src/HOL/Tools/string_syntax.ML
author haftmann
Sun, 06 May 2018 18:20:25 +0000
changeset 68099 305f9f3edf05
parent 68028 1f9f973eed2a
child 68939 bcce5967e10e
permissions -rw-r--r--
typo
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
     1
(*  Title:      HOL/Tools/string_syntax.ML
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
     2
    Author:     Makarius
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
     3
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
     4
Concrete syntax for characters and strings.
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
     5
*)
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
     6
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
     7
signature STRING_SYNTAX = sig
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
     8
  val hex: int -> string
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
     9
  val mk_bits_syntax: int -> int -> term list
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    10
  val dest_bits_syntax: term list -> int
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    11
  val plain_strings_of: string -> string list
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    12
  datatype character = Char of string | Ord of int
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    13
  val classify_character: int -> character
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    14
end
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    15
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    16
structure String_Syntax: STRING_SYNTAX =
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    17
struct
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    18
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    19
(* numeral *)
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    20
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    21
fun hex_digit n = if n = 10 then "A"
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    22
  else if n = 11 then "B"
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    23
  else if n = 12 then "C"
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    24
  else if n = 13 then "D"
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    25
  else if n = 14 then "E"
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    26
  else if n = 15 then "F"
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    27
  else string_of_int n;
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    28
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    29
fun hex_prefix ms = "0x" ^ implode (replicate (2 - length ms) "0" @ ms);
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    30
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    31
fun hex n = hex_prefix (map hex_digit (radixpand (16, n)));
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    32
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    33
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    34
(* booleans as bits *)
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    35
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    36
fun mk_bit_syntax b =
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    37
  Syntax.const (if b = 1 then @{const_syntax True} else @{const_syntax False});
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    38
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    39
fun mk_bits_syntax len = map mk_bit_syntax o Integer.radicify 2 len;
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    40
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    41
fun dest_bit_syntax (Const (@{const_syntax True}, _)) = 1 
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    42
  | dest_bit_syntax (Const (@{const_syntax False}, _)) = 0
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    43
  | dest_bit_syntax _ = raise Match;
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    44
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    45
val dest_bits_syntax = Integer.eval_radix 2 o map dest_bit_syntax;
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    46
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    47
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    48
(* char *)
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    49
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    50
fun mk_char_syntax i =
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    51
  list_comb (Syntax.const @{const_syntax Char}, mk_bits_syntax 8 i);
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    52
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    53
fun mk_char_syntax' c =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    54
  if Symbol.is_ascii c then mk_char_syntax (ord c)
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    55
  else if c = "\<newline>" then mk_char_syntax 10
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    56
  else error ("Bad character: " ^ quote c);
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    57
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    58
fun plain_strings_of str =
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    59
  map fst (Lexicon.explode_str (str, Position.none));
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    60
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    61
datatype character = Char of string | Ord of int;
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    62
40627
becf5d5187cc renamed raw "explode" function to "raw_explode" to emphasize its meaning;
wenzelm
parents: 35363
diff changeset
    63
val specials = raw_explode "\\\"`'";
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    64
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    65
fun classify_character i =
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    66
  let
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    67
    val c = chr i
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    68
  in
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    69
    if not (member (op =) specials c) andalso Symbol.is_ascii c andalso Symbol.is_printable c
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    70
    then Char c
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    71
    else if c = "\n"
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    72
    then Char "\<newline>"
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    73
    else Ord i
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    74
  end;
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    75
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    76
fun dest_char_syntax b0 b1 b2 b3 b4 b5 b6 b7 =
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    77
  classify_character (dest_bits_syntax [b0, b1, b2, b3, b4, b5, b6, b7])
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    78
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    79
fun dest_char_ast (Ast.Appl [Ast.Constant @{syntax_const "_Char"}, Ast.Constant s]) =
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    80
      plain_strings_of s
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    81
  | dest_char_ast _ = raise Match;
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    82
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    83
fun char_tr [(c as Const (@{syntax_const "_constrain"}, _)) $ t $ u] =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    84
      c $ char_tr [t] $ u
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    85
  | char_tr [Free (str, _)] =
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    86
      (case plain_strings_of str of
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    87
        [c] => mk_char_syntax' c
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    88
      | _ => error ("Single character expected: " ^ str))
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    89
  | char_tr ts = raise TERM ("char_tr", ts);
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    90
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    91
fun char_ord_tr [(c as Const (@{syntax_const "_constrain"}, _)) $ t $ u] =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    92
      c $ char_ord_tr [t] $ u
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    93
  | char_ord_tr [Const (num, _)] =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    94
      (mk_char_syntax o #value o Lexicon.read_num) num
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    95
  | char_ord_tr ts = raise TERM ("char_ord_tr", ts);
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
    96
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    97
fun char_tr' [b1, b2, b3, b4, b5, b6, b7, b8] =
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
    98
      (case dest_char_syntax b1 b2 b3 b4 b5 b6 b7 b8 of
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
    99
        Char s => Syntax.const @{syntax_const "_Char"} $
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   100
          Syntax.const (Lexicon.implode_str [s])
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
   101
      | Ord n => Syntax.const @{syntax_const "_Char_ord"} $
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
   102
          Syntax.free (hex n))
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   103
  | char_tr' _ = raise Match;
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   104
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   105
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   106
(* string *)
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   107
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   108
fun mk_string_syntax [] = Syntax.const @{const_syntax Nil}
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   109
  | mk_string_syntax (c :: cs) =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   110
      Syntax.const @{const_syntax Cons} $ mk_char_syntax' c $ mk_string_syntax cs;
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   111
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   112
fun mk_string_ast ss =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   113
  Ast.Appl [Ast.Constant @{syntax_const "_inner_string"},
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   114
    Ast.Variable (Lexicon.implode_str ss)];
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   115
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   116
fun string_tr [(c as Const (@{syntax_const "_constrain"}, _)) $ t $ u] =
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   117
      c $ string_tr [t] $ u
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   118
  | string_tr [Free (str, _)] =
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
   119
      mk_string_syntax (plain_strings_of str)
68099
haftmann
parents: 68028
diff changeset
   120
  | string_tr ts = raise TERM ("string_tr", ts);
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   121
35115
446c5063e4fd modernized translations;
wenzelm
parents: 31048
diff changeset
   122
fun list_ast_tr' [args] =
42224
578a51fae383 discontinued special treatment of structure Ast: no pervasive content, no inclusion in structure Syntax;
wenzelm
parents: 40627
diff changeset
   123
      Ast.Appl [Ast.Constant @{syntax_const "_String"},
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   124
        (mk_string_ast o maps dest_char_ast o Ast.unfold_ast @{syntax_const "_args"}) args]
51160
599ff65b85e2 systematic conversions between nat and nibble/char;
haftmann
parents: 46483
diff changeset
   125
  | list_ast_tr' _ = raise Match;
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   126
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   127
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   128
(* theory setup *)
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   129
58822
90a5e981af3e modernized setup;
wenzelm
parents: 55108
diff changeset
   130
val _ =
90a5e981af3e modernized setup;
wenzelm
parents: 55108
diff changeset
   131
  Theory.setup
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   132
   (Sign.parse_translation
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   133
     [(@{syntax_const "_Char"}, K char_tr),
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   134
      (@{syntax_const "_Char_ord"}, K char_ord_tr),
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   135
      (@{syntax_const "_String"}, K string_tr)] #>
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   136
    Sign.print_translation
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   137
     [(@{const_syntax Char}, K char_tr')] #>
58822
90a5e981af3e modernized setup;
wenzelm
parents: 55108
diff changeset
   138
    Sign.print_ast_translation
62597
b3f2b8c906a6 model characters directly as range 0..255
haftmann
parents: 58822
diff changeset
   139
     [(@{syntax_const "_list"}, K list_ast_tr')]);
21759
f4b20360751f Concrete syntax for hex chars and strings.
wenzelm
parents:
diff changeset
   140
68028
1f9f973eed2a proper datatype for 8-bit characters
haftmann
parents: 62597
diff changeset
   141
end