src/Pure/ML/ml_lex.ML
author paulson
Fri, 05 Oct 2007 09:59:03 +0200
changeset 24854 0ebcd575d3c6
parent 24596 f1333a841b26
child 27732 8dbf5761a24a
permissions -rw-r--r--
filtering out some package theorems
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     1
(*  Title:      Pure/ML/ml_lex.ML
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     2
    ID:         $Id$
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     3
    Author:     Makarius
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     4
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     5
Lexical syntax for SML.
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     6
*)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     7
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     8
signature ML_LEX =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
     9
sig
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    10
  datatype token_kind =
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    11
    Keyword | Ident | LongIdent | TypeVar | Word | Int | Real | Char | String |
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    12
    Space | Comment | Error of string | EOF
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    13
  eqtype token
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    14
  val stopper: token * (token -> bool)
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    15
  val is_regular: token -> bool
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    16
  val is_improper: token -> bool
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    17
  val pos_of: token -> string
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    18
  val kind_of: token -> token_kind
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    19
  val val_of: token -> string
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    20
  val keywords: string list
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    21
  val source: (Symbol.symbol, 'a) Source.source ->
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    22
    (token, int * (Symbol.symbol, 'a) Source.source) Source.source
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    23
end;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    24
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    25
structure ML_Lex: ML_LEX =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    26
struct
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    27
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    28
(** tokens **)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    29
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    30
(* datatype token *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    31
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    32
datatype token_kind =
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    33
  Keyword | Ident | LongIdent | TypeVar | Word | Int | Real | Char | String |
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    34
  Space | Comment | Error of string | EOF;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    35
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    36
datatype token = Token of int * (token_kind * string);
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    37
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    38
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    39
(* control tokens *)
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    40
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    41
val eof = Token (0, (EOF, ""));
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    42
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    43
fun is_eof (Token (_, (EOF, _))) = true
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    44
  | is_eof _ = false;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    45
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    46
val stopper = (eof, is_eof);
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    47
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    48
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    49
fun is_regular (Token (_, (Error _, _))) = false
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    50
  | is_regular (Token (_, (EOF, _))) = false
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    51
  | is_regular _ = true;
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    52
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    53
fun is_improper (Token (_, (Space, _))) = true
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    54
  | is_improper (Token (_, (Comment, _))) = true
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    55
  | is_improper _ = false;
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    56
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    57
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    58
(* token content *)
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    59
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    60
fun pos_of (Token (n, _)) = " (line " ^ string_of_int n ^ ")";
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    61
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    62
fun kind_of (Token (_, (k, _))) = k;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    63
fun val_of (Token (_, (_, x))) = x;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    64
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    65
fun token_leq (Token (_, (_, x)), Token (_, (_, x'))) = x <= x';
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    66
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    67
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    68
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    69
(** scanners **)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    70
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    71
(* diagnostics *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    72
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    73
fun lex_err msg ((n, cs), _) = "SML lexical error (line " ^ string_of_int n ^ "): " ^ msg cs;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    74
fun !!! msg scan = Scan.!! (lex_err (K msg)) scan;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    75
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    76
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    77
(* line numbering *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    78
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
    79
fun incr_line scan = Scan.depend (fn (n: int) => scan >> pair (n + 1));
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    80
val keep_line = Scan.lift;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    81
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    82
val scan_blank =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    83
  incr_line ($$ "\n") ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    84
  keep_line (Scan.one Symbol.is_ascii_blank);
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    85
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    86
val scan_blanks = Scan.repeat scan_blank >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    87
val scan_blanks1 = Scan.repeat1 scan_blank >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    88
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    89
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    90
(* keywords *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    91
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    92
val keywords = ["#", "(", ")", ",", "->", "...", ":", ":>", ";", "=",
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    93
  "=>", "[", "]", "_", "{", "|", "}", "abstype", "and", "andalso", "as",
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    94
  "case", "datatype", "do", "else", "end", "eqtype", "exception", "fn",
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    95
  "fun", "functor", "handle", "if", "in", "include", "infix", "infixr",
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    96
  "let", "local", "nonfix", "of", "op", "open", "orelse", "raise", "rec",
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    97
  "sharing", "sig", "signature", "struct", "structure", "then", "type",
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    98
  "val", "where", "while", "with", "withtype"];
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
    99
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   100
val scan_keyword = Scan.literal (Scan.make_lexicon (map explode keywords)) >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   101
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   102
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   103
(* identifiers *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   104
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   105
local
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   106
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   107
val scan_letdigs =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   108
  Scan.many (Symbol.is_ascii_letter orf Symbol.is_ascii_digit orf Symbol.is_ascii_quasi) >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   109
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   110
val scan_alphanumeric = Scan.one Symbol.is_ascii_letter ^^ scan_letdigs;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   111
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   112
val scan_symbolic = Scan.many1 (member (op =) (explode "!#$%&*+-/:<=>?@\\^`|~")) >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   113
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   114
in
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   115
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   116
val scan_ident = scan_alphanumeric || scan_symbolic;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   117
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   118
val scan_longident =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   119
  (Scan.repeat1 (scan_alphanumeric ^^ $$ ".") >> implode) ^^ (scan_ident || $$ "=");
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   120
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   121
val scan_typevar = $$ "'" ^^ scan_letdigs;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   122
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   123
end;
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   124
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   125
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   126
(* numerals *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   127
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   128
local
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   129
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   130
val scan_dec = Scan.many1 Symbol.is_ascii_digit >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   131
val scan_hex = Scan.many1 Symbol.is_ascii_hex >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   132
val scan_sign = Scan.optional ($$ "~") "";
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   133
val scan_decint = scan_sign ^^ scan_dec;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   134
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   135
in
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   136
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   137
val scan_word = Scan.this_string "0wx" ^^ scan_hex || Scan.this_string "0w" ^^ scan_dec;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   138
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   139
val scan_int = scan_sign ^^ (Scan.this_string "0x" ^^ scan_hex || scan_dec);
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   140
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   141
val scan_exp = ($$ "E" || $$ "e") ^^ scan_decint;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   142
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   143
val scan_real =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   144
  scan_decint ^^ $$ "." ^^ scan_dec ^^ Scan.optional scan_exp "" ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   145
  scan_decint ^^ scan_exp;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   146
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   147
end;
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   148
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   149
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   150
(* chars and strings *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   151
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   152
local
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   153
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   154
val scan_escape =
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   155
  Scan.one (member (op =) (explode "\"\\abtnvfr")) ||
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   156
  $$ "^" ^^ Scan.one (fn s => ord "@" <= ord s andalso ord s <= ord "_") ||
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   157
  Scan.one Symbol.is_ascii_digit ^^
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   158
    Scan.one Symbol.is_ascii_digit ^^
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   159
    Scan.one Symbol.is_ascii_digit;
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   160
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   161
val scan_str =
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   162
  keep_line (Scan.one (fn s => Symbol.is_printable s andalso s <> "\"" andalso s <> "\\")) ||
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   163
  keep_line ($$ "\\") ^^ !!! "bad escape character in string" (keep_line scan_escape);
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   164
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   165
val scan_gap = keep_line ($$ "\\") ^^ scan_blanks1 ^^ keep_line ($$ "\\");
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   166
val scan_gaps = Scan.repeat scan_gap >> implode;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   167
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   168
in
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   169
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   170
val scan_char =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   171
  keep_line ($$ "#" ^^ $$ "\"") ^^ scan_gaps ^^ scan_str ^^ scan_gaps ^^ keep_line ($$ "\"");
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   172
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   173
val scan_string =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   174
  keep_line ($$ "\"") ^^
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   175
  !!! "missing quote at end of string"
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   176
    ((Scan.repeat (scan_gap || scan_str) >> implode) ^^ keep_line ($$ "\""));
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   177
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   178
end;
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   179
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   180
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   181
(* scan nested comments *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   182
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   183
local
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   184
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   185
val scan_cmt =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   186
  Scan.lift scan_blank ||
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   187
  Scan.depend (fn (d: int) => keep_line ($$ "(" ^^ $$ "*") >> pair (d + 1)) ||
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   188
  Scan.depend (fn 0 => Scan.fail | d => keep_line ($$ "*" ^^ $$ ")") >> pair (d - 1)) ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   189
  Scan.lift (keep_line ($$ "*" --| Scan.ahead (~$$ ")"))) ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   190
  Scan.lift (keep_line (Scan.one (fn s => s <> "*" andalso Symbol.is_regular s)));
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   191
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   192
in
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   193
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   194
val scan_comment =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   195
  keep_line ($$ "(" ^^ $$ "*") ^^
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   196
  !!! "missing end of comment"
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   197
    (Scan.pass 0 (Scan.repeat scan_cmt >> implode) ^^ keep_line ($$ "*" ^^ $$ ")"));
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   198
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   199
end;
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   200
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   201
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   202
(* scan token *)
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   203
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   204
val scan =
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   205
  let
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   206
    val scanner = Scan.state :|-- (fn n =>
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   207
      let
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   208
        fun token k x = Token (n, (k, x));
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   209
      in
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   210
        scan_char >> token Char ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   211
        scan_string >> token String ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   212
        scan_blanks1 >> token Space ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   213
        scan_comment >> token Comment ||
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   214
        keep_line (Scan.max token_leq
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   215
          (scan_keyword >> token Keyword)
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   216
          (scan_word >> token Word ||
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   217
            scan_real >> token Real ||
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   218
            scan_int >> token Int ||
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   219
            scan_longident >> token LongIdent ||
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   220
            scan_ident >> token Ident ||
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   221
            scan_typevar >> token TypeVar))
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   222
      end);
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   223
  in !! (lex_err (fn cs => "bad input " ^ quote (Symbol.beginning 10 cs))) scanner end;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   224
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   225
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   226
(* token source *)
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   227
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   228
local
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   229
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   230
val is_junk = (not o Symbol.is_blank) andf Symbol.is_regular;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   231
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   232
fun recover msg = Scan.state :|-- (fn n =>
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   233
  keep_line (Scan.many is_junk) >> (fn cs => [Token (n, (Error msg, implode cs))]));
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   234
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   235
in
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   236
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   237
fun source src =
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   238
  Source.source' 1 Symbol.stopper (Scan.bulk scan) (SOME (false, recover)) src;
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   239
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   240
end;
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   241
24596
f1333a841b26 removed obsolete Selector token;
wenzelm
parents: 24579
diff changeset
   242
end;
24579
852fc50927b1 Lexical syntax for SML.
wenzelm
parents:
diff changeset
   243