Minetest logo

IRC log for #minetest-hub, 2018-06-16

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

All times shown according to UTC.

Time Nick Message
00:53 Tmanyo joined #minetest-hub
03:18 ssieb joined #minetest-hub
04:11 Tmanyo joined #minetest-hub
05:38 lumberJ joined #minetest-hub
06:02 lumberJ joined #minetest-hub
08:50 CWz joined #minetest-hub
08:51 IcyDiamond I think I need to create a material helper library
08:51 IcyDiamond I find myself often checking for mods to add some sort of recipe
08:52 IcyDiamond I wish minetest_game had consistency and materials as groups
08:53 IcyDiamond Can I do a pull req changing the groups or is that not a good idea right now
08:55 IcyDiamond I also feel like the engine needs aliases for item names
08:56 IcyDiamond This way changing the item names to the new convention wouldn't break anything
09:11 Gael-de-Sailly joined #minetest-hub
09:27 Fixer joined #minetest-hub
09:41 Krock joined #minetest-hub
10:22 CWz_ joined #minetest-hub
10:28 paramat joined #minetest-hub
11:01 IhrFussel joined #minetest-hub
11:02 IhrFussel 17 players @ 0.4 secs max_lag ... I could've NEVER imagined that simply moving protection checks to the bottom of the mobs ABM could make such a huge difference
11:12 entuland joined #minetest-hub
11:15 xerox123 hello
11:18 DI3HARD139 joined #minetest-hub
11:21 IhrFussel !tell tenplus1 I recommend moving the protection check in the mobs ABM to the very bottom right before actual spawning because it makes a HUGE difference if the server uses the areas mod with 1,000+ protections
11:21 MinetestBot IhrFussel: yeah, sure, whatever
11:29 Krock hi IhrFussel, xerox123
11:29 Krock IhrFussel, sounds more like an areas issue
11:31 IhrFussel Krock, no issue, the mod has to loop through all protections in order to check if the current position is inside one
11:31 Krock no, it could use the AreaStore stuff
11:31 IhrFussel It already does that
11:31 Krock i.e. check whether your server was build with spatial
11:35 IhrFussel Why don't you enable that by default then?
11:36 IhrFussel You make it server owners sometimes harder than it should be
11:38 IhrFussel Doesn't matter now, 0.4 secs with ~ 15 players is good with 180 mods
11:39 Darcidride joined #minetest-hub
11:40 IhrFussel Every server owner who wants to use AreaStore is screwed if they can't compile and use the PPA builds for example
11:52 CBugDCoder joined #minetest-hub
11:59 paramat joined #minetest-hub
12:00 IhrFussel But I don't really get why the protection checks for placing and digging seem to be faster
12:02 IhrFussel on_place() and on_dig() also just use minetest.is_protected() ... why would the call there be less laggy than when calling it from mobs ABM =/
12:11 IhrFussel Like seriously... the profiler reports MAX 1ms for on_dignode and on_placenode callbacks
12:26 Krock > if they can't compile
12:27 Krock You can compile Minetest on every machine which can run Minetest
12:27 Krock with cross-compiling even more could be done than that
12:30 rubenwardy Basically no one uses windows to host servers
12:30 rubenwardy At least not the sort that has 1000s of areas
12:31 IhrFussel You cannot and should not expect every server owner to know what compiling is even... do you only support server owners who are tech savvy?
12:31 Krock what do we support at all?
12:32 IhrFussel I doubt that all the popular servers on the list right now are run by admins who are well educated about tech stuff
12:33 IhrFussel And those server owners are in an disadvantage then ... You should make it as straightforward as possible for everyone to use optimized features
12:34 rubenwardy If you're using Linux, you should probably know how to run a single bash script
12:35 IhrFussel I DO know lua and compiling...my server is an entire custom experience ... But even I didn't know that you need to enable a custom build flag for AreaStore to work
12:35 IhrFussel I expected that to be builtin
12:36 BillyS How would I add to a callback for a node? Basically, I want to override the callback so that the original version is called, and then my bit is called after that
12:36 rubenwardy It does work, just uses a simpler implementation
12:37 BillyS Could I somehow get the original function by using minetest.registerd_node[<node>].<func>?
12:38 IhrFussel Is that implementation the reason why it takes 1ms to check protection in on_dignode() and on_placenode() but 100-500ms when checking for protection with no name provided?
12:38 rubenwardy Sounds like an areas bug to me
12:41 IhrFussel If you got a minute maybe read the function that is being run inside is_protected() https://pastebin.com/9Ykb6d56
12:44 IhrFussel Parent function that calls ^ https://pastebin.com/uaSCtVti
12:50 Krock BillyS, this time I've found a leak :D
12:50 Krock item duplication by rejoining
12:50 BillyS Ok, WHAT? :P
12:51 BillyS Krock: Explain plz
12:51 Krock it's unified_inventory I believe. checking whether it's been fixed yet
12:51 Krock not here yet
12:51 BillyS ah
12:53 Krock not fixed. https://github.com/minetest-mods/unified_inventory/blob/master/bags.lua#L161 this is one of the lines
13:11 aerozoic joined #minetest-hub
13:22 Krock programming done.. testing begins! :D
13:22 BillyS fsck
13:31 xerox123 fsck - check and repair a Linux filesystemâ„¢
13:31 BillyS â„¢
13:34 BillyS Griefern00bz on the server
13:34 BillyS rollbacking
13:48 Krock done
13:53 Ruslan1 joined #minetest-hub
13:55 ANAND Hello everyone!
13:56 Krock BillyS, cd ./mods/unified_inventory/ && wget https://github.com/minetest-mods/unified_inventory/pull/115.patch && git am -3 115.patch
13:56 Krock if you'd like to test it
13:57 BillyS k
13:58 Krock it's entirely backwards compatible so you won't notice any differences on the server
13:58 BillyS kk
13:58 Krock also the speed is near to identical :D
13:58 BillyS YAy
13:59 BillyS Added it
14:03 IhrFussel Good that only the bags.lua file needs to be changed for the fix, cause I modified the other files quite a bit
14:06 Krock maybe I'll find more issues with the other files :D
14:06 Krock then you've got to solve these merge conficts
14:09 aerozoic Is that bug easy to reproduce?
14:10 Krock yes, with CSM and the right formspecs it's quite easy
14:11 IhrFussel Only possible via CSMs again..I knew it
14:11 Krock I'm however using a modified client to get some additional information in Lua from the C++ client
14:12 Krock that makes finding the right inventory lists a bit easier
14:12 aerozoic Does it leave a trace in the logs that i can search for?
14:12 Krock no, this time there's not a single line
14:12 aerozoic ok, thx for the update
14:12 Krock however, the players have to rejoin in order to duplicate the items
14:13 Krock np. I appreciate it if you could test it on your server. Even if something's going wrong (doubt it, tested locally), then it only affects the online players
14:14 IhrFussel Well an easy excuse for that would be "sorry guys my connection is so bad today D:"
14:15 aerozoic What would be a test procedure? Just to make sure there is no server crash?
14:15 Krock "everything takes years to do. cant place BLOCKS or craft stuff. will rejoin until it works nicely"
14:15 IhrFussel I will replace my bags.lua with the patched one if someone can confirm that it works correctly
14:15 Krock aerozoic, to test the exploit?
14:15 Krock or to test the fix overall?
14:16 aerozoic Oh i thought you were talking to me.. "I appreciate it if you could test it on your server."
14:17 aerozoic What specifically would i test for?
14:19 aerozoic If it's to test whether the exploit is prevented, i would need to know how to do the exploit.
14:19 IhrFussel Krock, worst case if the patch doesn't work? Just a crash or more? If the worst that could happen is a server crash then I'll just apply it right now and restart the server in 10 minutes so you can try the exploit again
14:20 Krock IhrFussel, crash and cleared bag slots for the players. However, the items inside won't be affected
14:20 Krock that's the absolute worst case
14:21 IhrFussel No items cannot be erased?
14:21 IhrFussel So*
14:21 Krock bag contents: no. bag items: yes.
14:22 IhrFussel So worst case: craft new bag and your items are back
14:25 IhrFussel Okay file replaced, will restart now
14:26 IhrFussel Server boots up! Good sign
14:27 IhrFussel CRASH on join xP
14:28 Krock huh
14:28 IhrFussel One sec trying to post debug.txt error
14:29 IhrFussel ServerError: AsyncErr: ServerThread::run Lua: Runtime error from mod 'unified_inventory' in callback on_joinplayer(): ...st/worlds/oldserver/worldmods/unified_inventory/bags.lua:136: attempt to call global 'save_bags_metadata' (a nil value)
14:31 IhrFussel save_bags_metadata doesn't seem to exist in unified_inventory ... typo?
14:32 Krock IhrFussel, it exists. see diff
14:32 Krock are you using Lua or LuaJIT?
14:32 IhrFussel JIT
14:32 Krock you didn't copy the file correctly then
14:34 IhrFussel I clicked RAW in your file version
14:34 Krock also: would it work if you moved the save_bags_metadata function (bags.lua) above load_bags_metadata?
14:34 Krock that's line 140 and below
14:34 IhrFussel Copied from here...not correct? https://raw.githubusercontent.com/SmallJoker/unified_inventory/d0987c1303d0b8562a0c06c2a2f986ffabb94778/bags.lua
14:35 Krock that's correct
14:36 IhrFussel I moved the function to the top now, let's try
14:37 IhrFussel No crash on join
14:37 Krock LuaJIT is being weird
14:37 Krock pushed fix
14:37 Krock "fix" for a problem I cannot reproduce
14:41 IhrFussel Krock, can you confirm that it's fixed now?
14:41 Krock yes
14:41 Krock local slots are gone and it's saving the stuff correctly
14:41 Krock and also restores the bags which are used
14:43 Krock ANAND, that's not how you rebase a branch
14:43 Krock the heck happened with that PR
14:43 ANAND Sorry.... >.<
14:43 ANAND That's the first time I rebased
14:46 ANAND Krock: Shall I start a new PR altogether?
14:46 Fixer joined #minetest-hub
14:48 Krock ANAND, how did you rebase it?
14:50 Krock you could compare your changes with the original branch using git diff
14:50 ANAND No idea tbh... I did a lot of things and faced a few complications along the way. I do remember trying to 'git rebase master' with the modified branch checked out.
14:50 Krock uh well ok
14:50 ANAND I guess I should've asked... ._.
14:50 Krock git fetch upstream  (or however your minetest/minetest remote is called)
14:50 Krock then rebase onto the upstream master
14:51 Krock i.e. git rebase upstream/master
14:52 ANAND I did that; got an error that menu_lua_api.txt has been modified (or something), so I merged it, then tried to rebase it again
14:53 Krock no no. if it was modified, then you're asked to solve the conflicts
14:54 ANAND yes
14:54 Krock this means to open the conflicting file and searching for >>>>>>>>>>
14:54 Krock afterwards add the file and git rebase --continue
14:54 ANAND search for >>>>>> and then?
14:55 Krock both versions are denoted there. it's then your task to decide which parts that should be taken
14:55 Krock however, there are tools around which can do that for you (yet haven't used one)
14:56 ANAND So I delete the part that I don't want, and then rebase?
14:57 Krock afterwards add the file (git add <filename> and git rebase --continue
14:57 Krock uh well, I'm sure there's a video about this somewhere
14:59 ANAND Right, I'll take a look at a couple of tutorials on rebasing... Meanwhile, I'll just make a new PR to get rid of this mess
15:31 ANAND Krock: #7455 :3
15:31 ShadowBot https://github.com/minetest/minetest/issues/7455 -- Remove deprecated functions from menu_lua_api.txt by ClobberXD
15:37 IhrFussel VanessaE, someone on my server claims that you made anyone staff who asked for it :P
15:37 ANAND lol
15:38 VanessaE they're wrong.
15:39 Krock VanessaE, do you have push access to unified_inventory btw? Today I wrote a security fix an item duplication expoit
15:40 VanessaE I do not.
15:40 Krock too bad. thanks anyway :/
15:40 VanessaE you based your pull on the minetest-mods fork, but I have my own fork of it on gitlab.
15:40 Krock I'm starting early to ask because the development/response time there is very slow
15:41 VanessaE since I somehow became the maintainer of it, you may want to point at my fork
15:41 IhrFussel Now the reply of that someone was "/me slaps IhrFussel"
15:42 Krock uh well, it's hard to tell which one will be used by more people since none of them is a mirror
15:42 Krock IhrFussel, perma ban
15:43 IhrFussel Now he called me a WENNIE
15:43 VanessaE Krock: both were identical a week or so ago, though everyones' git-remote probably still points at the minetest-mods one.
15:43 Krock I checked the and the HEAD is still the same in both
15:44 paramat joined #minetest-hub
15:46 IhrFussel Apps still use "Ficker" as pre-defined name and each time someone joins with it, I have to tell them to change it
15:46 Krock blacklist?
15:47 IhrFussel Yeah maybe I should just disallow any names starting with Ficker
15:47 sniper338 joined #minetest-hub
15:48 sniper338 joined #minetest-hub
15:48 ANAND joined #minetest-hub
15:48 Krock Did you mean: any names that _contain_ this word?
15:49 IhrFussel I need to check first if anyone has an account with "Ficker" in the middle of their name
15:50 IhrFussel Nope, ok can block anything with Ficker in the name
16:12 IhrFussel How do I make this carpet rotatable? https://pastebin.com/dZyKhwJv
16:12 IhrFussel With screwdriver
16:31 IhrFussel Maybe you cannot rotate railllike nodes by definition?
16:31 IhrFussel raillike*
16:41 ANAND joined #minetest-hub
17:01 paramat i suspect so
17:02 IhrFussel Regarding the active_block_range github issue: Maybe it could already help to make it possible to not calculate the _wider parameter by default and only on request
17:02 IhrFussel Pretty sure that would speed up c++ a lot
17:03 paramat not sure, it may be lightweight
17:04 IhrFussel Hm...counting objects within a range of 3 mapblocks in each direction... I could imagine that to be pretty heavy on resources UNLESS it gets only calculated once per mapblock
17:30 IcyDiamond https://i.lunasqu.ee/Screenshot_18-06-16-20:29:29.png
17:30 IcyDiamond owo
17:30 IcyDiamond from left to right: generator, alloy furnace, pulverizer, furnacem power cell
17:30 IcyDiamond s/furnacem/furnace,/
17:31 Gael-de-Sailly joined #minetest-hub
17:35 paramat nice work
17:36 IcyDiamond thanks
17:39 IcyDiamond https://i.lunasqu.ee/Screenshot_18-06-16-20:39:01.png the ui will be changed but this is it rn
18:37 paramat joined #minetest-hub
18:50 Fixer joined #minetest-hub
19:21 pauloue joined #minetest-hub
19:59 pauloue joined #minetest-hub
20:46 Fixer_ joined #minetest-hub
21:23 longerstaff13 joined #minetest-hub
21:23 longerstaff13-m joined #minetest-hub
21:51 Miner_48er joined #minetest-hub
21:56 BillyS Shara: Sorry to bother, but what's the IRC channel for DL? :)
22:00 Shara I'll PM you
22:01 tumeninodes joined #minetest-hub
22:01 tumeninodes Hey folks o/  who's around?
22:05 tumeninodes So, I had been using Calinou's one-line install script for MT 0.5.0-dev (which is now 5.0.0-dev I guess) and I just had to switch to wiping, doing a new MT fork on github, then cloned that to home/minetest, then compiled and then added the official MT repo as upstream
22:06 tumeninodes My question is..., I had backed up my MTG game (which was done the same way a while back), and texture folder, game.conf, worlds folders
22:07 tumeninodes when I place those back into the new MT directory, all should be good right? This should not alter my remote/github links or anything so I can carry on as usual? I should not need to do any git changes for MTG directory?
22:16 shivajiva hmm MTG has a dev and stable versions but stable is marked as 0.4
22:17 tumeninodes yes, indicates 0.4 series i.e. 0.4.17.1
22:17 shivajiva so things have changed for 0.5
22:17 tumeninodes that is stable and 5.0.0 is dev
22:18 tumeninodes well, the version has been changed to 5.0.0 rather than 0.5.0
22:18 shivajiva ah okay the 0.5 dev bit confused me
22:19 shivajiva I've run mtg from 14 months ago on 0.4 stable so no issues at that point up to current stable
22:20 tumeninodes so you're still running 0.4.16 stable then (I think)
22:20 shivajiva 0.4.17-dev disappeared recently didn't it?
22:21 tumeninodes then 0.4.17 as a bug-fix release, then 0.4.17.1 to fix the bugs in the bug-fix release :P
22:21 shivajiva lol something like that
22:21 tumeninodes hmmm, as far as the 0.4.17-dev question, someone more qualified has to address that
22:22 tumeninodes coulda just added -new and improved tag haha
22:23 shivajiva mtg has a backport to 0.4.17 but mt only has 0.4 stable
22:25 tumeninodes because it is still part of the 0.4 family, I believe but that should be confirmed by someone else
22:25 tumeninodes 0.4.xx
22:25 shivajiva somewhere in the recent tree we have a 0.4.17-dev
22:26 tumeninodes I do not think MTG actually has a version attached to it, it goes off whatever MT is at
22:26 shivajiva of MT
22:27 shivajiva but last time I pulled and compiled 0.4-stable it kicked out 0.4.16
22:27 tumeninodes and gave you what? 0.4.17 or 0.4.17.1?
22:28 shivajiva 0.4.16
22:28 tumeninodes oh I misundersttod your use of "kicked"
22:28 shivajiva maybe I'm going senile, yea should have said spat out
22:28 tumeninodes hmmm, you should ping a core or game dev
22:31 tumeninodes for my .minetest build, I use the mint/ubuntu ppa (which is still at 0.4.17 awaiting a bump to 0.4.17.1)
22:31 tumeninodes for my /minetest build I just compiled dev and have 5.0.0-dev
22:32 tumeninodes so maybe the stable source on file has not been updated?
22:32 tumeninodes which would be odd
22:34 tumeninodes stable was bumped to 0.4.17 13 days ago https://github.com/minetest/minetest/commit/ecbb9310de9cde6bf22dacbb1e95da326daafa51
22:34 tumeninodes so there is no way is should have spat 0.4.16 at you. I would def ping someone
22:38 shivajiva not before I've rechecked but it was fresh install and a checkout of 0.4, wonder if it was on the day it changed
22:45 shivajiva last clean local clone I made in the past 28 days states on pull ' Your configuration specifies to merge with the ref 'refs/heads/backport-0.4' from the remote, but no such ref was fetched.' So that copy certainly was older than 13 days
22:47 shivajiva but now it has the correct tags
22:49 shivajiva the warning is because I'm on the branch that now has no ref
22:56 shivajiva 0.4.17.1 so it's just the time frame, nothing to report
22:58 xerox123 title says Minecraft... but the video is clearly Minetest: https://youtu.be/ztAg643gJBA
22:58 xerox123 !title
22:58 MinetestBot xerox123: Is a round Minecraft world possible? - YouTube
23:00 shivajiva Jeija
23:02 Mr_Pardison joined #minetest-hub
23:08 shivajiva minetest is advertised from 7:10 onwards :)
23:09 Mr_Pardison greetings.
23:09 shivajiva Hi Mr P
23:53 entuland there, have a slice: https://imgur.com/a/wzqp1tR

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