Retrieve random CC BY 4.0 articles from Elsevier as PDF

fetch_elsevier.R

library(rcrossref)
library(purrr)
# helper function to retrieve pdfs with curl, requires api key
elsevier_fetch <- function(doi) {
	u <- paste0("http://api.elsevier.com/content/article/doi/",
		doi,
		"?httpAccept=application%2Fpdf"
		)
	file_out <- URLencode(paste0(doi, ".pdf"), reserved = TRUE)
	file.create(file_out)
	download.file(u, destfile = file_out)
	Sys.sleep(2)
}

# get 100 random works from Elsevier (Crossref member 78) with
# open license http://creativecommons.org/licenses/by/4.0/
dois <- rcrossref::cr_r(sample = 100, filter = c(member = "78", 
	license.url	= "http://creativecommons.org/licenses/by/4.0/"))

purrr::map(dois, elsevier_fetch)