X-Git-Url: http://matita.cs.unibo.it/gitweb/?a=blobdiff_plain;f=matitaB%2Fcomponents%2Fcontent_pres%2FcicNotationLexer.ml;h=b7451bc780c8925f539aa72f589afdf764d9cf6c;hb=647b419e96770d90a82d7a9e5e8843566a9f93ee;hp=f5e8ba010b9a625b98297adf41a100adb7d220a4;hpb=dee464f8cd331524663167659d1fad01e558d4e1;p=helm.git diff --git a/matitaB/components/content_pres/cicNotationLexer.ml b/matitaB/components/content_pres/cicNotationLexer.ml index f5e8ba010..b7451bc78 100644 --- a/matitaB/components/content_pres/cicNotationLexer.ml +++ b/matitaB/components/content_pres/cicNotationLexer.ml @@ -82,12 +82,13 @@ let regexp hreftitle = "title=" qstring let regexp hrefclose = xmarkup "/" [ 'A' 'a' ] ymarkup let regexp tag_cont = ident_letter | xml_digit | "_" | "-" -let regexp tagname = ident_letter tag_cont* -let regexp opentag = xmarkup tagname -let regexp closetag = xmarkup "/" tagname ymarkup -let regexp attribute = tagname "=" qstring +let regexp gtagname = [ 'B' - 'Z' 'b' - 'z' ] | ident_letter tag_cont+ +let regexp attrname = ident_letter tag_cont* +let regexp gopentag = xmarkup gtagname +let regexp closetag = xmarkup "/" gtagname ymarkup +let regexp attribute = attrname "=" qstring let regexp generictag = - opentag (utf8_blank+ attribute)* ymarkup + gopentag (utf8_blank+ attribute)* ymarkup let regexp tex_token = '\\' ident @@ -129,11 +130,11 @@ let level1_keywords = "break"; "list0"; "list1"; "sep"; "opt"; - "term"; "ident"; "number"; + "term"; "ident"; "number"; "ref" ] @ level1_layouts let level2_meta_keywords = - [ "if"; "then"; "elCicNotationParser.se"; + [ "if"; "then"; "else"; "fold"; "left"; "right"; "rec"; "fail"; "default"; @@ -253,8 +254,9 @@ let read_unparsed_group token_name lexbuf = let end_cnum = level2_pattern_token_group 0 buffer lexbuf in return_with_loc (token_name, Buffer.contents buffer) begin_cnum end_cnum -let handle_keywords lexbuf k name = +let handle_keywords lexbuf k name = let s = Ulexing.utf8_lexeme lexbuf in + prerr_endline (Printf.sprintf "handling \"%s\" as a keyword or %s?" s name); if k s then return lexbuf ("", s) else @@ -329,43 +331,63 @@ let update_table loc desc href loctable = if desc <> None || href <> None then (let s,e = HExtlib.loc_of_floc loc in + prerr_endline (Printf.sprintf "*** [%d,%d] \"%s\",\"%s\"" s e (so_pp href) (so_pp desc)); + + Printf.printf "*** [%d,%d] \"%s\",\"%s\"" + s e (so_pp href) (so_pp desc); LocalizeEnv.add loc (href,desc) loctable) else loctable ;; let level2_ast_token loctable status = - let rec aux desc href = + prerr_endline ("X keywords = " ^ (String.concat ", " + (StringSet.elements status))); + (* start = starting point of last open A tag (default -1 = no tag) *) + let rec aux desc href start = lexer | let_rec -> return lexbuf ("LETREC","") | let_corec -> return lexbuf ("LETCOREC","") | we_proved -> return lexbuf ("WEPROVED","") | we_have -> return lexbuf ("WEHAVE","") - | utf8_blank+ -> ligatures_token (aux desc href) lexbuf + | utf8_blank+ -> ligatures_token (aux desc href start) lexbuf | meta -> let s = Ulexing.utf8_lexeme lexbuf in return lexbuf ("META", String.sub s 1 (String.length s - 1)) | implicit -> return lexbuf ("IMPLICIT", "") | placeholder -> return lexbuf ("PLACEHOLDER", "") - | hreftag -> aux_in_tag None None lexbuf - | hrefclose -> aux None None lexbuf + | hreftag -> + let start = Ulexing.lexeme_start lexbuf in + aux_attr None None start lexbuf + | hrefclose -> return lexbuf ("LEXING_ERROR", Ulexing.utf8_lexeme lexbuf) + (* ignore other tags *) | generictag - | closetag -> ligatures_token (aux desc href) lexbuf - | ident -> loctable := - update_table (loc_of_buf lexbuf) desc href !loctable; - handle_keywords lexbuf (fun x -> StringSet.mem x status) "IDENT" + | closetag -> ligatures_token (aux desc href start) lexbuf + | ident -> + if start <> ~-1 then + let idtok,_ = + handle_keywords lexbuf (fun x -> StringSet.mem x status) "IDENT" + in aux_close_tag desc href start idtok lexbuf +(* ("IDENT", Ulexing.utf8_lexeme lexbuf) lexbuf *) + else + handle_keywords lexbuf (fun x -> StringSet.mem x status) "IDENT" | variable_ident -> return lexbuf ("IDENT", Ulexing.utf8_lexeme lexbuf) - | pident -> loctable := - update_table (loc_of_buf lexbuf) desc href !loctable; - handle_keywords lexbuf (fun x -> StringSet.mem x status) "PIDENT" - | number -> loctable := - update_table (loc_of_buf lexbuf) desc href !loctable; - return lexbuf ("NUMBER", Ulexing.utf8_lexeme lexbuf) - | tex_token -> loctable := - update_table (loc_of_buf lexbuf) desc href !loctable; - return lexbuf (expand_macro lexbuf) + | pident -> if start <> ~-1 then + aux_close_tag desc href start ("PIDENT", Ulexing.utf8_lexeme lexbuf) lexbuf + else + handle_keywords lexbuf (fun x -> StringSet.mem x status) "PIDENT" + | number -> let token = "NUMBER", Ulexing.utf8_lexeme lexbuf + in + if start <> ~-1 then + aux_close_tag desc href start token lexbuf + else return lexbuf token + | tex_token -> let token = expand_macro lexbuf + in + if start <> ~-1 then + aux_close_tag desc href start token lexbuf + else return lexbuf token | nreference -> return lexbuf ("NREF", Ulexing.utf8_lexeme lexbuf) | uri -> return lexbuf ("URI", Ulexing.utf8_lexeme lexbuf) | qstring -> @@ -386,47 +408,78 @@ let level2_ast_token loctable status = Ulexing.utf8_sub_lexeme lexbuf 2 (Ulexing.lexeme_length lexbuf - 4) in return lexbuf ("NOTE", comment) *) - ligatures_token (aux desc href) lexbuf + ligatures_token (aux desc href start) lexbuf | begincomment -> return lexbuf ("BEGINCOMMENT","") | endcomment -> return lexbuf ("ENDCOMMENT","") | eof -> return_eoi lexbuf - | _ -> loctable := - update_table (loc_of_buf lexbuf) desc href !loctable; - return_symbol lexbuf (Ulexing.utf8_lexeme lexbuf) - - and aux_in_tag desc href = lexer - | utf8_blank+ -> ligatures_token (aux_in_tag desc href) lexbuf + | _ -> let token = "SYMBOL", (Ulexing.utf8_lexeme lexbuf) + in + if start <> ~-1 then + aux_close_tag desc href start token lexbuf + else return lexbuf token + + and aux_attr desc href start = lexer + | utf8_blank+ -> ligatures_token (aux_attr desc href start) lexbuf | href -> - aux_in_tag desc + aux_attr desc (Some (Ulexing.utf8_sub_lexeme lexbuf 6 (Ulexing.lexeme_length lexbuf - 7))) - lexbuf + start lexbuf | hreftitle -> - aux_in_tag + aux_attr (Some (Ulexing.utf8_sub_lexeme lexbuf 7 (Ulexing.lexeme_length lexbuf - 8))) - href lexbuf - | ">" -> aux desc href lexbuf - | _ -> aux None None lexbuf - in aux None None + href start lexbuf + | ymarkup -> aux desc href start lexbuf + | _ -> return lexbuf ("LEXING_ERROR", Ulexing.utf8_lexeme lexbuf) + +(* and aux_in_tag desc href = lexer + | ident -> loctable := + update_table (loc_of_buf lexbuf) desc href !loctable; + handle_keywords lexbuf (fun x -> StringSet.mem x status) "IDENT" + | variable_ident -> + return lexbuf ("IDENT", Ulexing.utf8_lexeme lexbuf) + | pident -> loctable := + update_table (loc_of_buf lexbuf) desc href !loctable; + handle_keywords lexbuf (fun x -> StringSet.mem x status) "PIDENT" + | number -> loctable := + update_table (loc_of_buf lexbuf) desc href !loctable; + return lexbuf ("NUMBER", Ulexing.utf8_lexeme lexbuf) + | tex_token -> loctable := + update_table (loc_of_buf lexbuf) desc href !loctable; + return lexbuf (expand_macro lexbuf) + | _ -> loctable := + update_table (loc_of_buf lexbuf) desc href !loctable; + return_symbol lexbuf (Ulexing.utf8_lexeme lexbuf) + *) + and aux_close_tag desc href start token = lexer + | hrefclose -> let _,b = Ulexing.loc lexbuf in + loctable := update_table (HExtlib.floc_of_loc (start,b)) desc href !loctable; + prerr_endline + (Printf.sprintf "adding loc (%d,%d) to table" start b); + return_with_loc token start b + | _ -> return lexbuf ("LEXING_ERROR", Ulexing.utf8_lexeme lexbuf) + in aux None None ~-1 let rec level1_pattern_token = lexer | utf8_blank+ -> ligatures_token level1_pattern_token lexbuf | number -> return lexbuf ("NUMBER", Ulexing.utf8_lexeme lexbuf) - | hreftag -> return lexbuf ("ATAG", "") +(* | hreftag -> return lexbuf ("ATAG", "") | hrefclose -> return lexbuf ("ATAGEND","") | generictag - | closetag -> ligatures_token level1_pattern_token lexbuf - | ident ->handle_keywords lexbuf (fun x -> List.mem x level1_keywords) "IDENT" + | closetag -> ligatures_token level1_pattern_token lexbuf *) + | qkeyword -> + return lexbuf ("QKEYWORD", remove_quotes (Ulexing.utf8_lexeme lexbuf)) + | csymbol -> prerr_endline ("lexing csymbol " ^ (Ulexing.utf8_lexeme lexbuf)); + return lexbuf ("CSYMBOL", remove_left_quote (Ulexing.utf8_lexeme lexbuf)) + | ident -> handle_keywords lexbuf (fun x -> List.mem x level1_keywords) "IDENT" | variable_ident -> return lexbuf ("IDENT", Ulexing.utf8_lexeme lexbuf) - | pident->handle_keywords lexbuf (fun x->List.mem x level1_keywords) "PIDENT" + | pident-> handle_keywords lexbuf (fun x->List.mem x level1_keywords) "PIDENT" | color -> return lexbuf ("COLOR", Ulexing.utf8_lexeme lexbuf) | percentage -> return lexbuf ("PERCENTAGE", Ulexing.utf8_lexeme lexbuf) | floatwithunit -> return lexbuf ("FLOATWITHUNIT", Ulexing.utf8_lexeme lexbuf) | tex_token -> return lexbuf (expand_macro lexbuf) - | qkeyword -> - return lexbuf ("QKEYWORD", remove_quotes (Ulexing.utf8_lexeme lexbuf)) | '(' -> return lexbuf ("LPAREN", "") | ')' -> return lexbuf ("RPAREN", "") | eof -> return_eoi lexbuf @@ -440,20 +493,22 @@ let level2_ast_token loctable status = type lexers = { level1_pattern_lexer : (string * string) Token.glexer; level2_ast_lexer : (string * string) Token.glexer; - level2_meta_lexer : (string * string) Token.glexer + level2_meta_lexer : (string * string) Token.glexer; } -let mk_lexers loctable keywords = - let initial_keywords = +let initial_keywords = [ "CProp"; "Prop"; "Type"; "Set"; "let"; "match"; "with"; "in"; "and"; "to"; "as"; "on"; "return"; "done" ] - in +;; + +let mk_lexers loctable keywords = + prerr_endline ("mk_lexers keywords: " ^ (String.concat ", " keywords)); let status = List.fold_right StringSet.add (initial_keywords @ keywords) StringSet.empty in { level1_pattern_lexer = mk_lexer level1_pattern_token; level2_ast_lexer = mk_lexer (level2_ast_token loctable status); - level2_meta_lexer = mk_lexer level2_meta_token + level2_meta_lexer = mk_lexer level2_meta_token; } ;;