README.md

In hrbrmstr/knitrengines: Collection of Language Engines for Use with the 'knitr' Package

knitrengines

Collection of Language Engines for Use with the ‘knitr’ Package

Description

The knitr package lets you use other languages in knitr ‘chunks’. This package is a collection of languages processors not built-in with knitr.

What’s Inside The Tin

The following functions are implemented:

• knitr_cawk_engine: knitr engine for the CSV-enhanced Awk language
• knitr_elixir_engine: knitr engine for the Elixir language
• knitr_gnuplot_engine: knitr engine for the Go language
• knitr_pygments_engine: knitr engine for the pygments syntax highlighting (w/o execution)

Come Again?

knitrengines is package to collect and seamlessly add new language engines to knitr

You can thank Wendy Smoak for this package as a comment on a DDS post sparked it.

The knitr package already has support for a plethora of languages besides R code chunks (26 as of the last update to this package). That is probably sufficient for the vast majority of users.

However, if you need to perform some processing in another languages and want to include it in your reproducible workflow, this package will allow you to incorporate those language code chunks provided there is a matching knitr language processor available.

To use one of these alternate code chunks, just ensure you have a call to library(knitrengines) in an R code chunk at/near the top of your R markdown file then use one of the available engines via the short name below as the package will auto-register them on attach.

This package contains support for the following engines:

• cawk: Support for Awk builds that include CSV support (see ?knitr_cawk_engine).
• elixir : Elixir language support
• pygments : Use pygments to stylize non-executable code blocks. See knitrengines::knitr_pygments_engine for all the “gotchas”
• gnuplot : Use gnuplot for plotting

You can contribe to the project and add support for other language code chunks by:

• forking this repo
• adding a new xyz_engine.r under the R directory ensuring you add yourself as an @author
• adding that engine to the list in zzz.r
• update DESCRIPTION and add yourself as a contributor and update the third dottedn number in the version string (i.e. 0.2.0 -> 0.2.1)
• update inst/examples/knitr_engine_test.Rmd and add a (small) example of your engine
• update the README.Rmd to include your new engine.
• submit a PR

The only real downside is that these language chunks do not have access to the variables in/across chunks, so you have to export the data from previous chunks to files (or databases, etc.) to access it (if needed).

Before you go creating other engines, these are the ones knitr already:

sort(names(knitr::knit_engines$get()))
##  [1] "asis"      "asy"       "awk"       "bash"      "block"     "block2"    "bslib"     "c"         "cat"      
## [10] "cc"        "coffee"    "comment"   "css"       "ditaa"     "dot"       "embed"     "exec"      "fortran"  
## [19] "fortran95" "gawk"      "glue"      "glue_sql"  "gluesql"   "go"        "groovy"    "haskell"   "highlight"
## [28] "js"        "julia"     "lein"      "mermaid"   "mysql"     "node"      "octave"    "ojs"       "perl"     
## [37] "psql"      "python"    "R"         "Rcpp"      "Rscript"   "ruby"      "sas"       "sass"      "scala"    
## [46] "scss"      "sed"       "sh"        "sql"       "stan"      "stata"     "targets"   "tikz"      "verbatim" 
## [55] "zsh"

Installation

remotes::install_github("hrbrmstr/knitrengines")

NOTE: To use the ‘remotes’ install options you will need to have the {remotes} package installed.

hrbrmstr/knitrengines documentation built on Sept. 17, 2023, 2:29 a.m.