jaro_winkler 1.3.5 → 1.3.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 32cfc9b7e44ed814f68170efdf3f799ed4641868
-  data.tar.gz: 81e3e5311cc315cf302d8a1f34509e263e6cb8f0
+  metadata.gz: d48fca750c919028bdf13a2145bf431919a2c924
+  data.tar.gz: 134da623e4f99a708cf83f7e2e961a8882ecffb8
 SHA512:
-  metadata.gz: 874bcc658f618d3b1843957945543c1a5f2c465333d75c7f805c3179aad2db897b25fd8471ac8f7e4cb02216d2a1d7a440664b8bc0f89444dcc88507d5d35cae
-  data.tar.gz: c6842017d743987b8ce40e6ec752e1184f455a5db56c5119a064b37558fb5a22c25e19a65e79d7b43a06406227bb5826665f8c72063fe90afe87972209c8e363
+  metadata.gz: 7476d0dcd726f7d6b2405c861b139f99b22c71e21e48607beab76c1a29a11b6e9398c33943e8bfcff042b18fdea2d66f858283df17bbfd4e351a3ca93ee8f05a
+  data.tar.gz: 1c0a9c5b3521e761c752bc19a09a24052da3a46357195b9a86c86815418fc55bbe090b44db738bf086a44aabb1056f660a1584f1c8c7d01d124faf20623041b4

data/README.md

@@ -1,7 +1,5 @@
 [![Build Status](https://travis-ci.org/tonytonyjan/jaro_winkler.svg?branch=master)](https://travis-ci.org/tonytonyjan/jaro_winkler)
 
-# About
-
 It's an implementation of [Jaro-Winkler distance](http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance) algorithm, it uses C extension and will fallback to pure Ruby version in JRuby. Both of them supports UTF-8 string.
 
 # Installation
@@ -35,7 +33,7 @@ weight      | number  | 0.1     | A constant scaling factor for how much the sco
 threshold   | number  | 0.7     | The prefix bonus is only added when the compared strings have a Jaro distance above the threshold.
 adj_table   | boolean | false   | The option is used to give partial credit for characters that may be errors due to known phonetic or character recognition errors. A typical example is to match the letter "O" with the number "0".
 
-# About Adjusting Table
+# Adjusting Table
 
 ## Default Table
 
@@ -65,15 +63,6 @@ where
 
 - `s` is the number of nonmatching but similar characters.
 
-## Difference Between v1.3.1 And v1.3.2.beta
-
-Version     | Algorithm
------------ | -----------------------------------------------------------------------
-v1.3.1      | One linked list to store sparse matrix and iterate to find similar character.
-v1.3.2.beta | One hash table with multiple linked lists for collision handling.
-
-In theory, the latter should work more efficient than the former (more test data needed).
-
 # Why This?
 
 There is also another similar gem named [fuzzy-string-match](https://github.com/kiyoka/fuzzy-string-match) which both provides C and Ruby version as well.
@@ -108,9 +97,9 @@ str_1      | str_2      | origin | jaro_winkler | fuzzystringmatch | hotwater |
 - The origin result is from the [original C implementation by the author of the algorithm](http://web.archive.org/web/20100227020019/http://www.census.gov/geo/msb/stand/strcmp.c).
 - Test data are borrowed from [fuzzy-string-match's rspec file](https://github.com/kiyoka/fuzzy-string-match/blob/master/test/basic_pure_spec.rb).
 
-## Benchmark
+# Benchmark
 
-### Pure Ruby
+## Pure Ruby
 
                  | user     | system   | total    | real
 ---------------- | -------- | -------- | -------- | ------------
@@ -120,7 +109,7 @@ fuzzystringmatch | 1.510000 | 0.000000 | 1.510000 | (  1.510136)
 - jaro_winkler (1.3.1)
 - fuzzy-string-match (0.9.6)
 
-### Native
+## Native
 
                  | user     | system   | total    | real
 ---------------- | -------- | -------- | -------- | ------------
@@ -136,4 +125,5 @@ amatch           | 0.960000 | 0.000000 | 0.960000 | (  0.961509)
 
 # Todo
 
-- Custom adjusting word table.
+- Custom adjusting word table.
+- The algorithm between C and Ruby are different.

data/ext/jaro_winkler/adj_matrix.c

@@ -1,6 +1,7 @@
-#include <stdlib.h>
 #include "adj_matrix.h"
-#include "codepoints.h"
+#include "code.h"
+
+#include <stdlib.h>
 
 const char *DEFAULT_ADJ_TABLE[] = {
   "A","E", "A","I", "A","O", "A","U", "B","V", "E","I", "E","O", "E","U", "I","O", "I","U", "O","U",
@@ -76,8 +77,11 @@ AdjMatrix* adj_matrix_default(){
     ret_matrix = adj_matrix_new(ADJ_MATRIX_DEFAULT_LENGTH);
     int length = sizeof(DEFAULT_ADJ_TABLE) / sizeof(char*);
     for(int i = 0; i < length; i += 2){
-      UnicodeHash h1 = unicode_hash_new(DEFAULT_ADJ_TABLE[i]), h2 = unicode_hash_new(DEFAULT_ADJ_TABLE[i + 1]);
-      adj_matrix_add(ret_matrix, h1.code, h2.code);
+      unsigned long long code_1, code_2;
+      int dummy_length;
+      utf_char_to_code((char*)DEFAULT_ADJ_TABLE[i], &code_1, &dummy_length);
+      utf_char_to_code((char*)DEFAULT_ADJ_TABLE[i+1], &code_2, &dummy_length);
+      adj_matrix_add(ret_matrix, code_1, code_2);
     }
     first_time = 0;
   }

data/ext/jaro_winkler/adj_matrix.h

@@ -1,5 +1,5 @@
 #ifndef ADJ_MATRIX_H
-#define ADJ_MATRIX_H 1
+#define ADJ_MATRIX_H
 #define ADJ_MATRIX_DEFAULT_LENGTH 958
 #define ADJ_MATRIX_SEED 9527
 
@@ -19,4 +19,4 @@ char       adj_matrix_find   (AdjMatrix *matrix, unsigned long long x, unsigned
 void       adj_matrix_free   (AdjMatrix *matrix);
 AdjMatrix* adj_matrix_default();
 
-#endif /* ADJ_MATRIX_H */
+#endif

data/ext/jaro_winkler/code.c

@@ -0,0 +1,29 @@
+#include <stdlib.h>
+#include <string.h>
+
+void utf_char_to_code(char *str, unsigned long long *ret_code, int *ret_byte_length){
+  unsigned char first_char = str[0];
+  if(first_char >= 252) *ret_byte_length = 6;      // 1111110x
+  else if(first_char >= 248) *ret_byte_length = 5; // 111110xx
+  else if(first_char >= 240) *ret_byte_length = 4; // 11110xxx
+  else if(first_char >= 224) *ret_byte_length = 3; // 1110xxxx
+  else if(first_char >= 192) *ret_byte_length = 2; // 110xxxxx
+  else *ret_byte_length = 1;
+  *ret_code = 0;
+  memcpy(ret_code, str, *ret_byte_length);
+}
+
+void string_to_codes(char *str, int length, unsigned long long **ret_codes, int *ret_length){
+  unsigned int code;
+  char byte_length;
+
+  *ret_codes = calloc(length, sizeof(long long));
+  *ret_length = 0;
+
+  for(int i = 0; i < length;){
+    int byte_length;
+    utf_char_to_code(&str[i], &(*ret_codes)[*ret_length], &byte_length);
+    *ret_length += 1;
+    i += byte_length;
+  }
+}

data/ext/jaro_winkler/code.h

@@ -0,0 +1,7 @@
+#ifndef CODE_H
+#define CODE_H
+
+void utf_char_to_code(char *str, unsigned long long *ret_code, int *ret_byte_length);
+void string_to_codes(char *str, int length, unsigned long long **ret_codes, int *ret_length);
+
+#endif

data/ext/jaro_winkler/jaro.c

@@ -0,0 +1,103 @@
+#include "jaro.h"
+#include "code.h"
+#include "adj_matrix.h"
+
+#include <string.h>
+#include <stdlib.h>
+#include <ctype.h>
+
+double jaro_winkler_distance(char* short_str, int short_str_len, char* long_str, int long_str_len, LibJaroOption *opt){
+  if(!short_str_len || !long_str_len) return 0.0;
+
+  if(short_str_len > long_str_len){
+    SWAP(short_str, long_str);
+    SWAP(short_str_len, long_str_len);
+  }
+
+  unsigned long long *short_codes, *long_codes;
+  int short_codes_len, long_codes_len;
+  string_to_codes(short_str, short_str_len, &short_codes, &short_codes_len);
+  string_to_codes(long_str, long_str_len, &long_codes, &long_codes_len);
+
+  if(opt->ignore_case){
+    for(int i = 0; i < short_codes_len; i++) short_codes[i] = tolower(short_codes[i]);
+    for(int i = 0; i < long_codes_len; i++) long_codes[i] = tolower(long_codes[i]);
+  }
+
+  int window_size = long_codes_len/2 - 1;
+  if(window_size < 0) window_size = 0;
+
+  char short_codes_flag[MAX_WORD_LENGTH];
+  char long_codes_flag[MAX_WORD_LENGTH];
+  memset(short_codes_flag, 0, MAX_WORD_LENGTH);
+  memset(long_codes_flag, 0, MAX_WORD_LENGTH);
+
+  // count number of matching characters
+  int match_count = 0;
+  for(int i = 0; i < short_codes_len; i++){
+    int left = (i >= window_size) ? i - window_size : 0;
+    int right = (i + window_size <= long_codes_len - 1) ? (i + window_size) : (long_codes_len - 1);
+    if(right > long_codes_len - 1) right = long_codes_len - 1;
+    for(int j = left; j <= right; j++){
+      if(!long_codes_flag[j] && short_codes[i] == long_codes[j]){
+        short_codes_flag[i] = long_codes_flag[j] = 1;
+        match_count++;
+        break;
+      }
+    }
+  }
+  if(!match_count){
+    free(short_codes); free(long_codes);
+    return 0.0;
+  }
+
+  // count number of transpositions
+  int transposition_count = 0, j = 0, k = 0;
+  for(int i = 0; i < short_codes_len; i++){
+    if(short_codes_flag[i]){
+      for(j = k; j < long_codes_len; j++){
+        if(long_codes_flag[j]){
+          k = j + 1;
+          break;
+        }
+      }
+      if(short_codes[i] != long_codes[j]) transposition_count++;
+    }
+  }
+
+  // count similarities in nonmatched characters
+  int similar_count = 0;
+  if(opt->adj_table && short_codes_len > match_count)
+    for(int i = 0; i < short_codes_len; i++)
+      if(!short_codes_flag[i])
+        for(int j = 0; j < long_codes_len; j++)
+          if(!long_codes_flag[j])
+            if(adj_matrix_find(adj_matrix_default(), short_codes[i], long_codes[j])){
+              similar_count += 3;
+              break;
+            }
+
+  // jaro distance
+  double jaro_distance;
+  double m = (double)match_count;
+  double t = (double)(transposition_count/2);
+  if(opt->adj_table) m = similar_count/10.0 + m;
+  jaro_distance = (m/short_codes_len + m/long_codes_len + (m-t)/m) / 3;
+
+  // jaro winkler distance
+  if(!opt){
+    static LibJaroOption default_opt = {.weight = DEFAULT_WEIGHT, .threshold = DEFAULT_THRESHOLD};
+    opt = &default_opt;
+  }
+  if(jaro_distance < opt->threshold){
+    free(short_codes); free(long_codes);
+    return jaro_distance;
+  }
+  else{
+    int prefix = 0;
+    int max_4 = short_codes_len > 4 ? 4 : short_codes_len;
+    for(prefix = 0; prefix < max_4 && short_codes[prefix] == long_codes[prefix]; prefix++);
+    free(short_codes); free(long_codes);
+    return jaro_distance + prefix*opt->weight*(1-jaro_distance);
+  }
+}

data/ext/jaro_winkler/jaro.h

@@ -0,0 +1,16 @@
+#ifndef LIBJARO_JARO_H
+#define LIBJARO_JARO_H
+
+#define SWAP(x, y) do{ __typeof__(x) SWAP = x; x = y; y = SWAP; }while(0)
+#define MAX_WORD_LENGTH 64
+#define DEFAULT_WEIGHT 0.1
+#define DEFAULT_THRESHOLD 0.7
+
+typedef struct LibJaroOption{
+  double weight, threshold;
+  char ignore_case, adj_table;
+} LibJaroOption;
+
+double jaro_winkler_distance(char *str1, int len1, char *str2, int len2, LibJaroOption *opt);
+
+#endif

data/ext/jaro_winkler/jaro_winkler.c

@@ -1,27 +1,28 @@
-#include "jaro_winkler.h"
-#include "distance.h"
+#include "ruby.h"
+#include "jaro.h"
 
 VALUE rb_mJaroWinkler;
+VALUE distance(int argc, VALUE *argv, VALUE self);
 
 void Init_jaro_winkler(void){
   rb_mJaroWinkler = rb_define_module("JaroWinkler");
-  rb_define_module_function(rb_mJaroWinkler, "c_distance", rb_distance, -1);
+  rb_define_module_function(rb_mJaroWinkler, "c_distance", distance, -1);
 }
 
-VALUE rb_distance(int argc, VALUE *argv, VALUE self){
+VALUE distance(int argc, VALUE *argv, VALUE self){
   VALUE s1, s2, opt;
   rb_scan_args(argc, argv, "2:", &s1, &s2, &opt);
-  Option c_opt = option_new();
+  LibJaroOption c_opt = {.weight = DEFAULT_WEIGHT, .threshold = DEFAULT_THRESHOLD, .ignore_case = 0, .adj_table = 0};
   if(TYPE(opt) == T_HASH){
-    VALUE weight      = rb_hash_aref(opt, ID2SYM(rb_intern("weight"))),
-          threshold   = rb_hash_aref(opt, ID2SYM(rb_intern("threshold"))),
+    VALUE weight = rb_hash_aref(opt, ID2SYM(rb_intern("weight"))),
+          threshold = rb_hash_aref(opt, ID2SYM(rb_intern("threshold"))),
           ignore_case = rb_hash_aref(opt, ID2SYM(rb_intern("ignore_case"))),
-          adj_table   = rb_hash_aref(opt, ID2SYM(rb_intern("adj_table")));
+          adj_table = rb_hash_aref(opt, ID2SYM(rb_intern("adj_table")));
     if(!NIL_P(weight)) c_opt.weight = NUM2DBL(weight);
     if(c_opt.weight > 0.25) rb_raise(rb_eRuntimeError, "Scaling factor should not exceed 0.25, otherwise the distance can become larger than 1.");
-    if(!NIL_P(threshold)) c_opt.threshold     = NUM2DBL(threshold);
-    if(!NIL_P(ignore_case)) c_opt.ignore_case = (TYPE(ignore_case)  == T_FALSE || NIL_P(ignore_case)) ? 0 : 1;
-    if(!NIL_P(adj_table)) c_opt.adj_table     = (TYPE(adj_table)    == T_FALSE || NIL_P(adj_table))   ? 0 : 1;
+    if(!NIL_P(threshold)) c_opt.threshold = NUM2DBL(threshold);
+    if(!NIL_P(ignore_case)) c_opt.ignore_case = (TYPE(ignore_case) == T_FALSE || NIL_P(ignore_case)) ? 0 : 1;
+    if(!NIL_P(adj_table)) c_opt.adj_table = (TYPE(adj_table) == T_FALSE || NIL_P(adj_table)) ? 0 : 1;
   }
-  return rb_float_new(distance(StringValuePtr(s1), RSTRING_LEN(s1), StringValuePtr(s2), RSTRING_LEN(s2), c_opt));
+  return rb_float_new(jaro_winkler_distance(StringValuePtr(s1), RSTRING_LEN(s1), StringValuePtr(s2), RSTRING_LEN(s2), &c_opt));
 }

data/lib/jaro_winkler/adjusting_table.rb

@@ -1,10 +1,19 @@
 module JaroWinkler
-  DEFAULT_ADJ_TABLE = Hash.new({})
+  DEFAULT_ADJ_TABLE = Hash.new
   [
     ['A', 'E'], ['A', 'I'], ['A', 'O'], ['A', 'U'], ['B', 'V'], ['E', 'I'], ['E', 'O'], ['E', 'U'], ['I', 'O'],
     ['I', 'U'], ['O', 'U'], ['I', 'Y'], ['E', 'Y'], ['C', 'G'], ['E', 'F'], ['W', 'U'], ['W', 'V'], ['X', 'K'],
     ['S', 'Z'], ['X', 'S'], ['Q', 'C'], ['U', 'V'], ['M', 'N'], ['L', 'I'], ['Q', 'O'], ['P', 'R'], ['I', 'J'],
     ['2', 'Z'], ['5', 'S'], ['8', 'B'], ['1', 'I'], ['1', 'L'], ['0', 'O'], ['0', 'Q'], ['C', 'K'], ['G', 'J'],
     ['E', ' '], ['Y', ' '], ['S', ' ']
-  ].each{ |s1, s2| DEFAULT_ADJ_TABLE[s1][s2] = DEFAULT_ADJ_TABLE[s2][s1] = true }
-end
+  ].each{ |s1, s2|
+    if not DEFAULT_ADJ_TABLE.has_key?(s1)
+      DEFAULT_ADJ_TABLE[s1] = Hash.new
+    end
+    if not DEFAULT_ADJ_TABLE.has_key?(s2)
+      DEFAULT_ADJ_TABLE[s2] = Hash.new
+    end
+    DEFAULT_ADJ_TABLE[s1][s2] = DEFAULT_ADJ_TABLE[s2][s1] = true
+  }
+  DEFAULT_ADJ_TABLE.default = Hash.new
+end

data/lib/jaro_winkler/version.rb

@@ -1,3 +1,3 @@
 module JaroWinkler
-  VERSION = "1.3.5"
+  VERSION = "1.3.6"
 end

data/spec/adjusting_table_spec.rb

@@ -0,0 +1,8 @@
+# spec/adjusting_table_spec.rb
+require 'jaro_winkler'
+
+describe JaroWinkler::DEFAULT_ADJ_TABLE do
+  it 'should not be empty' do
+    expect(JaroWinkler::DEFAULT_ADJ_TABLE).not_to be_empty
+  end
+end

data/spec/jaro_winkler_spec.rb

@@ -4,40 +4,29 @@ include JaroWinkler
 
 shared_examples 'common' do |strategy|
   it 'works' do
-    ary = [
-      ['henka'         , 'henkan'       , 0.9667] ,
-      ['al'            , 'al'           , 1.0]    ,
-      ['martha'        , 'marhta'       , 0.9611] ,
-      ['jones'         , 'johnson'      , 0.8323] ,
-      ['abcvwxyz'      , 'cabvwxyz'     , 0.9583] ,
-      ['dwayne'        , 'duane'        , 0.8400] ,
-      ['dixon'         , 'dicksonx'     , 0.8133] ,
-      ['fvie'          , 'ten'          , 0.0]    ,
-      ['tony'          , 'tony'         , 1.0]    ,
-      ['tonytonyjan'   , 'tonytonyjan'  , 1.0]    ,
-      ['x'             , 'x'            , 1.0]    ,
-      [''              , ''             , 0.0]    ,
-      ['tony'          , ''             , 0.0]    ,
-      [''              , 'tony'         , 0.0]    ,
-      ['tonytonyjan'   , 'tony'         , 0.8727] ,
-      ['tony'          , 'tonytonyjan'  , 0.8727] ,
-      # ['San Francisco' , 'Santa Monica' , 0.8180]
-    ]
-    ary.each do |s1, s2, ans|
-      expect(send(strategy, s1, s2)).to be_within(0.0001).of(ans)
-    end
+    expect(send(strategy, 'henka','henkan')).to be_within(0.0001).of(0.9667)
+    expect(send(strategy, 'al','al')).to be_within(0.0001).of(1.0)
+    expect(send(strategy, 'martha','marhta')).to be_within(0.0001).of(0.9611)
+    expect(send(strategy, 'jones','johnson')).to be_within(0.0001).of(0.8323)
+    expect(send(strategy, 'abcvwxyz','cabvwxyz')).to be_within(0.0001).of(0.9583)
+    expect(send(strategy, 'dwayne','duane')).to be_within(0.0001).of(0.8400)
+    expect(send(strategy, 'dixon','dicksonx')).to be_within(0.0001).of(0.8133)
+    expect(send(strategy, 'fvie','ten')).to be_within(0.0001).of(0.0)
+    expect(send(strategy, 'tony','tony')).to be_within(0.0001).of(1.0)
+    expect(send(strategy, 'tonytonyjan','tonytonyjan')).to be_within(0.0001).of(1.0)
+    expect(send(strategy, 'x','x')).to be_within(0.0001).of(1.0)
+    expect(send(strategy, '','')).to be_within(0.0001).of(0.0)
+    expect(send(strategy, 'tony','')).to be_within(0.0001).of(0.0)
+    expect(send(strategy, '','tony')).to be_within(0.0001).of(0.0)
+    expect(send(strategy, 'tonytonyjan','tony')).to be_within(0.0001).of(0.8727)
+    expect(send(strategy, 'tony','tonytonyjan')).to be_within(0.0001).of(0.8727)
   end
 
   it 'works with UTF-8' do
-    ary = [
-      ['變形金剛4:絕跡重生' , '變形金剛4: 絕跡重生' , 0.9818] ,
-      ['連勝文'             , '連勝丼'              , 0.8222] ,
-      ['馬英九'             , '馬英丸'              , 0.8222] ,
-      ['良い'               , 'いい'                , 0.6666] ,
-    ]
-    ary.each do |s1, s2, ans|
-      expect(send(strategy, s1, s2)).to be_within(0.0001).of(ans)
-    end
+    expect(send(strategy, '變形金剛4:絕跡重生','變形金剛4: 絕跡重生')).to be_within(0.0001).of(0.9818)
+    expect(send(strategy, '連勝文','連勝丼')).to be_within(0.0001).of(0.8222)
+    expect(send(strategy, '馬英九','馬英丸')).to be_within(0.0001).of(0.8222)
+    expect(send(strategy, '良い','いい')).to be_within(0.0001).of(0.6666)
   end
 
   it 'sets ignore_case' do
@@ -54,19 +43,14 @@ shared_examples 'common' do |strategy|
 
 
   it 'works with adjusting table' do
-    ary = [
-      ['HENKA'       , 'HENKAN'      , 0.9667] , # m=5, t=0, s=0
-      ['AL'          , 'AL'          , 1.0   ] , # m=2, t=0, s=0
-      ['MARTHA'      , 'MARHTA'      , 0.9611] , # m=6, t=1, s=0
-      ['JONES'       , 'JOHNSON'     , 0.8598] , # m=4, t=0, s=3
-      ['ABCVWXYZ'    , 'CABVWXYZ'    , 0.9583] , # m=8, t=1, s=0
-      ['DWAYNE'      , 'DUANE'       , 0.8730] , # m=4, t=0, s=3
-      ['DIXON'       , 'DICKSONX'    , 0.8393] , # m=4, t=0, s=3
-      ['FVIE'        , 'TEN'         , 0.0   ]
-    ]
-    ary.each do |s1, s2, ans|
-      expect(send(strategy, s1, s2, adj_table: true)).to be_within(0.0001).of(ans)
-    end
+    expect(send(strategy, 'HENKA', 'HENKAN', adj_table: true)).to be_within(0.0001).of(0.9667) # m=5, t=0, s=0
+    expect(send(strategy, 'AL', 'AL', adj_table: true)).to be_within(0.0001).of(1.0) # m=2, t=0, s=0
+    expect(send(strategy, 'MARTHA', 'MARHTA', adj_table: true)).to be_within(0.0001).of(0.9611) # m=6, t=1, s=0
+    expect(send(strategy, 'JONES', 'JOHNSON', adj_table: true)).to be_within(0.0001).of(0.8598) # m=4, t=0, s=3
+    expect(send(strategy, 'ABCVWXYZ', 'CABVWXYZ', adj_table: true)).to be_within(0.0001).of(0.9583) # m=8, t=1, s=0
+    expect(send(strategy, 'DWAYNE', 'DUANE', adj_table: true)).to be_within(0.0001).of(0.8730) # m=4, t=0, s=3
+    expect(send(strategy, 'DIXON', 'DICKSONX', adj_table: true)).to be_within(0.0001).of(0.8393) # m=4, t=0, s=3
+    expect(send(strategy, 'FVIE', 'TEN', adj_table: true)).to be_within(0.0001).of(0.0)
   end
 
   context 'with weight exceeding 0.25' do

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: jaro_winkler
 version: !ruby/object:Gem::Version
-  version: 1.3.5
+  version: 1.3.6
 platform: ruby
 authors:
 - Jian Weihang
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-04-02 00:00:00.000000000 Z
+date: 2015-06-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -131,19 +131,19 @@ files:
 - benchmark/pure.txt
 - ext/jaro_winkler/adj_matrix.c
 - ext/jaro_winkler/adj_matrix.h
-- ext/jaro_winkler/codepoints.c
-- ext/jaro_winkler/codepoints.h
-- ext/jaro_winkler/distance.c
-- ext/jaro_winkler/distance.h
+- ext/jaro_winkler/code.c
+- ext/jaro_winkler/code.h
 - ext/jaro_winkler/extconf.rb
+- ext/jaro_winkler/jaro.c
+- ext/jaro_winkler/jaro.h
 - ext/jaro_winkler/jaro_winkler.c
-- ext/jaro_winkler/jaro_winkler.h
 - ext/jaro_winkler/murmur_hash2.c
 - jaro_winkler.gemspec
 - lib/jaro_winkler.rb
 - lib/jaro_winkler/adjusting_table.rb
 - lib/jaro_winkler/fallback.rb
 - lib/jaro_winkler/version.rb
+- spec/adjusting_table_spec.rb
 - spec/jaro_winkler_spec.rb
 - spec/spec_helper.rb
 homepage: https://github.com/tonytonyjan/jaro_winkler
@@ -166,11 +166,12 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.5
+rubygems_version: 2.4.6
 signing_key: 
 specification_version: 4
 summary: Ruby & C implementation of Jaro-Winkler distance algorithm which both support
   UTF-8 string.
 test_files:
+- spec/adjusting_table_spec.rb
 - spec/jaro_winkler_spec.rb
 - spec/spec_helper.rb

data/ext/jaro_winkler/codepoints.c

@@ -1,29 +0,0 @@
-#include <string.h>
-#include <stdlib.h>
-#include "codepoints.h"
-
-UnicodeHash unicode_hash_new(const char *str){
-  UnicodeHash ret = {};
-  unsigned char first_char = str[0];
-  if(first_char >= 252) ret.byte_length = 6;      // 1111110x
-  else if(first_char >= 248) ret.byte_length = 5; // 111110xx
-  else if(first_char >= 240) ret.byte_length = 4; // 11110xxx
-  else if(first_char >= 224) ret.byte_length = 3; // 1110xxxx
-  else if(first_char >= 192) ret.byte_length = 2; // 110xxxxx
-  else ret.byte_length = 1;
-  memcpy(&ret.code, str, ret.byte_length);
-  return ret;
-}
-
-Codepoints codepoints_new(const char *str, int byte_len){
-  Codepoints ret = {};
-  ret.ary = malloc(byte_len * sizeof(long long));
-  ret.length = 0;
-  for(int i = 0; i < byte_len;){
-    UnicodeHash hash = unicode_hash_new(str + i);
-    ret.ary[ret.length] = hash.code;
-    ret.length++;
-    i += hash.byte_length;
-  }
-  return ret;
-}

data/ext/jaro_winkler/codepoints.h

@@ -1,17 +0,0 @@
-#ifndef CODEPOINTS_H
-#define CODEPOINTS_H 1
-
-typedef struct{
-  unsigned long long code;
-  unsigned int byte_length;
-} UnicodeHash;
-
-typedef struct{
-  unsigned long long *ary;
-  int length;
-} Codepoints;
-
-UnicodeHash unicode_hash_new(const char *str);
-Codepoints  codepoints_new  (const char *str, int byte_len);
-
-#endif /* CODEPOINTS_H */

data/ext/jaro_winkler/distance.c

@@ -1,76 +0,0 @@
-#include <stdlib.h>
-#include <ctype.h>
-#include "distance.h"
-#include "codepoints.h"
-#include "adj_matrix.h"
-
-Option option_new(){
-  Option opt;
-  opt.ignore_case = opt.adj_table = 0;
-  opt.weight = 0.1;
-  opt.threshold = 0.7;
-  return opt;
-}
-
-double distance(char *s1, int s1_byte_len, char *s2, int s2_byte_len, Option opt){
-  Codepoints code_ary_1 = codepoints_new(s1, s1_byte_len),
-             code_ary_2 = codepoints_new(s2, s2_byte_len);
-
-  if(opt.ignore_case){
-    for(int i = 0; i < code_ary_1.length; ++i) if(code_ary_1.ary[i] < 256 && islower(code_ary_1.ary[i])) code_ary_1.ary[i] -= 32;
-    for(int i = 0; i < code_ary_2.length; ++i) if(code_ary_2.ary[i] < 256 && islower(code_ary_2.ary[i])) code_ary_2.ary[i] -= 32;
-  }
-
-  // Guarantee the order
-  if(code_ary_1.length > code_ary_2.length){
-    unsigned long long *tmp = code_ary_1.ary; code_ary_1.ary = code_ary_2.ary; code_ary_2.ary = tmp;
-    int tmp2 = code_ary_1.length; code_ary_1.length = code_ary_2.length; code_ary_2.length = tmp2;
-  }
-
-  // Compute jaro distance
-  int window_size = code_ary_2.length / 2 - 1;
-  if(window_size < 0) window_size = 0;
-  double matches     = 0.0,
-         sim_matches = 0.0;
-  int transpositions = 0,
-      previous_index = -1,
-      max_index      = code_ary_2.length - 1;
-  for(int i = 0; i < code_ary_1.length; i++){
-    int left  = i - window_size;
-    int right = i + window_size;
-    if(left  < 0) left = 0;
-    if(right > max_index) right = max_index;
-    char matched     = 0,
-         found       = 0,
-         sim_matched = 0;
-    for(int j = left; j <= right; j++){
-      if(code_ary_1.ary[i] == code_ary_2.ary[j]){
-        matched = 1;
-        if(!found && j > previous_index){
-          previous_index = j;
-          found = 1;
-        }
-      }else if(opt.adj_table && adj_matrix_find(adj_matrix_default(), code_ary_1.ary[i], code_ary_2.ary[j])) sim_matched = 1;
-    } // for(int j = left; j <= right; j++){
-    if(matched){
-      matches++;
-      if(!found) transpositions++;
-    }else if(sim_matched) sim_matches += 3;
-  } // for(int i = 0; i < code_ary_1.length; i++){
-
-  // Don't divide transpositions by 2 since it's been counted directly by above code.
-  double similarity = matches;
-  if(opt.adj_table) similarity += sim_matches / 10;
-  double jaro_distance = matches == 0 ? 0 : (similarity / code_ary_1.length + similarity / code_ary_2.length + (matches - transpositions) / matches) / 3.0;
-
-  // calculate jaro-winkler distance
-  double threshold = opt.threshold, weight = opt.weight;
-  int prefix = 0;
-  int max_length = code_ary_1.length > 4 ? 4 : code_ary_1.length;
-  for(int i = 0; i < max_length; ++i){
-    if(code_ary_1.ary[i] == code_ary_2.ary[i]) prefix++;
-    else break;
-  }
-  free(code_ary_1.ary); free(code_ary_2.ary);
-  return jaro_distance < threshold ? jaro_distance : jaro_distance + ((prefix * weight) * (1 - jaro_distance));
-}

data/ext/jaro_winkler/distance.h

@@ -1,12 +0,0 @@
-#ifndef DISTANCE_H
-#define DISTANCE_H 1
-
-typedef struct{
-  double weight, threshold;
-  char ignore_case, adj_table;
-} Option;
-
-double distance(char *s1, int s1_byte_len, char *s2, int s2_byte_len, Option opt);
-Option option_new();
-
-#endif /* DISTANCE_H */

data/ext/jaro_winkler/jaro_winkler.h

@@ -1,8 +0,0 @@
-#ifndef JARO_WINKLER_H
-#define JARO_WINKLER_H 1
-
-#include "ruby.h"
-
-VALUE rb_distance(int argc, VALUE *argv, VALUE obj);
-
-#endif /* JARO_WINKLER_H */