write.fwf2: Reading and Writing Files in Fixed Width Format

In miceadds: Some Additional Multiple Imputation Functions, Especially for 'mice'

Reading and Writing Files in Fixed Width Format

Description

Reads and writes files in fixed width format. The functions are written for being more efficient than utils::read.fwf.

Usage

write.fwf2(dat, format.full, format.round, file)

read.fwf2( file, format.full, variables=NULL)

Arguments

dat	Data frame (or matrix). Variables can be numeric or strings. However, string length of string variables are not allowed to be larger than what is specified in format.full.
format.full	Vector with fixed width variable lengths
format.round	Vector with digits after decimals
file	File name
variables	Optional vector with variable names

See Also

utils::read.fwf

Examples

## Not run: 
#############################################################################
# EXAMPLE 1: Write and read a file in fixed width format
#############################################################################

# set working directory
path <- "P:/ARb/temp"
setwd(path)

# define a data frame
set.seed(9876)
dat <- data.frame( "x"=seq( 1, 21, len=5), "y"=stats::runif( 5 ),
            "z"=stats::rnorm( 5 ) )

# save data frame in fixed width format
format.full <- c(6, 6, 8 )
format.round <- c( 0, 2, 3 )
write.fwf2( dat, format.full=format.full, format.round=format.round,
                file="testdata" )

# read the data
dat1 <- miceadds::read.fwf2( file="testdata.dat", format.full=c(6,6,8),
               variables=c("x","y","z") )
# check differences between data frames
dat - dat1

#############################################################################
# EXAMPLE 2: Write datasets containing some string variables in fwf format
#############################################################################

n <- 5
dat <- data.frame( "x"=stats::runif(n, 0, 9 ), "y"=LETTERS[1:n] )
write.fwf2(dat, format.full=c(4,2), format.round=c(2,0),  file="testdata")

## End(Not run)

miceadds documentation built on May 29, 2024, 11:05 a.m.