| author | wenzelm |
| Sun, 11 Jan 2009 18:18:35 +0100 | |
| changeset 29448 | 34b9652b2f45 |
| parent 29318 | 6337d1cb2ba0 |
| child 29565 | 3f8b24fcfbd6 |
| permissions | -rw-r--r-- |
| 18 | 1 |
(* Title: Pure/Syntax/lexicon.ML |
| 0 | 2 |
ID: $Id$ |
| 18 | 3 |
Author: Tobias Nipkow and Markus Wenzel, TU Muenchen |
| 0 | 4 |
|
| 4703 | 5 |
Lexer for the inner Isabelle syntax (terms and types). |
| 18 | 6 |
*) |
| 0 | 7 |
|
8 |
signature LEXICON0 = |
|
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
9 |
sig |
| 0 | 10 |
val is_identifier: string -> bool |
| 14679 | 11 |
val is_ascii_identifier: string -> bool |
| 20165 | 12 |
val is_tid: string -> bool |
| 27773 | 13 |
val scan_id: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
14 |
val scan_longid: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
15 |
val scan_tid: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
16 |
val scan_nat: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
17 |
val scan_int: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
18 |
val scan_hex: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
19 |
val scan_bin: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
20 |
val scan_var: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
21 |
val scan_tvar: SymbolPos.T list -> SymbolPos.T list * SymbolPos.T list |
|
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
22 |
val implode_xstr: string list -> string |
|
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
23 |
val explode_xstr: string -> string list |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
24 |
val read_indexname: string -> indexname |
| 4703 | 25 |
val read_var: string -> term |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
26 |
val read_variable: string -> indexname option |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
27 |
val const: string -> term |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
28 |
val free: string -> term |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
29 |
val var: indexname -> term |
| 5860 | 30 |
val read_nat: string -> int option |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
31 |
val read_int: string -> int option |
|
24630
351a308ab58d
simplified type int (eliminated IntInf.int, integer);
wenzelm
parents:
24583
diff
changeset
|
32 |
val read_xnum: string -> {radix: int, leading_zeros: int, value: int}
|
| 28904 | 33 |
val read_float: string -> {mant: int, exp: int}
|
| 19002 | 34 |
val fixedN: string |
35 |
val constN: string |
|
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
36 |
end; |
| 0 | 37 |
|
38 |
signature LEXICON = |
|
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
39 |
sig |
| 18 | 40 |
include LEXICON0 |
41 |
val is_xid: string -> bool |
|
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
42 |
datatype token_kind = |
| 27887 | 43 |
Literal | IdentSy | LongIdentSy | VarSy | TFreeSy | TVarSy | |
| 28904 | 44 |
NumSy | FloatSy | XNumSy | StrSy | Space | Comment | EOF |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
45 |
datatype token = Token of token_kind * string * Position.range |
| 27806 | 46 |
val str_of_token: token -> string |
47 |
val pos_of_token: token -> Position.T |
|
| 27887 | 48 |
val is_proper: token -> bool |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
49 |
val mk_eof: Position.T -> token |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
50 |
val eof: token |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
51 |
val is_eof: token -> bool |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
52 |
val stopper: token Scan.stopper |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
53 |
val idT: typ |
| 3828 | 54 |
val longidT: typ |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
55 |
val varT: typ |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
56 |
val tidT: typ |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
57 |
val tvarT: typ |
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
58 |
val terminals: string list |
|
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
59 |
val is_terminal: string -> bool |
|
28413
ee73353fb87c
report_token/parse_token: back to context-less version;
wenzelm
parents:
28407
diff
changeset
|
60 |
val report_token: token -> unit |
| 18 | 61 |
val matching_tokens: token * token -> bool |
62 |
val valued_token: token -> bool |
|
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
63 |
val predef_term: string -> token option |
| 27773 | 64 |
val tokenize: Scan.lexicon -> bool -> SymbolPos.T list -> token list |
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
65 |
end; |
| 0 | 66 |
|
| 14679 | 67 |
structure Lexicon: LEXICON = |
| 0 | 68 |
struct |
69 |
||
| 18 | 70 |
(** is_identifier etc. **) |
71 |
||
| 16150 | 72 |
val is_identifier = Symbol.is_ident o Symbol.explode; |
| 18 | 73 |
|
| 14679 | 74 |
fun is_ascii_identifier s = |
75 |
let val cs = Symbol.explode s |
|
| 16150 | 76 |
in forall Symbol.is_ascii cs andalso Symbol.is_ident cs end; |
| 14679 | 77 |
|
| 18 | 78 |
fun is_xid s = |
| 4703 | 79 |
(case Symbol.explode s of |
| 16150 | 80 |
"_" :: cs => Symbol.is_ident cs |
81 |
| cs => Symbol.is_ident cs); |
|
| 18 | 82 |
|
|
330
2fda15dd1e0f
changed the way a grammar is generated to allow the new parser to work;
clasohm
parents:
237
diff
changeset
|
83 |
fun is_tid s = |
| 4703 | 84 |
(case Symbol.explode s of |
| 16150 | 85 |
"'" :: cs => Symbol.is_ident cs |
| 18 | 86 |
| _ => false); |
87 |
||
| 0 | 88 |
|
89 |
||
| 4703 | 90 |
(** basic scanners **) |
91 |
||
| 27773 | 92 |
open BasicSymbolPos; |
93 |
||
94 |
fun !!! msg = SymbolPos.!!! ("Inner lexical error: " ^ msg);
|
|
95 |
||
96 |
val scan_id = Scan.one (Symbol.is_letter o symbol) ::: Scan.many (Symbol.is_letdig o symbol); |
|
97 |
val scan_longid = scan_id @@@ (Scan.repeat1 ($$$ "." @@@ scan_id) >> flat); |
|
98 |
val scan_tid = $$$ "'" @@@ scan_id; |
|
| 4703 | 99 |
|
| 27773 | 100 |
val scan_nat = Scan.many1 (Symbol.is_digit o symbol); |
101 |
val scan_int = $$$ "-" @@@ scan_nat || scan_nat; |
|
| 28904 | 102 |
val scan_natdot = scan_nat @@@ $$$ "." @@@ scan_nat; |
103 |
val scan_float = $$$ "-" @@@ scan_natdot || scan_natdot; |
|
| 27773 | 104 |
val scan_hex = $$$ "0" @@@ $$$ "x" @@@ Scan.many1 (Symbol.is_ascii_hex o symbol); |
105 |
val scan_bin = $$$ "0" @@@ $$$ "b" @@@ Scan.many1 (fn (s, _) => s = "0" orelse s = "1"); |
|
| 4703 | 106 |
|
| 27773 | 107 |
val scan_id_nat = scan_id @@@ Scan.optional ($$$ "." @@@ scan_nat) []; |
108 |
val scan_var = $$$ "?" @@@ scan_id_nat; |
|
109 |
val scan_tvar = $$$ "?" @@@ $$$ "'" @@@ scan_id_nat; |
|
| 4703 | 110 |
|
111 |
||
112 |
||
| 18 | 113 |
(** datatype token **) |
| 0 | 114 |
|
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
115 |
datatype token_kind = |
| 27887 | 116 |
Literal | IdentSy | LongIdentSy | VarSy | TFreeSy | TVarSy | |
| 28904 | 117 |
NumSy | FloatSy | XNumSy | StrSy | Space | Comment | EOF; |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
118 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
119 |
datatype token = Token of token_kind * string * Position.range; |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
120 |
|
| 27806 | 121 |
fun str_of_token (Token (_, s, _)) = s; |
122 |
fun pos_of_token (Token (_, _, (pos, _))) = pos; |
|
123 |
||
| 27887 | 124 |
fun is_proper (Token (Space, _, _)) = false |
125 |
| is_proper (Token (Comment, _, _)) = false |
|
126 |
| is_proper _ = true; |
|
127 |
||
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
128 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
129 |
(* stopper *) |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
130 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
131 |
fun mk_eof pos = Token (EOF, "", (pos, Position.none)); |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
132 |
val eof = mk_eof Position.none; |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
133 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
134 |
fun is_eof (Token (EOF, _, _)) = true |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
135 |
| is_eof _ = false; |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
136 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
137 |
val stopper = Scan.stopper (K eof) is_eof; |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
138 |
|
| 0 | 139 |
|
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
140 |
(* terminal arguments *) |
| 0 | 141 |
|
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
142 |
val idT = Type ("id", []);
|
| 3828 | 143 |
val longidT = Type ("longid", []);
|
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
144 |
val varT = Type ("var", []);
|
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
145 |
val tidT = Type ("tid", []);
|
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
146 |
val tvarT = Type ("tvar", []);
|
| 0 | 147 |
|
|
29156
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
148 |
val terminal_kinds = |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
149 |
[("id", IdentSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
150 |
("longid", LongIdentSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
151 |
("var", VarSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
152 |
("tid", TFreeSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
153 |
("tvar", TVarSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
154 |
("num", NumSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
155 |
("float_token", FloatSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
156 |
("xnum", XNumSy),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
157 |
("xstr", StrSy)];
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
158 |
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
159 |
val terminals = map #1 terminal_kinds; |
| 20664 | 160 |
val is_terminal = member (op =) terminals; |
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
161 |
|
| 0 | 162 |
|
| 27887 | 163 |
(* markup *) |
164 |
||
165 |
val token_kind_markup = |
|
166 |
fn Literal => Markup.literal |
|
167 |
| IdentSy => Markup.ident |
|
168 |
| LongIdentSy => Markup.ident |
|
169 |
| VarSy => Markup.var |
|
170 |
| TFreeSy => Markup.tfree |
|
171 |
| TVarSy => Markup.tvar |
|
|
29318
6337d1cb2ba0
added numeral, which supercedes num, xnum, float;
wenzelm
parents:
29156
diff
changeset
|
172 |
| NumSy => Markup.numeral |
|
6337d1cb2ba0
added numeral, which supercedes num, xnum, float;
wenzelm
parents:
29156
diff
changeset
|
173 |
| FloatSy => Markup.numeral |
|
6337d1cb2ba0
added numeral, which supercedes num, xnum, float;
wenzelm
parents:
29156
diff
changeset
|
174 |
| XNumSy => Markup.numeral |
|
6337d1cb2ba0
added numeral, which supercedes num, xnum, float;
wenzelm
parents:
29156
diff
changeset
|
175 |
| StrSy => Markup.inner_string |
| 27887 | 176 |
| Space => Markup.none |
177 |
| Comment => Markup.inner_comment |
|
178 |
| EOF => Markup.none; |
|
179 |
||
|
28413
ee73353fb87c
report_token/parse_token: back to context-less version;
wenzelm
parents:
28407
diff
changeset
|
180 |
fun report_token (Token (kind, _, (pos, _))) = |
|
ee73353fb87c
report_token/parse_token: back to context-less version;
wenzelm
parents:
28407
diff
changeset
|
181 |
Position.report (token_kind_markup kind) pos; |
| 27887 | 182 |
|
183 |
||
| 18 | 184 |
(* matching_tokens *) |
| 0 | 185 |
|
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
186 |
fun matching_tokens (Token (Literal, x, _), Token (Literal, y, _)) = x = y |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
187 |
| matching_tokens (Token (k, _, _), Token (k', _, _)) = k = k'; |
| 0 | 188 |
|
189 |
||
| 18 | 190 |
(* valued_token *) |
| 0 | 191 |
|
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
192 |
fun valued_token (Token (Literal, _, _)) = false |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
193 |
| valued_token (Token (EOF, _, _)) = false |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
194 |
| valued_token _ = true; |
| 0 | 195 |
|
196 |
||
| 18 | 197 |
(* predef_term *) |
| 0 | 198 |
|
|
29156
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
199 |
fun predef_term s = |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
200 |
(case AList.lookup (op =) terminal_kinds s of |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
201 |
SOME sy => SOME (Token (sy, s, Position.no_range)) |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
202 |
| NONE => NONE); |
| 0 | 203 |
|
204 |
||
| 4703 | 205 |
(* xstr tokens *) |
| 18 | 206 |
|
| 14730 | 207 |
val scan_chr = |
| 27773 | 208 |
$$$ "\\" |-- $$$ "'" || |
209 |
Scan.one ((fn s => s <> "\\" andalso s <> "'" andalso Symbol.is_regular s) o symbol) >> single || |
|
210 |
$$$ "'" --| Scan.ahead (~$$$ "'"); |
|
| 14730 | 211 |
|
212 |
val scan_str = |
|
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
213 |
$$$ "'" @@@ $$$ "'" @@@ !!! "missing end of string" |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
214 |
((Scan.repeat scan_chr >> flat) @@@ $$$ "'" @@@ $$$ "'"); |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
215 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
216 |
val scan_str_body = |
| 27773 | 217 |
$$$ "'" |-- $$$ "'" |-- !!! "missing end of string" |
218 |
((Scan.repeat scan_chr >> flat) --| $$$ "'" --| $$$ "'"); |
|
| 14730 | 219 |
|
| 0 | 220 |
|
| 4703 | 221 |
fun implode_xstr cs = enclose "''" "''" (implode (map (fn "'" => "\\'" | c => c) cs)); |
| 18 | 222 |
|
| 4703 | 223 |
fun explode_xstr str = |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
224 |
(case Scan.read SymbolPos.stopper scan_str_body (SymbolPos.explode (str, Position.none)) of |
| 27773 | 225 |
SOME cs => map symbol cs |
| 5868 | 226 |
| _ => error ("Inner lexical error: literal string expected at " ^ quote str));
|
| 18 | 227 |
|
228 |
||
| 27806 | 229 |
|
| 18 | 230 |
(** tokenize **) |
231 |
||
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
232 |
fun token_leq (Token (_, s1, _), Token (_, s2, _)) = s1 <= s2; |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
233 |
fun token kind ss = Token (kind, SymbolPos.content ss, SymbolPos.range ss); |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
234 |
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
235 |
fun tokenize lex xids syms = |
| 18 | 236 |
let |
237 |
val scan_xid = |
|
| 27773 | 238 |
if xids then $$$ "_" @@@ scan_id || scan_id |
| 18 | 239 |
else scan_id; |
240 |
||
| 20096 | 241 |
val scan_num = scan_hex || scan_bin || scan_int; |
242 |
||
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
243 |
val scan_val = |
|
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
244 |
scan_tvar >> token TVarSy || |
|
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
245 |
scan_var >> token VarSy || |
|
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
246 |
scan_tid >> token TFreeSy || |
| 28904 | 247 |
scan_float >> token FloatSy || |
|
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
248 |
scan_num >> token NumSy || |
| 27773 | 249 |
$$$ "#" @@@ scan_num >> token XNumSy || |
|
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
250 |
scan_longid >> token LongIdentSy || |
|
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
251 |
scan_xid >> token IdentSy; |
| 18 | 252 |
|
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
253 |
val scan_lit = Scan.literal lex >> token Literal; |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
254 |
|
| 4703 | 255 |
val scan_token = |
| 27887 | 256 |
SymbolPos.scan_comment !!! >> token Comment || |
257 |
Scan.max token_leq scan_lit scan_val || |
|
258 |
scan_str >> token StrSy || |
|
259 |
Scan.many1 (Symbol.is_blank o symbol) >> token Space; |
|
| 18 | 260 |
in |
| 27773 | 261 |
(case Scan.error |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
262 |
(Scan.finite SymbolPos.stopper (Scan.repeat scan_token)) syms of |
| 27887 | 263 |
(toks, []) => toks |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
264 |
| (_, ss) => error ("Inner lexical error at: " ^ SymbolPos.content ss ^
|
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
265 |
Position.str_of (#1 (SymbolPos.range ss)))) |
| 18 | 266 |
end; |
267 |
||
268 |
||
269 |
||
270 |
(** scan variables **) |
|
271 |
||
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
272 |
(* scan_indexname *) |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
273 |
|
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
274 |
local |
| 18 | 275 |
|
| 27773 | 276 |
val scan_vname = |
| 18 | 277 |
let |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
278 |
fun nat n [] = n |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
279 |
| nat n (c :: cs) = nat (n * 10 + (ord c - ord "0")) cs; |
| 18 | 280 |
|
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
281 |
fun idxname cs ds = (implode (rev cs), nat 0 ds); |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
282 |
fun chop_idx [] ds = idxname [] ds |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
283 |
| chop_idx (cs as (_ :: "\\<^isub>" :: _)) ds = idxname cs ds |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
284 |
| chop_idx (cs as (_ :: "\\<^isup>" :: _)) ds = idxname cs ds |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
285 |
| chop_idx (c :: cs) ds = |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
286 |
if Symbol.is_digit c then chop_idx cs (c :: ds) |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
287 |
else idxname (c :: cs) ds; |
| 18 | 288 |
|
| 27773 | 289 |
val scan = (scan_id >> map symbol) -- |
290 |
Scan.optional ($$$ "." |-- scan_nat >> (nat 0 o map symbol)) ~1; |
|
| 18 | 291 |
in |
| 27773 | 292 |
scan >> |
293 |
(fn (cs, ~1) => chop_idx (rev cs) [] |
|
294 |
| (cs, i) => (implode cs, i)) |
|
| 18 | 295 |
end; |
296 |
||
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
297 |
in |
| 18 | 298 |
|
| 27773 | 299 |
val scan_indexname = $$$ "'" |-- scan_vname >> (fn (x, i) => ("'" ^ x, i)) || scan_vname;
|
|
15443
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
300 |
|
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
301 |
end; |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
302 |
|
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
303 |
|
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
304 |
(* indexname *) |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
305 |
|
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
306 |
fun read_indexname s = |
| 27773 | 307 |
(case Scan.read SymbolPos.stopper scan_indexname (SymbolPos.explode (s, Position.none)) of |
| 15531 | 308 |
SOME xi => xi |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
309 |
| _ => error ("Lexical error in variable name: " ^ quote s));
|
| 18 | 310 |
|
311 |
||
| 4703 | 312 |
(* read_var *) |
| 18 | 313 |
|
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
314 |
fun const c = Const (c, dummyT); |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
315 |
fun free x = Free (x, dummyT); |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
316 |
fun var xi = Var (xi, dummyT); |
|
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
317 |
|
| 4703 | 318 |
fun read_var str = |
| 18 | 319 |
let |
320 |
val scan = |
|
| 27773 | 321 |
$$$ "?" |-- scan_indexname --| Scan.ahead (Scan.one SymbolPos.is_eof) >> var || |
322 |
Scan.many (Symbol.is_regular o symbol) >> (free o implode o map symbol); |
|
323 |
in the (Scan.read SymbolPos.stopper scan (SymbolPos.explode (str, Position.none))) end; |
|
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
324 |
|
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
325 |
|
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
326 |
(* read_variable *) |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
327 |
|
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
328 |
fun read_variable str = |
| 27773 | 329 |
let val scan = $$$ "?" |-- scan_indexname || scan_indexname |
330 |
in Scan.read SymbolPos.stopper scan (SymbolPos.explode (str, Position.none)) end; |
|
| 4587 | 331 |
|
332 |
||
| 19002 | 333 |
(* specific identifiers *) |
| 5260 | 334 |
|
| 19002 | 335 |
val constN = "\\<^const>"; |
336 |
val fixedN = "\\<^fixed>"; |
|
337 |
||
| 5260 | 338 |
|
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
339 |
(* read numbers *) |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
340 |
|
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
341 |
local |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
342 |
|
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
343 |
fun nat cs = |
| 27773 | 344 |
Option.map (#1 o Library.read_int o map symbol) |
345 |
(Scan.read SymbolPos.stopper scan_nat cs); |
|
| 5860 | 346 |
|
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
347 |
in |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
348 |
|
| 27773 | 349 |
fun read_nat s = nat (SymbolPos.explode (s, Position.none)); |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
350 |
|
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
351 |
fun read_int s = |
| 27773 | 352 |
(case SymbolPos.explode (s, Position.none) of |
353 |
("-", _) :: cs => Option.map ~ (nat cs)
|
|
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
354 |
| cs => nat cs); |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
355 |
|
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
356 |
end; |
| 5860 | 357 |
|
358 |
||
| 20096 | 359 |
(* read_xnum: hex/bin/decimal *) |
| 9326 | 360 |
|
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
361 |
local |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
362 |
|
| 20096 | 363 |
val ten = ord "0" + 10; |
364 |
val a = ord "a"; |
|
365 |
val A = ord "A"; |
|
| 23802 | 366 |
val _ = a > A orelse sys_error "Bad ASCII"; |
| 20096 | 367 |
|
368 |
fun remap_hex c = |
|
369 |
let val x = ord c in |
|
370 |
if x >= a then chr (x - a + ten) |
|
371 |
else if x >= A then chr (x - A + ten) |
|
372 |
else c |
|
373 |
end; |
|
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
374 |
|
| 21781 | 375 |
fun leading_zeros ["0"] = 0 |
376 |
| leading_zeros ("0" :: cs) = 1 + leading_zeros cs
|
|
377 |
| leading_zeros _ = 0; |
|
378 |
||
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
379 |
in |
|
15965
f422f8283491
Use of IntInf.int instead of int in most numeric simprocs; avoids
paulson
parents:
15570
diff
changeset
|
380 |
|
| 9326 | 381 |
fun read_xnum str = |
382 |
let |
|
| 20096 | 383 |
val (sign, radix, digs) = |
384 |
(case Symbol.explode (perhaps (try (unprefix "#")) str) of |
|
385 |
"0" :: "x" :: cs => (1, 16, map remap_hex cs) |
|
386 |
| "0" :: "b" :: cs => (1, 2, cs) |
|
387 |
| "-" :: cs => (~1, 10, cs) |
|
388 |
| cs => (1, 10, cs)); |
|
|
29156
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
389 |
in |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
390 |
{radix = radix,
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
391 |
leading_zeros = leading_zeros digs, |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
392 |
value = sign * #1 (Library.read_radix_int radix digs)} |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
393 |
end; |
| 9326 | 394 |
|
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
395 |
end; |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
396 |
|
| 28904 | 397 |
fun read_float str = |
398 |
let |
|
399 |
val (sign, cs) = |
|
|
29156
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
400 |
(case Symbol.explode str of |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
401 |
"-" :: cs => (~1, cs) |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
402 |
| cs => (1, cs)); |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
403 |
val (intpart, fracpart) = |
| 28904 | 404 |
(case take_prefix Symbol.is_digit cs of |
|
29156
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
405 |
(intpart, "." :: fracpart) => (intpart, fracpart) |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
406 |
| _ => raise Fail "read_float"); |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
407 |
in |
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
408 |
{mant = sign * #1 (Library.read_int (intpart @ fracpart)),
|
|
89f76a58a378
renamed terminal category "float" to "float_token", to avoid name
wenzelm
parents:
28904
diff
changeset
|
409 |
exp = length fracpart} |
| 28904 | 410 |
end |
411 |
||
| 0 | 412 |
end; |