google/re2
Re2
re2 is a C++ regular expression library designed for high-performance text processing. It is a non-backtracking regex engine that provides linear-time pattern matching, ensuring that execution time remains proportional to the size of the input string regardless of the pattern used.
The library supports UTF-8 and Latin-1 text encodings for searching and extracting substrings. It includes capabilities for multi-pattern optimization, allowing multiple regular expressions to be combined into a single representation to scan text for several patterns in one pass.
The project covers core regex operations including pattern compilation, full string validation, and the extraction of specific data through numbered or named grouping expressions.
Features
- Linear-Time Regex Matchers - Prevents catastrophic backtracking by processing each input character a constant number of times regardless of the pattern.
- Text Processing - Provides optimized searching and substring extraction for text encoded in UTF-8 and Latin-1.
- High-Performance Text Processing - Ensures predictable execution time and memory usage when processing large volumes of text with regular expressions.
- Capture Group Implementation - Supports capturing specific substrings from text using named or numbered regular expression grouping expressions.
- C++ Regular Expression Libraries - Provides a fast, thread-safe C++ library for matching text patterns without exponential time complexity.
- Finite Automata Regex Engines - Uses deterministic finite automata (DFA) to ensure that text matching time is linear relative to the input size.
- Regular Expression Engines - Evaluates patterns against strings to identify or extract specific text sequences using a high-performance regex engine.
- Text Search - Enables matching a single piece of text against multiple regular expression patterns simultaneously.
- Regex - Implements an optimized internal bytecode representation for regular expression patterns to minimize repeated search overhead.
- NFA-Based Matching - Employs non-deterministic finite automata (NFA) simulation to handle complex patterns while avoiding exponential backtracking.
- Stateless Thread-Safe Matching - Separates compiled regex objects from execution state to allow multiple threads to perform matches simultaneously.
- Multi-Pattern Matching Algorithms - Combines multiple regular expressions into a single shared state machine to scan text for several patterns in one pass.
- Regex Encoding Support - Interprets patterns and input text as UTF-8 or Latin-1 to ensure accurate character matching across different encodings.
- Regular Expressions - Safe, finite-state machine based library for efficient regex execution.
