author | wenzelm |
Sat, 09 Aug 2008 00:09:35 +0200 | |
changeset 27800 | df444ddeff56 |
parent 27773 | a52166b228b9 |
child 27806 | ece79c0597fe |
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} |
19002 | 33 |
val fixedN: string |
34 |
val constN: string |
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
35 |
end; |
0 | 36 |
|
37 |
signature LEXICON = |
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
38 |
sig |
18 | 39 |
include LEXICON0 |
40 |
val is_xid: string -> bool |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
41 |
datatype token_kind = |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
42 |
Literal | IdentSy | LongIdentSy | VarSy | TFreeSy | TVarSy | NumSy | XNumSy | StrSy | EOF |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
43 |
datatype token = Token of token_kind * string * Position.range |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
44 |
val mk_eof: Position.T -> token |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
45 |
val eof: token |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
46 |
val is_eof: token -> bool |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
47 |
val stopper: token Scan.stopper |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
48 |
val idT: typ |
3828 | 49 |
val longidT: typ |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
50 |
val varT: typ |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
51 |
val tidT: typ |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
52 |
val tvarT: typ |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
53 |
val terminals: string list |
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
54 |
val is_terminal: string -> bool |
18 | 55 |
val str_of_token: token -> string |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
56 |
val pos_of_token: token -> string |
18 | 57 |
val display_token: token -> string |
58 |
val matching_tokens: token * token -> bool |
|
59 |
val valued_token: token -> bool |
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
60 |
val predef_term: string -> token option |
27773 | 61 |
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
|
62 |
end; |
0 | 63 |
|
14679 | 64 |
structure Lexicon: LEXICON = |
0 | 65 |
struct |
66 |
||
18 | 67 |
(** is_identifier etc. **) |
68 |
||
16150 | 69 |
val is_identifier = Symbol.is_ident o Symbol.explode; |
18 | 70 |
|
14679 | 71 |
fun is_ascii_identifier s = |
72 |
let val cs = Symbol.explode s |
|
16150 | 73 |
in forall Symbol.is_ascii cs andalso Symbol.is_ident cs end; |
14679 | 74 |
|
18 | 75 |
fun is_xid s = |
4703 | 76 |
(case Symbol.explode s of |
16150 | 77 |
"_" :: cs => Symbol.is_ident cs |
78 |
| cs => Symbol.is_ident cs); |
|
18 | 79 |
|
330
2fda15dd1e0f
changed the way a grammar is generated to allow the new parser to work;
clasohm
parents:
237
diff
changeset
|
80 |
fun is_tid s = |
4703 | 81 |
(case Symbol.explode s of |
16150 | 82 |
"'" :: cs => Symbol.is_ident cs |
18 | 83 |
| _ => false); |
84 |
||
0 | 85 |
|
86 |
||
4703 | 87 |
(** basic scanners **) |
88 |
||
27773 | 89 |
open BasicSymbolPos; |
90 |
||
91 |
fun !!! msg = SymbolPos.!!! ("Inner lexical error: " ^ msg); |
|
92 |
||
93 |
val scan_id = Scan.one (Symbol.is_letter o symbol) ::: Scan.many (Symbol.is_letdig o symbol); |
|
94 |
val scan_longid = scan_id @@@ (Scan.repeat1 ($$$ "." @@@ scan_id) >> flat); |
|
95 |
val scan_tid = $$$ "'" @@@ scan_id; |
|
4703 | 96 |
|
27773 | 97 |
val scan_nat = Scan.many1 (Symbol.is_digit o symbol); |
98 |
val scan_int = $$$ "-" @@@ scan_nat || scan_nat; |
|
99 |
val scan_hex = $$$ "0" @@@ $$$ "x" @@@ Scan.many1 (Symbol.is_ascii_hex o symbol); |
|
100 |
val scan_bin = $$$ "0" @@@ $$$ "b" @@@ Scan.many1 (fn (s, _) => s = "0" orelse s = "1"); |
|
4703 | 101 |
|
27773 | 102 |
val scan_id_nat = scan_id @@@ Scan.optional ($$$ "." @@@ scan_nat) []; |
103 |
val scan_var = $$$ "?" @@@ scan_id_nat; |
|
104 |
val scan_tvar = $$$ "?" @@@ $$$ "'" @@@ scan_id_nat; |
|
4703 | 105 |
|
106 |
||
107 |
||
18 | 108 |
(** datatype token **) |
0 | 109 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
110 |
datatype token_kind = |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
111 |
Literal | IdentSy | LongIdentSy | VarSy | TFreeSy | TVarSy | NumSy | XNumSy | StrSy | EOF; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
112 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
113 |
datatype token = Token of token_kind * string * Position.range; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
114 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
115 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
116 |
(* stopper *) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
117 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
118 |
fun mk_eof pos = Token (EOF, "", (pos, Position.none)); |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
119 |
val eof = mk_eof Position.none; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
120 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
121 |
fun is_eof (Token (EOF, _, _)) = true |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
122 |
| is_eof _ = false; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
123 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
124 |
val stopper = Scan.stopper (K eof) is_eof; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
125 |
|
0 | 126 |
|
127 |
||
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
128 |
(* terminal arguments *) |
0 | 129 |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
130 |
val idT = Type ("id", []); |
3828 | 131 |
val longidT = Type ("longid", []); |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
132 |
val varT = Type ("var", []); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
133 |
val tidT = Type ("tid", []); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
134 |
val tvarT = Type ("tvar", []); |
0 | 135 |
|
11697 | 136 |
val terminals = ["id", "longid", "var", "tid", "tvar", "num", "xnum", "xstr"]; |
20664 | 137 |
val is_terminal = member (op =) terminals; |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
138 |
|
0 | 139 |
|
18 | 140 |
(* str_of_token *) |
0 | 141 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
142 |
fun str_of_token (Token (_, s, _)) = s; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
143 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
144 |
fun pos_of_token (Token (_, _, (pos, _))) = Position.str_of pos; |
0 | 145 |
|
18 | 146 |
|
147 |
(* display_token *) |
|
0 | 148 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
149 |
fun display_token (Token (Literal, s, _)) = quote s |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
150 |
| display_token (Token (IdentSy, s, _)) = "id(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
151 |
| display_token (Token (LongIdentSy, s, _)) = "longid(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
152 |
| display_token (Token (VarSy, s, _)) = "var(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
153 |
| display_token (Token (TFreeSy, s, _)) = "tid(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
154 |
| display_token (Token (TVarSy, s, _)) = "tvar(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
155 |
| display_token (Token (NumSy, s, _)) = "num(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
156 |
| display_token (Token (XNumSy, s, _)) = "xnum(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
157 |
| display_token (Token (StrSy, s, _)) = "xstr(" ^ s ^ ")" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
158 |
| display_token (Token (EOF, _, _)) = ""; |
0 | 159 |
|
18 | 160 |
|
161 |
(* matching_tokens *) |
|
0 | 162 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
163 |
fun matching_tokens (Token (Literal, x, _), Token (Literal, y, _)) = x = y |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
164 |
| matching_tokens (Token (k, _, _), Token (k', _, _)) = k = k'; |
0 | 165 |
|
166 |
||
18 | 167 |
(* valued_token *) |
0 | 168 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
169 |
fun valued_token (Token (Literal, _, _)) = false |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
170 |
| valued_token (Token (EOF, _, _)) = false |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
171 |
| valued_token _ = true; |
0 | 172 |
|
173 |
||
18 | 174 |
(* predef_term *) |
0 | 175 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
176 |
fun predef_term "id" = SOME (Token (IdentSy, "id", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
177 |
| predef_term "longid" = SOME (Token (LongIdentSy, "longid", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
178 |
| predef_term "var" = SOME (Token (VarSy, "var", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
179 |
| predef_term "tid" = SOME (Token (TFreeSy, "tid", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
180 |
| predef_term "tvar" = SOME (Token (TVarSy, "tvar", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
181 |
| predef_term "num" = SOME (Token (NumSy, "num", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
182 |
| predef_term "xnum" = SOME (Token (XNumSy, "xnum", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
183 |
| predef_term "xstr" = SOME (Token (StrSy, "xstr", Position.no_range)) |
15531 | 184 |
| predef_term _ = NONE; |
0 | 185 |
|
186 |
||
4703 | 187 |
(* xstr tokens *) |
18 | 188 |
|
14730 | 189 |
val scan_chr = |
27773 | 190 |
$$$ "\\" |-- $$$ "'" || |
191 |
Scan.one ((fn s => s <> "\\" andalso s <> "'" andalso Symbol.is_regular s) o symbol) >> single || |
|
192 |
$$$ "'" --| Scan.ahead (~$$$ "'"); |
|
14730 | 193 |
|
194 |
val scan_str = |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
195 |
$$$ "'" @@@ $$$ "'" @@@ !!! "missing end of string" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
196 |
((Scan.repeat scan_chr >> flat) @@@ $$$ "'" @@@ $$$ "'"); |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
197 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
198 |
val scan_str_body = |
27773 | 199 |
$$$ "'" |-- $$$ "'" |-- !!! "missing end of string" |
200 |
((Scan.repeat scan_chr >> flat) --| $$$ "'" --| $$$ "'"); |
|
14730 | 201 |
|
0 | 202 |
|
4703 | 203 |
fun implode_xstr cs = enclose "''" "''" (implode (map (fn "'" => "\\'" | c => c) cs)); |
18 | 204 |
|
4703 | 205 |
fun explode_xstr str = |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
206 |
(case Scan.read SymbolPos.stopper scan_str_body (SymbolPos.explode (str, Position.none)) of |
27773 | 207 |
SOME cs => map symbol cs |
5868 | 208 |
| _ => error ("Inner lexical error: literal string expected at " ^ quote str)); |
18 | 209 |
|
210 |
||
211 |
(** tokenize **) |
|
212 |
||
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
213 |
fun token_leq (Token (_, s1, _), Token (_, s2, _)) = s1 <= s2; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
214 |
fun token kind ss = Token (kind, SymbolPos.content ss, SymbolPos.range ss); |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
215 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
216 |
fun tokenize lex xids syms = |
18 | 217 |
let |
218 |
val scan_xid = |
|
27773 | 219 |
if xids then $$$ "_" @@@ scan_id || scan_id |
18 | 220 |
else scan_id; |
221 |
||
20096 | 222 |
val scan_num = scan_hex || scan_bin || scan_int; |
223 |
||
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
224 |
val scan_val = |
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
225 |
scan_tvar >> token TVarSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
226 |
scan_var >> token VarSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
227 |
scan_tid >> token TFreeSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
228 |
scan_num >> token NumSy || |
27773 | 229 |
$$$ "#" @@@ scan_num >> token XNumSy || |
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
230 |
scan_longid >> token LongIdentSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
231 |
scan_xid >> token IdentSy; |
18 | 232 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
233 |
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
|
234 |
|
4703 | 235 |
val scan_token = |
27773 | 236 |
SymbolPos.scan_comment !!! >> K NONE || |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
237 |
Scan.max token_leq scan_lit scan_val >> SOME || |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
238 |
scan_str >> (SOME o token StrSy) || |
27773 | 239 |
Scan.one (Symbol.is_blank o symbol) >> K NONE; |
18 | 240 |
in |
27773 | 241 |
(case Scan.error |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
242 |
(Scan.finite SymbolPos.stopper (Scan.repeat scan_token)) syms of |
24245 | 243 |
(toks, []) => map_filter I toks |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
244 |
| (_, ss) => error ("Inner lexical error at: " ^ SymbolPos.content ss ^ |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
245 |
Position.str_of (#1 (SymbolPos.range ss)))) |
18 | 246 |
end; |
247 |
||
248 |
||
249 |
||
250 |
(** scan variables **) |
|
251 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
252 |
(* scan_indexname *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
253 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
254 |
local |
18 | 255 |
|
27773 | 256 |
val scan_vname = |
18 | 257 |
let |
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
258 |
fun nat n [] = n |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
259 |
| nat n (c :: cs) = nat (n * 10 + (ord c - ord "0")) cs; |
18 | 260 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
261 |
fun idxname cs ds = (implode (rev cs), nat 0 ds); |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
262 |
fun chop_idx [] ds = idxname [] ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
263 |
| chop_idx (cs as (_ :: "\\<^isub>" :: _)) ds = idxname cs ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
264 |
| chop_idx (cs as (_ :: "\\<^isup>" :: _)) ds = idxname cs ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
265 |
| chop_idx (c :: cs) ds = |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
266 |
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
|
267 |
else idxname (c :: cs) ds; |
18 | 268 |
|
27773 | 269 |
val scan = (scan_id >> map symbol) -- |
270 |
Scan.optional ($$$ "." |-- scan_nat >> (nat 0 o map symbol)) ~1; |
|
18 | 271 |
in |
27773 | 272 |
scan >> |
273 |
(fn (cs, ~1) => chop_idx (rev cs) [] |
|
274 |
| (cs, i) => (implode cs, i)) |
|
18 | 275 |
end; |
276 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
277 |
in |
18 | 278 |
|
27773 | 279 |
val scan_indexname = $$$ "'" |-- scan_vname >> (fn (x, i) => ("'" ^ x, i)) || scan_vname; |
15443
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
280 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
281 |
end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
282 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
283 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
284 |
(* indexname *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
285 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
286 |
fun read_indexname s = |
27773 | 287 |
(case Scan.read SymbolPos.stopper scan_indexname (SymbolPos.explode (s, Position.none)) of |
15531 | 288 |
SOME xi => xi |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
289 |
| _ => error ("Lexical error in variable name: " ^ quote s)); |
18 | 290 |
|
291 |
||
4703 | 292 |
(* read_var *) |
18 | 293 |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
294 |
fun const c = Const (c, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
295 |
fun free x = Free (x, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
296 |
fun var xi = Var (xi, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
297 |
|
4703 | 298 |
fun read_var str = |
18 | 299 |
let |
300 |
val scan = |
|
27773 | 301 |
$$$ "?" |-- scan_indexname --| Scan.ahead (Scan.one SymbolPos.is_eof) >> var || |
302 |
Scan.many (Symbol.is_regular o symbol) >> (free o implode o map symbol); |
|
303 |
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
|
304 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
305 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
306 |
(* read_variable *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
307 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
308 |
fun read_variable str = |
27773 | 309 |
let val scan = $$$ "?" |-- scan_indexname || scan_indexname |
310 |
in Scan.read SymbolPos.stopper scan (SymbolPos.explode (str, Position.none)) end; |
|
4587 | 311 |
|
312 |
||
19002 | 313 |
(* specific identifiers *) |
5260 | 314 |
|
19002 | 315 |
val constN = "\\<^const>"; |
316 |
val fixedN = "\\<^fixed>"; |
|
317 |
||
5260 | 318 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
319 |
(* read numbers *) |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
320 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
321 |
local |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
322 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
323 |
fun nat cs = |
27773 | 324 |
Option.map (#1 o Library.read_int o map symbol) |
325 |
(Scan.read SymbolPos.stopper scan_nat cs); |
|
5860 | 326 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
327 |
in |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
328 |
|
27773 | 329 |
fun read_nat s = nat (SymbolPos.explode (s, Position.none)); |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
330 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
331 |
fun read_int s = |
27773 | 332 |
(case SymbolPos.explode (s, Position.none) of |
333 |
("-", _) :: cs => Option.map ~ (nat cs) |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
334 |
| cs => nat cs); |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
335 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
336 |
end; |
5860 | 337 |
|
338 |
||
20096 | 339 |
(* read_xnum: hex/bin/decimal *) |
9326 | 340 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
341 |
local |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
342 |
|
20096 | 343 |
val ten = ord "0" + 10; |
344 |
val a = ord "a"; |
|
345 |
val A = ord "A"; |
|
23802 | 346 |
val _ = a > A orelse sys_error "Bad ASCII"; |
20096 | 347 |
|
348 |
fun remap_hex c = |
|
349 |
let val x = ord c in |
|
350 |
if x >= a then chr (x - a + ten) |
|
351 |
else if x >= A then chr (x - A + ten) |
|
352 |
else c |
|
353 |
end; |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
354 |
|
21781 | 355 |
fun leading_zeros ["0"] = 0 |
356 |
| leading_zeros ("0" :: cs) = 1 + leading_zeros cs |
|
357 |
| leading_zeros _ = 0; |
|
358 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
359 |
in |
15965
f422f8283491
Use of IntInf.int instead of int in most numeric simprocs; avoids
paulson
parents:
15570
diff
changeset
|
360 |
|
9326 | 361 |
fun read_xnum str = |
362 |
let |
|
20096 | 363 |
val (sign, radix, digs) = |
364 |
(case Symbol.explode (perhaps (try (unprefix "#")) str) of |
|
365 |
"0" :: "x" :: cs => (1, 16, map remap_hex cs) |
|
366 |
| "0" :: "b" :: cs => (1, 2, cs) |
|
367 |
| "-" :: cs => (~1, 10, cs) |
|
368 |
| cs => (1, 10, cs)); |
|
24630
351a308ab58d
simplified type int (eliminated IntInf.int, integer);
wenzelm
parents:
24583
diff
changeset
|
369 |
val value = sign * #1 (Library.read_radix_int radix digs); |
21781 | 370 |
in {radix = radix, leading_zeros = leading_zeros digs, value = value} end; |
9326 | 371 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
372 |
end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
373 |
|
0 | 374 |
end; |