Time  Nick       Message
11:50 IhrFussel  Can this crash the server? 2018-08-10 13:40:42: ERROR[Main]: ServerError: AsyncErr: ServerThread::run Lua: allow_put should return a number. name=Dodier523_bags
13:30 IhrFussel  My server now crashed TWICE cause of this error: ERROR[Main]: ServerError: AsyncErr: ServerThread::run Lua: allow_put should return a number. name=Dodier523_bags
13:32 IhrFussel  Relevant code, maybe someone can help here https://github.com/minetest-mods/unified_inventory/blob/master/bags.lua#L172
13:53 Bobr       not sure but its to do with it cant read the bags i think
13:55 rubenwardy IhrFussel: allow put should return a number
13:56 IhrFussel  It's not my code, I updated bags.lua in June after Krock fixed an exploit in it
13:57 IhrFussel  https://github.com/minetest-mods/unified_inventory/pull/115
13:57 rubenwardy It looks like the size of a bag decreased to a point where nothing more can fit in
13:58 rubenwardy If you look at all possible routes through that function, there's one which returns nil
13:58 rubenwardy The one where it exits out the end
13:58 rubenwardy NB: no return statement is basically the same as    return nil
13:59 IhrFussel  So a "return 0" is missing before "end," ?
13:59 rubenwardy (the only way to distinguish between return nil and no return is using #table.pack(func()) >
13:59 rubenwardy Mahbe
13:59 rubenwardy I'm not sure what the code is supposed to do
14:00 IhrFussel  Krock, please read the recent log if you haven't already...my server crashes cause of bags.lua sometimes ... could it be related to your exploit fix?
14:01 Krock      already looking at git blame
14:02 Krock      I see. there was a return 0 at the end
14:02 Krock      apparently forgot that one when I refactored
14:03 Krock      I'm however not sure how you would reproduce this bug
14:03 IhrFussel  It only happens with one user
14:03 Krock      there used to be a bag which was larger than the new one
14:04 Krock      i.e. too many slots occupied
14:05 Krock      pushing a fix in a minute
14:07 IhrFussel  Anything suspicious here? https://pastebin.com/XTMuRdQ1
14:08 Krock      IhrFussel: the detached inventory contents would be more relevant
14:08 Krock      the player file only provides half of the necessary information
14:08 IhrFussel  You mean the player attributes?
14:09 Krock      either that or the detached inventory contents in-game
14:09 IhrFussel  Huh? I thought those are only stored in the player file/DB
14:10 Krock      I don't see any player meta in that file
14:10 IhrFussel  "unified_inventory:bags":"return {\"unified_inventory:bag_large\", \"unified_inventory:bag_large\", \"unified_inventory:bag_medium\", \"unified_inventory:bag_large\"}"}
14:11 Krock      oh I see
14:11 Krock      they tried to swap the bags
14:11 Krock      like instant-upgrade
14:12 IhrFussel  You mean they wanted to replace one bag directly with the one they had on their "mouse"?
14:13 IhrFussel  And will the return 0 fix it too?
14:13 Krock      aaaagh
14:13 * Krock    slaps VanessaE for using merge commits
14:16 Krock      IhrFussel: yes yes
14:17 * Krock    pokes VanessaE to update gitlab/u-i
14:48 ANAND      How are PRs merged into MT?
14:48 ANAND      They're squashed, I think
14:50 ANAND      How does it show up as X authored and Y committed?
14:53 xerox123   that's how I see it
14:54 Krock      ANAND: squash, rebase and push
14:55 Krock      the squash process causes the "committed" addition
14:55 ANAND      Ah
14:55 Krock      since it's someone else who actually wrote or modified the commit
14:55 ANAND      Is this done from the Github UI itself?
14:55 Krock      yes, but everything can be done manually aswell
14:56 ANAND      oh nice :)
14:56 Krock      on fact, if the other branch is even with the minetest branch, then a simple pull & push suffice. in this case, the commit is not modified and thus, the original author kept
14:56 Krock      (without any "committed" addition)
14:57 ANAND      Which is not the case with PRs, right?
15:00 Krock      usually not
15:02 ANAND      ok thanks for clarifying :)
15:42 Krock      rubenwardy: Am I right with the assumption that the forum "Just 25.4 mm off from a topic. Maximum." allows any topic as long it's kept on-topic?
15:42 Krock      surely, without religious/racism content
15:42 rubenwardy Probablh
15:43 benrob0329 Honestly offtopic is a mess
15:43 benrob0329 The categories make less sense than most posts
15:43 Krock      Well, it was a bit worse before
15:43 rubenwardy It's annoying how the categories sre
15:43 rubenwardy C55 trying to be funny
15:44 Krock      not sure if he changed it
15:44 benrob0329 Krock: is it though?
15:44 Krock      > 1 inch
15:44 benrob0329 Is it better?
15:44 benrob0329 Problem: Offtopic is a shitposting mesa
15:44 Krock      benrob0329: well yeah, now it's at least clear what belongs a bit to Minetest and what not
15:45 benrob0329 Solution?: Nuke it and make categories that leave people more unsure of what goes in offtopic
15:45 Krock      and "On-topic" looks like a Meta Forum about Minetest and projects around
15:45 benrob0329 I mean hey, the shitposting stopped for a little while, but only because people were complaining about the new offtopic section
15:49 Krock      Proposal for better named categories: 'Related to "mine" and/or "test"' and
15:50 Krock      "Something way different lol pls dont shitpost ok"
15:52 benrob0329 Other offtopic sections: Encourage Users to make good posts in a more relaxed environment.
15:53 benrob0329 Our Offtopic: Encourage Users to not post offtopic.
15:53 Krock      well, it's not wrong. people should post on-topic on an off-topic topic
15:55 benrob0329 But, on-topic posts are what the rest of the forums are for..
15:55 benrob0329 Things relating to mine or test can be offtopic, but they often aren't entirely.
15:56 benrob0329 Its probably the category I have the least problems with
15:57 benrob0329 "Just 1 inch off from a topic. Maximum." Sounds like youre yelling at your child
17:17 paramat    yes i asked c55 if he could clean up the offtopic section mess, he said the subforum titles were meant to be a joke
17:19 paramat    i still think it's best to nuke it from orbit as it is not a worthwhile use of the limited resources. people can talk offtopic elsewhere. the quality of offtopic is very low too
17:24 benrob0329 The offtopic section of the forum is one of the most important IMO, a place where people in a fanbase can talk about other or related things while still interacting with said fanbase
17:24 benrob0329 But having a good offtopic section that encourages good topic is even more import an than having one at all
17:24 benrob0329 Otherwise its just shitposting
17:25 benrob0329 s/import an/important
17:58 paramat    yeah, it's potentially important, but ours is so low quality, consistently
18:00 roben1430  ours has low-quality categories, consistently
18:11 BillyS     Hmmmm
18:13 BillyS     How would one prevent users from placing a block if they don't have certain privleges?
18:13 BillyS     I suppose it would have to be removed as soon as they place it or something ..
18:14 sofar      the server blocks the placement, the client thinks it has placed it but then the client receives the placement-denied message
18:14 sofar      and then the block is put back in the inventory
18:14 sofar      the blocking is done in lua, e.g. protector mods do this
18:14 BillyS     But how does the server block it?
18:15 sofar      server receives placement event, and doesn't process it
18:15 sofar      it runs lua code for the placement event, the lua code discards the event and tells the server to send a packet back to the client that placement didn't happen
18:15 BillyS     Hmm ... looking in simple_protection, it looks like the lua overrides minetest.item_place
18:15 sofar      look at the protector_redo mod, for code
18:16 BillyS     Kay thx
18:16 sofar      or any other mod like areas
18:18 Krock      > not considering simple_protection as acceptable example
18:18 * Krock    feels sad man
18:19 Krock      however, I think this placement overwrite was for compat reasons.. not entirely sure :D
18:19 paramat    RDR2 'in game footage' yes but 90% of the trailer is the fixed story sequences :) game trailers are so full of deception, even stellar overload does it
18:20 paramat    MT's game trailer understates MT :)
18:20 IcyDiamond hi, how do you make a mesh made in blender have multiple textures on different objects
18:20 IcyDiamond like, how do i assign a texture to a material or whatever
18:21 IcyDiamond hmm, or do i have to make just a single texture and map UVs to it
18:21 * Krock    points at some Blender tutorials
18:21 IcyDiamond im talking about using the mesh in minetest
18:21 Krock      Jordach would have some knowledge there if they were online rn....
18:22 Krock      I've never seen more than one texture for the object
18:22 benrob0329 IcyDiamond: multiple materials in blender
18:22 Krock      but that doesn't mean that it's the limit
18:22 IcyDiamond benrob0329: yeah, but how do i assign textures to them in minetest
18:23 Krock      obj:set_properties({textures= {"foobar.png"} })
18:23 benrob0329 Textures is a list
18:24 benrob0329 Or, whatever lua calls that thing
18:24 Krock      ipair-table-thingy?
18:24 benrob0329 Tables
18:43 BillyS     If a block has a formspec defined as well as an on_rightclick function, which will be called when a player right-clicks it?
18:44 Krock      BillyS: https://github.com/minetest/minetest/blob/master/doc/lua_api.txt#L5824-L5830
18:45 BillyS     Hmmm ... okay
18:45 Krock      so on 5.0-dev you can rely on that it's triggered
18:45 BillyS     thx
19:02 * VanessaE pokes Krock with an updated u-i and because merge commits aren't bad :P
19:05 * Krock    pokes VanessaE with a sticky note saying: "Merge commits are bad because they mess up the history"
19:08 VanessaE   :P
19:09 nerzhul    hi Krock
19:09 VanessaE   well since it's obvious that my copy isn't going to be considered "upstream", I'll just delete it and point people back to the github one.
19:10 Krock      hello nerzhul :)
19:10 nerzhul    i hope i can take time to develop the postgresql playerbackend this weekend
19:11 nerzhul    to ensure we have it for 5.0.0
19:11 nerzhul    did we define a target for 5.0 feature freeze ?
19:11 Krock      https://github.com/orgs/minetest/teams/engine/discussions/13 nothing happened
19:11 VanessaE   or rather, I'll turn my copy into a fork.
19:12 Krock      VanessaE: or.. a mirror?
19:12 nerzhul    holidays, holidays :)
19:12 Krock      a month holidays. would be nice.
19:14 * Sokomine notices quite a lot of poking and hides as a precaution
19:15 Sokomine   VanessaE: er. whichever. just hope it will be clear which one will be the one that gets updates?
19:17 VanessaE   it's a mirror now.  and I updated the forum post to point to the github one.
19:22 Krock      roughly 695 days until we hit issue number 10000
19:22 Krock      (or PR number)
19:36 paramat    don't set a date for freeze/release, we know that ends badly. however i think we should try to release before year end, hopefully it will be sooner
19:38 Krock      "before year end" can't be sooner due to time travel limits
19:39 Krock      it's just a rough date so that we can see how much is going to fit int 5.0
19:39 Krock      *fit into
19:39 Krock      hi tenplus1
19:39 tenplus1   hi folks :)
19:39 Krock      BLAZING FAST
19:39 tenplus1   ehehe, hey Krock *whoosh!"
19:40 tenplus1   nssm updated to fix swords with missing functions
19:41 tenplus1   o/ pie billy
19:42 Krock      pielly
19:43 tenplus1   pielly cirrus
19:43 tenplus1   :P
19:44 tenplus1   o.O
19:45 tenplus1   what's new with you Krock ?
19:46 Krock      RL stuff here and then.. nothing special
19:47 tenplus1   same... got a player miffed that he has to craft a mortar&pestle to make flour when using farming redo ;D
19:47 tenplus1   lol, i really dont get that challenge at all...  nice take on the vid tho :D
19:49 paramat    also i think we should avoid adding unnecessary stuff into 5.0.0 for no good reason, therefore delaying it
19:50 tenplus1   hi paramat, how do you mean ?
19:50 tenplus1   what kinda stuff ?#
19:50 paramat    that was in response to "it's just a rough date so that we can see how much is going to fit int 5.0"
19:50 tenplus1   ahh gotcha
19:51 paramat    we only have a very few blockers left, once done, freeze
19:51 tenplus1   I do agree, adding fluff to the game doesnt help
19:51 * tenplus1 would still remove the different bush stems and only have 1
19:52 paramat    heh
19:52 Krock      paramat: not sure about the float network format. It could be worth a Blocker tag too, since it would be annoying to make it backwards compatible
19:52 paramat    i meant engine primarily
19:52 paramat    already added blocker to that
19:53 tenplus1   any major breakages ?
19:54 paramat    thought you knew the 5.0.0 breakage? :)
19:54 tenplus1   not them all, isnt there a few
19:54 tenplus1   map version and player model
19:55 paramat    server-client compat, player pos, erm ..
19:55 paramat    player models, also attachment positions to vehicles needs updating
19:56 tenplus1   how has the player pos been changed, this one's important to me
19:56 paramat    (player attachment pos)
19:56 tenplus1   isnt it still at foot level ?
19:57 Krock      oh, we have a new map version?
19:58 paramat    no new map version eheh
19:59 tenplus1   0.4.17 getpos is at player foot level, what is it in 5.0 ?
20:01 Krock      uhm.. same?
20:02 tenplus1   am paranoid about that part changing cause many mods rely on player pos
20:03 paramat    same, the change is elsewhere afaik
20:03 tenplus1   pah!
20:03 paramat    see the PR thread for details
20:04 tenplus1   falling.lua still has the replace issue
20:04 tenplus1   was considering including pull for my new version
20:04 tenplus1   https://notabug.org/TenPlus1/falling_item
20:29 tenplus1   nite all o/
21:39 IcyDiamond added links to my minetest forum and contentdb profiles on my personal website :)
21:58 Fixer      nice, ty
22:09 paramat    hopefully 5 months
22:26 Fixer      no need to rush it