X-Git-Url: http://matita.cs.unibo.it/gitweb/?a=blobdiff_plain;f=helm%2Focaml%2Fcic_disambiguation%2FcicTextualLexer2.ml;h=7606344d1b1feea9981fd442e1228084f48ff7b1;hb=12cc5b2b8e7f7bb0b5e315094b008a293a4df6b1;hp=5363d2180e72426f31bd329eeaf3f406da343db0;hpb=8004125685a99b6c0f2f95fd7f3fa09a4f5c9094;p=helm.git

diff --git a/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml b/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
index 5363d2180..7606344d1 100644
--- a/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
+++ b/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
@@ -30,31 +30,60 @@ 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 placeholder = '%'
+let regexp symbol_char =
+  [^ 'a' - 'z' 'A' - 'Z' '0' - '9'
+     ' ' '\t' '\n'
+     '\\' '(' '[' '{' ')' ']' '}' '?'
+  ]
+
+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 = [^ 'a' - 'z' 'A' - 'Z' '0' - '9' ' ' '\t' '\n' ]
-let regexp ident_cont = alpha | num | '_'
+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 lparen = [ '(' '[' '{' ]
-let regexp rparen = [ ')' ']' '}' ]
-let regexp meta = '?' num
+let regexp uri_step = (alpha (ident_cont|'-')*)
+let regexp meta = implicit num
+let regexp qstring = '"' [^ '"']* '"'
+let regexp uri_suffix = "con"|"ind"|"var"
+let regexp uri =
+  ("cic:/" | "theory:/")              (* schema *)
+  uri_step ('/' uri_step)*            (* path *)
+  ('.' uri_suffix)+                   (* ext *)
+  ("#xpointer(" num ('/' num)+ ")")?  (* xpointer *)
 (* let regexp catchall = .* *)
 
 let keywords = Hashtbl.create 17
 let _ =
   List.iter (fun keyword -> Hashtbl.add keywords keyword ("", keyword))
-    [ "Prop"; "Type"; "Set"; "let"; "rec"; "using"; "match"; "with" ]
+    [ "Prop"; "Type"; "Set"; "let"; "Let"; "rec"; "using"; "match"; "with";
+    "in"; "and"; "to"; "as"; "on"; "names"]
 
 let error lexbuf msg =
   raise (Error (Ulexing.lexeme_start lexbuf, Ulexing.lexeme_end lexbuf, msg))
 let error_at_end lexbuf msg =
   raise (Error (Ulexing.lexeme_end lexbuf, Ulexing.lexeme_end lexbuf, msg))
 
-let return lexbuf token = (token, Ulexing.loc lexbuf)
+let return lexbuf token =
+  let flocation_begin =
+    { Lexing.pos_fname = ""; Lexing.pos_lnum = -1; Lexing.pos_bol = -1;
+      Lexing.pos_cnum = Ulexing.lexeme_start lexbuf }
+  in
+  let flocation_end =
+    { flocation_begin with Lexing.pos_cnum = Ulexing.lexeme_end lexbuf }
+  in
+  (token, (flocation_begin, flocation_end))
 
 (*
 let parse_ext_ident ident =
@@ -85,45 +114,63 @@ let rec token' = lexer
 
 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
       (try
         return lexbuf (Hashtbl.find keywords lexeme)
       with Not_found -> return lexbuf ("IDENT", lexeme))
   | num -> return lexbuf ("NUM", Ulexing.utf8_lexeme lexbuf)
-  | lparen -> return lexbuf ("LPAREN", Ulexing.utf8_lexeme lexbuf)
-  | rparen -> return lexbuf ("RPAREN", Ulexing.utf8_lexeme lexbuf)
+  | 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)
   | tex_token ->
       let macro =
         Ulexing.utf8_sub_lexeme lexbuf 1 (Ulexing.lexeme_length lexbuf - 1)
       in
       (try
-        return lexbuf ("SYMBOL", Macro.expand macro)
-      with Macro.Macro_not_found _ ->
+        return lexbuf ("SYMBOL", Utf8Macro.expand macro)
+      with Utf8Macro.Macro_not_found _ ->
         return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf))
+  | 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_location
+  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 lex =
-  { 
-    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;
     Token.tok_text = Token.lexer_text;
     Token.tok_comm = None;
   }
+