| Time |
Nick |
Message |
| 00:16 |
|
behalebabo joined #minetest-hub |
| 01:25 |
|
fluxflux_ joined #minetest-hub |
| 03:44 |
|
Ruslan1 joined #minetest-hub |
| 04:33 |
|
olliy2 joined #minetest-hub |
| 05:28 |
ANAND |
Is ObjectRef:get_pos supposed to return nil for whatever reason? |
| 05:28 |
ANAND |
I have an entity's object, but obj:get_pos() returns nil |
| 05:29 |
ANAND |
Both obj and obj.get_pos exist, btw, otherwise MT would crash |
| 05:34 |
ANAND |
And I'm not able to reproduce it now, although the code is still intact |
| 05:54 |
ANAND |
But I noticed something else. obj:set_attach never attaches the object to the parent |
| 05:55 |
ANAND |
I'm periodically checking if the entity is detached, and re-attach it to the player |
| 05:55 |
ANAND |
re-attaching* |
| 05:59 |
|
olliy joined #minetest-hub |
| 07:15 |
|
CWz joined #minetest-hub |
| 07:34 |
|
hlqkj joined #minetest-hub |
| 07:45 |
|
pyrollo joined #minetest-hub |
| 08:37 |
|
ShadowNinja joined #minetest-hub |
| 08:52 |
|
hlqkj_ joined #minetest-hub |
| 10:53 |
|
Fixer joined #minetest-hub |
| 10:56 |
|
Curious joined #minetest-hub |
| 11:03 |
|
curiousguy joined #minetest-hub |
| 11:58 |
|
olliy joined #minetest-hub |
| 12:11 |
|
aerozoic joined #minetest-hub |
| 12:36 |
|
BuckarooBanzai joined #minetest-hub |
| 13:01 |
|
hlqkj joined #minetest-hub |
| 13:37 |
|
Fixer_ joined #minetest-hub |
| 14:33 |
|
CWz_ joined #minetest-hub |
| 15:37 |
|
HDMI_STECKDOSE joined #minetest-hub |
| 16:07 |
|
BuckarooBanzai0 joined #minetest-hub |
| 16:12 |
|
BuckarooBanzai8 joined #minetest-hub |
| 17:29 |
|
Krock joined #minetest-hub |
| 17:48 |
|
scr267 joined #minetest-hub |
| 18:24 |
|
BuckarooBanzai joined #minetest-hub |
| 18:27 |
|
BuckarooBanzai1 joined #minetest-hub |
| 18:34 |
|
tomraceror joined #minetest-hub |
| 18:35 |
|
tomraceror left #minetest-hub |
| 19:28 |
|
hlqkj joined #minetest-hub |
| 20:06 |
|
nore joined #minetest-hub |
| 21:49 |
rubenwardy |
The rotation of a player object doesn't update when attached to an entity. Any ideas on how to work around this? |
| 21:49 |
rubenwardy |
I tried object.rotation.y = player.look_horizontal, but there was no changes |
| 21:49 |
rubenwardy |
player does have player:set_rotation() |
| 21:49 |
rubenwardy |
*doesn't |
| 21:50 |
rubenwardy |
It would be nice for the idea of server- or client-control to be properly coded in, meaning that set_rotation would work in these situations when the player _object_ is server-controlled |
| 21:50 |
rubenwardy |
actually, I haven't tried that |
| 22:19 |
|
hlqkj joined #minetest-hub |
| 22:51 |
shivajiva |
anyone up for testing https://github.com/shivajiva101/authx |
| 22:53 |
rubenwardy |
you should consider splitting the init.lua file into api.lua, chatcmd.lua, gui.lua at least |
| 22:53 |
rubenwardy |
ideally db.lua too |
| 22:54 |
rubenwardy |
designing mods like that encourages you to make contained API functions, hopefully |
| 22:55 |
rubenwardy |
plus a 3000 line long file is pretty bad |
| 22:55 |
rubenwardy |
but anyway, nice work |
| 22:55 |
rubenwardy |
is it a new and improved schema? |
| 22:56 |
rubenwardy |
having an id field which is not PRIMARY KEY is problematic, and there appears to be a table with no primary keys |
| 22:59 |
shivajiva |
I have this aversion to creating global functions for some reason it seems more appealing to keep it tightly wrapped. The advantage of using auth as part of sban is reduced storage, what tables did you notice that on? |
| 23:01 |
shivajiva |
expired is an odd table due to the multiple names and id's possible |
| 23:03 |
shivajiva |
I suppose I could use a single id entry and serialize the records in a data field to tidy it up |
| 23:04 |
rubenwardy |
`id` should be the primary key of the current table. If it's the id of something else, then it should be tableId like playerId. If it's not then it should probably be removed as unnecessary |
| 23:04 |
rubenwardy |
for example, if you have name as a primary key then an id is unnecessary |
| 23:07 |
shivajiva |
the id is stored because it's tracking data for an account regardless of the index being on name though |
| 23:08 |
rubenwardy |
the point of a primary key is that it uniquely indentifies the current record, and is the key used in foreign keys |
| 23:09 |
rubenwardy |
oh, is the id actually a tracking id to merge multiple accounts? |
| 23:09 |
shivajiva |
yes |
| 23:09 |
rubenwardy |
ahh |
| 23:12 |
shivajiva |
via the assumption of name and ip's being common over a progression of connections I keep track with an id and assign where possible, violations are marked for further analysis by the db admin |
| 23:14 |
shivajiva |
pre emptive bans are possible via a separate table so the real data isn't contaminated |
| 23:14 |
shivajiva |
a 'blacklist' effectively |
| 23:15 |
shivajiva |
but transparent to the ban and unban commands if you have server priv |
| 23:17 |
rubenwardy |
I suggest using a different name to id for that |
| 23:17 |
rubenwardy |
idk what |
| 23:18 |
rubenwardy |
group or clump |
| 23:18 |
rubenwardy |
altSet |
| 23:18 |
rubenwardy |
heh |
| 23:18 |
rubenwardy |
it doesn't matter that much |
| 23:20 |
shivajiva |
if it makes it more readable I'm all for that |
| 23:22 |
shivajiva |
account would suffice instead of id |
| 23:26 |
shivajiva |
if I split the code into parts there's nothing to stop the global functions being used over the api functions so what's the incentive, always wondered that |
| 23:26 |
shivajiva |
if you are up for using a mods api you are up for reading and understanding the code anyway |
| 23:30 |
shivajiva |
this kind of mod needs to be secure in that respect |
