mirror of
https://github.com/Dadido3/noita-mapcap.git
synced 2024-12-22 02:17:33 +00:00
David Vogel
0126e706cb
- Add ability to capture while normally playing - Calculate capture area based on coordinate transformations - Improve and simplify captureScreenshot function - Move dynamic library wrappers into libraries folder - Update capture.dll to support cropping and resizing - Recompile capture.dll with newer PureBasic compiler that uses C backend - Increase capture.dll worker threads to 6 - Increase capture.dll queue size by one - Add Round and Rounded methods to Vec2 - Split magic number XML files for easier debugging - Fix some EmmyLua annotations - And and fix some comments
331 lines
9.2 KiB
Lua
331 lines
9.2 KiB
Lua
-- Copyright (c) 2022 David Vogel
|
|
--
|
|
-- This software is released under the MIT License.
|
|
-- https://opensource.org/licenses/MIT
|
|
|
|
-- Just 2D vector stuff. Mainly for Noita.
|
|
|
|
-- State: Some stuff is untested.
|
|
|
|
---Metatable of the Vec2 object that is returned by this lib.
|
|
---This will only contain a __call field pointing to the constructor.
|
|
---@class Vec2Meta
|
|
local libMetaTable = {}
|
|
|
|
---@class Vec2
|
|
---@field [1] number
|
|
---@field [2] number
|
|
---@field x number
|
|
---@field y number
|
|
local Vec2 = setmetatable({}, libMetaTable)
|
|
|
|
-----------------
|
|
-- Constructor --
|
|
-----------------
|
|
|
|
---Creates a new vector.
|
|
---
|
|
--- Vec2() -- Returns a vector with zeroed coordinates.
|
|
--- Vec2(v) -- Returns a copy of the given Vec2 object.
|
|
--- Vec2("1.2, 3.4") -- Returns a vector with x = 1.2 and y = 3.4.
|
|
--- Vec2(1.2, 3.4) -- Returns a vector with x = 1.2 and y = 3.4.
|
|
---@param ... any
|
|
---@return Vec2
|
|
function libMetaTable:__call(...)
|
|
local n = select("#", ...)
|
|
if n == 0 then
|
|
return setmetatable({ 0, 0 }, Vec2) -- Zero initialized vector.
|
|
elseif n == 1 then
|
|
local param = select(1, ...)
|
|
if type(param) == "string" then
|
|
local vector = {}
|
|
for field in string.gmatch(param, "[^,%s]+") do
|
|
table.insert(vector, tonumber(field))
|
|
end
|
|
assert(#vector == 2, string.format("parsed vector contains an invalid number of fields: %d, expected %d", #vector, 2))
|
|
return setmetatable(vector, Vec2) -- Vector initialized with the given coordinates.
|
|
elseif getmetatable(param) == Vec2 then
|
|
return Vec2(param[1], param[2])
|
|
end
|
|
error(string.format("unsupported argument type %q", type(param)))
|
|
elseif n == 2 then
|
|
assert(type(select(1, ...)) == "number", string.format("first argument has type %q, expects %q", type(select(1, ...)), "number"))
|
|
assert(type(select(2, ...)) == "number", string.format("first argument has type %q, expects %q", type(select(2, ...)), "number"))
|
|
return setmetatable({ ... }, Vec2) -- Vector initialized with the given coordinates.
|
|
end
|
|
|
|
error(string.format("called Vec2 constructor with %d argument(s)", n))
|
|
end
|
|
|
|
-----------------
|
|
-- Metamethods --
|
|
-----------------
|
|
|
|
---Handle special fields, like x and y.
|
|
---@param key any
|
|
---@return any
|
|
function Vec2:__index(key)
|
|
if type(key) == "number" then return rawget(self, key) end
|
|
if key == "x" then return rawget(self, 1) end
|
|
if key == "y" then return rawget(self, 2) end
|
|
return rawget(Vec2, key)
|
|
end
|
|
|
|
---Handle special fields, like x and y.
|
|
---@param key any
|
|
function Vec2:__newindex(key, value)
|
|
if type(key) == "number" then return rawset(self, key, value) end
|
|
if key == "x" then rawset(self, 1, value) end
|
|
if key == "y" then rawset(self, 2, value) end
|
|
-- There should no way to manipulate any other keys of the object or its metatable.
|
|
end
|
|
|
|
---Returns a string representation of this vector.
|
|
---This supports a round-trip via Vec2(tostring(v)) without loss of information.
|
|
---@return string
|
|
function Vec2:__tostring()
|
|
return string.format("%.16g, %.16g", self[1], self[2])
|
|
end
|
|
|
|
----------------------------
|
|
-- Mathematic metamethods --
|
|
----------------------------
|
|
|
|
---Returns a new vector that is the sum v1 + v2.
|
|
---
|
|
---This will not mutate any vector.
|
|
---@param v1 Vec2
|
|
---@param v2 Vec2
|
|
---@return Vec2
|
|
function Vec2.__add(v1, v2)
|
|
assert(getmetatable(v1) == Vec2 and getmetatable(v2) == Vec2, "wrong argument types. Expected two Vec2 objects")
|
|
return Vec2(v1[1] + v2[1], v1[2] + v2[2])
|
|
end
|
|
|
|
---Returns a new vector that is the difference v1 - v2.
|
|
---
|
|
---This will not mutate any vector.
|
|
---@param v1 Vec2
|
|
---@param v2 Vec2
|
|
---@return Vec2
|
|
function Vec2.__sub(v1, v2)
|
|
assert(getmetatable(v1) == Vec2 and getmetatable(v2) == Vec2, "wrong argument types. Expected two Vec2 objects")
|
|
return Vec2(v1[1] - v2[1], v1[2] - v2[2])
|
|
end
|
|
|
|
---Returns a new vector that is the multiplication of a vector with a scalar.
|
|
---
|
|
---This will not mutate any value.
|
|
---@param a number|Vec2
|
|
---@param b number|Vec2
|
|
---@return Vec2
|
|
function Vec2.__mul(a, b)
|
|
if type(a) == "number" and getmetatable(b) == Vec2 then
|
|
return Vec2(b[1] * a, b[2] * a)
|
|
elseif getmetatable(a) == Vec2 and type(b) == "number" then
|
|
return Vec2(a[1] * b, a[2] * b)
|
|
end
|
|
|
|
error(string.format("invalid combination of argument types for multiplication: %q and %q", type(a), type(b)))
|
|
end
|
|
|
|
---Returns a new vector that is the division of a vector by a scalar.
|
|
---
|
|
---This will not mutate any value.
|
|
---@param v Vec2
|
|
---@param s number
|
|
---@return Vec2
|
|
function Vec2.__div(v, s)
|
|
if getmetatable(v) == Vec2 and type(s) == "number" then
|
|
return Vec2(v[1] / s, v[2] / s)
|
|
end
|
|
|
|
error(string.format("invalid combination of argument types for division: %q and %q", type(v), type(s)))
|
|
end
|
|
|
|
---Returns the negated vector.
|
|
---
|
|
---This will not mutate any value.
|
|
---@return Vec2
|
|
function Vec2:__unm()
|
|
return Vec2(-self[1], -self[2])
|
|
end
|
|
|
|
---Returns whether the two vectors are equal.
|
|
---Will return false if one of the values is not a vector.
|
|
---@param v1 Vec2
|
|
---@param v2 Vec2
|
|
---@return boolean
|
|
function Vec2.__eq(v1, v2)
|
|
if getmetatable(v1) ~= Vec2 or getmetatable(v2) ~= Vec2 then return false end
|
|
return v1[1] == v2[1] and v1[2] == v2[2]
|
|
end
|
|
|
|
-------------
|
|
-- Methods --
|
|
-------------
|
|
|
|
---Adds v to the vector.
|
|
---
|
|
---This mutates self.
|
|
---@param v Vec2
|
|
function Vec2:Add(v)
|
|
assert(getmetatable(v) == Vec2, string.format("wrong argument type %q, expected Vec2 object", type(v)))
|
|
self[1], self[2] = self[1] + v[1], self[2] + v[2]
|
|
end
|
|
|
|
---Subtracts v from the vector.
|
|
---
|
|
---This mutates self.
|
|
---@param v Vec2
|
|
function Vec2:Sub(v)
|
|
assert(getmetatable(v) == Vec2, string.format("wrong argument type %q, expected Vec2 object", type(v)))
|
|
self[1], self[2] = self[1] - v[1], self[2] - v[2]
|
|
end
|
|
|
|
---Multiplies self with the given scalar.
|
|
---
|
|
---This mutates self.
|
|
---@param s number
|
|
function Vec2:Mul(s)
|
|
assert(type(s) == "number", string.format("wrong argument type %q, expected number", type(s)))
|
|
self[1], self[2] = self[1] * s, self[2] * s
|
|
end
|
|
|
|
---Divides self by the given scalar.
|
|
---
|
|
---This mutates self.
|
|
---@param s number
|
|
function Vec2:Div(s)
|
|
assert(type(s) == "number", string.format("wrong argument type %q, expected number", type(s)))
|
|
self[1], self[2] = self[1] / s, self[2] / s
|
|
end
|
|
|
|
---Returns a copy of self.
|
|
---@return Vec2
|
|
function Vec2:Copy()
|
|
return Vec2(self)
|
|
end
|
|
|
|
---Returns the vector fields as parameters.
|
|
---@return number
|
|
---@return number
|
|
function Vec2:Unpack()
|
|
return self[1], self[2]
|
|
end
|
|
|
|
---Sets the vector fields to the given coordinates.
|
|
---@param x number
|
|
---@param y number
|
|
function Vec2:Set(x, y)
|
|
assert(type(x) == "number", string.format("wrong argument type %q, expected number", type(x)))
|
|
assert(type(y) == "number", string.format("wrong argument type %q, expected number", type(y)))
|
|
self[1], self[2] = x, y
|
|
end
|
|
|
|
---Returns the squared length of the vector.
|
|
---@return number
|
|
function Vec2:LengthSqr()
|
|
return self[1] ^ 2 + self[2] ^ 2
|
|
end
|
|
|
|
---Returns the length of the vector.
|
|
---@return number
|
|
function Vec2:Length()
|
|
return math.sqrt(self:LengthSqr())
|
|
end
|
|
|
|
---Returns the squared distance of self to the given vector.
|
|
---@param v Vec2
|
|
---@return number
|
|
function Vec2:DistanceSqr(v)
|
|
return (v - self):LengthSqr()
|
|
end
|
|
|
|
---Returns the distance of self to the given vector.
|
|
---@param v Vec2
|
|
---@return number
|
|
function Vec2:Distance(v)
|
|
return (v - self):Length()
|
|
end
|
|
|
|
---Sets the length of the vector to 1.
|
|
---
|
|
---This mutates self.
|
|
function Vec2:Normalize()
|
|
local len = self:Length()
|
|
self:Div(len)
|
|
end
|
|
|
|
---Returns a copy of the vector with its length set to 1.
|
|
---@return Vec2
|
|
function Vec2:Normalized()
|
|
local len = self:Length()
|
|
return self / len
|
|
end
|
|
|
|
---Compares this vector to the given one.
|
|
---@param v Vec2
|
|
---@param tolerance number -- Tolerance per field.
|
|
---@return boolean
|
|
function Vec2:EqualTo(v, tolerance)
|
|
if math.abs(v[1] - self[1]) > tolerance then return false end
|
|
if math.abs(v[2] - self[2]) > tolerance then return false end
|
|
return true
|
|
end
|
|
|
|
---Round returns v rounded by the given method.
|
|
---@param x number
|
|
---@param method "nearest"|"floor"|"ceil"|"to-zero"|"away-zero"|nil -- Defaults to "nearest".
|
|
---@return integer
|
|
local function round(x, method)
|
|
method = method or "nearest"
|
|
|
|
if method == "nearest" then
|
|
return math.floor(x + 0.5)
|
|
elseif method == "floor" then
|
|
return math.floor(x)
|
|
elseif method == "ceil" then
|
|
return math.ceil(x)
|
|
elseif method == "to-zero" then
|
|
if x >= 0 then
|
|
return math.floor(x)
|
|
else
|
|
return math.ceil(x)
|
|
end
|
|
elseif method == "away-zero" then
|
|
if x >= 0 then
|
|
return math.ceil(x)
|
|
else
|
|
return math.floor(x)
|
|
end
|
|
end
|
|
|
|
error(string.format("invalid rounding method %q", method))
|
|
end
|
|
|
|
---Round rounds all vector fields individually by the given rounding method.
|
|
---@param method "nearest"|"floor"|"ceil"|"to-zero"|"away-zero"|nil -- Defaults to "nearest".
|
|
function Vec2:Round(method)
|
|
self[1], self[2] = round(self[1], method), round(self[2], method)
|
|
end
|
|
|
|
---Round rounds all vector fields individually by the given rounding method.
|
|
---@param method "nearest"|"floor"|"ceil"|"to-zero"|"away-zero"|nil -- Defaults to "nearest".
|
|
---@return Vec2
|
|
function Vec2:Rounded(method)
|
|
return Vec2(round(self[1], method), round(self[2], method))
|
|
end
|
|
|
|
-------------------------
|
|
-- JSON Implementation --
|
|
-------------------------
|
|
|
|
---MarshalJSON implements the JSON marshaler interface.
|
|
---@return string
|
|
function Vec2:MarshalJSON()
|
|
return string.format("[%.16g, %.16g]", self[1], self[2]) -- Encode as JSON array. -- TODO: Handle NaN, +Inf, -Inf, ... correctly
|
|
end
|
|
|
|
return Vec2
|