# twitter tsuffs
require(twitteR)
require(plyr)

# ADD STUFF

install.packages(c("devtools", "rjson", "bit64", "httr"))

#RESTART R session!

library(devtools)
install_github("twitteR", username = "geoffjentry")
library(twitteR)

# SIGN IN

api_key <- "c9d9pLSiwpnZa5ghQqc3VSKO0"
api_secret <- "dDplH1upTc6zXrXF8GOoHRG5bJwVPb5tWYPkakpvHq138Wa6qN"
access_token <- "101609102-f0hXebcRZNEwZTGqn6uBoK8TdOlvi2b8q4mrOcOH"
access_token_secret <- "1pHo0cvIcqlG5PhzULhnC9JFzEhzdfdDopsQF8wUnQL37"
setup_twitter_oauth(api_key,api_secret,access_token,access_token_secret)

# the most frequent 100 English words that contain AO (open o), accorind to the CMU pronuncing dictionary: 
# for
# on
# or
# all
# your
# also 
# want
# because 

# search all varient spellings from CMU pronuncing dictionary 
words <- c("awn", "awr", "awll", "yawr", "awlso", "wawnt", "becawse")
TwitterData <- NULL
for(i in 1:length(words)){
  word <- searchTwitter(words[i], n=100, lang = "en")
  word.df = do.call("rbind",lapply(word,as.data.frame))
  TwitterData <- rbind(TwitterData, word.df)
}
  
write.csv(TwitterData, "TwitterData.csv")

# now to do the same with Scottish Data
words <- c("dae", "tae", "whae", "yae")
TwitterData <- NULL
for(i in 1:length(words)){
  word <- searchTwitter(words[i], n=100, lang = "en")
  word.df = do.call("rbind",lapply(word,as.data.frame))
  TwitterData <- rbind(TwitterData, word.df)
}

write.csv(TwitterData, "scotsTwitterData.csv")