- clean no longer unregister URIs from the getter, but returns them

[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 2cb13cb33ade892562e006ca4292f030276e6b80..9534704edd466caeabd591cc9ed6ebf84ef8b7d2 100644 (file)
--- a/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
+++ b/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
@@ -30,22 +30,30 @@ exception Not_an_extended_ident
 let regexp alpha = [ 'a' - 'z' 'A' - 'Z' ]
 let regexp digit = [ '0' - '9' ]
 let regexp blank = [ ' ' '\t' '\n' ]
-
+let regexp paren = [ '(' '[' '{' ')' ']' '}' ]
+let regexp implicit = '?'
+let regexp symbol_char =
+  [^ 'a' - 'z' 'A' - 'Z' '0' - '9'
+     ' ' '\t' '\n'
+     '\\' '(' '[' '{' ')' ']' '}' '?'
+  ]
+let regexp comment_char = [^ '\n' ]
+let regexp comment = "%%" comment_char*
 let regexp blanks = blank+
 let regexp num = digit+
 let regexp tex_token = '\\' alpha+
-let regexp symbol = [^ 'a' - 'z' 'A' - 'Z' '0' - '9' ' ' '\t' '\n' ]
+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+)
 let regexp ident' = ((alpha | tex_token) ident_cont'*) | ('_' ident_cont'+)
-let regexp paren = [ '(' '[' '{' ')' ']' '}' ]
-let regexp implicit = '?'
-let regexp meta = '?' num
+let regexp meta = implicit num
 let regexp qstring = '"' [^ '"']* '"'
 let regexp uri =
-  (*      schema      *) (*     path     *) (*  ext   *) (*    xpointer     *)
-  ("cic:/" | "theory:/") ident ('/' ident)* ('.' ident)+ ('#' num ('/' num)*)?
+  ("cic:/" | "theory:/")              (* schema *)
+  ident ('/' ident)*                  (* path *)
+  ('.' ident)+                        (* ext *)
+  ("#xpointer(" num ('/' num)+ ")")?    (* xpointer *)
 (* let regexp catchall = .* *)
 
 let keywords = Hashtbl.create 17
@@ -123,6 +131,11 @@ let rec token = lexer
         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)
   | eof -> return lexbuf ("EOI", "")
   | _ -> error lexbuf "Invalid character"