Script that looks up and plots the language popularity based on the stack-overflow tag-count.

language-popularity-on-stack-overflow.R

require(RColorBrewer)
require(ggplot2)
require(scales)

# Thanks to drew conway for this handy function.
get.stack<-function(tok) {
    # Must check for XML install, thanks onertipaday!
    if (!require(XML)) install.packages('XML')
    library(XML)
    # Enter a SO tag as character string, and number of tags are returned
    tok<-gsub("(/| )","-",tok)
    tok<-gsub("#","%23",tok,fixed=TRUE)
    base.stack<-"http://stackoverflow.com/questions/tagged/"
    stack.tree<-htmlTreeParse(paste(base.stack,tok,sep=""),useInternalNodes=TRUE)
    tag.count<-getNodeSet(stack.tree,"//div[@class='module']/div[@class='summarycount al']")
    tag.num<-suppressWarnings(as.numeric(gsub(",","",xmlValue(tag.count[[1]]),fixed=TRUE)))
    if(is.na(tag.num)) {
        warning(paste("Something went wrong trying to parse '",tok,"'.\nNA returned",sep=""))
    }
    return(tag.num)
}

# Feel free to add more if you want.
languages <- c(
    'python', 'php', 'ruby', 'nodejs', 'erlang', 
    'coldfusion', 'java', 'haskell', 'c#', 'objc',
    'lisp', 'go', 'r', 'assembly', 'scala', 
    'actionscript', 'shell', 'javascript')

results <- data.frame()

for (lang in 1:length(languages)) {
    language         <- languages[lang]
    languageStackPop <- get.stack(language)
    results          <- rbind.data.frame(
        results, 
        data.frame(name = language, popularity = languageStackPop)
    )
}

ggplot(results, aes(x=reorder(name, popularity), y=popularity)) + 
    geom_bar() + 
    geom_text(data=results,aes(x=name,y=popularity,label=popularity),hjust=-0.25) +
    coord_flip() + 
    theme_bw() +
    ylab("Number of Questions on StackOverflow") +
    xlab("Language Name") +
    labs(title="Language Popularity by Number of tags on StackOverflow") +
    scale_y_continuous(labels=comma)