Initial R code for looking at sqlbits tracks

sqlbitsRcode.R

# get html from url
library(RCurl)
library(XML)
library(data.table)
library(ggplot2)

url <- "http://sqlbits.com/information/PublicSessions.aspx"
src<-getURL(url)

# transform html
h.s<-htmlParse(src)

#extract session data
sessions<-data.table(session=xpathSApply(h.s,"//h4",xmlValue),
                     presenter=xpathSApply(h.s,"//a[@id='SpeakerName']",xmlValue),
                     Level=xpathSApply(h.s,"//div[@style='float: right; font-weight: bold; margin-left: 1em']",xmlValue),
                     Track=xpathSApply(h.s,"//div[@style='float: right; font-weight: bold']",xmlValue))

#clean session data
sessions[,session:=gsub(" ","",gsub("\r\n","", session))][
  ,Level:=gsub(" ","",gsub("\r\n","", Level))][
    ,Track:=gsub(" ","",gsub("\r\n","", Track))]

#start charting for kicks
ggplot(sessions,aes(x=Track,y=..count..,fill=Level, group=Level))+
  geom_histogram()+
  theme_minimal()+
  scale_fill_brewer()+
  labs(title="Submitted sessions")