The wrapr R package supplies a number of substantial programming tools, including the S3/S4 compatible dot-pipe, unpack/pack object tools, and many more. It also supplies a number of formatting and parsing convenience tools:

• qc() (“quoting concatenate”): quotes strings, giving value-oriented interfaces much of the incidental convenience of non-standard evaluation (NSE) interfaces.
• map_to_char(): prints maps and vectors as executable code..
• let(): allows proper value-oriented programming over non-standard evaluation (NSE) interfaces.

I am excited to share one more such convenience interface: bc() (“blank concatenate”). bc() takes a single string argument, parses it, and builds up a vector of the described values.

bc() is easy to demonstrate.

library(wrapr)

x <- 1
y <- 2

ls()  # result not easy to paste back into R
# [1] "x" "y"

# the bc() fix
bc('"x" "y"')
# [1] "x" "y"

# the map_to_char() fix
map_to_char(ls())
# [1] "c('x', 'y')"

I myself find these solutions a bit more convenient than the usual dump(), deparse(), dput(), eval(), or parse().

bc() requires the outer quotes, but not the internal quotes. That is: bc('x y') and bc('x,y') are also equivalent to c("x", "y").

And that is some of the tools that make using R results and error messages to progress on projects by producing new R code easier. These are small things, but they can help keep one in the flow when working with data. Also, a thank you to Emil Erik Pula Bellamy Begtrup-Bright for suggesting the new functionality.

Categories: Tutorials

Tagged as: Programming R rstats

jmount

Data Scientist and trainer at Win Vector LLC. One of the authors of Practical Data Science with R.