readChar() usage on CRAN

cran_readchar_args.R

library(jsonlite)
library(data.table)
library(lintr)

read_page <- function(page) {
  tmp <- tempfile()
  on.exit(unlink(tmp))
  system2("curl",
    c("--location", "--request",
      "GET", sprintf("'https://api.github.com/search/code?q=readChar+org:cran+language:R+-path:.Rd&per_page=100&page=%d'", page),
      "--header", sprintf('"Authorization: Bearer %s"', Sys.getenv("GITHUB_PAT")),
      "--header", '"X-GitHub-Api-Version: 2022-11-28"'),
    stdout=tmp)
  jsonlite::fromJSON(tmp)
}

n_pages <- Inf
page <- 1L

results <- list()

while (page <= n_pages) {
  Sys.sleep(0.2) # rate-limiting
  page_data <- read_page(page)
  if (page == 1L) n_pages <- ceiling(page_data$total_count / 100)
  page <- page + 1L

  results[[page]] <- data.frame(
    file_path = page_data$items$path,
    repo = page_data$items$repository$full_name
  )
}

results <- rbindlist(results)
results[, raw_url := sprintf("https://raw.githubusercontent.com/%s/master/%s", repo, file_path)]
setorder(results, raw_url)

full_file_xpath_fmt <- "
  //SYMBOL_FUNCTION_CALL[text() = 'readChar']
    /parent::expr
    /following-sibling::expr[expr[%s]]
"
read_char_linter <- undesirable_function_linter(c(readChar = "any readChar() usage"))
file_size_linter <- make_linter_from_xpath(
  sprintf(full_file_xpath_fmt, "SYMBOL_FUNCTION_CALL[text() = 'file.size']"),
  "readChar() with file.size()")
file_info_linter <- make_linter_from_xpath(
  sprintf(full_file_xpath_fmt, "expr/SYMBOL_FUNCTION_CALL[text() = 'file.info']"),
  "readChar() with file.info()$size")

lint_info <- rbindlist(lapply(results$raw_url, function(url) {
  Sys.sleep(0.2)
  download.file(url, tmp <- tempfile(), quiet = TRUE)
  on.exit(unlink(tmp))
  
  lints <- as.data.table(lint(tmp,
    linters = list(
      n_calls = read_char_linter,
      n_file_size = file_size_linter(),
      n_file_info = file_info_linter())))
  lints[, raw_url := url]
  lints
}))

results <- merge(results,
  dcast(lint_info[linter != "error"], raw_url ~ linter, fun.aggregate = length),
  by = "raw_url")