How to download a Wikipedia image for a species page

Question

I routinely want images of organisms to complement datasets and it would be great if I could pull out a species image for, say a bottlenose dolphin, given the Genus and species. I would then use this image in a K12 educational Shiny App similar to this one for students to explore the dataset. I found a way to get the URL and some page info with the WikipediR package, but I can't figure out how to extract the URL for the image in the sidebar.

require(WikipediR)
page_info("en","wikipedia",page="Tursiops truncatus")

I know there's a way (i.e. here), but I don't really understand how to make this work in R.

I don't think you need to use WikipediaR package or even the Wikimedia API. I would just suggest using rvest, similar to this question here: https://stackoverflow.com/questions/36202414/r-download-image-using-rvest — Stedy, Sep 13 '17 at 02:56
Thanks, so much for pointing me in this direction! This led me to what I was looking for! Cheers — mattador, Sep 13 '17 at 18:14

mattador · Accepted Answer · 2017-09-14T22:00:27.623

Thanks to Stedy's suggestion, I found a solution. Note there are 2 similarly named Wikipedia interface packages for R. This one uses WikipediR, not WikipediaR.

require(WikipediR); require(rvest)

#titles= vector of page name(s)
#res= desired width in pixels (220 px thumbnail by default)
#savedest= save destination (w terminal '/'); wd by default

getwikipic<-function(titles,res,savedest){
    if(missing(res)){res=220}
    if(missing(savedest)){savedest=NA}
  lapply(titles, function (ttl,...){
  d<-page_info("en","wikipedia",page=ttl,clean_response=T)
  url<-d[[1]]$fullurl
  wikipage<-html_session(url)
  imginfo<-wikipage %>% html_nodes("tr:nth-child(2) img")
  img.url<- imginfo[1] %>% html_attr("src")
  img.url<-paste0("https:",img.url)
  if(is.na(savedest)){
    savefilename<-paste0(ttl,".jpg")
    }else{savefilename<-paste0(savedest,ttl,".jpg")}

  if(res!=220){img.url<-gsub(220,res,img.url)}  

  download.file(img.url,savefilename)
  return(paste0("orig.file: ",basename(img.url)))#tell user original filename (or error)

  },res,savedest)#End lapply
}#End function

Alternatively, I created a GitHub repo with the code here. You can source and run this quite simply in R.

devtools::source_url("https://raw.githubusercontent.com/drwilkins/getwikipic/master/getwikipic.R")

titles<-c("numbat")
getwikipic(titles,1024)

Downloads this to your working directory pic

since your code isn't that long, could you please post (i.e. cut and paste) it here in addition to providing the Github link? SO strongly encourages making answers as self-contained as possible ... — Ben Bolker, Sep 13 '17 at 18:40

How to download a Wikipedia image for a species page

1 Answers1