forked from DavidKinder/Inform6
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathsyntax.c
716 lines (598 loc) · 23.7 KB
/
syntax.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
/* ------------------------------------------------------------------------- */
/* "syntax" : Syntax analyser and compiler */
/* */
/* Part of Inform 6.33 */
/* copyright (c) Graham Nelson 1993 - 2013 */
/* */
/* ------------------------------------------------------------------------- */
#include "header.h"
static char *lexical_source;
int no_syntax_lines; /* Syntax line count */
static void begin_syntax_line(int statement_mode)
{ no_syntax_lines++;
next_token_begins_syntax_line = TRUE;
clear_expression_space();
if (statement_mode)
{ statements.enabled = TRUE;
conditions.enabled = TRUE;
local_variables.enabled = TRUE;
system_functions.enabled = TRUE;
misc_keywords.enabled = FALSE;
directive_keywords.enabled = FALSE;
directives.enabled = FALSE;
segment_markers.enabled = FALSE;
opcode_names.enabled = FALSE;
}
else
{ directives.enabled = TRUE;
segment_markers.enabled = TRUE;
statements.enabled = FALSE;
misc_keywords.enabled = FALSE;
directive_keywords.enabled = FALSE;
local_variables.enabled = FALSE;
system_functions.enabled = FALSE;
conditions.enabled = FALSE;
opcode_names.enabled = FALSE;
}
sequence_point_follows = TRUE;
if (debugfile_switch)
{ get_next_token();
statement_debug_location = get_token_location();
put_token_back();
}
}
extern void panic_mode_error_recovery(void)
{
/* Consume tokens until the next semicolon (or end of file).
This is typically called after a syntax error, in hopes of
getting parsing back on track. */
while ((token_type != EOF_TT)
&& ((token_type != SEP_TT)||(token_value != SEMICOLON_SEP)))
get_next_token();
}
extern void get_next_token_with_directives(void)
{
/* A higher-level version of get_next_token(), which detects and
obeys directives such as #ifdef/#ifnot/#endif. (The # sign is
required in this case.)
This is called while parsing a long construct, such as Class or
Object, where we want to support internal #ifdefs. (Although
function-parsing predates this and doesn't make use of it.)
(Technically this permits *any* #-directive, which means you
can define global variables or properties or what-have-you in
the middle of an object. You can do that in the middle of an
object, too. Don't. It's about as well-supported as Wile E.
Coyote one beat before the plummet-lines kick in.) */
int directives_save, segment_markers_save, statements_save;
while (TRUE)
{
get_next_token();
/* If the first token is not a '#', return it directly. */
if ((token_type != SEP_TT) || (token_value != HASH_SEP))
return;
/* Save the lexer flags, and set up for directive parsing. */
directives_save = directives.enabled;
segment_markers_save = segment_markers.enabled;
statements_save = statements.enabled;
directives.enabled = TRUE;
segment_markers.enabled = FALSE;
statements.enabled = FALSE;
conditions.enabled = FALSE;
local_variables.enabled = FALSE;
misc_keywords.enabled = FALSE;
system_functions.enabled = FALSE;
get_next_token();
if ((token_type == SEP_TT) && (token_value == OPEN_SQUARE_SEP))
{ error("It is illegal to nest a routine inside an object using '#['");
return;
}
if (token_type == DIRECTIVE_TT)
parse_given_directive(TRUE);
else
{ ebf_error("directive", token_text);
return;
}
/* Restore all the lexer flags. (We are squashing several of them
into a single save variable, which I think is safe because that's
what CKnight did.)
*/
directive_keywords.enabled = FALSE;
directives.enabled = directives_save;
segment_markers.enabled = segment_markers_save;
statements.enabled =
conditions.enabled =
local_variables.enabled =
misc_keywords.enabled =
system_functions.enabled = statements_save;
}
}
extern void parse_program(char *source)
{
lexical_source = source;
while (parse_directive(FALSE)) ;
}
extern int parse_directive(int internal_flag)
{
/* Internal_flag is FALSE if the directive is encountered normally,
TRUE if encountered with a # prefix inside a routine or object
definition.
Returns: TRUE if program continues, FALSE if end of file reached. */
int routine_symbol, rep_symbol;
int is_renamed;
begin_syntax_line(FALSE);
get_next_token();
if (token_type == EOF_TT) return(FALSE);
if ((token_type == SEP_TT) && (token_value == HASH_SEP))
get_next_token();
if ((token_type == SEP_TT) && (token_value == OPEN_SQUARE_SEP))
{ if (internal_flag)
{ error("It is illegal to nest routines using '#['");
return(TRUE);
}
directives.enabled = FALSE;
directive_keywords.enabled = FALSE;
segment_markers.enabled = FALSE;
/* The upcoming symbol is a definition; don't count it as a
top-level reference *to* the function. */
df_dont_note_global_symbols = TRUE;
get_next_token();
df_dont_note_global_symbols = FALSE;
if ((token_type != SYMBOL_TT)
|| ((!(sflags[token_value] & UNKNOWN_SFLAG))
&& (!(sflags[token_value] & REPLACE_SFLAG))))
{ ebf_error("routine name", token_text);
return(FALSE);
}
routine_symbol = token_value;
rep_symbol = routine_symbol;
is_renamed = find_symbol_replacement(&rep_symbol);
if ((sflags[routine_symbol] & REPLACE_SFLAG)
&& !is_renamed && (is_systemfile()))
{ /* The function is definitely being replaced (system_file
always loses priority in a replacement) but is not
being renamed to something else. Skip its definition
entirely. */
dont_enter_into_symbol_table = TRUE;
do
{ get_next_token();
} while (!((token_type == EOF_TT)
|| ((token_type==SEP_TT)
&& (token_value==CLOSE_SQUARE_SEP))));
dont_enter_into_symbol_table = FALSE;
if (token_type == EOF_TT) return FALSE;
}
else
{ /* Parse the function definition and assign its symbol. */
assign_symbol(routine_symbol,
parse_routine(lexical_source, FALSE,
(char *) symbs[routine_symbol], FALSE, routine_symbol),
ROUTINE_T);
slines[routine_symbol] = routine_starts_line;
}
if (is_renamed) {
/* This function was subject to a "Replace X Y" directive.
The first time we see a definition for symbol X, we
copy it to Y -- that's the "original" form of the
function. */
if (svals[rep_symbol] == 0) {
assign_symbol(rep_symbol, svals[routine_symbol], ROUTINE_T);
}
}
get_next_token();
if ((token_type != SEP_TT) || (token_value != SEMICOLON_SEP))
{ ebf_error("';' after ']'", token_text);
put_token_back();
}
return TRUE;
}
if ((token_type == SYMBOL_TT) && (stypes[token_value] == CLASS_T))
{ if (internal_flag)
{ error("It is illegal to nest an object in a routine using '#classname'");
return(TRUE);
}
sflags[token_value] |= USED_SFLAG;
make_object(FALSE, NULL, -1, -1, svals[token_value]);
return TRUE;
}
if (token_type != DIRECTIVE_TT)
{ /* If we're internal, we expect only a directive here. If
we're top-level, the possibilities are broader. */
if (internal_flag)
ebf_error("directive", token_text);
else
ebf_error("directive, '[' or class name", token_text);
panic_mode_error_recovery();
return TRUE;
}
return !(parse_given_directive(internal_flag));
}
static int switch_sign(void)
{
if ((token_type == SEP_TT)&&(token_value == COLON_SEP)) return 1;
if ((token_type == SEP_TT)&&(token_value == COMMA_SEP)) return 2;
if ((token_type==MISC_KEYWORD_TT)&&(token_value==TO_MK)) return 3;
return 0;
}
static assembly_operand spec_stack[32];
static int spec_type[32];
static void compile_alternatives_z(assembly_operand switch_value, int n,
int stack_level, int label, int flag)
{ switch(n)
{ case 1:
assemblez_2_branch(je_zc, switch_value,
spec_stack[stack_level],
label, flag); return;
case 2:
assemblez_3_branch(je_zc, switch_value,
spec_stack[stack_level], spec_stack[stack_level+1],
label, flag); return;
case 3:
assemblez_4_branch(je_zc, switch_value,
spec_stack[stack_level], spec_stack[stack_level+1],
spec_stack[stack_level+2],
label, flag); return;
}
}
static void compile_alternatives_g(assembly_operand switch_value, int n,
int stack_level, int label, int flag)
{
int the_zc = (flag) ? jeq_gc : jne_gc;
if (n == 1) {
assembleg_2_branch(the_zc, switch_value,
spec_stack[stack_level],
label);
}
else {
error("*** Cannot generate multi-equality tests in Glulx ***");
}
}
static void compile_alternatives(assembly_operand switch_value, int n,
int stack_level, int label, int flag)
{
if (!glulx_mode)
compile_alternatives_z(switch_value, n, stack_level, label, flag);
else
compile_alternatives_g(switch_value, n, stack_level, label, flag);
}
static void parse_switch_spec(assembly_operand switch_value, int label,
int action_switch)
{
int i, j, label_after = -1, spec_sp = 0;
int max_equality_args = ((!glulx_mode) ? 3 : 1);
sequence_point_follows = FALSE;
do
{ if (spec_sp == 32)
{ error("At most 32 values can be given in a single 'switch' case");
panic_mode_error_recovery();
return;
}
if (action_switch)
{ get_next_token();
spec_stack[spec_sp].type =
((!glulx_mode) ? LONG_CONSTANT_OT : CONSTANT_OT);
spec_stack[spec_sp].value = 0;
spec_stack[spec_sp].marker = 0;
spec_stack[spec_sp] = action_of_name(token_text);
if (spec_stack[spec_sp].value == -1)
{ spec_stack[spec_sp].value = 0;
ebf_error("action (or fake action) name", token_text);
}
}
else
spec_stack[spec_sp] =
code_generate(parse_expression(CONSTANT_CONTEXT), CONSTANT_CONTEXT, -1);
misc_keywords.enabled = TRUE;
get_next_token();
misc_keywords.enabled = FALSE;
spec_type[spec_sp++] = switch_sign();
switch(spec_type[spec_sp-1])
{ case 0:
if (action_switch)
ebf_error("',' or ':'", token_text);
else ebf_error("',', ':' or 'to'", token_text);
panic_mode_error_recovery();
return;
case 1: goto GenSpecCode;
case 3: if (label_after == -1) label_after = next_label++;
}
} while(TRUE);
GenSpecCode:
if ((spec_sp > max_equality_args) && (label_after == -1))
label_after = next_label++;
if (label_after == -1)
{ compile_alternatives(switch_value, spec_sp, 0, label, FALSE); return;
}
for (i=0; i<spec_sp;)
{
j=i; while ((j<spec_sp) && (spec_type[j] != 3)) j++;
if (j > i)
{ if (j-i > max_equality_args) j=i+max_equality_args;
if (j == spec_sp)
compile_alternatives(switch_value, j-i, i, label, FALSE);
else
compile_alternatives(switch_value, j-i, i, label_after, TRUE);
i=j;
}
else
{
if (!glulx_mode) {
if (i == spec_sp - 2)
{ assemblez_2_branch(jl_zc, switch_value, spec_stack[i],
label, TRUE);
assemblez_2_branch(jg_zc, switch_value, spec_stack[i+1],
label, TRUE);
}
else
{ assemblez_2_branch(jl_zc, switch_value, spec_stack[i],
next_label, TRUE);
assemblez_2_branch(jg_zc, switch_value, spec_stack[i+1],
label_after, FALSE);
assemble_label_no(next_label++);
}
}
else {
if (i == spec_sp - 2)
{ assembleg_2_branch(jlt_gc, switch_value, spec_stack[i],
label);
assembleg_2_branch(jgt_gc, switch_value, spec_stack[i+1],
label);
}
else
{ assembleg_2_branch(jlt_gc, switch_value, spec_stack[i],
next_label);
assembleg_2_branch(jle_gc, switch_value, spec_stack[i+1],
label_after);
assemble_label_no(next_label++);
}
}
i = i+2;
}
}
assemble_label_no(label_after);
}
extern int32 parse_routine(char *source, int embedded_flag, char *name,
int veneer_flag, int r_symbol)
{ int32 packed_address; int i; int debug_flag = FALSE;
int switch_clause_made = FALSE, default_clause_made = FALSE,
switch_label = 0;
debug_location_beginning beginning_debug_location =
get_token_location_beginning();
/* (switch_label needs no initialisation here, but it prevents some
compilers from issuing warnings) */
if ((source != lexical_source) || (veneer_flag))
{ lexical_source = source;
restart_lexer(lexical_source, name);
}
no_locals = 0;
for (i=0;i<MAX_LOCAL_VARIABLES-1;i++) local_variables.keywords[i] = "";
do
{ statements.enabled = TRUE;
dont_enter_into_symbol_table = TRUE;
get_next_token();
dont_enter_into_symbol_table = FALSE;
if ((token_type == SEP_TT) && (token_value == TIMES_SEP)
&& (no_locals == 0) && (!debug_flag))
{ debug_flag = TRUE; continue;
}
if (token_type != DQ_TT)
{ if ((token_type == SEP_TT)
&& (token_value == SEMICOLON_SEP)) break;
ebf_error("local variable name or ';'", token_text);
panic_mode_error_recovery();
break;
}
if (no_locals == MAX_LOCAL_VARIABLES-1)
{ error_numbered("Too many local variables for a routine; max is",
MAX_LOCAL_VARIABLES-1);
panic_mode_error_recovery();
break;
}
for (i=0;i<no_locals;i++)
if (strcmpcis(token_text, local_variables.keywords[i])==0)
error_named("Local variable defined twice:", token_text);
local_variables.keywords[no_locals++] = token_text;
} while(TRUE);
construct_local_variable_tables();
if ((trace_fns_setting==3)
|| ((trace_fns_setting==2) && (veneer_mode==FALSE))
|| ((trace_fns_setting==1) && (is_systemfile()==FALSE)))
debug_flag = TRUE;
if ((embedded_flag == FALSE) && (veneer_mode == FALSE) && debug_flag)
sflags[r_symbol] |= STAR_SFLAG;
packed_address = assemble_routine_header(no_locals, debug_flag,
name, embedded_flag, r_symbol);
do
{ begin_syntax_line(TRUE);
get_next_token();
if (token_type == EOF_TT)
{ ebf_error("']'", token_text);
assemble_routine_end
(embedded_flag,
get_token_location_end(beginning_debug_location));
put_token_back();
break;
}
if ((token_type == SEP_TT)
&& (token_value == CLOSE_SQUARE_SEP))
{ if (switch_clause_made && (!default_clause_made))
assemble_label_no(switch_label);
directives.enabled = TRUE;
sequence_point_follows = TRUE;
get_next_token();
assemble_routine_end
(embedded_flag,
get_token_location_end(beginning_debug_location));
put_token_back();
break;
}
if ((token_type == STATEMENT_TT) && (token_value == SDEFAULT_CODE))
{ if (default_clause_made)
error("Multiple 'default' clauses defined in same 'switch'");
default_clause_made = TRUE;
if (switch_clause_made)
{ if (!execution_never_reaches_here)
{ sequence_point_follows = FALSE;
if (!glulx_mode)
assemblez_0((embedded_flag)?rfalse_zc:rtrue_zc);
else
assembleg_1(return_gc,
((embedded_flag)?zero_operand:one_operand));
}
assemble_label_no(switch_label);
}
switch_clause_made = TRUE;
get_next_token();
if ((token_type == SEP_TT) &&
(token_value == COLON_SEP)) continue;
ebf_error("':' after 'default'", token_text);
panic_mode_error_recovery();
continue;
}
/* Only check for the form of a case switch if the initial token
isn't double-quoted text, as that would mean it was a print_ret
statement: this is a mild ambiguity in the grammar.
Action statements also cannot be cases. */
if ((token_type != DQ_TT) && (token_type != SEP_TT))
{ get_next_token();
if (switch_sign() > 0)
{ assembly_operand AO;
if (default_clause_made)
error("'default' must be the last 'switch' case");
if (switch_clause_made)
{ if (!execution_never_reaches_here)
{ sequence_point_follows = FALSE;
if (!glulx_mode)
assemblez_0((embedded_flag)?rfalse_zc:rtrue_zc);
else
assembleg_1(return_gc,
((embedded_flag)?zero_operand:one_operand));
}
assemble_label_no(switch_label);
}
switch_label = next_label++;
switch_clause_made = TRUE;
put_token_back(); put_token_back();
if (!glulx_mode) {
AO.type = VARIABLE_OT; AO.value = 249; AO.marker = 0;
}
else {
AO.type = GLOBALVAR_OT;
AO.value = MAX_LOCAL_VARIABLES+6; /* sw__var */
AO.marker = 0;
}
parse_switch_spec(AO, switch_label, TRUE);
continue;
}
else
{ put_token_back(); put_token_back(); get_next_token();
sequence_point_follows = TRUE;
}
}
parse_statement(-1, -1);
} while (TRUE);
return packed_address;
}
extern void parse_code_block(int break_label, int continue_label,
int switch_rule)
{ int switch_clause_made = FALSE, default_clause_made = FALSE, switch_label,
unary_minus_flag;
begin_syntax_line(TRUE);
get_next_token();
if (token_type == SEP_TT && token_value == OPEN_BRACE_SEP)
{ do
{ begin_syntax_line(TRUE);
get_next_token();
if (token_type == SEP_TT && token_value == CLOSE_BRACE_SEP)
{ if (switch_clause_made && (!default_clause_made))
assemble_label_no(switch_label);
return;
}
if (token_type == EOF_TT)
{ ebf_error("'}'", token_text); return; }
if (switch_rule != 0)
{
/* Within a 'switch' block */
if ((token_type==STATEMENT_TT)&&(token_value==SDEFAULT_CODE))
{ if (default_clause_made)
error("Multiple 'default' clauses defined in same 'switch'");
default_clause_made = TRUE;
if (switch_clause_made)
{ if (!execution_never_reaches_here)
{ sequence_point_follows = FALSE;
assemble_jump(break_label);
}
assemble_label_no(switch_label);
}
switch_clause_made = TRUE;
get_next_token();
if ((token_type == SEP_TT) &&
(token_value == COLON_SEP)) continue;
ebf_error("':' after 'default'", token_text);
panic_mode_error_recovery();
continue;
}
/* Decide: is this an ordinary statement, or the start
of a new case? */
if (token_type == DQ_TT) goto NotASwitchCase;
unary_minus_flag
= ((token_type == SEP_TT)&&(token_value == MINUS_SEP));
if (unary_minus_flag) get_next_token();
/* Now read the token _after_ any possible constant:
if that's a 'to', ',' or ':' then we have a case */
misc_keywords.enabled = TRUE;
get_next_token();
misc_keywords.enabled = FALSE;
if (switch_sign() > 0)
{ assembly_operand AO;
if (default_clause_made)
error("'default' must be the last 'switch' case");
if (switch_clause_made)
{ if (!execution_never_reaches_here)
{ sequence_point_follows = FALSE;
assemble_jump(break_label);
}
assemble_label_no(switch_label);
}
switch_label = next_label++;
switch_clause_made = TRUE;
put_token_back(); put_token_back();
if (unary_minus_flag) put_token_back();
AO = temp_var1;
parse_switch_spec(AO, switch_label, FALSE);
continue;
}
else
{ put_token_back(); put_token_back();
if (unary_minus_flag) put_token_back();
get_next_token();
}
}
if ((switch_rule != 0) && (!switch_clause_made))
ebf_error("switch value", token_text);
NotASwitchCase:
sequence_point_follows = TRUE;
parse_statement(break_label, continue_label);
}
while(TRUE);
}
if (switch_rule != 0)
ebf_error("braced code block after 'switch'", token_text);
parse_statement(break_label, continue_label);
return;
}
/* ========================================================================= */
/* Data structure management routines */
/* ------------------------------------------------------------------------- */
extern void init_syntax_vars(void)
{
}
extern void syntax_begin_pass(void)
{ no_syntax_lines = 0;
}
extern void syntax_allocate_arrays(void)
{
}
extern void syntax_free_arrays(void)
{
}
/* ========================================================================= */