get_emails.rb

# This code was used for the following blog post:
#
# http://www.alexejgossmann.com/email_data_analysis/
#
# Part of the code is based on: http://www.johnantony.com/sending-and-receiving-outlook-email-via-ruby/
# All private information (such as email addresses and passwords) was replaced with ***********

require 'mail'
require 'daru'

#--- configure delivery and retrieval methods

Mail.defaults do
  delivery_method :smtp, { 
                           :address              => 'smtp.office365.com',
                           :port                 => 587,
                           :domain               => 'tulane.edu',
                           :user_name            => '*******@tulane.edu',
                           :password             => '**********',
                           :authentication       => :login,
                           :enable_starttls_auto => true  
                           }

  retriever_method :imap, { :address             => 'outlook.office365.com',
                            :port                => 993,
                            :user_name           => '*******@tulane.edu',
                            :password            => '**********',
                            :enable_ssl          => true }
end
 
#--- retrieve all emails from INBOX

emails = Mail.all(read_only: true)
puts "Number of emails retrieved: #{emails.length}"

#--- extract all emails from PhD advisors

emails_from_advisors = emails.select do |email|
  /(*******@tulane.edu|*******@tulane.edu|*******@gmail.com|*******@tulane.edu)/i =~ email.from.to_s
end

puts "Number of emails received from all advisors: #{emails_from_advisors.length}"

#--- put the emails into a data frame and save it

n = emails_from_advisors.length
emails_from_advisors_df = Daru::DataFrame.new( { 'subject' => Array.new(n), 
                                     'from' => Array.new(n), 
                                     'to' => Array.new(n),
                                     'cc' => Array.new(n),
                                     'bcc' => Array.new(n),
                                     'date' => Array.new(n),
                                     'messageid' => Array.new(n),
                                     'replyto' => Array.new(n),
                                     'inreplyto' => Array.new(n),
                                     'body' => Array.new(n) } )

emails_from_advisors.each_with_index do |email, ind|
    # email fields defined at https://github.com/mikel/mail/tree/master/lib/mail/fields
    emails_from_advisors_df['subject'][ind] = email.subject.to_s.gsub(",", ";")
    # (commas are replaced with semicolons in order to write csv later)
    emails_from_advisors_df['from'][ind] = email.from.to_s.gsub(",", ";")
    emails_from_advisors_df['to'][ind] = email.to.to_s.gsub(",", ";")
    emails_from_advisors_df['cc'][ind] = email.cc.to_s.gsub(",", ";")
    emails_from_advisors_df['bcc'][ind] = email.bcc.to_s.gsub(",", ";")
    emails_from_advisors_df['date'][ind] = email.date.to_s.gsub(",", ";")
    emails_from_advisors_df['messageid'][ind] = email.message_id.to_s.gsub(",", ";")
    emails_from_advisors_df['replyto'][ind] = email.reply_to.to_s.gsub(",", ";")
    emails_from_advisors_df['inreplyto'][ind] = email.in_reply_to.to_s.gsub(",", ";")
    #emails_from_advisors_df['body'][ind] = email.body.to_s.gsub(",", ";")
end

emails_from_advisors_df.write_csv "emails_from_advisors.csv"

#--- retrieve all emails from "Sent Items"

sent = Mail.all(mailbox: "Sent Items", read_only: true)
puts "Number of emails retrieved: #{sent.length}"

#--- extract all emails to PhD advisors

emails_to_advisors = sent.select do |email| 
  /(*******@tulane.edu|*******@tulane.edu|*******@gmail.com|*******@tulane.edu)/i =~ email.to.to_s ||
  /(*******@tulane.edu|*******@tulane.edu|*******@gmail.com|*******@tulane.edu)/i =~ email.cc.to_s ||
  /(*******@tulane.edu|*******@tulane.edu|*******@gmail.com|*******@tulane.edu)/i =~ email.bcc.to_s
end

puts "Number of emails sent to all advisors: #{emails_to_advisors.length}"
 
#--- put the emails sent to PhD advisors into a data frame and save it

n = emails_to_advisors.length
emails_to_advisors_df = Daru::DataFrame.new( { 'subject' => Array.new(n), 
                                           'from' => Array.new(n), 
                                           'to' => Array.new(n),
                                           'cc' => Array.new(n),
                                           'bcc' => Array.new(n),
                                           'date' => Array.new(n),
                                           'messageid' => Array.new(n),
                                           'replyto' => Array.new(n),
                                           'inreplyto' => Array.new(n),
                                           'body' => Array.new(n) } )

emails_to_advisors.each_with_index do |email, ind|
    # email fields defined at https://github.com/mikel/mail/tree/master/lib/mail/fields
    emails_to_advisors_df['subject'][ind] = email.subject.to_s.gsub(",", ";")
    # (commas are replaced with semicolons in order to write csv below)
    emails_to_advisors_df['from'][ind] = email.from.to_s.gsub(",", ";")
    emails_to_advisors_df['to'][ind] = email.to.to_s.gsub(",", ";")
    emails_to_advisors_df['cc'][ind] = email.cc.to_s.gsub(",", ";")
    emails_to_advisors_df['bcc'][ind] = email.bcc.to_s.gsub(",", ";")
    emails_to_advisors_df['date'][ind] = email.date.to_s.gsub(",", ";")
    emails_to_advisors_df['messageid'][ind] = email.message_id.to_s.gsub(",", ";")
    emails_to_advisors_df['replyto'][ind] = email.reply_to.to_s.gsub(",", ";")
    emails_to_advisors_df['inreplyto'][ind] = email.in_reply_to.to_s.gsub(",", ";")
    #emails_to_advisors_df['body'][ind] = email.body.to_s.gsub(",", ";")
end

emails_to_advisors_df.write_csv "emails_to_advisors.csv"