Mustard

Enables substring matching using CharacterSet instances, useful for parsing mixed formats without separators, as demonstrated with 'hello2017year' splitting into letters and digits.

Implements the TokenizerType protocol for sophisticated matching like date formats or emoji detection, allowing domain-specific parsing logic without relying on rigid patterns.

Tokens include matched text, range, and additional context such as Date objects for date tokens, providing more insight than simple substring extraction.

Includes benchmarking and performance comparisons against alternatives like Scanner, ensuring efficient tokenization for demanding applications.