src/Pure/General/symbol.ML
author wenzelm
Thu, 01 Sep 2016 16:05:22 +0200
changeset 63750 9c8a366778e1
parent 62877 741560a5283b
child 63936 b87784e19a77
permissions -rw-r--r--
more robust persistent storage; tuned;
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
6118
caa439435666 fixed titles;
wenzelm
parents: 6116
diff changeset
     1
(*  Title:      Pure/General/symbol.ML
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
     2
    Author:     Markus Wenzel, TU Muenchen
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
     3
21897
d0c67d715deb fixed typo in comment
paulson
parents: 21858
diff changeset
     4
Generalized characters with infinitely many named symbols.
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
     5
*)
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
     6
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
     7
signature SYMBOL =
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
     8
sig
40509
0bc83ae22789 tuned signatures;
wenzelm
parents: 40131
diff changeset
     9
  type symbol = string
61865
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
    10
  val spaces: int -> string
26524
63953bec4c98 added some ASCII control symbols;
wenzelm
parents: 25644
diff changeset
    11
  val STX: symbol
63953bec4c98 added some ASCII control symbols;
wenzelm
parents: 25644
diff changeset
    12
  val DEL: symbol
62210
e068ea693678 tuned signature (according to Scala version);
wenzelm
parents: 61865
diff changeset
    13
  val open_: symbol
e068ea693678 tuned signature (according to Scala version);
wenzelm
parents: 61865
diff changeset
    14
  val close: symbol
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    15
  val space: symbol
62804
7b9c5416f30e tuned signature;
wenzelm
parents: 62529
diff changeset
    16
  val is_space: symbol -> bool
61579
634cd44bb1d3 symbolic syntax "\<comment> text";
wenzelm
parents: 61475
diff changeset
    17
  val comment: symbol
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    18
  val is_char: symbol -> bool
37533
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
    19
  val is_utf8: symbol -> bool
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    20
  val is_symbolic: symbol -> bool
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
    21
  val is_symbolic_char: symbol -> bool
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    22
  val is_printable: symbol -> bool
61470
c42960228a81 clarified Symbol.is_control;
wenzelm
parents: 61435
diff changeset
    23
  val is_control: symbol -> bool
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    24
  val eof: symbol
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    25
  val is_eof: symbol -> bool
27766
1ae745357856 export not_eof;
wenzelm
parents: 27745
diff changeset
    26
  val not_eof: symbol -> bool
27732
8dbf5761a24a abstract type Scan.stopper;
wenzelm
parents: 26632
diff changeset
    27
  val stopper: symbol Scan.stopper
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
    28
  val is_malformed: symbol -> bool
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
    29
  val malformed_msg: symbol -> string
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    30
  val is_ascii: symbol -> bool
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    31
  val is_ascii_letter: symbol -> bool
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    32
  val is_ascii_digit: symbol -> bool
24580
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
    33
  val is_ascii_hex: symbol -> bool
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    34
  val is_ascii_quasi: symbol -> bool
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    35
  val is_ascii_blank: symbol -> bool
34095
c2f176a38448 robust representation of low ASCII control characters within XML/YXML text;
wenzelm
parents: 33955
diff changeset
    36
  val is_ascii_control: symbol -> bool
50236
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
    37
  val is_ascii_letdig: symbol -> bool
20200
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
    38
  val is_ascii_lower: symbol -> bool
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
    39
  val is_ascii_upper: symbol -> bool
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
    40
  val to_ascii_lower: symbol -> symbol
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
    41
  val to_ascii_upper: symbol -> symbol
50238
98d35a7368bd more uniform Symbol.is_ascii_identifier in ML/Scala;
wenzelm
parents: 50237
diff changeset
    42
  val is_ascii_identifier: string -> bool
50236
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
    43
  val scan_ascii_id: string list -> string * string list
14834
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
    44
  val is_raw: symbol -> bool
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
    45
  val decode_raw: symbol -> string
14977
77d88064991a added escape, export encode_raw, default mode now trivial, tuned;
wenzelm
parents: 14961
diff changeset
    46
  val encode_raw: string -> string
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
    47
  datatype sym =
61475
5b58a17c440a tuned signature;
wenzelm
parents: 61470
diff changeset
    48
    Char of string | UTF8 of string | Sym of string | Control of string | Raw of string |
43485
33a24212a72d more tolerant Symbol.decode;
wenzelm
parents: 43418
diff changeset
    49
    Malformed of string | EOF
14873
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
    50
  val decode: symbol -> sym
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    51
  datatype kind = Letter | Digit | Quasi | Blank | Other
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    52
  val kind: symbol -> kind
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    53
  val is_letter: symbol -> bool
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    54
  val is_digit: symbol -> bool
12904
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
    55
  val is_quasi: symbol -> bool
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    56
  val is_blank: symbol -> bool
47850
c638127b4653 avoid interference of markup for literal tokens, which may contain slightly odd \<^bsub> \<^esub> counted as pseudo-markup (especially relevant for HTML output, e.g. of thm power3_eq_cube);
wenzelm
parents: 43947
diff changeset
    57
  val is_block_ctrl: symbol -> bool
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    58
  val is_quasi_letter: symbol -> bool
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    59
  val is_letdig: symbol -> bool
14728
wenzelm
parents: 14678
diff changeset
    60
  val beginning: int -> symbol list -> string
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
    61
  val source: (string, 'a) Source.source -> (symbol, (string, 'a) Source.source) Source.source
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    62
  val explode: string -> symbol list
50237
wenzelm
parents: 50236
diff changeset
    63
  val esc: symbol -> string
wenzelm
parents: 50236
diff changeset
    64
  val escape: string -> string
wenzelm
parents: 50236
diff changeset
    65
  val scanner: string -> (string list -> 'a * string list) -> symbol list -> 'a
50162
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
    66
  val split_words: symbol list -> string list
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
    67
  val explode_words: string -> string list
61435
636bb75e7683 tuned signature (cf. XML.trim_blanks);
wenzelm
parents: 58864
diff changeset
    68
  val trim_blanks: string -> string
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    69
  val bump_init: string -> string
12904
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
    70
  val bump_string: string -> string
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    71
  val length: symbol list -> int
6692
05c56f41e661 avoid string constants;
wenzelm
parents: 6640
diff changeset
    72
  val xsymbolsN: string
40131
7cbebd636e79 explicitly qualify type Output.output, which is a slightly odd internal feature;
wenzelm
parents: 37728
diff changeset
    73
  val output: string -> Output.output * int
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    74
end;
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    75
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    76
structure Symbol: SYMBOL =
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    77
struct
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
    78
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    79
(** type symbol **)
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    80
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    81
(*Symbols, which are considered the smallest entities of any Isabelle
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    82
  string, may be of the following form:
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    83
14834
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
    84
    (1) ASCII symbols: a
17823
4735c07399c8 raw symbols: allow non-ASCII chars (e.g. UTF-8);
wenzelm
parents: 17756
diff changeset
    85
    (2) regular symbols: \<ident>
14834
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
    86
    (3) control symbols: \<^ident>
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
    87
    (4) raw control symbols: \<^raw:...>, where "..." may be any printable
20205
7b2958d3d575 raw symbols: disallow dot to avoid confusion in NameSpace.unpack;
wenzelm
parents: 20200
diff changeset
    88
        character (excluding ".", ">"), or \<^raw000>
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    89
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    90
  Output is subject to the print_mode variable (default: verbatim),
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
    91
  actual interpretation in display is up to front-end tools.
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    92
*)
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    93
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    94
type symbol = string;
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
    95
26524
63953bec4c98 added some ASCII control symbols;
wenzelm
parents: 25644
diff changeset
    96
val STX = chr 2;
63953bec4c98 added some ASCII control symbols;
wenzelm
parents: 25644
diff changeset
    97
val DEL = chr 127;
63953bec4c98 added some ASCII control symbols;
wenzelm
parents: 25644
diff changeset
    98
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
    99
val open_ = "\<open>";
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   100
val close = "\<close>";
62210
e068ea693678 tuned signature (according to Scala version);
wenzelm
parents: 61865
diff changeset
   101
26524
63953bec4c98 added some ASCII control symbols;
wenzelm
parents: 25644
diff changeset
   102
val space = chr 32;
62804
7b9c5416f30e tuned signature;
wenzelm
parents: 62529
diff changeset
   103
fun is_space s = s = space;
17063
502105583951 tuned Symbol.spaces;
wenzelm
parents: 16138
diff changeset
   104
61865
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   105
local
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   106
  val small_spaces = Vector.tabulate (65, fn i => replicate_string i space);
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   107
in
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   108
  fun spaces n =
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   109
    if n < 64 then Vector.sub (small_spaces, n)
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   110
    else
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   111
      replicate_string (n div 64) (Vector.sub (small_spaces, 64)) ^
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   112
        Vector.sub (small_spaces, n mod 64);
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   113
end;
6dcc9e4f1aa6 tuned signature;
wenzelm
parents: 61707
diff changeset
   114
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   115
val comment = "\<comment>";
61579
634cd44bb1d3 symbolic syntax "\<comment> text";
wenzelm
parents: 61475
diff changeset
   116
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   117
fun is_char s = size s = 1;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   118
37533
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   119
fun is_utf8 s = size s > 0 andalso forall_string (fn c => ord c >= 128) s;
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   120
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
   121
fun raw_symbolic s =
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   122
  String.isPrefix "\092<" s andalso String.isSuffix ">" s andalso not (String.isPrefix "\092<^" s);
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
   123
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   124
fun is_symbolic s =
62210
e068ea693678 tuned signature (according to Scala version);
wenzelm
parents: 61865
diff changeset
   125
  s <> open_ andalso s <> close andalso raw_symbolic s;
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
   126
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
   127
val is_symbolic_char = member (op =) (raw_explode "!#$%&*+-/<=>?@^_|~");
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   128
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   129
fun is_printable s =
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   130
  if is_char s then ord space <= ord s andalso ord s <= ord "~"
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
   131
  else is_utf8 s orelse raw_symbolic s;
26632
90c0b075c0d3 added is_utf8_trailer;
wenzelm
parents: 26538
diff changeset
   132
61470
c42960228a81 clarified Symbol.is_control;
wenzelm
parents: 61435
diff changeset
   133
fun is_control s =
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   134
  String.isPrefix "\092<^" s andalso String.isSuffix ">" s;
61470
c42960228a81 clarified Symbol.is_control;
wenzelm
parents: 61435
diff changeset
   135
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   136
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   137
(* input source control *)
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   138
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   139
val eof = "";
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   140
fun is_eof s = s = eof;
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   141
fun not_eof s = s <> eof;
27732
8dbf5761a24a abstract type Scan.stopper;
wenzelm
parents: 26632
diff changeset
   142
val stopper = Scan.stopper (K eof) is_eof;
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   143
48774
c4bd5bb3ae69 further clarification of malformed symbols;
wenzelm
parents: 48773
diff changeset
   144
fun is_malformed s =
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   145
  String.isPrefix "\092<" s andalso not (String.isSuffix ">" s)
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   146
  orelse s = "\092<>" orelse s = "\092<^>";
48774
c4bd5bb3ae69 further clarification of malformed symbols;
wenzelm
parents: 48773
diff changeset
   147
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
   148
fun malformed_msg s = "Malformed symbolic character: " ^ quote s;
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   149
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   150
43418
c69e9fbb81a8 recovered markup for non-alphabetic keywords;
wenzelm
parents: 40627
diff changeset
   151
(* ASCII symbols *)
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   152
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   153
fun is_ascii s = is_char s andalso ord s < 128;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   154
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   155
fun is_ascii_letter s =
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   156
  is_char s andalso
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   157
   (ord "A" <= ord s andalso ord s <= ord "Z" orelse
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   158
    ord "a" <= ord s andalso ord s <= ord "z");
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   159
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   160
fun is_ascii_digit s =
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   161
  is_char s andalso ord "0" <= ord s andalso ord s <= ord "9";
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   162
24580
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
   163
fun is_ascii_hex s =
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
   164
  is_char s andalso
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
   165
   (ord "0" <= ord s andalso ord s <= ord "9" orelse
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
   166
    ord "A" <= ord s andalso ord s <= ord "F" orelse
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
   167
    ord "a" <= ord s andalso ord s <= ord "f");
916259859344 replaced Symbol.is_hex_letter to Symbol.is_ascii_hex;
wenzelm
parents: 23784
diff changeset
   168
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   169
fun is_ascii_quasi "_" = true
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   170
  | is_ascii_quasi "'" = true
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   171
  | is_ascii_quasi _ = false;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   172
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   173
val is_ascii_blank =
43845
d89353d17f54 added File.fold_pages for streaming of large files;
wenzelm
parents: 43777
diff changeset
   174
  fn " " => true | "\t" => true | "\n" => true | "\^K" => true | "\f" => true | "\^M" => true
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   175
    | _ => false;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   176
34095
c2f176a38448 robust representation of low ASCII control characters within XML/YXML text;
wenzelm
parents: 33955
diff changeset
   177
fun is_ascii_control s = is_char s andalso ord s < 32 andalso not (is_ascii_blank s);
c2f176a38448 robust representation of low ASCII control characters within XML/YXML text;
wenzelm
parents: 33955
diff changeset
   178
50236
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
   179
fun is_ascii_letdig s = is_ascii_letter s orelse is_ascii_digit s orelse is_ascii_quasi s;
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
   180
20200
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
   181
fun is_ascii_lower s = is_char s andalso (ord "a" <= ord s andalso ord s <= ord "z");
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
   182
fun is_ascii_upper s = is_char s andalso (ord "A" <= ord s andalso ord s <= ord "Z");
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
   183
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
   184
fun to_ascii_lower s = if is_ascii_upper s then chr (ord s + ord "a" - ord "A") else s;
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
   185
fun to_ascii_upper s = if is_ascii_lower s then chr (ord s + ord "A" - ord "a") else s;
c6fb50dbbc30 added is/to_ascii_lower/upper;
wenzelm
parents: 20192
diff changeset
   186
50238
98d35a7368bd more uniform Symbol.is_ascii_identifier in ML/Scala;
wenzelm
parents: 50237
diff changeset
   187
fun is_ascii_identifier s =
98d35a7368bd more uniform Symbol.is_ascii_identifier in ML/Scala;
wenzelm
parents: 50237
diff changeset
   188
  size s > 0 andalso is_ascii_letter (String.substring (s, 0, 1)) andalso
98d35a7368bd more uniform Symbol.is_ascii_identifier in ML/Scala;
wenzelm
parents: 50237
diff changeset
   189
  forall_string is_ascii_letdig s;
98d35a7368bd more uniform Symbol.is_ascii_identifier in ML/Scala;
wenzelm
parents: 50237
diff changeset
   190
50236
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
   191
val scan_ascii_id = Scan.one is_ascii_letter ^^ (Scan.many is_ascii_letdig >> implode);
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
   192
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   193
14956
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   194
(* encode_raw *)
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   195
20205
7b2958d3d575 raw symbols: disallow dot to avoid confusion in NameSpace.unpack;
wenzelm
parents: 20200
diff changeset
   196
fun raw_chr c =
48773
0e1bab274672 more liberal scanning of potentially malformed symbols;
wenzelm
parents: 48704
diff changeset
   197
  is_char c andalso
0e1bab274672 more liberal scanning of potentially malformed symbols;
wenzelm
parents: 48704
diff changeset
   198
    (ord space <= ord c andalso ord c <= ord "~" andalso c <> "." andalso c <> ">"
0e1bab274672 more liberal scanning of potentially malformed symbols;
wenzelm
parents: 48704
diff changeset
   199
      orelse ord c >= 128);
14956
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   200
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   201
fun encode_raw "" = ""
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   202
  | encode_raw str =
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   203
      let
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   204
        val raw0 = enclose "\092<^raw:" ">";
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   205
        val raw1 = raw0 o implode;
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   206
        val raw2 = enclose "\092<^raw" ">" o string_of_int o ord;
50162
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   207
33955
fff6f11b1f09 curried take/drop
haftmann
parents: 33548
diff changeset
   208
        fun encode cs = enc (take_prefix raw_chr cs)
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   209
        and enc ([], []) = []
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   210
          | enc (cs, []) = [raw1 cs]
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   211
          | enc ([], d :: ds) = raw2 d :: encode ds
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   212
          | enc (cs, d :: ds) = raw1 cs :: raw2 d :: encode ds;
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   213
      in
40627
becf5d5187cc renamed raw "explode" function to "raw_explode" to emphasize its meaning;
wenzelm
parents: 40523
diff changeset
   214
        if exists_string (not o raw_chr) str then implode (encode (raw_explode str))
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   215
        else raw0 str
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   216
      end;
14956
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   217
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   218
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   219
(* diagnostics *)
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   220
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   221
fun beginning n cs =
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   222
  let
33955
fff6f11b1f09 curried take/drop
haftmann
parents: 33548
diff changeset
   223
    val drop_blanks = #1 o take_suffix is_ascii_blank;
14956
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   224
    val all_cs = drop_blanks cs;
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   225
    val dots = if length all_cs > n then " ..." else "";
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   226
  in
33955
fff6f11b1f09 curried take/drop
haftmann
parents: 33548
diff changeset
   227
    (drop_blanks (take n all_cs)
14956
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   228
      |> map (fn c => if is_ascii_blank c then space else c)
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   229
      |> implode) ^ dots
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   230
  end;
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   231
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   232
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   233
(* decode_raw *)
14834
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
   234
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
   235
fun is_raw s =
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   236
  String.isPrefix "\092<^raw" s andalso String.isSuffix ">" s;
14834
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
   237
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
   238
fun decode_raw s =
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   239
  if not (is_raw s) then error (malformed_msg s)
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   240
  else if String.isPrefix "\092<^raw:" s then String.substring (s, 7, size s - 8)
40627
becf5d5187cc renamed raw "explode" function to "raw_explode" to emphasize its meaning;
wenzelm
parents: 40523
diff changeset
   241
  else chr (#1 (Library.read_int (raw_explode (String.substring (s, 6, size s - 7)))));
14834
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
   242
e47744683560 improved support for raw symbols;
wenzelm
parents: 14728
diff changeset
   243
14873
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   244
(* symbol variants *)
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   245
37533
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   246
datatype sym =
61475
5b58a17c440a tuned signature;
wenzelm
parents: 61470
diff changeset
   247
  Char of string | UTF8 of string | Sym of string | Control of string | Raw of string |
43485
33a24212a72d more tolerant Symbol.decode;
wenzelm
parents: 43418
diff changeset
   248
  Malformed of string | EOF;
14873
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   249
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   250
fun decode s =
43485
33a24212a72d more tolerant Symbol.decode;
wenzelm
parents: 43418
diff changeset
   251
  if s = "" then EOF
33a24212a72d more tolerant Symbol.decode;
wenzelm
parents: 43418
diff changeset
   252
  else if is_char s then Char s
37533
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   253
  else if is_utf8 s then UTF8 s
14873
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   254
  else if is_raw s then Raw (decode_raw s)
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
   255
  else if is_malformed s then Malformed s
61475
5b58a17c440a tuned signature;
wenzelm
parents: 61470
diff changeset
   256
  else if is_control s then Control (String.substring (s, 3, size s - 4))
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
   257
  else Sym (String.substring (s, 2, size s - 3));
14873
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   258
7efc14398e82 added datatype sym, val decode: symbol -> sym;
wenzelm
parents: 14834
diff changeset
   259
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   260
(* standard symbol kinds *)
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   261
14171
0cab06e3bbd0 Extended the notion of letter and digit, such that now one may use greek,
skalberg
parents: 13730
diff changeset
   262
local
50235
wenzelm
parents: 50233
diff changeset
   263
  val letter_symbols =
wenzelm
parents: 50233
diff changeset
   264
    Symtab.make_set [
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   265
      "\<A>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   266
      "\<B>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   267
      "\<C>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   268
      "\<D>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   269
      "\<E>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   270
      "\<F>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   271
      "\<G>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   272
      "\<H>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   273
      "\<I>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   274
      "\<J>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   275
      "\<K>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   276
      "\<L>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   277
      "\<M>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   278
      "\<N>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   279
      "\<O>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   280
      "\<P>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   281
      "\<Q>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   282
      "\<R>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   283
      "\<S>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   284
      "\<T>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   285
      "\<U>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   286
      "\<V>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   287
      "\<W>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   288
      "\<X>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   289
      "\<Y>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   290
      "\<Z>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   291
      "\<a>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   292
      "\<b>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   293
      "\<c>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   294
      "\<d>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   295
      "\<e>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   296
      "\<f>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   297
      "\<g>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   298
      "\<h>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   299
      "\<i>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   300
      "\<j>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   301
      "\<k>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   302
      "\<l>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   303
      "\<m>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   304
      "\<n>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   305
      "\<o>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   306
      "\<p>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   307
      "\<q>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   308
      "\<r>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   309
      "\<s>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   310
      "\<t>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   311
      "\<u>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   312
      "\<v>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   313
      "\<w>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   314
      "\<x>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   315
      "\<y>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   316
      "\<z>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   317
      "\<AA>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   318
      "\<BB>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   319
      "\<CC>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   320
      "\<DD>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   321
      "\<EE>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   322
      "\<FF>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   323
      "\<GG>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   324
      "\<HH>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   325
      "\<II>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   326
      "\<JJ>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   327
      "\<KK>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   328
      "\<LL>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   329
      "\<MM>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   330
      "\<NN>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   331
      "\<OO>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   332
      "\<PP>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   333
      "\<QQ>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   334
      "\<RR>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   335
      "\<SS>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   336
      "\<TT>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   337
      "\<UU>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   338
      "\<VV>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   339
      "\<WW>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   340
      "\<XX>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   341
      "\<YY>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   342
      "\<ZZ>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   343
      "\<aa>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   344
      "\<bb>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   345
      "\<cc>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   346
      "\<dd>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   347
      "\<ee>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   348
      "\<ff>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   349
      "\<gg>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   350
      "\<hh>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   351
      "\<ii>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   352
      "\<jj>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   353
      "\<kk>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   354
      "\<ll>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   355
      "\<mm>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   356
      "\<nn>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   357
      "\<oo>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   358
      "\<pp>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   359
      "\<qq>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   360
      "\<rr>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   361
      "\<ss>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   362
      "\<tt>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   363
      "\<uu>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   364
      "\<vv>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   365
      "\<ww>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   366
      "\<xx>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   367
      "\<yy>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   368
      "\<zz>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   369
      "\<alpha>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   370
      "\<beta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   371
      "\<gamma>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   372
      "\<delta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   373
      "\<epsilon>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   374
      "\<zeta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   375
      "\<eta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   376
      "\<theta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   377
      "\<iota>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   378
      "\<kappa>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   379
      (*"\<lambda>", sic!*)
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   380
      "\<mu>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   381
      "\<nu>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   382
      "\<xi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   383
      "\<pi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   384
      "\<rho>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   385
      "\<sigma>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   386
      "\<tau>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   387
      "\<upsilon>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   388
      "\<phi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   389
      "\<chi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   390
      "\<psi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   391
      "\<omega>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   392
      "\<Gamma>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   393
      "\<Delta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   394
      "\<Theta>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   395
      "\<Lambda>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   396
      "\<Xi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   397
      "\<Pi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   398
      "\<Sigma>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   399
      "\<Upsilon>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   400
      "\<Phi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   401
      "\<Psi>",
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   402
      "\<Omega>"
50235
wenzelm
parents: 50233
diff changeset
   403
    ];
14171
0cab06e3bbd0 Extended the notion of letter and digit, such that now one may use greek,
skalberg
parents: 13730
diff changeset
   404
in
50242
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   405
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   406
val is_letter_symbol = Symtab.defined letter_symbols;
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   407
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   408
end;
14173
a3690aeb79d4 Removed the extended digits again.
skalberg
parents: 14171
diff changeset
   409
50242
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   410
datatype kind = Letter | Digit | Quasi | Blank | Other;
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   411
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   412
fun kind s =
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   413
  if is_ascii_letter s then Letter
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   414
  else if is_ascii_digit s then Digit
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   415
  else if is_ascii_quasi s then Quasi
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   416
  else if is_ascii_blank s then Blank
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   417
  else if is_char s then Other
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   418
  else if is_letter_symbol s then Letter
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   419
  else Other;
56b9c792a98b support for sub-structured identifier syntax (inactive);
wenzelm
parents: 50239
diff changeset
   420
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   421
fun is_letter s = kind s = Letter;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   422
fun is_digit s = kind s = Digit;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   423
fun is_quasi s = kind s = Quasi;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   424
fun is_blank s = kind s = Blank;
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   425
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   426
val is_block_ctrl = member (op =) ["\<^bsub>", "\<^esub>", "\<^bsup>", "\<^esup>"];
47850
c638127b4653 avoid interference of markup for literal tokens, which may contain slightly odd \<^bsub> \<^esub> counted as pseudo-markup (especially relevant for HTML output, e.g. of thm power3_eq_cube);
wenzelm
parents: 43947
diff changeset
   427
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   428
fun is_quasi_letter s = let val k = kind s in k = Letter orelse k = Quasi end;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   429
fun is_letdig s = let val k = kind s in k = Letter orelse k = Digit orelse k = Quasi end;
11010
2c6559297be3 added strip_blanks;
wenzelm
parents: 10953
diff changeset
   430
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   431
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   432
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   433
(** symbol input **)
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   434
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   435
(* source *)
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   436
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   437
local
14561
c53396af770e * raw control symbols are of the form \<^raw:...> now.
schirmer
parents: 14559
diff changeset
   438
37533
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   439
fun is_plain s = is_ascii s andalso s <> "\^M" andalso s <> "\\";
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   440
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   441
fun is_utf8_trailer s = is_char s andalso 128 <= ord s andalso ord s < 192;
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   442
37728
5d2b3e827371 implode pseudo utf8, i.e. decode byte-stuffed low ASCII characters;
wenzelm
parents: 37534
diff changeset
   443
fun implode_pseudo_utf8 (cs as ["\192", c]) =
5d2b3e827371 implode pseudo utf8, i.e. decode byte-stuffed low ASCII characters;
wenzelm
parents: 37534
diff changeset
   444
      if ord c < 160 then chr (ord c - 128) else implode cs
5d2b3e827371 implode pseudo utf8, i.e. decode byte-stuffed low ASCII characters;
wenzelm
parents: 37534
diff changeset
   445
  | implode_pseudo_utf8 cs = implode cs;
5d2b3e827371 implode pseudo utf8, i.e. decode byte-stuffed low ASCII characters;
wenzelm
parents: 37534
diff changeset
   446
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   447
val scan_encoded_newline =
17756
d4a35f82fbb4 minor tweaks for Poplog/ML;
wenzelm
parents: 17412
diff changeset
   448
  $$ "\^M" -- $$ "\n" >> K "\n" ||
54734
b91afc3aa3e6 clarified Proof General legacy: special treatment of \<^newline> only in TTY mode;
wenzelm
parents: 54732
diff changeset
   449
  $$ "\^M" >> K "\n";
14956
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   450
70ec4b8aef8d prevent looping of error messages involving malformed symbols;
wenzelm
parents: 14908
diff changeset
   451
val scan_raw =
21858
05f57309170c avoid conflict with Alice keywords: renamed pack -> implode, unpack -> explode, any -> many, avoided assert;
wenzelm
parents: 21495
diff changeset
   452
  Scan.this_string "raw:" ^^ (Scan.many raw_chr >> implode) ||
05f57309170c avoid conflict with Alice keywords: renamed pack -> implode, unpack -> explode, any -> many, avoided assert;
wenzelm
parents: 21495
diff changeset
   453
  Scan.this_string "raw" ^^ (Scan.many1 is_ascii_digit >> implode);
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   454
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
   455
val scan_total =
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   456
  Scan.one is_plain ||
37728
5d2b3e827371 implode pseudo utf8, i.e. decode byte-stuffed low ASCII characters;
wenzelm
parents: 37534
diff changeset
   457
  Scan.one is_utf8 ::: Scan.many is_utf8_trailer >> implode_pseudo_utf8 ||
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   458
  scan_encoded_newline ||
48773
0e1bab274672 more liberal scanning of potentially malformed symbols;
wenzelm
parents: 48704
diff changeset
   459
  ($$ "\\" ^^ $$ "<" ^^
50236
476a3350589c clarified Symbol.scan_ascii_id;
wenzelm
parents: 50235
diff changeset
   460
    (($$ "^" ^^ Scan.optional (scan_raw || scan_ascii_id) "" || Scan.optional scan_ascii_id "") ^^
48774
c4bd5bb3ae69 further clarification of malformed symbols;
wenzelm
parents: 48773
diff changeset
   461
      Scan.optional ($$ ">") "")) ||
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   462
  Scan.one not_eof;
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   463
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   464
in
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   465
58864
505a8150368a recover via scanner;
wenzelm
parents: 58854
diff changeset
   466
fun source src = Source.source stopper (Scan.bulk scan_total) src;
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   467
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   468
end;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   469
14562
980da32f4617 proper handling of lines terminated by CRLF or CR;
wenzelm
parents: 14561
diff changeset
   470
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   471
(* explode *)
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   472
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   473
local
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   474
14562
980da32f4617 proper handling of lines terminated by CRLF or CR;
wenzelm
parents: 14561
diff changeset
   475
fun no_explode [] = true
980da32f4617 proper handling of lines terminated by CRLF or CR;
wenzelm
parents: 14561
diff changeset
   476
  | no_explode ("\\" :: "<" :: _) = false
17756
d4a35f82fbb4 minor tweaks for Poplog/ML;
wenzelm
parents: 17412
diff changeset
   477
  | no_explode ("\^M" :: _) = false
37533
d775bd70f571 explicit treatment of UTF8 character sequences as Isabelle symbols;
wenzelm
parents: 34095
diff changeset
   478
  | no_explode (c :: cs) = is_ascii c andalso no_explode cs;
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   479
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   480
in
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   481
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   482
fun sym_explode str =
40627
becf5d5187cc renamed raw "explode" function to "raw_explode" to emphasize its meaning;
wenzelm
parents: 40523
diff changeset
   483
  let val chs = raw_explode str in
14562
980da32f4617 proper handling of lines terminated by CRLF or CR;
wenzelm
parents: 14561
diff changeset
   484
    if no_explode chs then chs
40523
1050315f6ee2 simplified/robustified treatment of malformed symbols, which are now fully internalized (total Symbol.explode etc.);
wenzelm
parents: 40509
diff changeset
   485
    else Source.exhaust (source (Source.of_list chs))
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   486
  end;
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   487
23676
ea9b7e9c2301 scan: changed treatment of malformed symbols, passed to next stage;
wenzelm
parents: 23618
diff changeset
   488
end;
14994
7d8501843146 added chars_only, symbol_output;
wenzelm
parents: 14981
diff changeset
   489
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   490
50237
wenzelm
parents: 50236
diff changeset
   491
(* escape *)
wenzelm
parents: 50236
diff changeset
   492
wenzelm
parents: 50236
diff changeset
   493
val esc = fn s =>
wenzelm
parents: 50236
diff changeset
   494
  if is_char s then s
wenzelm
parents: 50236
diff changeset
   495
  else if is_utf8 s then translate_string (fn c => "\\" ^ string_of_int (ord c)) s
wenzelm
parents: 50236
diff changeset
   496
  else "\\" ^ s;
wenzelm
parents: 50236
diff changeset
   497
wenzelm
parents: 50236
diff changeset
   498
val escape = implode o map esc o sym_explode;
wenzelm
parents: 50236
diff changeset
   499
wenzelm
parents: 50236
diff changeset
   500
wenzelm
parents: 50236
diff changeset
   501
wenzelm
parents: 50236
diff changeset
   502
(** scanning through symbols **)
wenzelm
parents: 50236
diff changeset
   503
wenzelm
parents: 50236
diff changeset
   504
(* scanner *)
wenzelm
parents: 50236
diff changeset
   505
wenzelm
parents: 50236
diff changeset
   506
fun scanner msg scan syms =
wenzelm
parents: 50236
diff changeset
   507
  let
wenzelm
parents: 50236
diff changeset
   508
    fun message (ss, NONE) = (fn () => msg ^ ": " ^ quote (beginning 10 ss))
wenzelm
parents: 50236
diff changeset
   509
      | message (ss, SOME msg') = (fn () => msg ^ ", " ^ msg' () ^ ": " ^ quote (beginning 10 ss));
wenzelm
parents: 50236
diff changeset
   510
    val finite_scan = Scan.error (Scan.finite stopper (!! message scan));
wenzelm
parents: 50236
diff changeset
   511
  in
wenzelm
parents: 50236
diff changeset
   512
    (case finite_scan syms of
wenzelm
parents: 50236
diff changeset
   513
      (result, []) => result
wenzelm
parents: 50236
diff changeset
   514
    | (_, rest) => error (message (rest, NONE) ()))
wenzelm
parents: 50236
diff changeset
   515
  end;
wenzelm
parents: 50236
diff changeset
   516
wenzelm
parents: 50236
diff changeset
   517
50162
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   518
(* space-separated words *)
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   519
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   520
val scan_word =
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   521
  Scan.many1 is_ascii_blank >> K NONE ||
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   522
  Scan.many1 (fn s => not (is_ascii_blank s) andalso not_eof s) >> (SOME o implode);
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   523
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   524
val split_words = scanner "Bad text" (Scan.repeat scan_word >> map_filter I);
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   525
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   526
val explode_words = split_words o sym_explode;
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   527
e06eabc421e7 some support for breakable text and paragraphs;
wenzelm
parents: 48774
diff changeset
   528
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   529
(* blanks *)
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   530
61707
wenzelm
parents: 61579
diff changeset
   531
val trim_blanks = sym_explode #> trim is_blank #> implode;
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   532
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   533
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   534
(* bump string -- treat as base 26 or base 1 numbers *)
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   535
62529
8b7bdfc09f3b clarified treatment of fragments of Isabelle symbols during bootstrap;
wenzelm
parents: 62210
diff changeset
   536
fun symbolic_end (_ :: "\<^sub>" :: _) = true
53198
06b096cf89ca ignore trailing primes, e.g. rename \<alpha>' to \<alpha>'' instead of \<alpha>'a;
wenzelm
parents: 53016
diff changeset
   537
  | symbolic_end ("'" :: ss) = symbolic_end ss
55033
8e8243975860 support for nested text cartouches;
wenzelm
parents: 54734
diff changeset
   538
  | symbolic_end (s :: _) = raw_symbolic s
14908
wenzelm
parents: 14873
diff changeset
   539
  | symbolic_end [] = false;
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   540
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   541
fun bump_init str =
14908
wenzelm
parents: 14873
diff changeset
   542
  if symbolic_end (rev (sym_explode str)) then str ^ "'"
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   543
  else str ^ "a";
12904
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   544
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   545
fun bump_string str =
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   546
  let
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   547
    fun bump [] = ["a"]
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   548
      | bump ("z" :: ss) = "a" :: bump ss
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   549
      | bump (s :: ss) =
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   550
          if is_char s andalso ord "a" <= ord s andalso ord s < ord "z"
12904
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   551
          then chr (ord s + 1) :: ss
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   552
          else "a" :: s :: ss;
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   553
33955
fff6f11b1f09 curried take/drop
haftmann
parents: 33548
diff changeset
   554
    val (ss, qs) = apfst rev (take_suffix is_quasi (sym_explode str));
14908
wenzelm
parents: 14873
diff changeset
   555
    val ss' = if symbolic_end ss then "'" :: ss else bump ss;
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   556
  in implode (rev ss' @ qs) end;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   557
12904
c208d71702d1 added is_quasi;
wenzelm
parents: 12116
diff changeset
   558
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   559
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   560
(** symbol output **)
14977
77d88064991a added escape, export encode_raw, default mode now trivial, tuned;
wenzelm
parents: 14961
diff changeset
   561
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   562
(* length *)
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   563
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   564
fun sym_len s =
24593
1547ea587f5a added some int constraints (ML_Parse.fix_ints not active here);
wenzelm
parents: 24580
diff changeset
   565
  if not (is_printable s) then (0: int)
62877
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   566
  else if String.isPrefix "\092<long" s then 2
741560a5283b avoid malformed Isabelle symbols during bootstrap;
wenzelm
parents: 62804
diff changeset
   567
  else if String.isPrefix "\092<Long" s then 2
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   568
  else 1;
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   569
19473
wenzelm
parents: 19305
diff changeset
   570
fun sym_length ss = fold (fn s => fn n => sym_len s + n) ss 0;
14678
662b181cae05 clarification of ascii vs. symbolic entities;
wenzelm
parents: 14632
diff changeset
   571
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   572
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   573
(* print mode *)
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   574
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   575
val xsymbolsN = "xsymbols";
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   576
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   577
fun output s = (s, sym_length (sym_explode s));
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   578
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   579
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   580
(*final declarations of this structure!*)
29324
e07fc65e296b added output;
wenzelm
parents: 27903
diff changeset
   581
val explode = sym_explode;
6272
52d99d68d3be added output_width;
wenzelm
parents: 6225
diff changeset
   582
val length = sym_length;
6116
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   583
8ba2f25610f7 files scan.ML, source.ML, symbol.ML, pretty.ML moved to Pure/General;
wenzelm
parents:
diff changeset
   584
end;