anomia

The goal of anomia is to make it easy to give your R Markdown code chunks a name. Broadly, anomic aphasia is the inability to retrieve a word or the name of an object. Word selection anomia occurs when a person knows how to use an object and can correctly select the target object from a group of objects, and yet cannot name the object.

Installation

anomia works by converting an R Markdown document to XML, modifying the code blocks’ names, and then converting back to R Markdown. This package relies on tinkr which is not yet on CRAN.

To install anomia

remotes::install_github("ropensci/tinkr")
remotes::install_github("adknudson/anomia")

Examples

There are two primary ways to use this package. The first is to create a naming template and then use it to name the unnamed chunks in a .Rmd file. The second method is to give the naming spec in the YML header, and then use the RStudio plugin to (re)name chunks.

Using a Template

template <- anom_combo(
  pattern = list("colors",
                 c("animals", "fruits"),
                 "adverbs"),
  prefix = "001",
  suffix = NULL,
  sep = "-",
  case = "upper"
)

anom_name_chunks(path = "file.Rmd", template = template)

Using a YML Spec

The same arguments used to create a naming template can be used in the YML header file to specify a naming pattern. It will follow the hierarchy of anomia > type > arguments

---
title: "R Markdown Document"
output:
  html_document: default
anomia:
  combo:
    prefix: "001"
    pattern
      - colors
      - [animals, fruits]
      - adverbs
    sep: "-"
    case: "upper"
---

Why?

Naming your chunks is a good idea, especially if you use the caching feature while knitting. If no name is given, then knitr gives a default name of unnamed-chunk-XX. This seems fine, until a new chunks is inserted before others. This causes the names of chunks to be updated, which invalidates the cache and hence code chunks will be re-run to rebuild the cache. If a name is given, then rearranging chunks does not invalidate the cache.

Why not just name the chunks yourself? Because we’re programmers and any tedious task must be automated. Also not every chunk requires a meaningful name, so having a program generate names for you is a convenient way to not have to sweat the difficult task of coming up with a name on your own.