Re: [charmod-norm] Normalisation for Case-Insensitive Comparison

Thank you, Richard57.  I've been struggling trying to understand the "right" way to do this.  The Unicode Character Database's Case-Folding.txt file clearly states at the top:

# NOTE: case folding does not preserve normalization formats!

So the 3.1 Matching Algorithm which puts normalization first seems strange, because after case folding, the string may no longer be normalized, and this is clearly wrong for a string matching algorithm.

However, I haven't found a reference that normalization prior to case folding is or is not required for consist case folding results.  Do you have a reference for U+0345 being the only exception to requiring normalization before case folding?

-- 
GitHub Notification of comment by EricSharkey
Please view or discuss this issue at https://github.com/w3c/charmod-norm/issues/172#issuecomment-395839646 using your GitHub account

Received on Friday, 8 June 2018 17:58:10 UTC