README.rdoc

README¶ ↑

One thing every public relations professional has had to deal with is compiling coverage reports. This process typically involves searching for articles and then pasting information piece-by-piece into an excel grid. Now Coverage Reportr helps to simplify that process by automating most of the data entry and allowing multiple users to collaborate on a single report. You can even download your reports to .csv and send them as attachments or use the data for other applications.

One of the coolest features of Coverage Reportr is it’s ability to pull information about an article just from the URL and to save an instance as a row in a .csv file. These are all provided by the methods availble to the “hit” class model:

“‘ruby

class Hit < ActiveRecord::Base
  belongs_to :coverage_report
  require 'open-uri'
  require 'uri'
  require 'csv'

AlchemyAPI.key = ENV.fetch('ALCHEMY_API_KEY')

  #This method parses the URL string to extract just the name of the publication.
  def self.get_publication(url)
    host = URI.parse(url).host
    if host[0] == "w"
      URI.parse(url).host.split(".")[1..-2].first
    else
      URI.parse(url).host.split(".").first
    end
  end

  #Here Nokogiri helps find the article title by seeking the "title" css selector, it works most of the time.
  def self.get_title(url)
    Nokogiri::HTML(open(url)).css('title').first.content
  end

  #AlchemyAPI is a logic-based text reader that intelligently identifies the author of an article.
  def self.get_author(url)
    AlchemyAPI::AuthorExtraction.new.search(:url => url)['author']
  end

  #It can even tell you whether the article is positive or negative in tone, which is an important feature for PR people.
  def self.get_sentiment(url)
    AlchemyAPI::SentimentAnalysis.new.search(:url => url)['type']
  end

  #This method extracts the full body of an article, but it's currently not utilized in the current feature-set.
  def self.get_text(url)
    AlchemyAPI::TextExtraction.new.search(:url => url)
  end

  #This method enables coverage reports to be downloaded in .csv format.
  def self.to_csv(hits)
    column_names = [:hit_publication, :hit_title, :hit_url, :hit_author, :hit_date, :hit_sentiment]
    CSV.generate do |csv|
        csv << column_names
        hits.each do |hit|
          csv << column_names.map { |key| hit[key] }
        end
    end
  end
end

“‘

Coverage Reportr takes advantage of the below technologies:

• Ruby version 2.1.1

• Rails version 4.1.1

• Text analysis by AlchemyAPI: www.alchemyapi.com/api/

• Additional text analysis by Nokogiri: nokogiri.org/

• Coverage Reportr uses PostgreSQL for database management: www.postgresql.org/

• Coverage Reportr uses Skeleton for responsive (and good looking) design: www.getskeleton.com/

And it has a few dependencies:

• Nokogiri gem: rubygems.org/gems/nokogiri

• the AlchemyAPI gem written by johnallen3d: github.com/technekes/alchemy-api-rb

Deployment Instructions:

• When running locally - simply initiate the rails server and point your browser to localhost:3000

• Or visit Coverage Reportr on Heroku: coverage-reportr.herokuapp.com/

Coverage Reportr Backlog:

• No effective date extraction function, therefore it automatically populates with today’s date.

• New York Times prevents Nokogiri from redirecting due to required log-in. Need to integrate NYT article API and add methods specifically for that site. This may be the case with other sites, but I have not come across them yet.

• When the URL is analyzed, some sites redirect the user directly to the original page:

• ESPN.com

• Cnet.com