matrixStats (version 0.55.0)

indexByRow: Translates matrix indices by rows into indices by columns

Description

Translates matrix indices by rows into indices by columns.

Usage

indexByRow(dim, idxs = NULL, ...)

Arguments

dim

A numeric vector of length two specifying the length of the "template" matrix.

idxs

A vector of indices. If NULL, all indices are returned.

...

Not use.

Value

Returns an integer vector of indices.

Known limitations

The current implementation does not support long-vector indices, because both input and output indices are of type integers. This means that the indices in argument idxs can only be in range [1,2^31-1]. Using a greater value will be coerced to NA_integer_. Moreover, returned indices can only be in the same range [1,2^31-1].

Examples

# NOT RUN {
dim <- c(5, 4)
X <- matrix(NA_integer_, nrow = dim[1], ncol = dim[2])
Y <- t(X)
idxs <- seq_along(X)

# Assign by columns
X[idxs] <- idxs
print(X)

# Assign by rows
Y[indexByRow(dim(Y), idxs)] <- idxs
print(Y)

stopifnot(X == t(Y))
# }

Run the code above in your browser using DataCamp Workspace