author | wenzelm |
Fri, 15 Sep 2006 22:56:13 +0200 | |
changeset 20548 | 8ef25fe585a8 |
parent 20313 | bf9101cc4385 |
child 20664 | ffbc5a57191a |
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 |
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
13 |
val implode_xstr: string list -> string |
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
14 |
val explode_xstr: string -> string list |
4703 | 15 |
val scan_id: string list -> string * string list |
16 |
val scan_longid: string list -> string * string list |
|
17 |
val scan_var: string list -> string * string list |
|
18 |
val scan_tid: string list -> string * string list |
|
4902 | 19 |
val scan_tvar: string list -> string * string list |
4703 | 20 |
val scan_nat: string list -> string * string list |
21 |
val scan_int: string list -> string * string list |
|
20067
26bac504ef90
hex and binary numerals (contributed by Rafal Kolanski)
kleing
parents:
19918
diff
changeset
|
22 |
val scan_hex: string list -> string * string list |
26bac504ef90
hex and binary numerals (contributed by Rafal Kolanski)
kleing
parents:
19918
diff
changeset
|
23 |
val scan_bin: string list -> string * string list |
18 | 24 |
val string_of_vname: indexname -> string |
2583
690835a06cf2
added string_of_vname' (treats neg. index as free);
wenzelm
parents:
2363
diff
changeset
|
25 |
val string_of_vname': indexname -> string |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
26 |
val read_indexname: string -> indexname |
4703 | 27 |
val read_var: string -> term |
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
28 |
val read_variable: string -> indexname option |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
29 |
val const: string -> term |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
30 |
val free: string -> term |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
31 |
val var: indexname -> term |
5860 | 32 |
val read_nat: string -> int option |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
33 |
val read_int: string -> int option |
15965
f422f8283491
Use of IntInf.int instead of int in most numeric simprocs; avoids
paulson
parents:
15570
diff
changeset
|
34 |
val read_xnum: string -> IntInf.int |
7784 | 35 |
val read_idents: string -> string list |
19002 | 36 |
val fixedN: string |
37 |
val constN: string |
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
38 |
end; |
0 | 39 |
|
40 |
signature LEXICON = |
|
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
41 |
sig |
18 | 42 |
include LEXICON0 |
43 |
val is_xid: string -> bool |
|
44 |
datatype token = |
|
45 |
Token of string | |
|
46 |
IdentSy of string | |
|
3828 | 47 |
LongIdentSy of string | |
18 | 48 |
VarSy of string | |
49 |
TFreeSy of string | |
|
50 |
TVarSy of string | |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
51 |
NumSy of string | |
11697 | 52 |
XNumSy of string | |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
53 |
StrSy of string | |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
54 |
EndToken |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
55 |
val idT: typ |
3828 | 56 |
val longidT: typ |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
57 |
val varT: typ |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
58 |
val tidT: typ |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
59 |
val tvarT: typ |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
60 |
val terminals: string list |
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
61 |
val is_terminal: string -> bool |
18 | 62 |
val str_of_token: token -> string |
63 |
val display_token: token -> string |
|
64 |
val matching_tokens: token * token -> bool |
|
65 |
val valued_token: token -> bool |
|
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
66 |
val predef_term: string -> token option |
4703 | 67 |
val tokenize: Scan.lexicon -> bool -> string list -> token list |
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
68 |
end; |
0 | 69 |
|
14679 | 70 |
structure Lexicon: LEXICON = |
0 | 71 |
struct |
72 |
||
4247
9bba9251bb4d
added implode_xstr: string list -> string, explode_xstr: string -> string list;
wenzelm
parents:
3828
diff
changeset
|
73 |
|
18 | 74 |
(** is_identifier etc. **) |
75 |
||
16150 | 76 |
val is_identifier = Symbol.is_ident o Symbol.explode; |
18 | 77 |
|
14679 | 78 |
fun is_ascii_identifier s = |
79 |
let val cs = Symbol.explode s |
|
16150 | 80 |
in forall Symbol.is_ascii cs andalso Symbol.is_ident cs end; |
14679 | 81 |
|
18 | 82 |
fun is_xid s = |
4703 | 83 |
(case Symbol.explode s of |
16150 | 84 |
"_" :: cs => Symbol.is_ident cs |
85 |
| cs => Symbol.is_ident cs); |
|
18 | 86 |
|
330
2fda15dd1e0f
changed the way a grammar is generated to allow the new parser to work;
clasohm
parents:
237
diff
changeset
|
87 |
fun is_tid s = |
4703 | 88 |
(case Symbol.explode s of |
16150 | 89 |
"'" :: cs => Symbol.is_ident cs |
18 | 90 |
| _ => false); |
91 |
||
0 | 92 |
|
93 |
||
4703 | 94 |
(** basic scanners **) |
95 |
||
96 |
val scan_letter_letdigs = Scan.one Symbol.is_letter -- Scan.any Symbol.is_letdig >> op ::; |
|
97 |
val scan_digits1 = Scan.any1 Symbol.is_digit; |
|
20067
26bac504ef90
hex and binary numerals (contributed by Rafal Kolanski)
kleing
parents:
19918
diff
changeset
|
98 |
val scan_hex1 = Scan.any1 (Symbol.is_digit orf Symbol.is_hex_letter); |
26bac504ef90
hex and binary numerals (contributed by Rafal Kolanski)
kleing
parents:
19918
diff
changeset
|
99 |
val scan_bin1 = Scan.any1 (fn s => s = "0" orelse s = "1"); |
4703 | 100 |
|
101 |
val scan_id = scan_letter_letdigs >> implode; |
|
102 |
val scan_longid = scan_id ^^ (Scan.repeat1 ($$ "." ^^ scan_id) >> implode); |
|
103 |
val scan_tid = $$ "'" ^^ scan_id; |
|
104 |
||
105 |
val scan_nat = scan_digits1 >> implode; |
|
5513 | 106 |
val scan_int = $$ "-" ^^ scan_nat || scan_nat; |
4703 | 107 |
|
20091 | 108 |
val scan_hex = $$ "0" ^^ $$ "x" ^^ (scan_hex1 >> implode); |
109 |
val scan_bin = $$ "0" ^^ $$ "b" ^^ (scan_bin1 >> implode); |
|
20067
26bac504ef90
hex and binary numerals (contributed by Rafal Kolanski)
kleing
parents:
19918
diff
changeset
|
110 |
|
4703 | 111 |
val scan_id_nat = scan_id ^^ Scan.optional ($$ "." ^^ scan_nat) ""; |
112 |
val scan_var = $$ "?" ^^ scan_id_nat; |
|
4902 | 113 |
val scan_tvar = $$ "?" ^^ $$ "'" ^^ scan_id_nat; |
4703 | 114 |
|
115 |
||
116 |
||
18 | 117 |
(** string_of_vname **) |
0 | 118 |
|
14783 | 119 |
val string_of_vname = Term.string_of_vname; |
120 |
val string_of_vname' = Term.string_of_vname'; |
|
2583
690835a06cf2
added string_of_vname' (treats neg. index as free);
wenzelm
parents:
2363
diff
changeset
|
121 |
|
18 | 122 |
|
0 | 123 |
|
18 | 124 |
(** datatype token **) |
0 | 125 |
|
18 | 126 |
datatype token = |
127 |
Token of string | |
|
128 |
IdentSy of string | |
|
3828 | 129 |
LongIdentSy of string | |
18 | 130 |
VarSy of string | |
131 |
TFreeSy of string | |
|
132 |
TVarSy of string | |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
133 |
NumSy of string | |
11697 | 134 |
XNumSy of string | |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
135 |
StrSy of string | |
18 | 136 |
EndToken; |
0 | 137 |
|
138 |
||
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
139 |
(* terminal arguments *) |
0 | 140 |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
141 |
val idT = Type ("id", []); |
3828 | 142 |
val longidT = Type ("longid", []); |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
143 |
val varT = Type ("var", []); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
144 |
val tidT = Type ("tid", []); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
145 |
val tvarT = Type ("tvar", []); |
0 | 146 |
|
11697 | 147 |
val terminals = ["id", "longid", "var", "tid", "tvar", "num", "xnum", "xstr"]; |
237
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
148 |
|
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
149 |
fun is_terminal s = s mem terminals; |
a7d3e712767a
MAJOR INTERNAL CHANGE: extend and merge operations of syntax tables
wenzelm
parents:
164
diff
changeset
|
150 |
|
0 | 151 |
|
18 | 152 |
(* str_of_token *) |
0 | 153 |
|
18 | 154 |
fun str_of_token (Token s) = s |
155 |
| str_of_token (IdentSy s) = s |
|
3828 | 156 |
| str_of_token (LongIdentSy s) = s |
18 | 157 |
| str_of_token (VarSy s) = s |
158 |
| str_of_token (TFreeSy s) = s |
|
159 |
| str_of_token (TVarSy s) = s |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
160 |
| str_of_token (NumSy s) = s |
11697 | 161 |
| str_of_token (XNumSy s) = s |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
162 |
| str_of_token (StrSy s) = s |
376
d3d01131470f
extended signature SCANNER by some basic scanners and type lexicon;
wenzelm
parents:
330
diff
changeset
|
163 |
| str_of_token EndToken = "EOF"; |
0 | 164 |
|
18 | 165 |
|
166 |
(* display_token *) |
|
0 | 167 |
|
18 | 168 |
fun display_token (Token s) = quote s |
169 |
| display_token (IdentSy s) = "id(" ^ s ^ ")" |
|
3828 | 170 |
| display_token (LongIdentSy s) = "longid(" ^ s ^ ")" |
18 | 171 |
| display_token (VarSy s) = "var(" ^ s ^ ")" |
330
2fda15dd1e0f
changed the way a grammar is generated to allow the new parser to work;
clasohm
parents:
237
diff
changeset
|
172 |
| display_token (TFreeSy s) = "tid(" ^ s ^ ")" |
18 | 173 |
| display_token (TVarSy s) = "tvar(" ^ s ^ ")" |
11697 | 174 |
| display_token (NumSy s) = "num(" ^ s ^ ")" |
175 |
| display_token (XNumSy s) = "xnum(" ^ s ^ ")" |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
176 |
| display_token (StrSy s) = "xstr(" ^ s ^ ")" |
18 | 177 |
| display_token EndToken = ""; |
0 | 178 |
|
18 | 179 |
|
180 |
(* matching_tokens *) |
|
0 | 181 |
|
18 | 182 |
fun matching_tokens (Token x, Token y) = (x = y) |
183 |
| matching_tokens (IdentSy _, IdentSy _) = true |
|
3828 | 184 |
| matching_tokens (LongIdentSy _, LongIdentSy _) = true |
18 | 185 |
| matching_tokens (VarSy _, VarSy _) = true |
186 |
| matching_tokens (TFreeSy _, TFreeSy _) = true |
|
187 |
| matching_tokens (TVarSy _, TVarSy _) = true |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
188 |
| matching_tokens (NumSy _, NumSy _) = true |
11697 | 189 |
| matching_tokens (XNumSy _, XNumSy _) = true |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
190 |
| matching_tokens (StrSy _, StrSy _) = true |
18 | 191 |
| matching_tokens (EndToken, EndToken) = true |
192 |
| matching_tokens _ = false; |
|
0 | 193 |
|
194 |
||
18 | 195 |
(* valued_token *) |
0 | 196 |
|
18 | 197 |
fun valued_token (Token _) = false |
198 |
| valued_token (IdentSy _) = true |
|
3828 | 199 |
| valued_token (LongIdentSy _) = true |
18 | 200 |
| valued_token (VarSy _) = true |
201 |
| valued_token (TFreeSy _) = true |
|
202 |
| valued_token (TVarSy _) = true |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
203 |
| valued_token (NumSy _) = true |
11697 | 204 |
| valued_token (XNumSy _) = true |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
205 |
| valued_token (StrSy _) = true |
18 | 206 |
| valued_token EndToken = false; |
0 | 207 |
|
208 |
||
18 | 209 |
(* predef_term *) |
0 | 210 |
|
15531 | 211 |
fun predef_term "id" = SOME (IdentSy "id") |
212 |
| predef_term "longid" = SOME (LongIdentSy "longid") |
|
213 |
| predef_term "var" = SOME (VarSy "var") |
|
214 |
| predef_term "tid" = SOME (TFreeSy "tid") |
|
215 |
| predef_term "tvar" = SOME (TVarSy "tvar") |
|
216 |
| predef_term "num" = SOME (NumSy "num") |
|
217 |
| predef_term "xnum" = SOME (XNumSy "xnum") |
|
218 |
| predef_term "xstr" = SOME (StrSy "xstr") |
|
219 |
| predef_term _ = NONE; |
|
0 | 220 |
|
221 |
||
4703 | 222 |
(* xstr tokens *) |
18 | 223 |
|
14730 | 224 |
fun lex_err msg prfx (cs, _) = |
225 |
"Inner lexical error: " ^ msg ^ " at " ^ quote (prfx ^ Symbol.beginning 10 cs); |
|
226 |
||
227 |
val scan_chr = |
|
228 |
$$ "\\" |-- Scan.one Symbol.not_eof || |
|
19305 | 229 |
Scan.one (fn s => s <> "'" andalso Symbol.not_eof s) || |
230 |
$$ "'" --| Scan.ahead (~$$ "'"); |
|
14730 | 231 |
|
232 |
val scan_str = |
|
233 |
$$ "'" |-- $$ "'" |-- !! (lex_err "missing end of string" "''") |
|
234 |
(Scan.repeat scan_chr --| $$ "'" --| $$ "'"); |
|
235 |
||
0 | 236 |
|
4703 | 237 |
fun implode_xstr cs = enclose "''" "''" (implode (map (fn "'" => "\\'" | c => c) cs)); |
18 | 238 |
|
4703 | 239 |
fun explode_xstr str = |
5868 | 240 |
(case Scan.read Symbol.stopper scan_str (Symbol.explode str) of |
15531 | 241 |
SOME cs => cs |
5868 | 242 |
| _ => error ("Inner lexical error: literal string expected at " ^ quote str)); |
18 | 243 |
|
244 |
||
14730 | 245 |
(* nested comments *) |
246 |
||
247 |
val scan_cmt = |
|
248 |
Scan.depend (fn d => $$ "(" ^^ $$ "*" >> pair (d + 1)) || |
|
249 |
Scan.depend (fn 0 => Scan.fail | d => $$ "*" ^^ $$ ")" >> pair (d - 1)) || |
|
19305 | 250 |
Scan.lift ($$ "*" --| Scan.ahead (~$$ ")")) || |
251 |
Scan.lift (Scan.one (fn s => s <> "*" andalso Symbol.not_eof s)); |
|
14730 | 252 |
|
14783 | 253 |
val scan_comment = |
254 |
$$ "(" -- $$ "*" -- !! (lex_err "missing end of comment" "(*") |
|
14730 | 255 |
(Scan.pass 0 (Scan.repeat scan_cmt) -- $$ "*" -- $$ ")") |
14783 | 256 |
>> K (); |
14730 | 257 |
|
258 |
||
18 | 259 |
|
260 |
(** tokenize **) |
|
261 |
||
2363 | 262 |
fun tokenize lex xids chs = |
18 | 263 |
let |
264 |
val scan_xid = |
|
265 |
if xids then $$ "_" ^^ scan_id || scan_id |
|
266 |
else scan_id; |
|
267 |
||
20096 | 268 |
val scan_num = scan_hex || scan_bin || scan_int; |
269 |
||
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
270 |
val scan_val = |
4902 | 271 |
scan_tvar >> pair TVarSy || |
4703 | 272 |
scan_var >> pair VarSy || |
273 |
scan_tid >> pair TFreeSy || |
|
20096 | 274 |
scan_num >> pair NumSy || |
275 |
$$ "#" ^^ scan_num >> pair XNumSy || |
|
3828 | 276 |
scan_longid >> pair LongIdentSy || |
18 | 277 |
scan_xid >> pair IdentSy; |
278 |
||
4703 | 279 |
val scan_lit = Scan.literal lex >> (pair Token o implode); |
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
280 |
|
4703 | 281 |
val scan_token = |
15531 | 282 |
scan_comment >> K NONE || |
283 |
Scan.max (op <= o pairself snd) scan_lit scan_val >> (fn (tk, s) => SOME (tk s)) || |
|
284 |
scan_str >> (SOME o StrSy o implode_xstr) || |
|
285 |
Scan.one Symbol.is_blank >> K NONE; |
|
18 | 286 |
in |
4938 | 287 |
(case Scan.error (Scan.finite Symbol.stopper (Scan.repeat scan_token)) chs of |
19482
9f11af8f7ef9
tuned basic list operators (flat, maps, map_filter);
wenzelm
parents:
19305
diff
changeset
|
288 |
(toks, []) => map_filter I toks @ [EndToken] |
4703 | 289 |
| (_, cs) => error ("Inner lexical error at: " ^ quote (implode cs))) |
18 | 290 |
end; |
291 |
||
292 |
||
293 |
||
294 |
(** scan variables **) |
|
295 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
296 |
(* scan_indexname *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
297 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
298 |
local |
18 | 299 |
|
300 |
fun scan_vname chrs = |
|
301 |
let |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
302 |
fun nat n [] = n |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
303 |
| nat n (c :: cs) = nat (n * 10 + (ord c - ord "0")) cs; |
18 | 304 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
305 |
fun idxname cs ds = (implode (rev cs), nat 0 ds); |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
306 |
fun chop_idx [] ds = idxname [] ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
307 |
| chop_idx (cs as (_ :: "\\<^isub>" :: _)) ds = idxname cs ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
308 |
| chop_idx (cs as (_ :: "\\<^isup>" :: _)) ds = idxname cs ds |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
309 |
| chop_idx (c :: cs) ds = |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
310 |
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
|
311 |
else idxname (c :: cs) ds; |
18 | 312 |
|
313 |
val scan = |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
314 |
scan_letter_letdigs -- Scan.optional ($$ "." |-- scan_digits1 >> nat 0) ~1; |
18 | 315 |
in |
316 |
(case scan chrs of |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
317 |
((cs, ~1), cs') => (chop_idx (rev cs) [], cs') |
18 | 318 |
| ((cs, i), cs') => ((implode cs, i), cs')) |
319 |
end; |
|
320 |
||
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
321 |
in |
18 | 322 |
|
15443
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
323 |
val scan_indexname = |
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
324 |
$$ "'" |-- scan_vname >> (fn (x, i) => ("'" ^ x, i)) |
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
325 |
|| scan_vname; |
07f78cc82a73
indexname function now parses type variables as well; changed input
berghofe
parents:
14981
diff
changeset
|
326 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
327 |
end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
328 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
329 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
330 |
(* indexname *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
331 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
332 |
fun read_indexname s = |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
333 |
(case Scan.read Symbol.stopper scan_indexname (Symbol.explode s) of |
15531 | 334 |
SOME xi => xi |
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
335 |
| _ => error ("Lexical error in variable name: " ^ quote s)); |
18 | 336 |
|
337 |
||
4703 | 338 |
(* read_var *) |
18 | 339 |
|
550
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
340 |
fun const c = Const (c, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
341 |
fun free x = Free (x, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
342 |
fun var xi = Var (xi, dummyT); |
353eea6ec232
replaced id, var, tid, tvar by idT, varT, tidT, tvarT;
wenzelm
parents:
376
diff
changeset
|
343 |
|
4703 | 344 |
fun read_var str = |
18 | 345 |
let |
346 |
val scan = |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
347 |
$$ "?" |-- scan_indexname --| Scan.ahead (Scan.one Symbol.is_eof) >> var || |
4703 | 348 |
Scan.any Symbol.not_eof >> (free o implode); |
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
349 |
in the (Scan.read Symbol.stopper scan (Symbol.explode str)) end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
350 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
351 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
352 |
(* read_variable *) |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
353 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
354 |
fun read_variable str = |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
355 |
let val scan = $$ "?" |-- scan_indexname || scan_indexname |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
356 |
in Scan.read Symbol.stopper scan (Symbol.explode str) end; |
4587 | 357 |
|
358 |
||
19002 | 359 |
(* specific identifiers *) |
5260 | 360 |
|
19002 | 361 |
val constN = "\\<^const>"; |
362 |
val fixedN = "\\<^fixed>"; |
|
363 |
||
5260 | 364 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
365 |
(* read numbers *) |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
366 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
367 |
local |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
368 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
369 |
fun nat cs = |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
370 |
Option.map (#1 o Library.read_int) |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
371 |
(Scan.read Symbol.stopper scan_digits1 cs); |
5860 | 372 |
|
20313
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
373 |
in |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
374 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
375 |
val read_nat = nat o Symbol.explode; |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
376 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
377 |
fun read_int s = |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
378 |
(case Symbol.explode s of |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
379 |
"-" :: cs => Option.map ~ (nat cs) |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
380 |
| cs => nat cs); |
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
381 |
|
bf9101cc4385
renamed Syntax.indexname to Syntax.read_indexname;
wenzelm
parents:
20165
diff
changeset
|
382 |
end; |
5860 | 383 |
|
384 |
||
20096 | 385 |
(* read_xnum: hex/bin/decimal *) |
9326 | 386 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
387 |
local |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
388 |
|
20096 | 389 |
val ten = ord "0" + 10; |
390 |
val a = ord "a"; |
|
391 |
val A = ord "A"; |
|
392 |
val _ = assert (a > A) "Bad ASCII"; |
|
393 |
||
394 |
fun remap_hex c = |
|
395 |
let val x = ord c in |
|
396 |
if x >= a then chr (x - a + ten) |
|
397 |
else if x >= A then chr (x - A + ten) |
|
398 |
else c |
|
399 |
end; |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
400 |
|
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
401 |
in |
15965
f422f8283491
Use of IntInf.int instead of int in most numeric simprocs; avoids
paulson
parents:
15570
diff
changeset
|
402 |
|
9326 | 403 |
fun read_xnum str = |
404 |
let |
|
20096 | 405 |
val (sign, radix, digs) = |
406 |
(case Symbol.explode (perhaps (try (unprefix "#")) str) of |
|
407 |
"0" :: "x" :: cs => (1, 16, map remap_hex cs) |
|
408 |
| "0" :: "b" :: cs => (1, 2, cs) |
|
409 |
| "-" :: cs => (~1, 10, cs) |
|
410 |
| cs => (1, 10, cs)); |
|
411 |
in sign * #1 (Library.read_intinf radix digs) end; |
|
9326 | 412 |
|
15991
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
413 |
end; |
670f8e4b5a98
added read_variable: optional question mark on input;
wenzelm
parents:
15965
diff
changeset
|
414 |
|
9326 | 415 |
|
7784 | 416 |
(* read_ident(s) *) |
417 |
||
418 |
fun read_idents str = |
|
419 |
let |
|
420 |
val blanks = Scan.any Symbol.is_blank; |
|
421 |
val junk = Scan.any Symbol.not_eof; |
|
422 |
val idents = Scan.repeat1 (blanks |-- scan_id --| blanks) -- junk; |
|
423 |
in |
|
424 |
(case Scan.read Symbol.stopper idents (Symbol.explode str) of |
|
15531 | 425 |
SOME (ids, []) => ids |
426 |
| SOME (_, bad) => error ("Bad identifier: " ^ quote (implode bad)) |
|
427 |
| NONE => error ("No identifier found in: " ^ quote str)) |
|
7784 | 428 |
end; |
429 |
||
0 | 430 |
end; |