pragmatic_tokenizer 3.0.3 → 3.1.0

data/lib/pragmatic_tokenizer/version.rb

@@ -1,3 +1,3 @@
 module PragmaticTokenizer
-  VERSION = "3.0.3".freeze
+  VERSION = "3.1.0".freeze
 end

data/pragmatic_tokenizer.gemspec

@@ -1,5 +1,4 @@
-# coding: utf-8
-lib = File.expand_path('../lib', __FILE__)
+lib = File.expand_path('lib', __dir__)
 $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
 require 'pragmatic_tokenizer/version'
 
@@ -9,9 +8,9 @@ Gem::Specification.new do |spec|
   spec.authors       = ["Kevin S. Dias"]
   spec.email         = ["diasks2@gmail.com"]
 
-  spec.summary       = %q{A multilingual tokenizer}
-  spec.description   = %q{A multilingual tokenizer to split a string into tokens.}
-  spec.homepage      = "https://github.com/diasks2/pragmatic_tokenizer"
+  spec.summary       = 'A multilingual tokenizer'
+  spec.description   = 'A multilingual tokenizer to split a string into tokens.'
+  spec.homepage      = 'https://github.com/diasks2/pragmatic_tokenizer'
 
   spec.files         = `git ls-files -z`.split("\x0")
   spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
@@ -20,7 +19,7 @@ Gem::Specification.new do |spec|
 
   spec.add_runtime_dependency "unicode"
   spec.add_development_dependency "bundler", "~> 1.9"
-  spec.add_development_dependency "rake", "~> 10.0"
+  spec.add_development_dependency "rake", ">= 12.3.3"
   spec.add_development_dependency "rspec"
   spec.add_development_dependency "stackprof"
   spec.add_development_dependency "rubocop"

data/spec/languages/deutsch_spec.rb

@@ -175,7 +175,7 @@ describe PragmaticTokenizer do
           remove_stop_words: true,
           language:          'de'
       )
-      expect(pt.tokenize(text)).to eq(["der", "die", "lehrer_in", "und", "seine", "ihre", "schüler_innen", ".", "english", "."])
+      expect(pt.tokenize(text)).to eq(["lehrer_in", "schüler_innen", ".", "english", "."])
     end
 
     it 'removes English and German stopwords' do

data/spec/languages/english_spec.rb

@@ -443,6 +443,17 @@ describe PragmaticTokenizer do
         end
       end
 
+      context 'option (downcase)' do
+        it 'does not downcase URLs' do
+          skip "NOT IMPLEMENTED"
+          text = "Here are some domains and urls GOOGLE.com http://test.com/UPPERCASE."
+          pt = PragmaticTokenizer::Tokenizer.new(
+              downcase: :true
+          )
+          expect(pt.tokenize(text)).to eq(["here", "are", "some", "domains", "and", "urls", "GOOGLE.com", "http://test.com/UPPERCASE", "."])
+        end
+      end
+
       context 'option (domains)' do
         it 'tokenizes a string #001' do
           text = "Here are some domains and urls google.com https://www.google.com www.google.com."
@@ -485,6 +496,38 @@ describe PragmaticTokenizer do
       end
 
       context 'option (long_word_split)' do
+        it 'should not split twitter handles' do
+          text = "@john_doe"
+          pt = PragmaticTokenizer::Tokenizer.new(
+              long_word_split: 5
+          )
+          expect(pt.tokenize(text)).to eq(["@john_doe"])
+        end
+
+        it 'should not split emails' do
+          text = "john_doe@something.com"
+          pt = PragmaticTokenizer::Tokenizer.new(
+              long_word_split: 5
+          )
+          expect(pt.tokenize(text)).to eq(["john_doe@something.com"])
+        end
+
+        it 'should not split emails 2' do
+          text = "john_doe＠something.com"
+          pt = PragmaticTokenizer::Tokenizer.new(
+              long_word_split: 5
+          )
+          expect(pt.tokenize(text)).to eq(["john_doe＠something.com"])
+        end
+
+        it 'should not split urls' do
+          text = "http://test.com/some_path"
+          pt = PragmaticTokenizer::Tokenizer.new(
+              long_word_split: 5
+          )
+          expect(pt.tokenize(text)).to eq(["http://test.com/some_path"])
+        end
+
         it 'tokenizes a string #001' do
           text = "Some main-categories of the mathematics-test have sub-examples that most 14-year olds can't answer, therefor the implementation-instruction made in the 1990-years needs to be revised."
           pt = PragmaticTokenizer::Tokenizer.new(
@@ -1041,6 +1084,15 @@ describe PragmaticTokenizer do
           expect(pt.tokenize(text)).to eq(["short", "sentence", "explanations", "."])
         end
 
+        it 'removes stop words 2' do
+          text = 'This is a short sentence with explanations and stop words i.e. is a stop word as so is e.g. I think.'
+          pt = PragmaticTokenizer::Tokenizer.new(
+              language:          'en',
+              remove_stop_words: true
+          )
+          expect(pt.tokenize(text)).to eq(["short", "sentence", "explanations", "."])
+        end
+
         it 'removes user-supplied stop words' do
           text = 'This is a short sentence with explanations and stop words.'
           pt = PragmaticTokenizer::Tokenizer.new(

data/spec/languages/french_spec.rb

@@ -3,11 +3,11 @@ require 'spec_helper'
 describe PragmaticTokenizer do
   context 'Language: French (fr)' do
     it 'tokenizes a string #001' do
-      text = "L'art de l'univers, c'est un art"
+      text = "D'art de l'univers, c'est un art"
       pt = PragmaticTokenizer::Tokenizer.new(
           language: 'fr'
       )
-      expect(pt.tokenize(text)).to eq(["l'", "art", "de", "l'", "univers", ",", "c'est", "un", "art"])
+      expect(pt.tokenize(text)).to eq(["d'", "art", "de", "l'", "univers", ",", "c'" ,"est", "un", "art"])
     end
   end
 end

data/spec/performance_spec.rb

@@ -1,4 +1,3 @@
-# -*- encoding : utf-8 -*-
 require 'benchmark'
 require 'spec_helper'
 require 'stackprof'
@@ -29,6 +28,7 @@ describe PragmaticTokenizer do
   # 24.2
   # 23.2
   # 11.6
+  # 12.0
   # it 'is fast? (long strings)' do
   #   string = "Hello World. My name is Jonas. What is your name? My name is Jonas IV Smith. There it is! I found it. My name is Jonas E. Smith. Please turn to p. 55. Were Jane and co. at the party? They closed the deal with Pitt, Briggs & Co. at noon. Let's ask Jane and co. They should know. They closed the deal with Pitt, Briggs & Co. It closed yesterday. I can't see Mt. Fuji from here. St. Michael's Church is on 5th st. near the light. That is JFK Jr.'s book. I visited the U.S.A. last year. I live in the E.U. How about you? I live in the U.S. How about you? I work for the U.S. Government in Virginia. I have lived in the U.S. for 20 years. She has $100.00 in her bag. She has $100.00. It is in her bag. He teaches science (He previously worked for 5 years as an engineer.) at the local University. Her email is Jane.Doe@example.com. I sent her an email. The site is: https://www.example.50.com/new-site/awesome_content.html. Please check it out. She turned to him, 'This is great.' she said. She turned to him, \"This is great.\" she said. She turned to him, \"This is great.\" She held the book out to show him. Hello!! Long time no see. Hello?? Who is there? Hello!? Is that you? Hello?! Is that you? 1.) The first item 2.) The second item 1.) The first item. 2.) The second item. 1) The first item 2) The second item 1) The first item. 2) The second item. 1. The first item 2. The second item 1. The first item. 2. The second item. • 9. The first item • 10. The second item ⁃9. The first item ⁃10. The second item a. The first item b. The second item c. The third list item This is a sentence\ncut off in the middle because pdf. It was a cold \nnight in the city. features\ncontact manager\nevents, activities\n You can find it at N°. 1026.253.553. That is where the treasure is. She works at Yahoo! in the accounting department. We make a good team, you and I. Did you see Albert I. Jones yesterday? Thoreau argues that by simplifying one’s life, “the laws of the universe will appear less complex. . . .” \"Bohr [...] used the analogy of parallel stairways [...]\" (Smith 55). If words are left off at the end of a sentence, and that is all that is omitted, indicate the omission with ellipsis marks (preceded and followed by a space) and then indicate the end of the sentence with a period . . . . Next sentence. I never meant that.... She left the store. I wasn’t really ... well, what I mean...see . . . what I'm saying, the thing is . . . I didn’t mean it. One further habit which was somewhat weakened . . . was that of combining words into self-interpreting compounds. . . . The practice was not abandoned. . . ." * 1000
   #   puts "LENGTH: #{string.length}"

data/spec/spec_helper.rb

@@ -1,2 +1,2 @@
-$LOAD_PATH.unshift File.expand_path('../../lib', __FILE__)
+$LOAD_PATH.unshift File.expand_path('../lib', __dir__)
 require 'pragmatic_tokenizer'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pragmatic_tokenizer
 version: !ruby/object:Gem::Version
-  version: 3.0.3
+  version: 3.1.0
 platform: ruby
 authors:
 - Kevin S. Dias
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-02-16 00:00:00.000000000 Z
+date: 2020-07-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: unicode
@@ -42,16 +42,16 @@ dependencies:
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: '10.0'
+        version: 12.3.3
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: '10.0'
+        version: 12.3.3
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
@@ -111,7 +111,6 @@ files:
 - README.md
 - Rakefile
 - lib/pragmatic_tokenizer.rb
-- lib/pragmatic_tokenizer/full_stop_separator.rb
 - lib/pragmatic_tokenizer/languages.rb
 - lib/pragmatic_tokenizer/languages/arabic.rb
 - lib/pragmatic_tokenizer/languages/bulgarian.rb
@@ -140,6 +139,7 @@ files:
 - lib/pragmatic_tokenizer/languages/turkish.rb
 - lib/pragmatic_tokenizer/post_processor.rb
 - lib/pragmatic_tokenizer/pre_processor.rb
+- lib/pragmatic_tokenizer/regex.rb
 - lib/pragmatic_tokenizer/tokenizer.rb
 - lib/pragmatic_tokenizer/version.rb
 - pragmatic_tokenizer.gemspec
@@ -169,7 +169,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.1
+rubygems_version: 2.7.6
 signing_key: 
 specification_version: 4
 summary: A multilingual tokenizer

data/lib/pragmatic_tokenizer/full_stop_separator.rb

@@ -1,63 +0,0 @@
-# -*- encoding : utf-8 -*-
-
-module PragmaticTokenizer
-  # This class separates true full stops while ignoring
-  # periods that are part of an abbreviation
-  class FullStopSeparator
-
-    REGEXP_ENDS_WITH_DOT = /\A(.+)\.\z/
-    REGEXP_ONLY_LETTERS  = /\A[a-z]\z/i
-    REGEXP_UNKNOWN1      = /[a-z](?:\.[a-z])+\z/i
-    REGEXP_UNKNOWN2      = /\A(.*\w)\.\z/
-    DOT                  = '.'.freeze
-
-    def initialize(tokens:, abbreviations:, downcase:)
-      @tokens        = tokens
-      @abbreviations = abbreviations
-      @downcase      = downcase
-    end
-
-    def separate
-      create_cleaned_tokens
-      replace_last_token unless @cleaned_tokens.empty?
-      @cleaned_tokens
-    end
-
-    private
-
-      def create_cleaned_tokens
-        @cleaned_tokens = []
-        @tokens.each_with_index do |token, position|
-          if @tokens[position + 1] && token =~ REGEXP_ENDS_WITH_DOT
-            match = Regexp.last_match(1)
-            if unknown_method1(match)
-              @cleaned_tokens += [match, DOT]
-              next
-            end
-          end
-          @cleaned_tokens << token
-        end
-      end
-
-      def unknown_method1(token)
-        !abbreviation?(token) && token !~ REGEXP_ONLY_LETTERS && token !~ REGEXP_UNKNOWN1
-      end
-
-      def abbreviation?(token)
-        @abbreviations.include?(inverse_case(token))
-      end
-
-      def inverse_case(token)
-        @downcase ? token : Unicode.downcase(token)
-      end
-
-      def replace_last_token
-        last_token = @cleaned_tokens[-1]
-        return if abbreviation?(last_token.chomp(DOT)) || last_token !~ REGEXP_UNKNOWN2
-        @cleaned_tokens[-1] = Regexp.last_match(1)
-        @cleaned_tokens << DOT
-      end
-
-  end
-
-end