| Time |
Nick |
Message |
| 00:00 |
GreenDimond |
also Minetest |
| 00:00 |
sofar |
from lua? |
| 00:00 |
GreenDimond |
Yes |
| 00:00 |
sofar |
how much data is it? |
| 00:00 |
sofar |
5, 50, or 500000? |
| 00:01 |
GreenDimond |
Can be as little as 50 characters to as much as 5 million |
| 00:01 |
sofar |
how many elements? |
| 00:01 |
sofar |
not byte size |
| 00:01 |
GreenDimond |
What I currently store in one line would be an element right? |
| 00:01 |
GreenDimond |
each line is a serialized table, btw |
| 00:02 |
sofar |
if you define that to be "an element" then yes |
| 00:02 |
sofar |
e..g. think row in an sql table |
| 00:02 |
GreenDimond |
I have little knowledge of how databases work |
| 00:03 |
GreenDimond |
In theory the amount of elements could be anything, but at a sane level it would probably be no more than 10 million. |
| 00:03 |
GreenDimond |
Average of less than 5 mil |
| 00:05 |
GreenDimond |
could even be 1 element |
| 00:05 |
sofar |
I think you should consider using something like lsqlite3 or even an actual SQL backend then |
| 00:06 |
GreenDimond |
Alright, then the question of how is it read? Can I access indexes individually (similar to a table)? |
| 00:06 |
sofar |
Yes, you just query by e.g. index number, or some other data property |
| 00:06 |
GreenDimond |
and how fast is it |
| 00:07 |
GreenDimond |
Because I could be accessing thousands of indexes per second |
| 00:07 |
sofar |
Usec response times |
| 00:07 |
sofar |
Even large queries can be fast |
| 00:07 |
sofar |
Depends on how good you write queries and design your sql tables |
| 00:08 |
sofar |
It uses lsqlite3 |
| 00:08 |
sofar |
Itb* |
| 00:09 |
GreenDimond |
I suppose I should look at ITB and sauth/sban code |
| 00:10 |
rubenwardy |
bad uses of lsqlite3 can result in massive lag |
| 00:10 |
rubenwardy |
ie: sauth and sban are very sloq |
| 00:10 |
rubenwardy |
before the sauth optimisations |
| 00:10 |
GreenDimond |
Yeah I was sort of aware of that |
| 00:10 |
rubenwardy |
which essentially resulting in in-memory caching being used |
| 00:10 |
sofar |
There are alternatives, and there are ways to make mistakes :) |
| 00:11 |
GreenDimond |
Are databases more compressed too? |
| 00:11 |
sofar |
No |
| 00:11 |
GreenDimond |
Less? |
| 00:11 |
sofar |
No |
| 00:11 |
GreenDimond |
So there is no space gain |
| 00:11 |
rubenwardy |
I think it would be very beneficial to have an engine API for nonrelational data |
| 00:11 |
sofar |
Well generally SQL databases have a nontrivial overhead |
| 00:12 |
rubenwardy |
ie: arrays of records potentially with a key, but not necessarily |
| 00:12 |
GreenDimond |
But I suppose because it will no longer be easily accessible it doesnt need to be human-readable at that point, so minetest.compress will help. |
| 00:12 |
rubenwardy |
unsure on how this should be implemented |
| 00:12 |
sofar |
You can put compressed data in the DB, though |
| 00:12 |
rubenwardy |
but builtin caching would be good |
| 00:12 |
sofar |
My does this |
| 00:13 |
* rubenwardy |
reads up on recent security news for my exam tomorrow |
| 00:13 |
rubenwardy |
Fortnite had a vulnerabilty |
| 00:13 |
rubenwardy |
sounds about right |
| 00:13 |
sofar |
Hah |
| 00:13 |
rubenwardy |
dump of 3 billion user records found on hacker forums, announced tomorrow |
| 00:13 |
rubenwardy |
because time zones |
| 00:13 |
rubenwardy |
https://www.troyhunt.com/the-773-million-record-collection-1-data-reach/ |
| 00:14 |
rubenwardy |
3 billion records, but 800 million unique emails |
| 00:14 |
sofar |
All dances were stolen |
| 00:17 |
rubenwardy |
Troy Hunt himself was in that breach, apparently |
| 00:18 |
sofar |
Well it suck my 50 hidden emails were in there |
| 00:19 |
GreenDimond |
so sqlite3 has its own "language" |
| 00:20 |
sofar |
It's almost identical to MySQL and pgsql |
| 00:20 |
GreenDimond |
I meant its not Lua |
| 00:20 |
sofar |
For most simple uses it's the same |
| 00:21 |
sofar |
Sure, that's why there are bindings to lua |
| 00:21 |
rubenwardy |
...lool |
| 00:21 |
rubenwardy |
GreenDimond: you're gonna have to learn other languages some day |
| 00:21 |
rubenwardy |
and SQL is preetttyyy essential |
| 00:21 |
GreenDimond |
Fucking hell rubenwardy shut up |
| 00:21 |
rubenwardy |
:D |
| 00:21 |
rubenwardy |
sorry |
| 00:22 |
sofar |
We can harass, I mean help you with it |
| 00:22 |
rubenwardy |
It was meant to be a light tease, sorry if it came across rude or otherwise |
| 00:23 |
GreenDimond |
It was fine the first time |
| 00:23 |
GreenDimond |
But it wasnt |
| 00:23 |
GreenDimond |
especially considering I do know more than Lua |
| 00:24 |
GreenDimond |
When I ask what language things are in, it is not a "darn, it isnt lua :(" its a "can I use what I already know best or is it something else" |
| 00:24 |
GreenDimond |
Generally when it comes to Minetest, I assume things are Lua because thats the main language used in mods. |
| 00:25 |
rubenwardy |
well, lsqlite3 is a lua binding for sqlite, but you still need to use SQL because it's harder to express the necessary querying in Lua |
| 00:25 |
rubenwardy |
but yeah, ok, sorry |
| 00:27 |
rubenwardy |
there is actually an ORM for SQLite in Lua: http://itdxer.com/4DaysORM/ |
| 00:27 |
rubenwardy |
but it'll be sloowww |
| 00:27 |
GreenDimond |
Yeah that doesnt look as good |
| 00:28 |
GreenDimond |
So downside, users have to install sqlite3 for the mod to work. Upside, it'll be really freakin cool. |
| 00:28 |
GreenDimond |
Also upside, I wont have to scrub a text file backwards. |
| 00:28 |
rubenwardy |
the modding book describes a method to make the sqlite3 dependency option |
| 00:28 |
rubenwardy |
+optional |
| 00:29 |
rubenwardy |
basically by having a backend_something and backend_sqlite file, which both have the same methods but different storage methods |
| 00:29 |
rubenwardy |
however, extra work |
| 00:29 |
GreenDimond |
ill just stick to one |
| 00:34 |
GreenDimond |
different topic, can a coroutine loop be used to not hang minetest? |
| 00:35 |
GreenDimond |
In other words, a normal loop will hold Minetest until it exits, can I fix that with a coroutine |
| 00:38 |
sofar |
Don't think so |
| 01:04 |
|
ANAND joined #minetest-hub |
| 01:04 |
|
red-001 joined #minetest-hub |
| 01:43 |
|
kaeza joined #minetest-hub |
| 02:09 |
|
ANAND joined #minetest-hub |
| 02:13 |
kaeza |
Greetings |
| 02:14 |
ANAND |
Greetings! |
| 02:20 |
|
xerox123_ joined #minetest-hub |
| 02:33 |
|
Ruslan1 joined #minetest-hub |
| 02:36 |
|
ssieb_ joined #minetest-hub |
| 02:37 |
|
sniper33- joined #minetest-hub |
| 02:41 |
|
celeron55_ joined #minetest-hub |
| 02:47 |
|
Calinou joined #minetest-hub |
| 03:20 |
|
garywhite joined #minetest-hub |
| 04:02 |
|
tumeninodes joined #minetest-hub |
| 04:03 |
tumeninodes |
ughhhh. MS already frikken around with the layout of github webpages. They do this just to annoy me... I _know_ it |
| 05:28 |
jas_ |
a couple of people mistype their passwords and the next thing we know we all have to deal with this confirmation dialog. |
| 05:56 |
IcyDiamond |
Oops |
| 05:56 |
jas_ |
hi |
| 05:57 |
jas_ |
paramat: i think the dialog needs to be disabled on android, if you want android users |
| 05:57 |
jas_ |
maybe it's not that bad, i do tend to exagerate |
| 05:57 |
jas_ |
what's good icy? |
| 05:59 |
jas_ |
tumeni: i had a similar frustration when trying to make a new issue, it kept showing me a hundred related issues and seemed to be actively discouraging me from posting a new one in duplicate. stressful, really. |
| 06:01 |
jas_ |
(i'm sorry, i'm sure that's a fine feature) it's just when dialogs pop up and they're like "YOU GOT IT?" or, "OK?" and it's like "i never asked you anything" and you can't NOT click on it BAH i digress |
| 07:52 |
|
CWz joined #minetest-hub |
| 08:19 |
|
celeron55_ joined #minetest-hub |
| 08:57 |
|
twoelk joined #minetest-hub |
| 11:05 |
|
calcul0n joined #minetest-hub |
| 11:09 |
|
Fixer joined #minetest-hub |
| 11:26 |
|
Fixer_ joined #minetest-hub |
| 11:27 |
|
jluc joined #minetest-hub |
| 12:41 |
|
entuland joined #minetest-hub |
| 12:44 |
|
entuland joined #minetest-hub |
| 12:45 |
IcyDiamond |
https://i.lunasqu.ee/Screenshot_19-01-17-14:45:20.png wtf do i do now |
| 12:48 |
|
entuland_ joined #minetest-hub |
| 12:50 |
|
Unarelith joined #minetest-hub |
| 13:12 |
|
ANAND joined #minetest-hub |
| 13:15 |
rubenwardy |
IcyDiamond: do you have any blocks? |
| 13:24 |
|
jluc left #minetest-hub |
| 13:25 |
|
calcul0n joined #minetest-hub |
| 13:25 |
|
behalebabo joined #minetest-hub |
| 13:42 |
|
scr267 joined #minetest-hub |
| 13:44 |
|
_Xenon joined #minetest-hub |
| 13:51 |
|
scr267 joined #minetest-hub |
| 13:56 |
|
scr267 joined #minetest-hub |
| 14:41 |
|
aerozoic joined #minetest-hub |
| 14:43 |
|
cautiouspotato joined #minetest-hub |
| 15:44 |
|
behalebabo joined #minetest-hub |
| 16:22 |
rubenwardy |
ha, MT is everywhere |
| 16:22 |
rubenwardy |
found this: https://github.com/stevedonovan/LDoc/issues/307 |
| 16:47 |
|
Peppy_ joined #minetest-hub |
| 16:55 |
|
Peppy joined #minetest-hub |
| 17:13 |
|
Gael-de-Sailly joined #minetest-hub |
| 17:57 |
|
paramat joined #minetest-hub |
| 17:58 |
|
Darcidride_ joined #minetest-hub |
| 18:14 |
|
Darcidride_ joined #minetest-hub |
| 18:17 |
|
entuland joined #minetest-hub |
| 18:25 |
|
Darcidride_ joined #minetest-hub |
| 18:26 |
|
IhrFussel joined #minetest-hub |
| 18:47 |
rubenwardy |
blooodyy hell http://capitalismgame.rubenwardy.com/docs |
| 18:49 |
paramat |
capitalism game vs. communistcraft |
| 18:50 |
rubenwardy |
that joke is not original at all |
| 18:50 |
rubenwardy |
:P |
| 18:50 |
rubenwardy |
maybe I should rename it to "BusinessTest" |
| 18:51 |
rubenwardy |
that doc software is terrible |
| 18:52 |
rubenwardy |
it can be used to make good looking docs, for example this one: http://lua4z.com/doc/index.html |
| 18:52 |
rubenwardy |
but it's painful to do so |
| 19:06 |
paramat |
Busytest |
| 19:07 |
paramat |
or Busynest |
| 19:18 |
|
garywhite joined #minetest-hub |
| 19:19 |
paramat |
Businest |
| 19:47 |
|
Andrey01 joined #minetest-hub |
| 19:47 |
Andrey01 |
hi |
| 19:49 |
Andrey01 |
question: can i use other names of inventory list differing from 'main' and 'craft'? for example, 'my_list' name. |
| 19:49 |
Andrey01 |
I read rubenwardy`s manual and i have not understood either i can use other names or not. |
| 19:54 |
sofar |
you can make your own |
| 20:04 |
|
jas_ joined #minetest-hub |
| 20:17 |
Andrey01 |
sofar, i tried to create formspec with 'list' element that has 'main' name because Krock said i can not create with other name in this case, then inv:set_list(<new_listname>, {<empty>}) implemented |
| 20:17 |
Andrey01 |
so correctly? |
| 20:18 |
sofar |
if you *make* new inventories, you can use them |
| 20:18 |
Andrey01 |
i get warning: WARNING[Main]: GUIFormSpecMenu::drawList(): The inventory list "kwc1_2" @ "nodemeta:-192,3,-189" doesn't exist |
| 20:18 |
|
entuland_ joined #minetest-hub |
| 20:18 |
sofar |
not sure about node inventories |
| 20:20 |
Andrey01 |
no, when i created formspec string, i wrote same list name that i used in that method |
| 20:56 |
|
paramat joined #minetest-hub |
| 21:09 |
|
Glory7000 joined #minetest-hub |
| 21:18 |
|
scr267 joined #minetest-hub |
| 21:42 |
|
Ruslan1 joined #minetest-hub |
| 22:11 |
|
twoelk left #minetest-hub |
| 22:12 |
|
benrob0329 joined #minetest-hub |
| 23:02 |
|
DI3HARD139 joined #minetest-hub |
| 23:50 |
|
Fixer joined #minetest-hub |
| 23:56 |
|
kaeza joined #minetest-hub |
| 23:57 |
kaeza |
Greetings. |
| 23:57 |
rubenwardy |
p/ |
| 23:58 |
kaeza |
\q |
