Tree Database

[DorkeyDear]

Script Name: Tree-base Database
Script Description Database shaped like a tree
Author: DorkeyDear
Compile Test: Passed
Core Version: 2.6.5
Hosted by: Soldat Central - http://soldatcentral.com/

Full Description:
This is for scripters, not server owners (unless ofc ur a scripter )
Each node can have a value (string), and a list of children (each one a new node). The value must be a string. The root node does have a value.
An example has been included. As of right now, it is not safe to have a node refer to another node, but it is possible.
Note that this uses the binary functions, and is included.

(Size 2.95 KB)
- http://soldatcentral.com/index.php?page=script&f=122 -

** Script hosted by Soldat Central! Please visit the author's script page and Rate this script **

[JFK]

Wow curt, looks like a very efficient script for nested data structures. Hell lot more efficient as my XML functions.
Although I'd advice a little more explanation, it seems pretty ease to use as well.

[DorkeyDear]

I actually made this like 0.5 years ago, but I'm not sure why I had a problem making it. I looked at my old code as reference, and remade it from scratch in like an hour or so.
Note to self: modify binary functions to ability to write/read from a string, not only a file. (so that u can easily put other type of data into the tree in this script)

Explanation:
For those who don't know what a tree is in this case, it is just some sort of element (what I call a node) with sub-elements (sub-nodes, aka the "children" nodes of the "parent" node). Each of these nodes can store one string of information. I'll supply an example for stats:

Code: [Select]

Player Stats
  Soldier
    Kills:Total
      1246
    Kills:Weapon
      1
      3
      58
      542
      39
      41
      10
      17
      416
      0
      81
      28
      1
      9
    Deaths:Total
      1380
  Major
    Kills:Total
      4968
    Kills:Weapon
      120
      195
      301
      485
      198
      894
      1981
      199
      323
      82
      114
      28
      12
      36
    Deaths-Total
      9939
"Player Stats" is the root node (the first node, (should) contain everything)
Each child of this node is a player, value of which being the player's nickname. There are two for examples, Soldier and Major
In each player's node, you will find three example stat categories: Kills:Total, Kills:Weapon, and Deaths:Total. Each supply information as described. Kills:Total has one child, which is an integer saying the total number of kills this dude got. Same thing with Deaths:Total but with deaths. Kills:Weapon has 14 children, to represent the 14 weapons menu weapons (not all weapons, this was just an example so I didn't bother including every one), and the order respectively represents the same weapons as listed in the weapons menu.

Technically, we don't need to even put the nodes "Kills:Total", "Kills:Weapon", and "Deaths:Total" because we know the number of children in each one is a fixed number, and having these would just increase the file size. But having them is nice for clarity to help understand whats going on.

EDIT:
For those who want to know the file format, here it is:
[value : string][child location : ]
[children count : uint16][child location : array<uint16>]
type string is prefixed by uint16 representing the string's length.
type array is prefixed by uint16 representing the array's length (not size, but number of children)

This pattern just repeats throughout the database.
Another way to put it:
( [strlen : uint16] [str : char]* [children count : uint16] [child location : uint16]* )*
whereas * means it repeats.
I'm not sure if there is a standard or whatever to how I should style it; same goes along with string, I'm not sure if it is a standard to null-terminate it, but in case you need the null character inside the string (in case it is another database, possible a tree, or anything), I did not do that.