On Fri, 2005-08-26 at 19:05 +0200, Alain Frisch wrote: > For what concerns John's question about the integration of OCamlDuce in > OCaml, there are many answers. .. and time to explore them. I wasn't trying to suggest the merger should be done tomorrow (hence Ocaml 4.0 number ..) > 3) OCaml is a general purpose language, and the extension adds support for > a specific domain Tree pattern matching isn't all that specific is it? I mean one could argue regular patterns are a 'specific domain' .. but one would be missing the fact the finite state automata are in the basis of computing. Ocaml can't even do regular matching, how on earth you could call it a 'pattern matching language' I don't know. Yet CDuce is already extending patterns well beyond mere regular matches by providing fairly strong capturing ability. > constraint of being able to compile any existing OCaml program with > OCamlDuce, for instance, resulted in the introduction of explicit > delimiters {{..}} for all the new constructions, which is syntactically > heavy meaning {{ UGLY }} .. yes, it is ghastly. > and theoretically useless. 5) It's too early to say whether > OCamlDuce is useful or not compared to a simpler solution with two > compilers (OCaml, CDuce). Agree. However, whenever you have two tools you have to join them together somehow. One way of doing that is with Unix scripting, shell scripts and the like. This is very ugly and non-portable, but can usually be got to work. For example, using ocamllex/yacc is just horrible in practice, in Felix compiler there are so many files just to glue them together .. .. in Felix language itself both lexer and parser tools are built in to the language and the compiler does the gluing. Camlp4 provides some interesting ways to glue things in a much more structured way than Unix script. -- John Skaller