ebrowse.c 93.2 KB
Newer Older
1 2
/* ebrowse.c --- parsing files for the ebrowse C++ browser

3
Copyright (C) 1992-2020 Free Software Foundation, Inc.
4 5 6 7 8

This file is part of GNU Emacs.

GNU Emacs is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
9 10
the Free Software Foundation, either version 3 of the License, or (at
your option) any later version.
11 12 13 14 15 16 17

GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
18
along with GNU Emacs.  If not, see <https://www.gnu.org/licenses/>.  */
19

20

21
#include <config.h>
22
#include <stddef.h>
23 24 25 26
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <assert.h>
27
#include <getopt.h>
28

29 30 31 32
#include <flexmember.h>
#include <min-max.h>
#include <unlocked-io.h>

33 34 35 36 37
/* The SunOS compiler doesn't have SEEK_END.  */
#ifndef SEEK_END
#define SEEK_END 2
#endif

38 39
/* Files are read in chunks of this number of bytes.  */

40
enum { READ_CHUNK_SIZE = 100 * 1024 };
41

42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
/* Value is true if strings X and Y compare equal.  */

static bool
streq (char const *x, char const *y)
{
  return strcmp (x, y) == 0;
}

static bool
filename_eq (char const *x, char const *y)
{
#ifdef __MSDOS__
  return strcasecmp (x, y) == 0;
#elif defined WINDOWSNT
  return stricmp (x, y) == 0;
57
#else
58
  return streq (x, y);
59
#endif
60 61
}

62 63
/* The default output file name.  */

64
#define DEFAULT_OUTFILE "BROWSE"
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

/* A version string written to the output file.  Change this whenever
   the structure of the output file changes.  */

#define EBROWSE_FILE_VERSION "ebrowse 5.0"

/* The output file consists of a tree of Lisp objects, with major
   nodes built out of Lisp structures.  These are the heads of the
   Lisp structs with symbols identifying their type.  */

#define TREE_HEADER_STRUCT	"[ebrowse-hs "
#define TREE_STRUCT		"[ebrowse-ts "
#define MEMBER_STRUCT		"[ebrowse-ms "
#define CLASS_STRUCT		"[ebrowse-cs "

/* The name of the symbol table entry for global functions, variables,
   defines etc.  This name also appears in the browser display.  */

#define GLOBALS_NAME "*Globals*"

/* Token definitions.  */

enum token
{
  YYEOF = 0,			/* end of file */
  CSTRING = 256,		/* string constant */
  CCHAR,			/* character constant */
  CINT,				/* integral constant */
  CFLOAT,			/* real constant */

  ELLIPSIS,			/* ... */
  LSHIFTASGN,			/* <<= */
  RSHIFTASGN,			/* >>= */
  ARROWSTAR,			/* ->* */
  IDENT,			/* identifier */
  DIVASGN,			/* /= */
  INC,				/* ++ */
  ADDASGN,			/* += */
  DEC,				/* -- */
  ARROW,			/* -> */
  SUBASGN,			/* -= */
  MULASGN,			/* *= */
  MODASGN,			/* %= */
  LOR,				/* || */
  ORASGN,			/* |= */
  LAND,				/* && */
  ANDASGN,			/* &= */
  XORASGN,			/* ^= */
  POINTSTAR,			/* .* */
  DCOLON,			/* :: */
  EQ,				/* == */
  NE,				/* != */
  LE,				/* <= */
  LSHIFT,			/* << */
  GE,				/* >= */
  RSHIFT,			/* >> */

/* Keywords.  The undef's are there because these
   three symbols are very likely to be defined somewhere.  */
#undef BOOL
#undef TRUE
#undef FALSE

  ASM,				/* asm */
  AUTO,				/* auto */
  BREAK,			/* break */
  CASE,				/* case  */
  CATCH,			/* catch */
  CHAR,				/* char */
  CLASS,			/* class */
  CONST,			/* const */
  CONTINUE,			/* continue */
  DEFAULT,			/* default */
  DELETE,			/* delete */
  DO,				/* do */
  DOUBLE,			/* double */
  ELSE,				/* else */
  ENUM,				/* enum */
  EXTERN,			/* extern */
  FLOAT,			/* float */
  FOR,				/* for */
  FRIEND,			/* friend */
  GOTO,				/* goto */
  IF,				/* if */
  T_INLINE,			/* inline */
  INT,				/* int */
  LONG,				/* long */
  NEW,				/* new */
  OPERATOR,			/* operator */
  PRIVATE,			/* private */
  PROTECTED,			/* protected */
  PUBLIC,			/* public */
  REGISTER,			/* register */
  RETURN,			/* return */
  SHORT,			/* short */
  SIGNED,			/* signed */
  SIZEOF,			/* sizeof */
  STATIC,			/* static */
  STRUCT,			/* struct */
  SWITCH,			/* switch */
  TEMPLATE,			/* template */
  THIS,				/* this */
  THROW,			/* throw */
  TRY,				/* try */
  TYPEDEF,			/* typedef */
  UNION,			/* union */
  UNSIGNED,			/* unsigned */
  VIRTUAL,			/* virtual */
  VOID,				/* void */
  VOLATILE,			/* volatile */
  WHILE,			/* while */
  MUTABLE,			/* mutable */
  BOOL,				/* bool */
  TRUE,				/* true */
  FALSE,			/* false */
  SIGNATURE,			/* signature (GNU extension) */
  NAMESPACE,			/* namespace */
  EXPLICIT,			/* explicit */
  TYPENAME,			/* typename */
  CONST_CAST,			/* const_cast */
  DYNAMIC_CAST,			/* dynamic_cast */
  REINTERPRET_CAST,		/* reinterpret_cast */
  STATIC_CAST,			/* static_cast */
  TYPEID,			/* typeid */
  USING,			/* using */
190 191
  WCHAR,			/* wchar_t */
  FINAL                         /* final */
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
};

/* Storage classes, in a wider sense.  */

enum sc
{
  SC_UNKNOWN,
  SC_MEMBER,			/* Is an instance member.  */
  SC_STATIC,			/* Is static member.  */
  SC_FRIEND,			/* Is friend function.  */
  SC_TYPE			/* Is a type definition.  */
};

/* Member visibility.  */

enum visibility
{
  V_PUBLIC,
  V_PROTECTED,
  V_PRIVATE
};

/* Member flags.  */

#define F_VIRTUAL	1	/* Is virtual function.  */
#define F_INLINE	2	/* Is inline function.  */
#define F_CONST		4	/* Is const.  */
#define F_PURE		8	/* Is pure virtual function.  */
#define F_MUTABLE	16	/* Is mutable.  */
#define F_TEMPLATE	32	/* Is a template.  */
#define F_EXPLICIT	64	/* Is explicit constructor.  */
#define F_THROW		128	/* Has a throw specification.  */
#define F_EXTERNC	256	/* Is declared extern "C".  */
#define F_DEFINE	512	/* Is a #define.  */

227
/* Set and test a bit in an int.  */
228

229 230 231 232 233 234 235 236 237 238 239
static void
set_flag (int *f, int flag)
{
  *f |= flag;
}

static bool
has_flag (int f, int flag)
{
  return (f & flag) != 0;
}
240 241 242 243 244 245 246 247 248 249 250 251

/* Structure describing a class member.  */

struct member
{
  struct member *next;		/* Next in list of members.  */
  struct member *anext;		/* Collision chain in member_table.  */
  struct member **list;		/* Pointer to list in class.  */
  unsigned param_hash;		/* Hash value for parameter types.  */
  int vis;			/* Visibility (public, ...).  */
  int flags;			/* See F_* above.  */
  char *regexp;			/* Matching regular expression.  */
252
  const char *filename;		/* Don't free this shared string.  */
253 254
  int pos;			/* Buffer position of occurrence.  */
  char *def_regexp;		/* Regular expression matching definition.  */
255
  const char *def_filename;	/* File name of definition.  */
256
  int def_pos;			/* Buffer position of definition.  */
257
  char name[FLEXIBLE_ARRAY_MEMBER]; /* Member name.  */
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
};

/* Structures of this type are used to connect class structures with
   their super and subclasses.  */

struct link
{
  struct sym *sym;		/* The super or subclass.  */
  struct link *next;		/* Next in list or NULL.  */
};

/* Structure used to record namespace aliases.  */

struct alias
{
  struct alias *next;		/* Next in list.  */
274 275
  struct sym *namesp;		/* Namespace in which defined.  */
  struct link *aliasee;		/* List of aliased namespaces (A::B::C...).  */
276
  char name[FLEXIBLE_ARRAY_MEMBER]; /* Alias name.  */
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
};

/* The structure used to describe a class in the symbol table,
   or a namespace in all_namespaces.  */

struct sym
{
  int flags;			/* Is class a template class?.  */
  unsigned char visited;	/* Used to find circles.  */
  struct sym *next;             /* Hash collision list.  */
  struct link *subs;		/* List of subclasses.  */
  struct link *supers;		/* List of superclasses.  */
  struct member *vars;		/* List of instance variables.  */
  struct member *fns;		/* List of instance functions.  */
  struct member *static_vars;	/* List of static variables.  */
  struct member *static_fns;	/* List of static functions.  */
  struct member *friends;	/* List of friend functions.  */
  struct member *types;		/* List of local types.  */
  char *regexp;			/* Matching regular expression.  */
  int pos;			/* Buffer position.  */
297 298
  const char *filename;		/* File in which it can be found.  */
  const char *sfilename; 	/* File in which members can be found.  */
299
  struct sym *namesp;		/* Namespace in which defined. .  */
300
  char name[FLEXIBLE_ARRAY_MEMBER]; /* Name of the class.  */
301 302 303 304 305 306 307 308
};

/* Experimental: Print info for `--position-info'.  We print
   '(CLASS-NAME SCOPE MEMBER-NAME).  */

#define P_DEFN	1
#define P_DECL  2

309 310 311
static int info_where;
static struct sym *info_cls = NULL;
static struct member *info_member = NULL;
312 313 314 315 316

/* Experimental.  For option `--position-info', the buffer position we
   are interested in.  When this position is reached, print out
   information about what we know about that point.  */

317
static int info_position = -1;
318 319 320

/* Command line options structure for getopt_long.  */

321
static struct option const options[] =
322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
{
  {"append", 			no_argument, 	   NULL, 'a'},
  {"files", 			required_argument, NULL, 'f'},
  {"help", 			no_argument, 	   NULL, -2},
  {"min-regexp-length", 	required_argument, NULL, 'm'},
  {"max-regexp-length", 	required_argument, NULL, 'M'},
  {"no-nested-classes", 	no_argument, 	   NULL, 'n'},
  {"no-regexps", 		no_argument, 	   NULL, 'x'},
  {"no-structs-or-unions", 	no_argument, 	   NULL, 's'},
  {"output-file", 		required_argument, NULL, 'o'},
  {"position-info", 		required_argument, NULL, 'p'},
  {"search-path", 		required_argument, NULL, 'I'},
  {"verbose", 			no_argument, 	   NULL, 'v'},
  {"version", 			no_argument, 	   NULL, -3},
  {"very-verbose", 		no_argument, 	   NULL, 'V'},
  {NULL, 			0, 		   NULL, 0}
};

/* Semantic values of tokens.  Set by yylex..  */

342 343 344
static unsigned yyival;		/* Set for token CINT.  */
static char *yytext;		/* Set for token IDENT.  */
static char *yytext_end;
345 346 347

/* Output file.  */

348
static FILE *yyout;
349 350 351

/* Current line number.  */

352
static int yyline;
353 354 355

/* The name of the current input file.  */

356
static const char *filename;
357 358 359 360

/* Three character class vectors, and macros to test membership
   of characters.  */

361 362 363
static char is_ident[255];
static char is_digit[255];
static char is_white[255];
364 365 366 367 368 369 370

#define IDENTP(C)	is_ident[(unsigned char) (C)]
#define DIGITP(C)	is_digit[(unsigned char) (C)]
#define WHITEP(C)	is_white[(unsigned char) (C)]

/* Command line flags.  */

371 372 373 374 375 376
static int f_append;
static int f_verbose;
static int f_very_verbose;
static int f_structs = 1;
static int f_regexps = 1;
static int f_nested_classes = 1;
377 378 379 380 381

/* Maximum and minimum lengths of regular expressions matching a
   member, class etc., for writing them to the output file.  These are
   overridable from the command line.  */

382 383
static int min_regexp = 5;
static int max_regexp = 50;
384 385 386

/* Input buffer.  */

387 388 389
static char *inbuffer;
static char *in;
static size_t inbuffer_size;
390 391 392 393 394

/* Return the current buffer position in the input file.  */

#define BUFFER_POS() (in - inbuffer)

395
/* If current lookahead is CSTRING, the following points to the
396 397 398
   first character in the string constant.  Used for recognizing
   extern "C".  */

399
static char *string_start;
400 401 402 403 404 405 406 407

/* The size of the hash tables for classes.and members.  Should be
   prime.  */

#define TABLE_SIZE 1001

/* The hash table for class symbols.  */

408
static struct sym *class_table[TABLE_SIZE];
409 410 411 412 413

/* Hash table containing all member structures.  This is generally
   faster for member lookup than traversing the member lists of a
   `struct sym'.  */

414
static struct member *member_table[TABLE_SIZE];
415

416 417
/* Hash table for namespace aliases */

418
static struct alias *namespace_alias_table[TABLE_SIZE];
419

420 421 422
/* The special class symbol used to hold global functions,
   variables etc.  */

423
static struct sym *global_symbols;
424 425 426

/* The current namespace.  */

427
static struct sym *current_namespace;
428 429 430

/* The list of all known namespaces.  */

431
static struct sym *all_namespaces;
432 433 434

/* Stack of namespaces we're currently nested in, during the parse.  */

435 436 437
static struct sym **namespace_stack;
static int namespace_stack_size;
static int namespace_sp;
438 439 440

/* The current lookahead token.  */

441
static int tk = -1;
442 443 444 445 446

/* Structure describing a keyword.  */

struct kw
{
447
  const char *name;		/* Spelling.  */
448 449 450 451 452 453 454
  int tk;			/* Token value.  */
  struct kw *next;		/* Next in collision chain.  */
};

/* Keywords are lookup up in a hash table of their own.  */

#define KEYWORD_TABLE_SIZE 1001
455
static struct kw *keyword_table[KEYWORD_TABLE_SIZE];
456 457 458 459 460 461 462 463 464

/* Search path.  */

struct search_path
{
  char *path;
  struct search_path *next;
};

465 466
static struct search_path *search_path;
static struct search_path *search_path_tail;
467 468 469

/* Function prototypes.  */

470 471 472 473 474
static char *matching_regexp (void);
static struct sym *add_sym (const char *, struct sym *);
static void add_global_defn (char *, char *, int, unsigned, int, int, int);
static void add_global_decl (char *, char *, int, unsigned, int, int, int);
static struct member *add_member (struct sym *, char *, int, int, unsigned);
475
static void class_definition (struct sym *, const char *, int, int, int);
476 477
static char *operator_name (int *);
static void parse_qualified_param_ident_or_type (char **);
478 479 480 481 482 483 484 485

/***********************************************************************
			      Utilities
 ***********************************************************************/

/* Print an error in a printf-like style with the current input file
   name and line number.  */

486
static void
487
yyerror (const char *format, const char *s)
488 489
{
  fprintf (stderr, "%s:%d: ", filename, yyline);
490
  fprintf (stderr, format, s);
491 492 493 494 495
  putc ('\n', stderr);
}


/* Like malloc but print an error and exit if not enough memory is
Dave Love's avatar
Dave Love committed
496
   available.  */
497

498
static void * ATTRIBUTE_MALLOC
499
xmalloc (size_t nbytes)
500 501
{
  void *p = malloc (nbytes);
502 503
  if (p == NULL)
    {
504
      yyerror ("out of memory", NULL);
505
      exit (EXIT_FAILURE);
506 507
    }
  return p;
508 509 510 511 512
}


/* Like realloc but print an error and exit if out of memory.  */

513
static void *
514
xrealloc (void *p, size_t sz)
515 516
{
  p = realloc (p, sz);
517 518
  if (p == NULL)
    {
519
      yyerror ("out of memory", NULL);
520
      exit (EXIT_FAILURE);
521 522
    }
  return p;
523 524 525 526 527 528
}


/* Like strdup, but print an error and exit if not enough memory is
   available..  If S is null, return null.  */

529
static char *
530
xstrdup (char *s)
531 532
{
  if (s)
Paul Eggert's avatar
Paul Eggert committed
533
    return strcpy (xmalloc (strlen (s) + 1), s);
534 535 536 537 538 539 540 541 542 543 544 545
  return s;
}



/***********************************************************************
			       Symbols
 ***********************************************************************/

/* Initialize the symbol table.  This currently only sets up the
   special symbol for globals (`*Globals*').  */

546
static void
547
init_sym (void)
548 549 550 551 552 553 554 555 556 557 558 559
{
  global_symbols = add_sym (GLOBALS_NAME, NULL);
}


/* Add a symbol for class NAME to the symbol table.  NESTED_IN_CLASS
   is the class in which class NAME was found.  If it is null,
   this means the scope of NAME is the current namespace.

   If a symbol for NAME already exists, return that.  Otherwise
   create a new symbol and set it to default values.  */

560
static struct sym *
561
add_sym (const char *name, struct sym *nested_in_class)
562 563 564
{
  struct sym *sym;
  unsigned h;
565
  const char *s;
566 567 568 569 570 571 572
  struct sym *scope = nested_in_class ? nested_in_class : current_namespace;

  for (s = name, h = 0; *s; ++s)
    h = (h << 1) ^ *s;
  h %= TABLE_SIZE;

  for (sym = class_table[h]; sym; sym = sym->next)
573 574 575 576
    if (streq (name, sym->name)
	&& ((!sym->namesp && !scope)
	    || (sym->namesp && scope
		&& streq (sym->namesp->name, scope->name))))
577 578 579 580 581 582 583 584 585 586
      break;

  if (sym == NULL)
    {
      if (f_very_verbose)
	{
	  putchar ('\t');
	  puts (name);
	}

587
      sym = xmalloc (FLEXSIZEOF (struct sym, name, strlen (name) + 1));
588
      memset (sym, 0, offsetof (struct sym, name));
589 590 591 592 593 594 595 596 597 598 599 600
      strcpy (sym->name, name);
      sym->namesp = scope;
      sym->next = class_table[h];
      class_table[h] = sym;
    }

  return sym;
}


/* Add links between superclass SUPER and subclass SUB.  */

601
static void
602
add_link (struct sym *super, struct sym *sub)
603 604 605 606 607 608 609 610 611 612 613 614
{
  struct link *lnk, *lnk2, *p, *prev;

  /* See if a link already exists.  */
  for (p = super->subs, prev = NULL;
       p && strcmp (sub->name, p->sym->name) > 0;
       prev = p, p = p->next)
    ;

  /* Avoid duplicates.  */
  if (p == NULL || p->sym != sub)
    {
615 616
      lnk = (struct link *) xmalloc (sizeof *lnk);
      lnk2 = (struct link *) xmalloc (sizeof *lnk2);
617

618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640
      lnk->sym = sub;
      lnk->next = p;

      if (prev)
	prev->next = lnk;
      else
	super->subs = lnk;

      lnk2->sym = super;
      lnk2->next = sub->supers;
      sub->supers = lnk2;
    }
}


/* Find in class CLS member NAME.

   VAR non-zero means look for a member variable; otherwise a function
   is searched.  SC specifies what kind of member is searched---a
   static, or per-instance member etc.  HASH is a hash code for the
   parameter types of functions.  Value is a pointer to the member
   found or null if not found.  */

641
static struct member *
642
find_member (struct sym *cls, char *name, int var, int sc, unsigned int hash)
643 644 645 646 647 648 649 650 651 652 653 654
{
  struct member **list;
  struct member *p;
  unsigned name_hash = 0;
  char *s;
  int i;

  switch (sc)
    {
    case SC_FRIEND:
      list = &cls->friends;
      break;
655

656 657 658
    case SC_TYPE:
      list = &cls->types;
      break;
659

660 661 662
    case SC_STATIC:
      list = var ? &cls->static_vars : &cls->static_fns;
      break;
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
    default:
      list = var ? &cls->vars : &cls->fns;
      break;
    }

  for (s = name; *s; ++s)
    name_hash = (name_hash << 1) ^ *s;
  i = name_hash % TABLE_SIZE;

  for (p = member_table[i]; p; p = p->anext)
    if (p->list == list && p->param_hash == hash && streq (name, p->name))
      break;

  return p;
}


/* Add to class CLS information for the declaration of member NAME.
   REGEXP is a regexp matching the declaration, if non-null.  POS is
   the position in the source where the declaration is found.  HASH is
   a hash code for the parameter list of the member, if it's a
   function.  VAR non-zero means member is a variable or type.  SC
   specifies the type of member (instance member, static, ...).  VIS
   is the member's visibility (public, protected, private).  FLAGS is
   a bit set giving additional information about the member (see the
   F_* defines).  */

691
static void
692
add_member_decl (struct sym *cls, char *name, char *regexp, int pos, unsigned int hash, int var, int sc, int vis, int flags)
693 694 695 696 697 698 699 700
{
  struct member *m;

  m = find_member (cls, name, var, sc, hash);
  if (m == NULL)
    m = add_member (cls, name, var, sc, hash);

  /* Have we seen a new filename?  If so record that.  */
701
  if (!cls->filename || !filename_eq (cls->filename, filename))
702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
    m->filename = filename;

  m->regexp = regexp;
  m->pos = pos;
  m->flags = flags;

  switch (vis)
    {
    case PRIVATE:
      m->vis = V_PRIVATE;
      break;

    case PROTECTED:
      m->vis = V_PROTECTED;
      break;

    case PUBLIC:
      m->vis = V_PUBLIC;
      break;
    }

  info_where = P_DECL;
  info_cls = cls;
  info_member = m;
}


/* Add to class CLS information for the definition of member NAME.
   REGEXP is a regexp matching the declaration, if non-null.  POS is
   the position in the source where the declaration is found.  HASH is
   a hash code for the parameter list of the member, if it's a
   function.  VAR non-zero means member is a variable or type.  SC
   specifies the type of member (instance member, static, ...).  VIS
   is the member's visibility (public, protected, private).  FLAGS is
   a bit set giving additional information about the member (see the
   F_* defines).  */

739
static void
740
add_member_defn (struct sym *cls, char *name, char *regexp, int pos, unsigned int hash, int var, int sc, int flags)
741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763
{
  struct member *m;

  if (sc == SC_UNKNOWN)
    {
      m = find_member (cls, name, var, SC_MEMBER, hash);
      if (m == NULL)
	{
	  m = find_member (cls, name, var, SC_STATIC, hash);
	  if (m == NULL)
	    m = add_member (cls, name, var, sc, hash);
	}
    }
  else
    {
      m = find_member (cls, name, var, sc, hash);
      if (m == NULL)
	m = add_member (cls, name, var, sc, hash);
    }

  if (!cls->sfilename)
    cls->sfilename = filename;

764
  if (!filename_eq (cls->sfilename, filename))
765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780
    m->def_filename = filename;

  m->def_regexp = regexp;
  m->def_pos = pos;
  m->flags |= flags;

  info_where = P_DEFN;
  info_cls = cls;
  info_member = m;
}


/* Add a symbol for a define named NAME to the symbol table.
   REGEXP is a regular expression matching the define in the source,
   if it is non-null.  POS is the position in the file.  */

781
static void
782
add_define (char *name, char *regexp, int pos)
783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
{
  add_global_defn (name, regexp, pos, 0, 1, SC_FRIEND, F_DEFINE);
  add_global_decl (name, regexp, pos, 0, 1, SC_FRIEND, F_DEFINE);
}


/* Add information for the global definition of NAME.
   REGEXP is a regexp matching the declaration, if non-null.  POS is
   the position in the source where the declaration is found.  HASH is
   a hash code for the parameter list of the member, if it's a
   function.  VAR non-zero means member is a variable or type.  SC
   specifies the type of member (instance member, static, ...).  VIS
   is the member's visibility (public, protected, private).  FLAGS is
   a bit set giving additional information about the member (see the
   F_* defines).  */

799
static void
800
add_global_defn (char *name, char *regexp, int pos, unsigned int hash, int var, int sc, int flags)
801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829
{
  int i;
  struct sym *sym;

  /* Try to find out for which classes a function is a friend, and add
     what we know about it to them.  */
  if (!var)
    for (i = 0; i < TABLE_SIZE; ++i)
      for (sym = class_table[i]; sym; sym = sym->next)
	if (sym != global_symbols && sym->friends)
	  if (find_member (sym, name, 0, SC_FRIEND, hash))
	    add_member_defn (sym, name, regexp, pos, hash, 0,
			     SC_FRIEND, flags);

  /* Add to global symbols.  */
  add_member_defn (global_symbols, name, regexp, pos, hash, var, sc, flags);
}


/* Add information for the global declaration of NAME.
   REGEXP is a regexp matching the declaration, if non-null.  POS is
   the position in the source where the declaration is found.  HASH is
   a hash code for the parameter list of the member, if it's a
   function.  VAR non-zero means member is a variable or type.  SC
   specifies the type of member (instance member, static, ...).  VIS
   is the member's visibility (public, protected, private).  FLAGS is
   a bit set giving additional information about the member (see the
   F_* defines).  */

830
static void
831
add_global_decl (char *name, char *regexp, int pos, unsigned int hash, int var, int sc, int flags)
832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848
{
  /* Add declaration only if not already declared.  Header files must
     be processed before source files for this to have the right effect.
     I do not want to handle implicit declarations at the moment.  */
  struct member *m;
  struct member *found;

  m = found = find_member (global_symbols, name, var, sc, hash);
  if (m == NULL)
    m = add_member (global_symbols, name, var, sc, hash);

  /* Definition already seen => probably last declaration implicit.
     Override.  This means that declarations must always be added to
     the symbol table before definitions.  */
  if (!found)
    {
      if (!global_symbols->filename
849
	  || !filename_eq (global_symbols->filename, filename))
850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868
	m->filename = filename;

      m->regexp = regexp;
      m->pos = pos;
      m->vis = V_PUBLIC;
      m->flags = flags;

      info_where = P_DECL;
      info_cls = global_symbols;
      info_member = m;
    }
}


/* Add a symbol for member NAME to class CLS.
   VAR non-zero means it's a variable.  SC specifies the kind of
   member.  HASH is a hash code for the parameter types of a function.
   Value is a pointer to the member's structure.  */

869
static struct member *
870
add_member (struct sym *cls, char *name, int var, int sc, unsigned int hash)
871
{
872 873
  struct member *m = xmalloc (FLEXSIZEOF (struct member, name,
					  strlen (name) + 1));
874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899
  struct member **list;
  struct member *p;
  struct member *prev;
  unsigned name_hash = 0;
  int i;
  char *s;

  strcpy (m->name, name);
  m->param_hash = hash;

  m->vis = 0;
  m->flags = 0;
  m->regexp = NULL;
  m->filename = NULL;
  m->pos = 0;
  m->def_regexp = NULL;
  m->def_filename = NULL;
  m->def_pos = 0;

  assert (cls != NULL);

  switch (sc)
    {
    case SC_FRIEND:
      list = &cls->friends;
      break;
900

901 902 903
    case SC_TYPE:
      list = &cls->types;
      break;
904

905 906 907
    case SC_STATIC:
      list = var ? &cls->static_vars : &cls->static_fns;
      break;
908

909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940
    default:
      list = var ? &cls->vars : &cls->fns;
      break;
    }

  for (s = name; *s; ++s)
    name_hash = (name_hash << 1) ^ *s;
  i = name_hash % TABLE_SIZE;
  m->anext = member_table[i];
  member_table[i] = m;
  m->list = list;

  /* Keep the member list sorted.  It's cheaper to do it here than to
     sort them in Lisp.  */
  for (prev = NULL, p = *list;
       p && strcmp (name, p->name) > 0;
       prev = p, p = p->next)
    ;

  m->next = p;
  if (prev)
    prev->next = m;
  else
    *list = m;
  return m;
}


/* Given the root R of a class tree, step through all subclasses
   recursively, marking functions as virtual that are declared virtual
   in base classes.  */

941
static void
942
mark_virtual (struct sym *r)
943 944 945 946 947 948 949
{
  struct link *p;
  struct member *m, *m2;

  for (p = r->subs; p; p = p->next)
    {
      for (m = r->fns; m; m = m->next)
950
        if (has_flag (m->flags, F_VIRTUAL))
951 952 953
          {
            for (m2 = p->sym->fns; m2; m2 = m2->next)
              if (m->param_hash == m2->param_hash && streq (m->name, m2->name))
954
                set_flag (&m2->flags, F_VIRTUAL);
955 956 957 958 959 960 961 962 963 964
          }

      mark_virtual (p->sym);
    }
}


/* For all roots of the class tree, mark functions as virtual that
   are virtual because of a virtual declaration in a base class.  */

965
static void
966
mark_inherited_virtual (void)
967 968 969 970 971 972 973 974 975 976 977 978 979
{
  struct sym *r;
  int i;

  for (i = 0; i < TABLE_SIZE; ++i)
    for (r = class_table[i]; r; r = r->next)
      if (r->supers == NULL)
        mark_virtual (r);
}


/* Create and return a symbol for a namespace with name NAME.  */

980
static struct sym *
981
make_namespace (char *name, struct sym *context)
982
{
983
  struct sym *s = xmalloc (FLEXSIZEOF (struct sym, name, strlen (name) + 1));
984
  memset (s, 0, offsetof (struct sym, name));
985 986
  strcpy (s->name, name);
  s->next = all_namespaces;
987
  s->namesp = context;
988 989 990 991 992
  all_namespaces = s;
  return s;
}


Paul Eggert's avatar
Paul Eggert committed
993
/* Find the symbol for namespace NAME.  If not found, return NULL */
994

995
static struct sym *
996
check_namespace (char *name, struct sym *context)
997
{
998
  struct sym *p = NULL;
999

1000 1001
  for (p = all_namespaces; p; p = p->next)
    {
1002
      if (streq (p->name, name) && (p->namesp == context))
1003
	    break;
1004
    }
1005 1006

  return p;
1007
}
1008

1009 1010 1011
/* Find the symbol for namespace NAME.  If not found, add a new symbol
   for NAME to all_namespaces.  */

1012
static struct sym *
1013
find_namespace (char *name, struct sym *context)
1014 1015 1016
{
  struct sym *p = check_namespace (name, context);

1017
  if (p == NULL)
1018
    p = make_namespace (name, context);
1019 1020 1021

  return p;
}
1022

1023

1024 1025
/* Find namespace alias with name NAME. If not found return NULL. */

1026
static struct link *
1027
check_namespace_alias (char *name)
1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048
{
  struct link *p = NULL;
  struct alias *al;
  unsigned h;
  char *s;

  for (s = name, h = 0; *s; ++s)
    h = (h << 1) ^ *s;
  h %= TABLE_SIZE;

  for (al = namespace_alias_table[h]; al; al = al->next)
    if (streq (name, al->name) && (al->namesp == current_namespace))
      {
        p = al->aliasee;
        break;
      }

  return p;
}

/* Register the name NEW_NAME as an alias for namespace list OLD_NAME.  */
1049

1050
static void
1051
register_namespace_alias (char *new_name, struct link *old_name)
1052
{
1053 1054
  unsigned h;
  char *s;
1055 1056
  struct alias *al;

1057 1058 1059 1060 1061 1062 1063 1064
  for (s = new_name, h = 0; *s; ++s)
    h = (h << 1) ^ *s;
  h %= TABLE_SIZE;


  /* Is it already in the table of aliases?  */
  for (al = namespace_alias_table[h]; al; al = al->next)
    if (streq (new_name, al->name) && (al->namesp == current_namespace))
1065 1066
      return;

1067
  al = xmalloc (FLEXSIZEOF (struct alias, name, strlen (new_name) + 1));
1068
  strcpy (al->name, new_name);
1069 1070 1071 1072
  al->next = namespace_alias_table[h];
  al->namesp = current_namespace;
  al->aliasee = old_name;
  namespace_alias_table[h] = al;
1073 1074 1075 1076 1077
}


/* Enter namespace with name NAME.  */

1078
static void
1079
enter_namespace (char *name)
1080
{
1081
  struct sym *p = find_namespace (name, current_namespace);
1082 1083 1084 1085

  if (namespace_sp == namespace_stack_size)
    {
      int size = max (10, 2 * namespace_stack_size);
1086 1087 1088
      namespace_stack
	= (struct sym **) xrealloc ((void *)namespace_stack,
				    size * sizeof *namespace_stack);
1089 1090
      namespace_stack_size = size;
    }
1091

1092 1093 1094 1095 1096 1097 1098
  namespace_stack[namespace_sp++] = current_namespace;
  current_namespace = p;
}


/* Leave the current namespace.  */

1099
static void
1100
leave_namespace (void)
1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114
{
  assert (namespace_sp > 0);
  current_namespace = namespace_stack[--namespace_sp];
}



/***********************************************************************
		       Writing the Output File
 ***********************************************************************/

/* Write string S to the output file FP in a Lisp-readable form.
   If S is null, write out `()'.  */

1115
static void
1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
putstr (const char *s, FILE *fp)
{
  if (!s)
    {
      putc ('(', fp);
      putc (')', fp);
      putc (' ', fp);
    }
  else
    {
      putc ('"', fp);
      fputs (s, fp);
      putc ('"', fp);
      putc (' ', fp);
    }
}
1132 1133 1134

/* A dynamically allocated buffer for constructing a scope name.  */

1135 1136 1137
static char *scope_buffer;
static int scope_buffer_size;
static int scope_buffer_len;
1138 1139 1140 1141


/* Make sure scope_buffer has enough room to add LEN chars to it.  */

1142
static void
1143
ensure_scope_buffer_room (int len)
1144 1145 1146 1147
{
  if (scope_buffer_len + len >= scope_buffer_size)
    {
      int new_size = max (2 * scope_buffer_size, scope_buffer_len + len);
1148
      scope_buffer = (char *) xrealloc (scope_buffer, new_size);
1149 1150 1151 1152 1153 1154 1155 1156 1157
      scope_buffer_size = new_size;
    }
}


/* Recursively add the scope names of symbol P and the scopes of its
   namespaces to scope_buffer.  Value is a pointer to the complete
   scope name constructed.  */

1158
static char *
1159
sym_scope_1 (struct sym *p)
1160 1161
{
  int len;
1162

1163 1164 1165 1166 1167 1168
  if (p->namesp)
    sym_scope_1 (p->namesp);

  if (*scope_buffer)
    {
      ensure_scope_buffer_room (3);
Paul Eggert's avatar
Paul Eggert committed
1169
      strcpy (scope_buffer + scope_buffer_len, "::");
1170 1171 1172 1173 1174
      scope_buffer_len += 2;
    }

  len = strlen (p->name);
  ensure_scope_buffer_room (len + 1);
Paul Eggert's avatar
Paul Eggert committed
1175
  strcpy (scope_buffer + scope_buffer_len, p->name);
1176
  scope_buffer_len += len;
1177

1178
  if (has_flag (p->flags, F_TEMPLATE))
1179 1180
    {
      ensure_scope_buffer_room (3);
Paul Eggert's avatar
Paul Eggert committed
1181
      strcpy (scope_buffer + scope_buffer_len, "<>");
1182 1183
      scope_buffer_len += 2;
    }
1184

1185 1186 1187 1188 1189 1190 1191
  return scope_buffer;
}


/* Return the scope of symbol P in printed representation, i.e.
   as it would appear in a C*+ source file.  */

1192
static char *
1193
sym_scope (struct sym *p)
1194 1195 1196 1197
{
  if (!scope_buffer)
    {
      scope_buffer_size = 1024;
1198
      scope_buffer = (char *) xmalloc (scope_buffer_size);
1199
    }
1200

1201 1202
  *scope_buffer = '\0';
  scope_buffer_len = 0;
1203

1204 1205 1206 1207 1208 1209 1210 1211 1212 1213
  if (p->namesp)
    sym_scope_1 (p->namesp);

  return scope_buffer;
}


/* Dump the list of members M to file FP.  Value is the length of the
   list.  */

1214
static int
1215
dump_members (FILE *fp, struct member *m)
1216 1217 1218 1219 1220 1221 1222 1223
{
  int n;

  putc ('(', fp);

  for (n = 0; m; m = m->next, ++n)
    {
      fputs (MEMBER_STRUCT, fp);
1224 1225
      putstr (m->name, fp);
      putstr (NULL, fp);		/* FIXME? scope for globals */
1226
      fprintf (fp, "%u ", (unsigned) m->flags);
1227 1228
      putstr (m->filename, fp);
      putstr (m->regexp, fp);
1229 1230 1231
      fprintf (fp, "%u ", (unsigned) m->pos);
      fprintf (fp, "%u ", (unsigned) m->vis);
      putc (' ', fp);
1232 1233
      putstr (m->def_filename, fp);
      putstr (m->def_regexp, fp);
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
      fprintf (fp, "%u", (unsigned) m->def_pos);
      putc (']', fp);
      putc ('\n', fp);
    }

  putc (')', fp);
  putc ('\n', fp);
  return n;
}


/* Dump class ROOT to stream FP.  */

1247
static void
1248
dump_sym (FILE *fp, struct sym *root)
1249 1250
{
  fputs (CLASS_STRUCT, fp);
1251
  putstr (root->name, fp);
1252

1253 1254
  /* Print scope, if any.  */
  if (root->namesp)
1255
    putstr (sym_scope (root), fp);
1256
  else
1257
    putstr (NULL, fp);
1258

1259
  /* Print flags.  */
1260
  fprintf (fp, "%d", root->flags);
1261 1262
  putstr (root->filename, fp);
  putstr (root->regexp, fp);
1263
  fprintf (fp, "%u", (unsigned) root->pos);
1264
  putstr (root->sfilename, fp);
1265 1266 1267 1268 1269 1270 1271 1272
  putc (']', fp);
  putc ('\n', fp);
}


/* Dump class ROOT and its subclasses to file FP.  Value is the
   number of classes written.  */

1273
static int
1274
dump_tree (FILE *fp, struct sym *root)
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319
{
  struct link *lk;
  unsigned n = 0;

  dump_sym (fp, root);

  if (f_verbose)
    {
      putchar ('+');
      fflush (stdout);
    }

  putc ('(', fp);

  for (lk = root->subs; lk; lk = lk->next)
    {
      fputs (TREE_STRUCT, fp);
      n += dump_tree (fp, lk->sym);
      putc (']', fp);
    }

  putc (')', fp);

  dump_members (fp, root->vars);
  n += dump_members (fp, root->fns);
  dump_members (fp, root->static_vars);
  n += dump_members (fp, root->static_fns);
  n += dump_members (fp, root->friends);
  dump_members (fp, root->types);

  /* Superclasses.  */
  putc ('(', fp);
  putc (')', fp);

  /* Mark slot.  */
  putc ('(', fp);
  putc (')', fp);

  putc ('\n', fp);
  return n;
}


/* Dump the entire class tree to file FP.  */

1320
static void
1321
dump_roots (FILE *fp)
1322 1323 1324 1325 1326 1327 1328 1329 1330
{
  int i, n = 0;
  struct sym *r;

  /* Output file header containing version string, command line
     options etc.  */
  if (!f_append)
    {
      fputs (TREE_HEADER_STRUCT, fp);
1331
      putstr (EBROWSE_FILE_VERSION, fp);
1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393

      putc ('\"', fp);
      if (!f_structs)
	fputs (" -s", fp);
      if (f_regexps)
	fputs (" -x", fp);
      putc ('\"', fp);
      fputs (" ()", fp);
      fputs (" ()", fp);
      putc (']', fp);
    }

  /* Mark functions as virtual that are so because of functions
     declared virtual in base classes.  */
  mark_inherited_virtual ();

  /* Dump the roots of the graph.  */
  for (i = 0; i < TABLE_SIZE; ++i)
    for (r = class_table[i]; r; r = r->next)
      if (!r->supers)
        {
	  fputs (TREE_STRUCT, fp);
          n += dump_tree (fp, r);
	  putc (']', fp);
        }

  if (f_verbose)
    putchar ('\n');
}



/***********************************************************************
				Scanner
 ***********************************************************************/

#ifdef DEBUG
#define INCREMENT_LINENO 			\
do {						\
  if (f_very_verbose)				\
    {						\
      ++yyline;					\
      printf ("%d:\n", yyline);			\
    }						\
  else						\
    ++yyline;					\
} while (0)
#else
#define INCREMENT_LINENO	++yyline
#endif

/* Define two macros for accessing the input buffer (current input
   file).  GET(C) sets C to the next input character and advances the
   input pointer.  UNGET retracts the input pointer.  */

#define GET(C)	((C) = *in++)
#define UNGET() (--in)


/* Process a preprocessor line.  Value is the next character from the
   input buffer not consumed.  */

1394
static int
1395
process_pp_line (void)
1396
{
1397
  int in_comment = 0, in_string = 0;
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
  int c;
  char *p = yytext;

  /* Skip over white space.  The `#' has been consumed already.  */
  while (WHITEP (GET (c)))
    ;

  /* Read the preprocessor command (if any).  */
  while (IDENTP (c))
    {
      *p++ = c;
      GET (c);
    }

  /* Is it a `define'?  */
  *p = '\0';

  if (*yytext && streq (yytext, "define"))
    {
      p = yytext;
      while (WHITEP (c))
	GET (c);
      while (IDENTP (c))
	{
	  *p++ = c;
	  GET (c);
	}

      *p = '\0';

      if (*yytext)
	{
	  char *regexp = matching_regexp ();
	  int pos = BUFFER_POS ();
	  add_define (yytext, regexp, pos);
	}
    }
1435

1436
  while (c && (c != '\n' || in_comment || in_string))
1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449
    {
      if (c == '\\')
	GET (c);
      else if (c == '/' && !in_comment)
	{
	  if (GET (c) == '*')
	    in_comment = 1;
	}
      else if (c == '*' && in_comment)
	{
	  if (GET (c) == '/')
	    in_comment = 0;
	}
1450 1451
      else if (c == '"')
	in_string = !in_string;
1452

1453 1454 1455 1456 1457
      if (c == '\n')
	INCREMENT_LINENO;

      GET (c);
    }
1458

1459 1460 1461 1462 1463 1464
  return c;
}


/* Value is the next token from the input buffer.  */

1465
static int
1466
yylex (void)
1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507
{
  int c;
  char end_char;
  char *p;

  for (;;)
    {
      while (WHITEP (GET (c)))
        ;

      switch (c)
        {
        case '\n':
          INCREMENT_LINENO;
          break;

        case '\r':
          break;

        case 0:
          /* End of file.  */
          return YYEOF;

        case '\\':
          GET (c);
          break;

        case '"':
        case '\'':
          /* String and character constants.  */
          end_char = c;
          string_start = in;
          while (GET (c) && c != end_char)
            {
              switch (c)
                {
                case '\\':
                  /* Escape sequences.  */
                  if (!GET (c))
                    {
                      if (end_char == '\'')
1508
                        yyerror ("EOF in character constant", NULL);
1509
                      else
1510
                        yyerror ("EOF in string constant", NULL);
1511 1512 1513 1514 1515
                      goto end_string;
                    }
                  else switch (c)
                    {
                    case '\n':
1516
		      INCREMENT_LINENO;
1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574
                    case 'a':
                    case 'b':
                    case 'f':
                    case 'n':
                    case 'r':
                    case 't':
                    case 'v':
                      break;

                    case 'x':
                      {
                        /* Hexadecimal escape sequence.  */
                        int i;
                        for (i = 0; i < 2; ++i)
                          {
                            GET (c);

                            if (c >= '0' && c <= '7')
                              ;
                            else if (c >= 'a' && c <= 'f')
                              ;
                            else if (c >= 'A' && c <= 'F')
                              ;
                            else
                              {
                                UNGET ();
                                break;
                              }
                          }
                      }
                      break;

                    case '0':
                      {
                        /* Octal escape sequence.  */
                        int i;
                        for (i = 0; i < 3; ++i)
                          {
                            GET (c);

                            if (c >= '0' && c <= '7')
                              ;
                            else
                              {
                                UNGET ();
                                break;
                              }
                          }
                      }
                      break;

                    default:
                      break;
                    }
                  break;

                case '\n':
                  if (end_char == '\'')
1575
                    yyerror ("newline in character constant", NULL);
1576
                  else
1577
                    yyerror ("newline in string constant", NULL);
1578
                  INCREMENT_LINENO;
1579
                  break;
1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610

                default:
                  break;
                }
            }

        end_string:
          return end_char == '\'' ? CCHAR : CSTRING;

        case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': case 'g':
        case 'h': case 'i': case 'j': case 'k': case 'l': case 'm': case 'n':
        case 'o': case 'p': case 'q': case 'r': case 's': case 't': case 'u':
        case 'v': case 'w': case 'x': case 'y': case 'z':
        case 'A': case 'B': case 'C': case 'D': case 'E': case 'F': case 'G':
        case 'H': case 'I': case 'J': case 'K': case 'L': case 'M': case 'N':
        case 'O': case 'P': case 'Q': case 'R': case 'S': case 'T': case 'U':
        case 'V': case 'W': case 'X': case 'Y': case 'Z': case '_':
          {
            /* Identifier and keywords.  */
            unsigned hash;
            struct kw *k;

            p = yytext;
            *p++ = hash = c;

            while (IDENTP (GET (*p)))
	      {
		hash = (hash << 1) ^ *p++;
		if (p == yytext_end - 1)
		  {
		    int size = yytext_end - yytext;
1611
		    yytext = (char *) xrealloc (yytext, 2 * size);
1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657
		    yytext_end = yytext + 2 * size;
		    p = yytext + size - 1;
		  }
	      }

            UNGET ();
            *p = 0;

            for (k = keyword_table[hash % KEYWORD_TABLE_SIZE]; k; k = k->next)
              if (streq (k->name, yytext))
                return k->tk;

            return IDENT;
          }

        case '/':
          /* C and C++ comments, '/' and '/='.  */
          switch (GET (c))
            {
            case '*':
              while (GET (c))
                {
                  switch (c)
                    {
                    case '*':
                      if (GET (c) == '/')
                        goto comment_end;
                      UNGET ();
                      break;
                    case '\\':
                      GET (c);
                      break;
                    case '\n':
                      INCREMENT_LINENO;
                      break;
                    }
                }
            comment_end:;
              break;

            case '=':
              return DIVASGN;

            case '/':
	      while (GET (c) && c != '\n')
		;
1658 1659 1660 1661 1662
	      /* Don't try to read past the end of the input buffer if
		 the file ends in a C++ comment without a newline.  */
	      if (c == 0)
		return YYEOF;

1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
	      INCREMENT_LINENO;
	      break;

            default:
              UNGET ();
              return '/';
            }
          break;

        case '+':
          if (GET (c) == '+')
            return INC;
          else if (c == '=')
            return ADDASGN;
          UNGET ();
          return '+';

        case '-':
          switch (GET (c))
            {
            case '-':
              return DEC;
            case '>':
              if (GET (c) == '*')
                return ARROWSTAR;
              UNGET ();
              return ARROW;
            case '=':
              return SUBASGN;
            }
          UNGET ();
          return '-';

        case '*':
          if (GET (c) == '=')
            return MULASGN;
          UNGET ();
          return '*';

        case '%':
          if (GET (c) == '=')
            return MODASGN;
          UNGET ();
          return '%';

        case '|':
          if (GET (c) == '|')
            return LOR;
          else if (c == '=')
            return ORASGN;
          UNGET ();
          return '|';

        case '&':
          if (GET (c) == '&')
            return LAND;
          else if (c == '=')
            return ANDASGN;
          UNGET ();
          return '&';

        case '^':
          if (GET (c) == '=')
            return XORASGN;
          UNGET ();
          return '^';

        case '.':
          if (GET (c) == '*')
            return POINTSTAR;
          else if (c == '.')
            {
              if (GET (c) != '.')
1736
                yyerror ("invalid token '..' ('...' assumed)", NULL);
1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876
              UNGET ();
              return ELLIPSIS;
            }
          else if (!DIGITP (c))
            {
              UNGET ();
              return '.';
            }
          goto mantissa;

        case ':':
          if (GET (c) == ':')
            return DCOLON;
          UNGET ();
          return ':';

        case '=':
          if (GET (c) == '=')
            return EQ;
          UNGET ();
          return '=';

        case '!':
          if (GET (c) == '=')
            return NE;
          UNGET ();
          return '!';

        case '<':
          switch (GET (c))
            {
            case '=':
              return LE;
            case '<':
              if (GET (c) == '=')
                return LSHIFTASGN;
              UNGET ();
              return LSHIFT;
            }
          UNGET ();
          return '<';

        case '>':
          switch (GET (c))
            {
            case '=':
              return GE;
            case '>':
              if (GET (c) == '=')
                return RSHIFTASGN;
              UNGET ();
              return RSHIFT;
            }
          UNGET ();
          return '>';

        case '#':
          c = process_pp_line ();
          if (c == 0)
            return YYEOF;
          break;

        case '(': case ')': case '[': case ']': case '{': case '}':
        case ';': case ',': case '?': case '~':
          return c;

        case '0':
          yyival = 0;

          if (GET (c) == 'x' || c == 'X')
            {
              while (GET (c))
                {
                  if (DIGITP (c))
                    yyival = yyival * 16 + c - '0';
                  else if (c >= 'a' && c <= 'f')
                    yyival = yyival * 16 + c - 'a' + 10;
                  else if (c >= 'A' && c <= 'F')
                    yyival = yyival * 16 + c - 'A' + 10;
                  else
                    break;
                }

              goto int_suffixes;
            }
          else if (c == '.')
            goto mantissa;

          while (c >= '0' && c <= '7')
            {
              yyival = (yyival << 3) + c - '0';
              GET (c);
            }

        int_suffixes:
          /* Integer suffixes.  */
          while (isalpha (c))
            GET (c);
          UNGET ();
          return CINT;

        case '1': case '2': case '3': case '4': case '5': case '6':
        case '7': case '8': case '9':
          /* Integer or floating constant, part before '.'.  */
          yyival = c - '0';

          while (GET (c) && DIGITP (c))
            yyival = 10 * yyival + c - '0';

          if (c != '.')
            goto int_suffixes;

        mantissa:
          /* Digits following '.'.  */
          while (DIGITP (c))
            GET (c);

          /* Optional exponent.  */
          if (c == 'E' || c == 'e')
            {
              if (GET (c) == '-' || c == '+')
                GET (c);

              while (DIGITP (c))
                GET (c);
            }

          /* Optional type suffixes.  */
          while (isalpha (c))
            GET (c);
	  UNGET ();
          return CFLOAT;

        default:
          break;
        }
    }
}


1877 1878 1879 1880 1881 1882
/* Actually local to matching_regexp.  These variables must be in
   global scope for the case that `static' get's defined away.  */

static char *matching_regexp_buffer, *matching_regexp_end_buf;


1883 1884 1885 1886
/* Value is the string from the start of the line to the current
   position in the input buffer, or maybe a bit more if that string is
   shorter than min_regexp.  */

1887
static char *
1888
matching_regexp (void)
1889 1890 1891 1892 1893 1894 1895 1896
{
  char *p;
  char *s;
  char *t;

  if (!f_regexps)
    return NULL;

1897
  if (matching_regexp_buffer == NULL)
1898
    {
1899 1900
      matching_regexp_buffer = (char *) xmalloc (max_regexp);
      matching_regexp_end_buf = &matching_regexp_buffer[max_regexp] - 1;
1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911
    }

  /* Scan back to previous newline of buffer start.  */
  for (p = in - 1; p > inbuffer && *p != '\n'; --p)
    ;

  if (*p == '\n')
    {
      while (in - p < min_regexp && p > inbuffer)
        {
          /* Line probably not significant enough */
1912
          for (--p; p > inbuffer && *p != '\n'; --p)
1913 1914 1915 1916 1917 1918 1919 1920 1921
            ;
        }
      if (*p == '\n')
        ++p;
    }

  /* Copy from end to make sure significant portions are included.
     This implies that in the browser a regular expressing of the form
     `^.*{regexp}' has to be used.  */
1922 1923
  for (s = matching_regexp_end_buf - 1, t = in;
       s > matching_regexp_buffer && t > p;)
1924 1925 1926
    {
      *--s = *--t;

1927
      if (*s == '"' || *s == '\\')
1928 1929 1930
        *--s = '\\';
    }

1931
  *(matching_regexp_end_buf - 1) = '\0';
1932 1933 1934 1935 1936 1937
  return xstrdup (s);
}


/* Return a printable representation of token T.  */

1938
static const char *
1939
token_string (int t)
1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014