Name normalization is the general rewrite layer in a spelling file. It is not alias lookup: NameNormalizer applies section-local Prefix, Infix and Suffix rules directly to a mutable string and reports how many substitutions it made.
The rules are byte-exact and case-sensitive. Prefix and suffix normalization stop after the first matching rule in their respective lists; infix normalization replaces every occurrence of each loaded fragment.
The normalizer stores three ordered rule lists. Each rule is a quoted pair from the spelling file: source fragment and replacement fragment. The public addPrefix(), addInfix() and addSuffix() methods parse those quoted pairs and return a Core error code when the directive is malformed.