Like most parsing systems, GOLD uses the lalr algorithm to analyze syntax and a deterministic finite automotion to identify different lexical units. However, GOLD takes a different approach than common compiler-compilers.
The LALR and DFA algorithms are simple automatas - using lookup tables to determine both state transition and parsing actions. As a result, most of the computational work is performed before any text is parsed. It is the computation of these tables that is both time-consuming and complex. Once created, these parsing tables are essentially independent of any programming language. Data is simply data.
No comments:
Post a Comment