colapply: Applies a function to columns of a data.table
In AlbertoEAF/data.Rh: Data manipulation helpers (adds convenient operations)

Description Usage Arguments Value Examples

View source: R/hello.R

You can apply a function independently to many columns at the same time. You can apply a function to: - all columns - specified columns (with cols=c("col1","col2", ...)) - cols whose name match cols_pattern

1 2	colapply(data, fn, cols = c(), cols_pattern = c(), prefix = "", suffix = "", by = NULL, coltypes_filter_fn = NULL, inplace = F)

`data`	- must be a data.table
`fn`	- function to apply to the columns
`cols`	- a character vector. If used only columns with such names will be used.
`cols_pattern`	- a character specifying the cols name pattern to which the function will be applied
`prefix`	- the prefix to apply to the resulting column names
`suffix`	- the suffix to apply to the resulting column names
`by`	- same as in data.table
`inplace`	(default==FALSE) - TRUE/FALSE
`coltype_filter_fn`	- a function that filters the coltypes object (applied at the end). Takes a coltypes data.table and returns it filtered.

data.table with remapped columns

    capply(data.table(iris), typeof)
    capply(data.table(iris), mean, cols_pattern="Length|Width")
    capply(data.table(iris), unique, cols=c("Species"))
    capply(data.table(iris), unique, c("Species")) # the same - cols is the first optional argument
    capply(data.table(iris), unique, "Species") # if the vector has dimension 1, this is equivalent
    capply(data.table(iris), function(x){x^2}, cols_pattern="Length|Width", inplace=T) # square all numeric columns