Add more unicode related tests to longestCommonSubstring algorithm.

This commit is contained in:
Oleksii Trekhleb 2018-08-28 06:58:43 +03:00
parent 82ac89b16a
commit af64d12a23
2 changed files with 27 additions and 13 deletions

View File

@ -7,6 +7,16 @@ describe('longestCommonSubstring', () => {
expect(longestCommonSubstring('', 'ABC')).toBe('');
expect(longestCommonSubstring('ABABC', 'BABCA')).toBe('BABC');
expect(longestCommonSubstring('BABCA', 'ABCBA')).toBe('ABC');
expect(longestCommonSubstring(
'Algorithms and data structures implemented in JavaScript',
'Here you may find Algorithms and data structures that are implemented in JavaScript',
)).toBe('Algorithms and data structures ');
});
it('should handle unicode correctly', () => {
expect(longestCommonSubstring('𐌵𐌵**ABC', '𐌵𐌵--ABC')).toBe('ABC');
expect(longestCommonSubstring('𐌵𐌵**A', '𐌵𐌵--A')).toBe('𐌵𐌵');
expect(longestCommonSubstring('A买B时', '买B时GD')).toBe('买B时');
expect(longestCommonSubstring('After test买时 case', 'another_test买时')).toBe('test买时');
});
});

View File

@ -1,23 +1,27 @@
/**
* @param {string} s1
* @param {string} s2
* @param {string} string1
* @param {string} string2
* @return {string}
*/
export default function longestCommonSubstring(s1, s2) {
// transform s1 & s2 into arrays to allow handling unicodes as single caracter.
const [a1, a2] = [s1, s2].map(s => Array.from(s));
export default function longestCommonSubstring(string1, string2) {
// Convert strings to arrays to treat unicode symbols length correctly.
// For example:
// '𐌵'.length === 2
// [...'𐌵'].length === 1
const s1 = [...string1];
const s2 = [...string2];
// Init the matrix of all substring lengths to use Dynamic Programming approach.
const substringMatrix = Array(a2.length + 1).fill(null).map(() => {
return Array(a1.length + 1).fill(null);
const substringMatrix = Array(s2.length + 1).fill(null).map(() => {
return Array(s1.length + 1).fill(null);
});
// Fill the first row and first column with zeros to provide initial values.
for (let columnIndex = 0; columnIndex <= a1.length; columnIndex += 1) {
for (let columnIndex = 0; columnIndex <= s1.length; columnIndex += 1) {
substringMatrix[0][columnIndex] = 0;
}
for (let rowIndex = 0; rowIndex <= a2.length; rowIndex += 1) {
for (let rowIndex = 0; rowIndex <= s2.length; rowIndex += 1) {
substringMatrix[rowIndex][0] = 0;
}
@ -26,9 +30,9 @@ export default function longestCommonSubstring(s1, s2) {
let longestSubstringColumn = 0;
let longestSubstringRow = 0;
for (let rowIndex = 1; rowIndex <= a2.length; rowIndex += 1) {
for (let columnIndex = 1; columnIndex <= a1.length; columnIndex += 1) {
if (a1[columnIndex - 1] === a2[rowIndex - 1]) {
for (let rowIndex = 1; rowIndex <= s2.length; rowIndex += 1) {
for (let columnIndex = 1; columnIndex <= s1.length; columnIndex += 1) {
if (s1[columnIndex - 1] === s2[rowIndex - 1]) {
substringMatrix[rowIndex][columnIndex] = substringMatrix[rowIndex - 1][columnIndex - 1] + 1;
} else {
substringMatrix[rowIndex][columnIndex] = 0;
@ -53,7 +57,7 @@ export default function longestCommonSubstring(s1, s2) {
let longestSubstring = '';
while (substringMatrix[longestSubstringRow][longestSubstringColumn] > 0) {
longestSubstring = a1[longestSubstringColumn - 1] + longestSubstring;
longestSubstring = s1[longestSubstringColumn - 1] + longestSubstring;
longestSubstringRow -= 1;
longestSubstringColumn -= 1;
}