let regexp blank = [ ' ' '\t' '\n' ]
let regexp paren = [ '(' '[' '{' ')' ']' '}' ]
let regexp implicit = '?'
+let regexp placeholder = '%'
let regexp symbol_char =
[^ 'a' - 'z' 'A' - 'Z' '0' - '9'
' ' '\t' '\n'
'\\' '(' '[' '{' ')' ']' '}' '?'
]
-let regexp comment_char = [^ '\n' ]
-let regexp comment = "%%" comment_char*
+
+let regexp comment_char = [^'*'] | '*'[^')']
+let regexp note = "(*" ([^'*'] | "**") comment_char* "*)"
+
+let regexp commentbegin = "(**" blank
+let regexp commentend = "*)"
+
let regexp blanks = blank+
let regexp num = digit+
let regexp tex_token = '\\' alpha+
-let regexp symbol = symbol_char+
+let regexp symbol = symbol_char
let regexp ident_cont = alpha | num | '_' | '\''
let regexp ident_cont' = ident_cont | tex_token
let regexp ident = (alpha ident_cont*) | ('_' ident_cont+)
and token = lexer
*)
-let rec token = lexer
- | blanks -> token lexbuf
+
+let remove_quotes s = String.sub s 1 (String.length s - 2)
+
+let rec token comments = lexer
+ | blanks -> token comments lexbuf
| uri -> return lexbuf ("URI", Ulexing.utf8_lexeme lexbuf)
| ident ->
let lexeme = Ulexing.utf8_lexeme lexbuf in
| paren -> return lexbuf ("PAREN", Ulexing.utf8_lexeme lexbuf)
| meta -> return lexbuf ("META", Ulexing.utf8_lexeme lexbuf)
| implicit -> return lexbuf ("IMPLICIT", Ulexing.utf8_lexeme lexbuf)
+ | placeholder -> return lexbuf ("PLACEHOLDER", 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)
+ return lexbuf ("QSTRING", remove_quotes (Ulexing.utf8_lexeme lexbuf))
| symbol -> return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf)
| tex_token ->
let macro =
return lexbuf ("SYMBOL", Utf8Macro.expand macro)
with Utf8Macro.Macro_not_found _ ->
return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf))
- | comment ->
- let comment =
- Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 2)
- in
- return lexbuf ("COMMENT", comment)
+ | note ->
+ (*if comments then*)
+ let comment =
+ Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 4)
+ in
+ return lexbuf ("NOTE", comment)
+ (*else
+ token comments lexbuf*)
+ | commentbegin -> return lexbuf ("BEGINCOMMENT","")
+ | commentend -> return lexbuf ("ENDCOMMENT","")
| eof -> return lexbuf ("EOI", "")
| _ -> error lexbuf "Invalid character"
-let tok_func stream =
+let tok_func comments stream =
let lexbuf = Ulexing.from_utf8_stream stream in
Token.make_stream_and_flocation
(fun () ->
try
- token lexbuf
+ token comments lexbuf
with
| Ulexing.Error -> error_at_end lexbuf "Unexpected character"
| Ulexing.InvalidCodepoint i -> error_at_end lexbuf "Invalid code point")
-let cic_lexer =
- {
- Token.tok_func = tok_func;
+let cic_lexer ?(comments = false) () =
+ {
+ Token.tok_func = tok_func comments;
Token.tok_using = (fun _ -> ());
Token.tok_removing = (fun _ -> ());
Token.tok_match = Token.default_match;