When implementing many Unicode algorithms such as text segmentation, normalization, bidi processing, etc., fast access to character metadata is crucial to good performance. There over a million code ...