Elsevier invoice data check

script.R

library(tidyverse)
library(rcrossref)
library(janitor)
library(crminer)
my_dois <- readxl::read_xlsx("dois_elsevier.xlsx") %>%
  clean_names()
# call Crossref
cr_df <- cr_works(my_dois$digital_object_identifier_doi, .progress = "text")
# get CC licensed articles
cc_df <- cr_df$data %>%
  select(doi, license, link) %>%
  unnest(cols = c(license)) %>% 
  filter(grepl("creative", URL)) %>%
  rename(cc_license = URL, cc_content_version = content.version)
# get XML full-text links for CC-licensed articles
elsevier_tdm <- cc_df %>%
  unnest(cols = c(link)) %>%
  filter(content.type == "text/xml",
         intended.application == "text-mining",
         content.version == "vor")

# defining a parsing function
elsevier_parse <- function(tdm_url) {
  req <- crminer::crm_xml(tdm_url)
  email <- xml2::xml_find_first(req, "//ce:e-address") %>% 
    xml2::xml_text()
  oa_sponsor_name <- xml2::xml_find_first(req, "//d1:coredata//d1:openaccessSponsorName") %>% 
    xml2::xml_text()
  oa_sponsor_type <- xml2::xml_find_first(req, "//d1:coredata//d1:openaccessSponsorType") %>% 
    xml2::xml_text()
  oa_article <- xml2::xml_find_first(req, "//d1:coredata//d1:openaccessArticle") %>% 
    xml2::xml_text()
  oa_type <- xml2::xml_find_first(req, "//d1:coredata//d1:openaccessType") %>% 
    xml2::xml_text()
  oa_archive <- xml2::xml_find_first(req, "//d1:coredata//d1:openArchiveArticle") %>% 
    xml2::xml_text()
  data_frame(email,
             oa_sponsor_name, oa_sponsor_type, oa_article, oa_type, oa_archive, 
             tdm_url)
}
# apply function on every XML full-text link to mine open access invoice data.
elsevier_oa_info <- purrr::map(elsevier_tdm$URL, .f = purrr::safely(elsevier_parse))
# get data frame were the full-text API call was successful
elsevier_oa_info_df <- purrr::map_df(elsevier_oa_info, "result")
# merge it with data frame and export it 
els_df <- inner_join(elsevier_oa_info_df, elsevier_tdm, by = c("tdm_url" = "URL")) %>% 
  select(1:10) 
left_join(my_dois, els_df, by = c("digital_object_identifier_doi" = "doi")) %>%
  write_csv("elsevier_output.csv")