Symbols within the "Derived Property: Full_Composition_Exclusion" section in https://www.unicode.org/Public/UNIDATA/DerivedNormalizationProps.txt are not the best choice for detecting WorstFit ANSI transformations.
Unicode normalization legitimately and deliberately transforms the Kelvin compatibility code point (within the range described above) to a capital K. Whilst this may not always be desired, it is not an indicator of WorstFit ANSI transformation.
I suggest transformation of code points with the "Derived Property: Full_Composition_Exclusion" are not highlighted as an issue, or are at least categorized differently.
(Plenty of Unicode "confusables" can be found at https://util.unicode.org/UnicodeJsps/confusables.jsp)
Symbols within the "Derived Property: Full_Composition_Exclusion" section in https://www.unicode.org/Public/UNIDATA/DerivedNormalizationProps.txt are not the best choice for detecting WorstFit ANSI transformations.
Unicode normalization legitimately and deliberately transforms the Kelvin compatibility code point (within the range described above) to a capital K. Whilst this may not always be desired, it is not an indicator of WorstFit ANSI transformation.
I suggest transformation of code points with the "Derived Property: Full_Composition_Exclusion" are not highlighted as an issue, or are at least categorized differently.
(Plenty of Unicode "confusables" can be found at https://util.unicode.org/UnicodeJsps/confusables.jsp)