This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these template messages)(Learn how and when to remove this template message)
|Initial release||March 11, 2010|
2020-05-01 / May 1, 2020
|Type||Pattern matching library|
RE2 is a software library for regular expressions via a finite-state machine using automata theory, in contrast to almost all other regular expression libraries, which use backtracking implementations. It provides a C++ interface.
RE2 was implemented and is used by Google.
Comparison to PCRE
RE2 does not support back-references, which cannot be implemented efficiently. It is also slightly slower than PCRE for parenthetic capturing operations, but is much faster for matching in general.
PCRE can use a large recursive stack and have exponential runtime on certain patterns. RE2 uses a fixed stack and guarantees that run-time increases linearly (not exponentially) with the size of the input. The maximum memory allocated with RE2 can be configured if you have good knowledge of the workings of its code.
Google's RE2 has a slightly smaller set of features than PCRE, but has very predictable run-time and a maximum memory allotment, making it suitable for use in server applications which require precise boundaries on memory usage and computational time. PCRE, on the other hand, has almost all of the features that a regular expression library can have, but has unpredictable run-time and memory usage and can grow unbounded.
RE2 by Google is designed for the Linux operating system. It also runs on MS Windows using MSVC, MinGW, or Cygwin. It should also run on other operating systems that support CMake.
Use of RE2 in Google products
RegexExtract() does not use grouping.
- Cox, Russ (March 11, 2010). "RE2: a principled approach to regular expression matching". Google Open Source Blog. Retrieved 2020-05-29.
- "Releases". Github. Retrieved 2020-05-29.
- "Search and use find and replace". Retrieved 24 March 2020.