Two-level grammar

From Wikipedia, the free encyclopedia
Jump to: navigation, search

A two-level grammar is a formal grammar that is used to generate another formal grammar [1], such as one with an infinite rule set [2]. This is how a Van Wijngaarden grammar was used to specify Algol68 [3]. A context free grammar that defines the rules for a second grammar can yield an effectively infinite set of rules for the derived grammar. This makes such two-level grammars more powerful than a single layer of context free grammar, because generative two-level grammars have actually been shown to be Turing complete.[1]

Two-level grammar can also refer to a formal grammar for a two-level formal language, which is a formal language specified at two levels, for example, the levels of words and sentences.[citation needed]

Example[edit]

A well-known non-context-free language is

\{a^n b^n a^n | n \ge 1\}.

A two-level grammar for this language is the metagrammar

N ::= 1 | N1
X ::= a | b

together with grammar schema

Start ::=  \langle a^N \rangle\langle b^N \rangle\langle a^N \rangle
 \langle X^{N1} \rangle ::= \langle X^N \rangle X
 \langle X^1 \rangle ::= X

See also[edit]

References[edit]

  1. ^ Sintzoff, M. "Existence of van Wijngaarden syntax for every recursively enumerable set", Annales de la Société Scientifique de Bruxelles 2 (1967), 115-118.

External links[edit]

  • Petersson, Kent (1990), "Syntax and Semantics of Programming Languages", Draft Lecture Notes, PDF text.