Getting Content from the "Zeit-online"

gettingZeitOnlineResults.R

#devtools::install_github("chgrl/diezeit")
library(diezeit)

require(ggplot2)
library(plyr)
require(stringr)
require(ggthemes)

options(stringsAsFactors = FALSE)

## set key .. get your own here: 
#http://developer.zeit.de/quickstart/
key="blubb"
Sys.setenv(ZEIT_KEY=key)


############################# 
#helper function
dataframeFromResult <- function(l) {
  l1 <- lapply(l, function(x) {
    x[sapply(x, is.null)] <- NA
    unlist(x)
  })
  keys <- unique(unlist(lapply(l1, names)))
  l2 <- lapply(l1, '[', keys)
  l3 <- lapply(l2, setNames, keys)
  res <- data.frame(do.call(rbind, l3))
  return(res)
}

##################### function to get results!
getZeitResults <- function(term){
  ##limitSearch <- zeit_search("content", term, print=FALSE, limit=3)
  unlimited <- zeit_search("content", term, print=FALSE, limit=1000)
  
  dff <- dataframeFromResult(unlimited$matches)
  dff$date <- as.Date(str_split_fixed(dff$release_date, "T",2)[,1])
  return(dff)
}


terms <- c("data mining", "big data", "maschinelles lernen", "data scientist", "artificial intelligence", "open data")

for(i in terms){
  dff <- getZeitResults(i)
  dff$Term <- i
  
  if(exists("out")){
    out <- rbind(out, dff)
  }
  else{
    out <- dff
  }
  
}


ggplot(out, aes(date)) + geom_bar() +theme_economist()  + ylab("Number of articles") + facet_wrap(~Term, scales="free")
ggplot(out, aes(Term))+ geom_bar() +theme_economist() + ylab("Overall number of articles") + xlab("")