Time  Nick           Message
05:55 Taose          I'm curious what AI/NPC packs are recommended?
05:56 Taose          I don't quite want aggressive monsters just yet
07:21 jas_           Hijiri: i have a punch deflected that returned damage to mob/player and indeed use punch() for mobs and set_hp() for players.  i've already forgotten why, but "punch() didn't work for players" sounds right.  https://github.com/jastevenson303/warps/blob/dcbl/init.lua#L332
07:21 jas_           or maybe i didn't ever try to do punch on a player object; and only set_hp didn't work on entities (der?) ... i'll try now
07:22 jas_           (i just noticed a minetest.punch_node() in lua_api.txt, always finding new stuff in there :)
07:24 jas_           https://youtu.be/okaqXB6Ns5s
07:24 jas_           !title
07:24 MinetestBot    jas_: Leonard Cohen - Traveling Light - YouTube
07:27 jas_           Hijiri: hm, it seems to work as expected.
07:27 jas_           one observation is that set_hp() allows a number, whereas ObjectRef:punch() doesn't
07:34 jas_           but the direction here is different, punching other entities/players.  whereas with TNT i imagine it's the player/entities receiving the punch from the tnt(?)
07:36 Taose          ...turns out...really hard to update minetest using git after installing the minetest game etc...
07:36 Taose          (Unrelated histories)
07:38 jas_           actually of course obj:punch() takes toolcaps, and damage_groups, so can set damage amount there.  it's a good question Hijiri i think you're onto something
07:38 jas_           can just do a git pull inside games/minetest_game right Taose ?
07:39 Taose          Got a "Unrelated Histories" error
07:39 jas_           oh i never seen that error before
07:39 Taose          Neither have I
07:39 Taose          I don't do modding or fiddling for minetest
07:40 jas_           i do git pull for both minetest and minetest_game and do system-wide installs, but i dunno what unrelated history is.  maybe try fresh?
07:40 Taose          That's what I've had to do
07:40 jas_           interesting.  every time?  and minetest_game is inside minetest/games  ?
07:40 Taose          yep
07:40 jas_           (one repo inside another?  that's right)
07:40 jas_           how interesting.  well that sounds like a real chore
07:40 Taose          Currently having to reinstall all my mods
07:40 Taose          XD
07:40 jas_           so you do a system-wide install?
07:41 Taose          What do you mean by System-wide?
07:42 jas_           like, `make install' versus running inside minetest/ with -DRUN_IN_PLACE=1 ?
07:42 Taose          ah no I just follow the instructions
07:43 Taose          so I use the .cmake -DRUN_IN_PLACE command
07:43 jas_           so do you make && make install, or just make?
07:43 Taose          then use make -j15
07:43 jas_           is -DRUN_IN_PLACE the same as -DRUN_IN_PLACE=1 i wonder, or were you abbreviating?
07:43 * Taose        was abbr
07:43 jas_           doesn't -j15 mean you have 14 cpus?  (i dunno this command)
07:43 Taose          It should
07:44 Taose          but it doesn't work that way
07:44 jas_           you have 14 cpus?!
07:44 jas_           haha that would be sweet
07:44 Taose          It's only around -j50 that it seems to have issues
07:45 jas_           hmmm... unrelated histories.  i wonder if minetest.git is getting pulled, and you're not pulling minetest_game.git fresh accordingly?
07:45 jas_           because both need to be pulled when updating
07:45 jas_           but i don't think that'd be that error.  maybe i'll look it up now
07:46 jas_           what version of git?  i'm just reading https://stackoverflow.com/questions/37937984/git-refusing-to-merge-unrelated-histories
07:46 Taose          2.11
07:47 jas_           ye, me too.  "You can use --allow-unrelated-histories to force the merge to happen."
07:47 Taose          yeah didn't wanna do that
07:48 Taose          There were a few other errors popping up relating to an unfound font
07:48 Taose          I try to avoid forcing until I know what the problem is and I'm okay with forcing
07:48 jas_           it's very piculiar ... i never spelled this word b4.
07:48 jas_           peculiar
07:49 jas_           i never even heard of the error before.
07:49 Taose          and now I've lost my world
07:50 Taose          Well it's not lost, it's still there, it's just not showing up in the screen
07:50 Taose          actually I think it's gone
07:50 Taose          goddamit it
07:50 jas_           are you doing --depth 1?  (longshot, maybe that's why.)
07:51 Taose          nop
07:51 jas_           ye you should probably use symlinks if you're going to run_in_place and nuke your minetest dir haha
07:51 Taose          I am certainly starting to think that the devs haven't thought about these issues
07:52 Taose          okay
07:52 Taose          I've briefly lost my passion for minetest -_-
07:52 Taose          (you know what it's like after you've put in a weeks worth of work into something...)
07:53 jas_           yes sir, sounds like you could use some R&R
07:53 jas_           u can make an issue on github tracker where more eyes, better eyes can see
07:53 jas_           it may be something that's come up b4, and i'm the only one awake reading irc
07:53 jas_           and then take your vacation :)  and when you come back it'll be all fixed
09:57 jas_           *gasp* ! there's an on_walk_over() ? wooow
09:58 deltasquared   jas_: is that in -dev
09:58 deltasquared   then again, I guess mesecons would use it for pressure plates.
09:58 jas_           oh phew thought i chatted in the wrong channel lol... yes, i use current minetest dev version
09:59 jas_           and am currently reading: https://github.com/minetest/minetest_game/pull/1902/files
09:59 jas_           tenplus1 mentions a new "on_walk_over" nodedef, but i am not seeing it.
09:59 jas_           somehow i sat here and played minetest for a few hours without noticing and now my coffee is very cold... bbfn o/
11:49 MinetestBot    02[git] 04nerzhul -> 03minetest/minetest: Server: affect bind_addr on constructor instead of start() (#6474) 132afe629 https://git.io/vdOfQ (152017-09-28T11:47:30Z)
15:12 MinetestBot    02[git] 04orwell96 -> 03minetest/minetest: Add static_save property to luaentites to not save them statically. (… 1308846cd https://git.io/vdOBp (152017-09-28T15:11:51Z)
15:45 ClockGen       How can I get texture name of specific node?
15:46 ClockGen       I mean in code, tnt mod have similar thing to spawn debris particles
15:46 Krock          minetest.registered_nodes[node_name_here].tiles
15:46 ClockGen       But it uses dropped items instead of nodes
15:46 ClockGen       Thanks
16:34 ThomasMonroe   hey jcjordyn120
16:42 MinetestBot    02[git] 04raymoo -> 03minetest/minetest: Document orientation parameter of set_attach (#6473) 135258f3b https://git.io/vdOK2 (152017-09-28T16:40:47Z)
18:58 dogetest       Hello! Should I use _ or . for settings?
19:06 Krock          dogetest, doesn't matter as long there are no conflicts. Some mods use [ modname.setting_name ], others simply [ setting_name ]
19:07 Krock          modname_setting_name is an option too
19:12 dogetest       I going to keep using .
19:42 MinetestBot    02[git] 04SmallJoker -> 03minetest/minetest: Localplayer: Fix disable_jump effect and getStandingNodePos() 13f729b5d https://git.io/vdOjh (152017-09-28T19:36:53Z)
19:42 MinetestBot    02[git] 04HybridDog -> 03minetest/minetest: Lua_api.txt: Clarify arguments of functions, plus some cleanup 137e488b9 https://git.io/vdOjj (152017-09-28T19:32:16Z)
19:51 MinetestBot    02[git] 04paramat -> 03minetest/minetest_game: Beds: Strip colour bits from param2 in 'get look yaw' 13809e593 https://git.io/vd3v8 (152017-09-28T19:49:34Z)
20:03 dogetest       I written a mod and while it works when on mods/ it doesnt when games/subgame/mods/. Specifically, it seens that mod cant wrap nodes definitions and/or wrap register_node. https://github.com/dogetest/dogestate
20:05 deltasquared   hmm, I suspect the elevation bounds for rubber trees needs tweaking to cope with v7 worldgen's tendency to slope up quite quickly
20:06 sfan5          dogetest: you are relying on mod load order while that isn't guaranteed to be a particular order for non-depending mods
20:07 dogetest       sfan5: wouldnt wrapping register_node solve this?
20:07 deltasquared   I don't think you should be tampering with the API in such a way
20:08 sfan5          that's not it
20:08 deltasquared   this sounds like what #archlinux calls an !xyproblem
20:08 sfan5          dogetest: you need to wrap register_node before any mod uses it, but without depending on all other mods it's not guaranteed that yours is loaded first
20:09 deltasquared   mind if I ask *why* one would want to wrap register_node anyway
20:10 deltasquared   meanwhile, off to start my own instance
20:10 sfan5          because he wants to be able to deny an individual dig which register_on_dignode can't do
20:11 deltasquared   "deny an individual dig"... sorry, it's late, I can't parse that
20:11 dogetest       the code wraps each and every node and register_node too
20:14 dogetest       is it even possible to wrap register_node? it doenst raises any error when assigning a new value to it
20:15 deltasquared   perhaps not, though I would not at all be suprised if it were considered undefined or not allowed
20:16 deltasquared   dogetest: I mean *in theory* yes you could, assuming you could sort out the ordering issue somehow (which I don't see how)
20:16 sfan5          dogetest: it's hacky but definitely possible, i explained why it doesn't work already
20:17 deltasquared   though I personally would have done some setmetatable trickery to minetest (or it's C equiv) to make attempting to set values on it throw an error of the sorts "this mod is trying to mess with stuff it shouldn't"
20:17 dogetest       Sorry, but I didnt undestand
20:18 deltasquared   dogetest: you mean what I said? don't worry about it, it was just me rambling
20:18 sfan5          for overriding register_node to work your mod needs to be loaded as the very first mod
20:18 rubenwardy     https://gist.github.com/rubenwardy/9d59118bccc153a724925a7a7566f88d
20:18 dogetest       First I wrap each and every node's definition, then wrap register_node.
20:18 rubenwardy     dogetest ^^
20:18 sfan5          but your mod is not loaded first because every other mod would need to depend on it
20:18 dogetest       so ordershouldnt be a problem I think
20:19 deltasquared   dogetest: the function behaviour is fine. but you cannot retro-actively edit node defs
20:19 dogetest       are mods loaded in parallel?
20:19 sfan5          well either of those methods you're using is not working
20:19 deltasquared   not quite. you just can't edit node defs after they're registered... sfan5 am I right in thinking that the core engine only looks at the node def when it's registered and not after?
20:20 deltasquared   as in, you cannae just go into minetest.registered_nodes and muck around
20:20 dogetest       yes we can
20:20 sfan5          for usual stuff definitely
20:20 sfan5          but editing callbacks should work
20:20 dogetest       its seens the problem is within register_node wrapping
20:21 deltasquared   sfan5: I'm not quite sure he's got it yet... I'ma gonna go elsewhere before I get a bit testy
20:21 dogetest       xD
20:24 dogetest       https://github.com/dogetest/dogestate/blob/master/init.lua#L74
20:24 deltasquared   one moment.
20:25 deltasquared   sfan5: so you say altering the callback functions should be respected by the engine?
20:25 * deltasquared wonders what happened if you did that at runtime
20:25 deltasquared   explosions probably
20:25 sfan5          i don't know for sure but my guess is yes
20:26 sfan5          because: lua functions can't just be "stored" in C
20:26 deltasquared   I'm not really familiar with the lua-C interface myself
20:28 dogetest       rubenwardy: My code seens close to yours
20:29 Krock          you could store lua functions by generating their bytecode and memcpy it into an array or save to the disk :3
20:29 Krock          but the mod security doesn't like that
20:30 dogetest       So... cant override minetest.register_node ?
20:31 rubenwardy     you definitely can
20:31 rubenwardy     I've done it
20:31 Krock          all mods would have to depend on yours to ensure that yours is executed first
20:31 rubenwardy     try using override_items
20:31 Krock          or .. builtin
20:31 rubenwardy     also, some mods may override again
20:31 Krock          so override mods that override items :P
20:32 deltasquared   surely all those mods could do (unless they were seriously misguided) though is further wrap it in their own logic? which may or may not interfere
20:56 dogetest       Its really strange because I cant catch registered nodes at registered_items while on a subgame, but I can while on minetest/mods/. Using registered_nodes instead solves the problem
21:12 dogetest       woow, it got more strange now, because I cant get registered nodes at registered_items on minetest/mods too. But it was working anyway
21:14 dogetest       and still works
21:14 dogetest       oO
21:16 dogetest       oh, it works when running on minetest/mods, because register_node wrap works
21:33 redblade7      hi
21:34 redblade7      does anyone know what the successor bug of this is? https://github.com/minetest/minetest_game/issues/1784#event-1270523031
22:07 jas_           dogetest: i too noticed stark differences in mod loading behavior between external mods and mods internal to the subgame.
22:07 jas_           now that i think of it, this was the original impetus of my moving from worldmods/ to a subgame.  it fixed a dependency error iirc.