author | haftmann |
Tue, 23 Sep 2008 18:11:44 +0200 | |
changeset 28337 | 93964076e7b8 |
parent 27887 | 9f3fd48cf673 |
child 28407 | f9db1da584ac |
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 = |
27887 | 42 |
Literal | IdentSy | LongIdentSy | VarSy | TFreeSy | TVarSy | |
43 |
NumSy | XNumSy | StrSy | Space | Comment | EOF |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
44 |
datatype token = Token of token_kind * string * Position.range |
27806 | 45 |
val str_of_token: token -> string |
46 |
val pos_of_token: token -> Position.T |
|
27887 | 47 |
val is_proper: token -> bool |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
48 |
val mk_eof: Position.T -> token |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
49 |
val eof: token |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
50 |
val is_eof: token -> bool |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
51 |
val stopper: token Scan.stopper |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
52 |
val idT: typ |
3828 | 53 |
val longidT: typ |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
54 |
val varT: typ |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
55 |
val tidT: typ |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
56 |
val tvarT: typ |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
57 |
val terminals: string list |
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
58 |
val is_terminal: string -> bool |
27887 | 59 |
val report_token: token -> unit |
18 | 60 |
val matching_tokens: token * token -> bool |
61 |
val valued_token: token -> bool |
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
62 |
val predef_term: string -> token option |
27773 | 63 |
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
|
64 |
end; |
0 | 65 |
|
14679 | 66 |
structure Lexicon: LEXICON = |
0 | 67 |
struct |
68 |
||
18 | 69 |
(** is_identifier etc. **) |
70 |
||
16150 | 71 |
val is_identifier = Symbol.is_ident o Symbol.explode; |
18 | 72 |
|
14679 | 73 |
fun is_ascii_identifier s = |
74 |
let val cs = Symbol.explode s |
|
16150 | 75 |
in forall Symbol.is_ascii cs andalso Symbol.is_ident cs end; |
14679 | 76 |
|
18 | 77 |
fun is_xid s = |
4703 | 78 |
(case Symbol.explode s of |
16150 | 79 |
"_" :: cs => Symbol.is_ident cs |
80 |
| cs => Symbol.is_ident cs); |
|
18 | 81 |
|
330
2fda15dd1e0f
changed the way a grammar is generated to allow the new parser to work;
clasohm
parents:
237
diff
changeset
|
82 |
fun is_tid s = |
4703 | 83 |
(case Symbol.explode s of |
16150 | 84 |
"'" :: cs => Symbol.is_ident cs |
18 | 85 |
| _ => false); |
86 |
||
0 | 87 |
|
88 |
||
4703 | 89 |
(** basic scanners **) |
90 |
||
27773 | 91 |
open BasicSymbolPos; |
92 |
||
93 |
fun !!! msg = SymbolPos.!!! ("Inner lexical error: " ^ msg); |
|
94 |
||
95 |
val scan_id = Scan.one (Symbol.is_letter o symbol) ::: Scan.many (Symbol.is_letdig o symbol); |
|
96 |
val scan_longid = scan_id @@@ (Scan.repeat1 ($$$ "." @@@ scan_id) >> flat); |
|
97 |
val scan_tid = $$$ "'" @@@ scan_id; |
|
4703 | 98 |
|
27773 | 99 |
val scan_nat = Scan.many1 (Symbol.is_digit o symbol); |
100 |
val scan_int = $$$ "-" @@@ scan_nat || scan_nat; |
|
101 |
val scan_hex = $$$ "0" @@@ $$$ "x" @@@ Scan.many1 (Symbol.is_ascii_hex o symbol); |
|
102 |
val scan_bin = $$$ "0" @@@ $$$ "b" @@@ Scan.many1 (fn (s, _) => s = "0" orelse s = "1"); |
|
4703 | 103 |
|
27773 | 104 |
val scan_id_nat = scan_id @@@ Scan.optional ($$$ "." @@@ scan_nat) []; |
105 |
val scan_var = $$$ "?" @@@ scan_id_nat; |
|
106 |
val scan_tvar = $$$ "?" @@@ $$$ "'" @@@ scan_id_nat; |
|
4703 | 107 |
|
108 |
||
109 |
||
18 | 110 |
(** datatype token **) |
0 | 111 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
112 |
datatype token_kind = |
27887 | 113 |
Literal | IdentSy | LongIdentSy | VarSy | TFreeSy | TVarSy | |
114 |
NumSy | XNumSy | StrSy | Space | Comment | EOF; |
|
27800
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 |
datatype token = Token of token_kind * string * Position.range; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
117 |
|
27806 | 118 |
fun str_of_token (Token (_, s, _)) = s; |
119 |
fun pos_of_token (Token (_, _, (pos, _))) = pos; |
|
120 |
||
27887 | 121 |
fun is_proper (Token (Space, _, _)) = false |
122 |
| is_proper (Token (Comment, _, _)) = false |
|
123 |
| is_proper _ = true; |
|
124 |
||
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
125 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
126 |
(* stopper *) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
127 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
128 |
fun mk_eof pos = Token (EOF, "", (pos, Position.none)); |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
129 |
val eof = mk_eof Position.none; |
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 is_eof (Token (EOF, _, _)) = true |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
132 |
| is_eof _ = false; |
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 |
val stopper = Scan.stopper (K eof) is_eof; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
135 |
|
0 | 136 |
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
137 |
(* terminal arguments *) |
0 | 138 |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
139 |
val idT = Type ("id", []); |
3828 | 140 |
val longidT = Type ("longid", []); |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
141 |
val varT = Type ("var", []); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
142 |
val tidT = Type ("tid", []); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
143 |
val tvarT = Type ("tvar", []); |
0 | 144 |
|
11697 | 145 |
val terminals = ["id", "longid", "var", "tid", "tvar", "num", "xnum", "xstr"]; |
20664 | 146 |
val is_terminal = member (op =) terminals; |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
147 |
|
0 | 148 |
|
27887 | 149 |
(* markup *) |
150 |
||
151 |
val token_kind_markup = |
|
152 |
fn Literal => Markup.literal |
|
153 |
| IdentSy => Markup.ident |
|
154 |
| LongIdentSy => Markup.ident |
|
155 |
| VarSy => Markup.var |
|
156 |
| TFreeSy => Markup.tfree |
|
157 |
| TVarSy => Markup.tvar |
|
158 |
| NumSy => Markup.num |
|
159 |
| XNumSy => Markup.xnum |
|
160 |
| StrSy => Markup.xstr |
|
161 |
| Space => Markup.none |
|
162 |
| Comment => Markup.inner_comment |
|
163 |
| EOF => Markup.none; |
|
164 |
||
165 |
fun report_token (Token (kind, _, (pos, _))) = |
|
166 |
Position.report (token_kind_markup kind) pos; |
|
167 |
||
168 |
||
18 | 169 |
(* matching_tokens *) |
0 | 170 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
171 |
fun matching_tokens (Token (Literal, x, _), Token (Literal, y, _)) = x = y |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
172 |
| matching_tokens (Token (k, _, _), Token (k', _, _)) = k = k'; |
0 | 173 |
|
174 |
||
18 | 175 |
(* valued_token *) |
0 | 176 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
177 |
fun valued_token (Token (Literal, _, _)) = false |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
178 |
| valued_token (Token (EOF, _, _)) = false |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
179 |
| valued_token _ = true; |
0 | 180 |
|
181 |
||
18 | 182 |
(* predef_term *) |
0 | 183 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
184 |
fun predef_term "id" = SOME (Token (IdentSy, "id", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
185 |
| predef_term "longid" = SOME (Token (LongIdentSy, "longid", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
186 |
| predef_term "var" = SOME (Token (VarSy, "var", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
187 |
| predef_term "tid" = SOME (Token (TFreeSy, "tid", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
188 |
| predef_term "tvar" = SOME (Token (TVarSy, "tvar", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
189 |
| predef_term "num" = SOME (Token (NumSy, "num", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
190 |
| predef_term "xnum" = SOME (Token (XNumSy, "xnum", Position.no_range)) |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
191 |
| predef_term "xstr" = SOME (Token (StrSy, "xstr", Position.no_range)) |
15531 | 192 |
| predef_term _ = NONE; |
0 | 193 |
|
194 |
||
4703 | 195 |
(* xstr tokens *) |
18 | 196 |
|
14730 | 197 |
val scan_chr = |
27773 | 198 |
$$$ "\\" |-- $$$ "'" || |
199 |
Scan.one ((fn s => s <> "\\" andalso s <> "'" andalso Symbol.is_regular s) o symbol) >> single || |
|
200 |
$$$ "'" --| Scan.ahead (~$$$ "'"); |
|
14730 | 201 |
|
202 |
val scan_str = |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
203 |
$$$ "'" @@@ $$$ "'" @@@ !!! "missing end of string" |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
204 |
((Scan.repeat scan_chr >> flat) @@@ $$$ "'" @@@ $$$ "'"); |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
205 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
206 |
val scan_str_body = |
27773 | 207 |
$$$ "'" |-- $$$ "'" |-- !!! "missing end of string" |
208 |
((Scan.repeat scan_chr >> flat) --| $$$ "'" --| $$$ "'"); |
|
14730 | 209 |
|
0 | 210 |
|
4703 | 211 |
fun implode_xstr cs = enclose "''" "''" (implode (map (fn "'" => "\\'" | c => c) cs)); |
18 | 212 |
|
4703 | 213 |
fun explode_xstr str = |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
214 |
(case Scan.read SymbolPos.stopper scan_str_body (SymbolPos.explode (str, Position.none)) of |
27773 | 215 |
SOME cs => map symbol cs |
5868 | 216 |
| _ => error ("Inner lexical error: literal string expected at " ^ quote str)); |
18 | 217 |
|
218 |
||
27806 | 219 |
|
18 | 220 |
(** tokenize **) |
221 |
||
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
222 |
fun token_leq (Token (_, s1, _), Token (_, s2, _)) = s1 <= s2; |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
223 |
fun token kind ss = Token (kind, SymbolPos.content ss, SymbolPos.range ss); |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
224 |
|
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
225 |
fun tokenize lex xids syms = |
18 | 226 |
let |
227 |
val scan_xid = |
|
27773 | 228 |
if xids then $$$ "_" @@@ scan_id || scan_id |
18 | 229 |
else scan_id; |
230 |
||
20096 | 231 |
val scan_num = scan_hex || scan_bin || scan_int; |
232 |
||
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
233 |
val scan_val = |
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
234 |
scan_tvar >> token TVarSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
235 |
scan_var >> token VarSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
236 |
scan_tid >> token TFreeSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
237 |
scan_num >> token NumSy || |
27773 | 238 |
$$$ "#" @@@ scan_num >> token XNumSy || |
26007
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
239 |
scan_longid >> token LongIdentSy || |
3760d3ff4cce
basic scanners: produce symbol list instead of imploded string;
wenzelm
parents:
24630
diff
changeset
|
240 |
scan_xid >> token IdentSy; |
18 | 241 |
|
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
242 |
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
|
243 |
|
4703 | 244 |
val scan_token = |
27887 | 245 |
SymbolPos.scan_comment !!! >> token Comment || |
246 |
Scan.max token_leq scan_lit scan_val || |
|
247 |
scan_str >> token StrSy || |
|
248 |
Scan.many1 (Symbol.is_blank o symbol) >> token Space; |
|
18 | 249 |
in |
27773 | 250 |
(case Scan.error |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
251 |
(Scan.finite SymbolPos.stopper (Scan.repeat scan_token)) syms of |
27887 | 252 |
(toks, []) => toks |
27800
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
253 |
| (_, ss) => error ("Inner lexical error at: " ^ SymbolPos.content ss ^ |
df444ddeff56
datatype token: maintain range, tuned representation;
wenzelm
parents:
27773
diff
changeset
|
254 |
Position.str_of (#1 (SymbolPos.range ss)))) |
18 | 255 |
end; |
256 |
||
257 |
||
258 |
||
259 |
(** scan variables **) |
|
260 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
261 |
(* scan_indexname *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
262 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
263 |
local |
18 | 264 |
|
27773 | 265 |
val scan_vname = |
18 | 266 |
let |
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
267 |
fun nat n [] = n |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
268 |
| nat n (c :: cs) = nat (n * 10 + (ord c - ord "0")) cs; |
18 | 269 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
270 |
fun idxname cs ds = (implode (rev cs), nat 0 ds); |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
271 |
fun chop_idx [] ds = idxname [] ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
272 |
| chop_idx (cs as (_ :: "\\<^isub>" :: _)) ds = idxname cs ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
273 |
| chop_idx (cs as (_ :: "\\<^isup>" :: _)) ds = idxname cs ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
274 |
| chop_idx (c :: cs) ds = |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
275 |
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
|
276 |
else idxname (c :: cs) ds; |
18 | 277 |
|
27773 | 278 |
val scan = (scan_id >> map symbol) -- |
279 |
Scan.optional ($$$ "." |-- scan_nat >> (nat 0 o map symbol)) ~1; |
|
18 | 280 |
in |
27773 | 281 |
scan >> |
282 |
(fn (cs, ~1) => chop_idx (rev cs) [] |
|
283 |
| (cs, i) => (implode cs, i)) |
|
18 | 284 |
end; |
285 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
286 |
in |
18 | 287 |
|
27773 | 288 |
val scan_indexname = $$$ "'" |-- scan_vname >> (fn (x, i) => ("'" ^ x, i)) || scan_vname; |
15443
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
289 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
290 |
end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
291 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
292 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
293 |
(* indexname *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
294 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
295 |
fun read_indexname s = |
27773 | 296 |
(case Scan.read SymbolPos.stopper scan_indexname (SymbolPos.explode (s, Position.none)) of |
15531 | 297 |
SOME xi => xi |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
298 |
| _ => error ("Lexical error in variable name: " ^ quote s)); |
18 | 299 |
|
300 |
||
4703 | 301 |
(* read_var *) |
18 | 302 |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
303 |
fun const c = Const (c, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
304 |
fun free x = Free (x, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
305 |
fun var xi = Var (xi, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
306 |
|
4703 | 307 |
fun read_var str = |
18 | 308 |
let |
309 |
val scan = |
|
27773 | 310 |
$$$ "?" |-- scan_indexname --| Scan.ahead (Scan.one SymbolPos.is_eof) >> var || |
311 |
Scan.many (Symbol.is_regular o symbol) >> (free o implode o map symbol); |
|
312 |
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
|
313 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
314 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
315 |
(* read_variable *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
316 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
317 |
fun read_variable str = |
27773 | 318 |
let val scan = $$$ "?" |-- scan_indexname || scan_indexname |
319 |
in Scan.read SymbolPos.stopper scan (SymbolPos.explode (str, Position.none)) end; |
|
4587 | 320 |
|
321 |
||
19002 | 322 |
(* specific identifiers *) |
5260 | 323 |
|
19002 | 324 |
val constN = "\\<^const>"; |
325 |
val fixedN = "\\<^fixed>"; |
|
326 |
||
5260 | 327 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
328 |
(* read numbers *) |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
329 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
330 |
local |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
331 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
332 |
fun nat cs = |
27773 | 333 |
Option.map (#1 o Library.read_int o map symbol) |
334 |
(Scan.read SymbolPos.stopper scan_nat cs); |
|
5860 | 335 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
336 |
in |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
337 |
|
27773 | 338 |
fun read_nat s = nat (SymbolPos.explode (s, Position.none)); |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
339 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
340 |
fun read_int s = |
27773 | 341 |
(case SymbolPos.explode (s, Position.none) of |
342 |
("-", _) :: cs => Option.map ~ (nat cs) |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
343 |
| cs => nat cs); |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
344 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
345 |
end; |
5860 | 346 |
|
347 |
||
20096 | 348 |
(* read_xnum: hex/bin/decimal *) |
9326 | 349 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
350 |
local |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
351 |
|
20096 | 352 |
val ten = ord "0" + 10; |
353 |
val a = ord "a"; |
|
354 |
val A = ord "A"; |
|
23802 | 355 |
val _ = a > A orelse sys_error "Bad ASCII"; |
20096 | 356 |
|
357 |
fun remap_hex c = |
|
358 |
let val x = ord c in |
|
359 |
if x >= a then chr (x - a + ten) |
|
360 |
else if x >= A then chr (x - A + ten) |
|
361 |
else c |
|
362 |
end; |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
363 |
|
21781 | 364 |
fun leading_zeros ["0"] = 0 |
365 |
| leading_zeros ("0" :: cs) = 1 + leading_zeros cs |
|
366 |
| leading_zeros _ = 0; |
|
367 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
368 |
in |
15965
f422f8283491
Use of IntInf.int instead of int in most numeric simprocs; avoids
paulson
parents:
15570
diff
changeset
|
369 |
|
9326 | 370 |
fun read_xnum str = |
371 |
let |
|
20096 | 372 |
val (sign, radix, digs) = |
373 |
(case Symbol.explode (perhaps (try (unprefix "#")) str) of |
|
374 |
"0" :: "x" :: cs => (1, 16, map remap_hex cs) |
|
375 |
| "0" :: "b" :: cs => (1, 2, cs) |
|
376 |
| "-" :: cs => (~1, 10, cs) |
|
377 |
| cs => (1, 10, cs)); |
|
24630
351a308ab58d
simplified type int (eliminated IntInf.int, integer);
wenzelm
parents:
24583
diff
changeset
|
378 |
val value = sign * #1 (Library.read_radix_int radix digs); |
21781 | 379 |
in {radix = radix, leading_zeros = leading_zeros digs, value = value} end; |
9326 | 380 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
381 |
end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
382 |
|
0 | 383 |
end; |