Lexical Analysis deals with lexems and the length of an identifier is not defined in the programming languages mostly... i.e as such there is no constraint on the length of the identifiers so In the Lexical Analysis, regular expression can be used to model the structure of lexemes with variable length identifier included.
So option d