scrape MEP Twitter accounts listed by Wikipedia 'Freedom of Panorama' campaign 2015: https://meta.wikimedia.org/wiki/Freedom_of_Panorama_in_Europe_in_2015/Contact_your_MEP – note: several accounts are inactive or have been replaced by newer ones

code.r

library(dplyr)
library(readr)
library(rvest)

h = html("https://meta.wikimedia.org/wiki/Freedom_of_Panorama_in_Europe_in_2015/Contact_your_MEP")

l = html_nodes(h, xpath = "//a[contains(@href, 'Contact_your_MEP/')]") %>%
  html_attr("href") %>%
  unique

d = data_frame()

for(i in l) {

  h = html(paste0("https://meta.wikimedia.org", i))

  t = html_nodes(h, xpath = "//a[contains(@href, 'twitter.com')]") %>%
    html_attr("href")

  l = html_nodes(h, xpath = "//a[contains(@href, 'twitter.com')]/preceding::a[1]") %>%
    html_attr("href")

  n = html_nodes(h, xpath = "//a[contains(@href, 'twitter.com')]/preceding::a[1]") %>%
    html_text()

  l = l[ !grepl("/tweet|/hashtag", t) ]
  n = n[ !grepl("/tweet|/hashtag", t) ]
  t = t[ !grepl("/tweet|/hashtag", t) ]

  d = rbind(d, data_frame(country = substr(i, 62, 63),
                          name = n, url = l, twitter = t))

  cat(substr(i, 62, 63), ":", length(t), "accounts\n")

}

write_csv(d, "~/Desktop/meps_twitter_2015-07-02.csv")

meps.csv