X-Git-Url: http://matita.cs.unibo.it/gitweb/?a=blobdiff_plain;f=helm%2Focaml%2Fcic_notation%2FcicNotationLexer.ml;h=9c0efc8bb99a8b06c6b1e684637c1da562ba7c55;hb=08ecc780b3b0a4cac7ed72cf68c310e4eeffa2c1;hp=26dc5aacae4b9e8f83ec1f9863d1b700e13fdd6f;hpb=c27b932e5adcf89dc9de0e28f65e3370fe3e6b05;p=helm.git

diff --git a/helm/ocaml/cic_notation/cicNotationLexer.ml b/helm/ocaml/cic_notation/cicNotationLexer.ml
index 26dc5aaca..9c0efc8bb 100644
--- a/helm/ocaml/cic_notation/cicNotationLexer.ml
+++ b/helm/ocaml/cic_notation/cicNotationLexer.ml
@@ -29,19 +29,24 @@ exception Error of int * int * string
 
 let regexp number = xml_digit+
 
+  (* ZACK: breaks unicode's binder followed by an ascii letter without blank *)
+(* let regexp ident_letter = xml_letter *)
+
+let regexp ident_letter = [ 'a' - 'z' 'A' - 'Z' ]
+
 let regexp ident_decoration = '\'' | '!' | '?' | '`'
-let regexp ident_cont = xml_letter | xml_digit | '_'
-let regexp ident = xml_letter ident_cont* ident_decoration*
+let regexp ident_cont = ident_letter | xml_digit | '_'
+let regexp ident = ident_letter ident_cont* ident_decoration*
 
 let regexp tex_token = '\\' ident
 
 let regexp delim_begin = "\\["
 let regexp delim_end = "\\]"
 
-let regexp keyword = '"' ident '"'
 let regexp qkeyword = "'" ident "'"
 
 let regexp implicit = '?'
+let regexp placeholder = '%'
 let regexp meta = implicit number
 
 let regexp csymbol = '\'' ident
@@ -55,6 +60,11 @@ let regexp meta_ident = "$" ident
 let regexp meta_anonymous = "$_"
 let regexp qstring = '"' [^ '"']* '"'
 
+let regexp begincomment = "(**" xml_blank
+let regexp endcomment = "*)"
+let regexp comment_char = [^'*'] | '*'[^')']
+let regexp note = "(*" ([^'*'] | "**") comment_char* "*)"
+
 let level1_layouts = 
   [ "sub"; "sup";
     "below"; "above";
@@ -62,6 +72,14 @@ let level1_layouts =
     "sqrt"; "root"
   ]
 
+let level1_keywords =
+  [ "hbox"; "hvbox"; "hovbox"; "vbox";
+    "break";
+    "list0"; "list1"; "sep";
+    "opt";
+    "term"; "ident"; "number"
+  ] @ level1_layouts
+
 let level2_meta_keywords =
   [ "if"; "then"; "else";
     "fold"; "left"; "right"; "rec";
@@ -70,13 +88,15 @@ let level2_meta_keywords =
     "anonymous"; "ident"; "number"; "term"; "fresh"
   ]
 
-let level1_keywords =
-  [ "hbox"; "hvbox"; "hovbox"; "vbox";
-    "break";
-    "list0"; "list1"; "sep";
-    "opt";
-    "term"; "ident"; "number"
-  ]
+  (* (string, unit) Hashtbl.t, to exploit multiple bindings *)
+let level2_ast_keywords = Hashtbl.create 23
+let _ =
+  List.iter (fun k -> Hashtbl.add level2_ast_keywords k ())
+  [ "CProp"; "Prop"; "Type"; "Set"; "let"; "rec"; "corec"; "using"; "match";
+    "with"; "in"; "and"; "to"; "as"; "on"; "names" ]
+
+let add_level2_ast_keyword k = Hashtbl.add level2_ast_keywords k ()
+let remove_level2_ast_keyword k = Hashtbl.remove level2_ast_keywords k
 
 let regexp uri =
   ("cic:/" | "theory:/")              (* schema *)
@@ -138,8 +158,6 @@ let expand_macro lexbuf =
     ("SYMBOL", Utf8Macro.expand macro)
   with Utf8Macro.Macro_not_found _ -> "SYMBOL", Ulexing.utf8_lexeme lexbuf
 
-let keyword lexbuf = "KEYWORD", remove_quotes (Ulexing.utf8_lexeme lexbuf)
-
 let remove_quotes s = String.sub s 1 (String.length s - 2)
 let remove_left_quote s = String.sub s 1 (String.length s - 1)
 
@@ -173,29 +191,49 @@ let rec level2_meta_token = lexer
 	    return lexbuf ("IDENT", s)
 	end
   | "@{" -> read_unparsed_group "UNPARSED_AST" lexbuf
-  | ast_ident -> return lexbuf ("UNPARSED_AST", remove_left_quote (Ulexing.utf8_lexeme lexbuf))
-  | ast_csymbol ->  return lexbuf ("UNPARSED_AST", remove_left_quote (Ulexing.utf8_lexeme lexbuf))
+  | ast_ident ->
+      return lexbuf ("UNPARSED_AST",
+        remove_left_quote (Ulexing.utf8_lexeme lexbuf))
+  | ast_csymbol ->
+      return lexbuf ("UNPARSED_AST",
+        remove_left_quote (Ulexing.utf8_lexeme lexbuf))
   | eof -> return lexbuf ("EOI", "")
 
 let rec level2_ast_token = lexer
   | xml_blank+ -> level2_ast_token lexbuf
   | meta -> return lexbuf ("META", Ulexing.utf8_lexeme lexbuf)
-  | implicit -> return lexbuf ("IMPLICIT", Ulexing.utf8_lexeme lexbuf)
-  | ident -> return lexbuf ("IDENT", Ulexing.utf8_lexeme lexbuf)
+  | implicit -> return lexbuf ("IMPLICIT", "")
+  | placeholder -> return lexbuf ("PLACEHOLDER", "")
+  | ident ->
+      let lexeme = Ulexing.utf8_lexeme lexbuf in
+      if Hashtbl.mem level2_ast_keywords lexeme then
+        return lexbuf ("", lexeme)
+      else
+        return lexbuf ("IDENT", lexeme)
   | number -> return lexbuf ("NUMBER", Ulexing.utf8_lexeme lexbuf)
-  | keyword -> return lexbuf (keyword lexbuf)
   | tex_token -> return lexbuf (expand_macro lexbuf)
   | uri -> return lexbuf ("URI", Ulexing.utf8_lexeme lexbuf)
-  | qstring -> return lexbuf ("QSTRING", remove_quotes (Ulexing.utf8_lexeme lexbuf))
-  | csymbol -> return lexbuf ("CSYMBOL", Ulexing.utf8_lexeme lexbuf)
+  | qstring ->
+      return lexbuf ("QSTRING", remove_quotes (Ulexing.utf8_lexeme lexbuf))
+  | csymbol ->
+      return lexbuf ("CSYMBOL", remove_left_quote (Ulexing.utf8_lexeme lexbuf))
   | "${" -> read_unparsed_group "UNPARSED_META" lexbuf
   | "@{" -> read_unparsed_group "UNPARSED_AST" lexbuf
   | '(' -> return lexbuf ("LPAREN", "")
   | ')' -> return lexbuf ("RPAREN", "")
-  | meta_ident -> return lexbuf ("UNPARSED_META", remove_left_quote (Ulexing.utf8_lexeme lexbuf))
+  | meta_ident ->
+      return lexbuf ("UNPARSED_META",
+        remove_left_quote (Ulexing.utf8_lexeme lexbuf))
   | meta_anonymous -> return lexbuf ("UNPARSED_META", "anonymous")
-  | _ -> return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf)
+  | note -> 
+      let comment =
+        Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 4)
+      in
+      return lexbuf ("NOTE", comment)
+  | begincomment -> return lexbuf ("BEGINCOMMENT","")
+  | endcomment -> return lexbuf ("ENDCOMMENT","")
   | eof -> return lexbuf ("EOI", "")
+  | _ -> return lexbuf ("SYMBOL", Ulexing.utf8_lexeme lexbuf)
 
 let rec level1_pattern_token = lexer
   | xml_blank+ -> level1_pattern_token lexbuf
@@ -209,7 +247,8 @@ let rec level1_pattern_token = lexer
 	    return lexbuf ("IDENT", s)
 	end
   | tex_token -> return lexbuf (expand_macro lexbuf)
-  | qkeyword -> return lexbuf ("QKEYWORD", remove_quotes (Ulexing.utf8_lexeme lexbuf))
+  | qkeyword ->
+      return lexbuf ("QKEYWORD", remove_quotes (Ulexing.utf8_lexeme lexbuf))
   | '(' -> return lexbuf ("LPAREN", "")
   | ')' -> return lexbuf ("RPAREN", "")
   | eof -> return lexbuf ("EOI", "")