fix

[helm.git] / helm / ocaml / cic_disambiguation / cicTextualLexer2.ml

diff --git a/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml b/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
index 4cf0e4270a2fe29e53fdacac1aa017f83e055799..5127f21383b222e6bac568c79f73e777f6282582 100644 (file)
--- a/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
+++ b/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
@@ -32,20 +32,23 @@ let regexp digit = [ '0' - '9' ]
 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 dust = "%%" [^ '\n']* '\n'
-let regexp comment_char = [^ "*)" ]
-let regexp note = "(*" comment_char* "*)"
-let regexp commentbegin = "(**"
-let regexp commentend = "**)"
+
+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+)
@@ -124,6 +127,7 @@ let rec token comments = lexer
   | 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 ->
       return lexbuf ("QSTRING", remove_quotes (Ulexing.utf8_lexeme lexbuf))
   | symbol -> return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf)
@@ -135,11 +139,10 @@ let rec token comments = lexer
         return lexbuf ("SYMBOL", Utf8Macro.expand macro)
       with Utf8Macro.Macro_not_found _ ->
         return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf))
-  | dust -> token comments lexbuf
   | note -> 
       (*if comments then*)
         let comment =
-          Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 2)
+          Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 4)
         in
         return lexbuf ("NOTE", comment)
       (*else