On 4/27/06, Stephen Frost <sfrost@snowman.net> wrote:
> The answer to that can certainly be "performance" provided other factors
> (such as maintainability) don't change much. If you could show that
> then I think such a switch would be very seriously considered.
IMHO, switching parser-types (and parser generators) is more about
maintainability than performance itself. SQL is much nicer in
recursive descent where you don't have yacc/bison limitations such as
1 token of lookahead and non-ebnf grammars. The sort-of odd thing is
that PCCTS (like its much younger brother ANTLR) is intended on
generating ASTs whereas yacc/bison requires you to build the parse
tree manually (as we do).
Don't get me wrong, this was taken into consideration with PCCTS, but
it's not as optimal or beautiful as it would be to have PCCTS itself
generate the parse tree. Still, it's nicer to maintain than a
yacc/bison grammar.
--
Jonah H. Harris, Database Internals Architect
EnterpriseDB Corporation
732.331.1324