NOTE This project has not been updated since R 3.4.2.
It has now been archived, as it may still be useful as reference, but no further maintenance is planned.

Wmisc - because everyone should have his own misc package for R

Huh?

This is a collection of helper functions I wrote for myself.

Tokenizer: A convenient method to read a file step-by-step with wrapper functions to read a file as a whole. For smaller files (a few dozen Megabytes) this is as fast as base::readLines() and Kmisc::readlines(). Quick benchmarks showed that for files of this size access times on mass storage (including caching by the OS) are magnitudes higher than actual read times. For big files the use case for Tokenizer is read-and-process/parse, reducing memory load compared to the other methods.
Automat: A tool to implement State Machines for transparent stateful computation. Idea sparked by glyph/automat. See the Vignette for details.
Strings: A few string functions that were once faster than built-in alternatives and can still be used for convenience with R 3.x onward (although not for performance gains).

I want it!

install.packages(c("Wmisc"))

if you prefer the latest stable version from GitHub:

install.packages(c("devtools")) # if you haven't done that before
devtools::install_github("wamserma/R-wmisc")

On Windows you will also need to install RTools.

Building the Package manually

clone the git repository

git clone https://github.com/wamserma/R-wmisc.git

install the build dependencies

install.packages(c("devtools", "hash", "knitr", "microbenchmark", "rmarkdown", "roxygen2", "testthat"))

if you want, install the optional dependencies

install.packages(c("covr", "DiagrammeR", "lintr"))

open project in RStudio and hit Ctrl + Shift + B

It is suggested to use a GCC 5 or newer release to enable overflow checks in C, but the package will also build with the GCC 4.6 used in RTools 3.3.