Unicode NFD normalization and casefolding

This directory contains two libraries:

generator consumes Unicode's official UCD dataset and produces unicode_gen.rs. As of June 2024, this is around 25kiBi for UCD version 12.1.0.

unicode makes use of this data set to provide :

• NFD normalization
• Case folding
• Default ignorable character mappings.
• A convenience comparator casefold_cmp that combines the above three.

Updating unicode versions

See the update script in the scripts/ subdirectory. Updates to CIPD prebuilts will trigger regeneration of unicode_gen.rs.