EN
TypeScript - Soundex algorithm implementation
0 points
In this short article, we would like to show how to implement the Soundex algorithm in TypeScript.
Below algorithm calculates Soundex code for indicated string.
Soundex codes can be used to:
- check words similarity (as an alternative: fuzzy comparison with bigrams),
- build fuzzy searching in databases,
- implement fuzzy map data structure (e.g. FuzzyHashMap).
Note: that algorithm works well for English words.
Example implementation:
xxxxxxxxxx1
const findStartingCode = (word: string): string => {2
return word[0].toUpperCase();3
};4
5
const findLetterCode = (letter: string): string => {6
switch (letter.toUpperCase()) {7
case 'B':8
case 'F':9
case 'P':10
case 'V':11
return '1';12
13
case 'C':14
case 'G':15
case 'J':16
case 'K':17
case 'Q':18
case 'S':19
case 'X':20
case 'Z':21
return '2';22
23
case 'D':24
case 'T':25
return '3';26
27
case 'L':28
return '4';29
30
case 'M':31
case 'N':32
return '5';33
34
case 'R':35
return '6';36
37
default:38
return null;39
}40
};41
42
const calculateSoundexCode = (word: string): string => {43
if (word) {44
let wordCode = findStartingCode(word);45
let lastCode = findLetterCode(wordCode); // wordCode contains one letter at the begining46
for (let i = 1; i < word.length; ++i) {47
const letterCode = findLetterCode(word[i]);48
if (letterCode && letterCode != lastCode) {49
wordCode += letterCode;50
if (wordCode.length == 4) {51
break;52
}53
}54
lastCode = letterCode;55
}56
for (let i = wordCode.length; i < 4; ++i) {57
wordCode += '0';58
}59
return wordCode;60
}61
return null;62
};63
64
65
// Usage example:66
67
console.log(calculateSoundexCode('Robert')); // R16368
console.log(calculateSoundexCode('Rupert')); // R16369
console.log(calculateSoundexCode('Rubin')); // R15070
console.log(calculateSoundexCode('Ashcraft')); // A22671
console.log(calculateSoundexCode('Ashcroft')); // A22672
console.log(calculateSoundexCode('Tymczak')); // T52273
console.log(calculateSoundexCode('Pfister')); // P23674
console.log(calculateSoundexCode('Honeyman')); // H55575
Output:
xxxxxxxxxx1
R1632
R1633
R1504
A2265
A2266
T5227
P2368
H555