readthat

CRAN status Lifecycle: experimental Travis build status Codecov test coverage

Quickly read text/source from local files and web pages.

Installation

You can install the development released version of readthat from Github with:

remotes::install_github("mkearney/readthat")

Examples

Let’s say we want to read-in the source of the following websites:

## a vector of URLs
urls <- c(
  "https://mikewk.com",
  "https://cnn.com",
  "https://www.cnn.com/us"
)

Use readthat::read() to read the text/source of a single file/URL

## read single web/file (returns text vector)
x <- read(urls[1])

## preview output
substr(x, 1, 60)
#> [1] "<!DOCTYPE html>\n<html lang=\"en\">\n<head>\n\n  <meta charset=\"ut"

## use apply functions to read multiple pages
xx <- sapply(urls, read)

## preview output
lapply(xx, substr, 1, 60)
#> $`https://mikewk.com`
#> [1] "<!DOCTYPE html>\n<html lang=\"en\">\n<head>\n\n  <meta charset=\"ut"
#> 
#> $`https://cnn.com`
#> [1] "<!DOCTYPE html><html class=\"no-js\"><head><meta content=\"IE=e"
#> 
#> $`https://www.cnn.com/us`
#> [1] "<!DOCTYPE html><html class=\"no-js\"><head><meta content=\"IE=e"

Comparisons

Benchmark comparison for reading a text file:

## save a text file
writeLines(read(urls[1]), x <- tempfile())

## coompare read times
bm_file <- bench::mark(
  readr = readr::read_lines(x),
  readthat = read(x),
  readLines = readLines(x),
  check = FALSE
)

## view results
bm_file
#> # A tibble: 3 x 6
#>   expression      min   median `itr/sec` mem_alloc `gc/sec`
#>   <bch:expr> <bch:tm> <bch:tm>     <dbl> <bch:byt>    <dbl>
#> 1 readr       262.2µs  273.9µs     3517.    3.69MB     4.05
#> 2 readthat     83.7µs   86.2µs    10741.   28.65KB     4.04
#> 3 readLines   144.8µs  150.7µs     6311.   13.16KB     0

Benchmark comparison for reading a web page:

x <- "https://www.espn.com/nfl/scoreboard"
bm_html <- bench::mark(
  httr = httr::content(httr::GET(x), as = "text", encoding = "UTF-8"),
  xml2 = xml2::read_html(x),
  readthat = read(x),
  readLines = readLines(x, warn = FALSE),
  readr = readr::read_lines(x),
  check = FALSE,
  iterations = 25,
  filter_gc = TRUE
)
bm_html
#> # A tibble: 5 x 6
#>   expression      min   median `itr/sec` mem_alloc `gc/sec`
#>   <bch:expr> <bch:tm> <bch:tm>     <dbl> <bch:byt>    <dbl>
#> 1 httr         71.5ms    104ms      6.84     2.7MB    0.285
#> 2 xml2        187.9ms    200ms      4.50    1.85MB    1.42 
#> 3 readthat     48.6ms     52ms     14.1     23.9KB    0    
#> 4 readLines   375.9ms    472ms      1.95  620.33KB    0    
#> 5 readr       158.6ms    169ms      5.57   799.7KB    0.232