- changed license to lgpl

[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 21f0205203aea6e410b54d283a8ed60f1a92bf1f..6f179505d749c8782e7a83b2ff58edc1bde3b964 100644 (file)
--- a/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
+++ b/helm/ocaml/cic_disambiguation/cicTextualLexer2.ml
@@ -30,30 +30,52 @@ 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 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 paren = [ '(' '[' '{' ')' ']' '}' ]
-let regexp meta = '?' num
+let regexp meta = implicit num
+let regexp qstring = '"' [^ '"']* '"'
+let regexp uri =
+  ("cic:/" | "theory:/")              (* schema *)
+  ident ('/' ident)*                  (* path *)
+  ('.' ident)+                        (* 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" ]
 
 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 =
@@ -84,8 +106,9 @@ let rec token' = lexer
 
 and token = lexer
 *)
-let rec token = lexer
-  | blanks -> token lexbuf
+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
@@ -94,34 +117,48 @@ let rec token = lexer
   | num -> return lexbuf ("NUM", 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)
+  | 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.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))
+  | comment ->
+      if comments then
+        let comment =
+          Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 2)
+        in
+        return lexbuf ("COMMENT", comment)
+      else
+        token comments lexbuf
   | 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;
   }
+