This code lays out the steps necessary to download and analyze Twitter data from #isa2014.

gistfile1.r

library("scales")
library("Cairo")
library("ROAuth")
library("twitteR")
library("ggplot2")

#####################################
# Setting up the Twitter Connection #
#####################################

# the following header bits are taken from this blogpost: 
# http://davetang.org/muse/2013/04/06/using-the-r_twitter-package/

# houskeeping
rm(list=ls())

# necessary step for Windows
download.file(url="http://curl.haxx.se/ca/cacert.pem", destfile="cacert.pem")

#to get your consumerKey and consumerSecret see the twitteR documentation for instructions
cred <- OAuthFactory$new(consumerKey='',
                         consumerSecret='',
                         requestURL='https://api.twitter.com/oauth/request_token',
                         accessURL='https://api.twitter.com/oauth/access_token',
                         authURL='https://api.twitter.com/oauth/authorize')

# necessary step for Windows
cred$handshake(cainfo="cacert.pem")

# save for later use for Windows
save(cred, file="twitter authentication.Rdata")
registerTwitterOAuth(cred)

#################
# Analysis part #
#################

# NOT RUN:

# get all #isa2014 tweets & convert to data frame
# isa2014 <- searchTwitter("#isa2014", n=3500, cainfo="cacert.pem", since="2014-03-24", until="2014-03-31")
# isa2014df <- twListToDF(isa2014)
# 
# names <- unique(sort(isa2014df$screenName))
# 
# fulldf <- isa2014df[-(1:nrow(isa2014df)), ]
# 
# for(name in names[434:length(names)]) {
#   print(paste("Retrieving tweets from", name, "..."))
#   usertweets <- searchTwitter(paste(name, " #isa2014"), 500, cainfo="cacert.pem", since="2014-03-24", until="2014-03-31")
#   usertweets <- twListToDF(usertweets)
#   
#   fulldf <- rbind(fulldf, usertweets)
# }


# plot twenty busiest tweeps
df <- data.frame(table(isa2014df$screenName))
df <- tail(df[order(df$Freq), ], 20)

# get real names
for(i in 1:nrow(df)) {
  df[i, "realname"] <- getUser(df[i, "Var1"],  cainfo="cacert.pem")$name  
}

# create display
df$disp_name <- paste0(df$realname, " \n(@", df$Var1, ")")

CairoPNG("busiest_isa2014_tweeps_NEW.png", height=1600, width = 1000, pointsize=30)
par(mar=c(5, 8, 4, 2))
barplot(df$Freq, 
        names.arg=df$disp_name, 
        horiz=T, 
        las=1,
        main="20 Busiest #ISA2014 Tweeps",
        xlab="Tweet count",
        space=0.6, 
        col="navy", 
        border = FALSE, cex.names = .6)

dev.off()

#######################
# most popular tweets #
#######################

# add popular count
isa2014df$popular <- (isa2014df$favoriteCount + isa2014df$retweetCount)

# order & subset top 10
ordered <- isa2014df[order(isa2014df$popular, decreasing = TRUE), ]
top20tweets <- head(ordered[ordered$isRetweet == FALSE, ], 10)

# generate full twitter link
top20tweets$link <- paste0("https://twitter.com/", top20tweets$screenName, "/status/", top20tweets$id)

# write list of tweet links for c & p inclusion to wordpress
write.table(top20tweets[, "link"], row.names=FALSE, file="isa_twitter.txt", quote=F, col.names=FALSE)


####################################
# isatw14 "Twitter Panel" Analysis #
####################################

isatw14 <- searchTwitter("#isatw14", n=500, cainfo="cacert.pem")
isatw14df <- twListToDF(isatw14)

df_tw14 <- data.frame(table(isatw14df$screenName))
df_tw14 <- tail(df_tw14[order(df_tw14$Freq), ], 20)

# get real names
for(i in 1:nrow(df_tw14)) {
  df_tw14[i, "realname"] <- getUser(df_tw14[i, "Var1"],  cainfo="cacert.pem")$name  
}

# create display
df_tw14$disp_name <- paste0(df_tw14$realname, " \n(@", df_tw14$Var1, ")")

CairoPNG("busiest_twitter_panel_tweeps.png", height=1600, width = 1100, pointsize=30)
par(mar=c(5, 8, 4, 2))
barplot(df_tw14$Freq, 
        names.arg=df_tw14$disp_name, 
        horiz=T, 
        las=1,
        main="Busiest #isatw14 'The Twitter Panel' Tweeps",
        xlab="Tweet count",
        space=0.6, 
        col="navy", 
        border = FALSE, cex.names = .6)
dev.off()

# when do ISAlers tweet?

# some code taken from here:
# http://bommaritollc.com/2012/05/21/charting-twitter-time-series-data-with-tweet-and-unique-user-counts/

dt <- 30

isa2014df$date <- as.POSIXct(strptime(isa2014df$created, "%Y-%m-%d %H:%M:%S", tz="GMT"))

# fix dates
isa2014df$date <- isa2014df$date - (4*60*60)


minDate <- min(isa2014df$date)
maxDate <- max(isa2014df$date) + 60 * dt
dateBreaks <- seq(minDate, maxDate, by=60 * dt)

tweetCount <- hist(isa2014df$date, breaks=dateBreaks, plot=FALSE)                             
binBreaks <- tweetCount$breaks[1:length(tweetCount$breaks)-1]

plotData <- data.frame(dates=dateBreaks[1:length(dateBreaks)-1], tweets=as.numeric(tweetCount$count))

# time <- data.frame(xmin = as.POSIXct("2014-03-26 08:15"), xmax = as.POSIXct("2014-03-26 10:00"), ymin=-Inf, ymax=Inf)
                       
CairoPNG("ISA_Twitter_trend.png", width=1600, height=900, pointsize=30)
ggplot(plotData) +
  geom_bar(aes(x=dates, y=tweets), stat="identity") +
  scale_y_continuous("Number of tweets") +
  scale_x_datetime(breaks="3 hour", labels = date_format("%d. March %Hh")) +
  theme_bw() +
  theme(axis.text.x=element_text(hjust=1.1, angle=45), legend.key=element_blank())  +
  labs(x="", title="#isa2014 Twitter usage over time \n") # +
  # geom_rect(data=time, aes(xmin=xmin, xmax=xmax, ymin=ymin, ymax=ymax), color="grey20", alpha=0.5, inherit.aes = FALSE)
dev.off()