Permalink
6810 lines (5443 sloc) 258 KB
Minetest Lua Modding API Reference
==================================
* More information at <http://www.minetest.net/>
* Developer Wiki: <http://dev.minetest.net/>
Introduction
============
Content and functionality can be added to Minetest using Lua scripting
in run-time loaded mods.
A mod is a self-contained bunch of scripts, textures and other related
things, which is loaded by and interfaces with Minetest.
Mods are contained and ran solely on the server side. Definitions and media
files are automatically transferred to the client.
If you see a deficiency in the API, feel free to attempt to add the
functionality in the engine and API, and to document it here.
Programming in Lua
------------------
If you have any difficulty in understanding this, please read
[Programming in Lua](http://www.lua.org/pil/).
Startup
-------
Mods are loaded during server startup from the mod load paths by running
the `init.lua` scripts in a shared environment.
Paths
-----
* `RUN_IN_PLACE=1` (Windows release, local build)
* `$path_user`: `<build directory>`
* `$path_share`: `<build directory>`
* `RUN_IN_PLACE=0`: (Linux release)
* `$path_share`:
* Linux: `/usr/share/minetest`
* Windows: `<install directory>/minetest-0.4.x`
* `$path_user`:
* Linux: `$HOME/.minetest`
* Windows: `C:/users/<user>/AppData/minetest` (maybe)
Games
=====
Games are looked up from:
* `$path_share/games/<gameid>/`
* `$path_user/games/<gameid>/`
Where `<gameid>` is unique to each game.
The game directory can contain the following files:
* `game.conf`, with the following keys:
* `name`: Required, human readable name e.g. `name = Minetest`
* `description`: Short description to be shown in the content tab
* `disallowed_mapgens = <comma-separated mapgens>`
e.g. `disallowed_mapgens = v5,v6,flat`
These mapgens are removed from the list of mapgens for the game.
* `minetest.conf`:
Used to set default settings when running this game.
* `settingtypes.txt`:
In the same format as the one in builtin.
This settingtypes.txt will be parsed by the menu and the settings will be
displayed in the "Games" category in the advanced settings tab.
* If the game contains a folder called `textures` the server will load it as a
texturepack, overriding mod textures.
Any server texturepack will override mod textures and the game texturepack.
Menu images
-----------
Games can provide custom main menu images. They are put inside a `menu`
directory inside the game directory.
The images are named `$identifier.png`, where `$identifier` is one of
`overlay`, `background`, `footer`, `header`.
If you want to specify multiple images for one identifier, add additional
images named like `$identifier.$n.png`, with an ascending number $n starting
with 1, and a random image will be chosen from the provided ones.
Mods
====
Mod load path
-------------
Paths are relative to the directories listed in the [Paths] section above.
* `games/<gameid>/mods/`
* `mods/`
* `worlds/<worldname>/worldmods/`
World-specific games
--------------------
It is possible to include a game in a world; in this case, no mods or
games are loaded or checked from anywhere else.
This is useful for e.g. adventure worlds and happens if the `<worldname>/game/`
directory exists.
Mods should then be placed in `<worldname>/game/mods/`.
Modpacks
--------
Mods can be put in a subdirectory, if the parent directory, which otherwise
should be a mod, contains a file named `modpack.txt`. This file shall be
empty, except for lines starting with `#`, which are comments.
Mod directory structure
-----------------------
mods
├── modname
│   ├── mod.conf
│   ├── screenshot.png
│   ├── settingtypes.txt
│   ├── init.lua
│   ├── models
│   ├── textures
│   │   ├── modname_stuff.png
│   │   └── modname_something_else.png
│   ├── sounds
│   ├── media
│   ├── locale
│   └── <custom data>
└── another
### modname
The location of this directory can be fetched by using
`minetest.get_modpath(modname)`.
### mod.conf
A key-value store of mod details.
* `name`: The mod name. Allows Minetest to determine the mod name even if the
folder is wrongly named.
* `description`: Description of mod to be shown in the Mods tab of the main
menu.
* `depends`: A comma separated list of dependencies. These are mods that must be
loaded before this mod.
* `optional_depends`: A comma separated list of optional dependencies.
Like a dependency, but no error if the mod doesn't exist.
Note: to support 0.4.x, please also provide depends.txt.
### `screenshot.png`
A screenshot shown in the mod manager within the main menu. It should
have an aspect ratio of 3:2 and a minimum size of 300×200 pixels.
### `depends.txt`
**Deprecated:** you should use mod.conf instead.
This file is used if there are no dependencies in mod.conf.
List of mods that have to be loaded before loading this mod.
A single line contains a single modname.
Optional dependencies can be defined by appending a question mark
to a single modname. This means that if the specified mod
is missing, it does not prevent this mod from being loaded.
### `description.txt`
**Deprecated:** you should use mod.conf instead.
This file is used if there is no description in mod.conf.
A file containing a description to be shown in the Mods tab of the main menu.
### `settingtypes.txt`
A file in the same format as the one in builtin. It will be parsed by the
settings menu and the settings will be displayed in the "Mods" category.
### `init.lua`
The main Lua script. Running this script should register everything it
wants to register. Subsequent execution depends on minetest calling the
registered callbacks.
`minetest.settings` can be used to read custom or existing settings at load
time, if necessary. (See [`Settings`])
### `models`
Models for entities or meshnodes.
### `textures`, `sounds`, `media`
Media files (textures, sounds, whatever) that will be transferred to the
client and will be available for use by the mod.
### `locale`
Translation files for the clients. (See [Translations])
Naming conventions
------------------
Registered names should generally be in this format:
modname:<whatever>
`<whatever>` can have these characters:
a-zA-Z0-9_
This is to prevent conflicting names from corrupting maps and is
enforced by the mod loader.
Registered names can be overridden by prefixing the name with `:`. This can
be used for overriding the registrations of some other mod.
The `:` prefix can also be used for maintaining backwards compatibility.
### Example
In the mod `experimental`, there is the ideal item/node/entity name `tnt`.
So the name should be `experimental:tnt`.
Any mod can redefine `experimental:tnt` by using the name
:experimental:tnt
when registering it. That mod is required to have `experimental` as a
dependency.
Aliases
=======
Aliases can be added by using `minetest.register_alias(name, convert_to)` or
`minetest.register_alias_force(name, convert_to)`.
This converts anything called `name` to `convert_to`.
The only difference between `minetest.register_alias` and
`minetest.register_alias_force` is that if an item called `name` exists,
`minetest.register_alias` will do nothing while
`minetest.register_alias_force` will unregister it.
This can be used for maintaining backwards compatibility.
This can also set quick access names for things, e.g. if
you have an item called `epiclylongmodname:stuff`, you could do
minetest.register_alias("stuff", "epiclylongmodname:stuff")
and be able to use `/giveme stuff`.
Mapgen aliases
--------------
In a game, a certain number of these must be set to tell core mapgens which
of the game's nodes are to be used by the core mapgens. For example:
minetest.register_alias("mapgen_stone", "default:stone")
### Aliases needed for all mapgens except Mapgen V6
#### Base terrain
* mapgen_stone
* mapgen_water_source
* mapgen_river_water_source
#### Caves
Not required if cave liquid nodes are set in biome definitions.
* mapgen_lava_source
#### Dungeons
Not required if dungeon nodes are set in biome definitions.
* mapgen_cobble
* mapgen_stair_cobble
* mapgen_mossycobble
* mapgen_desert_stone
* mapgen_stair_desert_stone
* mapgen_sandstone
* mapgen_sandstonebrick
* mapgen_stair_sandstone_block
### Aliases needed for Mapgen V6
#### Terrain and biomes
* mapgen_stone
* mapgen_water_source
* mapgen_lava_source
* mapgen_dirt
* mapgen_dirt_with_grass
* mapgen_sand
* mapgen_gravel
* mapgen_desert_stone
* mapgen_desert_sand
* mapgen_dirt_with_snow
* mapgen_snowblock
* mapgen_snow
* mapgen_ice
#### Flora
* mapgen_tree
* mapgen_leaves
* mapgen_apple
* mapgen_jungletree
* mapgen_jungleleaves
* mapgen_junglegrass
* mapgen_pine_tree
* mapgen_pine_needles
#### Dungeons
* mapgen_cobble
* mapgen_stair_cobble
* mapgen_mossycobble
* mapgen_stair_desert_stone
### Setting the node used in Mapgen Singlenode
By default the world is filled with air nodes. To set a different node use, for
example:
minetest.register_alias("mapgen_singlenode", "default:stone")
Textures
========
Mods should generally prefix their textures with `modname_`, e.g. given
the mod name `foomod`, a texture could be called:
foomod_foothing.png
Textures are referred to by their complete name, or alternatively by
stripping out the file extension:
* e.g. `foomod_foothing.png`
* e.g. `foomod_foothing`
Texture modifiers
-----------------
There are various texture modifiers that can be used
to generate textures on-the-fly.
### Texture overlaying
Textures can be overlaid by putting a `^` between them.
Example:
default_dirt.png^default_grass_side.png
`default_grass_side.png` is overlaid over `default_dirt.png`.
The texture with the lower resolution will be automatically upscaled to
the higher resolution texture.
### Texture grouping
Textures can be grouped together by enclosing them in `(` and `)`.
Example: `cobble.png^(thing1.png^thing2.png)`
A texture for `thing1.png^thing2.png` is created and the resulting
texture is overlaid on top of `cobble.png`.
### Escaping
Modifiers that accept texture names (e.g. `[combine`) accept escaping to allow
passing complex texture names as arguments. Escaping is done with backslash and
is required for `^` and `:`.
Example: `cobble.png^[lowpart:50:color.png\^[mask\:trans.png`
The lower 50 percent of `color.png^[mask:trans.png` are overlaid
on top of `cobble.png`.
### Advanced texture modifiers
#### Crack
* `[crack:<n>:<p>`
* `[cracko:<n>:<p>`
* `[crack:<t>:<n>:<p>`
* `[cracko:<t>:<n>:<p>`
Parameters:
* `<t>`: tile count (in each direction)
* `<n>`: animation frame count
* `<p>`: current animation frame
Draw a step of the crack animation on the texture.
`crack` draws it normally, while `cracko` lays it over, keeping transparent
pixels intact.
Example:
default_cobble.png^[crack:10:1
#### `[combine:<w>x<h>:<x1>,<y1>=<file1>:<x2>,<y2>=<file2>:...`
* `<w>`: width
* `<h>`: height
* `<x>`: x position
* `<y>`: y position
* `<file>`: texture to combine
Creates a texture of size `<w>` times `<h>` and blits the listed files to their
specified coordinates.
Example:
[combine:16x32:0,0=default_cobble.png:0,16=default_wood.png
#### `[resize:<w>x<h>`
Resizes the texture to the given dimensions.
Example:
default_sandstone.png^[resize:16x16
#### `[opacity:<r>`
Makes the base image transparent according to the given ratio.
`r` must be between 0 (transparent) and 255 (opaque).
Example:
default_sandstone.png^[opacity:127
#### `[invert:<mode>`
Inverts the given channels of the base image.
Mode may contain the characters "r", "g", "b", "a".
Only the channels that are mentioned in the mode string will be inverted.
Example:
default_apple.png^[invert:rgb
#### `[brighten`
Brightens the texture.
Example:
tnt_tnt_side.png^[brighten
#### `[noalpha`
Makes the texture completely opaque.
Example:
default_leaves.png^[noalpha
#### `[makealpha:<r>,<g>,<b>`
Convert one color to transparency.
Example:
default_cobble.png^[makealpha:128,128,128
#### `[transform<t>`
* `<t>`: transformation(s) to apply
Rotates and/or flips the image.
`<t>` can be a number (between 0 and 7) or a transform name.
Rotations are counter-clockwise.
0 I identity
1 R90 rotate by 90 degrees
2 R180 rotate by 180 degrees
3 R270 rotate by 270 degrees
4 FX flip X
5 FXR90 flip X then rotate by 90 degrees
6 FY flip Y
7 FYR90 flip Y then rotate by 90 degrees
Example:
default_stone.png^[transformFXR90
#### `[inventorycube{<top>{<left>{<right>`
Escaping does not apply here and `^` is replaced by `&` in texture names
instead.
Create an inventory cube texture using the side textures.
Example:
[inventorycube{grass.png{dirt.png&grass_side.png{dirt.png&grass_side.png
Creates an inventorycube with `grass.png`, `dirt.png^grass_side.png` and
`dirt.png^grass_side.png` textures
#### `[lowpart:<percent>:<file>`
Blit the lower `<percent>`% part of `<file>` on the texture.
Example:
base.png^[lowpart:25:overlay.png
#### `[verticalframe:<t>:<n>`
* `<t>`: animation frame count
* `<n>`: current animation frame
Crops the texture to a frame of a vertical animation.
Example:
default_torch_animated.png^[verticalframe:16:8
#### `[mask:<file>`
Apply a mask to the base image.
The mask is applied using binary AND.
#### `[sheet:<w>x<h>:<x>,<y>`
Retrieves a tile at position x,y from the base image
which it assumes to be a tilesheet with dimensions w,h.
#### `[colorize:<color>:<ratio>`
Colorize the textures with the given color.
`<color>` is specified as a `ColorString`.
`<ratio>` is an int ranging from 0 to 255 or the word "`alpha`". If
it is an int, then it specifies how far to interpolate between the
colors where 0 is only the texture color and 255 is only `<color>`. If
omitted, the alpha of `<color>` will be used as the ratio. If it is
the word "`alpha`", then each texture pixel will contain the RGB of
`<color>` and the alpha of `<color>` multiplied by the alpha of the
texture pixel.
#### `[multiply:<color>`
Multiplies texture colors with the given color.
`<color>` is specified as a `ColorString`.
Result is more like what you'd expect if you put a color on top of another
color, meaning white surfaces get a lot of your new color while black parts
don't change very much.
Hardware coloring
-----------------
The goal of hardware coloring is to simplify the creation of
colorful nodes. If your textures use the same pattern, and they only
differ in their color (like colored wool blocks), you can use hardware
coloring instead of creating and managing many texture files.
All of these methods use color multiplication (so a white-black texture
with red coloring will result in red-black color).
### Static coloring
This method is useful if you wish to create nodes/items with
the same texture, in different colors, each in a new node/item definition.
#### Global color
When you register an item or node, set its `color` field (which accepts a
`ColorSpec`) to the desired color.
An `ItemStack`'s static color can be overwritten by the `color` metadata
field. If you set that field to a `ColorString`, that color will be used.
#### Tile color
Each tile may have an individual static color, which overwrites every
other coloring method. To disable the coloring of a face,
set its color to white (because multiplying with white does nothing).
You can set the `color` property of the tiles in the node's definition
if the tile is in table format.
### Palettes
For nodes and items which can have many colors, a palette is more
suitable. A palette is a texture, which can contain up to 256 pixels.
Each pixel is one possible color for the node/item.
You can register one node/item, which can have up to 256 colors.
#### Palette indexing
When using palettes, you always provide a pixel index for the given
node or `ItemStack`. The palette is read from left to right and from
top to bottom. If the palette has less than 256 pixels, then it is
stretched to contain exactly 256 pixels (after arranging the pixels
to one line). The indexing starts from 0.
Examples:
* 16x16 palette, index = 0: the top left corner
* 16x16 palette, index = 4: the fifth pixel in the first row
* 16x16 palette, index = 16: the pixel below the top left corner
* 16x16 palette, index = 255: the bottom right corner
* 2 (width) x 4 (height) palette, index = 31: the top left corner.
The palette has 8 pixels, so each pixel is stretched to 32 pixels,
to ensure the total 256 pixels.
* 2x4 palette, index = 32: the top right corner
* 2x4 palette, index = 63: the top right corner
* 2x4 palette, index = 64: the pixel below the top left corner
#### Using palettes with items
When registering an item, set the item definition's `palette` field to
a texture. You can also use texture modifiers.
The `ItemStack`'s color depends on the `palette_index` field of the
stack's metadata. `palette_index` is an integer, which specifies the
index of the pixel to use.
#### Linking palettes with nodes
When registering a node, set the item definition's `palette` field to
a texture. You can also use texture modifiers.
The node's color depends on its `param2`, so you also must set an
appropriate `paramtype2`:
* `paramtype2 = "color"` for nodes which use their full `param2` for
palette indexing. These nodes can have 256 different colors.
The palette should contain 256 pixels.
* `paramtype2 = "colorwallmounted"` for nodes which use the first
five bits (most significant) of `param2` for palette indexing.
The remaining three bits are describing rotation, as in `wallmounted`
paramtype2. Division by 8 yields the palette index (without stretching the
palette). These nodes can have 32 different colors, and the palette
should contain 32 pixels.
Examples:
* `param2 = 17` is 2 * 8 + 1, so the rotation is 1 and the third (= 2 + 1)
pixel will be picked from the palette.
* `param2 = 35` is 4 * 8 + 3, so the rotation is 3 and the fifth (= 4 + 1)
pixel will be picked from the palette.
* `paramtype2 = "colorfacedir"` for nodes which use the first
three bits of `param2` for palette indexing. The remaining
five bits are describing rotation, as in `facedir` paramtype2.
Division by 32 yields the palette index (without stretching the
palette). These nodes can have 8 different colors, and the
palette should contain 8 pixels.
Examples:
* `param2 = 17` is 0 * 32 + 17, so the rotation is 17 and the
first (= 0 + 1) pixel will be picked from the palette.
* `param2 = 35` is 1 * 32 + 3, so the rotation is 3 and the
second (= 1 + 1) pixel will be picked from the palette.
To colorize a node on the map, set its `param2` value (according
to the node's paramtype2).
### Conversion between nodes in the inventory and on the map
Static coloring is the same for both cases, there is no need
for conversion.
If the `ItemStack`'s metadata contains the `color` field, it will be
lost on placement, because nodes on the map can only use palettes.
If the `ItemStack`'s metadata contains the `palette_index` field, it is
automatically transferred between node and item forms by the engine,
when a player digs or places a colored node.
You can disable this feature by setting the `drop` field of the node
to itself (without metadata).
To transfer the color to a special drop, you need a drop table.
Example:
minetest.register_node("mod:stone", {
description = "Stone",
tiles = {"default_stone.png"},
paramtype2 = "color",
palette = "palette.png",
drop = {
items = {
-- assume that mod:cobblestone also has the same palette
{items = {"mod:cobblestone"}, inherit_color = true },
}
}
})
### Colored items in craft recipes
Craft recipes only support item strings, but fortunately item strings
can also contain metadata. Example craft recipe registration:
minetest.register_craft({
output = minetest.itemstring_with_palette("wool:block", 3),
type = "shapeless",
recipe = {
"wool:block",
"dye:red",
},
})
To set the `color` field, you can use `minetest.itemstring_with_color`.
Metadata field filtering in the `recipe` field are not supported yet,
so the craft output is independent of the color of the ingredients.
Soft texture overlay
--------------------
Sometimes hardware coloring is not enough, because it affects the
whole tile. Soft texture overlays were added to Minetest to allow
the dynamic coloring of only specific parts of the node's texture.
For example a grass block may have colored grass, while keeping the
dirt brown.
These overlays are 'soft', because unlike texture modifiers, the layers
are not merged in the memory, but they are simply drawn on top of each
other. This allows different hardware coloring, but also means that
tiles with overlays are drawn slower. Using too much overlays might
cause FPS loss.
For inventory and wield images you can specify overlays which
hardware coloring does not modify. You have to set `inventory_overlay`
and `wield_overlay` fields to an image name.
To define a node overlay, simply set the `overlay_tiles` field of the node
definition. These tiles are defined in the same way as plain tiles:
they can have a texture name, color etc.
To skip one face, set that overlay tile to an empty string.
Example (colored grass block):
minetest.register_node("default:dirt_with_grass", {
description = "Dirt with Grass",
-- Regular tiles, as usual
-- The dirt tile disables palette coloring
tiles = {{name = "default_grass.png"},
{name = "default_dirt.png", color = "white"}},
-- Overlay tiles: define them in the same style
-- The top and bottom tile does not have overlay
overlay_tiles = {"", "",
{name = "default_grass_side.png", tileable_vertical = false}},
-- Global color, used in inventory
color = "green",
-- Palette in the world
paramtype2 = "color",
palette = "default_foilage.png",
})
Sounds
======
Only Ogg Vorbis files are supported.
For positional playing of sounds, only single-channel (mono) files are
supported. Otherwise OpenAL will play them non-positionally.
Mods should generally prefix their sounds with `modname_`, e.g. given
the mod name "`foomod`", a sound could be called:
foomod_foosound.ogg
Sounds are referred to by their name with a dot, a single digit and the
file extension stripped out. When a sound is played, the actual sound file
is chosen randomly from the matching sounds.
When playing the sound `foomod_foosound`, the sound is chosen randomly
from the available ones of the following files:
* `foomod_foosound.ogg`
* `foomod_foosound.0.ogg`
* `foomod_foosound.1.ogg`
* (...)
* `foomod_foosound.9.ogg`
Examples of sound parameter tables:
-- Play locationless on all clients
{
gain = 1.0, -- default
fade = 0.0, -- default, change to a value > 0 to fade the sound in
pitch = 1.0, -- default
}
-- Play locationless to one player
{
to_player = name,
gain = 1.0, -- default
fade = 0.0, -- default, change to a value > 0 to fade the sound in
pitch = 1.0, -- default
}
-- Play locationless to one player, looped
{
to_player = name,
gain = 1.0, -- default
loop = true,
}
-- Play in a location
{
pos = {x = 1, y = 2, z = 3},
gain = 1.0, -- default
max_hear_distance = 32, -- default, uses an euclidean metric
}
-- Play connected to an object, looped
{
object = <an ObjectRef>,
gain = 1.0, -- default
max_hear_distance = 32, -- default, uses an euclidean metric
loop = true,
}
Looped sounds must either be connected to an object or played locationless to
one player using `to_player = name,`
`SimpleSoundSpec`
-----------------
* e.g. `""`
* e.g. `"default_place_node"`
* e.g. `{}`
* e.g. `{name = "default_place_node"}`
* e.g. `{name = "default_place_node", gain = 1.0}`
* e.g. `{name = "default_place_node", gain = 1.0, pitch = 1.0}`
Registered definitions
======================
Anything added using certain `minetest.register_*` functions gets added to
the global `minetest.registered_*` tables.
* `minetest.register_entity(name, entity definition)`
* added to `minetest.registered_entities[name]`
* `minetest.register_node(name, node definition)`
* added to `minetest.registered_items[name]`
* added to `minetest.registered_nodes[name]`
* `minetest.register_tool(name, item definition)`
* added to `minetest.registered_items[name]`
* `minetest.register_craftitem(name, item definition)`
* added to `minetest.registered_items[name]`
* `minetest.unregister_item(name)`
* Unregisters the item name from engine, and deletes the entry with key
`name` from `minetest.registered_items` and from the associated item
table according to its nature: `minetest.registered_nodes[]` etc
* `minetest.register_biome(biome definition)`
* returns an integer uniquely identifying the registered biome
* added to `minetest.registered_biome` with the key of `biome.name`
* if `biome.name` is nil, the key is the returned ID
* `minetest.unregister_biome(name)`
* Unregisters the biome name from engine, and deletes the entry with key
`name` from `minetest.registered_biome`
* `minetest.register_ore(ore definition)`
* returns an integer uniquely identifying the registered ore
* added to `minetest.registered_ores` with the key of `ore.name`
* if `ore.name` is nil, the key is the returned ID
* `minetest.register_decoration(decoration definition)`
* returns an integer uniquely identifying the registered decoration
* added to `minetest.registered_decorations` with the key of
`decoration.name`.
* if `decoration.name` is nil, the key is the returned ID
* `minetest.register_schematic(schematic definition)`
* returns an integer uniquely identifying the registered schematic
* added to `minetest.registered_schematic` with the key of `schematic.name`
* if `schematic.name` is nil, the key is the returned ID
* if the schematic is loaded from a file, schematic.name is set to the
filename.
* if the function is called when loading the mod, and schematic.name is a
relative path, then the current mod path will be prepended to the
schematic filename.
* `minetest.clear_registered_biomes()`
* clears all biomes currently registered
* `minetest.clear_registered_ores()`
* clears all ores currently registered
* `minetest.clear_registered_decorations()`
* clears all decorations currently registered
* `minetest.clear_registered_schematics()`
* clears all schematics currently registered
Note that in some cases you will stumble upon things that are not contained
in these tables (e.g. when a mod has been removed). Always check for
existence before trying to access the fields.
Example: If you want to check the drawtype of a node, you could do:
local function get_nodedef_field(nodename, fieldname)
if not minetest.registered_nodes[nodename] then
return nil
end
return minetest.registered_nodes[nodename][fieldname]
end
local drawtype = get_nodedef_field(nodename, "drawtype")
Nodes
=====
Nodes are the bulk data of the world: cubes and other things that take the
space of a cube. Huge amounts of them are handled efficiently, but they
are quite static.
The definition of a node is stored and can be accessed by using
minetest.registered_nodes[node.name]
See [Registered definitions].
Nodes are passed by value between Lua and the engine.
They are represented by a table:
{name="name", param1=num, param2=num}
`param1` and `param2` are 8-bit integers ranging from 0 to 255. The engine uses
them for certain automated functions. If you don't use these functions, you can
use them to store arbitrary values.
Node paramtypes
---------------
The functions of `param1` and `param2` are determined by certain fields in the
node definition.
`param1` is reserved for the engine when `paramtype != "none"`:
* `paramtype = "light"`
* The value stores light with and without sun in its upper and lower 4 bits
respectively.
* Required by a light source node to enable spreading its light.
* Required by the following drawtypes as they determine their visual
brightness from their internal light value:
* torchlike
* signlike
* firelike
* fencelike
* raillike
* nodebox
* mesh
* plantlike
* plantlike_rooted
`param2` is reserved for the engine when any of these are used:
* `liquidtype = "flowing"`
* The level and some flags of the liquid is stored in `param2`
* `drawtype = "flowingliquid"`
* The drawn liquid level is read from `param2`
* `drawtype = "torchlike"`
* `drawtype = "signlike"`
* `paramtype2 = "wallmounted"`
* The rotation of the node is stored in `param2`. You can make this value
by using `minetest.dir_to_wallmounted()`.
* `paramtype2 = "facedir"`
* The rotation of the node is stored in `param2`. Furnaces and chests are
rotated this way. Can be made by using `minetest.dir_to_facedir()`.
* Values range 0 - 23
* facedir / 4 = axis direction:
0 = y+, 1 = z+, 2 = z-, 3 = x+, 4 = x-, 5 = y-
* facedir modulo 4 = rotation around that axis
* `paramtype2 = "leveled"`
* Only valid for "nodebox" with 'type = "leveled"', and "plantlike_rooted".
* Leveled nodebox:
* The level of the top face of the nodebox is stored in `param2`.
* The other faces are defined by 'fixed = {}' like 'type = "fixed"'
nodeboxes.
* The nodebox height is (`param2` / 64) nodes.
* The maximum accepted value of `param2` is 127.
* Rooted plantlike:
* The height of the 'plantlike' section is stored in `param2`.
* The height is (`param2` / 16) nodes.
* `paramtype2 = "degrotate"`
* Only valid for "plantlike". The rotation of the node is stored in
`param2`.
* Values range 0 - 179. The value stored in `param2` is multiplied by two to
get the actual rotation in degrees of the node.
* `paramtype2 = "meshoptions"`
* Only valid for "plantlike". The value of `param2` becomes a bitfield which
can be used to change how the client draws plantlike nodes.
* Bits 0, 1 and 2 form a mesh selector.
Currently the following meshes are choosable:
* 0 = a "x" shaped plant (ordinary plant)
* 1 = a "+" shaped plant (just rotated 45 degrees)
* 2 = a "*" shaped plant with 3 faces instead of 2
* 3 = a "#" shaped plant with 4 faces instead of 2
* 4 = a "#" shaped plant with 4 faces that lean outwards
* 5-7 are unused and reserved for future meshes.
* Bits 3 through 7 are optional flags that can be combined and give these
effects:
* bit 3 (0x08) - Makes the plant slightly vary placement horizontally
* bit 4 (0x10) - Makes the plant mesh 1.4x larger
* bit 5 (0x20) - Moves each face randomly a small bit down (1/8 max)
* bits 6-7 are reserved for future use.
* `paramtype2 = "color"`
* `param2` tells which color is picked from the palette.
The palette should have 256 pixels.
* `paramtype2 = "colorfacedir"`
* Same as `facedir`, but with colors.
* The first three bits of `param2` tells which color is picked from the
palette. The palette should have 8 pixels.
* `paramtype2 = "colorwallmounted"`
* Same as `wallmounted`, but with colors.
* The first five bits of `param2` tells which color is picked from the
palette. The palette should have 32 pixels.
* `paramtype2 = "glasslikeliquidlevel"`
* Only valid for "glasslike_framed" or "glasslike_framed_optional"
drawtypes.
* `param2` values 0-63 define 64 levels of internal liquid, 0 being empty
and 63 being full.
* Liquid texture is defined using `special_tiles = {"modname_tilename.png"}`
Nodes can also contain extra data. See [Node Metadata].
Node drawtypes
--------------
There are a bunch of different looking node types.
Look for examples in `games/minimal` or `games/minetest_game`.
* `normal`
* A node-sized cube.
* `airlike`
* Invisible, uses no texture.
* `liquid`
* The cubic source node for a liquid.
* `flowingliquid`
* The flowing version of a liquid, appears with various heights and slopes.
* `glasslike`
* Often used for partially-transparent nodes.
* Only external sides of textures are visible.
* `glasslike_framed`
* All face-connected nodes are drawn as one volume within a surrounding
frame.
* The frame appearence is generated from the edges of the first texture
specified in `tiles`. The width of the edges used are 1/16th of texture
size: 1 pixel for 16x16, 2 pixels for 32x32 etc.
* The glass 'shine' (or other desired detail) on each node face is supplied
by the second texture specified in `tiles`.
* `glasslike_framed_optional`
* This switches between the above 2 drawtypes according to the menu setting
'Connected Glass'.
* `allfaces`
* Often used for partially-transparent nodes.
* External and internal sides of textures are visible.
* `allfaces_optional`
* Often used for leaves nodes.
* This switches between `normal`, `glasslike` and `allfaces` according to
the menu setting: Opaque Leaves / Simple Leaves / Fancy Leaves.
* With 'Simple Leaves' selected, the texture specified in `special_tiles`
is used instead, if present. This allows a visually thicker texture to be
used to compensate for how `glasslike` reduces visual thickness.
* `torchlike`
* A single vertical texture.
* If placed on top of a node, uses the first texture specified in `tiles`.
* If placed against the underside of a node, uses the second texture
specified in `tiles`.
* If placed on the side of a node, uses the third texture specified in
`tiles` and is perpendicular to that node.
* `signlike`
* A single texture parallel to, and mounted against, the top, underside or
side of a node.
* `plantlike`
* Two vertical and diagonal textures at right-angles to each other.
* See `paramtype2 = "meshoptions"` above for other options.
* `firelike`
* When above a flat surface, appears as 6 textures, the central 2 as
`plantlike` plus 4 more surrounding those.
* If not above a surface the central 2 do not appear, but the texture
appears against the faces of surrounding nodes if they are present.
* `fencelike`
* A 3D model suitable for a wooden fence.
* One placed node appears as a single vertical post.
* Adjacently-placed nodes cause horizontal bars to appear between them.
* `raillike`
* Often used for tracks for mining carts.
* Requires 4 textures to be specified in `tiles`, in order: Straight,
curved, t-junction, crossing.
* Each placed node automatically switches to a suitable rotated texture
determined by the adjacent `raillike` nodes, in order to create a
continuous track network.
* Becomes a sloping node if placed against stepped nodes.
* `nodebox`
* Often used for stairs and slabs.
* Allows defining nodes consisting of an arbitrary number of boxes.
* See [Node boxes] below for more information.
* `mesh`
* Uses models for nodes.
* Tiles should hold model materials textures.
* Only static meshes are implemented.
* For supported model formats see Irrlicht engine documentation.
* `plantlike_rooted`
* Enables underwater `plantlike` without air bubbles around the nodes.
* Consists of a base cube at the co-ordinates of the node plus a
`plantlike` extension above with a height of `param2 / 16` nodes.
* The `plantlike` extension visually passes through any nodes above the
base cube without affecting them.
* The base cube texture tiles are defined as normal, the `plantlike`
extension uses the defined special tile, for example:
`special_tiles = {{name = "default_papyrus.png", tileable_vertical = true}},`
`*_optional` drawtypes need less rendering time if deactivated
(always client-side).
Node boxes
----------
Node selection boxes are defined using "node boxes".
A nodebox is defined as any of:
{
-- A normal cube; the default in most things
type = "regular"
}
{
-- A fixed box (or boxes) (facedir param2 is used, if applicable)
type = "fixed",
fixed = box OR {box1, box2, ...}
}
{
-- A variable height box (or boxes) with the top face position defined
-- by the node parameter 'leveled = ', or if 'paramtype2 == "leveled"'
-- by param2.
-- Other faces are defined by 'fixed = {}' as with 'type = "fixed"'.
type = "leveled",
fixed = box OR {box1, box2, ...}
}
{
-- A box like the selection box for torches
-- (wallmounted param2 is used, if applicable)
type = "wallmounted",
wall_top = box,
wall_bottom = box,
wall_side = box
}
{
-- A node that has optional boxes depending on neighbouring nodes'
-- presence and type. See also `connects_to`.
type = "connected",
fixed = box OR {box1, box2, ...}
connect_top = box OR {box1, box2, ...}
connect_bottom = box OR {box1, box2, ...}
connect_front = box OR {box1, box2, ...}
connect_left = box OR {box1, box2, ...}
connect_back = box OR {box1, box2, ...}
connect_right = box OR {box1, box2, ...}
-- The following `disconnected_*` boxes are the opposites of the
-- `connect_*` ones above, i.e. when a node has no suitable neighbour
-- on the respective side, the corresponding disconnected box is drawn.
disconnected_top = box OR {box1, box2, ...}
disconnected_bottom = box OR {box1, box2, ...}
disconnected_front = box OR {box1, box2, ...}
disconnected_left = box OR {box1, box2, ...}
disconnected_back = box OR {box1, box2, ...}
disconnected_right = box OR {box1, box2, ...}
disconnected = box OR {box1, box2, ...} -- when there is *no* neighbour
disconnected_sides = box OR {box1, box2, ...} -- when there are *no*
-- neighbours to the sides
}
A `box` is defined as:
{x1, y1, z1, x2, y2, z2}
A box of a regular node would look like:
{-0.5, -0.5, -0.5, 0.5, 0.5, 0.5},
HUD
===
HUD element types
-----------------
The position field is used for all element types.
To account for differing resolutions, the position coordinates are the
percentage of the screen, ranging in value from `0` to `1`.
The name field is not yet used, but should contain a description of what the
HUD element represents. The direction field is the direction in which something
is drawn.
`0` draws from left to right, `1` draws from right to left, `2` draws from
top to bottom, and `3` draws from bottom to top.
The `alignment` field specifies how the item will be aligned. It ranges from
`-1` to `1`, with `0` being the center. `-1` is moved to the left/up, and `1`
is to the right/down. Fractional values can be used.
The `offset` field specifies a pixel offset from the position. Contrary to
position, the offset is not scaled to screen size. This allows for some
precisely positioned items in the HUD.
**Note**: `offset` _will_ adapt to screen DPI as well as user defined scaling
factor!
Below are the specific uses for fields in each type; fields not listed for that
type are ignored.
### `image`
Displays an image on the HUD.
* `scale`: The scale of the image, with 1 being the original texture size.
Only the X coordinate scale is used (positive values).
Negative values represent that percentage of the screen it
should take; e.g. `x=-100` means 100% (width).
* `text`: The name of the texture that is displayed.
* `alignment`: The alignment of the image.
* `offset`: offset in pixels from position.
### `text`
Displays text on the HUD.
* `scale`: Defines the bounding rectangle of the text.
A value such as `{x=100, y=100}` should work.
* `text`: The text to be displayed in the HUD element.
* `number`: An integer containing the RGB value of the color used to draw the
text. Specify `0xFFFFFF` for white text, `0xFF0000` for red, and so on.
* `alignment`: The alignment of the text.
* `offset`: offset in pixels from position.
### `statbar`
Displays a horizontal bar made up of half-images.
* `text`: The name of the texture that is used.
* `number`: The number of half-textures that are displayed.
If odd, will end with a vertically center-split texture.
* `direction`
* `offset`: offset in pixels from position.
* `size`: If used, will force full-image size to this value (override texture
pack image size)
### `inventory`
* `text`: The name of the inventory list to be displayed.
* `number`: Number of items in the inventory to be displayed.
* `item`: Position of item that is selected.
* `direction`
* `offset`: offset in pixels from position.
### `waypoint`
Displays distance to selected world position.
* `name`: The name of the waypoint.
* `text`: Distance suffix. Can be blank.
* `number:` An integer containing the RGB value of the color used to draw the
text.
* `world_pos`: World position of the waypoint.
Representations of simple things
================================
Position/vector
---------------
{x=num, y=num, z=num}
For helper functions see [Spatial Vectors].
`pointed_thing`
---------------
* `{type="nothing"}`
* `{type="node", under=pos, above=pos}`
* `{type="object", ref=ObjectRef}`
Exact pointing location (currently only `Raycast` supports these fields):
* `pointed_thing.intersection_point`: The absolute world coordinates of the
point on the selection box which is pointed at. May be in the selection box
if the pointer is in the box too.
* `pointed_thing.box_id`: The ID of the pointed selection box (counting starts
from 1).
* `pointed_thing.intersection_normal`: Unit vector, points outwards of the
selected selection box. This specifies which face is pointed at.
Is a null vector `{x = 0, y = 0, z = 0}` when the pointer is inside the
selection box.
Flag Specifier Format
=====================
Flags using the standardized flag specifier format can be specified in either
of two ways, by string or table.
The string format is a comma-delimited set of flag names; whitespace and
unrecognized flag fields are ignored. Specifying a flag in the string sets the
flag, and specifying a flag prefixed by the string `"no"` explicitly
clears the flag from whatever the default may be.
In addition to the standard string flag format, the schematic flags field can
also be a table of flag names to boolean values representing whether or not the
flag is set. Additionally, if a field with the flag name prefixed with `"no"`
is present, mapped to a boolean of any value, the specified flag is unset.
E.g. A flag field of value
{place_center_x = true, place_center_y=false, place_center_z=true}
is equivalent to
{place_center_x = true, noplace_center_y=true, place_center_z=true}
which is equivalent to
"place_center_x, noplace_center_y, place_center_z"
or even
"place_center_x, place_center_z"
since, by default, no schematic attributes are set.
Items
=====
Item types
----------
There are three kinds of items: nodes, tools and craftitems.
* Node: Can be placed in the world's voxel grid
* Tool: Has a wear property but cannot be stacked. The default use action is to
dig nodes or hit objects according to its tool capabilities.
* Craftitem: Cannot dig nodes or be placed
Amount and wear
---------------
All item stacks have an amount between 0 and 65535. It is 1 by
default. Tool item stacks can not have an amount greater than 1.
Tools use a wear (damage) value ranging from 0 to 65535. The
value 0 is the default and is used for unworn tools. The values
1 to 65535 are used for worn tools, where a higher value stands for
a higher wear. Non-tools always have a wear value of 0.
Item formats
------------
Items and item stacks can exist in three formats: Serializes, table format
and `ItemStack`.
When an item must be passed to a function, it can usually be in any of
these formats.
### Serialized
This is called "stackstring" or "itemstring". It is a simple string with
1-3 components: the full item identifier, an optional amount and an optional
wear value. Syntax:
<identifier> [<amount>[ <wear>]]
Examples:
* `'default:apple'`: 1 apple
* `'default:dirt 5'`: 5 dirt
* `'default:pick_stone'`: a new stone pickaxe
* `'default:pick_wood 1 21323'`: a wooden pickaxe, ca. 1/3 worn out
### Table format
Examples:
5 dirt nodes:
{name="default:dirt", count=5, wear=0, metadata=""}
A wooden pick about 1/3 worn out:
{name="default:pick_wood", count=1, wear=21323, metadata=""}
An apple:
{name="default:apple", count=1, wear=0, metadata=""}
### `ItemStack`
A native C++ format with many helper methods. Useful for converting
between formats. See the [Class reference] section for details.
Groups
======
In a number of places, there is a group table. Groups define the
properties of a thing (item, node, armor of entity, capabilities of
tool) in such a way that the engine and other mods can can interact with
the thing without actually knowing what the thing is.
Usage
-----
Groups are stored in a table, having the group names with keys and the
group ratings as values. For example:
-- Default dirt
groups = {crumbly=3, soil=1}
-- A more special dirt-kind of thing
groups = {crumbly=2, soil=1, level=2, outerspace=1}
Groups always have a rating associated with them. If there is no
useful meaning for a rating for an enabled group, it shall be `1`.
When not defined, the rating of a group defaults to `0`. Thus when you
read groups, you must interpret `nil` and `0` as the same value, `0`.
You can read the rating of a group for an item or a node by using
minetest.get_item_group(itemname, groupname)
Groups of items
---------------
Groups of items can define what kind of an item it is (e.g. wool).
Groups of nodes
---------------
In addition to the general item things, groups are used to define whether
a node is destroyable and how long it takes to destroy by a tool.
Groups of entities
------------------
For entities, groups are, as of now, used only for calculating damage.
The rating is the percentage of damage caused by tools with this damage group.
See [Entity damage mechanism].
object.get_armor_groups() --> a group-rating table (e.g. {fleshy=100})
object.set_armor_groups({fleshy=30, cracky=80})
Groups of tools
---------------
Groups in tools define which groups of nodes and entities they are
effective towards.
Groups in crafting recipes
--------------------------
An example: Make meat soup from any meat, any water and any bowl:
{
output = 'food:meat_soup_raw',
recipe = {
{'group:meat'},
{'group:water'},
{'group:bowl'},
},
-- preserve = {'group:bowl'}, -- Not implemented yet (TODO)
}
Another example: Make red wool from white wool and red dye:
{
type = 'shapeless',
output = 'wool:red',
recipe = {'wool:white', 'group:dye,basecolor_red'},
}
Special groups
--------------
* `immortal`: Disables the group damage system for an entity
* `punch_operable`: For entities; disables the regular damage mechanism for
players punching it by hand or a non-tool item, so that it can do something
else than take damage.
* `level`: Can be used to give an additional sense of progression in the game.
* A larger level will cause e.g. a weapon of a lower level make much less
damage, and get worn out much faster, or not be able to get drops
from destroyed nodes.
* `0` is something that is directly accessible at the start of gameplay
* There is no upper limit
* `dig_immediate`: Player can always pick up node without reducing tool wear
* `2`: the node always gets the digging time 0.5 seconds (rail, sign)
* `3`: the node always gets the digging time 0 seconds (torch)
* `disable_jump`: Player (and possibly other things) cannot jump from node
* `fall_damage_add_percent`: damage speed = `speed * (1 + value/100)`
* `bouncy`: value is bounce speed in percent
* `falling_node`: if there is no walkable block under the node it will fall
* `attached_node`: if the node under it is not a walkable block the node will be
dropped as an item. If the node is wallmounted the wallmounted direction is
checked.
* `soil`: saplings will grow on nodes in this group
* `connect_to_raillike`: makes nodes of raillike drawtype with same group value
connect to each other
* `slippery`: Players and items will slide on the node.
Slipperiness rises steadily with `slippery` value, starting at 1.
Known damage and digging time defining groups
---------------------------------------------
* `crumbly`: dirt, sand
* `cracky`: tough but crackable stuff like stone.
* `snappy`: something that can be cut using fine tools; e.g. leaves, small
plants, wire, sheets of metal
* `choppy`: something that can be cut using force; e.g. trees, wooden planks
* `fleshy`: Living things like animals and the player. This could imply
some blood effects when hitting.
* `explody`: Especially prone to explosions
* `oddly_breakable_by_hand`:
Can be added to nodes that shouldn't logically be breakable by the
hand but are. Somewhat similar to `dig_immediate`, but times are more
like `{[1]=3.50,[2]=2.00,[3]=0.70}` and this does not override the
speed of a tool if the tool can dig at a faster speed than this
suggests for the hand.
Examples of custom groups
-------------------------
Item groups are often used for defining, well, _groups of items_.
* `meat`: any meat-kind of a thing (rating might define the size or healing
ability or be irrelevant -- it is not defined as of yet)
* `eatable`: anything that can be eaten. Rating might define HP gain in half
hearts.
* `flammable`: can be set on fire. Rating might define the intensity of the
fire, affecting e.g. the speed of the spreading of an open fire.
* `wool`: any wool (any origin, any color)
* `metal`: any metal
* `weapon`: any weapon
* `heavy`: anything considerably heavy
Digging time calculation specifics
----------------------------------
Groups such as `crumbly`, `cracky` and `snappy` are used for this
purpose. Rating is `1`, `2` or `3`. A higher rating for such a group implies
faster digging time.
The `level` group is used to limit the toughness of nodes a tool can dig
and to scale the digging times / damage to a greater extent.
**Please do understand this**, otherwise you cannot use the system to it's
full potential.
Tools define their properties by a list of parameters for groups. They
cannot dig other groups; thus it is important to use a standard bunch of
groups to enable interaction with tools.
Tools
=====
Tools definition
----------------
Tools define:
* Full punch interval
* Maximum drop level
* For an arbitrary list of groups:
* Uses (until the tool breaks)
* Maximum level (usually `0`, `1`, `2` or `3`)
* Digging times
* Damage groups
### Full punch interval
When used as a weapon, the tool will do full damage if this time is spent
between punches. If e.g. half the time is spent, the tool will do half
damage.
### Maximum drop level
Suggests the maximum level of node, when dug with the tool, that will drop
it's useful item. (e.g. iron ore to drop a lump of iron).
This is not automated; it is the responsibility of the node definition
to implement this.
### Uses
Determines how many uses the tool has when it is used for digging a node,
of this group, of the maximum level. For lower leveled nodes, the use count
is multiplied by `3^leveldiff`.
* `uses=10, leveldiff=0`: actual uses: 10
* `uses=10, leveldiff=1`: actual uses: 30
* `uses=10, leveldiff=2`: actual uses: 90
### Maximum level
Tells what is the maximum level of a node of this group that the tool will
be able to dig.
### Digging times
List of digging times for different ratings of the group, for nodes of the
maximum level.
For example, as a Lua table, `times={2=2.00, 3=0.70}`. This would
result in the tool to be able to dig nodes that have a rating of `2` or `3`
for this group, and unable to dig the rating `1`, which is the toughest.
Unless there is a matching group that enables digging otherwise.
If the result digging time is 0, a delay of 0.15 seconds is added between
digging nodes; If the player releases LMB after digging, this delay is set to 0,
i.e. players can more quickly click the nodes away instead of holding LMB.
### Damage groups
List of damage for groups of entities. See [Entity damage mechanism].
Example definition of the capabilities of a tool
------------------------------------------------
tool_capabilities = {
full_punch_interval=1.5,
max_drop_level=1,
groupcaps={
crumbly={maxlevel=2, uses=20, times={[1]=1.60, [2]=1.20, [3]=0.80}}
}
damage_groups = {fleshy=2},
}
This makes the tool be able to dig nodes that fulfil both of these:
* Have the `crumbly` group
* Have a `level` group less or equal to `2`
Table of resulting digging times:
crumbly 0 1 2 3 4 <- level
-> 0 - - - - -
1 0.80 1.60 1.60 - -
2 0.60 1.20 1.20 - -
3 0.40 0.80 0.80 - -
level diff: 2 1 0 -1 -2
Table of resulting tool uses:
-> 0 - - - - -
1 180 60 20 - -
2 180 60 20 - -
3 180 60 20 - -
**Notes**:
* At `crumbly==0`, the node is not diggable.
* At `crumbly==3`, the level difference digging time divider kicks in and makes
easy nodes to be quickly breakable.
* At `level > 2`, the node is not diggable, because it's `level > maxlevel`
Entity damage mechanism
=======================
Damage calculation:
damage = 0
foreach group in cap.damage_groups:
damage += cap.damage_groups[group]
* limit(actual_interval / cap.full_punch_interval, 0.0, 1.0)
* (object.armor_groups[group] / 100.0)
-- Where object.armor_groups[group] is 0 for inexistent values
return damage
Client predicts damage based on damage groups. Because of this, it is able to
give an immediate response when an entity is damaged or dies; the response is
pre-defined somehow (e.g. by defining a sprite animation) (not implemented;
TODO).
Currently a smoke puff will appear when an entity dies.
The group `immortal` completely disables normal damage.
Entities can define a special armor group, which is `punch_operable`. This
group disables the regular damage mechanism for players punching it by hand or
a non-tool item, so that it can do something else than take damage.
On the Lua side, every punch calls:
entity:on_punch(puncher, time_from_last_punch, tool_capabilities, direction,
damage)
This should never be called directly, because damage is usually not handled by
the entity itself.
* `puncher` is the object performing the punch. Can be `nil`. Should never be
accessed unless absolutely required, to encourage interoperability.
* `time_from_last_punch` is time from last punch (by `puncher`) or `nil`.
* `tool_capabilities` can be `nil`.
* `direction` is a unit vector, pointing from the source of the punch to
the punched object.
* `damage` damage that will be done to entity
Return value of this function will determine if damage is done by this function
(retval true) or shall be done by engine (retval false)
To punch an entity/object in Lua, call:
object:punch(puncher, time_from_last_punch, tool_capabilities, direction)
* Return value is tool wear.
* Parameters are equal to the above callback.
* If `direction` equals `nil` and `puncher` does not equal `nil`, `direction`
will be automatically filled in based on the location of `puncher`.
Metadata
========
Node Metadata
-------------
The instance of a node in the world normally only contains the three values
mentioned in [Nodes]. However, it is possible to insert extra data into a node.
It is called "node metadata"; See `NodeMetaRef`.
Node metadata contains two things:
* A key-value store
* An inventory
Some of the values in the key-value store are handled specially:
* `formspec`: Defines a right-click inventory menu. See [Formspec].
* `infotext`: Text shown on the screen when the node is pointed at
Example:
local meta = minetest.get_meta(pos)
meta:set_string("formspec",
"size[8,9]"..
"list[context;main;0,0;8,4;]"..
"list[current_player;main;0,5;8,4;]")
meta:set_string("infotext", "Chest");
local inv = meta:get_inventory()
inv:set_size("main", 8*4)
print(dump(meta:to_table()))
meta:from_table({
inventory = {
main = {[1] = "default:dirt", [2] = "", [3] = "", [4] = "",
[5] = "", [6] = "", [7] = "", [8] = "", [9] = "",
[10] = "", [11] = "", [12] = "", [13] = "",
[14] = "default:cobble", [15] = "", [16] = "", [17] = "",
[18] = "", [19] = "", [20] = "default:cobble", [21] = "",
[22] = "", [23] = "", [24] = "", [25] = "", [26] = "",
[27] = "", [28] = "", [29] = "", [30] = "", [31] = "",
[32] = ""}
},
fields = {
formspec = "size[8,9]list[context;main;0,0;8,4;]list[current_player;main;0,5;8,4;]",
infotext = "Chest"
}
})
Item Metadata
-------------
Item stacks can store metadata too. See [`ItemStackMetaRef`].
Item metadata only contains a key-value store.
Some of the values in the key-value store are handled specially:
* `description`: Set the item stack's description. Defaults to
`idef.description`.
* `color`: A `ColorString`, which sets the stack's color.
* `palette_index`: If the item has a palette, this is used to get the
current color from the palette.
Example:
local meta = stack:get_meta()
meta:set_string("key", "value")
print(dump(meta:to_table()))
Formspec
========
Formspec defines a menu. Currently not much else than inventories are
supported. It is a string, with a somewhat strange format.
Spaces and newlines can be inserted between the blocks, as is used in the
examples.
Position and size units are inventory slots, `X` and `Y` position the formspec
element relative to the top left of the menu or container. `W` and `H` are its
width and height values.
When displaying text which can contain formspec code, e.g. text set by a player,
use `minetest.formspec_escape`.
For coloured text you can use `minetest.colorize`.
WARNING: Minetest allows you to add elements to every single formspec instance
using `player:set_formspec_prepend()`, which may be the reason backgrounds are
appearing when you don't expect them to. See [`no_prepend[]`].
Examples
--------
### Chest
size[8,9]
list[context;main;0,0;8,4;]
list[current_player;main;0,5;8,4;]
### Furnace
size[8,9]
list[context;fuel;2,3;1,1;]
list[context;src;2,1;1,1;]
list[context;dst;5,1;2,2;]
list[current_player;main;0,5;8,4;]
### Minecraft-like player inventory
size[8,7.5]
image[1,0.6;1,2;player.png]
list[current_player;main;0,3.5;8,4;]
list[current_player;craft;3,0;3,3;]
list[current_player;craftpreview;7,1;1,1;]
Elements
--------
### `size[<W>,<H>,<fixed_size>]`
* Define the size of the menu in inventory slots
* `fixed_size`: `true`/`false` (optional)
* deprecated: `invsize[<W>,<H>;]`
### `position[<X>,<Y>]`
* Must be used after `size` element.
* Defines the position on the game window of the formspec's `anchor` point.
* For X and Y, 0.0 and 1.0 represent opposite edges of the game window,
for example:
* [0.0, 0.0] sets the position to the top left corner of the game window.
* [1.0, 1.0] sets the position to the bottom right of the game window.
* Defaults to the center of the game window [0.5, 0.5].
### `anchor[<X>,<Y>]`
* Must be used after both `size` and `position` (if present) elements.
* Defines the location of the anchor point within the formspec.
* For X and Y, 0.0 and 1.0 represent opposite edges of the formspec,
for example:
* [0.0, 1.0] sets the anchor to the bottom left corner of the formspec.
* [1.0, 0.0] sets the anchor to the top right of the formspec.
* Defaults to the center of the formspec [0.5, 0.5].
* `position` and `anchor` elements need suitable values to avoid a formspec
extending off the game window due to particular game window sizes.
### `no_prepend[]`
* Must be used after the `size`, `position`, and `anchor` elements (if present).
* Disables player:set_formspec_prepend() from applying to this formspec.
### `container[<X>,<Y>]`
* Start of a container block, moves all physical elements in the container by
(X, Y).
* Must have matching `container_end`
* Containers can be nested, in which case the offsets are added
(child containers are relative to parent containers)
### `container_end[]`
* End of a container, following elements are no longer relative to this
container.
### `list[<inventory location>;<list name>;<X>,<Y>;<W>,<H>;]`
* Show an inventory list
### `list[<inventory location>;<list name>;<X>,<Y>;<W>,<H>;<starting item index>]`
* Show an inventory list
### `listring[<inventory location>;<list name>]`
* Allows to create a ring of inventory lists
* Shift-clicking on items in one element of the ring
will send them to the next inventory list inside the ring
* The first occurrence of an element inside the ring will
determine the inventory where items will be sent to
### `listring[]`
* Shorthand for doing `listring[<inventory location>;<list name>]`
for the last two inventory lists added by list[...]
### `listcolors[<slot_bg_normal>;<slot_bg_hover>]`
* Sets background color of slots as `ColorString`
* Sets background color of slots on mouse hovering
### `listcolors[<slot_bg_normal>;<slot_bg_hover>;<slot_border>]`
* Sets background color of slots as `ColorString`
* Sets background color of slots on mouse hovering
* Sets color of slots border
### `listcolors[<slot_bg_normal>;<slot_bg_hover>;<slot_border>;<tooltip_bgcolor>;<tooltip_fontcolor>]`
* Sets background color of slots as `ColorString`
* Sets background color of slots on mouse hovering
* Sets color of slots border
* Sets default background color of tooltips
* Sets default font color of tooltips
### `tooltip[<gui_element_name>;<tooltip_text>;<bgcolor>;<fontcolor>]`
* Adds tooltip for an element
* `<bgcolor>` tooltip background color as `ColorString` (optional)
* `<fontcolor>` tooltip font color as `ColorString` (optional)
### `tooltip[<X>,<Y>;<W>,<H>;<tooltip_text>;<bgcolor>;<fontcolor>]`
* Adds tooltip for an area. Other tooltips will take priority when present.
* `<bgcolor>` tooltip background color as `ColorString` (optional)
* `<fontcolor>` tooltip font color as `ColorString` (optional)
### `image[<X>,<Y>;<W>,<H>;<texture name>]`
* Show an image
### `item_image[<X>,<Y>;<W>,<H>;<item name>]`
* Show an inventory image of registered item/node
### `bgcolor[<color>;<fullscreen>]`
* Sets background color of formspec as `ColorString`
* If `true`, the background color is drawn fullscreen (does not affect the size
of the formspec).
### `background[<X>,<Y>;<W>,<H>;<texture name>]`
* Use a background. Inventory rectangles are not drawn then.
* Example for formspec 8x4 in 16x resolution: image shall be sized
8 times 16px times 4 times 16px.
### `background[<X>,<Y>;<W>,<H>;<texture name>;<auto_clip>]`
* Use a background. Inventory rectangles are not drawn then.
* Example for formspec 8x4 in 16x resolution:
image shall be sized 8 times 16px times 4 times 16px
* If `auto_clip` is `true`, the background is clipped to the formspec size
(`x` and `y` are used as offset values, `w` and `h` are ignored)
### `pwdfield[<X>,<Y>;<W>,<H>;<name>;<label>]`
* Textual password style field; will be sent to server when a button is clicked
* When enter is pressed in field, fields.key_enter_field will be sent with the
name of this field.
* Fields are a set height, but will be vertically centred on `H`
* `name` is the name of the field as returned in fields to `on_receive_fields`
* `label`, if not blank, will be text printed on the top left above the field
* See `field_close_on_enter` to stop enter closing the formspec
### `field[<X>,<Y>;<W>,<H>;<name>;<label>;<default>]`
* Textual field; will be sent to server when a button is clicked
* When enter is pressed in field, `fields.key_enter_field` will be sent with
the name of this field.
* Fields are a set height, but will be vertically centred on `H`
* `name` is the name of the field as returned in fields to `on_receive_fields`
* `label`, if not blank, will be text printed on the top left above the field
* `default` is the default value of the field
* `default` may contain variable references such as `${text}` which
will fill the value from the metadata value `text`
* **Note**: no extra text or more than a single variable is supported ATM.
* See `field_close_on_enter` to stop enter closing the formspec
### `field[<name>;<label>;<default>]`
* As above, but without position/size units
* When enter is pressed in field, `fields.key_enter_field` will be sent with
the name of this field.
* Special field for creating simple forms, such as sign text input
* Must be used without a `size[]` element
* A "Proceed" button will be added automatically
* See `field_close_on_enter` to stop enter closing the formspec
### `field_close_on_enter[<name>;<close_on_enter>]`
* <name> is the name of the field
* if <close_on_enter> is false, pressing enter in the field will submit the
form but not close it.
* defaults to true when not specified (ie: no tag for a field)
### `textarea[<X>,<Y>;<W>,<H>;<name>;<label>;<default>]`
* Same as fields above, but with multi-line input
* If the text overflows, a vertical scrollbar is added.
* If the name is empty, the textarea is read-only and
the background is not shown, which corresponds to a multi-line label.
### `label[<X>,<Y>;<label>]`
* The label formspec element displays the text set in `label`
at the specified position.
* The text is displayed directly without automatic line breaking,
so label should not be used for big text chunks.
### `vertlabel[<X>,<Y>;<label>]`
* Textual label drawn vertically
* `label` is the text on the label
### `button[<X>,<Y>;<W>,<H>;<name>;<label>]`
* Clickable button. When clicked, fields will be sent.
* Fixed button height. It will be vertically centred on `H`
* `label` is the text on the button
### `image_button[<X>,<Y>;<W>,<H>;<texture name>;<name>;<label>]`
* `texture name` is the filename of an image
### `image_button[<X>,<Y>;<W>,<H>;<texture name>;<name>;<label>;<noclip>;<drawborder>;<pressed texture name>]`
* `texture name` is the filename of an image
* `noclip=true` means the image button doesn't need to be within specified
formsize.
* `drawborder`: draw button border or not
* `pressed texture name` is the filename of an image on pressed state
### `item_image_button[<X>,<Y>;<W>,<H>;<item name>;<name>;<label>]`
* `item name` is the registered name of an item/node,
tooltip will be made out of its description
to override it use tooltip element
### `button_exit[<X>,<Y>;<W>,<H>;<name>;<label>]`
* When clicked, fields will be sent and the form will quit.
### `image_button_exit[<X>,<Y>;<W>,<H>;<texture name>;<name>;<label>]`
* When clicked, fields will be sent and the form will quit.
### `textlist[<X>,<Y>;<W>,<H>;<name>;<listelem 1>,<listelem 2>,...,<listelem n>]`
* Scrollable item list showing arbitrary text elements
* `name` fieldname sent to server on doubleclick value is current selected
element.
* `listelements` can be prepended by #color in hexadecimal format RRGGBB
(only).
* if you want a listelement to start with "#" write "##".
### `textlist[<X>,<Y>;<W>,<H>;<name>;<listelem 1>,<listelem 2>,...,<listelem n>;<selected idx>;<transparent>]`
* Scrollable itemlist showing arbitrary text elements
* `name` fieldname sent to server on doubleclick value is current selected
element.
* `listelements` can be prepended by #RRGGBB (only) in hexadecimal format
* if you want a listelement to start with "#" write "##"
* Index to be selected within textlist
* `true`/`false`: draw transparent background
* See also `minetest.explode_textlist_event`
(main menu: `core.explode_textlist_event`).
### `tabheader[<X>,<Y>;<name>;<caption 1>,<caption 2>,...,<caption n>;<current_tab>;<transparent>;<draw_border>]`
* Show a tab**header** at specific position (ignores formsize)
* `name` fieldname data is transferred to Lua
* `caption 1`...: name shown on top of tab
* `current_tab`: index of selected tab 1...
* `transparent` (optional): show transparent
* `draw_border` (optional): draw border
### `box[<X>,<Y>;<W>,<H>;<color>]`
* Simple colored box
* `color` is color specified as a `ColorString`.
If the alpha component is left blank, the box will be semitransparent.
### `dropdown[<X>,<Y>;<W>;<name>;<item 1>,<item 2>, ...,<item n>;<selected idx>]`
* Show a dropdown field
* **Important note**: There are two different operation modes:
1. handle directly on change (only changed dropdown is submitted)
2. read the value on pressing a button (all dropdown values are available)
* `x` and `y` position of dropdown
* Width of dropdown
* Fieldname data is transferred to Lua
* Items to be shown in dropdown
* Index of currently selected dropdown item
### `checkbox[<X>,<Y>;<name>;<label>;<selected>]`
* Show a checkbox
* `name` fieldname data is transferred to Lua
* `label` to be shown left of checkbox
* `selected` (optional): `true`/`false`
### `scrollbar[<X>,<Y>;<W>,<H>;<orientation>;<name>;<value>]`
* Show a scrollbar
* There are two ways to use it:
1. handle the changed event (only changed scrollbar is available)
2. read the value on pressing a button (all scrollbars are available)
* `orientation`: `vertical`/`horizontal`
* Fieldname data is transferred to Lua
* Value this trackbar is set to (`0`-`1000`)
* See also `minetest.explode_scrollbar_event`
(main menu: `core.explode_scrollbar_event`).
### `table[<X>,<Y>;<W>,<H>;<name>;<cell 1>,<cell 2>,...,<cell n>;<selected idx>]`
* Show scrollable table using options defined by the previous `tableoptions[]`
* Displays cells as defined by the previous `tablecolumns[]`
* `name`: fieldname sent to server on row select or doubleclick
* `cell 1`...`cell n`: cell contents given in row-major order
* `selected idx`: index of row to be selected within table (first row = `1`)
* See also `minetest.explode_table_event`
(main menu: `core.explode_table_event`).
### `tableoptions[<opt 1>;<opt 2>;...]`
* Sets options for `table[]`
* `color=#RRGGBB`
* default text color (`ColorString`), defaults to `#FFFFFF`
* `background=#RRGGBB`
* table background color (`ColorString`), defaults to `#000000`
* `border=<true/false>`
* should the table be drawn with a border? (default: `true`)
* `highlight=#RRGGBB`
* highlight background color (`ColorString`), defaults to `#466432`
* `highlight_text=#RRGGBB`
* highlight text color (`ColorString`), defaults to `#FFFFFF`
* `opendepth=<value>`
* all subtrees up to `depth < value` are open (default value = `0`)
* only useful when there is a column of type "tree"
### `tablecolumns[<type 1>,<opt 1a>,<opt 1b>,...;<type 2>,<opt 2a>,<opt 2b>;...]`
* Sets columns for `table[]`
* Types: `text`, `image`, `color`, `indent`, `tree`
* `text`: show cell contents as text
* `image`: cell contents are an image index, use column options to define
images.
* `color`: cell contents are a ColorString and define color of following
cell.
* `indent`: cell contents are a number and define indentation of following
cell.
* `tree`: same as indent, but user can open and close subtrees
(treeview-like).
* Column options:
* `align=<value>`
* for `text` and `image`: content alignment within cells.
Available values: `left` (default), `center`, `right`, `inline`
* `width=<value>`
* for `text` and `image`: minimum width in em (default: `0`)
* for `indent` and `tree`: indent width in em (default: `1.5`)
* `padding=<value>`: padding left of the column, in em (default `0.5`).
Exception: defaults to 0 for indent columns
* `tooltip=<value>`: tooltip text (default: empty)
* `image` column options:
* `0=<value>` sets image for image index 0
* `1=<value>` sets image for image index 1
* `2=<value>` sets image for image index 2
* and so on; defined indices need not be contiguous empty or
non-numeric cells are treated as `0`.
* `color` column options:
* `span=<value>`: number of following columns to affect
(default: infinite).
**Note**: do _not_ use a element name starting with `key_`; those names are
reserved to pass key press events to formspec!
Inventory
=========
Inventory locations
-------------------
* `"context"`: Selected node metadata (deprecated: `"current_name"`)
* `"current_player"`: Player to whom the menu is shown
* `"player:<name>"`: Any player
* `"nodemeta:<X>,<Y>,<Z>"`: Any node metadata
* `"detached:<name>"`: A detached inventory
Player Inventory lists
----------------------
* `main`: list containing the default inventory
* `craft`: list containing the craft input
* `craftpreview`: list containing the craft output
* `hand`: list containing an override for the empty hand
Colors
======
`ColorString`
-------------
`#RGB` defines a color in hexadecimal format.
`#RGBA` defines a color in hexadecimal format and alpha channel.
`#RRGGBB` defines a color in hexadecimal format.
`#RRGGBBAA` defines a color in hexadecimal format and alpha channel.
Named colors are also supported and are equivalent to
[CSS Color Module Level 4](http://dev.w3.org/csswg/css-color/#named-colors).
To specify the value of the alpha channel, append `#AA` to the end of the color
name (e.g. `colorname#08`). For named colors the hexadecimal string
representing the alpha value must (always) be two hexadecimal digits.
`ColorSpec`
-----------
A ColorSpec specifies a 32-bit color. It can be written in any of the following
forms:
* table form: Each element ranging from 0..255 (a, if absent, defaults to 255):
* `colorspec = {a=255, r=0, g=255, b=0}`
* numerical form: The raw integer value of an ARGB8 quad:
* `colorspec = 0xFF00FF00`
* string form: A ColorString (defined above):
* `colorspec = "green"`
Escape sequences
================
Most text can contain escape sequences, that can for example color the text.
There are a few exceptions: tab headers, dropdowns and vertical labels can't.
The following functions provide escape sequences:
* `minetest.get_color_escape_sequence(color)`:
* `color` is a ColorString
* The escape sequence sets the text color to `color`
* `minetest.colorize(color, message)`:
* Equivalent to:
`minetest.get_color_escape_sequence(color) ..
message ..
minetest.get_color_escape_sequence("#ffffff")`
* `minetest.get_background_escape_sequence(color)`
* `color` is a ColorString
* The escape sequence sets the background of the whole text element to
`color`. Only defined for item descriptions and tooltips.
* `minetest.strip_foreground_colors(str)`
* Removes foreground colors added by `get_color_escape_sequence`.
* `minetest.strip_background_colors(str)`
* Removes background colors added by `get_background_escape_sequence`.
* `minetest.strip_colors(str)`
* Removes all color escape sequences.
Spatial Vectors
===============
For the following functions, `v`, `v1`, `v2` are vectors,
`p1`, `p2` are positions:
* `vector.new(a[, b, c])`:
* Returns a vector.
* A copy of `a` if `a` is a vector.
* `{x = a, y = b, z = c}`, if all of `a`, `b`, `c` are defined numbers.
* `vector.direction(p1, p2)`:
* Returns a vector of length 1 with direction `p1` to `p2`.
* If `p1` and `p2` are identical, returns `{x = 0, y = 0, z = 0}`.
* `vector.distance(p1, p2)`:
* Returns zero or a positive number, the distance between `p1` and `p2`.
* `vector.length(v)`:
* Returns zero or a positive number, the length of vector `v`.
* `vector.normalize(v)`:
* Returns a vector of length 1 with direction of vector `v`.
* If `v` has zero length, returns `{x = 0, y = 0, z = 0}`.
* `vector.floor(v)`:
* Returns a vector, each dimension rounded down.
* `vector.round(v)`:
* Returns a vector, each dimension rounded to nearest integer.
* `vector.apply(v, func)`:
* Returns a vector where the function `func` has been applied to each
component.
* `vector.equals(v1, v2)`:
* Returns a boolean, `true` if the vectors are identical.
* `vector.sort(v1, v2)`:
* Returns in order minp, maxp vectors of the cuboid defined by `v1`, `v2`.
For the following functions `x` can be either a vector or a number:
* `vector.add(v, x)`:
* Returns a vector.
* `vector.subtract(v, x)`:
* Returns a vector.
* `vector.multiply(v, x)`:
* Returns a scaled vector or Schur product.
* `vector.divide(v, x)`:
* Returns a scaled vector or Schur quotient.
Helper functions
================
* `dump2(obj, name, dumped)`: returns a string which makes `obj`
human-readable, handles reference loops.
* `obj`: arbitrary variable
* `name`: string, default: `"_"`
* `dumped`: table, default: `{}`
* `dump(obj, dumped)`: returns a string which makes `obj` human-readable
* `obj`: arbitrary variable
* `dumped`: table, default: `{}`
* `math.hypot(x, y)`
* Get the hypotenuse of a triangle with legs x and y.
Useful for distance calculation.
* `math.sign(x, tolerance)`: returns `-1`, `0` or `1`
* Get the sign of a number.
* tolerance: number, default: `0.0`
* If the absolute value of `x` is within the `tolerance` or `x` is NaN,
`0` is returned.
* `string.split(str, separator, include_empty, max_splits, sep_is_pattern)`
* `separator`: string, default: `","`
* `include_empty`: boolean, default: `false`
* `max_splits`: number, if it's negative, splits aren't limited,
default: `-1`
* `sep_is_pattern`: boolean, it specifies whether separator is a plain
string or a pattern (regex), default: `false`
* e.g. `"a,b":split","` returns `{"a","b"}`
* `string:trim()`: returns the string without whitespace pre- and suffixes
* e.g. `"\n \t\tfoo bar\t ":trim()` returns `"foo bar"`
* `minetest.wrap_text(str, limit, as_table)`: returns a string or table
* Adds newlines to the string to keep it within the specified character
limit
* Note that the returned lines may be longer than the limit since it only
splits at word borders.
* `limit`: number, maximal amount of characters in one line
* `as_table`: boolean, if set to true, a table of lines instead of a string
is returned, default: `false`
* `minetest.pos_to_string(pos, decimal_places)`: returns string `"(X,Y,Z)"`
* `pos`: table {x=X, y=Y, z=Z}
* Converts the position `pos` to a human-readable, printable string
* `decimal_places`: number, if specified, the x, y and z values of
the position are rounded to the given decimal place.
* `minetest.string_to_pos(string)`: returns a position or `nil`
* Same but in reverse.
* If the string can't be parsed to a position, nothing is returned.
* `minetest.string_to_area("(X1, Y1, Z1) (X2, Y2, Z2)")`: returns two positions
* Converts a string representing an area box into two positions
* `minetest.formspec_escape(string)`: returns a string
* escapes the characters "[", "]", "\", "," and ";", which can not be used
in formspecs.
* `minetest.is_yes(arg)`
* returns true if passed 'y', 'yes', 'true' or a number that isn't zero.
* `minetest.is_nan(arg)`
* returns true when the passed number represents NaN.
* `minetest.get_us_time()`
* returns time with microsecond precision. May not return wall time.
* `table.copy(table)`: returns a table
* returns a deep copy of `table`
* `table.insert_all(table, other_table)`:
* Appends all values in `other_table` to `table` - uses `#table + 1` to
find new indices.
* `minetest.pointed_thing_to_face_pos(placer, pointed_thing)`: returns a
position.
* returns the exact position on the surface of a pointed node
Translations
============
Texts can be translated client-side with the help of `minetest.translate` and
translation files.
Translating a string
--------------------
Two functions are provided to translate strings: `minetest.translate` and
`minetest.get_translator`.
* `minetest.get_translator(textdomain)` is a simple wrapper around
`minetest.translate`, and `minetest.get_translator(textdomain)(str, ...)` is
equivalent to `minetest.translate(textdomain, str, ...)`.
It is intended to be used in the following way, so that it avoids verbose
repetitions of `minetest.translate`:
local S = minetest.get_translator(textdomain)
S(str, ...)
As an extra commodity, if `textdomain` is nil, it is assumed to be "" instead.
* `minetest.translate(textdomain, str, ...)` translates the string `str` with
the given `textdomain` for disambiguation. The textdomain must match the
textdomain specified in the translation file in order to get the string
translated. This can be used so that a string is translated differently in
different contexts.
It is advised to use the name of the mod as textdomain whenever possible, to
avoid clashes with other mods.
This function must be given a number of arguments equal to the number of
arguments the translated string expects.
Arguments are literal strings -- they will not be translated, so if you want
them to be, they need to come as outputs of `minetest.translate` as well.
For instance, suppose we want to translate "@1 Wool" with "@1" being replaced
by the translation of "Red". We can do the following:
local S = minetest.get_translator()
S("@1 Wool", S("Red"))
This will be displayed as "Red Wool" on old clients and on clients that do
not have localization enabled. However, if we have for instance a translation
file named `wool.fr.tr` containing the following:
@1 Wool=Laine @1
Red=Rouge
this will be displayed as "Laine Rouge" on clients with a French locale.
Operations on translated strings
--------------------------------
The output of `minetest.translate` is a string, with escape sequences adding
additional information to that string so that it can be translated on the
different clients. In particular, you can't expect operations like string.length
to work on them like you would expect them to, or string.gsub to work in the
expected manner. However, string concatenation will still work as expected
(note that you should only use this for things like formspecs; do not translate
sentences by breaking them into parts; arguments should be used instead), and
operations such as `minetest.colorize` which are also concatenation.
Translation file format
-----------------------
A translation file has the suffix `.[lang].tr`, where `[lang]` is the language
it corresponds to. It must be put into the `locale` subdirectory of the mod.
The file should be a text file, with the following format:
* Lines beginning with `# textdomain:` (the space is significant) can be used
to specify the text domain of all following translations in the file.
* All other empty lines or lines beginning with `#` are ignored.
* Other lines should be in the format `original=translated`. Both `original`
and `translated` can contain escape sequences beginning with `@` to insert
arguments, literal `@`, `=` or newline (See [Escapes] below).
There must be no extraneous whitespace around the `=` or at the beginning or
the end of the line.
Escapes
-------
Strings that need to be translated can contain several escapes, preceded by `@`.
* `@@` acts as a literal `@`.
* `@n`, where `n` is a digit between 1 and 9, is an argument for the translated
string that will be inlined when translated. Due to how translations are
implemented, the original translation string **must** have its arguments in
increasing order, without gaps or repetitions, starting from 1.
* `@=` acts as a literal `=`. It is not required in strings given to
`minetest.translate`, but is in translation files to avoid being confused
with the `=` separating the original from the translation.
* `@\n` (where the `\n` is a literal newline) acts as a literal newline.
As with `@=`, this escape is not required in strings given to
`minetest.translate`, but is in translation files.
* `@n` acts as a literal newline as well.
Perlin noise
============
Perlin noise creates a continuously-varying value depending on the input values.
Usually in Minetest the input values are either 2D or 3D co-ordinates in nodes.
The result is used during map generation to create the terrain shape, vary heat
and humidity to distribute biomes, vary the density of decorations or vary the
structure of ores.
Structure of perlin noise
-------------------------
An 'octave' is a simple noise generator that outputs a value between -1 and 1.
The smooth wavy noise it generates has a single characteristic scale, almost
like a 'wavelength', so on its own does not create fine detail.
Due to this perlin noise combines several octaves to create variation on
multiple scales. Each additional octave has a smaller 'wavelength' than the
previous.
This combination results in noise varying very roughly between -2.0 and 2.0 and
with an average value of 0.0, so `scale` and `offset` are then used to multiply
and offset the noise variation.
The final perlin noise variation is created as follows:
noise = offset + scale * (octave1 +
octave2 * persistence +
octave3 * persistence ^ 2 +
octave4 * persistence ^ 3 +
...)
Noise Parameters
----------------
Noise Parameters are commonly called `NoiseParams`.
### `offset`
After the multiplication by `scale` this is added to the result and is the final
step in creating the noise value.
Can be positive or negative.
### `scale`
Once all octaves have been combined, the result is multiplied by this.
Can be positive or negative.
### `spread`
For octave1, this is roughly the change of input value needed for a very large
variation in the noise value generated by octave1. It is almost like a
'wavelength' for the wavy noise variation.
Each additional octave has a 'wavelength' that is smaller than the previous
octave, to create finer detail. `spread` will therefore roughly be the typical
size of the largest structures in the final noise variation.
`spread` is a vector with values for x, y, z to allow the noise variation to be
stretched or compressed in the desired axes.
Values are positive numbers.
### `seed`
This is a whole number that determines the entire pattern of the noise
variation. Altering it enables different noise patterns to be created.
With other parameters equal, different seeds produce different noise patterns
and identical seeds produce identical noise patterns.
For this parameter you can randomly choose any whole number. Usually it is
preferable for this to be different from other seeds, but sometimes it is useful
to be able to create identical noise patterns.
When used in mapgen this is actually a 'seed offset', it is added to the
'world seed' to create the seed used by the noise, to ensure the noise has a
different pattern in different worlds.
### `octaves`
The number of simple noise generators that are combined.
A whole number, 1 or more.
Each additional octave adds finer detail to the noise but also increases the
noise calculation load.
3 is a typical minimum for a high quality, complex and natural-looking noise
variation. 1 octave has a slight 'gridlike' appearence.
Choose the number of octaves according to the `spread` and `lacunarity`, and the
size of the finest detail you require. For example:
if `spread` is 512 nodes, `lacunarity` is 2.0 and finest detail required is 16
nodes, octaves will be 6 because the 'wavelengths' of the octaves will be
512, 256, 128, 64, 32, 16 nodes.
Warning: If the 'wavelength' of any octave falls below 1 an error will occur.
### `persistence`
Each additional octave has an amplitude that is the amplitude of the previous
octave multiplied by `persistence`, to reduce the amplitude of finer details,
as is often helpful and natural to do so.
Since this controls the balance of fine detail to large-scale detail
`persistence` can be thought of as the 'roughness' of the noise.
A positive or negative non-zero number, often between 0.3 and 1.0.
A common medium value is 0.5, such that each octave has half the amplitude of
the previous octave.
This may need to be tuned when altering `lacunarity`; when doing so consider
that a common medium value is 1 / lacunarity.
### `lacunarity`
Each additional octave has a 'wavelength' that is the 'wavelength' of the
previous octave multiplied by 1 / lacunarity, to create finer detail.
'lacunarity' is often 2.0 so 'wavelength' often halves per octave.
A positive number no smaller than 1.0.
Values below 2.0 create higher quality noise at the expense of requiring more
octaves to cover a paticular range of 'wavelengths'.
### `flags`
Leave this field unset for no special handling.
Currently supported are `defaults`, `eased` and `absvalue`:
#### `defaults`
Specify this if you would like to keep auto-selection of eased/not-eased while
specifying some other flags.
#### `eased`
Maps noise gradient values onto a quintic S-curve before performing
interpolation. This results in smooth, rolling noise.
Disable this (`noeased`) for sharp-looking noise with a slightly gridded
appearence.
If no flags are specified (or defaults is), 2D noise is eased and 3D noise is
not eased.
Easing a 3D noise significantly increases the noise calculation load, so use
with restraint.
#### `absvalue`
The absolute value of each octave's noise variation is used when combining the
octaves. The final perlin noise variation is created as follows:
noise = offset + scale * (abs(octave1) +
abs(octave2) * persistence +
abs(octave3) * persistence ^ 2 +
abs(octave4) * persistence ^ 3 +
...)
### Format example
For 2D or 3D perlin noise or perlin noise maps:
np_terrain = {
offset = 0,
scale = 1,
spread = {x = 500, y = 500, z = 500},
seed = 571347,
octaves = 5,
persist = 0.63,
lacunarity = 2.0,
flags = "defaults, absvalue",
}
For 2D noise the Z component of `spread` is still defined but is ignored.
A single noise parameter table can be used for 2D or 3D noise.
Ores
====
Ore types
---------
These tell in what manner the ore is generated.
All default ores are of the uniformly-distributed scatter type.
### `scatter`
Randomly chooses a location and generates a cluster of ore.
If `noise_params` is specified, the ore will be placed if the 3D perlin noise
at that point is greater than the `noise_threshold`, giving the ability to
create a non-equal distribution of ore.
### `sheet`
Creates a sheet of ore in a blob shape according to the 2D perlin noise
described by `noise_params` and `noise_threshold`. This is essentially an
improved version of the so-called "stratus" ore seen in some unofficial mods.
This sheet consists of vertical columns of uniform randomly distributed height,
varying between the inclusive range `column_height_min` and `column_height_max`.
If `column_height_min` is not specified, this parameter defaults to 1.
If `column_height_max` is not specified, this parameter defaults to `clust_size`
for reverse compatibility. New code should prefer `column_height_max`.
The `column_midpoint_factor` parameter controls the position of the column at
which ore emanates from.
If 1, columns grow upward. If 0, columns grow downward. If 0.5, columns grow
equally starting from each direction.
`column_midpoint_factor` is a decimal number ranging in value from 0 to 1. If
this parameter is not specified, the default is 0.5.
The ore parameters `clust_scarcity` and `clust_num_ores` are ignored for this
ore type.
### `puff`
Creates a sheet of ore in a cloud-like puff shape.
As with the `sheet` ore type, the size and shape of puffs are described by
`noise_params` and `noise_threshold` and are placed at random vertical
positions within the currently generated chunk.
The vertical top and bottom displacement of each puff are determined by the
noise parameters `np_puff_top` and `np_puff_bottom`, respectively.
### `blob`
Creates a deformed sphere of ore according to 3d perlin noise described by
`noise_params`. The maximum size of the blob is `clust_size`, and
`clust_scarcity` has the same meaning as with the `scatter` type.
### `vein`
Creates veins of ore varying in density by according to the intersection of two
instances of 3d perlin noise with different seeds, both described by
`noise_params`.
`random_factor` varies the influence random chance has on placement of an ore
inside the vein, which is `1` by default. Note that modifying this parameter
may require adjusting `noise_threshold`.
The parameters `clust_scarcity`, `clust_num_ores`, and `clust_size` are ignored
by this ore type.
This ore type is difficult to control since it is sensitive to small changes.
The following is a decent set of parameters to work from:
noise_params = {
offset = 0,
scale = 3,
spread = {x=200, y=200, z=200},
seed = 5390,
octaves = 4,
persist = 0.5,
lacunarity = 2.0,
flags = "eased",
},
noise_threshold = 1.6
**WARNING**: Use this ore type *very* sparingly since it is ~200x more
computationally expensive than any other ore.
### `stratum`
Creates a single undulating ore stratum that is continuous across mapchunk
borders and horizontally spans the world.
The 2D perlin noise described by `noise_params` defines the Y co-ordinate of
the stratum midpoint. The 2D perlin noise described by `np_stratum_thickness`
defines the stratum's vertical thickness (in units of nodes). Due to being
continuous across mapchunk borders the stratum's vertical thickness is
unlimited.
If the noise parameter `noise_params` is omitted the ore will occur from y_min
to y_max in a simple horizontal stratum.
A parameter `stratum_thickness` can be provided instead of the noise parameter
`np_stratum_thickness`, to create a constant thickness.
Leaving out one or both noise parameters makes the ore generation less
intensive, useful when adding multiple strata.
`y_min` and `y_max` define the limits of the ore generation and for performance
reasons should be set as close together as possible but without clipping the
stratum's Y variation.
Each node in the stratum has a 1-in-`clust_scarcity` chance of being ore, so a
solid-ore stratum would require a `clust_scarcity` of 1.
The parameters `clust_num_ores`, `clust_size`, `noise_threshold` and
`random_factor` are ignored by this ore type.
Ore attributes
--------------
See section [Flag Specifier Format].
Currently supported flags:
`puff_cliffs`, `puff_additive_composition`.
### `puff_cliffs`
If set, puff ore generation will not taper down large differences in
displacement when approaching the edge of a puff. This flag has no effect for
ore types other than `puff`.
### `puff_additive_composition`
By default, when noise described by `np_puff_top` or `np_puff_bottom` results
in a negative displacement, the sub-column at that point is not generated. With
this attribute set, puff ore generation will instead generate the absolute
difference in noise displacement values. This flag has no effect for ore types
other than `puff`.
Decoration types
================
The varying types of decorations that can be placed.
`simple`
--------
Creates a 1 times `H` times 1 column of a specified node (or a random node from
a list, if a decoration list is specified). Can specify a certain node it must
spawn next to, such as water or lava, for example. Can also generate a
decoration of random height between a specified lower and upper bound.
This type of decoration is intended for placement of grass, flowers, cacti,
papyri, waterlilies and so on.
`schematic`
-----------
Copies a box of `MapNodes` from a specified schematic file (or raw description).
Can specify a probability of a node randomly appearing when placed.
This decoration type is intended to be used for multi-node sized discrete
structures, such as trees, cave spikes, rocks, and so on.
Schematics
==========
Schematic specifier
--------------------
A schematic specifier identifies a schematic by either a filename to a
Minetest Schematic file (`.mts`) or through raw data supplied through Lua,
in the form of a table. This table specifies the following fields:
* The `size` field is a 3D vector containing the dimensions of the provided
schematic. (required field)
* The `yslice_prob` field is a table of {ypos, prob} slice tables. A slice table
sets the probability of a particular horizontal slice of the schematic being
placed. (optional field)
`ypos` = 0 for the lowest horizontal slice of a schematic.
The default of `prob` is 255.
* The `data` field is a flat table of MapNode tables making up the schematic,
in the order of `[z [y [x]]]`. (required field)
Each MapNode table contains:
* `name`: the name of the map node to place (required)
* `prob` (alias `param1`): the probability of this node being placed
(default: 255)
* `param2`: the raw param2 value of the node being placed onto the map
(default: 0)
* `force_place`: boolean representing if the node should forcibly overwrite
any previous contents (default: false)
About probability values:
* A probability value of `0` or `1` means that node will never appear
(0% chance).
* A probability value of `254` or `255` means the node will always appear
(100% chance).
* If the probability value `p` is greater than `1`, then there is a
`(p / 256 * 100)` percent chance that node will appear when the schematic is
placed on the map.
Schematic attributes
--------------------
See section [Flag Specifier Format].
Currently supported flags: `place_center_x`, `place_center_y`, `place_center_z`,
`force_placement`.
* `place_center_x`: Placement of this decoration is centered along the X axis.
* `place_center_y`: Placement of this decoration is centered along the Y axis.
* `place_center_z`: Placement of this decoration is centered along the Z axis.
* `force_placement`: Schematic nodes other than "ignore" will replace existing
nodes.
Lua Voxel Manipulator
=====================
About VoxelManip
----------------
VoxelManip is a scripting interface to the internal 'Map Voxel Manipulator'
facility. The purpose of this object is for fast, low-level, bulk access to
reading and writing Map content. As such, setting map nodes through VoxelManip
will lack many of the higher level features and concepts you may be used to
with other methods of setting nodes. For example, nodes will not have their
construction and destruction callbacks run, and no rollback information is
logged.
It is important to note that VoxelManip is designed for speed, and *not* ease
of use or flexibility. If your mod requires a map manipulation facility that
will handle 100% of all edge cases, or the use of high level node placement
features, perhaps `minetest.set_node()` is better suited for the job.
In addition, VoxelManip might not be faster, or could even be slower, for your
specific use case. VoxelManip is most effective when setting large areas of map
at once - for example, if only setting a 3x3x3 node area, a
`minetest.set_node()` loop may be more optimal. Always profile code using both
methods of map manipulation to determine which is most appropriate for your
usage.
A recent simple test of setting cubic areas showed that `minetest.set_node()`
is faster than a VoxelManip for a 3x3x3 node cube or smaller.
Using VoxelManip
----------------
A VoxelManip object can be created any time using either:
`VoxelManip([p1, p2])`, or `minetest.get_voxel_manip([p1, p2])`.
If the optional position parameters are present for either of these routines,
the specified region will be pre-loaded into the VoxelManip object on creation.
Otherwise, the area of map you wish to manipulate must first be loaded into the
VoxelManip object using `VoxelManip:read_from_map()`.
Note that `VoxelManip:read_from_map()` returns two position vectors. The region
formed by these positions indicate the minimum and maximum (respectively)
positions of the area actually loaded in the VoxelManip, which may be larger
than the area requested. For convenience, the loaded area coordinates can also
be queried any time after loading map data with `VoxelManip:get_emerged_area()`.
Now that the VoxelManip object is populated with map data, your mod can fetch a
copy of this data using either of two methods. `VoxelManip:get_node_at()`,
which retrieves an individual node in a MapNode formatted table at the position
requested is the simplest method to use, but also the slowest.
Nodes in a VoxelManip object may also be read in bulk to a flat array table
using:
* `VoxelManip:get_data()` for node content (in Content ID form, see section
[Content IDs]),
* `VoxelManip:get_light_data()` for node light levels, and
* `VoxelManip:get_param2_data()` for the node type-dependent "param2" values.
See section [Flat array format] for more details.
It is very important to understand that the tables returned by any of the above
three functions represent a snapshot of the VoxelManip's internal state at the
time of the call. This copy of the data will not magically update itself if
another function modifies the internal VoxelManip state.
Any functions that modify a VoxelManip's contents work on the VoxelManip's
internal state unless otherwise explicitly stated.
Once the bulk data has been edited to your liking, the internal VoxelManip
state can be set using:
* `VoxelManip:set_data()` for node content (in Content ID form, see section
[Content IDs]),
* `VoxelManip:set_light_data()` for node light levels, and
* `VoxelManip:set_param2_data()` for the node type-dependent `param2` values.
The parameter to each of the above three functions can use any table at all in
the same flat array format as produced by `get_data()` etc. and is not required
to be a table retrieved from `get_data()`.
Once the internal VoxelManip state has been modified to your liking, the
changes can be committed back to the map by calling `VoxelManip:write_to_map()`
### Flat array format
Let
`Nx = p2.X - p1.X + 1`,
`Ny = p2.Y - p1.Y + 1`, and
`Nz = p2.Z - p1.Z + 1`.
Then, for a loaded region of p1..p2, this array ranges from `1` up to and
including the value of the expression `Nx * Ny * Nz`.
Positions offset from p1 are present in the array with the format of:
[
(0, 0, 0), (1, 0, 0), (2, 0, 0), ... (Nx, 0, 0),
(0, 1, 0), (1, 1, 0), (2, 1, 0), ... (Nx, 1, 0),
...
(0, Ny, 0), (1, Ny, 0), (2, Ny, 0), ... (Nx, Ny, 0),
(0, 0, 1), (1, 0, 1), (2, 0, 1), ... (Nx, 0, 1),
...
(0, Ny, 2), (1, Ny, 2), (2, Ny, 2), ... (Nx, Ny, 2),
...
(0, Ny, Nz), (1, Ny, Nz), (2, Ny, Nz), ... (Nx, Ny, Nz)
]
and the array index for a position p contained completely in p1..p2 is:
`(p.Z - p1.Z) * Ny * Nx + (p.Y - p1.Y) * Nx + (p.X - p1.X) + 1`
Note that this is the same "flat 3D array" format as
`PerlinNoiseMap:get3dMap_flat()`.
VoxelArea objects (see section [`VoxelArea`]) can be used to simplify calculation
of the index for a single point in a flat VoxelManip array.
### Content IDs
A Content ID is a unique integer identifier for a specific node type.
These IDs are used by VoxelManip in place of the node name string for
`VoxelManip:get_data()` and `VoxelManip:set_data()`. You can use
`minetest.get_content_id()` to look up the Content ID for the specified node
name, and `minetest.get_name_from_content_id()` to look up the node name string
for a given Content ID.
After registration of a node, its Content ID will remain the same throughout
execution of the mod.
Note that the node being queried needs to have already been been registered.
The following builtin node types have their Content IDs defined as constants:
* `minetest.CONTENT_UNKNOWN`: ID for "unknown" nodes
* `minetest.CONTENT_AIR`: ID for "air" nodes
* `minetest.CONTENT_IGNORE`: ID for "ignore" nodes
### Mapgen VoxelManip objects
Inside of `on_generated()` callbacks, it is possible to retrieve the same
VoxelManip object used by the core's Map Generator (commonly abbreviated
Mapgen). Most of the rules previously described still apply but with a few
differences:
* The Mapgen VoxelManip object is retrieved using:
`minetest.get_mapgen_object("voxelmanip")`
* This VoxelManip object already has the region of map just generated loaded
into it; it's not necessary to call `VoxelManip:read_from_map()` before using
a Mapgen VoxelManip.
* The `on_generated()` callbacks of some mods may place individual nodes in the
generated area using non-VoxelManip map modification methods. Because the
same Mapgen VoxelManip object is passed through each `on_generated()`
callback, it becomes necessary for the Mapgen VoxelManip object to maintain
consistency with the current map state. For this reason, calling any of the
following functions:
`minetest.add_node()`, `minetest.set_node()`, or `minetest.swap_node()`
will also update the Mapgen VoxelManip object's internal state active on the
current thread.
* After modifying the Mapgen VoxelManip object's internal buffer, it may be
necessary to update lighting information using either:
`VoxelManip:calc_lighting()` or `VoxelManip:set_lighting()`.
### Other API functions operating on a VoxelManip
If any VoxelManip contents were set to a liquid node,
`VoxelManip:update_liquids()` must be called for these liquid nodes to begin
flowing. It is recommended to call this function only after having written all
buffered data back to the VoxelManip object, save for special situations where
the modder desires to only have certain liquid nodes begin flowing.
The functions `minetest.generate_ores()` and `minetest.generate_decorations()`
will generate all registered decorations and ores throughout the full area
inside of the specified VoxelManip object.
`minetest.place_schematic_on_vmanip()` is otherwise identical to
`minetest.place_schematic()`, except instead of placing the specified schematic
directly on the map at the specified position, it will place the schematic
inside the VoxelManip.
### Notes
* Attempting to read data from a VoxelManip object before map is read will
result in a zero-length array table for `VoxelManip:get_data()`, and an
"ignore" node at any position for `VoxelManip:get_node_at()`.
* If either a region of map has not yet been generated or is out-of-bounds of
the map, that region is filled with "ignore" nodes.
* Other mods, or the core itself, could possibly modify the area of map
currently loaded into a VoxelManip object. With the exception of Mapgen
VoxelManips (see above section), the internal buffers are not updated. For
this reason, it is strongly encouraged to complete the usage of a particular
VoxelManip object in the same callback it had been created.
* If a VoxelManip object will be used often, such as in an `on_generated()`
callback, consider passing a file-scoped table as the optional parameter to
`VoxelManip:get_data()`, which serves as a static buffer the function can use
to write map data to instead of returning a new table each call. This greatly
enhances performance by avoiding unnecessary memory allocations.
Methods
-------
* `read_from_map(p1, p2)`: Loads a chunk of map into the VoxelManip object
containing the region formed by `p1` and `p2`.
* returns actual emerged `pmin`, actual emerged `pmax`
* `write_to_map([light])`: Writes the data loaded from the `VoxelManip` back to
the map.
* **important**: data must be set using `VoxelManip:set_data()` before
calling this.
* if `light` is true, then lighting is automatically recalculated.
The default value is true.
If `light` is false, no light calculations happen, and you should correct
all modified blocks with `minetest.fix_light()` as soon as possible.
Keep in mind that modifying the map where light is incorrect can cause
more lighting bugs.
* `get_node_at(pos)`: Returns a `MapNode` table of the node currently loaded in
the `VoxelManip` at that position
* `set_node_at(pos, node)`: Sets a specific `MapNode` in the `VoxelManip` at
that position.
* `get_data([buffer])`: Retrieves the node content data loaded into the
`VoxelManip` object.
* returns raw node data in the form of an array of node content IDs
* if the param `buffer` is present, this table will be used to store the
result instead.
* `set_data(data)`: Sets the data contents of the `VoxelManip` object
* `update_map()`: Does nothing, kept for compatibility.
* `set_lighting(light, [p1, p2])`: Set the lighting within the `VoxelManip` to
a uniform value.
* `light` is a table, `{day=<0...15>, night=<0...15>}`
* To be used only by a `VoxelManip` object from
`minetest.get_mapgen_object`.
* (`p1`, `p2`) is the area in which lighting is set, defaults to the whole
area if left out.
* `get_light_data()`: Gets the light data read into the `VoxelManip` object
* Returns an array (indices 1 to volume) of integers ranging from `0` to
`255`.
* Each value is the bitwise combination of day and night light values
(`0` to `15` each).
* `light = day + (night * 16)`
* `set_light_data(light_data)`: Sets the `param1` (light) contents of each node
in the `VoxelManip`.
* expects lighting data in the same format that `get_light_data()` returns
* `get_param2_data([buffer])`: Gets the raw `param2` data read into the
`VoxelManip` object.
* Returns an array (indices 1 to volume) of integers ranging from `0` to
`255`.
* If the param `buffer` is present, this table will be used to store the
result instead.
* `set_param2_data(param2_data)`: Sets the `param2` contents of each node in
the `VoxelManip`.
* `calc_lighting([p1, p2], [propagate_shadow])`: Calculate lighting within the
`VoxelManip`.
* To be used only by a `VoxelManip` object from
`minetest.get_mapgen_object`.
* (`p1`, `p2`) is the area in which lighting is set, defaults to the whole
area if left out or nil.
* `propagate_shadow` is an optional boolean deciding whether shadows in a
generated mapchunk above are propagated down into the mapchunk, defaults
to `true` if left out.
* `update_liquids()`: Update liquid flow
* `was_modified()`: Returns `true` or `false` if the data in the voxel
manipulator had been modified since the last read from map, due to a call to
`minetest.set_data()` on the loaded area elsewhere.
* `get_emerged_area()`: Returns actual emerged minimum and maximum positions.
`VoxelArea`
-----------
A helper class for voxel areas.
It can be created via `VoxelArea:new{MinEdge=pmin, MaxEdge=pmax}`.
The coordinates are *inclusive*, like most other things in Minetest.
### Methods
* `getExtent()`: returns a 3D vector containing the size of the area formed by
`MinEdge` and `MaxEdge`.
* `getVolume()`: returns the volume of the area formed by `MinEdge` and
`MaxEdge`.
* `index(x, y, z)`: returns the index of an absolute position in a flat array
starting at `1`.
* `x`, `y` and `z` must be integers to avoid an incorrect index result.
* The position (x, y, z) is not checked for being inside the area volume,
being outside can cause an incorrect index result.
* Useful for things like `VoxelManip`, raw Schematic specifiers,
`PerlinNoiseMap:get2d`/`3dMap`, and so on.
* `indexp(p)`: same functionality as `index(x, y, z)` but takes a vector.
* As with `index(x, y, z)`, the components of `p` must be integers, and `p`
is not checked for being inside the area volume.
* `position(i)`: returns the absolute position vector corresponding to index
`i`.
* `contains(x, y, z)`: check if (`x`,`y`,`z`) is inside area formed by
`MinEdge` and `MaxEdge`.
* `containsp(p)`: same as above, except takes a vector
* `containsi(i)`: same as above, except takes an index `i`
* `iter(minx, miny, minz, maxx, maxy, maxz)`: returns an iterator that returns
indices.
* from (`minx`,`miny`,`minz`) to (`maxx`,`maxy`,`maxz`) in the order of
`[z [y [x]]]`.
* `iterp(minp, maxp)`: same as above, except takes a vector
Mapgen objects
==============
A mapgen object is a construct used in map generation. Mapgen objects can be
used by an `on_generate` callback to speed up operations by avoiding
unnecessary recalculations, these can be retrieved using the
`minetest.get_mapgen_object()` function. If the requested Mapgen object is
unavailable, or `get_mapgen_object()` was called outside of an `on_generate()`
callback, `nil` is returned.
The following Mapgen objects are currently available:
### `voxelmanip`
This returns three values; the `VoxelManip` object to be used, minimum and
maximum emerged position, in that order. All mapgens support this object.
### `heightmap`
Returns an array containing the y coordinates of the ground levels of nodes in
the most recently generated chunk by the current mapgen.
### `biomemap`
Returns an array containing the biome IDs of nodes in the most recently
generated chunk by the current mapgen.
### `heatmap`
Returns an array containing the temperature values of nodes in the most
recently generated chunk by the current mapgen.
### `humiditymap`
Returns an array containing the humidity values of nodes in the most recently
generated chunk by the current mapgen.
### `gennotify`
Returns a table mapping requested generation notification types to arrays of
positions at which the corresponding generated structures are located within
the current chunk. To set the capture of positions of interest to be recorded
on generate, use `minetest.set_gen_notify()`.
For decorations, the returned positions are the ground surface 'place_on'
nodes, not the decorations themselves. A 'simple' type decoration is often 1
node above the returned position and possibly displaced by 'place_offset_y'.
Possible fields of the table returned are:
* `dungeon`
* `temple`
* `cave_begin`
* `cave_end`
* `large_cave_begin`
* `large_cave_end`
* `decoration`
Decorations have a key in the format of `"decoration#id"`, where `id` is the
numeric unique decoration ID.
Registered entities
===================
Functions receive a "luaentity" as `self`:
* It has the member `.name`, which is the registered name `("mod:thing")`
* It has the member `.object`, which is an `ObjectRef` pointing to the object
* The original prototype stuff is visible directly via a metatable
Callbacks:
* `on_activate(self, staticdata, dtime_s)`
* Called when the object is instantiated.
* `dtime_s` is the time passed since the object was unloaded, which can be
used for updating the entity state.
* `on_step(self, dtime)`
* Called on every server tick, after movement and collision processing.
`dtime` is usually 0.1 seconds, as per the `dedicated_server_step` setting
in `minetest.conf`.
* `on_punch(self, puncher, time_from_last_punch, tool_capabilities, dir)`
* Called when somebody punches the object.
* Note that you probably want to handle most punches using the automatic
armor group system.
* `puncher`: an `ObjectRef` (can be `nil`)
* `time_from_last_punch`: Meant for disallowing spamming of clicks
(can be `nil`).
* `tool_capabilities`: capability table of used tool (can be `nil`)
* `dir`: unit vector of direction of punch. Always defined. Points from the
puncher to the punched.
* `on_death(self, killer)`
* Called when the object dies.
* `killer`: an `ObjectRef` (can be `nil`)
* `on_rightclick(self, clicker)`
* `on_attach_child(self, child)`
* `child`: an `ObjectRef` of the child that attaches
* `on_detach_child(self, child)`
* `child`: an `ObjectRef` of the child that detaches
* `on_detach(self, parent)`
* `parent`: an `ObjectRef` (can be `nil`) from where it got detached
* This happens before the parent object is removed from the world
* `get_staticdata(self)`
* Should return a string that will be passed to `on_activate` when the
object is instantiated the next time.
L-system trees
==============
Tree definition
---------------
treedef={
axiom, --string initial tree axiom
rules_a, --string rules set A
rules_b, --string rules set B
rules_c, --string rules set C
rules_d, --string rules set D
trunk, --string trunk node name
leaves, --string leaves node name
leaves2, --string secondary leaves node name
leaves2_chance,--num chance (0-100) to replace leaves with leaves2
angle, --num angle in deg
iterations, --num max # of iterations, usually 2 -5
random_level, --num factor to lower nr of iterations, usually 0 - 3
trunk_type, --string single/double/crossed) type of trunk: 1 node,
-- 2x2 nodes or 3x3 in cross shape
thin_branches, --boolean true -> use thin (1 node) branches
fruit, --string fruit node name
fruit_chance, --num chance (0-100) to replace leaves with fruit node
seed, --num random seed, if no seed is provided, the engine
will create one.
}
Key for special L-System symbols used in axioms
-----------------------------------------------
* `G`: move forward one unit with the pen up
* `F`: move forward one unit with the pen down drawing trunks and branches
* `f`: move forward one unit with the pen down drawing leaves (100% chance)
* `T`: move forward one unit with the pen down drawing trunks only
* `R`: move forward one unit with the pen down placing fruit
* `A`: replace with rules set A
* `B`: replace with rules set B
* `C`: replace with rules set C
* `D`: replace with rules set D
* `a`: replace with rules set A, chance 90%
* `b`: replace with rules set B, chance 80%
* `c`: replace with rules set C, chance 70%
* `d`: replace with rules set D, chance 60%
* `+`: yaw the turtle right by `angle` parameter
* `-`: yaw the turtle left by `angle` parameter
* `&`: pitch the turtle down by `angle` parameter
* `^`: pitch the turtle up by `angle` parameter
* `/`: roll the turtle to the right by `angle` parameter
* `*`: roll the turtle to the left by `angle` parameter
* `[`: save in stack current state info
* `]`: recover from stack state info
Example
-------
Spawn a small apple tree:
pos = {x=230,y=20,z=4}
apple_tree={
axiom="FFFFFAFFBF",
rules_a="[&&&FFFFF&&FFFF][&&&++++FFFFF&&FFFF][&&&----FFFFF&&FFFF]",
rules_b="[&&&++FFFFF&&FFFF][&&&--FFFFF&&FFFF][&&&------FFFFF&&FFFF]",
trunk="default:tree",
leaves="default:leaves",
angle=30,
iterations=2,
random_level=0,
trunk_type="single",
thin_branches=true,
fruit_chance=10,
fruit="default:apple"
}
minetest.spawn_tree(pos,apple_tree)
'minetest' namespace reference
==============================
Utilities
---------
* `minetest.get_current_modname()`: returns the currently loading mod's name,
when loading a mod.
* `minetest.get_modpath(modname)`: returns e.g.
`"/home/user/.minetest/usermods/modname"`.
* Useful for loading additional `.lua` modules or static data from mod
* `minetest.get_modnames()`: returns a list of installed mods
* Return a list of installed mods, sorted alphabetically
* `minetest.get_worldpath()`: returns e.g. `"/home/user/.minetest/world"`
* Useful for storing custom data
* `minetest.is_singleplayer()`
* `minetest.features`: Table containing API feature flags
{
glasslike_framed = true,
nodebox_as_selectionbox = true,
chat_send_player_param3 = true,
get_all_craft_recipes_works = true,
-- The transparency channel of textures can optionally be used on
-- nodes
use_texture_alpha = true,
-- Tree and grass ABMs are no longer done from C++
no_legacy_abms = true,
-- Texture grouping is possible using parentheses
texture_names_parens = true,
-- Unique Area ID for AreaStore:insert_area
area_store_custom_ids = true,
-- add_entity supports passing initial staticdata to on_activate
add_entity_with_staticdata = true,
-- Chat messages are no longer predicted
no_chat_message_prediction = true,
-- The transparency channel of textures can optionally be used on
-- objects (ie: players and lua entities)
object_use_texture_alpha = true,
-- Object selectionbox is settable independently from collisionbox
object_independent_selectionbox = true,
}
* `minetest.has_feature(arg)`: returns `boolean, missing_features`
* `arg`: string or table in format `{foo=true, bar=true}`
* `missing_features`: `{foo=true, bar=true}`
* `minetest.get_player_information(player_name)`: Table containing information
about a player. Example return value:
{
address = "127.0.0.1", -- IP address of client
ip_version = 4, -- IPv4 / IPv6
min_rtt = 0.01, -- minimum round trip time
max_rtt = 0.2, -- maximum round trip time
avg_rtt = 0.02, -- average round trip time
min_jitter = 0.01, -- minimum packet time jitter
max_jitter = 0.5, -- maximum packet time jitter
avg_jitter = 0.03, -- average packet time jitter
connection_uptime = 200, -- seconds since client connected
protocol_version = 32, -- protocol version used by client
-- following information is available on debug build only!!!
-- DO NOT USE IN MODS
--ser_vers = 26, -- serialization version used by client
--major = 0, -- major version number
--minor = 4, -- minor version number
--patch = 10, -- patch version number
--vers_string = "0.4.9-git", -- full version string
--state = "Active" -- current client state
}
* `minetest.mkdir(path)`: returns success.
* Creates a directory specified by `path`, creating parent directories
if they don't exist.
* `minetest.get_dir_list(path, [is_dir])`: returns list of entry names
* is_dir is one of:
* nil: return all entries,
* true: return only subdirectory names, or
* false: return only file names.
* `minetest.safe_file_write(path, content)`: returns boolean indicating success
* Replaces contents of file at path with new contents in a safe (atomic)
way. Use this instead of below code when writing e.g. database files:
`local f = io.open(path, "wb"); f:write(content); f:close()`
* `minetest.get_version()`: returns a table containing components of the
engine version. Components:
* `project`: Name of the project, eg, "Minetest"
* `string`: Simple version, eg, "1.2.3-dev"
* `hash`: Full git version (only set if available),
eg, "1.2.3-dev-01234567-dirty".
Use this for informational purposes only. The information in the returned
table does not represent the capabilities of the engine, nor is it
reliable or verifiable. Compatible forks will have a different name and
version entirely. To check for the presence of engine features, test
whether the functions exported by the wanted features exist. For example:
`if minetest.check_for_falling then ... end`.
* `minetest.sha1(data, [raw])`: returns the sha1 hash of data
* `data`: string of data to hash
* `raw`: return raw bytes instead of hex digits, default: false
Logging
-------
* `minetest.debug(...)`
* Equivalent to `minetest.log(table.concat({...}, "\t"))`
* `minetest.log([level,] text)`
* `level` is one of `"none"`, `"error"`, `"warning"`, `"action"`,
`"info"`, or `"verbose"`. Default is `"none"`.
Registration functions
----------------------
Call these functions only at load time!
* `minetest.register_entity(name, entity definition)`
* `minetest.register_abm(abm definition)`
* `minetest.register_lbm(lbm definition)`
* `minetest.register_node(name, node definition)`
* `minetest.register_tool(name, item definition)`
* `minetest.register_craftitem(name, item definition)`
* `minetest.unregister_item(name)`
* `minetest.register_alias(name, convert_to)`
* Also use this to set the 'mapgen aliases' needed in a game for the core
* mapgens. See [Mapgen aliases] section above.
* `minetest.register_alias_force(name, convert_to)`
* `minetest.register_craft(recipe)`
* Check recipe table syntax for different types below.
* `minetest.clear_craft(recipe)`
* Will erase existing craft based either on output item or on input recipe.
* Specify either output or input only. If you specify both, input will be
ignored. For input use the same recipe table syntax as for
`minetest.register_craft(recipe)`. For output specify only the item,
without a quantity.
* If no erase candidate could be found, Lua exception will be thrown.
* **Warning**! The type field ("shaped", "cooking" or any other) will be
ignored if the recipe contains output. Erasing is then done independently
from the crafting method.
* `minetest.register_ore(ore definition)`
* `minetest.register_biome(biome definition)`
* `minetest.register_decoration(decoration definition)`
* `minetest.override_item(name, redefinition)`
* Overrides fields of an item registered with register_node/tool/craftitem.
* Note: Item must already be defined, (opt)depend on the mod defining it.
* Example: `minetest.override_item("default:mese", {light_source=LIGHT_MAX})`
* `minetest.clear_registered_ores()`
* `minetest.clear_registered_biomes()`
* `minetest.clear_registered_decorations()`
Global callback registration functions
--------------------------------------
Call these functions only at load time!
* `minetest.register_globalstep(func(dtime))`
* Called every server step, usually interval of 0.1s
* `minetest.register_on_mods_loaded(func())`
* Called after mods have finished loading and before the media is cached or the
aliases handled.
* `minetest.register_on_shutdown(func())`
* Called before server shutdown
* **Warning**: If the server terminates abnormally (i.e. crashes), the
registered callbacks **will likely not be run**. Data should be saved at
semi-frequent intervals as well as on server shutdown.
* `minetest.register_on_placenode(func(pos, newnode, placer, oldnode, itemstack, pointed_thing))`
* Called when a node has been placed
* If return `true` no item is taken from `itemstack`
* `placer` may be any valid ObjectRef or nil.
* **Not recommended**; use `on_construct` or `after_place_node` in node
definition whenever possible.
* `minetest.register_on_dignode(func(pos, oldnode, digger))`
* Called when a node has been dug.
* **Not recommended**; Use `on_destruct` or `after_dig_node` in node
definition whenever possible.
* `minetest.register_on_punchnode(func(pos, node, puncher, pointed_thing))`
* Called when a node is punched
* `minetest.register_on_generated(func(minp, maxp, blockseed))`
* Called after generating a piece of world. Modifying nodes inside the area
is a bit faster than usually.
* `minetest.register_on_newplayer(func(ObjectRef))`
* Called after a new player has been created
* `minetest.register_on_punchplayer(func(player, hitter, time_from_last_punch, tool_capabilities, dir, damage))`
* Called when a player is punched
* `player`: ObjectRef - Player that was punched
* `hitter`: ObjectRef - Player that hit
* `time_from_last_punch`: Meant for disallowing spamming of clicks
(can be nil).
* `tool_capabilities`: Capability table of used tool (can be nil)
* `dir`: Unit vector of direction of punch. Always defined. Points from
the puncher to the punched.
* `damage`: Number that represents the damage calculated by the engine
* should return `true` to prevent the default damage mechanism
* `minetest.register_on_player_hpchange(func(player, hp_change, reason), modifier)`
* Called when the player gets damaged or healed
* `player`: ObjectRef of the player
* `hp_change`: the amount of change. Negative when it is damage.
* `reason`: a PlayerHPChangeReason table.
* The `type` field will have one of the following values:
* `set_hp`: A mod or the engine called `set_hp` without
giving a type - use this for custom damage types.
* `punch`: Was punched. `reason.object` will hold the puncher, or nil if none.
* `fall`
* `node_damage`: damage_per_second from a neighbouring node.
* `drown`
* `respawn`
* Any of the above types may have additional fields from mods.
* `reason.from` will be `mod` or `engine`.
* `modifier`: when true, the function should return the actual `hp_change`.
Note: modifiers only get a temporary hp_change that can be modified by later modifiers.
modifiers can return true as a second argument to stop the execution of further functions.
Non-modifiers receive the final hp change calculated by the modifiers.
* `minetest.register_on_dieplayer(func(ObjectRef, reason))`
* Called when a player dies
* `reason`: a PlayerHPChangeReason table, see register_on_player_hpchange
* `minetest.register_on_respawnplayer(func(ObjectRef))`
* Called when player is to be respawned
* Called _before_ repositioning of player occurs
* return true in func to disable regular player placement
* `minetest.register_on_prejoinplayer(func(name, ip))`
* Called before a player joins the game
* If it returns a string, the player is disconnected with that string as
reason.
* `minetest.register_on_joinplayer(func(ObjectRef))`
* Called when a player joins the game
* `minetest.register_on_leaveplayer(func(ObjectRef, timed_out))`
* Called when a player leaves the game
* `timed_out`: True for timeout, false for other reasons.
* `minetest.register_on_auth_fail(func(name, ip))`
* Called when a client attempts to log into an account but supplies the
wrong password.
* `ip`: The IP address of the client.
* `name`: The account the client attempted to log into.
* `minetest.register_on_cheat(func(ObjectRef, cheat))`
* Called when a player cheats
* `cheat`: `{type=<cheat_type>}`, where `<cheat_type>` is one of:
* `moved_too_fast`
* `interacted_too_far`
* `interacted_while_dead`
* `finished_unknown_dig`
* `dug_unbreakable`
* `dug_too_fast`
* `minetest.register_on_chat_message(func(name, message))`
* Called always when a player says something
* Return `true` to mark the message as handled, which means that it will
not be sent to other players.
* `minetest.register_on_player_receive_fields(func(player, formname, fields))`
* Called when a button is pressed in player's inventory form
* Newest functions are called first
* If function returns `true`, remaining functions are not called
* `minetest.register_on_craft(func(itemstack, player, old_craft_grid, craft_inv))`
* Called when `player` crafts something
* `itemstack` is the output
* `old_craft_grid` contains the recipe (Note: the one in the inventory is
cleared).
* `craft_inv` is the inventory with the crafting grid
* Return either an `ItemStack`, to replace the output, or `nil`, to not
modify it.
* `minetest.register_craft_predict(func(itemstack, player, old_craft_grid, craft_inv))`
* The same as before, except that it is called before the player crafts, to
make craft prediction, and it should not change anything.
* `minetest.register_allow_player_inventory_action(func(player, inventory, action, inventory_info))`
* Determinates how much of a stack may be taken, put or moved to a
player inventory.
* `player` (type `ObjectRef`) is the player who modified the inventory
`inventory` (type `InvRef`).
* List of possible `action` (string) values and their
`inventory_info` (table) contents:
* `move`: `{from_list=string, to_list=string, from_index=number, to_index=number, count=number}`
* `put`: `{listname=string, index=number, stack=ItemStack}`
* `take`: Same as `put`
* Return a numeric value to limit the amount of items to be taken, put or
moved. A value of `-1` for `take` will make the source stack infinite.
* `minetest.register_on_player_inventory_action(func(player, inventory, action, inventory_info))`
* Called after a take, put or move event from/to/in a player inventory
* Function arguments: see `minetest.register_allow_player_inventory_action`
* Does not accept or handle any return value.
* `minetest.register_on_protection_violation(func(pos, name))`
* Called by `builtin` and mods when a player violates protection at a
position (eg, digs a node or punches a protected entity).
* The registered functions can be called using
`minetest.record_protection_violation`.
* The provided function should check that the position is protected by the
mod calling this function before it prints a message, if it does, to
allow for multiple protection mods.
* `minetest.register_on_item_eat(func(hp_change, replace_with_item, itemstack, user, pointed_thing))`
* Called when an item is eaten, by `minetest.item_eat`
* Return `true` or `itemstack` to cancel the default item eat response
(i.e.: hp increase).
* `minetest.register_on_priv_grant(function(name, granter, priv))`
* Called when `granter` grants the priv `priv` to `name`.
* Note that the callback will be called twice if it's done by a player,
once with granter being the player name, and again with granter being nil.
* `minetest.register_on_priv_revoke(function(name, revoker, priv))`
* Called when `revoker` revokes the priv `priv` from `name`.
* Note that the callback will be called twice if it's done by a player,
once with revoker being the player name, and again with revoker being nil.
* `minetest.register_can_bypass_userlimit(function(name, ip))`
* Called when `name` user connects with `ip`.
* Return `true` to by pass the player limit
* `minetest.register_on_modchannel_message(func(channel_name, sender, message))`
* Called when an incoming mod channel message is received
* You should have joined some channels to receive events.
* If message comes from a server mod, `sender` field is an empty string.
Other registration functions
----------------------------
* `minetest.register_chatcommand(cmd, chatcommand definition)`
* Adds definition to `minetest.registered_chatcommands`
* `minetest.override_chatcommand(name, redefinition)`
* Overrides fields of a chatcommand registered with `register_chatcommand`.
* `minetest.unregister_chatcommand(name)`
* Unregisters a chatcommands registered with `register_chatcommand`.
* `minetest.register_privilege(name, definition)`
* `definition`: `"description text"`
* `definition`:
`{description = "description text", give_to_singleplayer = boolean}`
the default of `give_to_singleplayer` is true.
* To allow players with `basic_privs` to grant, see `basic_privs`
minetest.conf setting.
* `on_grant(name, granter_name)`: Called when given to player `name` by
`granter_name`.
`granter_name` will be nil if the priv was granted by a mod.
* `on_revoke(name, revoker_name)`: Called when taken from player `name` by
`revoker_name`.
`revoker_name` will be nil if the priv was revoked by a mod
* Note that the above two callbacks will be called twice if a player is
responsible, once with the player name, and then with a nil player name.
* Return true in the above callbacks to stop register_on_priv_grant or
revoke being called.
* `minetest.register_authentication_handler(authentication handler definition)`
* Registers an auth handler that overrides the builtin one
* This function can be called by a single mod once only.
Setting-related
---------------
* `minetest.settings`: Settings object containing all of the settings from the
main config file (`minetest.conf`).
* `minetest.setting_get_pos(name)`: Loads a setting from the main settings and
parses it as a position (in the format `(1,2,3)`). Returns a position or nil.
Authentication
--------------
* `minetest.string_to_privs(str)`: returns `{priv1=true,...}`
* `minetest.privs_to_string(privs)`: returns `"priv1,priv2,..."`
* Convert between two privilege representations
* `minetest.get_player_privs(name) -> {priv1=true,...}`
* `minetest.check_player_privs(player_or_name, ...)`:
returns `bool, missing_privs`
* A quickhand for checking privileges.
* `player_or_name`: Either a Player object or the name of a player.
* `...` is either a list of strings, e.g. `"priva", "privb"` or
a table, e.g. `{ priva = true, privb = true }`.
* `minetest.check_password_entry(name, entry, password)`
* Returns true if the "password entry" for a player with name matches given
password, false otherwise.
* The "password entry" is the password representation generated by the
engine as returned as part of a `get_auth()` call on the auth handler.
* Only use this function for making it possible to log in via password from
external protocols such as IRC, other uses are frowned upon.
* `minetest.get_password_hash(name, raw_password)`
* Convert a name-password pair to a password hash that Minetest can use.
* The returned value alone is not a good basis for password checks based
on comparing the password hash in the database with the password hash
from the function, with an externally provided password, as the hash
in the db might use the new SRP verifier format.
* For this purpose, use `minetest.check_password_entry` instead.
* `minetest.get_player_ip(name)`: returns an IP address string for the player
`name`.
* The player needs to be online for this to be successful.
* `minetest.get_auth_handler()`: Return the currently active auth handler
* See the [Authentication handler definition]
* Use this to e.g. get the authentication data for a player:
`local auth_data = minetest.get_auth_handler().get_auth(playername)`
* `minetest.notify_authentication_modified(name)`
* Must be called by the authentication handler for privilege changes.
* `name`: string; if omitted, all auth data should be considered modified
* `minetest.set_player_password(name, password_hash)`: Set password hash of
player `name`.
* `minetest.set_player_privs(name, {priv1=true,...})`: Set privileges of player
`name`.
* `minetest.auth_reload()`
* See `reload()` in authentication handler definition
`minetest.set_player_password`, `minetest_set_player_privs`,
`minetest_get_player_privs` and `minetest.auth_reload` call the authentication
handler.
Chat
----
* `minetest.chat_send_all(text)`
* `minetest.chat_send_player(name, text)`
Environment access
------------------
* `minetest.set_node(pos, node)`
* `minetest.add_node(pos, node)`: alias to `minetest.set_node`
* Set node at position `pos`
* `node`: table `{name=string, param1=number, param2=number}`
* If param1 or param2 is omitted, it's set to `0`.
* e.g. `minetest.set_node({x=0, y=10, z=0}, {name="default:wood"})`
* `minetest.bulk_set_node({pos1, pos2, pos3, ...}, node)`
* Set node on all positions set in the first argument.
* e.g. `minetest.bulk_set_node({{x=0, y=1, z=1}, {x=1, y=2, z=2}}, {name="default:stone"})`
* For node specification or position syntax see `minetest.set_node` call
* Faster than set_node due to single call, but still considerably slower
than Lua Voxel Manipulators (LVM) for large numbers of nodes.
Unlike LVMs, this will call node callbacks. It also allows setting nodes
in spread out positions which would cause LVMs to waste memory.
For setting a cube, this is 1.3x faster than set_node whereas LVM is 20
times faster.
* `minetest.swap_node(pos, node)`
* Set node at position, but don't remove metadata
* `minetest.remove_node(pos)`
* By default it does the same as `minetest.set_node(pos, {name="air"})`
* `minetest.get_node(pos)`
* Returns the node at the given position as table in the format
`{name="node_name", param1=0, param2=0}`,
returns `{name="ignore", param1=0, param2=0}` for unloaded areas.
* `minetest.get_node_or_nil(pos)`
* Same as `get_node` but returns `nil` for unloaded areas.
* `minetest.get_node_light(pos, timeofday)`
* Gets the light value at the given position. Note that the light value
"inside" the node at the given position is returned, so you usually want
to get the light value of a neighbor.
* `pos`: The position where to measure the light.
* `timeofday`: `nil` for current time, `0` for night, `0.5` for day
* Returns a number between `0` and `15` or `nil`
* `minetest.place_node(pos, node)`
* Place node with the same effects that a player would cause
* `minetest.dig_node(pos)`
* Dig node with the same effects that a player would cause
* Returns `true` if successful, `false` on failure (e.g. protected location)
* `minetest.punch_node(pos)`
* Punch node with the same effects that a player would cause
* `minetest.spawn_falling_node(pos)`
* Change node into falling node
* Returns `true` if successful, `false` on failure
* `minetest.find_nodes_with_meta(pos1, pos2)`
* Get a table of positions of nodes that have metadata within a region
{pos1, pos2}.
* `minetest.get_meta(pos)`
* Get a `NodeMetaRef` at that position
* `minetest.get_node_timer(pos)`
* Get `NodeTimerRef`
* `minetest.add_entity(pos, name, [staticdata])`: Spawn Lua-defined entity at
position.
* Returns `ObjectRef`, or `nil` if failed
* `minetest.add_item(pos, item)`: Spawn item
* Returns `ObjectRef`, or `nil` if failed
* `minetest.get_player_by_name(name)`: Get an `ObjectRef` to a player
* `minetest.get_objects_inside_radius(pos, radius)`: returns a list of
ObjectRefs.
* `radius`: using an euclidean metric
* `minetest.set_timeofday(val)`
* `val` is between `0` and `1`; `0` for midnight, `0.5` for midday
* `minetest.get_timeofday()`
* `minetest.get_gametime()`: returns the time, in seconds, since the world was
created.
* `minetest.get_day_count()`: returns number days elapsed since world was
created.
* accounts for time changes.
* `minetest.find_node_near(pos, radius, nodenames, [search_center])`: returns
pos or `nil`.
* `radius`: using a maximum metric
* `nodenames`: e.g. `{"ignore", "group:tree"}` or `"default:dirt"`
* `search_center` is an optional boolean (default: `false`)
If true `pos` is also checked for the nodes
* `minetest.find_nodes_in_area(pos1, pos2, nodenames)`: returns a list of
positions.
* `nodenames`: e.g. `{"ignore", "group:tree"}` or `"default:dirt"`
* First return value: Table with all node positions
* Second return value: Table with the count of each node with the node name
as index.
* Area volume is limited to 4,096,000 nodes
* `minetest.find_nodes_in_area_under_air(pos1, pos2, nodenames)`: returns a
list of positions.
* `nodenames`: e.g. `{"ignore", "group:tree"}` or `"default:dirt"`
* Return value: Table with all node positions with a node air above
* Area volume is limited to 4,096,000 nodes
* `minetest.get_perlin(noiseparams)`
* `minetest.get_perlin(seeddiff, octaves, persistence, scale)`
* Return world-specific perlin noise (`int(worldseed)+seeddiff`)
* `minetest.get_voxel_manip([pos1, pos2])`
* Return voxel manipulator object.
* Loads the manipulator from the map if positions are passed.
* `minetest.set_gen_notify(flags, {deco_ids})`
* Set the types of on-generate notifications that should be collected.
* `flags` is a flag field with the available flags:
* dungeon
* temple
* cave_begin
* cave_end
* large_cave_begin
* large_cave_end
* decoration
* The second parameter is a list of IDs of decorations which notification
is requested for.
* `minetest.get_gen_notify()`
* Returns a flagstring and a table with the `deco_id`s.
* `minetest.get_decoration_id(decoration_name)`
* Returns the decoration ID number for the provided decoration name string,
or `nil` on failure.
* `minetest.get_mapgen_object(objectname)`
* Return requested mapgen object if available (see [Mapgen objects])
* `minetest.get_heat(pos)`
* Returns the heat at the position, or `nil` on failure.
* `minetest.get_humidity(pos)`
* Returns the humidity at the position, or `nil` on failure.
* `minetest.get_biome_data(pos)`
* Returns a table containing:
* `biome` the biome id of the biome at that position
* `heat` the heat at the position
* `humidity` the humidity at the position
* Or returns `nil` on failure.
* `minetest.get_biome_id(biome_name)`
* Returns the biome id, as used in the biomemap Mapgen object and returned
by `minetest.get_biome_data(pos)`, for a given biome_name string.
* `minetest.get_biome_name(biome_id)`
* Returns the biome name string for the provided biome id, or `nil` on
failure.
* If no biomes have been registered, such as in mgv6, returns `default`.
* `minetest.get_mapgen_params()`
* Deprecated: use `minetest.get_mapgen_setting(name)` instead.
* Returns a table containing:
* `mgname`
* `seed`
* `chunksize`
* `water_level`
* `flags`
* `minetest.set_mapgen_params(MapgenParams)`
* Deprecated: use `minetest.set_mapgen_setting(name, value, override)`
instead.
* Set map generation parameters.
* Function cannot be called after the registration period; only
initialization and `on_mapgen_init`.
* Takes a table as an argument with the fields:
* `mgname`
* `seed`
* `chunksize`
* `water_level`
* `flags`
* Leave field unset to leave that parameter unchanged.
* `flags` contains a comma-delimited string of flags to set, or if the
prefix `"no"` is attached, clears instead.
* `flags` is in the same format and has the same options as `mg_flags` in
`minetest.conf`.
* `minetest.get_mapgen_setting(name)`
* Gets the *active* mapgen setting (or nil if none exists) in string
format with the following order of precedence:
1) Settings loaded from map_meta.txt or overrides set during mod
execution.
2) Settings set by mods without a metafile override
3) Settings explicitly set in the user config file, minetest.conf
4) Settings set as the user config default
* `minetest.get_mapgen_setting_noiseparams(name)`
* Same as above, but returns the value as a NoiseParams table if the
setting `name` exists and is a valid NoiseParams.
* `minetest.set_mapgen_setting(name, value, [override_meta])`
* Sets a mapgen param to `value`, and will take effect if the corresponding
mapgen setting is not already present in map_meta.txt.
* `override_meta` is an optional boolean (default: `false`). If this is set
to true, the setting will become the active setting regardless of the map
metafile contents.
* Note: to set the seed, use `"seed"`, not `"fixed_map_seed"`.
* `minetest.set_mapgen_setting_noiseparams(name, value, [override_meta])`
* Same as above, except value is a NoiseParams table.
* `minetest.set_noiseparams(name, noiseparams, set_default)`
* Sets the noiseparams setting of `name` to the noiseparams table specified
in `noiseparams`.
* `set_default` is an optional boolean (default: `true`) that specifies
whether the setting should be applied to the default config or current
active config.
* `minetest.get_noiseparams(name)`
* Returns a table of the noiseparams for name.
* `minetest.generate_ores(vm, pos1, pos2)`
* Generate all registered ores within the VoxelManip `vm` and in the area
from `pos1` to `pos2`.
* `pos1` and `pos2` are optional and default to mapchunk minp and maxp.
* `minetest.generate_decorations(vm, pos1, pos2)`
* Generate all registered decorations within the VoxelManip `vm` and in the
area from `pos1` to `pos2`.
* `pos1` and `pos2` are optional and default to mapchunk minp and maxp.
* `minetest.clear_objects([options])`
* Clear all objects in the environment
* Takes an optional table as an argument with the field `mode`.
* mode = `"full"` : Load and go through every mapblock, clearing
objects (default).
* mode = `"quick"`: Clear objects immediately in loaded mapblocks,
clear objects in unloaded mapblocks only when the
mapblocks are next activated.
* `minetest.emerge_area(pos1, pos2, [callback], [param])`
* Queue all blocks in the area from `pos1` to `pos2`, inclusive, to be
asynchronously fetched from memory, loaded from disk, or if inexistent,
generates them.
* If `callback` is a valid Lua function, this will be called for each block
emerged.
* The function signature of callback is:
`function EmergeAreaCallback(blockpos, action, calls_remaining, param)`
* `blockpos` is the *block* coordinates of the block that had been
emerged.
* `action` could be one of the following constant values:
* `minetest.EMERGE_CANCELLED`
* `minetest.EMERGE_ERRORED`
* `minetest.EMERGE_FROM_MEMORY`
* `minetest.EMERGE_FROM_DISK`
* `minetest.EMERGE_GENERATED`
* `calls_remaining` is the number of callbacks to be expected after
this one.
* `param` is the user-defined parameter passed to emerge_area (or
nil if the parameter was absent).
* `minetest.delete_area(pos1, pos2)`
* delete all mapblocks in the area from pos1 to pos2, inclusive
* `minetest.line_of_sight(pos1, pos2)`: returns `boolean, pos`
* Checks if there is anything other than air between pos1 and pos2.
* Returns false if something is blocking the sight.
* Returns the position of the blocking node when `false`
* `pos1`: First position
* `pos2`: Second position
* `minetest.raycast(pos1, pos2, objects, liquids)`: returns `Raycast`
* Creates a `Raycast` object.
* `pos1`: start of the ray
* `pos2`: end of the ray
* `objects`: if false, only nodes will be returned. Default is `true`.
* `liquids`: if false, liquid nodes won't be returned. Default is `false`.
* `minetest.find_path(pos1,pos2,searchdistance,max_jump,max_drop,algorithm)`
* returns table containing path
* returns a table of 3D points representing a path from `pos1` to `pos2` or
`nil`.
* `pos1`: start position
* `pos2`: end position
* `searchdistance`: number of blocks to search in each direction using a
maximum metric.
* `max_jump`: maximum height difference to consider walkable
* `max_drop`: maximum height difference to consider droppable
* `algorithm`: One of `"A*_noprefetch"` (default), `"A*"`, `"Dijkstra"`
* `minetest.spawn_tree (pos, {treedef})`
* spawns L-system tree at given `pos` with definition in `treedef` table
* `minetest.transforming_liquid_add(pos)`
* add node to liquid update queue
* `minetest.get_node_max_level(pos)`
* get max available level for leveled node
* `minetest.get_node_level(pos)`
* get level of leveled node (water, snow)
* `minetest.set_node_level(pos, level)`
* set level of leveled node, default `level` equals `1`
* if `totallevel > maxlevel`, returns rest (`total-max`).
* `minetest.add_node_level(pos, level)`
* increase level of leveled node by level, default `level` equals `1`
* if `totallevel > maxlevel`, returns rest (`total-max`)
* can be negative for decreasing
* `minetest.fix_light(pos1, pos2)`: returns `true`/`false`
* resets the light in a cuboid-shaped part of
the map and removes lighting bugs.
* Loads the area if it is not loaded.
* `pos1` is the corner of the cuboid with the least coordinates
(in node coordinates), inclusive.
* `pos2` is the opposite corner of the cuboid, inclusive.
* The actual updated cuboid might be larger than the specified one,
because only whole map blocks can be updated.
The actual updated area consists of those map blocks that intersect
with the given cuboid.
* However, the neighborhood of the updated area might change
as well, as light can spread out of the cuboid, also light
might be removed.
* returns `false` if the area is not fully generated,
`true` otherwise
* `minetest.check_single_for_falling(pos)`
* causes an unsupported `group:falling_node` node to fall and causes an
unattached `group:attached_node` node to fall.
* does not spread these updates to neighbours.
* `minetest.check_for_falling(pos)`
* causes an unsupported `group:falling_node` node to fall and causes an
unattached `group:attached_node` node to fall.
* spread these updates to neighbours and can cause a cascade
of nodes to fall.
* `minetest.get_spawn_level(x, z)`
* Returns a player spawn y co-ordinate for the provided (x, z)
co-ordinates, or `nil` for an unsuitable spawn point.
* For most mapgens a 'suitable spawn point' is one with y between
`water_level` and `water_level + 16`, and in mgv7 well away from rivers,
so `nil` will be returned for many (x, z) co-ordinates.
* The spawn level returned is for a player spawn in unmodified terrain.
* The spawn level is intentionally above terrain level to cope with
full-node biome 'dust' nodes.
Mod channels
------------
You can find mod channels communication scheme in `doc/mod_channels.png`.
* `minetest.mod_channel_join(channel_name)`
* Server joins channel `channel_name`, and creates it if necessary. You
should listen for incoming messages with
`minetest.register_on_modchannel_message`
Inventory
---------
`minetest.get_inventory(location)`: returns an `InvRef`
* `location` = e.g.
* `{type="player", name="celeron55"}`
* `{type="node", pos={x=, y=, z=}}`
* `{type="detached", name="creative"}`
* `minetest.create_detached_inventory(name, callbacks, [player_name])`: returns
an `InvRef`.
* `callbacks`: See [Detached inventory callbacks]
* `player_name`: Make detached inventory available to one player
exclusively, by default they will be sent to every player (even if not
used).
Note that this parameter is mostly just a workaround and will be removed
in future releases.
* Creates a detached inventory. If it already exists, it is cleared.
* `minetest.do_item_eat(hp_change, replace_with_item, itemstack, user, pointed_thing)`:
returns left over ItemStack.
* See `minetest.item_eat` and `minetest.register_on_item_eat`
Formspec
--------
* `minetest.show_formspec(playername, formname, formspec)`
* `playername`: name of player to show formspec
* `formname`: name passed to `on_player_receive_fields` callbacks.
It should follow the `"modname:<whatever>"` naming convention
* `formspec`: formspec to display
* `minetest.close_formspec(playername, formname)`
* `playername`: name of player to close formspec
* `formname`: has to exactly match the one given in `show_formspec`, or the
formspec will not close.
* calling `show_formspec(playername, formname, "")` is equal to this
expression.
* to close a formspec regardless of the formname, call
`minetest.close_formspec(playername, "")`.
**USE THIS ONLY WHEN ABSOLUTELY NECESSARY!**
* `minetest.formspec_escape(string)`: returns a string
* escapes the characters "[", "]", "\", "," and ";", which can not be used
in formspecs.
* `minetest.explode_table_event(string)`: returns a table
* returns e.g. `{type="CHG", row=1, column=2}`
* `type` is one of:
* `"INV"`: no row selected
* `"CHG"`: selected
* `"DCL"`: double-click
* `minetest.explode_textlist_event(string)`: returns a table
* returns e.g. `{type="CHG", index=1}`
* `type` is one of:
* `"INV"`: no row selected
* `"CHG"`: selected
* `"DCL"`: double-click
* `minetest.explode_scrollbar_event(string)`: returns a table
* returns e.g. `{type="CHG", value=500}`
* `type` is one of:
* `"INV"`: something failed
* `"CHG"`: has been changed
* `"VAL"`: not changed
Item handling
-------------
* `minetest.inventorycube(img1, img2, img3)`
* Returns a string for making an image of a cube (useful as an item image)
* `minetest.get_pointed_thing_position(pointed_thing, above)`
* Get position of a `pointed_thing` (that you can get from somewhere)
* `minetest.dir_to_facedir(dir, is6d)`
* Convert a vector to a facedir value, used in `param2` for
`paramtype2="facedir"`.
* passing something non-`nil`/`false` for the optional second parameter
causes it to take the y component into account.
* `minetest.facedir_to_dir(facedir)`
* Convert a facedir back into a vector aimed directly out the "back" of a
node.
* `minetest.dir_to_wallmounted(dir)`
* Convert a vector to a wallmounted value, used for
`paramtype2="wallmounted"`.
* `minetest.wallmounted_to_dir(wallmounted)`
* Convert a wallmounted value back into a vector aimed directly out the
"back" of a node.
* `minetest.dir_to_yaw(dir)`
* Convert a vector into a yaw (angle)
* `minetest.yaw_to_dir(yaw)`
* Convert yaw (angle) to a vector
* `minetest.is_colored_paramtype(ptype)`
* Returns a boolean. Returns `true` if the given `paramtype2` contains
color information (`color`, `colorwallmounted` or `colorfacedir`).
* `minetest.strip_param2_color(param2, paramtype2)`
* Removes everything but the color information from the
given `param2` value.
* Returns `nil` if the given `paramtype2` does not contain color
information.
* `minetest.get_node_drops(nodename, toolname)`
* Returns list of item names.
* **Note**: This will be removed or modified in a future version.
* `minetest.get_craft_result(input)`: returns `output, decremented_input`
* `input.method` = `"normal"` or `"cooking"` or `"fuel"`
* `input.width` = for example `3`
* `input.items` = for example
`{stack1, stack2, stack3, stack4, stack 5, stack 6, stack 7, stack 8, stack 9}`
* `output.item` = `ItemStack`, if unsuccessful: empty `ItemStack`
* `output.time` = a number, if unsuccessful: `0`
* `output.replacements` = list of `ItemStack`s that couldn't be placed in
`decremented_input.items`
* `decremented_input` = like `input`
* `minetest.get_craft_recipe(output)`: returns input
* returns last registered recipe for output item (node)
* `output` is a node or item type such as `"default:torch"`
* `input.method` = `"normal"` or `"cooking"` or `"fuel"`
* `input.width` = for example `3`
* `input.items` = for example
`{stack1, stack2, stack3, stack4, stack 5, stack 6, stack 7, stack 8, stack 9}`
* `input.items` = `nil` if no recipe found
* `minetest.get_all_craft_recipes(query item)`: returns a table or `nil`
* returns indexed table with all registered recipes for query item (node)
or `nil` if no recipe was found.
* recipe entry table:
* `method`: 'normal' or 'cooking' or 'fuel'
* `width`: 0-3, 0 means shapeless recipe
* `items`: indexed [1-9] table with recipe items
* `output`: string with item name and quantity
* Example query for `"default:gold_ingot"` will return table:
{
[1]={method = "cooking", width = 3, output = "default:gold_ingot",
items = {1 = "default:gold_lump"}},
[2]={method = "normal", width = 1, output = "default:gold_ingot 9",
items = {1 = "default:goldblock"}}
}
* `minetest.handle_node_drops(pos, drops, digger)`
* `drops`: list of itemstrings
* Handles drops from nodes after digging: Default action is to put them
into digger's inventory.
* Can be overridden to get different functionality (e.g. dropping items on
ground)
* `minetest.itemstring_with_palette(item, palette_index)`: returns an item