nlp-pure 0.0.2 → 0.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: c4f6247383c48cd71d5ccebf6cc937023d1a880b
-  data.tar.gz: b33e1b19f2bfb5d49c6082f10699f06e2052e32c
+  metadata.gz: 7adc921eb51b54bd646cc0c7c57edbfd47a4d7fb
+  data.tar.gz: 54930de1e0cd9f5507bde731e057f22942f1d212
 SHA512:
-  metadata.gz: 7dee5b3c6947d08ef7e8b92a7332baeaaf8785969f98673613a7de2287dbba65dfb06a6f05990036013abaf01235f0017cb75cdf64be27770cc0c06046e30d99
-  data.tar.gz: a01ebec79d05301998d3618c1cd4e1b9cfa23982b36834cee17c529559b2bf36b5a32ea11d843469574924c42be245f192a636a3ee7f288490f7516e39f0a589
+  metadata.gz: a0ee82f8e519e712d36f89af779e9b42aed4887e0b87fa5c77e6b422f5b8ab6d66afc702c650f3bd9149a80d7c2fbb1434504fb53e132164c5dafb8c520b5102
+  data.tar.gz: db7bf9e6178c2e25a5d5bcdde763dbeb8fc9d3130800a0bfd41a8a6abae1ab64ad012ca27bd7334c241e8c4486566cd6f60a8f9c173086c53747f3296d7e9fed

data/CHANGELOG.md

@@ -1,3 +1,11 @@
+# 0.0.4
+
+Fixed bug in `NlpPure::Segmenting::DefaultWord` where ellipses without spaces would not segment.
+
+# 0.0.3
+
+Fixed bug in `NlpPure::Segmenting::DefaultWord` where double hyphens and spaced dashes would segment as empty words.
+
 # 0.0.2
 
 Added `NlpPure::Segmenting::DefaultWord` module for segmenting text into words.

data/README.md

@@ -2,7 +2,7 @@
 
 [![Code Climate](https://codeclimate.com/github/parhamr/nlp-pure/badges/gpa.svg)](https://codeclimate.com/github/parhamr/nlp-pure)
 [![Build Status](https://travis-ci.org/parhamr/nlp-pure.svg?branch=master)](https://travis-ci.org/parhamr/nlp-pure) 
-[![Coverage Status](https://coveralls.io/repos/parhamr/nlp-pure/badge.png?branch=master)](https://coveralls.io/r/parhamr/nlp-pure) 
+[![Coverage Status](https://coveralls.io/repos/parhamr/nlp-pure/badge.png?branch=master)](https://coveralls.io/r/parhamr/nlp-pure?branch=master) 
 
 Natural language processing algorithms implemented in pure Ruby with minimal dependencies.
 
@@ -44,7 +44,7 @@ $ gem install nlp-pure
 
 ```
 $ bundle exec irb
-irb(main):001:0> require_relative './lib/nlp_pure/segmenting/default_word'
+irb(main):001:0> require 'nlp_pure/segmenting/default_word'
 => true
 irb(main):002:0> NlpPure::Segmenting::DefaultWord.parse 'The quick brown fox jumps over the lazy dog.'
 => ["The", "quick", "brown", "fox", "jumps", "over", "the", "lazy", "dog."]

data/lib/nlp_pure/segmenting/default_word.rb

@@ -5,8 +5,9 @@ module NlpPure
     #
     module DefaultWord
       DEFAULT_OPTIONS = {
-        # hyphen, en dash, em dash, and string
-        split: /[\-–—\s]/
+        # 3+ periods as pseudo-ellipsis (with optional whitespace)
+        # OR hyphen, en dash, em dash, and whitespace
+        split: /\s?\.{3,}+\s?|[\s\-–—…]+/
       }.freeze
       def self.parse(*args)
         unless args.nil? || args.empty?
@@ -22,5 +23,3 @@ module NlpPure
     end
   end
 end
-
-require_relative '../segmenting'

data/lib/nlp_pure/version.rb

@@ -1,5 +1,5 @@
 # encoding: utf-8
 #
 module NlpPure
-  VERSION = '0.0.2'
+  VERSION = '0.0.4'
 end

data/spec/lib/segmenting/default_word_spec.rb

@@ -14,6 +14,13 @@ describe NlpPure::Segmenting::DefaultWord do
       let(:english_simple_sentence) { 'The quick brown fox jumps over the lazy dog.' }
       let(:english_hyphen_sentence) { 'The New York-based company hired new staff.' }
       let(:english_dash_sentence) { 'The quick brown fox—full of energy—jumps over the lazy dog.' }
+      let(:english_spaced_dash_sentence) { 'The quick brown fox — full of energy — jumps over the lazy dog.' }
+      let(:english_twohyphen_sentence) { 'The quick brown fox--full of energy--jumps over the lazy dog.' }
+      let(:english_ellipsis_sentence) { 'The quick brown fox…jumps over the lazy dog.' }
+      let(:english_spaced_ellipsis_sentence) { 'The quick brown fox … jumps over the lazy dog.' }
+      let(:english_period_ellipses_sentence) { 'The quick brown fox...jumps over the lazy dog.' }
+      let(:english_trailing_ellipses_sentence) { 'The quick brown fox jumps over the lazy dog …' }
+      let(:english_spaced_period_ellipses_sentence) { 'The quick brown fox ... jumps over the lazy dog.' }
       let(:english_abbreviation_sentence) { 'The U.S.A. is a member of NATO.' }
       let(:english_simple_paragraph) { 'Mary had a little lamb. The lamb’s fleece was white as snow. Everywhere that Mary went, the lamb was sure to go.' }
       let(:english_simple_line_breaks) { "Mary had a little lamb,\nHis fleece was white as snow,\nAnd everywhere that Mary went,\nThe lamb was sure to go." }
@@ -51,10 +58,38 @@ describe NlpPure::Segmenting::DefaultWord do
           expect(NlpPure::Segmenting::DefaultWord.parse(english_hyphen_sentence).length).to eq(8)
         end
 
+        it 'correctly segments double-hyphen dashes' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_twohyphen_sentence).length).to eq(12)
+        end
+
         it 'correctly segments dashes' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_spaced_dash_sentence).length).to eq(12)
+        end
+
+        it 'correctly segments spaced dashes' do
           expect(NlpPure::Segmenting::DefaultWord.parse(english_dash_sentence).length).to eq(12)
         end
 
+        it 'correctly segments ellipses' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_ellipsis_sentence).length).to eq(9)
+        end
+
+        it 'correctly segments spaced ellipses' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_spaced_ellipsis_sentence).length).to eq(9)
+        end
+
+        it 'correctly segments period-ellipses' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_period_ellipses_sentence).length).to eq(9)
+        end
+
+        it 'correctly segments spaced period-ellipses' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_spaced_period_ellipses_sentence).length).to eq(9)
+        end
+
+        it 'correctly segments with trailing, spaced ellipses' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_trailing_ellipses_sentence).length).to eq(9)
+        end
+
         it 'does not segment abbreviations' do
           expect(NlpPure::Segmenting::DefaultWord.parse(english_abbreviation_sentence).length).to eq(7)
         end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nlp-pure
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.0.4
 platform: ruby
 authors:
 - Reid Parham