| Time |
Nick |
Message |
| 00:09 |
Viper |
Hi, converted player backed to sqlite, now on startup i get ==> /!\ You are using old auth file backend. This backend is deprecated and will be removed in a future release /!\ |
| 00:12 |
MisterE |
so i run a newly build minetest on openbsd |
| 00:12 |
MisterE |
I get this error |
| 00:12 |
MisterE |
2021-04-07 17:10:54: ERROR[Main]: Failed to get paths by executable location, trying cwd |
| 00:12 |
MisterE |
2021-04-07 17:10:54: ERROR[Main]: Game specified in default_game [minetest] is invalid. |
| 00:12 |
MisterE |
2021-04-07 17:10:55: ERROR[Main]: ServerError: Supplied invalid gamespec |
| 00:12 |
MisterE |
and the server doesnt run |
| 00:16 |
Viper |
MisterE: what is folder name in minetest/games/????? <== |
| 00:16 |
MisterE |
actually, I just figured it out |
| 00:17 |
MisterE |
The other errors are not a big deal; I had to specify a gameid |
| 00:17 |
MisterE |
Thank you :) |
| 00:17 |
Viper |
yw MisterE |
| 00:20 |
|
Out`Of`Control joined #minetest |
| 00:23 |
|
Out`Of`Control joined #minetest |
| 00:25 |
|
Hawk777 joined #minetest |
| 00:34 |
|
katp32 left #minetest |
| 00:35 |
|
erlehmann joined #minetest |
| 00:44 |
|
Verticen joined #minetest |
| 01:37 |
|
LazyJ joined #minetest |
| 02:07 |
|
Surfer2011 joined #minetest |
| 02:10 |
|
Taoki joined #minetest |
| 02:10 |
|
Sven_vB joined #minetest |
| 02:30 |
|
Out`Of`Control joined #minetest |
| 02:51 |
|
Out`Of`Control joined #minetest |
| 03:50 |
|
Sven_vB joined #minetest |
| 04:00 |
|
MTDiscord joined #minetest |
| 04:30 |
|
specing_ joined #minetest |
| 05:58 |
|
Flabb joined #minetest |
| 06:01 |
|
absurb_ joined #minetest |
| 06:21 |
|
macc24 joined #minetest |
| 06:25 |
|
FeXoR joined #minetest |
| 06:30 |
|
olliy joined #minetest |
| 07:05 |
|
hlqkj joined #minetest |
| 08:00 |
|
ShadowNinja joined #minetest |
| 08:25 |
|
v32itas joined #minetest |
| 08:32 |
|
qjq joined #minetest |
| 08:58 |
|
calcul0n_ joined #minetest |
| 09:32 |
|
galex-713 joined #minetest |
| 09:33 |
MTDiscord |
<jordan4ibanez> Are the floatlands enabled in v7 mapgen yet? |
| 09:38 |
|
MadScientist joined #minetest |
| 09:41 |
|
proller joined #minetest |
| 09:55 |
|
hlqkj joined #minetest |
| 10:04 |
|
riff-IRC joined #minetest |
| 10:29 |
|
Peppy joined #minetest |
| 10:39 |
|
Andrew__ joined #minetest |
| 10:39 |
Andrew__ |
what does staff of light in ethereal do |
| 10:56 |
|
RetroBoy64 joined #minetest |
| 10:57 |
RetroBoy64 |
Hello everyone, how do I make it so more mobs spawn on the world? |
| 10:58 |
RetroBoy64 |
I have mods installed by the way |
| 11:00 |
|
RetroBoy64 left #minetest |
| 11:00 |
|
RetroBoy64 joined #minetest |
| 11:10 |
|
tech_exorcist joined #minetest |
| 11:12 |
Andrew__ |
RetroBoy64: They just spawn. Note that the "mobs" mod only provides a API for mobs, you need something like "mobs_monster" to make the actual mob entities. |
| 11:14 |
|
fmlatghor joined #minetest |
| 11:14 |
|
specing joined #minetest |
| 11:19 |
|
Fixer joined #minetest |
| 11:22 |
RetroBoy64 |
I have mobs animals as well |
| 11:23 |
RetroBoy64 |
But I rarely ever see them on the world, it's not like mod doesn't work either, becaue on the creative mod I can spawn mods all day |
| 11:23 |
Andrew__ |
hmm weird |
| 11:23 |
Andrew__ |
whats your worldgen |
| 11:26 |
RetroBoy64 |
let me take a look |
| 11:27 |
RetroBoy64 |
how do I view the world gen (seed)? There is no visible option on the menu |
| 11:31 |
RetroBoy64 |
They are a bit more common on water but I almost never see them on land |
| 11:33 |
|
qj joined #minetest |
| 11:35 |
|
qjq joined #minetest |
| 11:38 |
|
Verticen joined #minetest |
| 11:39 |
RetroBoy64 |
Should I come back later andrew? |
| 11:41 |
|
RetroBoy64 left #minetest |
| 11:52 |
|
Sven_vB joined #minetest |
| 11:58 |
|
proller joined #minetest |
| 12:17 |
Andrew__ |
idk :\ |
| 12:34 |
|
calcul0n__ joined #minetest |
| 12:37 |
Andrew__ |
I have a problem with 185.57.229.162 who is trying to hack my server |
| 12:49 |
|
tech_exorcist joined #minetest |
| 13:14 |
|
kamdard joined #minetest |
| 13:26 |
|
hlqkj joined #minetest |
| 13:30 |
MTDiscord |
<IhrFussel> Is it one of those auto bots that try to brute force access via ssh? |
| 13:31 |
Andrew__ |
No, they accessed via MT not ssh |
| 13:33 |
MTDiscord |
<IhrFussel> How do you know this? It's much more likely that one of your mods is missing some kind of check if you mean the world was modified or admin items/commands were used |
| 13:47 |
|
expert975 joined #minetest |
| 13:51 |
|
DS-minetest joined #minetest |
| 14:03 |
|
galex-713 joined #minetest |
| 14:17 |
|
galex-713 joined #minetest |
| 14:19 |
|
Sven_vB joined #minetest |
| 14:23 |
|
specing joined #minetest |
| 14:55 |
|
est31 joined #minetest |
| 14:59 |
|
expert975 joined #minetest |
| 15:17 |
|
Hawk777 joined #minetest |
| 16:41 |
|
sec^nd joined #minetest |
| 16:53 |
|
Talkless joined #minetest |
| 16:54 |
|
sec^nd joined #minetest |
| 17:03 |
|
Peppy joined #minetest |
| 17:07 |
|
sec^nd joined #minetest |
| 17:13 |
|
sec^nd joined #minetest |
| 17:32 |
|
ar10ch joined #minetest |
| 17:46 |
ar10ch |
in a formspec, can field values be locked so that only privileged users can change them? like by pressing a button to lock / unlock and checking player privs at that stage? |
| 17:50 |
MTDiscord |
<Warr1024> I'm pretty sure you get to decide what formspec elements to send to a player and can determine which fields to make actually editable at that time. |
| 17:51 |
MTDiscord |
<Warr1024> Keep in mind that players can modify formspec responses arbitrarily, including completely forging entire responses, so you can't trust them and need to fully validate them when received server-side. |
| 17:54 |
ar10ch |
Warr1024: ok thanks, it occurs to me to have a button to a "2nd page" and check the privs at that point |
| 17:54 |
Sokomine |
ar10ch: you can send a specific formspec to each user. in that you can offer whatever options you want. but: never trust what the user sends to the server. always check user input |
| 17:54 |
Sokomine |
oh. warr1024 replied already |
| 17:55 |
ar10ch |
Sokomine: thanks :) |
| 17:55 |
Sokomine |
well. question of my own: has a mod for "chatting" with npc (rpg-like dialogs) be written by now? i planned to do so - but had very little time (still trying to catch up with the forum) |
| 17:56 |
MTDiscord |
<Warr1024> Using the "formspec" metadata on a node does not give you the opportunity to do the checks, but you can do things like handling the right-click server-side in that case. I think there MAY be a network delay cost to that, but I would think that should be rarely an issue in practice. |
| 17:58 |
Sokomine |
yes, if you set the metadata and let the client do all the showing of the formspec then it's usually faster - but you have no choice as to what to show to the specific user. what is better depends on the situation. if it has to be specific, send a specific formspec to the user. otherwise don't. i'm still highly unpleased about the change where locked chests got made "secret" and thus got horribly slow |
| 17:58 |
Sokomine |
and unusable for me on many servers |
| 17:58 |
DS-minetest |
Sokomine: I've seen npc chatting on the Your Land server |
| 17:58 |
Sokomine |
oh. that sounds nice. havn't visited that one |
| 18:00 |
ar10ch |
A while ago I thought about using robots as "tutors", picking out keywords in input strings, so something along the lines of npc-chatting sounds great. |
| 18:01 |
ar10ch |
I dont know how to send "personalised" formspecs yet, but if it works - then probably hiding the fields part of the formspec behind a priv-checking button should suit the use-case |
| 18:02 |
Sokomine |
i'm thinking more along the line of old point and click adventures where you can select one option. not them spamming chat or trying to react to freely typed sentences |
| 18:02 |
ar10ch |
ah ok |
| 18:02 |
ar10ch |
Ah i know what u mean now |
| 18:03 |
MTDiscord |
<Warr1024> I had briefly toyed with the idea of formspec dialog, but I really hated the way it blocked the 3D scene behind it, I had rather little control over how formspecs were laid out on screen dependent on dimensions, and controlling the scene was problematic in general. |
| 18:04 |
MTDiscord |
<Warr1024> Some time I might want to play around with a formspec-less approach though, e.g. using tangible items or ents in 3D space for speaking, and maybe just floating ents or even waypoint HUDs for listening... |
| 18:05 |
Sokomine |
warr1024: the blocking out of the 3d scene will remain. there's a bit more control over how formspecs are laid out now. it became possible to write such chats some time ago |
| 18:09 |
|
MTDiscord joined #minetest |
| 18:12 |
erlehmann |
hey i just want your attention. in every version of mesecons you can put a button at the world border to crash the server. https://git.minetest.land/MineClone2/MineClone2/issues/1494 |
| 18:12 |
erlehmann |
the issue contains a fix |
| 18:12 |
Hawk777 |
Maybe that should be reported to Mesecons? |
| 18:13 |
erlehmann |
i reported it to mineclone2 |
| 18:13 |
Hawk777 |
Also, is it only buttons, or anything else? |
| 18:13 |
erlehmann |
but it was marked as UNCONFIRMED |
| 18:13 |
Hawk777 |
And two hours later someone confirmed it. |
| 18:13 |
erlehmann |
no, one developer of minclone2 (jordan4ibanez) now uses it to crash servers instead of fixing it |
| 18:13 |
erlehmann |
he uploaded a video to youtube even of repeatedly crashing a server people are on |
| 18:14 |
erlehmann |
he just put a button on the world border and kept pressing it |
| 18:14 |
erlehmann |
the issue is still unconfirmed |
| 18:14 |
DS-minetest |
link or it didn't happen |
| 18:14 |
Hawk777 |
Sure the unconfirmed *label* is there, but the confirmation message from cora is also *only three hours old*. |
| 18:15 |
sfan5 |
I guess fast travel is viable in MCL2 since getting to the world border via normal means is quite a feat in MTG? |
| 18:15 |
erlehmann |
sfan5, 7 minutes via nether |
| 18:15 |
erlehmann |
https://youtub.be/IR44kDUxSQQ (from their discord) |
| 18:16 |
sfan5 |
!title |
| 18:16 |
MinetestBot |
urllib.error.URLError: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:852)> (file "/usr/lib/python3.6/urllib/request.py", line 1327, in do_open) |
| 18:16 |
sfan5 |
uh oh |
| 18:16 |
sfan5 |
weird |
| 18:16 |
erlehmann |
sorry |
| 18:16 |
erlehmann |
it is this https://youtu.be/IR44kDUxSQQ |
| 18:16 |
erlehmann |
typo |
| 18:16 |
erlehmann |
so yeah |
| 18:16 |
erlehmann |
this developer in particular has commit rights to master branch |
| 18:16 |
erlehmann |
and *could* review the fix |
| 18:16 |
DS-minetest |
!title |
| 18:16 |
MinetestBot |
No title found. |
| 18:17 |
DS-minetest |
!title https://youtu.be/IR44kDUxSQQ |
| 18:17 |
MinetestBot |
No title found. |
| 18:17 |
sfan5 |
yeah it's broken |
| 18:17 |
erlehmann |
so i just want to warn all of you |
| 18:17 |
erlehmann |
if you have mesecons or mineclone2 |
| 18:17 |
erlehmann |
pull the fix |
| 18:17 |
Hawk777 |
Is there a reason you don’t want to report the issue to Mesecons? |
| 18:17 |
MTDiscord |
<Warr1024> https://xkcd.com/242/ |
| 18:17 |
erlehmann |
and DO NOT rely on mcl2 devs to fix it |
| 18:17 |
erlehmann |
Hawk777 i have reported the issue to a project and devs started abusing it instead of fixing it |
| 18:18 |
Hawk777 |
Well if you don’t report the issue to Mesecons, how can you expect the Mesecons devs to fix it? |
| 18:18 |
MTDiscord |
<Warr1024> If the devs wanted to be able to exploit their own code instead of fixing it, you'd think they could have just added an actual back-door anyway. |
| 18:19 |
ar10ch |
yeah but do the mineclone "redstone" buttons do the same thing? |
| 18:19 |
erlehmann |
look whatever you report it |
| 18:19 |
erlehmann |
its a fork with different skins |
| 18:19 |
|
Fixer joined #minetest |
| 18:19 |
erlehmann |
so don't worry |
| 18:20 |
Hawk777 |
I’m busy right now. I’m not going to take the time to try and repro the bug to check whether it happens with Mesecons or only Mineclone, and then make a ticket about it. It sounded like you cared a lot about the issue, so I thought you would care enough to report it to Mesecons (especially if Mineclone just forks Mesecons, it would be easier on both parties if the fix came from upstream). |
| 18:22 |
DS-minetest |
I find it quite sad that every attempt to replicate redstone in minetest that I've seen so far just uses mesecons and changes the textures. afaik mesecons and redstone are very different, redstone doesn't even have the same concept of conductors |
| 18:22 |
MTDiscord |
<Warr1024> Chances are good there are related bugs as well, like different kinds of input controls or devices that cause crashes near the map border, too. It's awfully rare to run into a "forgot that ignores can be within reach of players" kinds of bugs that's not part of a whole class of bugs... |
| 18:22 |
Hawk777 |
Maybe if I remember on the weekend I might try it out. Or I might forget about it by then. |
| 18:23 |
Hawk777 |
It doesn’t really affect me personally; I play on a closed-group server and we’re nowhere near the map border anyway. |
| 18:23 |
MTDiscord |
<Warr1024> Said jordan4ibanez actually had a rather impressively redstone-esque redstone remake of his own in Crafter... |
| 18:24 |
DS-minetest |
ah, right, I've forgot that one. though I've never seen it in-game, and I found it weird how it modified the abm timer settings, urgh. |
| 18:24 |
MTDiscord |
<Warr1024> NodeCore's nc_optics does a decent job capturing the original spirit of MC's redstone but the actual mechanics of it are quite different. |
| 18:25 |
MTDiscord |
<Warr1024> You can't really do something quite like redstone using ABMs anyway. The original stuff ran on a 20Hz fixed step. NC optics run at 12Hz. Most of the rest that I've seen do stuff either instantaneously or after like a full second delay. |
| 18:26 |
DS-minetest |
I guess I should try out nc in creative then some time. on that one nc server, I only came so far to produce glass, no more |
| 18:26 |
MTDiscord |
<Warr1024> producing glass is very close to what you need to do optics |
| 18:26 |
MTDiscord |
<Warr1024> i.e. you just need to produce the right kind of glass (depends on how you cool it) and then chip/chop that glass into optic components. |
| 18:27 |
DS-minetest |
yeah, the chip/chop thing is the problematic part here |
| 18:27 |
erlehmann |
look |
| 18:27 |
erlehmann |
there is only one single bug |
| 18:27 |
MTDiscord |
<Warr1024> you'd need fairly high-tier tools for it at least. |
| 18:27 |
erlehmann |
the function that depowers the world border |
| 18:27 |
erlehmann |
does not check for getting nil |
| 18:27 |
erlehmann |
from the ignore blocks at the border |
| 18:27 |
erlehmann |
bc whoever wrote it thought they could forceload all the blocks |
| 18:27 |
DS-minetest |
"there is only one single bug" lmao xD |
| 18:27 |
erlehmann |
to depower them |
| 18:27 |
erlehmann |
so it happens every time you depower the world border |
| 18:28 |
erlehmann |
regardless how you do it |
| 18:28 |
MTDiscord |
<Warr1024> I mean, it's possible for one true underlying bug to have a ton of presentations/symptoms; it just depends on how you count bugs. |
| 18:28 |
erlehmann |
if you want to be evil, put a mob that can jump in an enclosure and put a pressure plate under it |
| 18:28 |
MTDiscord |
<Warr1024> Like, "the creator of this web service clearly doesn't know a damn thing about SQL injection" is "just one" bug too :-) |
| 18:28 |
erlehmann |
then try to keep the part of the map where it is in active via some mens |
| 18:28 |
erlehmann |
means |
| 18:30 |
DS-minetest |
yeah, but I think it's quite hard to adequately summarize all bugs in all minetest related stuff, or at least only those from mesecons, to one bug |
| 18:33 |
ar10ch |
on the matter of world borders: is it possible to reduce the worlds radius? would this achieve any improvement in performance (less to load / generate) ? |
| 18:34 |
DS-minetest |
FYI, I haven't managed to cause a crash at world boarder by turning anything off there |
| 18:34 |
DS-minetest |
maybe something weird in the mineclone mesecons fork calls mesecon.turnoff somewhere |
| 18:36 |
erlehmann |
ah cool |
| 18:36 |
erlehmann |
DS-minetest, good to know |
| 18:36 |
erlehmann |
DS-minetest, i am relieved |
| 18:37 |
DS-minetest |
there are easier ways to crash a server using mesecons anyway |
| 18:37 |
MTDiscord |
<Warr1024> You can reduce the size of a world by setting mapgen_limit, before the areas in question get generated. |
| 18:37 |
erlehmann |
ok! |
| 18:37 |
erlehmann |
thank you |
| 18:37 |
MTDiscord |
<Warr1024> This does not prevent players from reaching the world border, it just makes that border be closer to the center. |
| 18:38 |
MTDiscord |
<Warr1024> it does help reduce the disk space that a world can occupy, which is really helpful if you want to offer downloads or keep rotating backups |
| 18:38 |
DS-minetest |
erlehmann: it is possible that I didn't try hard enough or just did something wrong (I didn't find clear reproduction instructions). it would be better if you tested it yourself, too |
| 18:38 |
MTDiscord |
<Warr1024> And you can also raise the mapgen_limit later if you decide you do want to expand, and it will generate the new terrain just fine as far as I've seen. |
| 18:39 |
ar10ch |
Warr1024: that's great, thanks! I was just about to ask the last bit too lol |
| 18:39 |
erlehmann |
DS-minetest, it would be easier to check if mesecons actually had the fix i think |
| 18:40 |
MTDiscord |
<Warr1024> If you set it in minetest.conf, then any new world you create will inherit the mapgen_limit setting, but it's "baked" into worlds in I think map_meta.txt, so that's where you'd have to change it if you want to modify existing worlds. |
| 18:40 |
DS-minetest |
erlehmann: what fix are you referring to? |
| 18:40 |
erlehmann |
DS-minetest, the issue contains a fix for the mcl2 code to make it not crash by checking for nil when forceloading a node to depower it |
| 18:40 |
erlehmann |
maybe mesecons already does this |
| 18:41 |
MTDiscord |
<Warr1024> I think you can also set mapgen_limit insanely low on an existing world, like lower it to 128 nodes or something, and it will de facto mean "anywhere already explored is fine, but no new mapgen anywhere else". The ungenerated areas will have "ignores" in them and you can't walk into those places. |
| 18:41 |
DS-minetest |
erlehmann: but how would this help in reproduction attempts? |
| 18:41 |
erlehmann |
you can actually very easily get into these places, the hard part is getting out again |
| 18:41 |
erlehmann |
DS-minetest, it shows exactly what to trigger, a depowering of the world border |
| 18:42 |
erlehmann |
i will test it! |
| 18:43 |
DS-minetest |
erlehmann: not really exact. it just shows a diff in .turnoff. a stack trace would be more helpful |
| 18:55 |
erlehmann |
DS-minetest, i have confirmed the bug is not in mesecons. but i found a hilarious dupe in the process. |
| 18:56 |
|
qj joined #minetest |
| 19:00 |
erlehmann |
DS-minetest, MTDiscord sorry for overreacting that much |
| 19:00 |
erlehmann |
you can still crash all mcl2 servers with this |
| 19:00 |
Hawk777 |
IMO still worth reporting the dupe bug to Mesecons. It’s less serious but should still be fixed. |
| 19:01 |
Hawk777 |
And I’m sorry if I came off as dismissive; I just found it frustrating that it looked like there was an important bug in Mesecons but you didn’t seem to want to report it to Mesecons. |
| 19:02 |
Hawk777 |
Good to hear it’s not actually there though. |
| 19:06 |
|
MTDiscord joined #minetest |
| 19:08 |
MTDiscord |
<Warr1024> Might be interesting to dig through the mesecons history, or check the blame on that bit of code, and see if they actually found and fixed the same bug pre-fork. |
| 19:10 |
DS-minetest |
as said before, a stack trace would be helpful. normally you don't call .turnoff for a non-mesecons node, afaik |
| 19:12 |
DS-minetest |
edit: forget it, I'm stupid |
| 19:12 |
erlehmann |
oh wow |
| 19:13 |
erlehmann |
its even worse |
| 19:13 |
erlehmann |
the asshole developer made redstone machines at the world border blow up instead of merging the fix |
| 19:14 |
|
homthack joined #minetest |
| 19:15 |
DS-minetest |
hm, just checked the history, the ignore check was always there in mesecons, so mcl2's faul |
| 19:16 |
|
I_am_6r1d joined #minetest |
| 19:20 |
ar10ch |
tbh i only tested the mcl game i downloaded way back very briefly but things seem to work - the server ive been on has been messed with horribly. the boat sinks, the protected doors/chests cannot be removed and mcl_weather causes so much lag that the game becomes unplayable. I dont know how many versions of mcl2 are being run on servers, but if theyre all like that then more bugs are little surprise |
| 19:24 |
|
qj joined #minetest |
| 19:28 |
|
qj joined #minetest |
| 19:32 |
MTDiscord |
<Danil_2461> so i have some questions: 1. who made player_api? 2. is it okay to use it in my game as long as i credit the creator and not violate the rules of the license? |
| 19:34 |
Krock |
@Danil_2461 it originates from "default" |
| 19:34 |
Krock |
many people worked on it |
| 19:35 |
Krock |
but since it's licensed as MIT you only have to attribute the source, and if it refers to a git history, it also documents the authors |
| 19:35 |
Krock |
I'd say just copypaste it. |
| 19:38 |
MTDiscord |
<Jonathon> Player_api from mtg is lgplv2 |
| 19:38 |
Krock |
weird |
| 19:39 |
MTDiscord |
<Jonathon> Aka only use it if you want gpl virus in your work |
| 19:39 |
specing |
LGPLv2 and GPL virus? What are you smoking? |
| 19:40 |
Krock |
I wonder why I thought that it's MIT |
| 19:40 |
Krock |
well then. you gotta need a LGPL2.1+ compatible license |
| 19:41 |
Krock |
mtg_craftguide is sublicensed as MIT |
| 19:41 |
Krock |
s/sub// |
| 19:41 |
Krock |
binoculars too |
| 19:42 |
Krock |
actually most of it is MIT |
| 19:43 |
MTDiscord |
<Jonathon> Mit is nice since others can use a "higher" license if they want without being forced to |
| 19:46 |
MTDiscord |
<jordan4ibanez> I merged your fix erlhemann, I'm still new to being a core member of the project, cut me a break |
| 19:46 |
DS-minetest |
why does .item_place_node not use core.set_last_run_mod when calling callbacks? |
| 19:46 |
DS-minetest |
züh züh züh |
| 19:52 |
MTDiscord |
<jordan4ibanez> But I just reverted it, so that someone else can merge it properly, fixed |
| 20:10 |
MTDiscord |
<Warr1024> Making machines explode instead of work as expected is a perfectly valid way to address the problem too |
| 20:11 |
MTDiscord |
<Warr1024> I guess the "as expected" part sort of depends on whether you're the type of developer wherein people should be expecting explosions |
| 20:12 |
DS-minetest |
the problem with explosions is that they shouldn't violate protection of other players |
| 20:13 |
MTDiscord |
<Warr1024> Sure, though again, that depends on what you're going for design-wise. |
| 20:14 |
MTDiscord |
<Warr1024> I had this debate in another forum already, but it seems that people seem to think "protection should render my creation completely inert except for changes of my own volition", which is fine for like a creative-type game. |
| 20:15 |
MTDiscord |
<Warr1024> For more survival-focused gameplay though, either protection may still subject players to the game's mechanics and consequences thereof, or protection support might not even be a priority. |
| 20:15 |
MTDiscord |
<Warr1024> If it's MCL2 it's a bit of a hard problem since my understanding of that game is that it's really targeting a mix of play styles. |
| 20:24 |
|
Fixer_ joined #minetest |
| 20:27 |
|
shirfty joined #minetest |
| 20:29 |
erlehmann |
<Warr1024> Making machines explode instead of work as expected is a perfectly valid way to address the problem too |
| 20:29 |
erlehmann |
that is wrong |
| 20:30 |
|
delta23 joined #minetest |
| 20:30 |
erlehmann |
Warr1024 the explosion might actually trigger another depowering, reccursively exploding until the lua stack overflows |
| 20:31 |
erlehmann |
i am not sure how much redstone you would need for it |
| 20:31 |
erlehmann |
but it does not really matter |
| 20:31 |
erlehmann |
bc if i understand it right mesecons keeps all mapblocks active where there is something |
| 20:32 |
erlehmann |
Warr1024 there can be issues where explosion is the correct thing |
| 20:32 |
erlehmann |
but in this case i think it was just “haha i am a dev and a griefer” |
| 20:38 |
ar10ch |
Yay - got me "personalised" priv-based formspecs |
| 20:38 |
|
Wuzzy joined #minetest |
| 20:40 |
ar10ch |
only thing I have not done is save the inputs, but I'm guessing they get saved as meta strings? |
| 20:56 |
|
proller joined #minetest |
| 20:58 |
|
Taoki joined #minetest |
| 21:02 |
Hawk777 |
erlehmann, it used to keep a lot of mapblocks active, but for about 4½ years, Mesecons proper has used voxelmanipulators instead so that long wires can run through inactive mapblocks (I actually wrote that code)—the mapblocks will be loaded, but not activated. I don’t know whether Mineclone has that change or not. |
| 21:04 |
erlehmann |
Hawk777, ah ok! |
| 21:04 |
erlehmann |
i see |
| 21:05 |
erlehmann |
Hawk777, could you maybe look at the proposed fix here? https://git.minetest.land/MineClone2/MineClone2/issues/1494 |
| 21:05 |
Hawk777 |
It was this PR <https://github.com/minetest-mods/mesecons/pull/279>. |
| 21:06 |
erlehmann |
Hawk777, query |
| 21:06 |
Hawk777 |
I’m not sure why you call “get_node_force” twice rather than just checking “if n ~= nil”. |
| 21:10 |
specing |
Hawk777: Has any thought been put into storing wiring/machine networks separately from block data? |
| 21:13 |
Hawk777 |
Certainly. See the extensive discussions on <https://github.com/minetest-mods/mesecons/issues/277> and <https://github.com/minetest-mods/mesecons/pull/279>. The resolution was that it would probably improve performance but also be a huge amount of work to get right (including in the face of server crashes half way through updating data structures), and nobody has done it. |
| 21:14 |
specing |
thanks |
| 21:17 |
Hawk777 |
erlehmann: one other thing, get_node_force in older versions (where it used active blocks instead of voxelmanipulators) could sometimes return nil even for legitimate parts of the world, specifically if there aren’t enough force-activation slots available (I forget what they’re actually called, equivalent to what Minecraft calls chunkloading tickets I think). There was some logic somewhere about deferring wire updates to a lat |
| 21:17 |
Hawk777 |
ck in that situation (which is no longer present in current Mesecons because, with voxelmanipulators, it’s unnecessary), but it’s worth thinking through if your proposed change would mean that, in that situation, you could hit that and have part of a wire be on and part be off. Although that code as written looks like it would previously have just crashed in that situation, so probably no worse? |
| 21:17 |
Hawk777 |
I’m not actually certain *why* current Mesecons doesn’t crash under the circumstances you described. |
| 21:18 |
erlehmann |
<Hawk777> I’m not sure why you call “get_node_force” twice rather than just checking “if n ~= nil”. |
| 21:18 |
erlehmann |
see that is why i wanted a code review |
| 21:42 |
|
Verticen joined #minetest |
| 21:44 |
|
deepman joined #minetest |
| 21:55 |
|
delta23 joined #minetest |
| 22:00 |
|
Sven_vB joined #minetest |
| 22:17 |
|
Taoki joined #minetest |
| 22:27 |
|
Taoki joined #minetest |
| 22:38 |
|
ar10ch joined #minetest |
| 22:41 |
ar10ch |
what is formname? If i label my formspec as "form1", what does formname do? Is it a sub of the formspec? |
| 22:42 |
MTDiscord |
<Jonathon> its for when you use minetest.register_on_player_receive_fields |
| 22:42 |
ar10ch |
ah |
| 22:42 |
MTDiscord |
<Jonathon> see receiving formspec submissions https://rubenwardy.com/minetest_modding_book/en/players/formspecs.html |
| 22:44 |
ar10ch |
i had a working formspec with buttons and input fields. Teleportation was a simple test. Then I added a priv-check and since then 2 different formspecs show for privileged / unprivileged users but the buttons no longer work. Gutted! Had hoped formname would be the key |
| 22:51 |
|
Sven_vB joined #minetest |
| 22:55 |
ar10ch |
Hmm.. what about minetest.show_formspec? I want to show my formspec in the event that the condition is met, so what is formname then? |
| 22:56 |
MTDiscord |
<Warr1024> Kind of makes me wonder why form name even needs to be a thing in the API; you could just as easily embed a hidden field in the form anyway to identify it, and the work you have to do to figure out if it's the form you want is more or less the same either way. |
| 22:57 |
MTDiscord |
<Warr1024> basically for formname, just make up a reasonable name for it, though prefixing it with your mod's technical name might be a good way to make sure it doesn't collide with anybody else's name. |
| 22:59 |
ar10ch |
I set it to various things - it usually works, but yeah I should follow a naming convention before I finish this exercise. I think I'll accept my progress so far until tomo and figure out why adding the priv-check feature until then |
| 23:00 |
ar10ch |
*why adding the priv feature broke it lol |
| 23:00 |
MTDiscord |
<Warr1024> minetest.get_current_modname() .. ":whatever" is usually one of the safer ways to go for naming stuff. |
| 23:03 |
|
Ruud joined #minetest |
| 23:04 |
ar10ch |
The idea that prompted me toward this exercise was that I was teleporting players with their chests (intact) but the point of that was to be something to aim for as a reward on a tough survival server, So I thought a formspec and privs had to be added |
| 23:05 |
MTDiscord |
<Warr1024> teleporting players without chests would be awkward |
| 23:06 |
MTDiscord |
<Warr1024> lotta important organs in there |
| 23:06 |
ar10ch |
haha |
| 23:06 |
ar10ch |
that took me a min |
| 23:06 |
MTDiscord |
<Warr1024> though I guess if you're going for "tough survival" then trying to do it without heart/lungs is about as tough as it gets |
| 23:08 |
DS-minetest |
hmmm, you are giving me ideas |
| 23:08 |
ar10ch |
yeah knife-project the player model in blender |
| 23:09 |
ar10ch |
walk around like a polo mint |
| 23:09 |
ar10ch |
a square polo mint |
| 23:10 |
* DS-minetest |
likes to implement weird mod ideas (https://github.com/Desour/crying_obsydian) |
| 23:12 |
MTDiscord |
<Warr1024> Haha, I like that |
| 23:12 |
MTDiscord |
<Warr1024> Crying obsidian is a callback to an MC feature that was never properly implemented, got removed, and then got readded later, right? |
| 23:12 |
ar10ch |
Just reading thru - looks interesting |
| 23:12 |
MTDiscord |
<Warr1024> It looks like that mod adds sound to it... :-) |
| 23:13 |
DS-minetest |
umm, it's crying obsidian |
| 23:13 |
ar10ch |
haha |
| 23:13 |
DS-minetest |
idk, what it has to do with mc |
| 23:13 |
DS-minetest |
¯\_(ツ)_/¯ |
| 23:14 |
Hawk777 |
MC does have something with the same name, not at all the same thing though. https://minecraft.fandom.com/wiki/Crying_Obsidian |
| 23:14 |
MTDiscord |
<Warr1024> https://minecraft.fandom.com/wiki/Crying_Obsidian#Trivia |
| 23:15 |
MTDiscord |
<Warr1024> I remember when it was added the first time, and I think I remember it being removed |
| 23:15 |
MTDiscord |
<Warr1024> it had never actually been used at the time though. I didn't know what the plans they had for it were. |
| 23:15 |
MTDiscord |
<Warr1024> I suppose they caved to fan pressure to add it in later though |
| 23:16 |
MTDiscord |
<Warr1024> They added it back in last year ... only about 7 years too late for me to try it out in MC. |
| 23:18 |
|
jess joined #minetest |
| 23:21 |
ar10ch |
hold on my brain cells are trying to regroup - hiding somewhere in a corner... |
| 23:23 |
ar10ch |
will anyone be offended if i paste my working code, with a hope to seeing how to add the priv check? I can paste the broken-button version after too |
| 23:25 |
DS-minetest |
use a pastebin service |
| 23:26 |
ar10ch |
yup |
| 23:26 |
ar10ch |
https://pastebin.pl/view/5b4a60e2 |
| 23:29 |
DS-minetest |
(node meta is typically set on_construct) |
| 23:30 |
ar10ch |
the code shown works fine - its just as i add the priv check that the buttons stop working. |
| 23:30 |
ar10ch |
hold on |
| 23:30 |
ar10ch |
ill paste |
| 23:30 |
DS-minetest |
if you want to have different formspecs depending on the player, you have to send the formspec string every time when the node's form is accessed. => use .show_formspec in on_rightclick and not formspec in meta |
| 23:33 |
ar10ch |
this works (to show different formspecs for privileged / unprivileged players) but the buttons are broken |
| 23:33 |
ar10ch |
https://pastebin.pl/view/ee36ef0d |
| 23:35 |
ar10ch |
ah... maybe i get what ur saying - so no meta string for formspec? |
| 23:37 |
ar10ch |
oh wait - is it broken because they all have the same names? oh duh |
| 23:39 |
Hawk777 |
That pastebin site is dead for me. |
| 23:40 |
ar10ch |
the last one? |
| 23:40 |
Hawk777 |
Yeah. |
| 23:40 |
Hawk777 |
Actually both now I think. For me all of pastebin.pl seems to be dead. Was working a few minutes ago. |
| 23:41 |
ar10ch |
yeah i 404'd. Think I broke pastebin. Sorry :( |
| 23:45 |
Hawk777 |
Looks like it’s back now. |
| 23:53 |
ar10ch |
https://pastebin.pl/view/578099cd |
| 23:54 |
ar10ch |
https://pastebin.pl/view/f644776d |
