racc 1.4.6
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitattributes +2 -0
- data/.gitignore +7 -0
- data/COPYING +515 -0
- data/ChangeLog +846 -0
- data/DEPENDS +4 -0
- data/README.en.rdoc +86 -0
- data/README.ja.rdoc +96 -0
- data/Rakefile +15 -0
- data/TODO +5 -0
- data/bin/racc +308 -0
- data/bin/racc2y +195 -0
- data/bin/y2racc +339 -0
- data/doc/en/NEWS.en.rdoc +282 -0
- data/doc/en/command.en.html +78 -0
- data/doc/en/debug.en.rdoc +20 -0
- data/doc/en/grammar.en.rdoc +230 -0
- data/doc/en/index.en.html +10 -0
- data/doc/en/parser.en.rdoc +74 -0
- data/doc/en/usage.en.html +92 -0
- data/doc/ja/NEWS.ja.rdoc +307 -0
- data/doc/ja/command.ja.html +94 -0
- data/doc/ja/debug.ja.rdoc +36 -0
- data/doc/ja/grammar.ja.rdoc +348 -0
- data/doc/ja/index.ja.html +10 -0
- data/doc/ja/parser.ja.rdoc +125 -0
- data/doc/ja/usage.ja.html +414 -0
- data/ext/racc/cparse/MANIFEST +4 -0
- data/ext/racc/cparse/cparse.c +824 -0
- data/ext/racc/cparse/depend +1 -0
- data/ext/racc/cparse/extconf.rb +7 -0
- data/fastcache/extconf.rb +2 -0
- data/fastcache/fastcache.c +185 -0
- data/lib/racc.rb +6 -0
- data/lib/racc/compat.rb +40 -0
- data/lib/racc/debugflags.rb +59 -0
- data/lib/racc/exception.rb +15 -0
- data/lib/racc/grammar.rb +1115 -0
- data/lib/racc/grammarfileparser.rb +559 -0
- data/lib/racc/info.rb +16 -0
- data/lib/racc/iset.rb +91 -0
- data/lib/racc/logfilegenerator.rb +214 -0
- data/lib/racc/parser.rb +439 -0
- data/lib/racc/parserfilegenerator.rb +511 -0
- data/lib/racc/pre-setup +13 -0
- data/lib/racc/sourcetext.rb +34 -0
- data/lib/racc/state.rb +971 -0
- data/lib/racc/statetransitiontable.rb +316 -0
- data/lib/racc/static.rb +5 -0
- data/misc/dist.sh +31 -0
- data/sample/array.y +67 -0
- data/sample/array2.y +59 -0
- data/sample/calc-ja.y +66 -0
- data/sample/calc.y +65 -0
- data/sample/conflict.y +15 -0
- data/sample/hash.y +60 -0
- data/sample/lalr.y +17 -0
- data/sample/lists.y +57 -0
- data/sample/syntax.y +46 -0
- data/sample/yyerr.y +46 -0
- data/setup.rb +1587 -0
- data/tasks/doc.rb +12 -0
- data/tasks/email.rb +55 -0
- data/tasks/file.rb +37 -0
- data/tasks/gem.rb +37 -0
- data/tasks/test.rb +16 -0
- data/test/assets/chk.y +126 -0
- data/test/assets/conf.y +16 -0
- data/test/assets/digraph.y +29 -0
- data/test/assets/echk.y +118 -0
- data/test/assets/err.y +60 -0
- data/test/assets/expect.y +7 -0
- data/test/assets/firstline.y +4 -0
- data/test/assets/ichk.y +102 -0
- data/test/assets/intp.y +546 -0
- data/test/assets/mailp.y +437 -0
- data/test/assets/newsyn.y +25 -0
- data/test/assets/noend.y +4 -0
- data/test/assets/nonass.y +41 -0
- data/test/assets/normal.y +27 -0
- data/test/assets/norule.y +4 -0
- data/test/assets/nullbug1.y +25 -0
- data/test/assets/nullbug2.y +15 -0
- data/test/assets/opt.y +123 -0
- data/test/assets/percent.y +35 -0
- data/test/assets/recv.y +97 -0
- data/test/assets/rrconf.y +14 -0
- data/test/assets/scan.y +72 -0
- data/test/assets/syntax.y +50 -0
- data/test/assets/unterm.y +5 -0
- data/test/assets/useless.y +12 -0
- data/test/assets/yyerr.y +46 -0
- data/test/bench.y +36 -0
- data/test/helper.rb +88 -0
- data/test/infini.y +8 -0
- data/test/scandata/brace +7 -0
- data/test/scandata/gvar +1 -0
- data/test/scandata/normal +4 -0
- data/test/scandata/percent +18 -0
- data/test/scandata/slash +10 -0
- data/test/src.intp +34 -0
- data/test/start.y +20 -0
- data/test/test_chk_y.rb +51 -0
- data/test/test_grammar_file_parser.rb +15 -0
- data/test/test_racc_command.rb +155 -0
- data/test/test_scan_y.rb +51 -0
- data/test/testscanner.rb +51 -0
- data/web/racc.en.rhtml +42 -0
- data/web/racc.ja.rhtml +51 -0
- metadata +166 -0
@@ -0,0 +1,824 @@
|
|
1
|
+
/*
|
2
|
+
|
3
|
+
cparse.c -- Racc Runtime Core
|
4
|
+
|
5
|
+
Copyright (c) 1999-2006 Minero Aoki
|
6
|
+
|
7
|
+
This library is free software.
|
8
|
+
You can distribute/modify this program under the same terms of ruby.
|
9
|
+
|
10
|
+
$originalId: cparse.c,v 1.8 2006/07/06 11:39:46 aamine Exp $
|
11
|
+
|
12
|
+
*/
|
13
|
+
|
14
|
+
#include <ruby.h>
|
15
|
+
|
16
|
+
/* -----------------------------------------------------------------------
|
17
|
+
Important Constants
|
18
|
+
----------------------------------------------------------------------- */
|
19
|
+
|
20
|
+
#define RACC_VERSION "1.4.5"
|
21
|
+
|
22
|
+
#define DEFAULT_TOKEN -1
|
23
|
+
#define ERROR_TOKEN 1
|
24
|
+
#define FINAL_TOKEN 0
|
25
|
+
|
26
|
+
#define vDEFAULT_TOKEN INT2FIX(DEFAULT_TOKEN)
|
27
|
+
#define vERROR_TOKEN INT2FIX(ERROR_TOKEN)
|
28
|
+
#define vFINAL_TOKEN INT2FIX(FINAL_TOKEN)
|
29
|
+
|
30
|
+
/* -----------------------------------------------------------------------
|
31
|
+
File Local Variables
|
32
|
+
----------------------------------------------------------------------- */
|
33
|
+
|
34
|
+
static VALUE RaccBug;
|
35
|
+
static VALUE CparseParams;
|
36
|
+
|
37
|
+
static ID id_yydebug;
|
38
|
+
static ID id_nexttoken;
|
39
|
+
static ID id_onerror;
|
40
|
+
static ID id_noreduce;
|
41
|
+
static ID id_errstatus;
|
42
|
+
|
43
|
+
static ID id_d_shift;
|
44
|
+
static ID id_d_reduce;
|
45
|
+
static ID id_d_accept;
|
46
|
+
static ID id_d_read_token;
|
47
|
+
static ID id_d_next_state;
|
48
|
+
static ID id_d_e_pop;
|
49
|
+
|
50
|
+
/* -----------------------------------------------------------------------
|
51
|
+
Utils
|
52
|
+
----------------------------------------------------------------------- */
|
53
|
+
|
54
|
+
/* For backward compatibility */
|
55
|
+
#ifndef ID2SYM
|
56
|
+
# define ID2SYM(i) ULONG2NUM(i)
|
57
|
+
#endif
|
58
|
+
#ifndef SYM2ID
|
59
|
+
# define SYM2ID(v) ((ID)NUM2ULONG(v))
|
60
|
+
#endif
|
61
|
+
#ifndef SYMBOL_P
|
62
|
+
# define SYMBOL_P(v) FIXNUM_P(v)
|
63
|
+
#endif
|
64
|
+
#ifndef LONG2NUM
|
65
|
+
# define LONG2NUM(i) INT2NUM(i)
|
66
|
+
#endif
|
67
|
+
|
68
|
+
static ID value_to_id _((VALUE v));
|
69
|
+
static inline long num_to_long _((VALUE n));
|
70
|
+
|
71
|
+
static ID
|
72
|
+
value_to_id(VALUE v)
|
73
|
+
{
|
74
|
+
if (! SYMBOL_P(v)) {
|
75
|
+
rb_raise(rb_eTypeError, "not symbol");
|
76
|
+
}
|
77
|
+
return SYM2ID(v);
|
78
|
+
}
|
79
|
+
|
80
|
+
static inline long
|
81
|
+
num_to_long(VALUE n)
|
82
|
+
{
|
83
|
+
return NUM2LONG(n);
|
84
|
+
}
|
85
|
+
|
86
|
+
#define AREF(s, idx) \
|
87
|
+
((0 <= idx && idx < RARRAY_LEN(s)) ? RARRAY_PTR(s)[idx] : Qnil)
|
88
|
+
|
89
|
+
/* -----------------------------------------------------------------------
|
90
|
+
Parser Stack Interfaces
|
91
|
+
----------------------------------------------------------------------- */
|
92
|
+
|
93
|
+
static VALUE get_stack_tail _((VALUE stack, long len));
|
94
|
+
static void cut_stack_tail _((VALUE stack, long len));
|
95
|
+
|
96
|
+
static VALUE
|
97
|
+
get_stack_tail(VALUE stack, long len)
|
98
|
+
{
|
99
|
+
if (len < 0) return Qnil; /* system error */
|
100
|
+
if (len > RARRAY_LEN(stack)) len = RARRAY_LEN(stack);
|
101
|
+
return rb_ary_new4(len, RARRAY_PTR(stack) + RARRAY_LEN(stack) - len);
|
102
|
+
}
|
103
|
+
|
104
|
+
static void
|
105
|
+
cut_stack_tail(VALUE stack, long len)
|
106
|
+
{
|
107
|
+
while (len > 0) {
|
108
|
+
rb_ary_pop(stack);
|
109
|
+
len--;
|
110
|
+
}
|
111
|
+
}
|
112
|
+
|
113
|
+
#define STACK_INIT_LEN 64
|
114
|
+
#define NEW_STACK() rb_ary_new2(STACK_INIT_LEN)
|
115
|
+
#define PUSH(s, i) rb_ary_store(s, RARRAY_LEN(s), i)
|
116
|
+
#define POP(s) rb_ary_pop(s)
|
117
|
+
#define LAST_I(s) \
|
118
|
+
((RARRAY_LEN(s) > 0) ? RARRAY_PTR(s)[RARRAY_LEN(s) - 1] : Qnil)
|
119
|
+
#define GET_TAIL(s, len) get_stack_tail(s, len)
|
120
|
+
#define CUT_TAIL(s, len) cut_stack_tail(s, len)
|
121
|
+
|
122
|
+
/* -----------------------------------------------------------------------
|
123
|
+
struct cparse_params
|
124
|
+
----------------------------------------------------------------------- */
|
125
|
+
|
126
|
+
struct cparse_params {
|
127
|
+
VALUE value_v; /* VALUE version of this struct */
|
128
|
+
|
129
|
+
VALUE parser; /* parser object */
|
130
|
+
|
131
|
+
int lex_is_iterator;
|
132
|
+
VALUE lexer; /* scanner object */
|
133
|
+
ID lexmid; /* name of scanner method (must be an iterator) */
|
134
|
+
|
135
|
+
/* State transition tables (immutable)
|
136
|
+
Data structure is from Dragon Book 4.9 */
|
137
|
+
/* action table */
|
138
|
+
VALUE action_table;
|
139
|
+
VALUE action_check;
|
140
|
+
VALUE action_default;
|
141
|
+
VALUE action_pointer;
|
142
|
+
/* goto table */
|
143
|
+
VALUE goto_table;
|
144
|
+
VALUE goto_check;
|
145
|
+
VALUE goto_default;
|
146
|
+
VALUE goto_pointer;
|
147
|
+
|
148
|
+
long nt_base; /* NonTerminal BASE index */
|
149
|
+
VALUE reduce_table; /* reduce data table */
|
150
|
+
VALUE token_table; /* token conversion table */
|
151
|
+
|
152
|
+
/* parser stacks and parameters */
|
153
|
+
VALUE state;
|
154
|
+
long curstate;
|
155
|
+
VALUE vstack;
|
156
|
+
VALUE tstack;
|
157
|
+
VALUE t;
|
158
|
+
long shift_n;
|
159
|
+
long reduce_n;
|
160
|
+
long ruleno;
|
161
|
+
|
162
|
+
long errstatus; /* nonzero in error recovering mode */
|
163
|
+
long nerr; /* number of error */
|
164
|
+
|
165
|
+
int use_result_var;
|
166
|
+
|
167
|
+
VALUE retval; /* return value of parser routine */
|
168
|
+
long fin; /* parse result status */
|
169
|
+
#define CP_FIN_ACCEPT 1
|
170
|
+
#define CP_FIN_EOT 2
|
171
|
+
#define CP_FIN_CANTPOP 3
|
172
|
+
|
173
|
+
int debug; /* user level debug */
|
174
|
+
int sys_debug; /* system level debug */
|
175
|
+
|
176
|
+
long i; /* table index */
|
177
|
+
};
|
178
|
+
|
179
|
+
/* -----------------------------------------------------------------------
|
180
|
+
Parser Main Routines
|
181
|
+
----------------------------------------------------------------------- */
|
182
|
+
|
183
|
+
static VALUE racc_cparse _((VALUE parser, VALUE arg, VALUE sysdebug));
|
184
|
+
static VALUE racc_yyparse _((VALUE parser, VALUE lexer, VALUE lexmid,
|
185
|
+
VALUE arg, VALUE sysdebug));
|
186
|
+
|
187
|
+
static void call_lexer _((struct cparse_params *v));
|
188
|
+
static VALUE lexer_i _((VALUE block_args, VALUE data, VALUE self));
|
189
|
+
|
190
|
+
static VALUE assert_array _((VALUE a));
|
191
|
+
static long assert_integer _((VALUE n));
|
192
|
+
static VALUE assert_hash _((VALUE h));
|
193
|
+
static VALUE initialize_params _((VALUE vparams, VALUE parser, VALUE arg,
|
194
|
+
VALUE lexer, VALUE lexmid));
|
195
|
+
static void cparse_params_mark _((void *ptr));
|
196
|
+
|
197
|
+
static void parse_main _((struct cparse_params *v,
|
198
|
+
VALUE tok, VALUE val, int resume));
|
199
|
+
static void extract_user_token _((struct cparse_params *v,
|
200
|
+
VALUE block_args, VALUE *tok, VALUE *val));
|
201
|
+
static void shift _((struct cparse_params* v, long act, VALUE tok, VALUE val));
|
202
|
+
static int reduce _((struct cparse_params* v, long act));
|
203
|
+
static VALUE reduce0 _((VALUE block_args, VALUE data, VALUE self));
|
204
|
+
|
205
|
+
#ifdef DEBUG
|
206
|
+
# define D_puts(msg) if (v->sys_debug) puts(msg)
|
207
|
+
# define D_printf(fmt,arg) if (v->sys_debug) printf(fmt,arg)
|
208
|
+
#else
|
209
|
+
# define D_puts(msg)
|
210
|
+
# define D_printf(fmt,arg)
|
211
|
+
#endif
|
212
|
+
|
213
|
+
static VALUE
|
214
|
+
racc_cparse(VALUE parser, VALUE arg, VALUE sysdebug)
|
215
|
+
{
|
216
|
+
volatile VALUE vparams;
|
217
|
+
struct cparse_params *v;
|
218
|
+
|
219
|
+
vparams = Data_Make_Struct(CparseParams, struct cparse_params,
|
220
|
+
cparse_params_mark, -1, v);
|
221
|
+
D_puts("starting cparse");
|
222
|
+
v->sys_debug = RTEST(sysdebug);
|
223
|
+
vparams = initialize_params(vparams, parser, arg, Qnil, Qnil);
|
224
|
+
v->lex_is_iterator = Qfalse;
|
225
|
+
parse_main(v, Qnil, Qnil, 0);
|
226
|
+
|
227
|
+
return v->retval;
|
228
|
+
}
|
229
|
+
|
230
|
+
static VALUE
|
231
|
+
racc_yyparse(VALUE parser, VALUE lexer, VALUE lexmid, VALUE arg, VALUE sysdebug)
|
232
|
+
{
|
233
|
+
volatile VALUE vparams;
|
234
|
+
struct cparse_params *v;
|
235
|
+
|
236
|
+
vparams = Data_Make_Struct(CparseParams, struct cparse_params,
|
237
|
+
cparse_params_mark, -1, v);
|
238
|
+
v->sys_debug = RTEST(sysdebug);
|
239
|
+
D_puts("start C yyparse");
|
240
|
+
vparams = initialize_params(vparams, parser, arg, lexer, lexmid);
|
241
|
+
v->lex_is_iterator = Qtrue;
|
242
|
+
D_puts("params initialized");
|
243
|
+
parse_main(v, Qnil, Qnil, 0);
|
244
|
+
call_lexer(v);
|
245
|
+
if (!v->fin) {
|
246
|
+
rb_raise(rb_eArgError, "%s() is finished before EndOfToken",
|
247
|
+
rb_id2name(v->lexmid));
|
248
|
+
}
|
249
|
+
|
250
|
+
return v->retval;
|
251
|
+
}
|
252
|
+
|
253
|
+
#ifdef HAVE_RB_BLOCK_CALL
|
254
|
+
static void
|
255
|
+
call_lexer(struct cparse_params *v)
|
256
|
+
{
|
257
|
+
rb_block_call(v->lexer, v->lexmid, 0, NULL, lexer_i, v->value_v);
|
258
|
+
}
|
259
|
+
#else
|
260
|
+
static VALUE
|
261
|
+
lexer_iter(VALUE data)
|
262
|
+
{
|
263
|
+
struct cparse_params *v;
|
264
|
+
|
265
|
+
Data_Get_Struct(data, struct cparse_params, v);
|
266
|
+
rb_funcall(v->lexer, v->lexmid, 0);
|
267
|
+
return Qnil;
|
268
|
+
}
|
269
|
+
|
270
|
+
static void
|
271
|
+
call_lexer(struct cparse_params *v)
|
272
|
+
{
|
273
|
+
rb_iterate(lexer_iter, v->value_v, lexer_i, v->value_v);
|
274
|
+
}
|
275
|
+
#endif
|
276
|
+
|
277
|
+
static VALUE
|
278
|
+
lexer_i(VALUE block_args, VALUE data, VALUE self)
|
279
|
+
{
|
280
|
+
struct cparse_params *v;
|
281
|
+
VALUE tok, val;
|
282
|
+
|
283
|
+
Data_Get_Struct(data, struct cparse_params, v);
|
284
|
+
if (v->fin)
|
285
|
+
rb_raise(rb_eArgError, "extra token after EndOfToken");
|
286
|
+
extract_user_token(v, block_args, &tok, &val);
|
287
|
+
parse_main(v, tok, val, 1);
|
288
|
+
if (v->fin && v->fin != CP_FIN_ACCEPT)
|
289
|
+
rb_iter_break();
|
290
|
+
return Qnil;
|
291
|
+
}
|
292
|
+
|
293
|
+
static VALUE
|
294
|
+
assert_array(VALUE a)
|
295
|
+
{
|
296
|
+
Check_Type(a, T_ARRAY);
|
297
|
+
return a;
|
298
|
+
}
|
299
|
+
|
300
|
+
static VALUE
|
301
|
+
assert_hash(VALUE h)
|
302
|
+
{
|
303
|
+
Check_Type(h, T_HASH);
|
304
|
+
return h;
|
305
|
+
}
|
306
|
+
|
307
|
+
static long
|
308
|
+
assert_integer(VALUE n)
|
309
|
+
{
|
310
|
+
return NUM2LONG(n);
|
311
|
+
}
|
312
|
+
|
313
|
+
static VALUE
|
314
|
+
initialize_params(VALUE vparams, VALUE parser, VALUE arg, VALUE lexer, VALUE lexmid)
|
315
|
+
{
|
316
|
+
struct cparse_params *v;
|
317
|
+
|
318
|
+
Data_Get_Struct(vparams, struct cparse_params, v);
|
319
|
+
v->value_v = vparams;
|
320
|
+
v->parser = parser;
|
321
|
+
v->lexer = lexer;
|
322
|
+
if (! NIL_P(lexmid))
|
323
|
+
v->lexmid = value_to_id(lexmid);
|
324
|
+
|
325
|
+
v->debug = RTEST(rb_ivar_get(parser, id_yydebug));
|
326
|
+
|
327
|
+
Check_Type(arg, T_ARRAY);
|
328
|
+
if (!(13 <= RARRAY_LEN(arg) && RARRAY_LEN(arg) <= 14))
|
329
|
+
rb_raise(RaccBug, "[Racc Bug] wrong arg.size %ld", RARRAY_LEN(arg));
|
330
|
+
v->action_table = assert_array (RARRAY_PTR(arg)[ 0]);
|
331
|
+
v->action_check = assert_array (RARRAY_PTR(arg)[ 1]);
|
332
|
+
v->action_default = assert_array (RARRAY_PTR(arg)[ 2]);
|
333
|
+
v->action_pointer = assert_array (RARRAY_PTR(arg)[ 3]);
|
334
|
+
v->goto_table = assert_array (RARRAY_PTR(arg)[ 4]);
|
335
|
+
v->goto_check = assert_array (RARRAY_PTR(arg)[ 5]);
|
336
|
+
v->goto_default = assert_array (RARRAY_PTR(arg)[ 6]);
|
337
|
+
v->goto_pointer = assert_array (RARRAY_PTR(arg)[ 7]);
|
338
|
+
v->nt_base = assert_integer(RARRAY_PTR(arg)[ 8]);
|
339
|
+
v->reduce_table = assert_array (RARRAY_PTR(arg)[ 9]);
|
340
|
+
v->token_table = assert_hash (RARRAY_PTR(arg)[10]);
|
341
|
+
v->shift_n = assert_integer(RARRAY_PTR(arg)[11]);
|
342
|
+
v->reduce_n = assert_integer(RARRAY_PTR(arg)[12]);
|
343
|
+
if (RARRAY_LEN(arg) > 13) {
|
344
|
+
v->use_result_var = RTEST(RARRAY_PTR(arg)[13]);
|
345
|
+
}
|
346
|
+
else {
|
347
|
+
v->use_result_var = Qtrue;
|
348
|
+
}
|
349
|
+
|
350
|
+
v->tstack = v->debug ? NEW_STACK() : Qnil;
|
351
|
+
v->vstack = NEW_STACK();
|
352
|
+
v->state = NEW_STACK();
|
353
|
+
v->curstate = 0;
|
354
|
+
PUSH(v->state, INT2FIX(0));
|
355
|
+
v->t = INT2FIX(FINAL_TOKEN + 1); /* must not init to FINAL_TOKEN */
|
356
|
+
v->nerr = 0;
|
357
|
+
v->errstatus = 0;
|
358
|
+
rb_ivar_set(parser, id_errstatus, LONG2NUM(v->errstatus));
|
359
|
+
|
360
|
+
v->retval = Qnil;
|
361
|
+
v->fin = 0;
|
362
|
+
|
363
|
+
v->lex_is_iterator = Qfalse;
|
364
|
+
|
365
|
+
rb_iv_set(parser, "@vstack", v->vstack);
|
366
|
+
if (v->debug) {
|
367
|
+
rb_iv_set(parser, "@tstack", v->tstack);
|
368
|
+
}
|
369
|
+
else {
|
370
|
+
rb_iv_set(parser, "@tstack", Qnil);
|
371
|
+
}
|
372
|
+
|
373
|
+
return vparams;
|
374
|
+
}
|
375
|
+
|
376
|
+
static void
|
377
|
+
cparse_params_mark(void *ptr)
|
378
|
+
{
|
379
|
+
struct cparse_params *v = (struct cparse_params*)ptr;
|
380
|
+
|
381
|
+
rb_gc_mark(v->value_v);
|
382
|
+
rb_gc_mark(v->parser);
|
383
|
+
rb_gc_mark(v->lexer);
|
384
|
+
rb_gc_mark(v->action_table);
|
385
|
+
rb_gc_mark(v->action_check);
|
386
|
+
rb_gc_mark(v->action_default);
|
387
|
+
rb_gc_mark(v->action_pointer);
|
388
|
+
rb_gc_mark(v->goto_table);
|
389
|
+
rb_gc_mark(v->goto_check);
|
390
|
+
rb_gc_mark(v->goto_default);
|
391
|
+
rb_gc_mark(v->goto_pointer);
|
392
|
+
rb_gc_mark(v->reduce_table);
|
393
|
+
rb_gc_mark(v->token_table);
|
394
|
+
rb_gc_mark(v->state);
|
395
|
+
rb_gc_mark(v->vstack);
|
396
|
+
rb_gc_mark(v->tstack);
|
397
|
+
rb_gc_mark(v->t);
|
398
|
+
rb_gc_mark(v->retval);
|
399
|
+
}
|
400
|
+
|
401
|
+
static void
|
402
|
+
extract_user_token(struct cparse_params *v, VALUE block_args,
|
403
|
+
VALUE *tok, VALUE *val)
|
404
|
+
{
|
405
|
+
if (NIL_P(block_args)) {
|
406
|
+
/* EOF */
|
407
|
+
*tok = Qfalse;
|
408
|
+
*val = rb_str_new("$", 1);
|
409
|
+
return;
|
410
|
+
}
|
411
|
+
|
412
|
+
if (TYPE(block_args) != T_ARRAY) {
|
413
|
+
rb_raise(rb_eTypeError,
|
414
|
+
"%s() %s %s (must be Array[2])",
|
415
|
+
v->lex_is_iterator ? rb_id2name(v->lexmid) : "next_token",
|
416
|
+
v->lex_is_iterator ? "yielded" : "returned",
|
417
|
+
rb_class2name(CLASS_OF(block_args)));
|
418
|
+
}
|
419
|
+
if (RARRAY_LEN(block_args) != 2) {
|
420
|
+
rb_raise(rb_eArgError,
|
421
|
+
"%s() %s wrong size of array (%ld for 2)",
|
422
|
+
v->lex_is_iterator ? rb_id2name(v->lexmid) : "next_token",
|
423
|
+
v->lex_is_iterator ? "yielded" : "returned",
|
424
|
+
RARRAY_LEN(block_args));
|
425
|
+
}
|
426
|
+
*tok = AREF(block_args, 0);
|
427
|
+
*val = AREF(block_args, 1);
|
428
|
+
}
|
429
|
+
|
430
|
+
#define SHIFT(v,act,tok,val) shift(v,act,tok,val)
|
431
|
+
#define REDUCE(v,act) do {\
|
432
|
+
switch (reduce(v,act)) { \
|
433
|
+
case 0: /* normal */ \
|
434
|
+
break; \
|
435
|
+
case 1: /* yyerror */ \
|
436
|
+
goto user_yyerror; \
|
437
|
+
case 2: /* yyaccept */ \
|
438
|
+
D_puts("u accept"); \
|
439
|
+
goto accept; \
|
440
|
+
default: \
|
441
|
+
break; \
|
442
|
+
} \
|
443
|
+
} while (0)
|
444
|
+
|
445
|
+
static void
|
446
|
+
parse_main(struct cparse_params *v, VALUE tok, VALUE val, int resume)
|
447
|
+
{
|
448
|
+
long i; /* table index */
|
449
|
+
long act; /* action type */
|
450
|
+
VALUE act_value; /* action type, VALUE version */
|
451
|
+
int read_next = 1; /* true if we need to read next token */
|
452
|
+
VALUE tmp;
|
453
|
+
|
454
|
+
if (resume)
|
455
|
+
goto resume;
|
456
|
+
|
457
|
+
while (1) {
|
458
|
+
D_puts("");
|
459
|
+
D_puts("---- enter new loop ----");
|
460
|
+
D_puts("");
|
461
|
+
|
462
|
+
D_printf("(act) k1=%ld\n", v->curstate);
|
463
|
+
tmp = AREF(v->action_pointer, v->curstate);
|
464
|
+
if (NIL_P(tmp)) goto notfound;
|
465
|
+
D_puts("(act) pointer[k1] ok");
|
466
|
+
i = NUM2LONG(tmp);
|
467
|
+
|
468
|
+
D_printf("read_next=%d\n", read_next);
|
469
|
+
if (read_next && (v->t != vFINAL_TOKEN)) {
|
470
|
+
if (v->lex_is_iterator) {
|
471
|
+
D_puts("resuming...");
|
472
|
+
if (v->fin) rb_raise(rb_eArgError, "token given after EOF");
|
473
|
+
v->i = i; /* save i */
|
474
|
+
return;
|
475
|
+
resume:
|
476
|
+
D_puts("resumed");
|
477
|
+
i = v->i; /* load i */
|
478
|
+
}
|
479
|
+
else {
|
480
|
+
D_puts("next_token");
|
481
|
+
tmp = rb_funcall(v->parser, id_nexttoken, 0);
|
482
|
+
extract_user_token(v, tmp, &tok, &val);
|
483
|
+
}
|
484
|
+
/* convert token */
|
485
|
+
v->t = rb_hash_aref(v->token_table, tok);
|
486
|
+
if (NIL_P(v->t)) {
|
487
|
+
v->t = vERROR_TOKEN;
|
488
|
+
}
|
489
|
+
D_printf("(act) t(k2)=%ld\n", NUM2LONG(v->t));
|
490
|
+
if (v->debug) {
|
491
|
+
rb_funcall(v->parser, id_d_read_token,
|
492
|
+
3, v->t, tok, val);
|
493
|
+
}
|
494
|
+
}
|
495
|
+
read_next = 0;
|
496
|
+
|
497
|
+
i += NUM2LONG(v->t);
|
498
|
+
D_printf("(act) i=%ld\n", i);
|
499
|
+
if (i < 0) goto notfound;
|
500
|
+
|
501
|
+
act_value = AREF(v->action_table, i);
|
502
|
+
if (NIL_P(act_value)) goto notfound;
|
503
|
+
act = NUM2LONG(act_value);
|
504
|
+
D_printf("(act) table[i]=%ld\n", act);
|
505
|
+
|
506
|
+
tmp = AREF(v->action_check, i);
|
507
|
+
if (NIL_P(tmp)) goto notfound;
|
508
|
+
if (NUM2LONG(tmp) != v->curstate) goto notfound;
|
509
|
+
D_printf("(act) check[i]=%ld\n", NUM2LONG(tmp));
|
510
|
+
|
511
|
+
D_puts("(act) found");
|
512
|
+
act_fixed:
|
513
|
+
D_printf("act=%ld\n", act);
|
514
|
+
goto handle_act;
|
515
|
+
|
516
|
+
notfound:
|
517
|
+
D_puts("(act) not found: use default");
|
518
|
+
act_value = AREF(v->action_default, v->curstate);
|
519
|
+
act = NUM2LONG(act_value);
|
520
|
+
goto act_fixed;
|
521
|
+
|
522
|
+
|
523
|
+
handle_act:
|
524
|
+
if (act > 0 && act < v->shift_n) {
|
525
|
+
D_puts("shift");
|
526
|
+
if (v->errstatus > 0) {
|
527
|
+
v->errstatus--;
|
528
|
+
rb_ivar_set(v->parser, id_errstatus, LONG2NUM(v->errstatus));
|
529
|
+
}
|
530
|
+
SHIFT(v, act, v->t, val);
|
531
|
+
read_next = 1;
|
532
|
+
}
|
533
|
+
else if (act < 0 && act > -(v->reduce_n)) {
|
534
|
+
D_puts("reduce");
|
535
|
+
REDUCE(v, act);
|
536
|
+
}
|
537
|
+
else if (act == -(v->reduce_n)) {
|
538
|
+
goto error;
|
539
|
+
error_recovered:
|
540
|
+
; /* goto label requires stmt */
|
541
|
+
}
|
542
|
+
else if (act == v->shift_n) {
|
543
|
+
D_puts("accept");
|
544
|
+
goto accept;
|
545
|
+
}
|
546
|
+
else {
|
547
|
+
rb_raise(RaccBug, "[Racc Bug] unknown act value %ld", act);
|
548
|
+
}
|
549
|
+
|
550
|
+
if (v->debug) {
|
551
|
+
rb_funcall(v->parser, id_d_next_state,
|
552
|
+
2, LONG2NUM(v->curstate), v->state);
|
553
|
+
}
|
554
|
+
}
|
555
|
+
/* not reach */
|
556
|
+
|
557
|
+
|
558
|
+
accept:
|
559
|
+
if (v->debug) rb_funcall(v->parser, id_d_accept, 0);
|
560
|
+
v->retval = RARRAY_PTR(v->vstack)[0];
|
561
|
+
v->fin = CP_FIN_ACCEPT;
|
562
|
+
return;
|
563
|
+
|
564
|
+
|
565
|
+
error:
|
566
|
+
D_printf("error detected, status=%ld\n", v->errstatus);
|
567
|
+
if (v->errstatus == 0) {
|
568
|
+
v->nerr++;
|
569
|
+
rb_funcall(v->parser, id_onerror,
|
570
|
+
3, v->t, val, v->vstack);
|
571
|
+
}
|
572
|
+
user_yyerror:
|
573
|
+
if (v->errstatus == 3) {
|
574
|
+
if (v->t == vFINAL_TOKEN) {
|
575
|
+
v->retval = Qfalse;
|
576
|
+
v->fin = CP_FIN_EOT;
|
577
|
+
return;
|
578
|
+
}
|
579
|
+
read_next = 1;
|
580
|
+
}
|
581
|
+
v->errstatus = 3;
|
582
|
+
rb_ivar_set(v->parser, id_errstatus, LONG2NUM(v->errstatus));
|
583
|
+
|
584
|
+
/* check if we can shift/reduce error token */
|
585
|
+
D_printf("(err) k1=%ld\n", v->curstate);
|
586
|
+
D_printf("(err) k2=%d (error)\n", ERROR_TOKEN);
|
587
|
+
while (1) {
|
588
|
+
tmp = AREF(v->action_pointer, v->curstate);
|
589
|
+
if (NIL_P(tmp)) goto error_pop;
|
590
|
+
D_puts("(err) pointer[k1] ok");
|
591
|
+
|
592
|
+
i = NUM2LONG(tmp) + ERROR_TOKEN;
|
593
|
+
D_printf("(err) i=%ld\n", i);
|
594
|
+
if (i < 0) goto error_pop;
|
595
|
+
|
596
|
+
act_value = AREF(v->action_table, i);
|
597
|
+
if (NIL_P(act_value)) {
|
598
|
+
D_puts("(err) table[i] == nil");
|
599
|
+
goto error_pop;
|
600
|
+
}
|
601
|
+
act = NUM2LONG(act_value);
|
602
|
+
D_printf("(err) table[i]=%ld\n", act);
|
603
|
+
|
604
|
+
tmp = AREF(v->action_check, i);
|
605
|
+
if (NIL_P(tmp)) {
|
606
|
+
D_puts("(err) check[i] == nil");
|
607
|
+
goto error_pop;
|
608
|
+
}
|
609
|
+
if (NUM2LONG(tmp) != v->curstate) {
|
610
|
+
D_puts("(err) check[i] != k1");
|
611
|
+
goto error_pop;
|
612
|
+
}
|
613
|
+
|
614
|
+
D_puts("(err) found: can handle error token");
|
615
|
+
break;
|
616
|
+
|
617
|
+
error_pop:
|
618
|
+
D_puts("(err) act not found: can't handle error token; pop");
|
619
|
+
|
620
|
+
if (RARRAY_LEN(v->state) <= 1) {
|
621
|
+
v->retval = Qnil;
|
622
|
+
v->fin = CP_FIN_CANTPOP;
|
623
|
+
return;
|
624
|
+
}
|
625
|
+
POP(v->state);
|
626
|
+
POP(v->vstack);
|
627
|
+
v->curstate = num_to_long(LAST_I(v->state));
|
628
|
+
if (v->debug) {
|
629
|
+
POP(v->tstack);
|
630
|
+
rb_funcall(v->parser, id_d_e_pop,
|
631
|
+
3, v->state, v->tstack, v->vstack);
|
632
|
+
}
|
633
|
+
}
|
634
|
+
|
635
|
+
/* shift/reduce error token */
|
636
|
+
if (act > 0 && act < v->shift_n) {
|
637
|
+
D_puts("e shift");
|
638
|
+
SHIFT(v, act, ERROR_TOKEN, val);
|
639
|
+
}
|
640
|
+
else if (act < 0 && act > -(v->reduce_n)) {
|
641
|
+
D_puts("e reduce");
|
642
|
+
REDUCE(v, act);
|
643
|
+
}
|
644
|
+
else if (act == v->shift_n) {
|
645
|
+
D_puts("e accept");
|
646
|
+
goto accept;
|
647
|
+
}
|
648
|
+
else {
|
649
|
+
rb_raise(RaccBug, "[Racc Bug] unknown act value %ld", act);
|
650
|
+
}
|
651
|
+
goto error_recovered;
|
652
|
+
}
|
653
|
+
|
654
|
+
static void
|
655
|
+
shift(struct cparse_params *v, long act, VALUE tok, VALUE val)
|
656
|
+
{
|
657
|
+
PUSH(v->vstack, val);
|
658
|
+
if (v->debug) {
|
659
|
+
PUSH(v->tstack, tok);
|
660
|
+
rb_funcall(v->parser, id_d_shift,
|
661
|
+
3, tok, v->tstack, v->vstack);
|
662
|
+
}
|
663
|
+
v->curstate = act;
|
664
|
+
PUSH(v->state, LONG2NUM(v->curstate));
|
665
|
+
}
|
666
|
+
|
667
|
+
static int
|
668
|
+
reduce(struct cparse_params *v, long act)
|
669
|
+
{
|
670
|
+
VALUE code;
|
671
|
+
v->ruleno = -act * 3;
|
672
|
+
code = rb_catch("racc_jump", reduce0, v->value_v);
|
673
|
+
v->errstatus = num_to_long(rb_ivar_get(v->parser, id_errstatus));
|
674
|
+
return NUM2INT(code);
|
675
|
+
}
|
676
|
+
|
677
|
+
static VALUE
|
678
|
+
reduce0(VALUE val, VALUE data, VALUE self)
|
679
|
+
{
|
680
|
+
struct cparse_params *v;
|
681
|
+
VALUE reduce_to, reduce_len, method_id;
|
682
|
+
long len;
|
683
|
+
ID mid;
|
684
|
+
VALUE tmp, tmp_t = Qundef, tmp_v = Qundef;
|
685
|
+
long i, k1, k2;
|
686
|
+
VALUE goto_state;
|
687
|
+
|
688
|
+
Data_Get_Struct(data, struct cparse_params, v);
|
689
|
+
reduce_len = RARRAY_PTR(v->reduce_table)[v->ruleno];
|
690
|
+
reduce_to = RARRAY_PTR(v->reduce_table)[v->ruleno+1];
|
691
|
+
method_id = RARRAY_PTR(v->reduce_table)[v->ruleno+2];
|
692
|
+
len = NUM2LONG(reduce_len);
|
693
|
+
mid = value_to_id(method_id);
|
694
|
+
|
695
|
+
/* call action */
|
696
|
+
if (len == 0) {
|
697
|
+
tmp = Qnil;
|
698
|
+
if (mid != id_noreduce)
|
699
|
+
tmp_v = rb_ary_new();
|
700
|
+
if (v->debug)
|
701
|
+
tmp_t = rb_ary_new();
|
702
|
+
}
|
703
|
+
else {
|
704
|
+
if (mid != id_noreduce) {
|
705
|
+
tmp_v = GET_TAIL(v->vstack, len);
|
706
|
+
tmp = RARRAY_PTR(tmp_v)[0];
|
707
|
+
}
|
708
|
+
else {
|
709
|
+
tmp = RARRAY_PTR(v->vstack)[ RARRAY_LEN(v->vstack) - len ];
|
710
|
+
}
|
711
|
+
CUT_TAIL(v->vstack, len);
|
712
|
+
if (v->debug) {
|
713
|
+
tmp_t = GET_TAIL(v->tstack, len);
|
714
|
+
CUT_TAIL(v->tstack, len);
|
715
|
+
}
|
716
|
+
CUT_TAIL(v->state, len);
|
717
|
+
}
|
718
|
+
if (mid != id_noreduce) {
|
719
|
+
if (v->use_result_var) {
|
720
|
+
tmp = rb_funcall(v->parser, mid,
|
721
|
+
3, tmp_v, v->vstack, tmp);
|
722
|
+
}
|
723
|
+
else {
|
724
|
+
tmp = rb_funcall(v->parser, mid,
|
725
|
+
2, tmp_v, v->vstack);
|
726
|
+
}
|
727
|
+
}
|
728
|
+
|
729
|
+
/* then push result */
|
730
|
+
PUSH(v->vstack, tmp);
|
731
|
+
if (v->debug) {
|
732
|
+
PUSH(v->tstack, reduce_to);
|
733
|
+
rb_funcall(v->parser, id_d_reduce,
|
734
|
+
4, tmp_t, reduce_to, v->tstack, v->vstack);
|
735
|
+
}
|
736
|
+
|
737
|
+
/* calculate transition state */
|
738
|
+
if (RARRAY_LEN(v->state) == 0)
|
739
|
+
rb_raise(RaccBug, "state stack unexpectedly empty");
|
740
|
+
k2 = num_to_long(LAST_I(v->state));
|
741
|
+
k1 = num_to_long(reduce_to) - v->nt_base;
|
742
|
+
D_printf("(goto) k1=%ld\n", k1);
|
743
|
+
D_printf("(goto) k2=%ld\n", k2);
|
744
|
+
|
745
|
+
tmp = AREF(v->goto_pointer, k1);
|
746
|
+
if (NIL_P(tmp)) goto notfound;
|
747
|
+
|
748
|
+
i = NUM2LONG(tmp) + k2;
|
749
|
+
D_printf("(goto) i=%ld\n", i);
|
750
|
+
if (i < 0) goto notfound;
|
751
|
+
|
752
|
+
goto_state = AREF(v->goto_table, i);
|
753
|
+
if (NIL_P(goto_state)) {
|
754
|
+
D_puts("(goto) table[i] == nil");
|
755
|
+
goto notfound;
|
756
|
+
}
|
757
|
+
D_printf("(goto) table[i]=%ld (goto_state)\n", NUM2LONG(goto_state));
|
758
|
+
|
759
|
+
tmp = AREF(v->goto_check, i);
|
760
|
+
if (NIL_P(tmp)) {
|
761
|
+
D_puts("(goto) check[i] == nil");
|
762
|
+
goto notfound;
|
763
|
+
}
|
764
|
+
if (tmp != LONG2NUM(k1)) {
|
765
|
+
D_puts("(goto) check[i] != table[i]");
|
766
|
+
goto notfound;
|
767
|
+
}
|
768
|
+
D_printf("(goto) check[i]=%ld\n", NUM2LONG(tmp));
|
769
|
+
|
770
|
+
D_puts("(goto) found");
|
771
|
+
transit:
|
772
|
+
PUSH(v->state, goto_state);
|
773
|
+
v->curstate = NUM2LONG(goto_state);
|
774
|
+
return INT2FIX(0);
|
775
|
+
|
776
|
+
notfound:
|
777
|
+
D_puts("(goto) not found: use default");
|
778
|
+
/* overwrite `goto-state' by default value */
|
779
|
+
goto_state = AREF(v->goto_default, k1);
|
780
|
+
goto transit;
|
781
|
+
}
|
782
|
+
|
783
|
+
/* -----------------------------------------------------------------------
|
784
|
+
Ruby Interface
|
785
|
+
----------------------------------------------------------------------- */
|
786
|
+
|
787
|
+
void
|
788
|
+
Init_cparse(void)
|
789
|
+
{
|
790
|
+
VALUE Racc, Parser;
|
791
|
+
ID id_racc = rb_intern("Racc");
|
792
|
+
|
793
|
+
if (rb_const_defined(rb_cObject, id_racc)) {
|
794
|
+
Racc = rb_const_get(rb_cObject, id_racc);
|
795
|
+
Parser = rb_const_get_at(Racc, rb_intern("Parser"));
|
796
|
+
}
|
797
|
+
else {
|
798
|
+
Racc = rb_define_module("Racc");
|
799
|
+
Parser = rb_define_class_under(Racc, "Parser", rb_cObject);
|
800
|
+
}
|
801
|
+
rb_define_private_method(Parser, "_racc_do_parse_c", racc_cparse, 2);
|
802
|
+
rb_define_private_method(Parser, "_racc_yyparse_c", racc_yyparse, 4);
|
803
|
+
rb_define_const(Parser, "Racc_Runtime_Core_Version_C",
|
804
|
+
rb_str_new2(RACC_VERSION));
|
805
|
+
rb_define_const(Parser, "Racc_Runtime_Core_Id_C",
|
806
|
+
rb_str_new2("$originalId: cparse.c,v 1.8 2006/07/06 11:39:46 aamine Exp $"));
|
807
|
+
|
808
|
+
CparseParams = rb_define_class_under(Racc, "CparseParams", rb_cObject);
|
809
|
+
|
810
|
+
RaccBug = rb_eRuntimeError;
|
811
|
+
|
812
|
+
id_yydebug = rb_intern("@yydebug");
|
813
|
+
id_nexttoken = rb_intern("next_token");
|
814
|
+
id_onerror = rb_intern("on_error");
|
815
|
+
id_noreduce = rb_intern("_reduce_none");
|
816
|
+
id_errstatus = rb_intern("@racc_error_status");
|
817
|
+
|
818
|
+
id_d_shift = rb_intern("racc_shift");
|
819
|
+
id_d_reduce = rb_intern("racc_reduce");
|
820
|
+
id_d_accept = rb_intern("racc_accept");
|
821
|
+
id_d_read_token = rb_intern("racc_read_token");
|
822
|
+
id_d_next_state = rb_intern("racc_next_state");
|
823
|
+
id_d_e_pop = rb_intern("racc_e_pop");
|
824
|
+
}
|