let regexp ident' = ((alpha | tex_token) ident_cont'*) | ('_' ident_cont'+)
let regexp paren = [ '(' '[' '{' ')' ']' '}' ]
let regexp meta = '?' num
+let regexp qstring = '"' [^ '"']* '"'
+let regexp uri =
+ (* schema *) (* path *) (* ext *) (* xpointer *)
+ ("cic:/" | "theory:/") ident ('/' ident)* ('.' ident)+ ('#' num ('/' num)*)?
(* let regexp catchall = .* *)
let keywords = Hashtbl.create 17
*)
let rec token = lexer
| blanks -> token lexbuf
+ | uri -> return lexbuf ("URI", Ulexing.utf8_lexeme lexbuf)
| ident ->
let lexeme = Ulexing.utf8_lexeme lexbuf in
(try
| num -> return lexbuf ("NUM", Ulexing.utf8_lexeme lexbuf)
| paren -> return lexbuf ("PAREN", Ulexing.utf8_lexeme lexbuf)
| meta -> return lexbuf ("META", Ulexing.utf8_lexeme lexbuf)
+ | qstring ->
+ let lexeme = Ulexing.utf8_lexeme lexbuf in
+ let s = String.sub lexeme 1 (String.length lexeme - 2) in
+ return lexbuf ("QSTRING", s)
| symbol -> return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf)
| tex_token ->
let macro =
| Ulexing.Error -> error_at_end lexbuf "Unexpected character"
| Ulexing.InvalidCodepoint i -> error_at_end lexbuf "Invalid code point")
-let lex =
+let cic_lexer =
{
Token.tok_func = tok_func;
Token.tok_using = (fun _ -> ());
Token.tok_text = Token.lexer_text;
Token.tok_comm = None;
}
+