sentino 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 02701d80b54badf1f285fead072a79954eac92c6
-  data.tar.gz: 7fca4ebe8d6c983fc7b58b5b724bc3298b112e2a
+  metadata.gz: 4f201f48f4b522c8ff4686c0a5450dcf8586e3d8
+  data.tar.gz: b1f59dd4d9a35600a127c79f7c45dc31c02ea480
 SHA512:
-  metadata.gz: fea7d2eada29dae5731cebb8cf072805f1850ffd231d788014b181a21fe7c07c57fca73db99af059ab18e84aa8501797fb29a938aa0d43f8fad1701340ee27cb
-  data.tar.gz: 432b18f8d105cac91eb29ae88f1a6d5651cf058fdb351531062e9501cf2e7169872ab377f01f27260de29ba9692e79de7c2194ad532bee11e3a3b5fda9d84321
+  metadata.gz: 2be893ed10a18d19ce449b5d68fcd02fae74b1a8637e96df5090a743c093c907d943c3cf0ad19fe65441a1276af635ac82eca1a687df11f53476352f0d6f7c6e
+  data.tar.gz: dfc68b4dd6f1d1cc7c7c0e6cd73ec06d3aef6248a64e5790e8f32b0c25af34b0fa10f61bb0636adc3843262ad0dc95361a52fed54e5151dc30dc84b67b5d6462

data/VERSION

@@ -1 +1 @@
-0.1.0
+0.2.0

data/lib/sentino/pattern.rb

@@ -0,0 +1,54 @@
+module Sentino
+  class Pattern
+    def initialize(reference)
+      @reference = prepare_string(reference)
+    end
+
+    def match(string)
+      check(string)
+    end
+
+
+private
+    def prepare_string(string)
+      words = string.split(" ").map{|s| s.downcase}
+      matchings = []
+      for word in words
+        matchings += create_matchings(word) if word.length > 3
+      end
+      matchings
+    end
+
+    def create_matchings(word, start=0, word_length=4)
+      list = []
+      max_length = word.length
+      return list if max_length < 4
+      if word_length == max_length
+        return [word]
+      elsif word_length + start < max_length
+        list << word[start..word_length-1+start]
+        list+=create_matchings(word, start+1, word_length)
+      elsif word_length + start == max_length
+        list << word[start..word_length-1+start]
+        list+=create_matchings(word, 0, word_length+1)
+      end
+      return list
+    end
+
+    def check(string)
+      counts = []
+      for mathcer in @reference
+        counts << count_matchings(string, mathcer)
+      end
+      puts counts
+      total_length = counts.map{|x| x == 0 ? 1 : x}.reduce(:+)
+      matching_length = counts.reduce(:+)
+      matching_length.to_f/total_length.to_f
+    end
+
+    def count_matchings(string, substring)
+      string.scan(/(?=#{substring})/).count
+    end
+
+  end
+end

data/lib/sentino.rb

@@ -2,4 +2,5 @@ $LOAD_PATH.unshift(File.dirname(__FILE__))
 
 module Sentino
   autoload :Word, 'sentino/word'
+  autoload :Pattern, 'sentino/pattern'
 end

data/sentino.gemspec

@@ -0,0 +1,68 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
+# -*- encoding: utf-8 -*-
+# stub: sentino 0.2.0 ruby lib
+
+Gem::Specification.new do |s|
+  s.name = "sentino"
+  s.version = "0.2.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.require_paths = ["lib"]
+  s.authors = ["Ferhat Ziba"]
+  s.date = "2016-06-10"
+  s.description = "Compare and give as result a matching result between 0 and 1"
+  s.email = "fero.ziba@gmail.com"
+  s.extra_rdoc_files = [
+    "LICENSE.txt",
+    "README.rdoc"
+  ]
+  s.files = [
+    ".document",
+    "Gemfile",
+    "Gemfile.lock",
+    "LICENSE.txt",
+    "README.rdoc",
+    "Rakefile",
+    "VERSION",
+    "lib/sentino.rb",
+    "lib/sentino/pattern.rb",
+    "lib/sentino/word.rb",
+    "sentino.gemspec",
+    "test/helper.rb",
+    "test/test_sentino.rb"
+  ]
+  s.homepage = "http://github.com/fero46/sentino"
+  s.licenses = ["MIT"]
+  s.rubygems_version = "2.4.8"
+  s.summary = "Compare Strings by word"
+
+  if s.respond_to? :specification_version then
+    s.specification_version = 4
+
+    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
+      s.add_development_dependency(%q<shoulda>, [">= 0"])
+      s.add_development_dependency(%q<rdoc>, ["~> 3.12"])
+      s.add_development_dependency(%q<bundler>, ["~> 1.0"])
+      s.add_development_dependency(%q<jeweler>, ["~> 2.0.1"])
+      s.add_development_dependency(%q<simplecov>, [">= 0"])
+      s.add_development_dependency(%q<test-unit>, [">= 0"])
+    else
+      s.add_dependency(%q<shoulda>, [">= 0"])
+      s.add_dependency(%q<rdoc>, ["~> 3.12"])
+      s.add_dependency(%q<bundler>, ["~> 1.0"])
+      s.add_dependency(%q<jeweler>, ["~> 2.0.1"])
+      s.add_dependency(%q<simplecov>, [">= 0"])
+      s.add_dependency(%q<test-unit>, [">= 0"])
+    end
+  else
+    s.add_dependency(%q<shoulda>, [">= 0"])
+    s.add_dependency(%q<rdoc>, ["~> 3.12"])
+    s.add_dependency(%q<bundler>, ["~> 1.0"])
+    s.add_dependency(%q<jeweler>, ["~> 2.0.1"])
+    s.add_dependency(%q<simplecov>, [">= 0"])
+    s.add_dependency(%q<test-unit>, [">= 0"])
+  end
+end
+

data/test/test_sentino.rb

@@ -9,12 +9,28 @@ class TestSentino < Test::Unit::TestCase
     assert_equal 1.0, sentino.match(target_1)
   end
 
-  should "should give 75 % identical Words" do 
+  should "should give 75 % identical Words" do
     reference = "Der erste String Test"
     target_1 = reference.gsub("erste", "zweite")
     sentino = Sentino::Word.new(reference)
     assert_equal 0.75, sentino.match(target_1)
+  end
+
+  should "give a higher value for baby then for tee" do
+    reference = "Babyprodukte / Babynahrung / Kekse & Tee / Babynahrung ab 4. Monat"
+    tee = "Geträn­ke Kaffee & Tee Tee Grüner & weißer Tee Weißer Tee"
+    baby = "Baby Babynahrung Kindersnacks Babykekse"
+    sentino = Sentino::Pattern.new(reference)
+    assert_true sentino.match(tee) < sentino.match(baby)
+  end
 
+  should "give a higher value for suger then for obst" do
+    reference = "Nahrungsmittel / Zucker & Süßstoff / Süßstoff / Süßstoff"
+    obst = "Obst & Gemüse Gemüse Zwiebeln & Knoblauch Zwiebeln"
+    suger = "Lebens­mittel Mehl Stärke & Co. Zucker & Süße Süßstoff"
+    sentino = Sentino::Pattern.new(reference)
+    assert_true sentino.match(obst) < sentino.match(suger)
   end
 
+
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: sentino
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - Ferhat Ziba
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-04-08 00:00:00.000000000 Z
+date: 2016-06-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: shoulda
@@ -110,7 +110,9 @@ files:
 - Rakefile
 - VERSION
 - lib/sentino.rb
+- lib/sentino/pattern.rb
 - lib/sentino/word.rb
+- sentino.gemspec
 - test/helper.rb
 - test/test_sentino.rb
 homepage: http://github.com/fero46/sentino
@@ -133,7 +135,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.5.1
+rubygems_version: 2.4.8
 signing_key: 
 specification_version: 4
 summary: Compare Strings by word