent on progress screen by orwell96
22:06 nore allowing CSM to do punches or the like is not an issue I think
22:06 nore I mean, it's nothing that can't already be done by a few trivial engine changes
22:06 sofar I think we should do all the important non-interaction stuff first
22:06 nore and the server still checks for minimum punch delay and the like
22:07 nore sofar: yes, of course, what I meant is that allowing interaction is not an issue
22:07 nerzhul sofar, agreed, when we will get all interesting data, new needs will appear
22:07 nore at some point I guess map prediction will be asked for as well
22:08 nore but, I think we can wait for that
22:09 nore (btw, I'm wondering if juhdanad soft not overlay could be used for text on nodes...)
22:09 VanessaE nore: it could, probably, but not live.
22:10 VanessaE omeone suggested recently regarding the threat of auto-aim-auto-punch/shoot, that the server should be able to detect these things (it already knows which way you're facing, and surely it can rate-limit the punching as well as time-limit turn-and-punch to something similar to human speeds
22:10 VanessaE +s
22:10 Dumbeldor Sofar: I hava squash ! :) #5480
22:10 ShadowBot https://github.com/minetest/minetest/issues/5480 -- [CSM] Localplayer api lua by Dumbeldor
22:10 Dumbeldor have *
22:12 sofar merci :)
22:13 sofar well
22:14 sofar Dumbeldor: somehow you got parts of the sound PR in there now
22:14 Dumbeldor Avec plaisir :p
22:14 sofar Dumbeldor: plus, you should update the commit message and title
22:15 sofar not sure whether it's github, but it doesn't look right
22:15 red-001 github rebase is a mess
22:16 nerzhul red-001, sometimes the diff is wrong, but here it's really wrong :p
22:17 sofar yeap
22:17 sofar Dumbeldor: domage :)
22:17 Dumbeldor Thanks sofar, i look ^^
22:17 sofar Dumbeldor: git commit --amend ; git push dumbeldor -f; #  ;)
22:18 sofar try and remove the sound changes from docs/client_lua_api.md
22:19 sofar Dumbeldor: aussi, you forgot to add `get_name` to the docs
22:21 Dumbeldor Yep
22:21 Dumbeldor wait a minute :)
22:22 Dumbeldor sofar: it's good
22:24 sofar better
22:24 sofar the lua doc needs some help, though
22:25 sofar commas are missing in the example tables
22:25 Dumbeldor Okay
22:25 sofar you're explaining `pos` but should just refer to existing documentation that explains `pos` etc :)
22:26 Dumbeldor Ok, I'm going to sleep, if you see anything else put a comment I continue tomorrow. See you soon. Thank you
22:26 sofar will do
22:26 red-001 Dumbeldor, could you make #5523 client-side?
22:26 ShadowBot https://github.com/minetest/minetest/issues/5523 -- API lua get max lag by Dumbeldor
22:27 Dumbeldor I would talk with nrzkt. I think it's server side.
22:28 sofar yeah, I commented that we want that client-side only
22:28 rubenwardy max lag isn't actually lag
22:28 rubenwardy it's the active server step
22:28 sofar rubenwardy: what is it then?
22:28 sofar oh
22:28 red-001 just use getClient(L)->getRTT()
22:28 rubenwardy and shouldn't be exposed to clients
22:28 rubenwardy s/clients/mods/g
22:28 sofar so it should always be 0.05sec or something like that?
22:28 rubenwardy yeah, unless it's mor
22:28 rubenwardy e
22:28 sofar 0.1 whatever
22:29 red-001 isn't RTT the ping?
22:29 Fixer RTT(ping to server) thing on F5 graph is stuck after login btw
22:29 Fixer i even posted a bugreport
22:29 rubenwardy max_lag >= dedicated_server_step, may be larger due to updating taking longer than dedicated_server_step
22:29 Fixer it was working ok in much earlier builds
22:29 rubenwardy this is afaik
22:29 troller joined #minetest-dev
22:30 rubenwardy it was a while ago that I read this bit
22:30 Fixer there were explanation on forum about max_lag
22:30 Fixer https://forum.minetest.net/viewtopic.php?f=6&amp;t=10884#p167292
22:30 GreenDimond joined #minetest-dev
22:33 red-001 did copy and paste get broken again?
22:35 Zeno` hasn't worked for me for ages
22:36 Zeno` not on this computer anyway, so I've been assuming it's a plasma issue
22:36 Fixer don't remember
22:36 GreenDimond ctrlc/ctrlv works for chat on my PC..
22:37 Fixer depends on os
22:37 GreenDimond I have Xubuntu :P
22:37 Fixer autocomplete still broken btw
22:38 red-001 arch linux here
22:38 red-001 I will test on windows tomorrow
22:39 Fixer rubenwardy: https://github.com/minetest/minetest/issues/3307 and https://github.com/minetest/minetest/issues/4151 should be closed I think since that workaround fix was merged just after 0.4.15 release iirc
23:26 kaeza joined #minetest-dev
23:45 Tmanyo joined #minetest-dev
23:46 waldo_ joined #minetest-dev
23:54 rubenwardy joined #minetest-dev

| Channels | #minetest-dev index | Today | | Google Search | Plaintext