Regular Expression Matching
Last Updated : 21 Sep, 2024
Given a text t and a pattern p where t consists of only lowercase English alphabets while p consists of lowercase English alphabets as well as special characters '.' and '*', the task is to implement a function to test regular expression such that:
'.' Matches any single character.'*' Matches zero or more of the preceding character.
Note: For each appearance of the character '*', there will be a previous valid character to match.
Examples:
Input: t = "aaa", p = "a"
Output: false
Explanation: "a" does not match the entire string "aaa".
Input: t = "abb", p = "a.*"
Output: true
Explanation: replace . with b then p becomes ab* now replace * with one preceeding character hence p becomes abb.
Input: t = "", p = "a*b*"
Output: true
Explanation: Note that * can match 0 occurrences also.
Naive Recursive Solution:
We can begin matching both pattern from one side. We begin from the right side. Following cases arise, when we match character by character.
Case 1 (If Last Characters are Same) : We move to the next character in both text t and pattern p
Case 2 (If Last Character of Patter is '.') : We move to the next character in both text t and pattern p
Case 2 (If Last Character of Patter is '*') : There must be at-least two characters in the pattern. If not, we return false. If yes, the following two cases arise.
a) The '*' and character preceding it match with 0 characters of the text. We move two characters in pattern and do not move in text.
b) The '*' and character preceding it match with 1 or more characters of the text. We match the preceding character of pattern with the current character of text. If match, then we move one character ahead in text and do not move in pattern.
We return false if both a) and b) are not true.
Below is implementation of the idea.
C++ #include <iostream> using namespace std; bool isMatchRec(string t, string p, int n, int m) { // If pattern is empty, then text must also be // empty if (m == 0) { return n == 0; } // If text is empty, then pattern can have characters // followed by *s if (n == 0) { return (m >= 2 && p[m - 1] == '*') && isMatchRec(t, p, n, m - 2); } // If last characters of both string and pattern // match, or pattern has '.' if (t[n - 1] == p[m - 1] || p[m - 1] == '.') { return isMatchRec(t, p, n - 1, m - 1); } // Handle '*' in the pattern if (p[m - 1] == '*' && m > 1) { // Check if '*' can represent zero occurrences // of the preceding character bool zero = isMatchRec(t, p, n, m - 2); // Check if '*' can represent one or more occurrences // of the preceding character bool oneOrMore = (p[m - 2] == t[n - 1] || p[m - 2] == '.') && isMatchRec(t, p, n - 1, m); return zero || oneOrMore; } // If no match return false; } // Wrapper function bool isMatch(string t, string p) { return isMatchRec(t, p, t.size(), p.size()); } int main() { cout << boolalpha << isMatch("aab", "a.*") << endl; cout << boolalpha << isMatch("aa", "a") << endl; cout << boolalpha << isMatch("aa", "a*") << endl; cout << boolalpha << isMatch("ab", ".*") << endl; cout << boolalpha << isMatch("mississippi", "mis*is*p*.") << endl; return 0; } public class GfG { public static boolean isMatchRec(String t, String p, int n, int m) { // If pattern is empty, then text must also be // empty if (m == 0) { return n == 0; } // If text is empty, then pattern can have characters // followed by *s if (n == 0) { return (m >= 2 && p.charAt(m - 1) == '*') && isMatchRec(t, p, n, m - 2); } // If last characters of both string and pattern // match, or pattern has '.' if (t.charAt(n - 1) == p.charAt(m - 1) || p.charAt(m - 1) == '.') { return isMatchRec(t, p, n - 1, m - 1); } // Handle '*' in the pattern if (p.charAt(m - 1) == '*' && m > 1) { // Check if '*' can represent zero occurrences // of the preceding character boolean zero = isMatchRec(t, p, n, m - 2); // Check if '*' can represent one or more occurrences // of the preceding character boolean oneOrMore = (p.charAt(m - 2) == t.charAt(n - 1) || p.charAt(m - 2) == '.') && isMatchRec(t, p, n - 1, m); return zero || oneOrMore; } // If no match return false; } // Wrapper function public static boolean isMatch(String t, String p) { return isMatchRec(t, p, t.length(), p.length()); } public static void main(String[] args) { System.out.println(isMatch("aab", "a.*")); System.out.println(isMatch("aa", "a")); System.out.println(isMatch("aa", "a*")); System.out.println(isMatch("ab", ".*"); System.out.println(isMatch("mississippi", "mis*is*p*.")); } } def is_match_rec(t, p, n, m): # If pattern is empty, then text must also be empty if m == 0: return n == 0 # If text is empty, then pattern can have characters followed by *s if n == 0: return (m >= 2 and p[m - 1] == '*') and is_match_rec(t, p, n, m - 2) # If last characters of both string and pattern match, or pattern has '.' if t[n - 1] == p[m - 1] or p[m - 1] == '.': return is_match_rec(t, p, n - 1, m - 1) # Handle '*' in the pattern if p[m - 1] == '*' and m > 1: # Check if '*' can represent zero occurrences of the preceding character zero = is_match_rec(t, p, n, m - 2) # Check if '*' can represent one or more occurrences of the preceding character one_or_more = (p[m - 2] == t[n - 1] or p[m - 2] == '.') and is_match_rec(t, p, n - 1, m) return zero or one_or_more # If no match return False # Wrapper function def is_match(t, p): return is_match_rec(t, p, len(t), len(p)) # Example usage print(is_match('aab', 'a.*')) print(is_match('aa', 'a')) print(is_match('aa', 'a*')) print(is_match('ab', '.*')) print(is_match('mississippi', 'mis*is*p*.')) function is_match_rec(t, p, n, m) { // If pattern is empty, then text must also be empty if (m === 0) { return n === 0; } // If text is empty, then pattern can have characters followed by *s if (n === 0) { return (m >= 2 && p[m - 1] === '*') && is_match_rec(t, p, n, m - 2); } // If last characters of both string and pattern match, or pattern has '.' if (t[n - 1] === p[m - 1] || p[m - 1] === '.') { return is_match_rec(t, p, n - 1, m - 1); } // Handle '*' in the pattern if (p[m - 1] === '*' && m > 1) { // Check if '*' can represent zero occurrences of the preceding character const zero = is_match_rec(t, p, n, m - 2); // Check if '*' can represent one or more occurrences of the preceding character const one_or_more = (p[m - 2] === t[n - 1] || p[m - 2] === '.') && is_match_rec(t, p, n - 1, m); return zero || one_or_more; } // If no match return false; } // Wrapper function function is_match(t, p) { return is_match_rec(t, p, t.length, p.length); } // Example usage console.log(is_match('aab', 'a.*')); console.log(is_match('aa', 'a')); console.log(is_match('aa', 'a*')); console.log(is_match('ab', '.*')); console.log(is_match('mississippi', 'mis*is*p*.')); Outputtrue false true true false
Dynamic Programming Solution
The above recursive solution has exponential time complexity in the worst case. Please note that we make two recursive calls in the last if condition. We can clearly notice overlapping subproblems here as we make calls for (n-1, m-1), (n, m-2) and/or (n-1, m). So we can use Dynamic Programming to solve this problem.
- Create a boolean 2D
dp array of size (n + 1) * (m + 1). Please note that the range of values in the recursion goes from 0 to text length (or n) and 0 to pattern length (or m) - dp[i][j] is going to be true if first i characters of text match with first j characters of pattern.
- If both strings are empty, then it’s a match, thus,
dp[0][0] = true. - For other cases, we simply follow the above recursive solution.
C++ #include <iostream> #include <vector> using namespace std; bool isMatch(string t, string p) { int n = t.size(); int m = p.size(); // DP table where dp[i][j] means whether first i characters in t // match the first j characters in p vector<vector<bool>> dp(n + 1, vector<bool>(m + 1, false)); // Empty pattern matches empty text dp[0][0] = true; // Deals with patterns like a*, a*b*, a*b*c* etc, where '*' // can eliminate preceding character for (int j = 1; j <= m; ++j) { if (p[j - 1] == '*' && j > 1) { dp[0][j] = dp[0][j - 2]; } } // Fill the table for (int i = 1; i <= n; ++i) { for (int j = 1; j <= m; ++j) { // Characters match if (p[j - 1] == '.' || t[i - 1] == p[j - 1]) { dp[i][j] = dp[i - 1][j - 1]; } else if (p[j - 1] == '*' && j > 1) { // Two cases: // 1. '*' represents zero occurrence of the preceding character // 2. '*' represents one or more occurrence of the preceding character dp[i][j] = dp[i][j - 2] || (dp[i - 1][j] && (p[j - 2] == t[i - 1] || p[j - 2] == '.')); } } } return dp[n][m]; } int main() { cout << boolalpha << isMatch("aab", "a.*") << endl; cout << boolalpha << isMatch("aa", "a") << endl; cout << boolalpha << isMatch("aa", "a*") << endl; cout << boolalpha << isMatch("ab", ".*") << endl; cout << boolalpha << isMatch("mississippi", "mis*is*p*.") << endl; return 0; } import java.util.*; public class GfG { public static boolean isMatch(String t, String p) { int n = t.length(); int m = p.length(); // DP table where dp[i][j] means whether first i characters in t // match the first j characters in p boolean[][] dp = new boolean[n + 1][m + 1]; // Empty pattern matches empty text dp[0][0] = true; // Deals with patterns like a*, a*b*, a*b*c* etc, where '*' // can eliminate the preceding character for (int j = 1; j <= m; ++j) { if (p.charAt(j - 1) == '*' && j > 1) { dp[0][j] = dp[0][j - 2]; } } // Fill the table for (int i = 1; i <= n; ++i) { for (int j = 1; j <= m; ++j) { // Characters match if (p.charAt(j - 1) == '.' || t.charAt(i - 1) == p.charAt(j - 1)) { dp[i][j] = dp[i - 1][j - 1]; } // Handle '*' in the pattern else if (p.charAt(j - 1) == '*' && j > 1) { // Two cases: // 1. '*' represents zero occurrence of the preceding character // 2. '*' represents one or more occurrence of the preceding character dp[i][j] = dp[i][j - 2] || (dp[i - 1][j] && (p.charAt(j - 2) == t.charAt(i - 1) || p.charAt(j - 2) == '.')); } } } return dp[n][m]; } public static void main(String[] args) { System.out.println(isMatch("aab", "a.*")); System.out.println(isMatch("aa", "a")); System.out.println(isMatch("aa", "a*")); System.out.println(isMatch("ab", ".*")); System.out.println(isMatch("mississippi", "mis*is*p*.")); } } def isMatch(t: str, p: str) -> bool: n = len(t) m = len(p) # DP table where dp[i][j] means whether first i characters in t # match the first j characters in p dp = [[False] * (m + 1) for _ in range(n + 1)] # Empty pattern matches empty text dp[0][0] = True # Deals with patterns like a*, a*b*, a*b*c* etc, where '*' # can eliminate the preceding character for j in range(1, m + 1): if p[j - 1] == '*' and j > 1: dp[0][j] = dp[0][j - 2] # Fill the table for i in range(1, n + 1): for j in range(1, m + 1): # Characters match if p[j - 1] == '.' or t[i - 1] == p[j - 1]: dp[i][j] = dp[i - 1][j - 1] # Handle '*' in the pattern elif p[j - 1] == '*' and j > 1: # Two cases: # 1. '*' represents zero occurrence of the preceding character # 2. '*' represents one or more occurrence of the preceding character dp[i][j] = dp[i][j - 2] or (dp[i - 1][j] and (p[j - 2] == t[i - 1] or p[j - 2] == '.')) return dp[n][m] if __name__ == "__main__": print(isMatch("aab", "a.*")) print(isMatch("aa", "a")) print(isMatch("aa", "a*")) print(isMatch("ab", ".*")) print(isMatch("mississippi", "mis*is*p*.")) function isMatch(t, p) { const n = t.length; const m = p.length; // DP table where dp[i][j] means whether first i characters in t // match the first j characters in p const dp = Array.from({ length: n + 1 }, () => Array(m + 1).fill(false)); // Empty pattern matches empty text dp[0][0] = true; // Deals with patterns like a*, a*b*, a*b*c* etc, where '*' // can eliminate the preceding character for (let j = 1; j <= m; ++j) { if (p[j - 1] === '*' && j > 1) { dp[0][j] = dp[0][j - 2]; } } // Fill the table for (let i = 1; i <= n; ++i) { for (let j = 1; j <= m; ++j) { // Characters match if (p[j - 1] === '.' || t[i - 1] === p[j - 1]) { dp[i][j] = dp[i - 1][j - 1]; } // Handle '*' in the pattern else if (p[j - 1] === '*' && j > 1) { // Two cases: // 1. '*' represents zero occurrence of the preceding character // 2. '*' represents one or more occurrence of the preceding character dp[i][j] = dp[i][j - 2] || (dp[i - 1][j] && (p[j - 2] === t[i - 1] || p[j - 2] === '.')); } } } return dp[n][m]; } // Testing the function console.log(isMatch("aab", "a.*")); // true console.log(isMatch("aa", "a")); // false console.log(isMatch("aa", "a*")); // true console.log(isMatch("ab", ".*")); // true console.log(isMatch("mississippi", "mis*is*p*.")); // false Outputtrue false true true false
Illustration
Let’s take an example t = "aab" and p = "c*a*b" and create a DP table.
| | | c | * | a | * | b |
|---|
| | 0 | 1 | 2 | 3 | 4 | 5 |
| 0 | TRUE | FALSE | TRUE | FALSE | TRUE | FALSE |
a | 1 | FALSE | FALSE | FALSE | TRUE | TRUE | FALSE |
a | 2 | FALSE | FALSE | FALSE | FALSE | TRUE | FALSE |
b | 3 | FALSE | FALSE | FALSE | FALSE | FALSE | TRUE |
- First column — it means
p is empty and it will match to s only if s is also empty which we have stored in dp[0][0]. Thus, remaining values of dp[0][i] will be false. - First row — this is not so easy. It means which
p matches empty t. The answer is either an empty pattern or a pattern that represents an empty string such as "a*", "x*y*", "l*m*n*" and so on. In the above example, if t = "" and p = "c*", then due to *, c can be replaced by 0 cs which gives us an empty string. Hence, dp[0][2] = true. - For non-empty strings, let’s say that t
[i - 1] == p[j - 1] this means the (i - 1)th and (j - 1)th characters are same. This means, we have to check if the remaining strings are a match or not. If they are a match, then the current substrings will be a match, otherwise they won’t be a match i.e., dp[i][j] = dp[i - 1][j - 1]. We’re taking (i - 1)th and (j - 1)th characters to offset empty strings as we’re assuming our strings start from index 1. - If
p[j - 1] == ".", then it means any single character can be matched. Therefore, here also, we will have to check if the remaining string is a match or not. Thus, dp[i][j] = dp[i - 1][j - 1]. - If
p[j - 1] == "*", then it means either it’s represents an empty string (0 characters), thus dp[i][j] = dp[i][j - 2] or t[i - 1] == p[j - 2] || p[j - 2] == ".", then current character of string equals the char preceding '*' in pattern so the result is dp[i-1][j].
Time Complexity: O(m×n)
Auxiliary Space: O(m×n)
Similar Reads
Regular Expressions In R Regular expressions (regex) are powerful tools used in programming languages like R for pattern matching within text data. They enable us to search for specific patterns, extract information, and manipulate strings efficiently. Here, we'll explore the fundamentals of regular expressions in R Program
5 min read
Regular Expressions in Scala Regular Expressions explain a common pattern utilized to match a series of input data so, it is helpful in Pattern Matching in numerous programming languages. In Scala Regular Expressions are generally termed as Scala Regex. Regex is a class which is imported from the package scala.util.matching.Reg
5 min read
Properties of Regular Expressions Regular expressions, often called regex or regexp, are a powerful tool used to search, match, and manipulate text. They are essentially patterns made up of characters and symbols that allow you to define a search pattern for text. In this article, we will see the basic properties of regular expressi
7 min read
Regular Expression to DFA The main function of regular expressions is to define patterns for matching strings; automata theory provides a structured pattern recognition of these patterns through Finite Automata. A very common method to construct a Deterministic Finite Automaton (DFA) based on any given regular expression is
6 min read
std::regex_match, std::regex_replace() | Regex (Regular Expression) In C++ Regex is the short form for “Regular expressionâ€, which is often used in this way in programming languages and many different libraries. It is supported in C++11 onward compilers.Function Templates used in regex regex_match() -This function return true if the regular expression is a match against th
3 min read