regEZ is a modern, header only implementation of a general regex engine in C++23. It works with any container with custom tokens, fully constexpr compatible!

Yes, this does not work only for strings, but for any container. This means what you could make some efficient pattern matching and queries on exotic datasets.

This is an example with the usual string regex syntx we all know and love:

// Define the regex vocabulary
auto vocab = regez::Vocabulary()
    .set(regez::regez_open_group, '(')
    .set(regez::regez_close_group, ')')
    .set(regez::regez_open_match, '[')
    .set(regez::regez_close_match, ']')
    .set(regez::regez_or, '|')
    .set(regez::regez_any, '*')
    .set(regez::regez_concat, '.')
    .set(regez::regez_escape, '\\');

// Create a regex object
regez::Regex<std::string> r(std::string("(a|b)*c"), vocab);
// Match the regex to a string
bool is_matched = regez.match(std::string("Nice"));

The regex will allow the following tokens:

• regez_concat: concatenation operator

• regez_or: "objectA OR objectB" matches either "objectA" or "objectB"

• regez_any: "objectA ANY" matches any number of occurences of "objectA", even 0

• regez_open_group: opens a new group, eg. "("

• regez_close_group: closes a group, eg. ")"

• regez_one_or_more: "+"

• regez_escape: escape any of the previous tokens

The library is currently in developement and It's not intended for production use. If you want to contribute, check out the issues!

TODO

To build the tests, run cmake with REGEZ_BUILD_TESTS=ON option.

cmake -B build -D REGEZ_BUILD_TESTS=ON
cmake --build build

The project falls under MIT license.