This document details the high-level design of the lexer.
The core of the lexer is the XmlChar class. This is a UTF-32 codepoint value
class modelling the XML Char symbol.
The XmlCharReader is a CharSequence to UTF-32 reader from a given start and
end offset that keeps track of the current offset.
The CharacterClass interface matches a single Char in a given collection of
values. This is designed to model regular expression character classes. This
allows c in NameChar style code checks for these character classes and for them
to be used in Kotlin when statements.
A set of ASCII character constants are also defined (Colon, EqualsSign,
LeftSquareBracket, etc.) to make it easier to refer to those codepoints in
the codebase.