src/Pure/Syntax/lexicon.ML
author wenzelm
Wed, 31 Jan 2018 14:02:37 +0100
changeset 67555 c550e38dd131
parent 67554 c2151a6bfd57
child 67556 5f86e2a9c59c
permissions -rw-r--r--
more efficient tokens_match_ord based on token_kind_index; tuned;
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     1
(*  Title:      Pure/Syntax/lexicon.ML
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     2
    Author:     Tobias Nipkow and Markus Wenzel, TU Muenchen
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
     3
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
     4
Lexer for the inner Isabelle syntax (terms and types).
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     5
*)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
     6
42290
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
     7
signature LEXICON =
4247
9bba9251bb4d added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents: 3828
diff changeset
     8
sig
42476
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
     9
  structure Syntax:
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    10
  sig
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    11
    val const: string -> term
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    12
    val free: string -> term
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    13
    val var: indexname -> term
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    14
  end
67426
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    15
  val scan_id: Symbol_Pos.T list scanner
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    16
  val scan_longid: Symbol_Pos.T list scanner
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    17
  val scan_tid: Symbol_Pos.T list scanner
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    18
  val scan_hex: Symbol_Pos.T list scanner
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    19
  val scan_bin: Symbol_Pos.T list scanner
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    20
  val scan_var: Symbol_Pos.T list scanner
6311cf9dc943 clarified signature;
wenzelm
parents: 67425
diff changeset
    21
  val scan_tvar: Symbol_Pos.T list scanner
50239
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
    22
  val is_tid: string -> bool
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
    23
  datatype token_kind =
67548
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
    24
    Literal | Ident | Long_Ident | Var | Type_Ident | Type_Var | Num | Float | Str |
67550
3b666615e3ce explicit dummy token;
wenzelm
parents: 67549
diff changeset
    25
    String | Cartouche | Space | Comment of Comment.kind option | Dummy | EOF
67554
c2151a6bfd57 more abstract type;
wenzelm
parents: 67552
diff changeset
    26
  eqtype token
67412
wenzelm
parents: 67388
diff changeset
    27
  val kind_of_token: token -> token_kind
27806
ece79c0597fe pos_of_token: Position.T;
wenzelm
parents: 27800
diff changeset
    28
  val str_of_token: token -> string
ece79c0597fe pos_of_token: Position.T;
wenzelm
parents: 27800
diff changeset
    29
  val pos_of_token: token -> Position.T
67551
4a087b9a29c5 clarified signature;
wenzelm
parents: 67550
diff changeset
    30
  val end_pos_of_token: token -> Position.T
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
    31
  val is_proper: token -> bool
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
    32
  val dummy: token
67552
679253fef277 clarified signature;
wenzelm
parents: 67551
diff changeset
    33
  val literal: string -> token
679253fef277 clarified signature;
wenzelm
parents: 67551
diff changeset
    34
  val is_literal: token -> bool
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
    35
  val tokens_match_ord: token * token -> order
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
    36
  val mk_eof: Position.T -> token
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
    37
  val eof: token
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
    38
  val is_eof: token -> bool
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
    39
  val stopper: token Scan.stopper
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    40
  val terminals: string list
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    41
  val is_terminal: string -> bool
49821
d15fe10593ff clarified output token markup (see also bc22daeed49e);
wenzelm
parents: 48992
diff changeset
    42
  val literal_markup: string -> Markup.T
44736
c2a3f1c84179 bulk reports for improved message throughput;
wenzelm
parents: 44706
diff changeset
    43
  val report_of_token: token -> Position.report
39510
d9f5f01faa1b Syntax.read_asts error: report token ranges within message -- no side-effect here;
wenzelm
parents: 39507
diff changeset
    44
  val reported_token_range: Proof.context -> token -> string
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    45
  val valued_token: token -> bool
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    46
  val predef_term: string -> token option
55108
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
    47
  val implode_string: Symbol.symbol list -> string
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
    48
  val explode_string: string * Position.T -> Symbol_Pos.T list
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
    49
  val implode_str: Symbol.symbol list -> string
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
    50
  val explode_str: string * Position.T -> Symbol_Pos.T list
30573
49899f26fbd1 de-camelized Symbol_Pos;
wenzelm
parents: 29565
diff changeset
    51
  val tokenize: Scan.lexicon -> bool -> Symbol_Pos.T list -> token list
42290
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    52
  val read_indexname: string -> indexname
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    53
  val read_var: string -> term
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    54
  val read_variable: string -> indexname option
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    55
  val read_nat: string -> int option
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    56
  val read_int: string -> int option
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
    57
  val read_num: string -> {radix: int, leading_zeros: int, value: int}
42290
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    58
  val read_float: string -> {mant: int, exp: int}
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    59
  val mark_class: string -> string val unmark_class: string -> string
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    60
  val mark_type: string -> string val unmark_type: string -> string
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    61
  val mark_const: string -> string val unmark_const: string -> string
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    62
  val mark_fixed: string -> string val unmark_fixed: string -> string
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    63
  val unmark:
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    64
   {case_class: string -> 'a,
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    65
    case_type: string -> 'a,
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    66
    case_const: string -> 'a,
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    67
    case_fixed: string -> 'a,
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    68
    case_default: string -> 'a} -> string -> 'a
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    69
  val is_marked: string -> bool
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    70
  val dummy_type: term
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
    71
  val fun_type: term
4247
9bba9251bb4d added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents: 3828
diff changeset
    72
end;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    73
14679
6ed90bd68eda added is_ascii_identifier;
wenzelm
parents: 13808
diff changeset
    74
structure Lexicon: LEXICON =
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    75
struct
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    76
42476
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    77
(** syntaxtic terms **)
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    78
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    79
structure Syntax =
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    80
struct
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    81
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    82
fun const c = Const (c, dummyT);
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    83
fun free x = Free (x, dummyT);
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    84
fun var xi = Var (xi, dummyT);
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    85
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    86
end;
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    87
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    88
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
    89
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
    90
(** basic scanners **)
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
    91
30573
49899f26fbd1 de-camelized Symbol_Pos;
wenzelm
parents: 29565
diff changeset
    92
open Basic_Symbol_Pos;
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
    93
55105
75815b3b38a1 tuned -- more direct err_prefix;
wenzelm
parents: 55035
diff changeset
    94
val err_prefix = "Inner lexical error: ";
75815b3b38a1 tuned -- more direct err_prefix;
wenzelm
parents: 55035
diff changeset
    95
75815b3b38a1 tuned -- more direct err_prefix;
wenzelm
parents: 55035
diff changeset
    96
fun !!! msg = Symbol_Pos.!!! (fn () => err_prefix ^ msg);
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
    97
50239
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
    98
val scan_id = Symbol_Pos.scan_ident;
61476
1884c40f1539 tuned signature;
wenzelm
parents: 59196
diff changeset
    99
val scan_longid = scan_id @@@ Scan.repeats1 ($$$ "." @@@ scan_id);
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   100
val scan_tid = $$$ "'" @@@ scan_id;
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   101
40525
14a2e686bdac eliminated slightly odd pervasive Symbol_Pos.symbol;
wenzelm
parents: 40290
diff changeset
   102
val scan_hex = $$$ "0" @@@ $$$ "x" @@@ Scan.many1 (Symbol.is_ascii_hex o Symbol_Pos.symbol);
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   103
val scan_bin = $$$ "0" @@@ $$$ "b" @@@ Scan.many1 (fn (s, _) => s = "0" orelse s = "1");
62782
057e8dbe4326 clarified modules;
wenzelm
parents: 62751
diff changeset
   104
val scan_num = scan_hex || scan_bin || Symbol_Pos.scan_nat;
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   105
62782
057e8dbe4326 clarified modules;
wenzelm
parents: 62751
diff changeset
   106
val scan_id_nat = scan_id @@@ Scan.optional ($$$ "." @@@ Symbol_Pos.scan_nat) [];
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   107
val scan_var = $$$ "?" @@@ scan_id_nat;
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   108
val scan_tvar = $$$ "?" @@@ $$$ "'" @@@ scan_id_nat;
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   109
50239
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   110
fun is_tid s =
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   111
  (case try (unprefix "'") s of
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   112
    SOME s' => Symbol_Pos.is_identifier s'
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   113
  | NONE => false);
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   114
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   115
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   116
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   117
(** tokens **)
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   118
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   119
(* datatype token_kind *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   120
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   121
datatype token_kind =
67548
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   122
  Literal | Ident | Long_Ident | Var | Type_Ident | Type_Var | Num | Float | Str |
67550
3b666615e3ce explicit dummy token;
wenzelm
parents: 67549
diff changeset
   123
  String | Cartouche | Space | Comment of Comment.kind option | Dummy | EOF;
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   124
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   125
val token_kinds =
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   126
  Vector.fromList
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   127
   [Literal, Ident, Long_Ident, Var, Type_Ident, Type_Var, Num, Float, Str,
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   128
    String, Cartouche, Space, Comment NONE, Comment (SOME Comment.Comment),
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   129
    Comment (SOME Comment.Cancel), Comment (SOME Comment.Latex), Dummy, EOF];
67539
1b8aad1909b7 tuned data structure and operations;
wenzelm
parents: 67532
diff changeset
   130
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   131
fun token_kind i = Vector.sub (token_kinds, i);
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   132
fun token_kind_index k = #1 (the (Vector.findi (fn (_, k') => k = k') token_kinds));
67539
1b8aad1909b7 tuned data structure and operations;
wenzelm
parents: 67532
diff changeset
   133
1b8aad1909b7 tuned data structure and operations;
wenzelm
parents: 67532
diff changeset
   134
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   135
(* datatype token *)
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   136
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   137
datatype token = Token of int * string * Position.range;
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   138
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   139
fun index_of_token (Token (i, _, _)) = i;
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   140
val kind_of_token = index_of_token #> token_kind;
27806
ece79c0597fe pos_of_token: Position.T;
wenzelm
parents: 27800
diff changeset
   141
fun str_of_token (Token (_, s, _)) = s;
ece79c0597fe pos_of_token: Position.T;
wenzelm
parents: 27800
diff changeset
   142
fun pos_of_token (Token (_, _, (pos, _))) = pos;
67551
4a087b9a29c5 clarified signature;
wenzelm
parents: 67550
diff changeset
   143
fun end_pos_of_token (Token (_, _, (_, end_pos))) = end_pos;
27806
ece79c0597fe pos_of_token: Position.T;
wenzelm
parents: 27800
diff changeset
   144
67425
7d4a088dbc0e clarified modules: uniform notion of formal comments;
wenzelm
parents: 67413
diff changeset
   145
val is_proper = kind_of_token #> (fn Space => false | Comment _ => false | _ => true);
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   146
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   147
val dummy = Token (token_kind_index Dummy, "", Position.no_range);
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   148
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   149
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   150
(* literals *)
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   151
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   152
val literal_index = token_kind_index Literal;
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   153
fun literal s = Token (literal_index, s, Position.no_range);
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   154
fun is_literal tok = index_of_token tok = literal_index;
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   155
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   156
fun tokens_match_ord toks =
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   157
  let val is = apply2 index_of_token toks in
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   158
    (case int_ord is of
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   159
      EQUAL => if #1 is = literal_index then fast_string_ord (apply2 str_of_token toks) else EQUAL
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   160
    | ord => ord)
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   161
  end;
67550
3b666615e3ce explicit dummy token;
wenzelm
parents: 67549
diff changeset
   162
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   163
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   164
(* stopper *)
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   165
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   166
val eof_index = token_kind_index EOF;
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   167
fun mk_eof pos = Token (eof_index, "", (pos, Position.none));
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   168
val eof = mk_eof Position.none;
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   169
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   170
fun is_eof tok = index_of_token tok = eof_index;
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   171
val stopper = Scan.stopper (K eof) is_eof;
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   172
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   173
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   174
(* terminal arguments *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   175
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   176
val terminal_kinds =
67548
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   177
 [("id", Ident),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   178
  ("longid", Long_Ident),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   179
  ("var", Var),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   180
  ("tid", Type_Ident),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   181
  ("tvar", Type_Var),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   182
  ("num_token", Num),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   183
  ("float_token", Float),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   184
  ("str_token", Str),
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   185
  ("string_token", String),
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54732
diff changeset
   186
  ("cartouche", Cartouche)];
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   187
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   188
val terminals = map #1 terminal_kinds;
20664
ffbc5a57191a member (op =);
wenzelm
parents: 20313
diff changeset
   189
val is_terminal = member (op =) terminals;
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   190
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   191
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   192
(* markup *)
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   193
49821
d15fe10593ff clarified output token markup (see also bc22daeed49e);
wenzelm
parents: 48992
diff changeset
   194
fun literal_markup s =
52189
816c88acd269 more literal tokens, e.g. "EX!";
wenzelm
parents: 51612
diff changeset
   195
  if Symbol.is_ascii_identifier s orelse exists Symbol.is_letter (Symbol.explode s)
816c88acd269 more literal tokens, e.g. "EX!";
wenzelm
parents: 51612
diff changeset
   196
  then Markup.literal
816c88acd269 more literal tokens, e.g. "EX!";
wenzelm
parents: 51612
diff changeset
   197
  else Markup.delimiter;
49821
d15fe10593ff clarified output token markup (see also bc22daeed49e);
wenzelm
parents: 48992
diff changeset
   198
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   199
val token_kind_markup =
67548
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   200
 fn Type_Ident => Markup.tfree
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   201
  | Type_Var => Markup.tvar
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   202
  | Num => Markup.numeral
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   203
  | Float => Markup.numeral
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   204
  | Str => Markup.inner_string
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   205
  | String => Markup.inner_string
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54732
diff changeset
   206
  | Cartouche => Markup.inner_cartouche
67440
e5ba0ca1e465 clarified markup;
wenzelm
parents: 67426
diff changeset
   207
  | Comment _ => Markup.inner_comment
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54732
diff changeset
   208
  | _ => Markup.empty;
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   209
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   210
fun report_of_token tok =
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   211
  let
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   212
    val markup =
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   213
      if is_literal tok then literal_markup (str_of_token tok)
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   214
      else token_kind_markup (kind_of_token tok);
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   215
  in (pos_of_token tok, markup) end;
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   216
39510
d9f5f01faa1b Syntax.read_asts error: report token ranges within message -- no side-effect here;
wenzelm
parents: 39507
diff changeset
   217
fun reported_token_range ctxt tok =
39168
e3ac771235f7 report token range after inner parse error -- often provides important clues about misunderstanding concerning lexical phase;
wenzelm
parents: 38474
diff changeset
   218
  if is_proper tok
50201
c26369c9eda6 Isabelle-specific implementation of quasi-abstract markup elements -- back to module arrangement before d83797ef0d2d;
wenzelm
parents: 49821
diff changeset
   219
  then Context_Position.reported_text ctxt (pos_of_token tok) Markup.token_range ""
39510
d9f5f01faa1b Syntax.read_asts error: report token ranges within message -- no side-effect here;
wenzelm
parents: 39507
diff changeset
   220
  else "";
39168
e3ac771235f7 report token range after inner parse error -- often provides important clues about misunderstanding concerning lexical phase;
wenzelm
parents: 38474
diff changeset
   221
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   222
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   223
(* valued_token *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   224
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   225
fun valued_token tok =
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   226
  not (is_literal tok orelse is_eof tok);
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   227
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   228
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   229
(* predef_term *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   230
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   231
fun predef_term s =
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   232
  (case AList.lookup (op =) terminal_kinds s of
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   233
    SOME k => SOME (Token (token_kind_index k, s, Position.no_range))
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   234
  | NONE => NONE);
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   235
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   236
55108
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   237
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   238
(** string literals **)
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   239
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   240
fun explode_literal scan_body (str, pos) =
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   241
  (case Scan.read Symbol_Pos.stopper scan_body (Symbol_Pos.explode (str, pos)) of
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   242
    SOME ss => ss
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   243
  | _ => error (err_prefix ^ "malformed string literal " ^ quote str ^ Position.here pos));
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   244
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   245
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   246
(* string *)
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   247
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   248
val scan_string = Scan.trace (Symbol_Pos.scan_string_qq err_prefix) >> #2;
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   249
val scan_string_body = Symbol_Pos.scan_string_qq err_prefix >> (#1 o #2);
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   250
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   251
fun implode_string ss = quote (implode (map (fn "\"" => "\\\"" | s => s) ss));
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   252
val explode_string = explode_literal scan_string_body;
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   253
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   254
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   255
(* str *)
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   256
14730
59ab60c6fcc6 support nested comments;
wenzelm
parents: 14679
diff changeset
   257
val scan_chr =
55107
1a29ea173bf9 tuned signature;
wenzelm
parents: 55106
diff changeset
   258
  $$ "\\" |-- $$$ "'" ||
40525
14a2e686bdac eliminated slightly odd pervasive Symbol_Pos.symbol;
wenzelm
parents: 40290
diff changeset
   259
  Scan.one
58854
b979c781c2db discontinued obsolete \<^sync> marker;
wenzelm
parents: 58421
diff changeset
   260
    ((fn s => s <> "\\" andalso s <> "'" andalso Symbol.not_eof s) o
40525
14a2e686bdac eliminated slightly odd pervasive Symbol_Pos.symbol;
wenzelm
parents: 40290
diff changeset
   261
      Symbol_Pos.symbol) >> single ||
55107
1a29ea173bf9 tuned signature;
wenzelm
parents: 55106
diff changeset
   262
  $$$ "'" --| Scan.ahead (~$$ "'");
14730
59ab60c6fcc6 support nested comments;
wenzelm
parents: 14679
diff changeset
   263
59ab60c6fcc6 support nested comments;
wenzelm
parents: 14679
diff changeset
   264
val scan_str =
55106
080c0006e917 tuned error messages, more accurate position;
wenzelm
parents: 55105
diff changeset
   265
  Scan.ahead ($$ "'" -- $$ "'") |--
080c0006e917 tuned error messages, more accurate position;
wenzelm
parents: 55105
diff changeset
   266
    !!! "unclosed string literal"
61476
1884c40f1539 tuned signature;
wenzelm
parents: 59196
diff changeset
   267
      ($$$ "'" @@@ $$$ "'" @@@ Scan.repeats scan_chr @@@ $$$ "'" @@@ $$$ "'");
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   268
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   269
val scan_str_body =
55107
1a29ea173bf9 tuned signature;
wenzelm
parents: 55106
diff changeset
   270
  Scan.ahead ($$ "'" |-- $$ "'") |--
55106
080c0006e917 tuned error messages, more accurate position;
wenzelm
parents: 55105
diff changeset
   271
    !!! "unclosed string literal"
61476
1884c40f1539 tuned signature;
wenzelm
parents: 59196
diff changeset
   272
      ($$ "'" |-- $$ "'" |-- Scan.repeats scan_chr --| $$ "'" --| $$ "'");
14730
59ab60c6fcc6 support nested comments;
wenzelm
parents: 14679
diff changeset
   273
55108
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   274
fun implode_str ss = enclose "''" "''" (implode (map (fn "'" => "\\'" | s => s) ss));
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   275
val explode_str = explode_literal scan_str_body;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   276
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   277
27806
ece79c0597fe pos_of_token: Position.T;
wenzelm
parents: 27800
diff changeset
   278
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   279
(** tokenize **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   280
67412
wenzelm
parents: 67388
diff changeset
   281
val token_leq = op <= o apply2 str_of_token;
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   282
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   283
fun token kind =
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   284
  let val i = token_kind_index kind
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   285
  in fn ss => Token (i, Symbol_Pos.content ss, Symbol_Pos.range ss) end;
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   286
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   287
fun tokenize lex xids syms =
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   288
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   289
    val scan_xid =
55962
fbd0e768bc8f special identifier "__" (i.e. empty name with internal suffix) serves as wild-card for completion;
wenzelm
parents: 55624
diff changeset
   290
      (if xids then $$$ "_" @@@ scan_id || scan_id else scan_id) ||
fbd0e768bc8f special identifier "__" (i.e. empty name with internal suffix) serves as wild-card for completion;
wenzelm
parents: 55624
diff changeset
   291
      $$$ "_" @@@ $$$ "_";
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   292
550
353eea6ec232 replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents: 376
diff changeset
   293
    val scan_val =
67548
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   294
      scan_tvar >> token Type_Var ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   295
      scan_var >> token Var ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   296
      scan_tid >> token Type_Ident ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   297
      Symbol_Pos.scan_float >> token Float ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   298
      scan_num >> token Num ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   299
      scan_longid >> token Long_Ident ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   300
      scan_xid >> token Ident;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   301
27800
df444ddeff56 datatype token: maintain range, tuned representation;
wenzelm
parents: 27773
diff changeset
   302
    val scan_lit = Scan.literal lex >> token Literal;
550
353eea6ec232 replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents: 376
diff changeset
   303
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   304
    val scan =
55105
75815b3b38a1 tuned -- more direct err_prefix;
wenzelm
parents: 55035
diff changeset
   305
      Symbol_Pos.scan_cartouche err_prefix >> token Cartouche ||
67425
7d4a088dbc0e clarified modules: uniform notion of formal comments;
wenzelm
parents: 67413
diff changeset
   306
      Symbol_Pos.scan_comment err_prefix >> token (Comment NONE) ||
7d4a088dbc0e clarified modules: uniform notion of formal comments;
wenzelm
parents: 67413
diff changeset
   307
      Comment.scan >> (fn (kind, ss) => token (Comment (SOME kind)) ss) ||
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   308
      Scan.max token_leq scan_lit scan_val ||
67548
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   309
      scan_string >> token String ||
c0f1667c1943 clarified signature;
wenzelm
parents: 67539
diff changeset
   310
      scan_str >> token Str ||
40525
14a2e686bdac eliminated slightly odd pervasive Symbol_Pos.symbol;
wenzelm
parents: 40290
diff changeset
   311
      Scan.many1 (Symbol.is_blank o Symbol_Pos.symbol) >> token Space;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   312
  in
67555
c550e38dd131 more efficient tokens_match_ord based on token_kind_index;
wenzelm
parents: 67554
diff changeset
   313
    (case Scan.error (Scan.finite Symbol_Pos.stopper (Scan.repeat scan)) syms of
27887
9f3fd48cf673 token_kind: add Space, Comment;
wenzelm
parents: 27806
diff changeset
   314
      (toks, []) => toks
55108
0b7a0c1fdf7e inner syntax token language allows regular quoted strings;
wenzelm
parents: 55107
diff changeset
   315
    | (_, ss) =>
55624
d52409077135 tuned messages;
wenzelm
parents: 55108
diff changeset
   316
        error ("Inner lexical error" ^ Position.here (#1 (Symbol_Pos.range ss)) ^
d52409077135 tuned messages;
wenzelm
parents: 55108
diff changeset
   317
          Markup.markup Markup.no_report ("\nat " ^ quote (Symbol_Pos.content ss))))
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   318
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   319
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   320
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   321
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   322
(** scan variables **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   323
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   324
(* scan_indexname *)
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   325
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   326
local
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   327
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   328
val scan_vname =
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   329
  let
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   330
    fun nat n [] = n
64275
ac2abc987cf9 accomodate Poly/ML repository version, which treats singleton strings as boxed;
wenzelm
parents: 62819
diff changeset
   331
      | nat n (c :: cs) = nat (n * 10 + (ord c - Char.ord #"0")) cs;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   332
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   333
    fun idxname cs ds = (implode (rev cs), nat 0 ds);
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   334
    fun chop_idx [] ds = idxname [] ds
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 61476
diff changeset
   335
      | chop_idx (cs as (_ :: "\<^sub>" :: _)) ds = idxname cs ds
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   336
      | chop_idx (c :: cs) ds =
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   337
          if Symbol.is_digit c then chop_idx cs (c :: ds)
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   338
          else idxname (c :: cs) ds;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   339
50239
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   340
    val scan =
fb579401dc26 tuned signature;
wenzelm
parents: 50238
diff changeset
   341
      (scan_id >> map Symbol_Pos.symbol) --
62782
057e8dbe4326 clarified modules;
wenzelm
parents: 62751
diff changeset
   342
      Scan.optional ($$ "." |-- Symbol_Pos.scan_nat >> (nat 0 o map Symbol_Pos.symbol)) ~1;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   343
  in
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   344
    scan >>
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   345
      (fn (cs, ~1) => chop_idx (rev cs) []
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   346
        | (cs, i) => (implode cs, i))
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   347
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   348
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   349
in
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   350
55107
1a29ea173bf9 tuned signature;
wenzelm
parents: 55106
diff changeset
   351
val scan_indexname = $$ "'" |-- scan_vname >> (fn (x, i) => ("'" ^ x, i)) || scan_vname;
15443
07f78cc82a73 indexname function now parses type variables as well; changed input
berghofe
parents: 14981
diff changeset
   352
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   353
end;
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   354
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   355
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   356
(* indexname *)
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   357
20313
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   358
fun read_indexname s =
62751
24e2b098bf44 tuned signature;
wenzelm
parents: 62663
diff changeset
   359
  (case Scan.read Symbol_Pos.stopper scan_indexname (Symbol_Pos.explode0 s) of
15531
08c8dad8e399 Deleted Library.option type.
skalberg
parents: 15443
diff changeset
   360
    SOME xi => xi
20313
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   361
  | _ => error ("Lexical error in variable name: " ^ quote s));
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   362
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   363
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   364
(* read_var *)
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   365
4703
a50ab39756db adapted to symbols, scan;
wenzelm
parents: 4587
diff changeset
   366
fun read_var str =
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   367
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   368
    val scan =
55107
1a29ea173bf9 tuned signature;
wenzelm
parents: 55106
diff changeset
   369
      $$ "?" |-- scan_indexname --| Scan.ahead (Scan.one Symbol_Pos.is_eof)
42476
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
   370
        >> Syntax.var ||
58854
b979c781c2db discontinued obsolete \<^sync> marker;
wenzelm
parents: 58421
diff changeset
   371
      Scan.many (Symbol.not_eof o Symbol_Pos.symbol)
42476
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
   372
        >> (Syntax.free o implode o map Symbol_Pos.symbol);
62751
24e2b098bf44 tuned signature;
wenzelm
parents: 62663
diff changeset
   373
  in the (Scan.read Symbol_Pos.stopper scan (Symbol_Pos.explode0 str)) end;
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   374
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   375
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   376
(* read_variable *)
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   377
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   378
fun read_variable str =
55107
1a29ea173bf9 tuned signature;
wenzelm
parents: 55106
diff changeset
   379
  let val scan = $$ "?" |-- scan_indexname || scan_indexname
62751
24e2b098bf44 tuned signature;
wenzelm
parents: 62663
diff changeset
   380
  in Scan.read Symbol_Pos.stopper scan (Symbol_Pos.explode0 str) end;
4587
6bce9ef27d7e added read_var;
wenzelm
parents: 4247
diff changeset
   381
6bce9ef27d7e added read_var;
wenzelm
parents: 4247
diff changeset
   382
20313
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   383
(* read numbers *)
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   384
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   385
local
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   386
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   387
fun nat cs =
40525
14a2e686bdac eliminated slightly odd pervasive Symbol_Pos.symbol;
wenzelm
parents: 40290
diff changeset
   388
  Option.map (#1 o Library.read_int o map Symbol_Pos.symbol)
62782
057e8dbe4326 clarified modules;
wenzelm
parents: 62751
diff changeset
   389
    (Scan.read Symbol_Pos.stopper Symbol_Pos.scan_nat cs);
5860
ed11c9890852 added read_nat;
wenzelm
parents: 5513
diff changeset
   390
20313
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   391
in
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   392
62751
24e2b098bf44 tuned signature;
wenzelm
parents: 62663
diff changeset
   393
fun read_nat s = nat (Symbol_Pos.explode0 s);
20313
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   394
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   395
fun read_int s =
62751
24e2b098bf44 tuned signature;
wenzelm
parents: 62663
diff changeset
   396
  (case Symbol_Pos.explode0 s of
27773
a52166b228b9 improved position handling due to SymbolPos.T;
wenzelm
parents: 27121
diff changeset
   397
    ("-", _) :: cs => Option.map ~ (nat cs)
20313
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   398
  | cs => nat cs);
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   399
bf9101cc4385 renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents: 20165
diff changeset
   400
end;
5860
ed11c9890852 added read_nat;
wenzelm
parents: 5513
diff changeset
   401
ed11c9890852 added read_nat;
wenzelm
parents: 5513
diff changeset
   402
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   403
(* read_num: hex/bin/decimal *)
9326
1625c1f172b3 added read_xnum;
wenzelm
parents: 9289
diff changeset
   404
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   405
local
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   406
64275
ac2abc987cf9 accomodate Poly/ML repository version, which treats singleton strings as boxed;
wenzelm
parents: 62819
diff changeset
   407
val ten = Char.ord #"0" + 10;
ac2abc987cf9 accomodate Poly/ML repository version, which treats singleton strings as boxed;
wenzelm
parents: 62819
diff changeset
   408
val a = Char.ord #"a";
ac2abc987cf9 accomodate Poly/ML repository version, which treats singleton strings as boxed;
wenzelm
parents: 62819
diff changeset
   409
val A = Char.ord #"A";
35428
bd7d6f65976e more systematic mark/unmark operations;
wenzelm
parents: 35262
diff changeset
   410
val _ = a > A orelse raise Fail "Bad ASCII";
20096
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   411
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   412
fun remap_hex c =
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   413
  let val x = ord c in
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   414
    if x >= a then chr (x - a + ten)
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   415
    else if x >= A then chr (x - A + ten)
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   416
    else c
7058714024b3 uniform treatment of num/xnum;
wenzelm
parents: 20091
diff changeset
   417
  end;
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   418
21781
8314ebb5364d read_xnum: return leading_zeros, radix;
wenzelm
parents: 21774
diff changeset
   419
fun leading_zeros ["0"] = 0
8314ebb5364d read_xnum: return leading_zeros, radix;
wenzelm
parents: 21774
diff changeset
   420
  | leading_zeros ("0" :: cs) = 1 + leading_zeros cs
8314ebb5364d read_xnum: return leading_zeros, radix;
wenzelm
parents: 21774
diff changeset
   421
  | leading_zeros _ = 0;
8314ebb5364d read_xnum: return leading_zeros, radix;
wenzelm
parents: 21774
diff changeset
   422
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   423
in
15965
f422f8283491 Use of IntInf.int instead of int in most numeric simprocs; avoids
paulson
parents: 15570
diff changeset
   424
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   425
fun read_num str =
9326
1625c1f172b3 added read_xnum;
wenzelm
parents: 9289
diff changeset
   426
  let
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   427
    val (radix, digs) =
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   428
      (case Symbol.explode str of
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   429
        "0" :: "x" :: cs => (16, map remap_hex cs)
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   430
      | "0" :: "b" :: cs => (2, cs)
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   431
      | cs => (10, cs));
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   432
  in
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   433
   {radix = radix,
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   434
    leading_zeros = leading_zeros digs,
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   435
    value = #1 (Library.read_radix_int radix digs)}
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   436
  end;
9326
1625c1f172b3 added read_xnum;
wenzelm
parents: 9289
diff changeset
   437
15991
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   438
end;
670f8e4b5a98 added read_variable: optional question mark on input;
wenzelm
parents: 15965
diff changeset
   439
28904
3ef9489eeef5 New lexical item "float".
nipkow
parents: 28413
diff changeset
   440
fun read_float str =
3ef9489eeef5 New lexical item "float".
nipkow
parents: 28413
diff changeset
   441
  let
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   442
    val cs = Symbol.explode str;
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   443
    val (intpart, fracpart) =
67522
9e712280cc37 clarified take/drop/chop prefix/suffix;
wenzelm
parents: 67440
diff changeset
   444
      (case chop_prefix Symbol.is_digit cs of
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   445
        (intpart, "." :: fracpart) => (intpart, fracpart)
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   446
      | _ => raise Fail "read_float");
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   447
  in
58421
37cbbd8eb460 discontinued old "xnum" token category;
wenzelm
parents: 58410
diff changeset
   448
   {mant = #1 (Library.read_int (intpart @ fracpart)),
29156
89f76a58a378 renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents: 28904
diff changeset
   449
    exp = length fracpart}
42046
6341c23baf10 added Lexicon.encode_position, Lexicon.decode_position;
wenzelm
parents: 40525
diff changeset
   450
  end;
6341c23baf10 added Lexicon.encode_position, Lexicon.decode_position;
wenzelm
parents: 40525
diff changeset
   451
42290
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   452
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   453
(* marked logical entities *)
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   454
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   455
fun marker s = (prefix s, unprefix s);
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   456
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 61476
diff changeset
   457
val (mark_class, unmark_class) = marker "\<^class>";
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 61476
diff changeset
   458
val (mark_type, unmark_type) = marker "\<^type>";
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 61476
diff changeset
   459
val (mark_const, unmark_const) = marker "\<^const>";
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 61476
diff changeset
   460
val (mark_fixed, unmark_fixed) = marker "\<^fixed>";
42290
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   461
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   462
fun unmark {case_class, case_type, case_const, case_fixed, case_default} s =
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   463
  (case try unmark_class s of
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   464
    SOME c => case_class c
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   465
  | NONE =>
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   466
      (case try unmark_type s of
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   467
        SOME c => case_type c
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   468
      | NONE =>
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   469
          (case try unmark_const s of
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   470
            SOME c => case_const c
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   471
          | NONE =>
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   472
              (case try unmark_fixed s of
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   473
                SOME c => case_fixed c
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   474
              | NONE => case_default s))));
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   475
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   476
val is_marked =
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   477
  unmark {case_class = K true, case_type = K true, case_const = K true,
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   478
    case_fixed = K true, case_default = K false};
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   479
42476
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
   480
val dummy_type = Syntax.const (mark_type "dummy");
d0bc1268ef09 clarified auxiliary structure Lexicon.Syntax;
wenzelm
parents: 42290
diff changeset
   481
val fun_type = Syntax.const (mark_type "fun");
42290
b1f544c84040 discontinued special treatment of structure Lexicon;
wenzelm
parents: 42264
diff changeset
   482
59196
73a6403637b3 more toplevel pretty printing;
wenzelm
parents: 58854
diff changeset
   483
73a6403637b3 more toplevel pretty printing;
wenzelm
parents: 58854
diff changeset
   484
(* toplevel pretty printing *)
73a6403637b3 more toplevel pretty printing;
wenzelm
parents: 58854
diff changeset
   485
62663
bea354f6ff21 clarified modules;
wenzelm
parents: 62529
diff changeset
   486
val _ =
62819
d3ff367a16a0 careful export of type-dependent functions, without losing their special status;
wenzelm
parents: 62782
diff changeset
   487
  ML_system_pp (fn _ => fn _ =>
62663
bea354f6ff21 clarified modules;
wenzelm
parents: 62529
diff changeset
   488
    Pretty.to_polyml o Pretty.str_list "{" "}" o map quote o Scan.dest_lexicon);
59196
73a6403637b3 more toplevel pretty printing;
wenzelm
parents: 58854
diff changeset
   489
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   490
end;