New approach to metaphone computation

rafaelk · rafaelk · commit 665d2f26ded1 · 2010-11-12T23:03:52.000+01:00
diff --git a/_workbench/strings/metaphone.js b/_workbench/strings/metaphone.js
@@ -0,0 +1,99 @@
+function metaphone (str) {
+    // The code below is based on description from Wikipedia (http://en.wikipedia.org/wiki/Metaphone)
+    // There are some modifications applied, like
+    // - changing the order of rules
+    // - changing the rules to match PHP algorithm
+    // modifications are based on PHP metaphone source code
+    
+    // changing the input string to lower case
+    // all rules replace lower-case characters with upper-case, so following rules won't be applied to already computed parts
+    str = ('' + str).toLowerCase();
+    
+    var rules = [
+        // 1. Drop duplicate adjacent letters, except for G
+        // php also doesn't remove 'C' duplicates
+        /([^cg])\1+/g, '',
+        
+        // 2. If the word begins with 'KN', 'GN', 'PN', 'AE', 'WR', drop the first letter
+        
+        /^[gkp]n/, 'N',
+        /^ae/, 'E',
+        /^wr/, 'R',
+        
+        // 3. Drop 'B' if after 'M' and if it is at the end of the word
+        // php ignores the "end of word" part of the rule
+        /mb/g, '',
+        
+        // 9. 'CK' transforms to 'K'
+        // Applying the rule here, cause rule 4) replaces all 'C's with 'K'
+        /ck/g, 'K',
+                
+        // 4. 'C' transforms to 'X' if followed by 'IA' or 'H' (unless in latter case, it is part of '-SCH-', in which case it transforms to 'K'). 'C' transforms to 'S' if followed by 'I', 'E', or 'Y'. Otherwise, 'C' transforms to 'K'
+        /\Bsch/g, 'SK', // check if replace both S and H or only the C. Check if '\B' works correctly for multiple words
+        /c(?=ia)/g, 'X',
+        /[cs]h/g, 'X', // PHP also drops the 'H' character
+        /c(?=[eiy])/g, 'S',
+        /c/g, 'K',
+        
+        // 5. 'DG' transforms to 'J' if followed by 'GE', 'GY', or 'GI'. Otherwise, 'D' transforms to 'T
+        /dg(?=[eiy])/g, 'J',
+        /d/g, 'T',
+        
+        // 6. Drop 'G' if followed by 'H' and 'H' is not at the end or before a vowel. Drop 'G' if followed by 'N' or 'NED' and is at the end
+        /gh(?![aeiou]|$)/g, 'H',
+        /g(?=n(ed)?)$/g, '',
+        
+        // 7. 'G' transforms to 'J' if before 'I', 'E', or 'Y', and it is not in 'GG'. Otherwise, 'G' transforms to 'K'. Reduce 'GG' to 'G'
+        /(^|[^g])g(?=[eiy])/g, '$1J',
+        /g+/g, 'K', // check if that's the way GG is replaced with K, or it should be G in result string
+        
+        // 8. Drop 'H' if after vowel and not before a vowel
+        /([aeiou])h(?![aeiou])/g, '$1',
+        
+        // 10. 'PH' transforms to 'F'
+        /ph/g, 'F',
+        
+        // 11. 'Q' transforms to 'K'
+        /q/g, 'K',
+        
+        // 12. 'S' transforms to 'X' if followed by 'H', 'IO', or 'IA'
+        // php also drops the 'H' in 'SH' case, which is done in 'CH' rule
+        /s(?=i[oa])/g, 'X',
+        
+        // 13. 'T' transforms to 'X' if followed by 'IA' or 'IO'. 'T' transforms to '0' if followed by 'H'. Drop 'T' if followed by 'CH'
+        // php replaces 'TH' with '0', not only the 'T'
+        /t(?=i[ao])/g, 'X',
+        /th/g, '0',
+        /t(?=ch)/g, '', // check if this is valid, cause 'CH' case is already replaced with 'X'
+        
+        // 14. 'V' transforms to 'F'
+        /v/g, 'F',
+        
+        // 15. 'WH' transforms to 'W' if at the beginning. Drop 'W' if not followed by a vowel
+        // 17. Drop 'Y' if not followed by a vowel
+        /^wh/, 'W',
+        /[wy](?![aeiou])/g, '',
+        
+        // 16. 'X' transforms to 'S' if at the beginning. Otherwise, 'X' transforms to 'KS'
+        /^x/, 'S',
+        /x/g, 'KS',
+        
+        // 18. 'Z' transforms to 'S'
+        /z/g, 'S',
+        
+        // 19. Drop all vowels unless it is the beginning
+        /(^[aeiou])?[aeiou]?/g, '$1',
+        
+        // Remove all non-ascii characters
+        // doing case-insensitive replace, so that upper-case ASCII characters remain untouched
+        /[^a-z0]/ig, ''
+    ];
+    
+    for (var i = 0, l = rules.length; i < l; i += 2) {
+        str = str.replace(rules[i], rules[i + 1]);
+    }
+    
+    // transforming result to upper-case
+    return str.toUpperCase();
+}
+
