src/Pure/Syntax/lexicon.ML
author wenzelm
Wed, 18 May 1994 15:24:39 +0200
changeset 376 d3d01131470f
parent 330 2fda15dd1e0f
child 550 353eea6ec232
permissions -rw-r--r--
extended signature SCANNER by some basic scanners and type lexicon; various minor internal changes;
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
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
     2
    ID:         $Id$
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     3
    Author:     Tobias Nipkow and Markus Wenzel, TU Muenchen
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
     4
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
     5
Scanner combinators and Isabelle's main lexer (used for terms and types).
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     6
*)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
     7
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     8
infix 5 -- ^^;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
     9
infix 3 >>;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    10
infix 0 ||;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    11
164
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    12
signature SCANNER =
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    13
sig
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    14
  exception LEXICAL_ERROR
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    15
  val >> : ('a -> 'b * 'c) * ('b -> 'd) -> 'a -> 'd * 'c
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    16
  val || : ('a -> 'b) * ('a -> 'b) -> 'a -> 'b
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    17
  val -- : ('a -> 'b * 'c) * ('c -> 'd * 'e) -> 'a -> ('b * 'd) * 'e
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    18
  val ^^ : ('a -> string * 'b) * ('b -> string * 'c) -> 'a -> string * 'c
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    19
  val $$ : ''a -> ''a list -> ''a * ''a list
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    20
  val scan_empty: 'a list -> 'b list * 'a list
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    21
  val scan_one: ('a -> bool) -> 'a list -> 'a * 'a list
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    22
  val scan_any: ('a -> bool) -> 'a list -> 'a list * 'a list
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    23
  val scan_any1: ('a -> bool) -> 'a list -> 'a list * 'a list
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    24
  val scan_end: 'a list -> 'b list * 'a list
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    25
  val optional: ('a -> 'b * 'a) -> 'a -> 'b option * 'a
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    26
  val repeat: ('a -> 'b * 'a) -> 'a -> 'b list * 'a
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    27
  val repeat1: ('a -> 'b * 'a) -> 'a -> 'b list * 'a
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    28
  val max_of: ('a -> ('b * string) * 'a) -> ('a -> ('b * string) * 'a)
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    29
    -> 'a -> ('b * string) option * 'a
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    30
  val scan_id: string list -> string * string list
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    31
  val scan_tid: string list -> string * string list
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    32
  val scan_nat: string list -> string * string list
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    33
  type lexicon
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    34
  val dest_lexicon: lexicon -> string list
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    35
  val empty_lexicon: lexicon
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    36
  val extend_lexicon: lexicon -> string list -> lexicon
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    37
  val make_lexicon: string list -> lexicon
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    38
  val merge_lexicons: lexicon -> lexicon -> lexicon
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
    39
  val scan_literal: lexicon -> string list -> string * string list
164
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    40
end;
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    41
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    42
signature LEXICON0 =
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    43
sig
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    44
  val is_identifier: string -> bool
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    45
  val string_of_vname: indexname -> string
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    46
  val scan_varname: string list -> indexname * string list
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    47
  val scan_var: string -> term
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    48
end;
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    49
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    50
signature LEXICON =
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    51
sig
164
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
    52
  include SCANNER
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    53
  include LEXICON0
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    54
  val is_xid: string -> bool
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
    55
  val is_tid: string -> bool
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    56
  datatype token =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    57
    Token of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    58
    IdentSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    59
    VarSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    60
    TFreeSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    61
    TVarSy of string |
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    62
    EndToken
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    63
  val id: string
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    64
  val var: string
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
    65
  val tid: string
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    66
  val tvar: string
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    67
  val terminals: string list
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    68
  val is_terminal: string -> bool
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    69
  val str_of_token: token -> string
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    70
  val display_token: token -> string
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    71
  val matching_tokens: token * token -> bool
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
    72
  val token_assoc: (token option * 'a list) list * token -> 'a list
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    73
  val valued_token: token -> bool
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
    74
  val predef_term: string -> token option
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    75
  val tokenize: lexicon -> bool -> string -> token list
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    76
end;
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    77
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    78
functor LexiconFun(): LEXICON =
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    79
struct
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    80
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    81
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    82
(** is_identifier etc. **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    83
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    84
fun is_ident [] = false
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    85
  | is_ident (c :: cs) = is_letter c andalso forall is_letdig cs;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    86
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    87
val is_identifier = is_ident o explode;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    88
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    89
fun is_xid s =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    90
  (case explode s of
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    91
    "_" :: cs => is_ident cs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    92
  | cs => is_ident cs);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    93
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
    94
fun is_tid s =
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    95
  (case explode s of
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    96
    "'" :: cs => is_ident cs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    97
  | _ => false);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
    98
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
    99
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   100
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   101
(** string_of_vname **)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   102
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   103
fun string_of_vname (x, i) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   104
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   105
    val si = string_of_int i;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   106
  in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   107
    if is_digit (last_elem (explode x)) then "?" ^ x ^ "." ^ si
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   108
    else if i = 0 then "?" ^ x
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   109
    else "?" ^ x ^ si
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   110
  end;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   111
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   112
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   113
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   114
(** datatype token **)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   115
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   116
datatype token =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   117
  Token of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   118
  IdentSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   119
  VarSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   120
  TFreeSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   121
  TVarSy of string |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   122
  EndToken;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   123
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   124
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   125
(* terminal arguments *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   126
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   127
val id = "id";
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   128
val var = "var";
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   129
val tid = "tid";
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   130
val tvar = "tvar";
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   131
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   132
val terminals = [id, var, tid, tvar];
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   133
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   134
fun is_terminal s = s mem terminals;
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   135
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   136
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   137
(* str_of_token *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   138
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   139
fun str_of_token (Token s) = s
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   140
  | str_of_token (IdentSy s) = s
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   141
  | str_of_token (VarSy s) = s
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   142
  | str_of_token (TFreeSy s) = s
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   143
  | str_of_token (TVarSy s) = s
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   144
  | str_of_token EndToken = "EOF";
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   145
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   146
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   147
(* display_token *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   148
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   149
fun display_token (Token s) = quote s
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   150
  | display_token (IdentSy s) = "id(" ^ s ^ ")"
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   151
  | display_token (VarSy s) = "var(" ^ s ^ ")"
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   152
  | display_token (TFreeSy s) = "tid(" ^ s ^ ")"
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   153
  | display_token (TVarSy s) = "tvar(" ^ s ^ ")"
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   154
  | display_token EndToken = "";
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   155
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   156
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   157
(* matching_tokens *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   158
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   159
fun matching_tokens (Token x, Token y) = (x = y)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   160
  | matching_tokens (IdentSy _, IdentSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   161
  | matching_tokens (VarSy _, VarSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   162
  | matching_tokens (TFreeSy _, TFreeSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   163
  | matching_tokens (TVarSy _, TVarSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   164
  | matching_tokens (EndToken, EndToken) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   165
  | matching_tokens _ = false;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   166
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   167
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   168
(* token_assoc *)
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   169
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   170
fun token_assoc (list, key) =
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   171
  let
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   172
    fun assoc [] = []
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   173
      | assoc ((keyi, xi) :: pairs) =
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   174
          if is_none keyi orelse matching_tokens (the keyi, key) then
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   175
            assoc pairs @ xi
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   176
          else assoc pairs;
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   177
  in assoc list end;
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   178
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   179
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   180
(* valued_token *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   181
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   182
fun valued_token (Token _) = false
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   183
  | valued_token (IdentSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   184
  | valued_token (VarSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   185
  | valued_token (TFreeSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   186
  | valued_token (TVarSy _) = true
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   187
  | valued_token EndToken = false;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   188
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   189
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   190
(* predef_term *)
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   191
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   192
fun predef_term name =
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   193
  if name = id then Some (IdentSy name)
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   194
  else if name = var then Some (VarSy name)
330
2fda15dd1e0f changed the way a grammar is generated to allow the new parser to work;
clasohm
parents: 237
diff changeset
   195
  else if name = tid then Some (TFreeSy name)
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   196
  else if name = tvar then Some (TVarSy name)
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   197
  else None;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   198
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   199
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   200
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   201
(** datatype lexicon **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   202
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   203
datatype lexicon =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   204
  Empty |
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   205
  Branch of string * string * lexicon * lexicon * lexicon;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   206
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   207
val no_token = "";
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   208
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   209
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   210
(* dest_lexicon *)
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   211
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   212
fun dest_lexicon Empty = []
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   213
  | dest_lexicon (Branch (_, "", lt, eq, gt)) =
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   214
      dest_lexicon eq @ dest_lexicon lt @ dest_lexicon gt
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   215
  | dest_lexicon (Branch (_, str, lt, eq, gt)) =
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   216
      str :: (dest_lexicon eq @ dest_lexicon lt @ dest_lexicon gt);
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   217
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   218
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   219
(* empty, extend, make, merge lexicons *)
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   220
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   221
val empty_lexicon = Empty;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   222
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   223
fun extend_lexicon lexicon strs =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   224
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   225
    fun ext (lex, s) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   226
      let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   227
        fun add (Branch (d, a, lt, eq, gt)) (chs as c :: cs) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   228
              if c < d then Branch (d, a, add lt chs, eq, gt)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   229
              else if c > d then Branch (d, a, lt, eq, add gt chs)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   230
              else Branch (d, if null cs then s else a, lt, add eq cs, gt)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   231
          | add Empty [c] =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   232
              Branch (c, s, Empty, Empty, Empty)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   233
          | add Empty (c :: cs) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   234
              Branch (c, no_token, Empty, add Empty cs, Empty)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   235
          | add lex [] = lex;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   236
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   237
        val cs = explode s;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   238
      in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   239
        if exists is_blank cs then
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   240
          sys_error ("extend_lexicon: blank in delimiter " ^ quote s)
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   241
        else add lex cs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   242
      end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   243
  in
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   244
    foldl ext (lexicon, strs \\ dest_lexicon lexicon)
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   245
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   246
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   247
val make_lexicon = extend_lexicon empty_lexicon;
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   248
237
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   249
fun merge_lexicons lex1 lex2 =
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   250
  let
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   251
    val strs1 = dest_lexicon lex1;
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   252
    val strs2 = dest_lexicon lex2;
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   253
  in
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   254
    if strs2 subset strs1 then lex1
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   255
    else if strs1 subset strs2 then lex2
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   256
    else extend_lexicon lex1 strs2
a7d3e712767a MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents: 164
diff changeset
   257
  end;
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   258
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   259
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   260
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   261
(** scanners **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   262
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   263
exception LEXICAL_ERROR;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   264
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   265
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   266
(* scanner combinators *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   267
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   268
fun (scan >> f) cs = apfst f (scan cs);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   269
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   270
fun (scan1 || scan2) cs = scan1 cs handle LEXICAL_ERROR => scan2 cs;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   271
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   272
fun (scan1 -- scan2) cs =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   273
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   274
    val (x, cs') = scan1 cs;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   275
    val (y, cs'') = scan2 cs';
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   276
  in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   277
    ((x, y), cs'')
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   278
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   279
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   280
fun (scan1 ^^ scan2) = scan1 -- scan2 >> op ^;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   281
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   282
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   283
(* generic scanners *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   284
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   285
fun $$ _ [] = raise LEXICAL_ERROR
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   286
  | $$ a (c :: cs) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   287
      if a = c then (c, cs) else raise LEXICAL_ERROR;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   288
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   289
fun scan_empty cs = ([], cs);
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   290
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   291
fun scan_one _ [] = raise LEXICAL_ERROR
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   292
  | scan_one pred (c :: cs) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   293
      if pred c then (c, cs) else raise LEXICAL_ERROR;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   294
164
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
   295
fun scan_any _ [] = ([], [])
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
   296
  | scan_any pred (chs as c :: cs) =
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
   297
      if pred c then apfst (cons c) (scan_any pred cs)
43506f0a98ae added SCANNER;
wenzelm
parents: 18
diff changeset
   298
      else ([], chs);
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   299
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   300
fun scan_any1 pred = scan_one pred -- scan_any pred >> op ::;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   301
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   302
fun scan_rest cs = (cs, []);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   303
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   304
fun scan_end [] = ([], [])
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   305
  | scan_end _ = raise LEXICAL_ERROR;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   306
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   307
fun optional scan = scan >> Some || scan_empty >> K None;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   308
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   309
fun repeat scan cs = (scan -- repeat scan >> op :: || scan_empty) cs;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   310
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   311
fun repeat1 scan = scan -- repeat scan >> op ::;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   312
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   313
fun max_of scan1 scan2 cs =
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   314
  (case (optional scan1 cs, optional scan2 cs) of
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   315
    (tok1, (None, _)) => tok1
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   316
  | ((None, _), tok2) => tok2
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   317
  | (tok1 as (Some (_, s1), _), tok2 as (Some (_, s2), _)) =>
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   318
      if size s1 >= size s2 then tok1 else tok2);
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   319
18
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
(* other scanners *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   322
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   323
val scan_letter_letdigs = scan_one is_letter -- scan_any is_letdig >> op ::;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   324
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   325
val scan_digits1 = scan_any1 is_digit;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   326
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   327
val scan_id = scan_letter_letdigs >> implode;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   328
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   329
val scan_id_nat =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   330
  scan_id ^^ ($$ "." ^^ (scan_digits1 >> implode) || scan_empty >> K "");
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   331
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   332
val scan_tid = $$ "'" ^^ scan_id;
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   333
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   334
val scan_nat = scan_digits1 >> implode;
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   335
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   336
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   337
(* scan_literal *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   338
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   339
fun scan_literal lex chrs =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   340
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   341
    fun scan_lit _ s_cs [] = s_cs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   342
      | scan_lit Empty s_cs _ = s_cs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   343
      | scan_lit (Branch (d, a, lt, eq, gt)) s_cs (chs as c :: cs) =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   344
          if c < d then scan_lit lt s_cs chs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   345
          else if c > d then scan_lit gt s_cs chs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   346
          else scan_lit eq (if a = no_token then s_cs else Some (a, cs)) cs;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   347
  in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   348
    (case scan_lit lex None chrs of
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   349
      None => raise LEXICAL_ERROR
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   350
    | Some s_cs => s_cs)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   351
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   352
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   353
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   354
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   355
(** tokenize **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   356
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   357
fun tokenize lex xids str =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   358
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   359
    val scan_xid =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   360
      if xids then $$ "_" ^^ scan_id || scan_id
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   361
      else scan_id;
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
    val scan_lit = scan_literal lex >> pair Token;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   364
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   365
    val scan_ident =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   366
      $$ "?" ^^ $$ "'" ^^ scan_id_nat >> pair TVarSy ||
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   367
      $$ "?" ^^ scan_id_nat >> pair VarSy ||
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   368
      $$ "'" ^^ scan_id >> pair TFreeSy ||
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   369
      scan_xid >> pair IdentSy;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   370
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   371
    fun scan_tokens [] rev_toks = rev (EndToken :: rev_toks)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   372
      | scan_tokens (chs as c :: cs) rev_toks =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   373
          if is_blank c then scan_tokens cs rev_toks
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   374
          else
376
d3d01131470f extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents: 330
diff changeset
   375
            (case max_of scan_lit scan_ident chs of
18
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   376
              (None, _) => error ("Lexical error at: " ^ quote (implode chs))
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   377
            | (Some (tk, s), chs') => scan_tokens chs' (tk s :: rev_toks));
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   378
  in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   379
    scan_tokens (explode str) []
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   380
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   381
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   382
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   383
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   384
(** scan variables **)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   385
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   386
(* scan_vname *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   387
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   388
fun scan_vname chrs =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   389
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   390
    fun nat_of_chs n [] = n
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   391
      | nat_of_chs n (c :: cs) = nat_of_chs (n * 10 + (ord c - ord "0")) cs;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   392
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   393
    val nat_of = nat_of_chs 0;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   394
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   395
    fun split_vname chs =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   396
      let val (cs, ds) = take_suffix is_digit chs
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   397
      in (implode cs, nat_of ds) end
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   398
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   399
    val scan =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   400
      scan_letter_letdigs --
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   401
        ($$ "." -- scan_digits1 >> (nat_of o #2) || scan_empty >> K ~1);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   402
  in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   403
    (case scan chrs of
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   404
      ((cs, ~1), cs') => (split_vname cs, cs')
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   405
    | ((cs, i), cs') => ((implode cs, i), cs'))
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   406
  end;
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   407
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   408
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   409
(* scan_varname *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   410
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   411
fun scan_varname chs =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   412
  scan_vname chs handle LEXICAL_ERROR
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   413
    => error ("scan_varname: bad varname " ^ quote (implode chs));
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   414
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   415
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   416
(* scan_var *)
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   417
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   418
fun scan_var str =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   419
  let
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   420
    fun tvar (x, i) = Var (("'" ^ x, i), dummyT);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   421
    fun var x_i = Var (x_i, dummyT);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   422
    fun free x = Free (x, dummyT);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   423
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   424
    val scan =
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   425
      $$ "?" -- $$ "'" -- scan_vname -- scan_end >> (tvar o #2 o #1) ||
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   426
      $$ "?" -- scan_vname -- scan_end >> (var o #2 o #1) ||
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   427
      scan_rest >> (free o implode);
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   428
  in
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   429
    #1 (scan (explode str))
c9ec452ff08f lots of internal cleaning and tuning;
wenzelm
parents: 0
diff changeset
   430
  end;
0
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   431
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   432
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   433
end;
a5a9c433f639 Initial revision
clasohm
parents:
diff changeset
   434