miscset (version 0.4)

gapply: Apply a Function by a Grid on a List

Description

gapply applies a function on a data.frame by using named list entries as grid. Support of multicore processing via mclapply. levels is a method to retrieve the grid of a gapply object. as.data.frame is a simplified rbind version which adds the grid to the output.

Usage

gapply(d, by, fun, ..., drop = TRUE, cores = 1)
## S3 method for class 'gapply':
levels(x)
## S3 method for class 'gapply':
as.data.frame(x, ...)
## S3 method for class 'gapply':
as.data.table(x, keyv = NULL, ...)

Arguments

d

A data.frame or data.table object.

by

A character vector with colnames for the grid to subset.

fun

A function to apply on d.

...

Additional arguments to fun or to as.data.frame method.

drop

Drop grid columns for fun.

cores

Number of multicores as mc.cores in mclapply.

x

Object of class gapply.

keyv

Character vector for data.table keys.

Details

The output of gapply is a list inheriting from the S3 class gapply which stores the grid of the original data. Support for subsetting grid entries is not yet supported. The function fun receives the subset of the data frame as first argument.

See Also

mclapply, setkeyv

Examples

#

f <- function (x) c(conc.diff = diff(range(x$conc)), uptake.sum=sum(x$uptake))
d <- gapply(CO2, c('Type', 'Treatment'), f)
d
levels(d)
as.data.frame(d)

#