scrape_urls {archiveRetriever}

scrape_urls: Scraping Urls from the Internet Archive

Description

scrape_urls scrapes Urls of mementos and lower-level web pages stored in the Internet Archive using XPaths as default

Usage

scrape_urls(
  Urls,
  Paths,
  collapse = TRUE,
  startnum = 1,
  attachto = NULL,
  CSS = FALSE,
  archiveDate = FALSE,
  ignoreErrors = FALSE,
  stopatempty = TRUE,
  emptylim = 10,
  encoding = "UTF-8",
  lengthwarning = TRUE
)

Arguments

Urls

A character vector of the memento of the Internet Archive

Paths

A named character vector of the content to be scraped from the memento. Takes XPath expressions as default.

collapse

Collapse matching html nodes

startnum

Specify the starting number for scraping the Urls. Important when scraping breaks during process.

attachto

Scraper attaches new content to existing object in working memory. Object should stem from same scraping process.

CSS

Use CSS selectors as input for the Paths

archiveDate

Retrieve the archiving date

ignoreErrors

Ignore errors for some Urls and proceed scraping

stopatempty

Stop if scraping does not succeed

emptylim

Specify the number of Urls not being scraped until break-off

encoding

Specify a default encoding for the homepage. Default is 'UTF-8'

lengthwarning

Warning function for large number of URLs appears. Set FALSE to disable default warning.

Value

This function scrapes the content of mementos or lower-level web pages from the Internet Archive. It returns a tibble including Urls and the scraped content. However, a memento being stored in the Internet Archive does not guarantee that the information from the homepage can be actually scraped. As the Internet Archive is an internet resource, it is always possible that a request fails due to connectivity problems. One easy and obvious solution is to re-try the function.

Examples

## Not run: 
scrape_urls(
Urls = "https://web.archive.org/web/20201001000859/https://www.nytimes.com/section/politics",
Paths = c(title = "//article/div/h2//text()", teaser = "//article/div/p/text()"),
collapse = FALSE, archiveDate = TRUE)


## End(Not run)