ocr_challenge 1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 53334886a5c1796fb3838f8eb11809185340b6a8
+  data.tar.gz: 94e10225727f7d9415f0f00695758d32d534a994
+SHA512:
+  metadata.gz: 807ce07f9ea34cb158d63705570ff4bece9c96d833c241940947f7540a78a09fdb8fa1ec12fc00bea3dc96a5dd9437411aa7d1b3ce3ab6ae5adcb6d8dd44ddc7
+  data.tar.gz: 2c710c70ab22749f553827c8dd79bf29578532af3a2d0bad0a271db379b2778ac28cd81962c91e471d64876ae2bcaecb4b66bf61dfeff5b1de05d0b782602f69

data/.gitignore

@@ -0,0 +1,3 @@
+.idea
+coverage/
+*.gem

data/Gemfile

@@ -0,0 +1,3 @@
+source 'https://rubygems.org'
+gemspec
+

data/Gemfile.lock

@@ -0,0 +1,36 @@
+PATH
+  remote: .
+  specs:
+    ocr_challenge (1.0)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    diff-lcs (1.2.5)
+    docile (1.1.5)
+    faker (1.4.3)
+      i18n (~> 0.5)
+    i18n (0.6.11)
+    multi_json (1.10.1)
+    rspec (2.99.0)
+      rspec-core (~> 2.99.0)
+      rspec-expectations (~> 2.99.0)
+      rspec-mocks (~> 2.99.0)
+    rspec-core (2.99.2)
+    rspec-expectations (2.99.2)
+      diff-lcs (>= 1.1.3, < 2.0)
+    rspec-mocks (2.99.2)
+    simplecov (0.9.1)
+      docile (~> 1.1.0)
+      multi_json (~> 1.0)
+      simplecov-html (~> 0.8.0)
+    simplecov-html (0.8.0)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  faker (~> 1.4)
+  ocr_challenge!
+  rspec (~> 2.99)
+  simplecov (~> 0.9)

data/LICENSE

@@ -0,0 +1,22 @@
+The MIT License (MIT)
+
+Copyright (c) 2014 Alexander Vanadio
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+

data/README.md

@@ -0,0 +1,81 @@
+## Business Card OCR Challenge - [defined here](http://www.asymmetrik.com/programming-challenges/business-card-ocr.html)
+
+### Overview
+This gem allows you to take a new line delimited String, and pull out names, email addresses, and phone numbers. The String input can come from a file, STDIN, the output of another program, etc.
+
+This concept has been specifically applied to business cards, but can easily be used more generically.
+
+### Installation
+
+```shell
+gem install ocr_challenge
+```
+or in your Gemfile
+
+```ruby
+gem 'ocr_challenge'
+```
+
+### Usage
+
+1.) Create a contact with your input String
+
+```ruby
+require 'ocr_challenge'
+include OcrChallenge
+
+text = """
+    Alexander Vanadio\n
+    execdd17@gmail.com\n
+    (123)-456-7890\n
+"""
+
+contact = IBusinessCardParser.get_contact_info(text)
+```
+
+2.) Get the information through your contact instance
+
+```ruby
+contact.get_name            # =>  "Name: Alexander Vanadio"
+contact.get_email_address   # => "Email: execdd17@gmail.com"
+contact.get_phone_number    # => "Phone: 123-456-7890"
+
+contact.to_s # => "Name: Alexander Vanadio\nEmail: execdd17@gmail.com\nPhone: 123-456-7890"
+```
+
+### Advanced Usage
+
+Let's use a more complicated String input. The IBusinessCard parser will attempt to find all email addresses and phone numbers, but not fax numbers. Once it does, you can get them directly from your IContactInfo instance.
+
+```ruby
+require 'ocr_challenge'
+include OcrChallenge
+
+text = """
+    Alexander Vanadio\n
+    Software Engineer
+    My Company Name
+    execdd17@gmail.com\n
+    anotherEmail@gmail.com\n
+    Phone: 1-(123)-456-7890\n
+    Cell: 123.444.7890\n
+    Fax: 892-234-5467
+"""
+
+contact = IBusinessCardParser.get_contact_info(text)
+
+contact.to_s # => "Name: Alexander Vanadio\nEmail: anotherEmail@gmail.com\nEmail: execdd17@gmail.com\nPhone Number: 123-444-7890\nPhone Number: 123-456-7890\n"
+
+# you can also access the names, email_addresses, and phone_numbers directly
+contact.names
+contact.email_addresses
+contact.phone_numbers
+```
+
+### Running Tests and Code Coverage
+
+```bash
+cd orc_challenge
+rspec
+firefox coverage/index.html
+```

data/lib/ocr_challenge/basic_land_and_cell_number_parser.rb

@@ -0,0 +1,26 @@
+# This module attempts to parse phone numbers which I w ill define as
+# BOTH land lines and cell phone numbers. It will attempt to filter out fax numbers
+# NOTE: This assumes the including class will have a 'lines' instance variable
+module OcrChallenge::BasicLandAndCellNumberParser
+  include OcrChallenge::BasicTenDigitTelecomParser
+
+  FAX_REGEX = /fax/i
+
+  def parse_phone_numbers
+    matching_lines = get_matches
+
+    phone_number_lines = matching_lines.reject do |line|
+      is_fax_number?(line)
+    end
+
+    phone_number_lines.map do |line|
+      format(line, '-')
+    end.sort
+  end
+
+  private
+
+  def is_fax_number?(line)
+    line =~ FAX_REGEX
+  end
+end

data/lib/ocr_challenge/basic_ten_digit_telecom_parser.rb

@@ -0,0 +1,27 @@
+# This will find numbers that are grouped in 3 chunks of consecutive numbers,
+# with chunk sizes of 3, 3, and 4 respectively, regardless of what is in
+# between the chunks. It will also accept a leading 1.
+# An example is 1-(234) 435-3567
+module OcrChallenge::BasicTenDigitTelecomParser
+
+  BASIC_NUMBER_REGEX = /([1][^\d]*)?[\d]{3}[^\d]*[\d]{3}[^\d]*[\d]{4}/
+
+  def get_matches
+    lines.select do |line|
+      line =~ BASIC_NUMBER_REGEX
+    end
+  end
+
+  private
+
+  # remove every character that isn't a number, and leading 1 if present
+  def scrub_line(line)
+    line.gsub!(/[^\d]/, '')
+    line.size == 11 ? line[(1..-1)] : line
+  end
+
+  def format(line, delimiter)
+    line = scrub_line(line)
+    line[(0..2)] + delimiter + line[(3..5)] + delimiter + line[(6..9)]
+  end
+end

data/lib/ocr_challenge/email_parser.rb

@@ -0,0 +1,18 @@
+module OcrChallenge::EmailParser
+
+  # taken from: http://www.regular-expressions.info/email.html
+  EMAIL_REGEX = /\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}\b/i
+
+  def parse_email_addresses
+    lines.inject(Array.new) do |email_addresses, line|
+      line = line.strip
+
+      if offset = line =~ EMAIL_REGEX
+        email_addresses << line[(offset..-1)]
+      end
+
+      email_addresses
+    end.sort
+  end
+
+end

data/lib/ocr_challenge/i_business_card_parser.rb

@@ -0,0 +1,18 @@
+class OcrChallenge::IBusinessCardParser
+  include OcrChallenge::BasicLandAndCellNumberParser
+  include OcrChallenge::EmailParser
+  include OcrChallenge::NameParser
+
+  def self.get_contact_info(document)
+    parser = new(document)
+    IContactInfo.new(parser)
+  end
+
+  def initialize(document)
+    @lines = document.split("\n").reject { |line| line.empty? }
+  end
+
+  private
+
+  attr_reader :lines
+end

data/lib/ocr_challenge/i_contact_info.rb

@@ -0,0 +1,45 @@
+class OcrChallenge::IContactInfo
+
+  attr_reader :names, :email_addresses, :phone_numbers
+
+  def initialize(parser, list_of_names_dir='names')
+    @parser           = parser
+    @names            = @parser.parse_names(list_of_names_dir)
+    @email_addresses  = @parser.parse_email_addresses
+    @phone_numbers    = @parser.parse_phone_numbers
+  end
+
+  # NOTE: perhaps unlikely, but a business card may have more than one name. For
+  # example, maybe there were multiple points of contact for a given company card.
+  # Since the challenge did not specify, I take the first one.
+  def get_name
+    "Name: #{names.first}"
+  end
+
+  # NOTE: the programming challenge does not account for multiple email addresses,
+  # so I take the first one
+  def get_email_address
+    "Email: #{email_addresses.first}"
+  end
+
+  # NOTE: the programming challenge does not take into account that a contact
+  # can have multiple phone numbers, so I take the firs one
+  def get_phone_number
+    "Phone: #{phone_numbers.first}"
+  end
+
+  def to_s
+    contact_as_string = ""
+
+    {Name: names, Email: email_addresses,
+     "Phone Number" => phone_numbers}.each_pair do |label, values|
+
+      values.each do |value|
+        contact_as_string << "#{label}: #{value}\n"
+      end
+    end
+
+    contact_as_string
+  end
+
+end

data/lib/ocr_challenge/name_parser.rb

@@ -0,0 +1,28 @@
+require 'pathname'
+
+# It turns out that identifying names in a blob of text is hard. I decided to
+# use a dictionary of names in combination with eliminating lines with digits.
+module OcrChallenge::NameParser
+
+  # Note: the name files are expected to be new line separated names
+  def parse_names(dir_path)
+
+    #TODO: catch IO exception
+    names_dir = Pathname.new(dir_path)
+    name_files= names_dir.children
+
+    preprocessed_lines = lines.map(&:strip).reject do |line|
+      line =~ /\d/    # names shouldn't have digits in them
+    end
+
+    # compare the current line with all the names available in the name files
+    preprocessed_lines.select do |line|
+      name_files.any? do |file|
+        name_lines = file.readlines
+        name_lines.any? do |name_line|
+          line.downcase =~ /\b#{name_line.downcase.chomp}\b/
+        end
+      end
+    end
+  end
+end

data/lib/ocr_challenge/version.rb

@@ -0,0 +1,3 @@
+module OcrChallenge
+  VERSION = 1.0
+end

data/lib/ocr_challenge.rb

@@ -0,0 +1,12 @@
+$:.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+
+# the project namespace
+module OcrChallenge; end
+
+require 'ocr_challenge/basic_ten_digit_telecom_parser'
+require 'ocr_challenge/basic_land_and_cell_number_parser'
+require 'ocr_challenge/name_parser'
+require 'ocr_challenge/email_parser'
+require 'ocr_challenge/i_contact_info'
+require 'ocr_challenge/i_business_card_parser'
+require 'ocr_challenge/version'