documentally 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: f7ba9de32ab5018f410bf3e823c9f8a219b52fc0
+  data.tar.gz: 723aba4f6bc91aaa0b8c189b4026af67b5bee329
+SHA512:
+  metadata.gz: 1ed6ff89c2a3866d00289a21592b03ce14722730dbb1709ceb3a4d14549bfe3e3e2c74cfa71a2518b57c8944f27bf821766c5425fd34b231f85e1020a973832a
+  data.tar.gz: 7081161224242150f5d9cf2a15d31edfeb765bb6bdc8150e9328fb3316e9995bf8b2885e59f2aae438e9bd79909f237aa768eec4008f24504a50dddf6859f527

data/README.md

@@ -0,0 +1,23 @@
+# Documentally
+
+Documentally uses [TF-IDF](http://en.wikipedia.org/wiki/tf-idf) to allow you to easily search and compare text documents.
+
+It's still under development, so the documentation is basically just the tests right now, but an example use would be:
+
+```ruby
+term_lists = [['doc_1', ['foo', 'foo', 'bar']],
+              ['doc_2', ['bar', 'baz', 'baz']],
+              ['doc_3', ['bar', 'baz', 'foo']]]
+
+corpus = Documentally::Corpus.new(term_lists)
+
+query = Documentally::Document.new('query', ['baz'])
+
+corpus.search(query) # => ['doc_2']
+```
+
+## Installation
+
+```sh
+gem install documentally
+```

data/lib/documentally.rb

@@ -0,0 +1,5 @@
+module Documentally
+end
+
+require 'documentally/document'
+require 'documentally/corpus'

data/lib/documentally/corpus.rb

@@ -0,0 +1,23 @@
+class Documentally::Corpus
+  attr_reader :documents
+
+  def initialize(term_lists)
+    all_terms = term_lists.map(&:last).inject(&:+)
+    master_document = Documentally::Document.new('master', all_terms)
+
+    @documents = term_lists.map { |name, term_list| Documentally::Document.new(name, term_list) }
+    documents.each do |document|
+      document.normalize!(master_document)
+    end
+  end
+
+  def search(query, take: 1)
+    order_documents_by(query).take(take)
+  end
+
+  private
+
+  def order_documents_by(query)
+    documents.sort_by { |document| document.similarity(query) }.reverse
+  end
+end

data/lib/documentally/document.rb

@@ -0,0 +1,40 @@
+class Documentally::Document
+  attr_accessor :term_hash
+  attr_reader :name
+
+  def initialize(name, terms)
+    @name = name
+    @term_hash = Hash.new(0.0)
+
+    terms.each do |term|
+      @term_hash[term] += 1
+    end
+  end
+
+  def to_s
+    name.to_s
+  end
+
+  def terms
+    term_hash.keys
+  end
+
+  def frequency(term)
+    term_hash[term]
+  end
+
+  def ==(other)
+    union_of_terms = (terms + other.terms).uniq
+    union_of_terms.all? { |term| frequency(term) == other.frequency(term) }
+  end
+
+  def similarity(query)
+    terms.map { |term| frequency(term) * query.frequency(term) }.inject(&:+)
+  end
+
+  def normalize!(corpus)
+    terms.each do |term|
+      term_hash[term] /= corpus.frequency(term)
+    end
+  end
+end

metadata

@@ -0,0 +1,48 @@
+--- !ruby/object:Gem::Specification
+name: documentally
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Harry Schwartz
+- Dan Honey
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-02-09 00:00:00.000000000 Z
+dependencies: []
+description: 
+email: hello@harryrschwartz.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- README.md
+- lib/documentally.rb
+- lib/documentally/corpus.rb
+- lib/documentally/document.rb
+homepage: http://github.com/hrs/documentally
+licenses:
+- GPL
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.2.2
+signing_key: 
+specification_version: 4
+summary: Simple TF-IDF document search library.
+test_files: []