Version française
Home     About     Download     Resources     Contact us    
Browse thread
[Caml-list] Camlp4 questions
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: -- (:)
From: Harrison, John R <johnh@i...>
Subject: [Caml-list] Camlp4 questions
I've recently been porting HOL Light from CAML Light to OCaml. Since HOL
relies heavily on the use of uppercase value identifiers and infix
alphanumeric identifiers, I had to use Camlp4 to make some syntax
modifications. This works fine, and I was also able to fix up the lexer
so that quotations are accepted using `...` rather than <<...>>. (The
former is standard for HOL, and looks more natural to the Anglo-Saxon
eye, James Joyce excepted.) Nevertheless, my modifications were made by
fairly naive brutish hacking, and I wondered if there are more elegant
ways of doing things. In particular:

1. There is an example of adding new infixes in section 4.3.1 of the
   Camlp4 documentation. However following this example still left 
   two problems. First, the level of precedence was unexpectedly
   high, with "f x o g y" not parsing as "(f x) o (g y)" as expected.
   I changed this by using AFTER "<" instead of AFTER "apply", which
   seemed to work, but is that the recommended solution? Secondly,
   enclosing the new infix in parentheses to suppress its infix
   status, e.g. "(o)", doesn't work; it still expects an operator
   name. It was easy enough to hack the parser a bit more, but is
   there a nice solution using simple extensions?

2. Most of my changes are at the lexical level. Yet the only way I
   could figure out of changing the lexer was to rebuild the CAML
   parser in "camlp4/etc/pa_o.ml" from scratch with a new lexer and
   additional modifications. Is it possible to change the lexer but
   otherwise make only simple extensions? This is for the toplevel,
   by the way.

John.
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr