Add docs and arch tests to BMI.
[deliverable/binutils-gdb.git] / gdb / f-exp.y
CommitLineData
c906108c 1/* YACC parser for Fortran expressions, for GDB.
197e01b6 2 Copyright (C) 1986, 1989, 1990, 1991, 1993, 1994, 1995, 1996, 2000, 2001,
7b6bb8da 3 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
0fb0cc75 4 Free Software Foundation, Inc.
4fcf66da 5
c906108c
SS
6 Contributed by Motorola. Adapted from the C parser by Farooq Butt
7 (fmbutt@engage.sps.mot.com).
8
5b1ba0e5 9 This file is part of GDB.
c906108c 10
5b1ba0e5
NS
11 This program is free software; you can redistribute it and/or modify
12 it under the terms of the GNU General Public License as published by
13 the Free Software Foundation; either version 3 of the License, or
14 (at your option) any later version.
c906108c 15
5b1ba0e5
NS
16 This program is distributed in the hope that it will be useful,
17 but WITHOUT ANY WARRANTY; without even the implied warranty of
18 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 GNU General Public License for more details.
c906108c 20
5b1ba0e5
NS
21 You should have received a copy of the GNU General Public License
22 along with this program. If not, see <http://www.gnu.org/licenses/>. */
c906108c
SS
23
24/* This was blantantly ripped off the C expression parser, please
25 be aware of that as you look at its basic structure -FMB */
26
27/* Parse a F77 expression from text in a string,
28 and return the result as a struct expression pointer.
29 That structure contains arithmetic operations in reverse polish,
30 with constants represented by operations that are followed by special data.
31 See expression.h for the details of the format.
32 What is important here is that it can be built up sequentially
33 during the process of parsing; the lower levels of the tree always
34 come first in the result.
35
36 Note that malloc's and realloc's in this file are transformed to
37 xmalloc and xrealloc respectively by the same sed command in the
38 makefile that remaps any other malloc/realloc inserted by the parser
39 generator. Doing this with #defines and trying to control the interaction
40 with include files (<malloc.h> and <stdlib.h> for example) just became
41 too messy, particularly when such includes can be inserted at random
42 times by the parser generator. */
43
44%{
45
46#include "defs.h"
47#include "gdb_string.h"
48#include "expression.h"
49#include "value.h"
50#include "parser-defs.h"
51#include "language.h"
52#include "f-lang.h"
53#include "bfd.h" /* Required by objfiles.h. */
54#include "symfile.h" /* Required by objfiles.h. */
55#include "objfiles.h" /* For have_full_symbols and have_partial_symbols */
fe898f56 56#include "block.h"
0f6e1ba6 57#include <ctype.h>
c906108c 58
3e79cecf
UW
59#define parse_type builtin_type (parse_gdbarch)
60#define parse_f_type builtin_f_type (parse_gdbarch)
61
c906108c
SS
62/* Remap normal yacc parser interface names (yyparse, yylex, yyerror, etc),
63 as well as gratuitiously global symbol names, so we can have multiple
64 yacc generated parsers in gdb. Note that these are only the variables
65 produced by yacc. If other parser generators (bison, byacc, etc) produce
66 additional global names that conflict at link time, then those parser
67 generators need to be fixed instead of adding those names to this list. */
68
69#define yymaxdepth f_maxdepth
70#define yyparse f_parse
71#define yylex f_lex
72#define yyerror f_error
73#define yylval f_lval
74#define yychar f_char
75#define yydebug f_debug
76#define yypact f_pact
77#define yyr1 f_r1
78#define yyr2 f_r2
79#define yydef f_def
80#define yychk f_chk
81#define yypgo f_pgo
82#define yyact f_act
83#define yyexca f_exca
84#define yyerrflag f_errflag
85#define yynerrs f_nerrs
86#define yyps f_ps
87#define yypv f_pv
88#define yys f_s
89#define yy_yys f_yys
90#define yystate f_state
91#define yytmp f_tmp
92#define yyv f_v
93#define yy_yyv f_yyv
94#define yyval f_val
95#define yylloc f_lloc
96#define yyreds f_reds /* With YYDEBUG defined */
97#define yytoks f_toks /* With YYDEBUG defined */
06891d83
JT
98#define yyname f_name /* With YYDEBUG defined */
99#define yyrule f_rule /* With YYDEBUG defined */
c906108c
SS
100#define yylhs f_yylhs
101#define yylen f_yylen
102#define yydefred f_yydefred
103#define yydgoto f_yydgoto
104#define yysindex f_yysindex
105#define yyrindex f_yyrindex
106#define yygindex f_yygindex
107#define yytable f_yytable
108#define yycheck f_yycheck
109
110#ifndef YYDEBUG
f461f5cf 111#define YYDEBUG 1 /* Default to yydebug support */
c906108c
SS
112#endif
113
f461f5cf
PM
114#define YYFPRINTF parser_fprintf
115
a14ed312 116int yyparse (void);
c906108c 117
a14ed312 118static int yylex (void);
c906108c 119
a14ed312 120void yyerror (char *);
c906108c 121
a14ed312 122static void growbuf_by_size (int);
c906108c 123
a14ed312 124static int match_string_literal (void);
c906108c
SS
125
126%}
127
128/* Although the yacc "value" of an expression is not used,
129 since the result is stored in the structure being created,
130 other node types do have values. */
131
132%union
133 {
134 LONGEST lval;
135 struct {
136 LONGEST val;
137 struct type *type;
138 } typed_val;
139 DOUBLEST dval;
140 struct symbol *sym;
141 struct type *tval;
142 struct stoken sval;
143 struct ttype tsym;
144 struct symtoken ssym;
145 int voidval;
146 struct block *bval;
147 enum exp_opcode opcode;
148 struct internalvar *ivar;
149
150 struct type **tvec;
151 int *ivec;
152 }
153
154%{
155/* YYSTYPE gets defined by %union */
a14ed312 156static int parse_number (char *, int, int, YYSTYPE *);
c906108c
SS
157%}
158
159%type <voidval> exp type_exp start variable
160%type <tval> type typebase
161%type <tvec> nonempty_typelist
162/* %type <bval> block */
163
164/* Fancy type parsing. */
165%type <voidval> func_mod direct_abs_decl abs_decl
166%type <tval> ptype
167
168%token <typed_val> INT
169%token <dval> FLOAT
170
171/* Both NAME and TYPENAME tokens represent symbols in the input,
172 and both convey their data as strings.
173 But a TYPENAME is a string that happens to be defined as a typedef
174 or builtin type name (such as int or char)
175 and a NAME is any other symbol.
176 Contexts where this distinction is not important can use the
177 nonterminal "name", which matches either NAME or TYPENAME. */
178
179%token <sval> STRING_LITERAL
180%token <lval> BOOLEAN_LITERAL
181%token <ssym> NAME
182%token <tsym> TYPENAME
2a5e440c 183%type <sval> name
c906108c 184%type <ssym> name_not_typename
c906108c
SS
185
186/* A NAME_OR_INT is a symbol which is not known in the symbol table,
187 but which would parse as a valid number in the current input radix.
188 E.g. "c" when input_radix==16. Depending on the parse, it will be
189 turned into a name or into a number. */
190
191%token <ssym> NAME_OR_INT
192
193%token SIZEOF
194%token ERROR
195
196/* Special type cases, put in to allow the parser to distinguish different
197 legal basetypes. */
198%token INT_KEYWORD INT_S2_KEYWORD LOGICAL_S1_KEYWORD LOGICAL_S2_KEYWORD
ce4b0682 199%token LOGICAL_S8_KEYWORD
c906108c
SS
200%token LOGICAL_KEYWORD REAL_KEYWORD REAL_S8_KEYWORD REAL_S16_KEYWORD
201%token COMPLEX_S8_KEYWORD COMPLEX_S16_KEYWORD COMPLEX_S32_KEYWORD
202%token BOOL_AND BOOL_OR BOOL_NOT
203%token <lval> CHARACTER
204
205%token <voidval> VARIABLE
206
207%token <opcode> ASSIGN_MODIFY
208
209%left ','
210%left ABOVE_COMMA
211%right '=' ASSIGN_MODIFY
212%right '?'
213%left BOOL_OR
214%right BOOL_NOT
215%left BOOL_AND
216%left '|'
217%left '^'
218%left '&'
219%left EQUAL NOTEQUAL
220%left LESSTHAN GREATERTHAN LEQ GEQ
221%left LSH RSH
222%left '@'
223%left '+' '-'
2a5e440c 224%left '*' '/'
bd49c137 225%right STARSTAR
2a5e440c 226%right '%'
c906108c
SS
227%right UNARY
228%right '('
229
230\f
231%%
232
233start : exp
234 | type_exp
235 ;
236
237type_exp: type
238 { write_exp_elt_opcode(OP_TYPE);
239 write_exp_elt_type($1);
240 write_exp_elt_opcode(OP_TYPE); }
241 ;
242
243exp : '(' exp ')'
244 { }
245 ;
246
247/* Expressions, not including the comma operator. */
248exp : '*' exp %prec UNARY
249 { write_exp_elt_opcode (UNOP_IND); }
ef944135 250 ;
c906108c
SS
251
252exp : '&' exp %prec UNARY
253 { write_exp_elt_opcode (UNOP_ADDR); }
ef944135 254 ;
c906108c
SS
255
256exp : '-' exp %prec UNARY
257 { write_exp_elt_opcode (UNOP_NEG); }
258 ;
259
260exp : BOOL_NOT exp %prec UNARY
261 { write_exp_elt_opcode (UNOP_LOGICAL_NOT); }
262 ;
263
264exp : '~' exp %prec UNARY
265 { write_exp_elt_opcode (UNOP_COMPLEMENT); }
266 ;
267
268exp : SIZEOF exp %prec UNARY
269 { write_exp_elt_opcode (UNOP_SIZEOF); }
270 ;
271
272/* No more explicit array operators, we treat everything in F77 as
273 a function call. The disambiguation as to whether we are
274 doing a subscript operation or a function call is done
275 later in eval.c. */
276
277exp : exp '('
278 { start_arglist (); }
279 arglist ')'
280 { write_exp_elt_opcode (OP_F77_UNDETERMINED_ARGLIST);
281 write_exp_elt_longcst ((LONGEST) end_arglist ());
282 write_exp_elt_opcode (OP_F77_UNDETERMINED_ARGLIST); }
283 ;
284
285arglist :
286 ;
287
288arglist : exp
289 { arglist_len = 1; }
290 ;
291
0b4e1325
WZ
292arglist : subrange
293 { arglist_len = 1; }
ef944135 294 ;
c906108c
SS
295
296arglist : arglist ',' exp %prec ABOVE_COMMA
297 { arglist_len++; }
298 ;
299
0b4e1325
WZ
300/* There are four sorts of subrange types in F90. */
301
302subrange: exp ':' exp %prec ABOVE_COMMA
303 { write_exp_elt_opcode (OP_F90_RANGE);
304 write_exp_elt_longcst (NONE_BOUND_DEFAULT);
305 write_exp_elt_opcode (OP_F90_RANGE); }
306 ;
307
308subrange: exp ':' %prec ABOVE_COMMA
309 { write_exp_elt_opcode (OP_F90_RANGE);
310 write_exp_elt_longcst (HIGH_BOUND_DEFAULT);
311 write_exp_elt_opcode (OP_F90_RANGE); }
c906108c
SS
312 ;
313
0b4e1325
WZ
314subrange: ':' exp %prec ABOVE_COMMA
315 { write_exp_elt_opcode (OP_F90_RANGE);
316 write_exp_elt_longcst (LOW_BOUND_DEFAULT);
317 write_exp_elt_opcode (OP_F90_RANGE); }
318 ;
319
320subrange: ':' %prec ABOVE_COMMA
321 { write_exp_elt_opcode (OP_F90_RANGE);
322 write_exp_elt_longcst (BOTH_BOUND_DEFAULT);
323 write_exp_elt_opcode (OP_F90_RANGE); }
324 ;
c906108c
SS
325
326complexnum: exp ',' exp
327 { }
328 ;
329
330exp : '(' complexnum ')'
c806c55a
UW
331 { write_exp_elt_opcode(OP_COMPLEX);
332 write_exp_elt_type (parse_f_type->builtin_complex_s16);
333 write_exp_elt_opcode(OP_COMPLEX); }
c906108c
SS
334 ;
335
336exp : '(' type ')' exp %prec UNARY
337 { write_exp_elt_opcode (UNOP_CAST);
338 write_exp_elt_type ($2);
339 write_exp_elt_opcode (UNOP_CAST); }
340 ;
341
2a5e440c
WZ
342exp : exp '%' name
343 { write_exp_elt_opcode (STRUCTOP_STRUCT);
344 write_exp_string ($3);
345 write_exp_elt_opcode (STRUCTOP_STRUCT); }
346 ;
347
c906108c
SS
348/* Binary operators in order of decreasing precedence. */
349
350exp : exp '@' exp
351 { write_exp_elt_opcode (BINOP_REPEAT); }
352 ;
353
bd49c137
WZ
354exp : exp STARSTAR exp
355 { write_exp_elt_opcode (BINOP_EXP); }
356 ;
357
c906108c
SS
358exp : exp '*' exp
359 { write_exp_elt_opcode (BINOP_MUL); }
360 ;
361
362exp : exp '/' exp
363 { write_exp_elt_opcode (BINOP_DIV); }
364 ;
365
c906108c
SS
366exp : exp '+' exp
367 { write_exp_elt_opcode (BINOP_ADD); }
368 ;
369
370exp : exp '-' exp
371 { write_exp_elt_opcode (BINOP_SUB); }
372 ;
373
374exp : exp LSH exp
375 { write_exp_elt_opcode (BINOP_LSH); }
376 ;
377
378exp : exp RSH exp
379 { write_exp_elt_opcode (BINOP_RSH); }
380 ;
381
382exp : exp EQUAL exp
383 { write_exp_elt_opcode (BINOP_EQUAL); }
384 ;
385
386exp : exp NOTEQUAL exp
387 { write_exp_elt_opcode (BINOP_NOTEQUAL); }
388 ;
389
390exp : exp LEQ exp
391 { write_exp_elt_opcode (BINOP_LEQ); }
392 ;
393
394exp : exp GEQ exp
395 { write_exp_elt_opcode (BINOP_GEQ); }
396 ;
397
398exp : exp LESSTHAN exp
399 { write_exp_elt_opcode (BINOP_LESS); }
400 ;
401
402exp : exp GREATERTHAN exp
403 { write_exp_elt_opcode (BINOP_GTR); }
404 ;
405
406exp : exp '&' exp
407 { write_exp_elt_opcode (BINOP_BITWISE_AND); }
408 ;
409
410exp : exp '^' exp
411 { write_exp_elt_opcode (BINOP_BITWISE_XOR); }
412 ;
413
414exp : exp '|' exp
415 { write_exp_elt_opcode (BINOP_BITWISE_IOR); }
416 ;
417
418exp : exp BOOL_AND exp
419 { write_exp_elt_opcode (BINOP_LOGICAL_AND); }
420 ;
421
422
423exp : exp BOOL_OR exp
424 { write_exp_elt_opcode (BINOP_LOGICAL_OR); }
425 ;
426
427exp : exp '=' exp
428 { write_exp_elt_opcode (BINOP_ASSIGN); }
429 ;
430
431exp : exp ASSIGN_MODIFY exp
432 { write_exp_elt_opcode (BINOP_ASSIGN_MODIFY);
433 write_exp_elt_opcode ($2);
434 write_exp_elt_opcode (BINOP_ASSIGN_MODIFY); }
435 ;
436
437exp : INT
438 { write_exp_elt_opcode (OP_LONG);
439 write_exp_elt_type ($1.type);
440 write_exp_elt_longcst ((LONGEST)($1.val));
441 write_exp_elt_opcode (OP_LONG); }
442 ;
443
444exp : NAME_OR_INT
445 { YYSTYPE val;
446 parse_number ($1.stoken.ptr, $1.stoken.length, 0, &val);
447 write_exp_elt_opcode (OP_LONG);
448 write_exp_elt_type (val.typed_val.type);
449 write_exp_elt_longcst ((LONGEST)val.typed_val.val);
450 write_exp_elt_opcode (OP_LONG); }
451 ;
452
453exp : FLOAT
454 { write_exp_elt_opcode (OP_DOUBLE);
3e79cecf 455 write_exp_elt_type (parse_f_type->builtin_real_s8);
c906108c
SS
456 write_exp_elt_dblcst ($1);
457 write_exp_elt_opcode (OP_DOUBLE); }
458 ;
459
460exp : variable
461 ;
462
463exp : VARIABLE
464 ;
465
466exp : SIZEOF '(' type ')' %prec UNARY
467 { write_exp_elt_opcode (OP_LONG);
3e79cecf 468 write_exp_elt_type (parse_f_type->builtin_integer);
c906108c
SS
469 CHECK_TYPEDEF ($3);
470 write_exp_elt_longcst ((LONGEST) TYPE_LENGTH ($3));
471 write_exp_elt_opcode (OP_LONG); }
472 ;
473
474exp : BOOLEAN_LITERAL
475 { write_exp_elt_opcode (OP_BOOL);
476 write_exp_elt_longcst ((LONGEST) $1);
477 write_exp_elt_opcode (OP_BOOL);
478 }
479 ;
480
481exp : STRING_LITERAL
482 {
483 write_exp_elt_opcode (OP_STRING);
484 write_exp_string ($1);
485 write_exp_elt_opcode (OP_STRING);
486 }
487 ;
488
489variable: name_not_typename
490 { struct symbol *sym = $1.sym;
491
492 if (sym)
493 {
494 if (symbol_read_needs_frame (sym))
495 {
905e0470
PM
496 if (innermost_block == 0
497 || contained_in (block_found,
498 innermost_block))
c906108c
SS
499 innermost_block = block_found;
500 }
501 write_exp_elt_opcode (OP_VAR_VALUE);
502 /* We want to use the selected frame, not
503 another more inner frame which happens to
504 be in the same block. */
505 write_exp_elt_block (NULL);
506 write_exp_elt_sym (sym);
507 write_exp_elt_opcode (OP_VAR_VALUE);
508 break;
509 }
510 else
511 {
512 struct minimal_symbol *msymbol;
710122da 513 char *arg = copy_name ($1.stoken);
c906108c
SS
514
515 msymbol =
516 lookup_minimal_symbol (arg, NULL, NULL);
517 if (msymbol != NULL)
c841afd5 518 write_exp_msymbol (msymbol);
c906108c
SS
519 else if (!have_full_symbols () && !have_partial_symbols ())
520 error ("No symbol table is loaded. Use the \"file\" command.");
521 else
522 error ("No symbol \"%s\" in current context.",
523 copy_name ($1.stoken));
524 }
525 }
526 ;
527
528
529type : ptype
530 ;
531
532ptype : typebase
533 | typebase abs_decl
534 {
535 /* This is where the interesting stuff happens. */
536 int done = 0;
537 int array_size;
538 struct type *follow_type = $1;
539 struct type *range_type;
540
541 while (!done)
542 switch (pop_type ())
543 {
544 case tp_end:
545 done = 1;
546 break;
547 case tp_pointer:
548 follow_type = lookup_pointer_type (follow_type);
549 break;
550 case tp_reference:
551 follow_type = lookup_reference_type (follow_type);
552 break;
553 case tp_array:
554 array_size = pop_type_int ();
555 if (array_size != -1)
556 {
557 range_type =
558 create_range_type ((struct type *) NULL,
3e79cecf
UW
559 parse_f_type->builtin_integer,
560 0, array_size - 1);
c906108c
SS
561 follow_type =
562 create_array_type ((struct type *) NULL,
563 follow_type, range_type);
564 }
565 else
566 follow_type = lookup_pointer_type (follow_type);
567 break;
568 case tp_function:
569 follow_type = lookup_function_type (follow_type);
570 break;
571 }
572 $$ = follow_type;
573 }
574 ;
575
576abs_decl: '*'
577 { push_type (tp_pointer); $$ = 0; }
578 | '*' abs_decl
579 { push_type (tp_pointer); $$ = $2; }
580 | '&'
581 { push_type (tp_reference); $$ = 0; }
582 | '&' abs_decl
583 { push_type (tp_reference); $$ = $2; }
584 | direct_abs_decl
585 ;
586
587direct_abs_decl: '(' abs_decl ')'
588 { $$ = $2; }
589 | direct_abs_decl func_mod
590 { push_type (tp_function); }
591 | func_mod
592 { push_type (tp_function); }
593 ;
594
595func_mod: '(' ')'
596 { $$ = 0; }
597 | '(' nonempty_typelist ')'
8dbb1c65 598 { free ($2); $$ = 0; }
c906108c
SS
599 ;
600
601typebase /* Implements (approximately): (type-qualifier)* type-specifier */
602 : TYPENAME
603 { $$ = $1.type; }
604 | INT_KEYWORD
3e79cecf 605 { $$ = parse_f_type->builtin_integer; }
c906108c 606 | INT_S2_KEYWORD
3e79cecf 607 { $$ = parse_f_type->builtin_integer_s2; }
c906108c 608 | CHARACTER
3e79cecf 609 { $$ = parse_f_type->builtin_character; }
ce4b0682
SDJ
610 | LOGICAL_S8_KEYWORD
611 { $$ = parse_f_type->builtin_logical_s8; }
c906108c 612 | LOGICAL_KEYWORD
3e79cecf 613 { $$ = parse_f_type->builtin_logical; }
c906108c 614 | LOGICAL_S2_KEYWORD
3e79cecf 615 { $$ = parse_f_type->builtin_logical_s2; }
c906108c 616 | LOGICAL_S1_KEYWORD
3e79cecf 617 { $$ = parse_f_type->builtin_logical_s1; }
c906108c 618 | REAL_KEYWORD
3e79cecf 619 { $$ = parse_f_type->builtin_real; }
c906108c 620 | REAL_S8_KEYWORD
3e79cecf 621 { $$ = parse_f_type->builtin_real_s8; }
c906108c 622 | REAL_S16_KEYWORD
3e79cecf 623 { $$ = parse_f_type->builtin_real_s16; }
c906108c 624 | COMPLEX_S8_KEYWORD
3e79cecf 625 { $$ = parse_f_type->builtin_complex_s8; }
c906108c 626 | COMPLEX_S16_KEYWORD
3e79cecf 627 { $$ = parse_f_type->builtin_complex_s16; }
c906108c 628 | COMPLEX_S32_KEYWORD
3e79cecf 629 { $$ = parse_f_type->builtin_complex_s32; }
c906108c
SS
630 ;
631
c906108c
SS
632nonempty_typelist
633 : type
634 { $$ = (struct type **) malloc (sizeof (struct type *) * 2);
635 $<ivec>$[0] = 1; /* Number of types in vector */
636 $$[1] = $1;
637 }
638 | nonempty_typelist ',' type
639 { int len = sizeof (struct type *) * (++($<ivec>1[0]) + 1);
640 $$ = (struct type **) realloc ((char *) $1, len);
641 $$[$<ivec>$[0]] = $3;
642 }
643 ;
644
2a5e440c
WZ
645name : NAME
646 { $$ = $1.stoken; }
647 ;
648
c906108c
SS
649name_not_typename : NAME
650/* These would be useful if name_not_typename was useful, but it is just
651 a fake for "variable", so these cause reduce/reduce conflicts because
652 the parser can't tell whether NAME_OR_INT is a name_not_typename (=variable,
653 =exp) or just an exp. If name_not_typename was ever used in an lvalue
654 context where only a name could occur, this might be useful.
655 | NAME_OR_INT
656 */
657 ;
658
659%%
660
661/* Take care of parsing a number (anything that starts with a digit).
662 Set yylval and return the token type; update lexptr.
663 LEN is the number of characters in it. */
664
665/*** Needs some error checking for the float case ***/
666
667static int
668parse_number (p, len, parsed_float, putithere)
710122da
DC
669 char *p;
670 int len;
c906108c
SS
671 int parsed_float;
672 YYSTYPE *putithere;
673{
710122da
DC
674 LONGEST n = 0;
675 LONGEST prevn = 0;
676 int c;
677 int base = input_radix;
c906108c
SS
678 int unsigned_p = 0;
679 int long_p = 0;
680 ULONGEST high_bit;
681 struct type *signed_type;
682 struct type *unsigned_type;
683
684 if (parsed_float)
685 {
686 /* It's a float since it contains a point or an exponent. */
687 /* [dD] is not understood as an exponent by atof, change it to 'e'. */
688 char *tmp, *tmp2;
689
4fcf66da 690 tmp = xstrdup (p);
c906108c
SS
691 for (tmp2 = tmp; *tmp2; ++tmp2)
692 if (*tmp2 == 'd' || *tmp2 == 'D')
693 *tmp2 = 'e';
694 putithere->dval = atof (tmp);
695 free (tmp);
696 return FLOAT;
697 }
698
699 /* Handle base-switching prefixes 0x, 0t, 0d, 0 */
700 if (p[0] == '0')
701 switch (p[1])
702 {
703 case 'x':
704 case 'X':
705 if (len >= 3)
706 {
707 p += 2;
708 base = 16;
709 len -= 2;
710 }
711 break;
712
713 case 't':
714 case 'T':
715 case 'd':
716 case 'D':
717 if (len >= 3)
718 {
719 p += 2;
720 base = 10;
721 len -= 2;
722 }
723 break;
724
725 default:
726 base = 8;
727 break;
728 }
729
730 while (len-- > 0)
731 {
732 c = *p++;
0f6e1ba6
AC
733 if (isupper (c))
734 c = tolower (c);
735 if (len == 0 && c == 'l')
736 long_p = 1;
737 else if (len == 0 && c == 'u')
738 unsigned_p = 1;
c906108c
SS
739 else
740 {
0f6e1ba6
AC
741 int i;
742 if (c >= '0' && c <= '9')
743 i = c - '0';
744 else if (c >= 'a' && c <= 'f')
745 i = c - 'a' + 10;
c906108c
SS
746 else
747 return ERROR; /* Char not a digit */
0f6e1ba6
AC
748 if (i >= base)
749 return ERROR; /* Invalid digit in this base */
750 n *= base;
751 n += i;
c906108c 752 }
c906108c
SS
753 /* Portably test for overflow (only works for nonzero values, so make
754 a second check for zero). */
755 if ((prevn >= n) && n != 0)
756 unsigned_p=1; /* Try something unsigned */
757 /* If range checking enabled, portably test for unsigned overflow. */
758 if (RANGE_CHECK && n != 0)
759 {
760 if ((unsigned_p && (unsigned)prevn >= (unsigned)n))
761 range_error("Overflow on numeric constant.");
762 }
763 prevn = n;
764 }
765
766 /* If the number is too big to be an int, or it's got an l suffix
767 then it's a long. Work out if this has to be a long by
768 shifting right and and seeing if anything remains, and the
769 target int size is different to the target long size.
770
771 In the expression below, we could have tested
3e79cecf 772 (n >> gdbarch_int_bit (parse_gdbarch))
c906108c
SS
773 to see if it was zero,
774 but too many compilers warn about that, when ints and longs
775 are the same size. So we shift it twice, with fewer bits
776 each time, for the same result. */
777
3e79cecf 778 if ((gdbarch_int_bit (parse_gdbarch) != gdbarch_long_bit (parse_gdbarch)
9a76efb6 779 && ((n >> 2)
3e79cecf 780 >> (gdbarch_int_bit (parse_gdbarch)-2))) /* Avoid shift warning */
c906108c
SS
781 || long_p)
782 {
3e79cecf
UW
783 high_bit = ((ULONGEST)1) << (gdbarch_long_bit (parse_gdbarch)-1);
784 unsigned_type = parse_type->builtin_unsigned_long;
785 signed_type = parse_type->builtin_long;
c906108c
SS
786 }
787 else
788 {
3e79cecf
UW
789 high_bit = ((ULONGEST)1) << (gdbarch_int_bit (parse_gdbarch)-1);
790 unsigned_type = parse_type->builtin_unsigned_int;
791 signed_type = parse_type->builtin_int;
c906108c
SS
792 }
793
794 putithere->typed_val.val = n;
795
796 /* If the high bit of the worked out type is set then this number
797 has to be unsigned. */
798
799 if (unsigned_p || (n & high_bit))
800 putithere->typed_val.type = unsigned_type;
801 else
802 putithere->typed_val.type = signed_type;
803
804 return INT;
805}
806
807struct token
808{
809 char *operator;
810 int token;
811 enum exp_opcode opcode;
812};
813
814static const struct token dot_ops[] =
815{
816 { ".and.", BOOL_AND, BINOP_END },
817 { ".AND.", BOOL_AND, BINOP_END },
818 { ".or.", BOOL_OR, BINOP_END },
819 { ".OR.", BOOL_OR, BINOP_END },
820 { ".not.", BOOL_NOT, BINOP_END },
821 { ".NOT.", BOOL_NOT, BINOP_END },
822 { ".eq.", EQUAL, BINOP_END },
823 { ".EQ.", EQUAL, BINOP_END },
824 { ".eqv.", EQUAL, BINOP_END },
825 { ".NEQV.", NOTEQUAL, BINOP_END },
826 { ".neqv.", NOTEQUAL, BINOP_END },
827 { ".EQV.", EQUAL, BINOP_END },
828 { ".ne.", NOTEQUAL, BINOP_END },
829 { ".NE.", NOTEQUAL, BINOP_END },
830 { ".le.", LEQ, BINOP_END },
831 { ".LE.", LEQ, BINOP_END },
832 { ".ge.", GEQ, BINOP_END },
833 { ".GE.", GEQ, BINOP_END },
834 { ".gt.", GREATERTHAN, BINOP_END },
835 { ".GT.", GREATERTHAN, BINOP_END },
836 { ".lt.", LESSTHAN, BINOP_END },
837 { ".LT.", LESSTHAN, BINOP_END },
838 { NULL, 0, 0 }
839};
840
841struct f77_boolean_val
842{
843 char *name;
844 int value;
845};
846
847static const struct f77_boolean_val boolean_values[] =
848{
849 { ".true.", 1 },
850 { ".TRUE.", 1 },
851 { ".false.", 0 },
852 { ".FALSE.", 0 },
853 { NULL, 0 }
854};
855
856static const struct token f77_keywords[] =
857{
858 { "complex_16", COMPLEX_S16_KEYWORD, BINOP_END },
859 { "complex_32", COMPLEX_S32_KEYWORD, BINOP_END },
860 { "character", CHARACTER, BINOP_END },
861 { "integer_2", INT_S2_KEYWORD, BINOP_END },
862 { "logical_1", LOGICAL_S1_KEYWORD, BINOP_END },
863 { "logical_2", LOGICAL_S2_KEYWORD, BINOP_END },
ce4b0682 864 { "logical_8", LOGICAL_S8_KEYWORD, BINOP_END },
c906108c
SS
865 { "complex_8", COMPLEX_S8_KEYWORD, BINOP_END },
866 { "integer", INT_KEYWORD, BINOP_END },
867 { "logical", LOGICAL_KEYWORD, BINOP_END },
868 { "real_16", REAL_S16_KEYWORD, BINOP_END },
869 { "complex", COMPLEX_S8_KEYWORD, BINOP_END },
870 { "sizeof", SIZEOF, BINOP_END },
871 { "real_8", REAL_S8_KEYWORD, BINOP_END },
872 { "real", REAL_KEYWORD, BINOP_END },
873 { NULL, 0, 0 }
874};
875
876/* Implementation of a dynamically expandable buffer for processing input
877 characters acquired through lexptr and building a value to return in
878 yylval. Ripped off from ch-exp.y */
879
880static char *tempbuf; /* Current buffer contents */
881static int tempbufsize; /* Size of allocated buffer */
882static int tempbufindex; /* Current index into buffer */
883
884#define GROWBY_MIN_SIZE 64 /* Minimum amount to grow buffer by */
885
886#define CHECKBUF(size) \
887 do { \
888 if (tempbufindex + (size) >= tempbufsize) \
889 { \
890 growbuf_by_size (size); \
891 } \
892 } while (0);
893
894
895/* Grow the static temp buffer if necessary, including allocating the first one
896 on demand. */
897
898static void
899growbuf_by_size (count)
900 int count;
901{
902 int growby;
903
904 growby = max (count, GROWBY_MIN_SIZE);
905 tempbufsize += growby;
906 if (tempbuf == NULL)
907 tempbuf = (char *) malloc (tempbufsize);
908 else
909 tempbuf = (char *) realloc (tempbuf, tempbufsize);
910}
911
912/* Blatantly ripped off from ch-exp.y. This routine recognizes F77
913 string-literals.
914
915 Recognize a string literal. A string literal is a nonzero sequence
916 of characters enclosed in matching single quotes, except that
917 a single character inside single quotes is a character literal, which
918 we reject as a string literal. To embed the terminator character inside
919 a string, it is simply doubled (I.E. 'this''is''one''string') */
920
921static int
922match_string_literal ()
923{
924 char *tokptr = lexptr;
925
926 for (tempbufindex = 0, tokptr++; *tokptr != '\0'; tokptr++)
927 {
928 CHECKBUF (1);
929 if (*tokptr == *lexptr)
930 {
931 if (*(tokptr + 1) == *lexptr)
932 tokptr++;
933 else
934 break;
935 }
936 tempbuf[tempbufindex++] = *tokptr;
937 }
938 if (*tokptr == '\0' /* no terminator */
939 || tempbufindex == 0) /* no string */
940 return 0;
941 else
942 {
943 tempbuf[tempbufindex] = '\0';
944 yylval.sval.ptr = tempbuf;
945 yylval.sval.length = tempbufindex;
946 lexptr = ++tokptr;
947 return STRING_LITERAL;
948 }
949}
950
951/* Read one token, getting characters through lexptr. */
952
953static int
954yylex ()
955{
956 int c;
957 int namelen;
958 unsigned int i,token;
959 char *tokstart;
960
961 retry:
065432a8
PM
962
963 prev_lexptr = lexptr;
964
c906108c
SS
965 tokstart = lexptr;
966
967 /* First of all, let us make sure we are not dealing with the
968 special tokens .true. and .false. which evaluate to 1 and 0. */
969
970 if (*lexptr == '.')
971 {
972 for (i = 0; boolean_values[i].name != NULL; i++)
973 {
bf896cb0
AC
974 if (strncmp (tokstart, boolean_values[i].name,
975 strlen (boolean_values[i].name)) == 0)
c906108c
SS
976 {
977 lexptr += strlen (boolean_values[i].name);
978 yylval.lval = boolean_values[i].value;
979 return BOOLEAN_LITERAL;
980 }
981 }
982 }
983
bd49c137 984 /* See if it is a special .foo. operator. */
c906108c
SS
985
986 for (i = 0; dot_ops[i].operator != NULL; i++)
bf896cb0 987 if (strncmp (tokstart, dot_ops[i].operator, strlen (dot_ops[i].operator)) == 0)
c906108c
SS
988 {
989 lexptr += strlen (dot_ops[i].operator);
990 yylval.opcode = dot_ops[i].opcode;
991 return dot_ops[i].token;
992 }
993
bd49c137
WZ
994 /* See if it is an exponentiation operator. */
995
996 if (strncmp (tokstart, "**", 2) == 0)
997 {
998 lexptr += 2;
999 yylval.opcode = BINOP_EXP;
1000 return STARSTAR;
1001 }
1002
c906108c
SS
1003 switch (c = *tokstart)
1004 {
1005 case 0:
1006 return 0;
1007
1008 case ' ':
1009 case '\t':
1010 case '\n':
1011 lexptr++;
1012 goto retry;
1013
1014 case '\'':
1015 token = match_string_literal ();
1016 if (token != 0)
1017 return (token);
1018 break;
1019
1020 case '(':
1021 paren_depth++;
1022 lexptr++;
1023 return c;
1024
1025 case ')':
1026 if (paren_depth == 0)
1027 return 0;
1028 paren_depth--;
1029 lexptr++;
1030 return c;
1031
1032 case ',':
1033 if (comma_terminates && paren_depth == 0)
1034 return 0;
1035 lexptr++;
1036 return c;
1037
1038 case '.':
1039 /* Might be a floating point number. */
1040 if (lexptr[1] < '0' || lexptr[1] > '9')
1041 goto symbol; /* Nope, must be a symbol. */
1042 /* FALL THRU into number case. */
1043
1044 case '0':
1045 case '1':
1046 case '2':
1047 case '3':
1048 case '4':
1049 case '5':
1050 case '6':
1051 case '7':
1052 case '8':
1053 case '9':
1054 {
1055 /* It's a number. */
1056 int got_dot = 0, got_e = 0, got_d = 0, toktype;
710122da 1057 char *p = tokstart;
c906108c
SS
1058 int hex = input_radix > 10;
1059
1060 if (c == '0' && (p[1] == 'x' || p[1] == 'X'))
1061 {
1062 p += 2;
1063 hex = 1;
1064 }
1065 else if (c == '0' && (p[1]=='t' || p[1]=='T' || p[1]=='d' || p[1]=='D'))
1066 {
1067 p += 2;
1068 hex = 0;
1069 }
1070
1071 for (;; ++p)
1072 {
1073 if (!hex && !got_e && (*p == 'e' || *p == 'E'))
1074 got_dot = got_e = 1;
1075 else if (!hex && !got_d && (*p == 'd' || *p == 'D'))
1076 got_dot = got_d = 1;
1077 else if (!hex && !got_dot && *p == '.')
1078 got_dot = 1;
1079 else if (((got_e && (p[-1] == 'e' || p[-1] == 'E'))
1080 || (got_d && (p[-1] == 'd' || p[-1] == 'D')))
1081 && (*p == '-' || *p == '+'))
1082 /* This is the sign of the exponent, not the end of the
1083 number. */
1084 continue;
1085 /* We will take any letters or digits. parse_number will
1086 complain if past the radix, or if L or U are not final. */
1087 else if ((*p < '0' || *p > '9')
1088 && ((*p < 'a' || *p > 'z')
1089 && (*p < 'A' || *p > 'Z')))
1090 break;
1091 }
1092 toktype = parse_number (tokstart, p - tokstart, got_dot|got_e|got_d,
1093 &yylval);
1094 if (toktype == ERROR)
1095 {
1096 char *err_copy = (char *) alloca (p - tokstart + 1);
1097
1098 memcpy (err_copy, tokstart, p - tokstart);
1099 err_copy[p - tokstart] = 0;
1100 error ("Invalid number \"%s\".", err_copy);
1101 }
1102 lexptr = p;
1103 return toktype;
1104 }
1105
1106 case '+':
1107 case '-':
1108 case '*':
1109 case '/':
1110 case '%':
1111 case '|':
1112 case '&':
1113 case '^':
1114 case '~':
1115 case '!':
1116 case '@':
1117 case '<':
1118 case '>':
1119 case '[':
1120 case ']':
1121 case '?':
1122 case ':':
1123 case '=':
1124 case '{':
1125 case '}':
1126 symbol:
1127 lexptr++;
1128 return c;
1129 }
1130
f55ee35c 1131 if (!(c == '_' || c == '$' || c ==':'
c906108c
SS
1132 || (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')))
1133 /* We must have come across a bad character (e.g. ';'). */
1134 error ("Invalid character '%c' in expression.", c);
1135
1136 namelen = 0;
1137 for (c = tokstart[namelen];
f55ee35c 1138 (c == '_' || c == '$' || c == ':' || (c >= '0' && c <= '9')
c906108c
SS
1139 || (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z'));
1140 c = tokstart[++namelen]);
1141
1142 /* The token "if" terminates the expression and is NOT
1143 removed from the input stream. */
1144
1145 if (namelen == 2 && tokstart[0] == 'i' && tokstart[1] == 'f')
1146 return 0;
1147
1148 lexptr += namelen;
1149
1150 /* Catch specific keywords. */
1151
1152 for (i = 0; f77_keywords[i].operator != NULL; i++)
45ccdddc
TT
1153 if (strlen (f77_keywords[i].operator) == namelen
1154 && strncmp (tokstart, f77_keywords[i].operator, namelen) == 0)
c906108c
SS
1155 {
1156 /* lexptr += strlen(f77_keywords[i].operator); */
1157 yylval.opcode = f77_keywords[i].opcode;
1158 return f77_keywords[i].token;
1159 }
1160
1161 yylval.sval.ptr = tokstart;
1162 yylval.sval.length = namelen;
1163
1164 if (*tokstart == '$')
1165 {
1166 write_dollar_variable (yylval.sval);
1167 return VARIABLE;
1168 }
1169
1170 /* Use token-type TYPENAME for symbols that happen to be defined
1171 currently as names of types; NAME for other symbols.
1172 The caller is not constrained to care about the distinction. */
1173 {
1174 char *tmp = copy_name (yylval.sval);
1175 struct symbol *sym;
1176 int is_a_field_of_this = 0;
1177 int hextype;
1178
1179 sym = lookup_symbol (tmp, expression_context_block,
176620f1 1180 VAR_DOMAIN,
3e79cecf 1181 parse_language->la_language == language_cplus
2570f2b7 1182 ? &is_a_field_of_this : NULL);
c906108c
SS
1183 if (sym && SYMBOL_CLASS (sym) == LOC_TYPEDEF)
1184 {
1185 yylval.tsym.type = SYMBOL_TYPE (sym);
1186 return TYPENAME;
1187 }
54a5b07d 1188 yylval.tsym.type
3e79cecf
UW
1189 = language_lookup_primitive_type_by_name (parse_language,
1190 parse_gdbarch, tmp);
54a5b07d 1191 if (yylval.tsym.type != NULL)
c906108c
SS
1192 return TYPENAME;
1193
1194 /* Input names that aren't symbols but ARE valid hex numbers,
1195 when the input radix permits them, can be names or numbers
1196 depending on the parse. Note we support radixes > 16 here. */
1197 if (!sym
1198 && ((tokstart[0] >= 'a' && tokstart[0] < 'a' + input_radix - 10)
1199 || (tokstart[0] >= 'A' && tokstart[0] < 'A' + input_radix - 10)))
1200 {
1201 YYSTYPE newlval; /* Its value is ignored. */
1202 hextype = parse_number (tokstart, namelen, 0, &newlval);
1203 if (hextype == INT)
1204 {
1205 yylval.ssym.sym = sym;
1206 yylval.ssym.is_a_field_of_this = is_a_field_of_this;
1207 return NAME_OR_INT;
1208 }
1209 }
1210
1211 /* Any other kind of symbol */
1212 yylval.ssym.sym = sym;
1213 yylval.ssym.is_a_field_of_this = is_a_field_of_this;
1214 return NAME;
1215 }
1216}
1217
1218void
1219yyerror (msg)
1220 char *msg;
1221{
065432a8
PM
1222 if (prev_lexptr)
1223 lexptr = prev_lexptr;
1224
c906108c
SS
1225 error ("A %s in expression, near `%s'.", (msg ? msg : "error"), lexptr);
1226}
This page took 1.522486 seconds and 4 git commands to generate.