nlp-pure 0.0.5 → 0.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8ae3951baabcafe913e157a575e3dc718a646f16
-  data.tar.gz: 14a6567449629a482bdc8863ffbfd04ae72af61b
+  metadata.gz: c5bbc92e65c96837a6e53f28248e15d48a35abe1
+  data.tar.gz: 79f767942ba8723a3f5f6eb04ea0ec4498e02591
 SHA512:
-  metadata.gz: f1766d42dd2916bdb0491448a9db0122b86f31325e3f12ce94c0d6b403cf5ecf50e4e95139f018f76896c1ef432e71a11dc36d8c2c597dc0870f400fb56bfeae
-  data.tar.gz: b3baa2f16339813070ffa978e03e8972046baebba815259d695da517baedc2830a55d0cfa75dcd234e307d2f09684194c02aa83a3b919f3db08be1426eb71537
+  metadata.gz: 9e00458afc1dadd851ea8ccd4e312ec19c6b775b455ebf6d5e599480dde8f333704a8c9601f62970bdefe26ffea7bd509bb3cd52314775d642865587d94e7214
+  data.tar.gz: 72abbd773eb915a11f76526b9bdfb37cbcd05c258aab45fd3c7e18c9fc1591c84c97cc3f99641ecee20ad27ea47d10daf9e35128d572d95aeb17aeda809e8a93

data/.travis.yml

@@ -1,14 +1,14 @@
 language: ruby
 sudo: false
 cache: bundler
+# NOTE: these run in order
 rvm:
-  - 2.2
-  - 2.1
-  - 2.0.0
   - jruby
   - rbx-2
+  - 2.0.0
+  - 2.1
+  - 2.2
 matrix:
   allow_failures:
     - rvm: rbx-2
     - rvm: jruby
-bundler_args: --without development

data/CHANGELOG.md

@@ -1,3 +1,9 @@
+# 0.1.0
+
+Officially leaving a non-semantic versioning scheme.
+
+Added benchmarking test.
+
 # 0.0.5
 
 Fixed bug in `NlpPure::Segmenting::DefaultWord` where leading ellipses could produce extra segmented words.

data/CONTRIBUTING.md

@@ -1,3 +1,5 @@
+# Contributing
+
 Pull requests are welcomed! Here’s a quick guide:
 
 1. Fork the repo.
@@ -13,11 +15,23 @@ a test!
 
 5. Push to your fork and submit a pull request.
 
-Syntax:
+
+## Project Goals
+
+* Accuracy over speed
+* One installation step (through `gem` or `bundle`)
+* Minimal runtime dependencies (beyond the standard libraries)
+* Effective collaboration (and minimized interpersonal conflict)
+* Sustainability and maintainability (this isn’t a full-time project)
+
+
+## Style Guide
+
+See also: `rake rubocop`
 
 * Two spaces, no tabs.
 * No trailing whitespace. Blank lines should not have any space.
-* Prefer &&/|| over and/or.
-* MyClass.my_method(my_arg) not my_method( my_arg ) or my_method my_arg.
-* a = b and not a=b.
+* Prefer `&& ||` over `and or`.
+* Use `MyClass.my_method(my_arg)` not `my_method( my_arg )` or `my_method my_arg`.
+* Prefer `a = b` to `a=b`.
 * Follow the conventions you see used in the source already.

data/README.md

@@ -10,11 +10,16 @@ NOTE: this is not affiliated with, endorsed by, or in any way connected with [Pu
 
 This project aims to provide functionality similar to [Treat](https://github.com/louismullie/treat), [open-nlp](https://github.com/louismullie/open-nlp), and [stanford-core-nlp](https://rubygems.org/gems/stanford-core-nlp) but with fewer dependencies. The code is tested against English language but the algorithm implementations aim to be flexible for other languages.
 
+## Table of Contents
 
-## Requirements
-
-TODO
-
+* [Installation](#installation)
+* [Usage](#usage)
+** [Word Segmentation](#word-segmentation)
+* [Supported Ruby Versions](#supported-ruby-versions)
+* [Versioning](#versioning)
+* [Contributing](CONTRIBUTING.md)
+* [License](LICENSE)
+* [See Also](#see-also)
 
 ## Installation
 
@@ -89,3 +94,36 @@ Constraint](http://docs.rubygems.org/read/chapter/16#page74) with two digits of
 ```ruby
 spec.add_dependency 'nlp-pure', '~> 0.1'
 ```
+
+
+## See Also
+
+[Search “nlp” at ruby-toolbox.com](https://www.ruby-toolbox.com/search?q=nlp)
+
+* APIs
+** [alchemy_api](https://github.com/dbalatero/alchemy_api)
+** [napi-ruby](https://github.com/Maluuba/napi-ruby)
+** [poliqarpr](https://github.com/apohllo/poliqarpr)
+** [wlapi](https://github.com/arbox/wlapi)
+* Bindings and Toolkits
+** [open-nlp](https://github.com/louismullie/open-nlp)
+** [stanford-core-nlp](https://github.com/louismullie/stanford-core-nlp)
+** [treat](https://github.com/louismullie/treat)
+* Classification
+** [linnaeus](https://github.com/djcp/linnaeus)
+** [maxent_string_classifier](https://github.com/mccraigmccraig/maxent_string_classifier)
+* N-Grams
+** [ruby-ngram](https://github.com/tkellen/ruby-ngram)
+* Specific Languages
+** Polish
+*** [nlp](https://github.com/knife/nlp)
+* Stopwords
+** [clarifier](https://github.com/meducation/clarifier)
+** [stopwords](https://github.com/brez/stopwords)
+** [stopwords-filter](https://github.com/brenes/stopwords-filter)
+* Tokenization
+** [rseg](https://rubygems.org/gems/rseg)
+** [Tokenizer](https://github.com/arbox/tokenizer)
+* Word Counters
+** [words_counted](https://github.com/abitdodgy/words_counted)
+

data/lib/nlp_pure/segmenting/default_word.rb

@@ -16,13 +16,15 @@ module NlpPure
         ]
       }.freeze
 
-      def self.parse(*args)
+      module_function
+
+      def parse(*args)
         unless args.nil? || args.empty?
-          clean_input(args[0]).split(options[:split])
+          clean_input(args[0]).split(options.fetch(:split, nil))
         end
       end
 
-      def self.clean_input(text = nil)
+      def clean_input(text = nil)
         input = text.to_s
         # perform replacements to work around the limitations of the splitting regexp
         options.fetch(:gsub, []).each do |gsub_pair|
@@ -33,7 +35,7 @@ module NlpPure
       end
 
       # NOTE: exposed as a method for easy mock/stub
-      def self.options
+      def options
         DEFAULT_OPTIONS
       end
     end

data/lib/nlp_pure/version.rb

@@ -1,5 +1,5 @@
 # encoding: utf-8
 #
 module NlpPure
-  VERSION = '0.0.5'
+  VERSION = '0.1.0'
 end

data/spec/lib/segmenting/default_word_spec.rb

@@ -7,6 +7,12 @@ describe NlpPure::Segmenting::DefaultWord do
     it 'is defined' do
       expect(defined?(NlpPure::Segmenting::DefaultWord)).to be_truthy
     end
+
+    describe '::DEFAULT_OPTIONS' do
+      it 'is Hash' do
+        expect(NlpPure::Segmenting::DefaultWord::DEFAULT_OPTIONS).to be_a Hash
+      end
+    end
   end
 
   describe '.parse' do
@@ -27,6 +33,26 @@ describe NlpPure::Segmenting::DefaultWord do
       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." }
 
+      context '(with nil options)' do
+        before do
+          expect(NlpPure::Segmenting::DefaultWord).to receive(:options).at_least(:once).and_return(nil)
+        end
+
+        it 'raises NoMethodError' do
+          expect { NlpPure::Segmenting::DefaultWord.parse(english_simple_sentence) }.to raise_error
+        end
+      end
+
+      context '(with blank options)' do
+        before do
+          expect(NlpPure::Segmenting::DefaultWord).to receive(:options).at_least(:once).and_return({})
+        end
+
+        it 'returns Array' do
+          expect(NlpPure::Segmenting::DefaultWord.parse(english_simple_sentence)).to be_an Array
+        end
+      end
+
       context '(with default options)' do
         context 'with `nil` argument' do
           it 'does not raise error' do
@@ -107,6 +133,74 @@ describe NlpPure::Segmenting::DefaultWord do
         it 'correctly counts with line breaks' do
           expect(NlpPure::Segmenting::DefaultWord.parse(english_simple_line_breaks).length).to eq(22)
         end
+
+        context 'benchmarking' do
+          before do
+            require 'benchmark'
+          end
+
+          it 'takes time', benchmarking: true do
+            expect(
+              Benchmark.realtime do
+                1000.times do
+                  NlpPure::Segmenting::DefaultWord.parse(english_simple_line_breaks)
+                end
+              end
+            ).to be < 0.1
+          end
+        end
+      end
+    end
+  end
+
+  describe '.clean_input' do
+    context 'English' do
+      let(:english_leading_ellipsis_sentence) { ' … the quick brown fox jumps over the lazy dog.' }
+
+      context '(with nil options)' do
+        before do
+          expect(NlpPure::Segmenting::DefaultWord).to receive(:options).at_least(:once).and_return(nil)
+        end
+
+        it 'raises NoMethodError' do
+          expect { NlpPure::Segmenting::DefaultWord.clean_input(english_leading_ellipsis_sentence) }.to raise_error
+        end
+      end
+
+      context '(with blank options)' do
+        before do
+          expect(NlpPure::Segmenting::DefaultWord).to receive(:options).at_least(:once).and_return({})
+        end
+
+        it 'only strips whitespace' do
+          expect(NlpPure::Segmenting::DefaultWord.clean_input(english_leading_ellipsis_sentence)).to eq english_leading_ellipsis_sentence.strip
+        end
+      end
+
+      context '(with default options)' do
+        context 'with `nil` argument' do
+          it 'does not raise error' do
+            expect { NlpPure::Segmenting::DefaultWord.clean_input(nil) }.to_not raise_error
+          end
+
+          it 'returns empty String' do
+            expect(NlpPure::Segmenting::DefaultWord.clean_input(nil)).to eq ''
+          end
+        end
+
+        context 'without arguments' do
+          it 'does not raise error' do
+            expect { NlpPure::Segmenting::DefaultWord.clean_input }.to_not raise_error
+          end
+
+          it 'returns nil' do
+            expect(NlpPure::Segmenting::DefaultWord.clean_input).to eq ''
+          end
+        end
+
+        it 'modifies the input' do
+          expect(NlpPure::Segmenting::DefaultWord.clean_input(english_leading_ellipsis_sentence)).to_not eq english_leading_ellipsis_sentence
+        end
       end
     end
   end

data/spec/spec_helper.rb

@@ -2,7 +2,12 @@
 require 'rspec'
 require 'coveralls'
 
-Coveralls.wear!
+Coveralls.wear! do
+  add_filter '/vendor/'
+  add_filter '/test/'
+  add_filter '/tmp/'
+  add_filter '/spec/'
+end
 
 RSpec.configure do |config|
   config.expect_with :rspec do |c|

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: nlp-pure
 version: !ruby/object:Gem::Version
-  version: 0.0.5
+  version: 0.1.0
 platform: ruby
 authors:
 - Reid Parham
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-02-15 00:00:00.000000000 Z
+date: 2015-02-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake