diagrams-core-1.3.0.3: Core libraries for diagrams EDSL

Diagrams.Core.Transform

Description

Diagrams defines the core library of primitives forming the basis of an embedded domain-specific language for describing and rendering diagrams.

The `Transform` module defines generic transformations parameterized by any vector space.

Synopsis

# Transformations

## Invertible linear transformations

data u :-: v infixr 7 Source

`(v1 :-: v2)` is a linear map paired with its inverse.

Constructors

 (u -> v) :-: (v -> u) infixr 7

Instances

 Monoid ((:-:) v v) Invertible linear maps from a vector space to itself form a monoid under composition. Semigroup ((:-:) a a)

(<->) :: (u -> v) -> (v -> u) -> u :-: v Source

Create an invertible linear map from two functions which are assumed to be linear inverses.

linv :: (u :-: v) -> v :-: u Source

Invert a linear map.

lapp :: (u :-: v) -> u -> v Source

Apply a linear map to a vector.

## General transformations

data Transformation v n Source

General (affine) transformations, represented by an invertible linear map, its transpose, and a vector representing a translation component.

By the transpose of a linear map we mean simply the linear map corresponding to the transpose of the map's matrix representation. For example, any scale is its own transpose, since scales are represented by matrices with zeros everywhere except the diagonal. The transpose of a rotation is the same as its inverse.

The reason we need to keep track of transposes is because it turns out that when transforming a shape according to some linear map L, the shape's normal vectors transform according to L's inverse transpose. (For a more detailed explanation and proof, see https://wiki.haskell.org/Diagrams/Dev/Transformations.) This is exactly what we need when transforming bounding functions, which are defined in terms of perpendicular (i.e. normal) hyperplanes.

For more general, non-invertible transformations, see `Diagrams.Deform` (in `diagrams-lib`).

Constructors

 Transformation (v n :-: v n) (v n :-: v n) (v n)

Instances

 (Additive v, Num n) => Monoid (Transformation v n) (Additive v, Num n) => Semigroup (Transformation v n) Transformations are closed under composition; `t1 <> t2` is the transformation which performs first `t2`, then `t1`. (Additive v, Num n) => HasOrigin (Transformation v n) (Additive v, Num n) => Transformable (Transformation v n) (Transformable a, (~) (* -> *) (V a) v, (~) * (N a) n) => Action (Transformation v n) a Transformations can act on transformable things. type N (Transformation v n) = n type V (Transformation v n) = v

inv :: (Functor v, Num n) => Transformation v n -> Transformation v n Source

Invert a transformation.

transp :: Transformation v n -> v n :-: v n Source

Get the transpose of a transformation (ignoring the translation component).

transl :: Transformation v n -> v n Source

Get the translational component of a transformation.

dropTransl :: (Additive v, Num n) => Transformation v n -> Transformation v n Source

Drop the translational component of a transformation, leaving only the linear part.

apply :: Transformation v n -> v n -> v n Source

Apply a transformation to a vector. Note that any translational component of the transformation will not affect the vector, since vectors are invariant under translation.

papply :: (Additive v, Num n) => Transformation v n -> Point v n -> Point v n Source

Apply a transformation to a point.

fromLinear :: (Additive v, Num n) => (v n :-: v n) -> (v n :-: v n) -> Transformation v n Source

Create a general affine transformation from an invertible linear transformation and its transpose. The translational component is assumed to be zero.

fromOrthogonal :: (Additive v, Num n) => (v n :-: v n) -> Transformation v n Source

An orthogonal linear map is one whose inverse is also its transpose.

fromSymmetric :: (Additive v, Num n) => (v n :-: v n) -> Transformation v n Source

A symmetric linear map is one whose transpose is equal to its self.

basis :: (Additive t, Traversable t, Num a) => [t a]

Produce a default basis for a vector space. If the dimensionality of the vector space is not statically known, see `basisFor`.

dimension :: forall a v. (V a ~ v, Additive v, Traversable v) => a -> Int Source

Get the dimension of an object whose vector space is an instance of `HasLinearMap`, e.g. transformations, paths, diagrams, etc.

onBasis :: (Additive v, Traversable v, Num n) => Transformation v n -> ([v n], v n) Source

Get the matrix equivalent of the linear transform, (as a list of columns) and the translation vector. This is mostly useful for implementing backends.

listRep :: Foldable v => v n -> [n] Source

Convert a vector v to a list of scalars.

matrixRep :: (Additive v, Traversable v, Num n) => Transformation v n -> [[n]] Source

Convert the linear part of a `Transformation` to a matrix representation as a list of column vectors which are also lists.

matrixHomRep :: (Additive v, Traversable v, Num n) => Transformation v n -> [[n]] Source

Convert a `Transformation v` to a homogeneous matrix representation. The final list is the translation. The representation leaves off the last row of the matrix as it is always [0,0, ... 1] and this representation is the defacto standard for backends.

determinant :: (Additive v, Traversable v, Num n) => Transformation v n -> n Source

The determinant of (the linear part of) a `Transformation`.

isReflection :: (Additive v, Traversable v, Num n, Ord n) => Transformation v n -> Bool Source

Determine whether a `Transformation` includes a reflection component, that is, whether it reverses orientation.

avgScale :: (Additive v, Traversable v, Floating n) => Transformation v n -> n Source

Compute the "average" amount of scaling performed by a transformation. Satisfies the properties

```  avgScale (scaling k) == k
avgScale (t1 <> t2)  == avgScale t1 * avgScale t2
```

eye :: (HasBasis v, Num n) => v (v n) Source

Identity matrix.

# The Transformable class

class (HasBasis v, Traversable v) => HasLinearMap v Source

`HasLinearMap` is a poor man's class constraint synonym, just to help shorten some of the ridiculously long constraint sets.

Instances

 (HasBasis v, Traversable v) => HasLinearMap v

class (Additive v, Representable v, Rep v ~ E v) => HasBasis v Source

An `Additive` vector space whose representation is made up of basis elements.

Instances

 (Additive v, Representable v, (~) * (Rep v) (E v)) => HasBasis v

class Transformable t where Source

Type class for things `t` which can be transformed.

Methods

transform :: Transformation (V t) (N t) -> t -> t Source

Apply a transformation to an object.

Instances

 Transformable t => Transformable [t] (Transformable t, Ord t) => Transformable (Set t) Transformable m => Transformable (Deletable m) (Num (N t), Additive (V t), Transformable t) => Transformable (TransInv t) ((~) (* -> *) (V t) v, (~) * (N t) n, (~) (* -> *) (V t) (V s), (~) * (N t) (N s), Functor v, Num n, Transformable t, Transformable s) => Transformable (s -> t) (Transformable t, Transformable s, (~) (* -> *) (V t) (V s), (~) * (N t) (N s)) => Transformable (t, s) Transformable t => Transformable (Map k t) (Additive v, Num n) => Transformable (Point v n) (Additive v, Num n) => Transformable (Transformation v n) (Additive v, Traversable v, Floating n) => Transformable (Style v n) (Additive v, Traversable v, Floating n) => Transformable (Attribute v n) `TAttribute`s are transformed directly, `MAttribute`s have their local scale multiplied by the average scale of the transform. Plain `Attribute`s are unaffected. (Additive v, Num n) => Transformable (Trace v n) (Metric v, Floating n) => Transformable (Envelope v n) (Transformable t, Transformable s, Transformable u, (~) (* -> *) (V s) (V t), (~) * (N s) (N t), (~) (* -> *) (V s) (V u), (~) * (N s) (N u)) => Transformable (t, s, u) (Additive v, Num n) => Transformable (Query v n m) Transformable (Prim b v n) The `Transformable` instance for `Prim` just pushes calls to `transform` down through the `Prim` constructor. (Metric v, Floating n) => Transformable (SubMap b v n m) (Metric v, Floating n) => Transformable (Subdiagram b v n m) (OrderedField n, Metric v, Semigroup m) => Transformable (QDiagram b v n m) Diagrams can be transformed by transforming each of their components appropriately.

# Translational invariance

newtype TransInv t Source

`TransInv` is a wrapper which makes a transformable type translationally invariant; the translational component of transformations will no longer affect things wrapped in `TransInv`.

Constructors

 TransInv t

Instances

 Eq t => Eq (TransInv t) Ord t => Ord (TransInv t) Show t => Show (TransInv t) Monoid t => Monoid (TransInv t) Semigroup t => Semigroup (TransInv t) Wrapped (TransInv t) HasOrigin (TransInv t) (Num (N t), Additive (V t), Transformable t) => Transformable (TransInv t) Qualifiable a => Qualifiable (TransInv a) Traced t => Traced (TransInv t) Enveloped t => Enveloped (TransInv t) Rewrapped (TransInv t) (TransInv t') type Unwrapped (TransInv t) = t type N (TransInv t) = N t type V (TransInv t) = V t

# Vector space independent transformations

Most transformations are specific to a particular vector space, but a few can be defined generically over any vector space.

translation :: v n -> Transformation v n Source

Create a translation.

translate :: (Num (N t), Transformable t) => Vn t -> t -> t Source

Translate by a vector.

scaling :: (Additive v, Fractional n) => n -> Transformation v n Source

Create a uniform scaling transformation.

scale :: (InSpace v n a, Eq n, Fractional n, Transformable a) => n -> a -> a Source

Scale uniformly in every dimension by the given scalar.