turkish_stemmer 0.1.0

data/benchmarks/turkish_word_recognition.rb

@@ -0,0 +1,26 @@
+# coding : utf-8
+require 'benchmark'
+require 'turkish_stemmer'
+
+Benchmark.bmbm(7) do |x|
+
+  x.report('regex') do
+    TurkishStemmer.class_eval do
+      def self.turkish?(word)
+        !! word.match(TurkishStemmer::ALPHABET)
+      end
+    end
+
+    100_000.times { TurkishStemmer.turkish?("aaa") }
+  end
+
+  x.report('loop') do
+    TurkishStemmer.class_eval do
+      def self.turkish?(word)
+        !! word.chars.to_a.all? { |c| "abcçdefgğhıijklmnoöprsştuüvyz".include?(c) }
+      end
+    end
+
+    100_000.times { TurkishStemmer.turkish?("aaaa") }
+  end
+end

data/config/derivational_states.yml

@@ -0,0 +1,10 @@
+a:
+  transitions:
+    - suffix: :s1
+      state:  :b
+
+  final_state: false
+
+b:
+  transitions: []
+  final_state: true

data/config/derivational_suffixes.yml

@@ -0,0 +1,6 @@
+s1:
+  name: "-lU"
+  regex: "lı|li|lu|lü"
+  optional_letter: false
+  check_harmony: true
+

data/config/nominal_verb_states.yml

@@ -0,0 +1,121 @@
+# The order of the transitions is very crusial.
+a:
+  transitions:
+    # Transitions to state B
+    - suffix: :s1
+      state:  :b
+    - suffix: :s2
+      state:  :b
+    - suffix: :s4
+      state:  :b
+    - suffix: :s3
+      state:  :b
+    # Transitions to state C
+    - suffix: :s5
+      state: :c
+    # Transitions to state D
+    - suffix: :s6
+      state: :d
+    - suffix: :s7
+      state: :d
+    - suffix: :s8
+      state: :d
+    - suffix: :s9
+      state: :d
+    # Transitions to state E
+    - suffix: :s10
+      state: :e
+    # Transitions to state F
+    - suffix: :s12
+      state: :f
+    - suffix: :s13
+      state: :f
+    - suffix: :s14
+      state: :f
+    - suffix: :s15
+      state: :f
+    # Transitions to state H
+    - suffix: :s11
+      state: :h
+
+  final_state: false
+
+b:
+  transitions:
+    - suffix: :s14
+      state: :f
+
+  final_state: true
+
+c:
+  transitions:
+    - suffix: :s10
+      state: :f
+    - suffix: :s12
+      state: :f
+    - suffix: :s13
+      state: :f
+    - suffix: :s14
+      state: :f
+
+  final_state: true
+
+d:
+  transitions:
+    - suffix: :s12
+      state: :f
+    - suffix: :s13
+      state: :f
+
+  final_state: false
+
+e:
+  transitions:
+    # Transitions to state G
+    - suffix: :s1
+      state: :g
+    - suffix: :s2
+      state: :g
+    - suffix: :s3
+      state: :g
+    - suffix: :s4
+      state: :g
+    - suffix: :s5
+      state: :g
+    # Transitions to state F
+    - suffix: :s14
+      state: :f
+  final_state: true
+
+f:
+  transitions: []
+
+  final_state: true
+
+
+g:
+  transitions:
+    - suffix: :s14
+      state: :f
+
+  final_state: false
+
+h:
+  transitions:
+    # Transitions to state F
+    - suffix: :s14
+      state: :f
+    # Transitions to state G
+    - suffix: :s1
+      state: :g
+    - suffix: :s2
+      state: :g
+    - suffix: :s3
+      state: :g
+    - suffix: :s4
+      state: :g
+    - suffix: :s5
+      state: :g
+
+  final_state: false
+

data/config/nominal_verb_suffixes.yml

@@ -0,0 +1,90 @@
+s1:
+  name: "-(y)Um"
+  regex: "ım|im|um|üm"
+  optional_letter: "y"
+  check_harmony: true
+
+s2:
+  name: "-sUn"
+  regex: "sın|sin|sun|sün"
+  optional_letter: false
+  check_harmony: true
+
+s3:
+  name: "-(y)Uz"
+  regex: "ız|iz|uz|üz"
+  optional_letter: "y"
+  check_harmony: true
+
+s4:
+  name: "-sUnUz"
+  regex: "sınız|siniz|sunuz|sünüz"
+  optional_letter: false
+  check_harmony: true
+
+s5:
+  name: "-lAr"
+  regex: "lar|ler"
+  optional_letter: false
+  check_harmony: true
+
+s6:
+  name: "-m"
+  regex: "m"
+  optional_letter: false
+  check_harmony: true
+
+s7:
+  name: "-n"
+  regex: "n"
+  optional_letter: false
+  check_harmony: true
+
+s8:
+  name: "-k"
+  regex: "k"
+  optional_letter: false
+  check_harmony: true
+
+s9:
+  name: "-nUz"
+  regex: "nız|niz|nuz|nüz"
+  optional_letter: false
+  check_harmony: true
+
+s10:
+  name: "-DUr"
+  regex: "tır|tir|tur|tür|dır|dir|dur|dür"
+  optional_letter: false
+  check_harmony: true
+
+s11:
+  name: "-cAsInA"
+  regex: "casına|çasına|cesine|çesine"
+  optional_letter: false
+  check_harmony: true
+
+s12:
+  name: "-(y)DU"
+  regex: "dı|di|du|dü|tı|ti|tu|tü"
+  optional_letter: "y"
+  check_harmony: true
+
+s13:
+  name: "-(y)sA"
+  regex: "sa|se"
+  optional_letter: "y"
+  check_harmony: true
+
+s14:
+  name: "-(y)mUş"
+  regex: "muş|miş|müş|mış"
+  optional_letter: "y"
+  check_harmony: true
+
+s15:
+  name: "-(y)ken"
+  regex: "ken"
+  optional_letter: "y"
+  check_harmony: true
+

data/config/noun_states.yml

@@ -0,0 +1,177 @@
+# The order of the transitions is very crusial.
+a:
+  transitions:
+    - suffix: :s16
+      state: :c
+    - suffix: :s7
+      state: :k
+    - suffix: :s3
+      state: :h
+    - suffix: :s5
+      state: :h
+    - suffix: :s1
+      state: :l
+    - suffix: :s14
+      state: :f
+    - suffix: :s15
+      state: :g
+    - suffix: :s17
+      state: :e
+    - suffix: :s10
+      state: :e
+    - suffix: :s19
+      state: :m
+    - suffix: :s4
+      state: :h
+    - suffix: :s9
+      state: :c
+    - suffix: :s12
+      state: :f
+    - suffix: :s13
+      state: :b
+    - suffix: :s18
+      state: :d
+    - suffix: :s2
+      state: :h
+    - suffix: :s6
+      state: :h
+    - suffix: :s8
+      state: :b
+    - suffix: :s11
+      state: :b
+
+  final_state: true
+
+b:
+  transitions:
+    - suffix: :s3
+      state: :h
+    - suffix: :s5
+      state: :h
+    - suffix: :s1
+      state: :l
+    - suffix: :s4
+      state: :h
+    - suffix: :s2
+      state: :h
+
+  final_state: true
+
+c:
+  transitions:
+    # Transitions to state K
+    - suffix: :s7
+      state: :k
+    # Transitions to state H
+    - suffix: :s6
+      state: :h
+
+  final_state: false
+
+d:
+  transitions:
+    # Transitions to state F
+    - suffix: :s14
+      state: :f
+    # Transitions to state E
+    - suffix: :s10
+      state: :e
+    # Transitions to state B
+    - suffix: :s13
+      state: :b
+
+  final_state: false
+
+e:
+  transitions:
+    - suffix: :s7
+      state: :k
+    - suffix: :s3
+      state: :h
+    - suffix: :s5
+      state: :h
+    - suffix: :s1
+      state: :l
+    - suffix: :s4
+      state: :h
+    - suffix: :s18
+      state: :d
+    - suffix: :s2
+      state: :h
+    - suffix: :s6
+      state: :h
+
+  final_state: true
+
+f:
+  transitions:
+    # Transitions to state K
+    - suffix: :s7
+      state: :k
+    # Transitions to state D
+    - suffix: :s18
+      state: :d
+    # Transitions to state H
+    - suffix: :s6
+      state: :h
+
+  final_state: false
+
+
+g:
+  transitions:
+    - suffix: :s5
+      state: :h
+    - suffix: :s3
+      state: :h
+    - suffix: :s1
+      state: :l
+    - suffix: :s4
+      state: :h
+    - suffix: :s18
+      state: :d
+    - suffix: :s2
+      state: :h
+
+  final_state: true
+
+h:
+  transitions:
+    # Transitions to state L
+    - suffix: :s1
+      state: :l
+
+  final_state: true
+
+k:
+  transitions: []
+
+  final_state: true
+
+l:
+  transitions:
+    # Transitions to state D
+    - suffix: :s18
+      state: :d
+
+  final_state: true
+
+m:
+  transitions:
+    - suffix: :s7
+      state: :k
+    - suffix: :s3
+      state: :h
+    - suffix: :s5
+      state: :h
+    - suffix: :s1
+      state: :l
+    - suffix: :s4
+      state: :h
+    - suffix: :s2
+      state: :h
+    - suffix: :s6
+      state: :h
+
+  final_state: true
+

data/config/noun_suffixes.yml

@@ -0,0 +1,113 @@
+s1:
+  name: "-lAr"
+  regex: "lar|ler"
+  optional_letter: false
+  check_harmony: true
+
+s2:
+  name: "-(U)m"
+  regex: "m"
+  optional_letter: "ı|i|u|ü"
+  check_harmony: true
+
+s3:
+  name: "-(U)mUz"
+  regex: "mız|miz|muz|müz"
+  optional_letter: "ı|i|u|ü"
+  check_harmony: true
+
+s4:
+  name: "-Un"
+  regex: "ın|in|un|ün"
+  optional_letter: false
+  check_harmony: true
+
+s5:
+  name: "-(U)nUz"
+  regex: "nız|niz|nuz|nüz"
+  optional_letter: "ı|i|u|ü"
+  check_harmony: true
+
+s6:
+  name: "-(s)U"
+  regex: "ı|i|u|ü"
+  optional_letter: "s"
+  check_harmony: true
+
+s7:
+  name: "-lArI"
+  regex: "ları|leri"
+  optional_letter: false
+  check_harmony: true
+
+s8:
+  name: "-(y)U"
+  regex: "ı|i|u|ü"
+  optional_letter: "y"
+  check_harmony: true
+
+s9:
+  name: "-nU"
+  regex: "nı|ni|nu|nü"
+  optional_letter: false
+  check_harmony: true
+
+s10:
+  name: "-(n)Un"
+  regex: "ın|in|un|ün"
+  optional_letter: "n"
+  check_harmony: true
+
+s11:
+  name: "-(y)A"
+  regex: "a|e"
+  optional_letter: "y"
+  check_harmony: true
+
+s12:
+  name: "-nA"
+  regex: "na|ne"
+  optional_letter: false
+  check_harmony: true
+
+s13:
+  name: "-DA"
+  regex: "da|de|ta|te"
+  optional_letter: false
+  check_harmony: true
+
+s14:
+  name: "-nDA"
+  regex: "nta|nte|nda|nde"
+  optional_letter: false
+  check_harmony: true
+
+s15:
+  name: "-DAn"
+  regex: "dan|tan|den|ten"
+  optional_letter: false
+  check_harmony: true
+
+s16:
+  name: "-nDAn"
+  regex: "ndan|ntan|nden|nten"
+  optional_letter: false
+  check_harmony: true
+
+s17:
+  name: "-(y)lA"
+  regex: "la|le"
+  optional_letter: "y"
+  check_harmony: true
+
+s18:
+  name: "-ki"
+  regex: "ki"
+  optional_letter: false
+  check_harmony: false
+
+s19:
+  name: "-(n)cA"
+  regex: "ca|ce"
+  optional_letter: "n"
+  check_harmony: true