This fork was created to support Unicode characters, based on the fork which was created to prove that StringScore provides reasonable performance. The question about it was raised on the Stackoverflow question Objective-c: Fast Fuzzy Search Match.

Mainly, the setup code is pulled out of the scoreAgainst: method, thus 3 new methods are added to NSString+StringScore:

- (NSCharacterSet *)invalidCharacterSet;
- (NSString *)decomposedStringWithInvalidCharacterSet:(NSCharacterSet *)invalidCharacterSet;

- (CGFloat) scoreAgainst:(NSString *)anotherString fuzziness:(NSNumber *)fuzziness options:(NSStringScoreOption)options
     invalidCharacterSet:(NSCharacterSet *)invalidCharacterSet decomposedString:(NSString *)string;

The first two can be used to generate the data that is needed for the scoring algorithm. It will speed up the operation if scoreAgainst: is called repeatedly on the same string.

The other changes focus on speeding up the algorithm itself by generating fewer NSString objects.

Author: Nicholas Bruning

Special thanks to Joshaven Potter for providing the basis for this library.

Licensed under the MIT license.