Detailed Description

template<class Blob_>
class tatami_chunked::SimpleSparseChunkWrapper< Blob_ >

Create a sparse chunk for a CustomSparseChunkedMatrix.

Wraps a sparse blob in a simple chunk interface for use inside a CustomSparseChunkedMatrix. Each blob should hold a (possibly compressed) 2-dimensional sparse array of numeric values. The wrapper satisfies the MockSimpleSparseChunk interface, but is even simpler; extraction of any data involves realization of the entire blob, with no optimization for subsets of interest along either dimension.

The Blob_ class should provide the following:

typedef value_type, specifying the type of the value in the array.
typedef index_type, specifying the type of the index in the submatrix.
A nrow() const method, returning the number of rows in the array as an integer.
A ncol() const method, returning the numbr of columns in the array as an integer.
A bool is_csr() const method, indicating whether the realized data is in compressed sparse row (CSR) format.
A void inflate(std::vector<value_type>& values, std::vector<Index_>& indices, std::vector<size_t>& pointers) const method that fills values, indices and pointers, each with the corresponding field for compressed sparse matrix format. (The type of Index_ is the same as that used in the CustomSparseChunkedMatrix.) This should constitute a CSR matrix if is_csr() returns true, and a CSC matrix otherwise.

Template Parameters

Blob_ Class to represent a simple chunk.

The documentation for this class was generated from the following file:

tatami_chunked/mock_sparse_chunk.hpp