Package 'rmdwc'

Count text elements in Jupyter Notebook files

Description

This function extracts text from specific cell types (e.g., markdown) in one or more .ipynb files and counts the number of characters, words, and lines. It optionally excludes certain patterns (e.g., code fences). The function uses a helper function rmdcount() to perform the counting on the extracted text.

Usage

ipynbcount(
  files,
  celltype = c("markdown"),
  space = "[[:space:]]",
  word = "[[:space:]]+",
  line = "\n",
  exclude = "```\\{.*?```"
)

Arguments

files	character: vector of paths to .ipynb (Jupyter Notebook) files.
celltype	character: vector indicating which cell types to include (default is 'markdown'). Valid values include 'markdown' and 'code'.
space	character: pattern to split a text at spaces (default: '[[:space:]]')
word	character: pattern to split a text at word boundaries (default: '[[:space:]]+')
line	character: pattern to split lines (default: '\n')
exclude	character: pattern to exclude text parts, e.g. code chunks (default: '```\\{.*?```')

Details

This function assumes that the notebook files are valid JSON and contain a list of cells under the cells field. It temporarily writes the extracted content to a file to reuse the rmdcount() logic.

Value

A data frame with counts of characters, words, and lines for each file. Additional columns include file (base name) and path (directory).

Examples

file <- system.file('ipynb/example_data_analysis.ipynb', package="rmdwc")
ipynbcount(file)                                   # without code
ipynbcount(file, celltype=c("markdown", "code"))   # with code

---

Word, character and non-whitespace characters count

Description

rmdcount counts lines, words, bytes, characters and non-whitespace characters in R Markdown files excluding code chunks. txtcount counts lines, words, bytes, characters and non-whitespace characters in plain text files.
Note that the counts may differ a bit from unix wc and Libre Office because it depends on the definition of a line, a word and a character.

Usage

rmdcount(
  files = NULL,
  space = "[[:space:]]",
  word = "[[:space:]]+",
  line = "\n",
  exclude = "```\\{.*?```"
)

txtcount(
  files = NULL,
  space = "[[:space:]]",
  word = "[[:space:]]+",
  line = "\n"
)

Arguments

files	character: file name(s)
space	character: pattern to split a text at spaces (default: '[[:space:]]')
word	character: pattern to split a text at word boundaries (default: '[[:space:]]+')
line	character: pattern to split lines (default: '\n')
exclude	character: pattern to exclude text parts, e.g. code chunks (default: '```\\{.*?```')

Details

We define:

Line

the number of lines. It differs from unix wc -l since wc counts the number of newlines.

Word

it is considered to be a character or characters delimited by white space. However, a "word" is in general a fuzzy concept, for example is "3.141593" a word? Therefore different programs may count differently, for more details see the discussion to the Libreoffice bug Word count gives wrong results - Another Example Comment 5.

The following approach is used to detect lines, words, characters and non-whitespace characters.

lines

strsplit(rmd, line)[[1]] with line='\n'

bytes

charToRaw(rmd)

words

strsplit(rmd, word)[[1]] with word='[[:space:]]+'

characters

strsplit(rmd, '')[[1]]

non-whitespace characters

strsplit(gsub(space, '', rmd), '')[[1]] with space='[[:space:]]'

If txtcount is used then code chunks are deleted with gsub('```\\{.*?```', '', rmd) before counting.

Value

a data frame with following elements

file

basename of file

lines

number of lines

words

number of words

bytes

number of bytes

chars

number of characters

nonws

number of non-whitespace characters

path

path of file

Examples

# count excluding code chunks
files <- system.file('rmarkdown/rstudio_pdf.Rmd', package="rmdwc")
rmdcount(files)
# count including code chunks
txtcount(files) # or rmdcount(files, exclude='')
# count for a set of R Markdown docs
files <- list.files(path=system.file('rmarkdown', package="rmdwc"), 
                    pattern="*.Rmd", full.names=TRUE)
rmdcount(files)
# use of rmdcount() in a R Markdown document 
if (interactive()) {
  files <- system.file('rmarkdown/rstudio_pdf.Rmd', package="rmdwc")
  file.edit(files) # SAVE(!) the file and knit it 
}
# count including code chunks
files <- system.file('rmarkdown/rstudio_pdf.Rmd', package="rmdwc")
txtcount(files)

---

rmdcountAddin

Description

Applies rmdcount to the current R Markdown document

Usage

rmdcountAddin()

Value

nothing

Examples

if (interactive()) rmdcountAddin()

---

Word-, character and non-whitespace characters count for a text

Description

Counts words, characters and non-whitespace characters in a string. Is used in rmdcount, see details there.

Usage

rmdwcl(rmd, space = "[[:space:]]", word = "[[:space:]]+", line = "\n")

Arguments

rmd	character: R Markdown document as string
space	character: pattern to split a text at spaces (default: '[[:space:]]')
word	character: pattern to split a text at word boundaries (default: '[[:space:]]+')
line	character: pattern to split lines (default: '\n')

Value

a list

Examples

file  <- system.file('rmarkdown/rstudio_pdf.Rmd', package="rmdwc")
fcont <- readChar(file, file.info(file)$size)
rmdwcl(fcont)

---