sentimental 1.2.1 → 1.3.0

data/data/slang.json

@@ -0,0 +1,57 @@
+{
+  "%-(": -1.0,
+  ")-:": -1.0,
+  "):": -1.0,
+  ")o:": -1.0,
+  "8-0": -1.0,
+  "8/": -1.0,
+  "8\\": -1.0,
+  "8c": -1.0,
+  ":'(": -1.0,
+  ":'-(": -1.0,
+  ":(": -1.0,
+  ":*(": -1.0,
+  ":,(": -1.0,
+  ":-(": -1.0,
+  ":-/": -1.0,
+  ":-S": -1.0,
+  ":-\\": -1.0,
+  ":-|": -0.5,
+  ":/": -0.5,
+  ":O": -0.25,
+  ":S": -0.25,
+  ":\\": -0.25,
+  ":|": -0.25,
+  "=(": -1.0,
+  ">:(": -1.0,
+  "D:": -1.0,
+  "sux": -1.0,
+  "(o;": 1.0,
+  "8-)": 1.0,
+  ";)": 1.0,
+  ";o)": 1.0,
+  "%-)": 1.0,
+  "(-:": 1.0,
+  ":-)": 1.0,
+  "(:": 1.0,
+  "(o:": 1.0,
+  "8)": 1.0,
+  ":)": 1.0,
+  ":-D": 1.0,
+  ":-P": 1.0,
+  ":D": 1.0,
+  ":P": 1.0,
+  ":]": 1.0,
+  ":o)": 1.0,
+  ":p": 1.0,
+  ";^)": 1.0,
+  "<3": 1.0,
+  "&lt;3": 1.0,
+  "=)": 1.0,
+  "=]": 1.0,
+  ">:)": 1.0,
+  ">:D": 1.0,
+  ">=D": 1.0,
+  "^_^": 1.0,
+  "}:)": 1.0
+}

data/lib/file_reader.rb

@@ -0,0 +1,19 @@
+require 'json'
+
+module FileReader
+  def hash_from_txt(filename)
+    new_words = {}
+    File.open(filename) do |file|
+      file.each_line do |line|
+        parsed_line = line.chomp.scan(/^([^\s]+)\s+(.+)/).first
+        next unless parsed_line
+        new_words[parsed_line[1]] = parsed_line[0].to_f
+      end
+    end
+    new_words
+  end
+
+  def hash_from_json(filename)
+    JSON.parse(File.read(filename))
+  end
+end

data/lib/sentimental.rb

@@ -1,12 +1,12 @@
+require_relative 'file_reader'
+
 class Sentimental
+  include FileReader
+
   attr_accessor :threshold, :word_scores, :neutral_regexps, :ngrams
 
   def initialize(threshold: 0, word_scores: nil, neutral_regexps: [], ngrams: 1)
-    if ngrams >= 1
-      @ngrams = ngrams.to_i
-    else
-      @ngrams = 1
-    end
+    @ngrams = ngrams.to_i.abs if ngrams.to_i >= 1
     @word_scores = word_scores || {}
     @word_scores.default = 0.0
     @threshold = threshold
@@ -14,10 +14,10 @@ class Sentimental
   end
 
   def score(string)
-    return 0 if neutral_regexps.any? {|regexp| string =~ regexp}
+    return 0 if neutral_regexps.any? { |regexp| string =~ regexp }
 
-    extract_words_with_n_grams(string).inject(0) do |score, token| 
-      score += word_scores[token]
+    extract_words_with_n_grams(string).inject(0) do |score, token|
+      score + word_scores[token]
     end
   end
 
@@ -38,24 +38,21 @@ class Sentimental
   end
 
   def load_defaults
-    ['sentiwords', 'sentislang'].each do |filename|
-      load_from(File.dirname(__FILE__) + "/../data/#{filename}.txt")
+    %w(slang en_words).each do |filename|
+      load_from_json(File.dirname(__FILE__) + "/../data/#{filename}.json")
     end
   end
 
   def load_from(filename)
-    File.open(filename) do |file|
-      file.each_line do |line|
-        if parsed_line = (line.chomp.scan(/^([^\s]+)\s+(.+)/).first)
-          sentiscore = parsed_line[0]
-          text = parsed_line[1]
-          word_scores[text] = sentiscore.to_f
-        end
-      end
-    end
+    word_scores.merge!(hash_from_txt(filename))
   end
-  
-  alias_method :load_senti_file, :load_from
+
+  def load_from_json(filename)
+    word_scores.merge!(hash_from_json(filename))
+  end
+
+  alias load_senti_file load_from
+  alias load_senti_json load_from_json
 
   private
 
@@ -71,9 +68,9 @@ class Sentimental
   end
 
   def ngramify(words, max_size)
-    return [words.join(" ")] if words.size <= max_size
+    return [words.join(' ')] if words.size <= max_size
     tail = words.last(words.size - 1)
-    
-    [words.first(max_size).join(" ")] + ngramify(tail, max_size)
+
+    [words.first(max_size).join(' ')] + ngramify(tail, max_size)
   end
 end

data/sentimental.gemspec

@@ -1,19 +1,22 @@
 Gem::Specification.new do |spec|
   spec.name         = 'sentimental'
-  spec.version      = '1.2.1'
+  spec.version      = '1.3.0'
   spec.summary      = 'Simple sentiment analysis'
   spec.description  = 'A simple sentiment analysis gem'
   spec.authors      = ['Jeff Emminger', 'Christopher MacLellan', 'Denis Pasin']
   spec.email        = ['jeff@7compass.com', 'denis@hellojam.fr']
   spec.homepage     = 'https://github.com/7compass/sentimental'
   spec.license      = 'MIT'
-  
+
   spec.files         = `git ls-files`.split($/)
   spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ['lib']
-  
+
   spec.add_development_dependency "bundler", "~> 1.3"
   spec.add_development_dependency "rake"
   spec.add_development_dependency "rspec", ">= 3.0.0"
+  spec.add_development_dependency "rubocop", "~> 0.40", ">= 0.40.0"
+
+  spec.add_runtime_dependency "json", "~> 1.8", ">= 1.8.3"
 end

data/spec/file_reader_spec.rb

@@ -0,0 +1,31 @@
+require_relative 'spec_helper'
+require_relative '../lib/file_reader'
+include FileReader
+
+describe FileReader do
+  describe '#hash_from_txt' do
+    subject(:answer) { FileReader.hash_from_txt('spec/test_data/test.txt') }
+
+    it 'returns a hash from a txt file' do
+      expect(answer.empty?).to be_falsey
+    end
+
+    it 'contains the data' do
+      expect(answer['TEST']).to eq 1.0
+      expect(answer[':(']).to eq(-1.0)
+    end
+  end
+
+  describe '#hash_from_json' do
+    subject(:answer) { FileReader.hash_from_json('spec/test_data/test.json') }
+
+    it 'returns a hash from a json file' do
+      expect(answer.empty?).to be_falsey
+    end
+
+    it 'contains the data' do
+      expect(answer['TEST']).to eq 1.0
+      expect(answer[':(']).to eq(-1.0)
+    end
+  end
+end

data/spec/sentimental_spec.rb

@@ -1,145 +1,143 @@
-require_relative "../lib/sentimental"
+require_relative '../lib/sentimental'
 
 describe Sentimental do
-
   before :each do
     analyzer.load_defaults
   end
 
   let(:analyzer) { Sentimental.new(threshold: 0.1) }
 
-  describe "#sentiment" do
-    it "returns :positive when the score > threshold" do
-      expect(analyzer.sentiment("I love ruby <3")).to be :positive
+  describe '#sentiment' do
+    it 'returns :positive when the score > threshold' do
+      expect(analyzer.sentiment('I love ruby <3')).to be :positive
     end
 
-    it "returns :negative when the score < -threshold" do
-      expect(analyzer.sentiment("I hate javascript")).to be :negative
+    it 'returns :negative when the score < -threshold' do
+      expect(analyzer.sentiment('I hate javascript')).to be :negative
     end
 
-    it "returns :positive when -threshold < score < threshold" do
-      expect(analyzer.sentiment("je en sais pas")).to be :neutral
+    it 'returns :positive when -threshold < score < threshold' do
+      expect(analyzer.sentiment('je en sais pas')).to be :neutral
     end
   end
 
-  describe "#classify" do
-    it "is true when in the class" do
-      expect(analyzer.classify("I love ruby")).to be_truthy
+  describe '#classify' do
+    it 'is true when in the class' do
+      expect(analyzer.classify('I love ruby')).to be_truthy
     end
 
-    it "is false otherwise" do
-      expect(analyzer.classify("je ne sais pas")).to be_falsy
-      expect(analyzer.classify("i hate java")).to be_falsy
+    it 'is false otherwise' do
+      expect(analyzer.classify('je ne sais pas')).to be_falsy
+      expect(analyzer.classify('i hate java')).to be_falsy
     end
   end
 
-  describe "initialization" do
+  describe 'initialization' do
     subject do
       Sentimental.new(
         threshold: 0.2,
-        word_scores: {"non" => -1.0},
-        neutral_regexps: [/.*/],
+        word_scores: { 'non' => -1.0 },
+        neutral_regexps: [/.*/]
       )
     end
 
-    it "takes multiple init params" do
+    it 'takes multiple init params' do
       expect(subject.threshold).to eq 0.2
-      expect(subject.word_scores["non"]).to eq -1.0
-      expect(subject.neutral_regexps).to include /.*/
+      expect(subject.word_scores['non']).to eq(-1.0)
+      expect(subject.neutral_regexps).to include(/.*/)
     end
   end
 
-  describe "neutral regexp" do
-    context "when there is some neutral regexp" do
-      let(:text_neutral) {"Do you love ruby?"}
-      let(:text) {"I love ruby"}
+  describe 'neutral regexp' do
+    context 'when there is some neutral regexp' do
+      let(:text_neutral) { 'Do you love ruby?' }
+      let(:text) { 'I love ruby' }
 
       before do
         analyzer.neutral_regexps << /\?\s*$/
       end
 
-      it "scores it to 0" do
+      it 'scores it to 0' do
         expect(analyzer.score(text_neutral)).to eq 0
         expect(analyzer.score(text)).not_to eq 0
       end
     end
   end
 
-  describe "n-grams" do
+  describe 'n-grams' do
     let(:word_scores) { nil }
     subject do
       Sentimental.new(word_scores: word_scores, ngrams: 3)
     end
 
-    it "is initialized by ngrams param" do
+    it 'is initialized by ngrams param' do
       expect(subject.ngrams).to eq 3
     end
-    
-    context "there is n-grams in the dictionary" do
-      let(:word_scores) {{"happy hour" => 1.0, "not happy hour" => -5.0}}
-      let(:text) { "why not happy hour, but happy so hour?" }
 
-      it "update scores regarding to n-grams" do
-        expect(subject.score(text)).to eq -4
+    context 'there is n-grams in the dictionary' do
+      let(:word_scores) { { 'happy hour' => 1.0, 'not happy hour' => -5.0 } }
+      let(:text) { 'why not happy hour, but happy so hour?' }
+
+      it 'update scores regarding to n-grams' do
+        expect(subject.score(text)).to eq(-4)
       end
     end
 
     context "there's n-grams longer than specified in dictionary" do
-      let(:word_scores) {{"happy hour" => 1.0, "not so happy hour" => -5.0}}
-      let(:text) { "why not so happy hour ?" }
+      let(:word_scores) { { 'happy hour' => 1.0, 'not so happy hour' => -5.0 } }
+      let(:text) { 'why not so happy hour ?' }
 
-      it "ignores the lines" do
+      it 'ignores the lines' do
         expect(subject.score(text)).to eq 1
       end
     end
   end
 
-  describe "scoring in a normal context" do
+  describe 'scoring in a normal context' do
     subject do
       analyzer.score(text)
     end
 
-    context "when the text is postive" do
-      let(:text) {'I love ruby'}
+    context 'when the text is postive' do
+      let(:text) { 'I love ruby' }
 
       it 'returns a positive score' do
         expect(subject).to be > 0
       end
     end
 
-    context "when the text is neutral" do
-      let(:text) {'I like ruby'}
+    context 'when the text is neutral' do
+      let(:text) { 'I like ruby' }
 
       it 'returns a neutral score' do
         expect(subject).to eq 0
       end
     end
 
-    context "when the text is negative" do
-      let(:text) {'I hate ruby'}
+    context 'when the text is negative' do
+      let(:text) { 'I hate ruby' }
 
       it 'returns a negative score' do
         expect(subject).to be < 0
       end
     end
 
-    context "when the text has smiley" do
-      let(:text) {'I love ruby'}
-      let(:text_with_smiley) {'I love ruby :-)'}
+    context 'when the text has smiley' do
+      let(:text) { 'I love ruby' }
+      let(:text_with_smiley) { 'I love ruby :-)' }
 
       it 'scores it' do
         expect(analyzer.score(text_with_smiley)).to be > analyzer.score(text)
       end
     end
 
-    context "when the text has punctuation" do
-      let(:text) {'I love ruby'}
-      let(:text_with_punctuation) {'I love, ruby'}
+    context 'when the text has punctuation' do
+      let(:text) { 'I love ruby' }
+      let(:text_with_punctuation) { 'I love, ruby' }
 
       it 'removes it' do
         expect(analyzer.score(text_with_punctuation)).to eq analyzer.score(text)
       end
     end
-
   end
 end

data/spec/spec_helper.rb

@@ -0,0 +1,12 @@
+def create_test_files
+  Dir.mkdir('spec/test_data') unless Dir.exist?('spec/test_data')
+  File.open('spec/test_data/test.txt', 'w+') do |file|
+    file.puts('1.0 TEST')
+    file.puts('-1.0 :(')
+  end
+  File.open('spec/test_data/test.json', 'w+') do |file|
+    file.puts('{"TEST":1.0,":(":-1.0}')
+  end
+end
+
+create_test_files

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: sentimental
 version: !ruby/object:Gem::Version
-  version: 1.2.1
+  version: 1.3.0
 platform: ruby
 authors:
 - Jeff Emminger
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-04-22 00:00:00.000000000 Z
+date: 2016-05-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -54,6 +54,46 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 3.0.0
+- !ruby/object:Gem::Dependency
+  name: rubocop
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.40'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.40.0
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.40'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.40.0
+- !ruby/object:Gem::Dependency
+  name: json
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.8'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.8.3
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.8'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.8.3
 description: A simple sentiment analysis gem
 email:
 - jeff@7compass.com
@@ -63,16 +103,23 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".gitignore"
+- ".rspec"
+- ".rubocop.yml"
 - Gemfile
 - Gemfile.lock
 - LICENSE.txt
 - README.md
 - Rakefile
-- data/sentislang.txt
-- data/sentiwords.txt
+- data/JSON_builder.rb
+- data/en_words.json
+- data/sentiwords_fr.txt
+- data/slang.json
+- lib/file_reader.rb
 - lib/sentimental.rb
 - sentimental.gemspec
+- spec/file_reader_spec.rb
 - spec/sentimental_spec.rb
+- spec/spec_helper.rb
 homepage: https://github.com/7compass/sentimental
 licenses:
 - MIT
@@ -98,4 +145,6 @@ signing_key:
 specification_version: 4
 summary: Simple sentiment analysis
 test_files:
+- spec/file_reader_spec.rb
 - spec/sentimental_spec.rb
+- spec/spec_helper.rb