regularexpression

A regular expression, often abbreviated regex or regexp, is a sequence of characters that defines a search pattern. It is used to locate, match, or manipulate strings according to specific rules, and is implemented in many programming languages, text editors, and data processing tools.

A regex is built from literals and metacharacters. Literals match themselves, while metacharacters represent classes, repetitions, or structural ideas. Common elements include character classes like [abc] or [a-z], quantifiers such as *, +, ?, and {m,n}, grouping with parentheses, alternation with |, and anchors like ^ and $. The dot . matches any single character (except newline in many engines), and backslashes escape special meanings or introduce escapes like \d for digits or \w for word characters. Unicode and locale support vary by engine.

Typical operations include matching a pattern against text, extracting parts via capture groups, replacing matches, and splitting strings. Patterns can be anchored to enforce position in the string, and quantifiers can be greedy or non-greedy, affecting how much text is consumed. Backreferences allow later parts of a pattern to refer to earlier matches.

Regex engines differ in implementation and performance. Some use backtracking, which is flexible but can suffer from exponential time in complex patterns; others rely on finite automata, offering speed and predictability but sometimes limited expressiveness. Unicode support, escaping rules, and syntax variations vary across languages and tools.

Historically rooted in formal language theory by Kleene, regular expressions were popularized by Unix tools and later by languages such as Perl, Java, Python, JavaScript, and .NET. They remain a fundamental tool for validation, parsing, tokenization, and text transformation, though they require careful craftsmanship to avoid pitfalls like unintended greediness or excessive backtracking.