mirror of
https://github.com/HybridDog/vector_extras.git
synced 2024-11-15 06:50:35 +01:00
155 lines
6.8 KiB
Markdown
155 lines
6.8 KiB
Markdown
|
|
||
|
# Vector helpers added by this mod
|
||
|
|
||
|
## Helpers which return many positions for a shape, e.g. a line
|
||
|
|
||
|
### Line functions
|
||
|
|
||
|
These may be deprecated since raycasting has been added to minetest.
|
||
|
See e.g. `minetest.line_of_sight`.
|
||
|
|
||
|
* `vector.line([pos, dir[, range][, alt]])`: returns a table of vectors
|
||
|
* `dir` is either a direction (when range is a number) or
|
||
|
the start position (when range is the end position).
|
||
|
* If alt is true, an old path calculation is used.
|
||
|
* `vector.twoline(x, y)`: can return e.g. `{{0,0}, {0,1}}`
|
||
|
* This is a lower-level function than `vector.line`; it can be used for
|
||
|
a 2D line.
|
||
|
* `vector.threeline(x, y, z)`: can return e.g. `{{0,0,0}, {0,1,0}}`
|
||
|
* Similar to `vector.twoline`; this one is for the 3D case.
|
||
|
* The parameters should be integers.
|
||
|
* `vector.rayIter(pos, dir)`: returns an iterator for a for loop
|
||
|
* `pos` can have non-integer values
|
||
|
* `vector.fine_line([pos, dir[, range], scale])`: returns a table of vectors
|
||
|
* Like `vector.line` but allows non-integer positions
|
||
|
* It uses `vector.rayIter`.
|
||
|
|
||
|
|
||
|
### Flood Fill
|
||
|
|
||
|
* `vector.search_2d(go_test, x0, y0, allow_revisit, give_map)`: returns e.g.
|
||
|
`{{0,0}, {0,1}}`
|
||
|
* This function uses a Flood Fill algorithm, so it can be used to detect
|
||
|
positions connected to each other in 2D.
|
||
|
* `go_test(x, y)` should be a function which returns true iff the algorithm
|
||
|
can "fill" at the position `(x, y)`.
|
||
|
* `(x0, y0)` defines the start position.
|
||
|
* If `allow_revisit` is false (the default), the function
|
||
|
invokes `go_test` only once at every potential position.
|
||
|
* If `give_map` is true (default is false), the function returns the
|
||
|
marked table, whose indices are 2D vector indices, instead of a list of
|
||
|
2D positions.
|
||
|
* `vector.search_3d(can_go, startpos, apply_move, moves)`: returns FIXME
|
||
|
* FIXME
|
||
|
|
||
|
|
||
|
### Other Shapes
|
||
|
|
||
|
* `vector.explosion_table(r)`: returns e.g. `{{pos1}, {pos2, true}}`
|
||
|
* The returned list of positions and boolean represents a sphere;
|
||
|
if the boolean is true, the position is on the outer side of the sphere.
|
||
|
* It might be used for explosion calculations; but `vector.explosion_perlin`
|
||
|
should make more realistic holes.
|
||
|
* `vector.explosion_perlin(rmin, rmax[, nparams])`: returns e.g.
|
||
|
`{{pos1}, {pos2, true}}`
|
||
|
* This function is similar to `vector.explosion_table`; the positions
|
||
|
do not represent a sphere but a more complex hole which is calculated
|
||
|
with the help of perlin noise.
|
||
|
* `rmin` and `rmax` represent the minimum and maximum radius,
|
||
|
and `nparams` (which has a default value) are parameters for the perlin
|
||
|
noise.
|
||
|
* `vector.circle(r)`: returns a table of vectors
|
||
|
* The returned positions represent a circle of radius `r` along the x and z
|
||
|
directions; the y coordinates are all zero.
|
||
|
* `vector.ring(r)`: returns a table of vectors
|
||
|
* This function is similar to `vector.circle`; the positions are all
|
||
|
touching each other (i.e. they are connected on whole surfaces and not
|
||
|
only infinitely thin edges), so it is called `ring` instead of `circle`
|
||
|
* `r` can be a non-integer number.
|
||
|
* `vector.throw_parabola(pos, vel, gravity, point_count, time)`
|
||
|
* FIXME: should return positions along a parabola so that moving objects
|
||
|
collisions can be calculated
|
||
|
* `vector.triangle(pos1, pos2, pos3)`: returns a table of positions, a number
|
||
|
and a table with barycentric coordinates
|
||
|
* This function calculates integer positions for a triangle defined by
|
||
|
`pos1`, `pos2` and `pos3`, so it can be used to place polygons in
|
||
|
minetest.
|
||
|
* The returned number is the number of positions.
|
||
|
* The barycentric coordinates are specified in a table with three elements;
|
||
|
the first one corresponds to `pos1`, etc.
|
||
|
|
||
|
|
||
|
## Helpers for various vector calculations
|
||
|
|
||
|
* `vector.sort_positions(ps[, preferred_coords])`
|
||
|
* Sorts a table of vectors `ps` along the coordinates specified in the
|
||
|
table `preferred_coords` in-place.
|
||
|
* If `preferred_coords` is omitted, it sorts along z, y and x in this order,
|
||
|
where z has the highest priority.
|
||
|
* `vector.maxnorm(v)`: returns the Tschebyshew norm of `v`
|
||
|
* `vector.sumnorm(v)`: returns the Manhattan norm of `v`
|
||
|
* `vector.pnorm(v, p)`: returns the `p` norm of `v`
|
||
|
* `vector.inside(pos, minp, maxp)`: returns a boolean
|
||
|
* Returns true iff `pos` is within the closed AABB defined by `minp`
|
||
|
and `maxp`.
|
||
|
* `vector.minmax(pos1, pos2)`: returns two vectors
|
||
|
* This does the same as `worldedit.sort_pos`.
|
||
|
* The components of the second returned vector are all bigger or equal to
|
||
|
those of the first one.
|
||
|
* `vector.move(pos1, pos2, length)`: returns a vector
|
||
|
* Go from `pos1` `length` metres to `pos2` and then round to the nearest
|
||
|
integer position.
|
||
|
* Made for rubenwardy
|
||
|
* `vector.from_number(i)`: returns `{x=i, y=i, z=i}`
|
||
|
* `vector.chunkcorner(pos)`: returns a vector
|
||
|
* Returns the mapblock position of the mapblock which contains
|
||
|
the integer position `pos`
|
||
|
* `vector.point_distance_minmax(p1, p2)`: returns two numbers
|
||
|
* Returns the minimum and maximum of the absolute component-wise distances
|
||
|
* `vector.collision(p1, p2)` FIXME
|
||
|
* `vector.update_minp_maxp(minp, maxp, pos)`
|
||
|
* Can change `minp` and `maxp` so that `pos` is within the AABB defined by
|
||
|
`minp` and `maxp`
|
||
|
* `vector.unpack(v)`: returns three numbers
|
||
|
* Returns `v.z, v.y, v.x`
|
||
|
* `vector.get_max_coord(v)`: returns a string
|
||
|
* Returns `"x"`, `"y"` or `"z"`, depending on which component has the
|
||
|
biggest value
|
||
|
* `vector.get_max_coords(v)`: returns three strings
|
||
|
* Similar to `vector.get_max_coord`; it returns the coordinates in the order
|
||
|
of their component values
|
||
|
* Example: `vector.get_max_coords{x=1, y=5, z=3}` returns `"y", "z", "x"`
|
||
|
* `vector.serialize(v)`: returns a string
|
||
|
* In comparison to `minetest.serialize`, this function uses a more compact
|
||
|
string for the serialization.
|
||
|
|
||
|
|
||
|
## Minetest-specific helper functions
|
||
|
|
||
|
* `vector.straightdelay([length, vel[, acc]])`: returns a number
|
||
|
* Returns the time an object takes to move `length` if it has velocity `vel`
|
||
|
and acceleration `acc`
|
||
|
* `vector.sun_dir([time])`: returns a vector or nil
|
||
|
* Returns the vector which points to the sun
|
||
|
* If `time` is omitted, it uses the current time.
|
||
|
* This function does not yet support the moon;
|
||
|
at night it simply returns `nil`.
|
||
|
|
||
|
|
||
|
## Helpers which I don't recommend to use now
|
||
|
|
||
|
* `vector.pos_to_string(pos)`: returns a string
|
||
|
* It is similar to `minetest.pos_to_string`; it uses a different format:
|
||
|
`"("..pos.x.."|"..pos.y.."|"..pos.z..")"`
|
||
|
* `vector.zero`
|
||
|
* The zero vector `{x=0, y=0, z=0}`
|
||
|
* `vector.quickadd(pos, [z],[y],[x])`
|
||
|
* Adds values to the vector components in-place
|
||
|
|
||
|
|
||
|
## Deprecated helpers
|
||
|
|
||
|
* `vector.plane`
|
||
|
* should be removed soon; it should have done the same as vector.triangle
|
||
|
|