NFKD

NFKD stands for Normalization Form Compatibility Decomposition. It is one of the Unicode normalization forms defined to allow consistent comparison and processing of text. NFKD applies the compatibility decomposition mapping to each character in a string, replacing characters with their canonical components, and does not reassemble the result into composed characters. The effect is a fully decomposed sequence that reflects compatibility variants such as ligatures and certain presentation forms.

In contrast to NFD (Canonical Decomposition) and NFC (Canonical Decomposition followed by Canonical Composition), NFKD specifically targets compatibility mappings. These mappings convert characters that have compatibility equivalents, such as ligatures, ligature-based forms, and certain formatted characters, into more basic components. After decomposition, no recomposition into composed characters occurs, which can change how the text is stored or displayed but preserves a form suitable for robust equivalence checks across visually similar forms.

Common examples of NFKD decomposition include ligatures like U+FB01 (fi) decomposing to the sequence f, i, and U+FB00 (ff) decomposing to f, f. The compatibility znak for the ohm sign U+2126 may map to the Greek capital letter Omega U+03A9. NFKD is also used to help implement search, indexing, and normalization pipelines that require stripping formatting differences and compatibility variants.

Caveats include that NFKD can alter the semantic distinctions between characters and is not reversible. It is most appropriate when the goal is to compare text across visually similar but technically distinct forms rather than to preserve original presentation details.