Transf¶

template<size_t N = 0, typename Scalar = std::conditional_t<N == 0, uint32_t, typename SmallestInteger<N>::type>> class Transf : public PTransf<0, std::conditional_t<0 == 0, uint32_t, typename SmallestInteger<0>::type>>¶

Defined in transf.hpp.

A transformation \(f\) is just a function defined on the whole of \(\{0, 1, \ldots, n - 1\}\) for some integer \(n\) called the degree of \(f\). A transformation is stored as a vector of the images of \(\{0, 1, \ldots, n - 1\}\), i.e. \(\{(0)f, (1)f, \ldots, (n - 1)f\}\).

If N is 0 (the default), then the degree of a Transf instance can be defined at runtime, and if N is not 0, then the degree is fixed at compile time.

If N is 0, then the default value of Scalar is uint32_t. If N is not 0, then the default value of Scalar is the smallest integer type able to hold N. See also SmallestInteger.

Note

Transf has the same member functions as StaticPTransf and DynamicPTransf, this isn’t currently reflected by the contents of this page.

Template Parameters

N – the degree (default: 0)
Scalar – an unsigned integer type (the type of the image values)

Member types¶

`container_type`	Type of the underlying container.
`value_type`	Type of the image values.

Static member functions¶

`identity() const`	Returns the identity transformation on degree() points.
`identity(size_t)`	Returns the identity transformation on the given number of points.
`make(T&&)`	Construct from a container and validate.
`make(std::initializer_list<value_type>&&)`	Construct from a container and validate.

Member functions¶

product_inplace(Transf const&, Transf const&)

Multiply two transformations and store the product in this.