a few lines of code to load packages, download and scrape data, convert, sort, etc.

cleaning_characters.r

# removes characters from multiple data frame columns
# example removes * , % characters

dw <- data.frame(gsub("\\*|,|%", "", as.matrix(dw)), stringsAsFactors = FALSE)

download_DTA_export_CSV.r

# Download Quality of Government Basic dataset.
file = "data/qog-cs.txt"
if(!file.exists(file)) {
  if(!file.exists(dta <- "data/qog-cs.dta")) {
    url = "http://www.qogdata.pol.gu.se/data/qog_std_cs.dta"
    download(url, dta, mode = "wb")
  }
  write.csv(read.dta(dta), file)
}
# Open local copy.
qog <- read.csv(file, stringsAsFactors = FALSE, header = TRUE)

geocode_QOG_data.r

# basic countrycode scheme
# example with Quality of Government data

# Add geographic continents using UN country codes.
continent <- factor(countrycode(qog$ccodealp, "iso3c", "continent"))

load_packages.r

# installs them first if needed
# you could add quiet options

# Load packages.
packages <- c("ggplot2", "RCurl")
packages <- lapply(packages, FUN = function(x) {
  if(!require(x, character.only = TRUE)) {
    install.packages(x)
    library(x, character.only = TRUE)
  }
})

merge_with_map.r

# merge map data and dataset
# averaging over the map data seems to fill it better than merging

# Transpose data to map dataset.
stateShapes$Swing   <- by(dw$Obama_Swing, uniqueStates, mean)[stateShapes$region]

plot_from_datawrapper.r

## http://lemonde.fr/societe/article/2013/01/09/qui-approuve-le-mariage-homosexuel_1814480_3224.html

library(ggplot2)
library(reshape)

## Fetch data
str(x <- read.csv("http://datawrapper.de//chart/xw9NF/data", sep="\t"))
head(x)

## Variable names
names(x) <- c("party",gsub("X","",names(x)[2:7]))

## Reshape
head(x <- melt(x, id = "party", variable_name = "t"))

## Line chart
ggplot(x, aes(x=t, y=value, group=party, color=party)) + geom_line()

## didn't find how to code it with gvisLineChart

plot_ngrams.r

library(ngramr)
x = c("intellectual property rights", "property rights", "patents", "copyright", "digital copyright")
ggram(paste0("(", x, "*5)"),
      year_start = 1980,
      corpus = "eng_2012", 
      geom = "point")

plot_QOG_countries.r

require(countrycode)
require(ggplot2)
qog <- read.csv("data/qog_cs.csv", sep = ";")
qog$continent <- countrycode(qog$ccodealp, "iso3c", "continent")
qplot(data = qog, y = wdi_fr, x = log(wdi_gdpc), 
      color = continent, label = ccodealp, geom = "text")

reorder_factors.r

# orders factors based on another variable
# see ?reorder for other options

dw$State <- with(dw, reorder(State, Obama_Swing), ordered = TRUE)

reorder_variables.r

# http://stackoverflow.com/questions/12544888/is-there-an-equivalent-r-function-to-stata-order-command/12547400#12547400

move <- function(data,variable,before) {
  m <- data[variable]
  r <- data[names(data)!=variable]
  i <- match(before,names(data))
  pre <- r[1:i-1]
  post <- r[i:length(names(r))]
  cbind(pre,m,post)
}

# example
library(MASS)
data(painters)
str(painters)

# Move 'Expression' variable before 'Drawing' variable.
new <- move(painters,"Expression","Drawing")
View(new)

save_CSV_data.r

# uses downloader to save CSV spreadsheet
# example with Quality of Government data

# Download Quality of Government Basic dataset.
file = "data/QOG.Basic.CS.csv"
if(!file.exists(file)) {
  url = "http://www.qog.pol.gu.se/digitalAssets/1373/1373417_qog_basic_cs_csv_120608.csv"
  download(url, file, mode = "wb")
}
# Open local copy.
qog <- read.csv(file, stringsAsFactors = FALSE, header = TRUE)

save_google_spreadsheet.r

# uses RCUrl to download from CSV link
# example with Daily Kos data

# Create a filename for the dataset.
file = "data/DKOS.US.0812.txt"
# Store the address of the spreadsheet.
link = "https://docs.google.com/spreadsheet/pub?key=0Av8O-dN2giY6dEFCOFZ4ZnlKS0x3M3Y0WHd5aWFDWkE&output=csv"
# Download dataset.
if (!file.exists(file)) {
  message("Dowloading the data...")
  # Download and read HTML spreadsheet.
  html <- textConnection(getURL(link, ssl.verifypeer = FALSE))
  # Convert and export CSV spreadsheet.
  write.csv(read.csv(html), file)
}
# Open file.
dkos <- read.csv(file, stringsAsFactors = FALSE)
# Check result.
str(dkos)

save_PDF_file.r

# uses downloader to save documentation files
# example with Quality of Government codebook

# Download Quality of Government Basic codebook.
file = "data/QOG.Basic.codebook.pdf"
if(!file.exists(file)) {
  url = "http://www.qogdata.pol.gu.se/codebook/codebook_basic_20120608.pdf"
  download(url, file, mode = "wb")
}

scrape_bea_table.r

# gets real income growth data from the BEA website (for US and each state separately)

library(downloader)
url = "http://www.bea.gov/iTable/download.cfm?ext=csv&fid=95B00AEA0D0F4551F3FFE2DEF6A88527CAA3FD0670540E68A18BEE95735B3DBD9AAF7DB4AD82265972EC1C8BB81FC4EBCACF89FB20F318A84DD56C7E821137F8"
file = "real_inc.csv"
download(url, file)
file <- read.csv(file, skip = 4)
head(file)

scrape_html_data.r

# for vlad
# first take at your mission
library(foreach)
library(RCurl)
library(XML)

scrape <- function(x) {
  # for security wrap in try(textConnection())
  page <- getURL(x)
  html <- htmlTreeParse(page, useInternal = TRUE, encoding = "UTF-8")
  # extract title
  node <- getNodeSet(html, path = "//td[@class='headline']")
  text <- xmlValue(node[[1]])
  # results = address + title
  return(c(x,text))
}

url = "http://www.cdep.ro/pls/proiecte/upl_pck.proiect?cam=1&idp="

# test run on first 10 pages
data <- foreach(i = 1:10) %do% scrape(paste(url,i,sep=""))

scrape_wikipedia_table.r

# there is a more complex way with readHTMLTable(getNode(htmlParse, ...))
# example with electoral college, 5th table

# Electoral college votes, 2012.
url = "http://en.wikipedia.org/wiki/Electoral_College_(United_States)"
# Extract fifth table.
college <- readHTMLTable(url, which = 5, stringsAsFactors = FALSE)
# Keep first and last columns, removing total electors.
college <- data.frame(State = college[, 1], 
                      College = as.numeric(college[, 35]))
# Merge to main dataset.
dw <- merge(dw, college, by = "State")
# Check result.
str(dw)

tsv.r

# from Brendan O'Connor's "Better I/O routines"

util$read.tsv <- function(..., header=F, sep='\t', quote='', comment='', na.strings='', stringsAsFactors=FALSE) {
  # read.table() wrapper with default settings for no-nonsense, pure TSV
  # Typical use case is output from another program.
  # (R's defaults are more geared for human-readable datafiles, which is less
  # feasible for large-scale data anyway.)
  # These options are substantially faster than read.table() defaults.
  #   (see e.g. LINK)
  # stringsAsFactors is the devil.
 
  args = list(...)
  args$header = header
  if (!is.null(args$col.names)) {
    # read.delim() is not smart about this.  Yikes.
    args$header = FALSE
  }
  args$sep = sep
  args$quote = quote
  args$comment = comment
  args$stringsAsFactors = stringsAsFactors
  args$na.strings = na.strings
  do.call(read.delim, args)
}
 
util$write.tsv <- function(..., header=NA, col.names=F, row.names=F, sep='\t', na='', quote=F) {
  # 'header' to 'col.names' naming consistency with read.table()
  if (is.finite(header)) col.names = header
  write.table(..., col.names=col.names, row.names=row.names, sep=sep, na=na, quote=quote)
}