cigrainger

import sys
import nltk
f = open('students.csv').read()
while f.split('\n')[0] != 'name,email':
    morecsv = raw_input("This does not look like the correct file. Would you like to see more? y/n?")
    while morecsv != 'y' and 'n':
        morecsv = raw_input("The prompt was y/n only. Would you like to see more? y/n?")
    if morecsv == 'y':
        print f.split('\n')[:10]
        newcsv = raw_input("Would you like to overwrite this file to create the student list? y/n?")

cigrainger

import sys
import nltk
f = open('students.csv').read()
while f.split('\n')[0] != 'name,email':
    morecsv = raw_input("This does not look like the correct file. Would you like to see more? y/n?")
    while morecsv != 'y' and 'n':
        morecsv = raw_input("The prompt was y/n only. Would you like to see more? y/n?")
    if morecsv == 'y':
        print f.split('\n')[:10]
        newcsv = raw_input("Would you like to overwrite this file to create the student list? y/n?")

cigrainger

def fuzzynames(x):
    fn = []
    for key in students.iterkeys():
        fn.append(nltk.metrics.edit_distance(x, key))
    if any(i > 5 for i in fn):
        while True:
            continue = raw_input("%s looks pretty close to %s. Are you sure this is a new student? " % (x,i))
            if continue.strip() == 'y' or 'n':
                break
        if continue == 'y':

cigrainger

def newstudent():
    newname = raw_input("What is the student's name? ")
    # newname = fuzzynames(newname)
    if newname == 'quit':
        return 'Okay! Nevermind.'
    if newname in students.keys():
        addnew = raw_input("%s is already in the list. Are you sure you want to add them again? y/n? " % newname)
        while addnew != 'y' and addnew!= 'n':
            addnew = raw_input("This is y/n only. %s is already in the list. Are you sure you want to add them again? y/n? " % newname)
        if addnew == 'y':

cigrainger

# Load required packages
library(reshape2)
library(zoo)
library(data.table)
library(ggplot2)
library(lattice)
library(foreach)
library(doParallel)

# Import data and prepare data frames

cigrainger

prep.comp <- function(x){
  if (names(x)[1]!='Week'){
    stop("The first column should be 'Week'. Check the data.")
  }
  x[['Week']] <- as.character(x[['Week']])
  x[['weekstart']] <- strsplit(x[['Week']],split=' - ',fixed=TRUE)[[1]]
  x[['weekend']] <- strsplit(x[['Week']],split=' - ',fixed=TRUE)[[2]]
  x[['weekstart']] <- as.Date(x[['weekstart']],format='%Y-%m-%d')
  x[['weekend']] <- as.Date(x[['weekend']],format='%Y-%m-%d')
  return(x)

cigrainger

setwd("/Users/thiemo/Dropbox/mafundo/")
is_installed <- function(mypkg) is.element(mypkg, installed.packages()[,1])

load_or_install<-function(package_names)
{

  for(package_name in package_names)
  {
    if(!is_installed(package_name))

cigrainger

library(foreign)
library(dplyr)
library(zoo)

df1 <- read.dta('~/AeroFS/Googlestuff/searchtechnology-collapsed.dta')
df2 <- read.dta('~/AeroFS/Googlestuff/mergedquarterly.dta')
load('googledates.rdata')

df1 <- left_join(df1,monthlookup,by='time')
df1$time <- NULL

cigrainger

prep.comp <- function(x,y){
  require(dplyr)
  if (names(x)[1]!='Week'){
    stop("The first column should be 'Week'. Check the data.")
  }
  countryname <- names(x)[2]
  x[['Week']] <- as.character(x[['Week']])
  x[['weekstart']] <- substr(x[['Week']],1,11)
  x[['weekend']] <- substr(x[['Week']],13,24)
  x[['weekstart']] <- as.Date(x[['weekstart']],format='%Y-%m-%d')

cigrainger

# Create function to get data from weekly to monthly
weekly.monthly <- function(x){
  require(dplyr)
  if (names(x)[1]!='Week'){
    stop("The first column should be 'Week'. Check the data.")
  }
  countryname <- names(x)[2]
  x[['Week']] <- as.character(x[['Week']])
  x[['weekstart']] <- substr(x[['Week']],1,11)
  x[['weekend']] <- substr(x[['Week']],13,24)