cp-demangle.c

功能较全面的反汇编器：反汇编器ht-2.0.15.tar.gz

  { "sz", NL ("sizeof "),   1 },
  { NULL, NULL, 0,          0 }
};

static struct demangle_component *
d_operator_name (struct d_info *di)
{
  char c1;
  char c2;

  c1 = d_next_char (di);
  c2 = d_next_char (di);
  if (c1 == 'v' && IS_DIGIT (c2))
    return d_make_extended_operator (di, c2 - '0', d_source_name (di));
  else if (c1 == 'c' && c2 == 'v')
    return d_make_comp (di, DEMANGLE_COMPONENT_CAST,
			cplus_demangle_type (di), NULL);
  else
    {
      /* LOW is the inclusive lower bound.  */
      int low = 0;
      /* HIGH is the exclusive upper bound.  We subtract one to ignore
	 the sentinel at the end of the array.  */
      int high = ((sizeof (cplus_demangle_operators)
		   / sizeof (cplus_demangle_operators[0]))
		  - 1);

      while (1)
	{
	  int i;
	  const struct demangle_operator_info *p;

	  i = low + (high - low) / 2;
	  p = cplus_demangle_operators + i;

	  if (c1 == p->code[0] && c2 == p->code[1])
	    return d_make_operator (di, p);

	  if (c1 < p->code[0] || (c1 == p->code[0] && c2 < p->code[1]))
	    high = i;
	  else
	    low = i + 1;
	  if (low == high)
	    return NULL;
	}
    }
}

static struct demangle_component *
d_make_character (struct d_info *di, int c)
{
  struct demangle_component *p;
  p = d_make_empty (di);
  if (p != NULL)
    {
      p->type = DEMANGLE_COMPONENT_CHARACTER;
      p->u.s_character.character = c;
    }
  return p;
}

static struct demangle_component *
d_java_resource (struct d_info *di)
{
  struct demangle_component *p = NULL;
  struct demangle_component *next = NULL;
  long len, i;
  char c;
  const char *str;

  len = d_number (di);
  if (len <= 1)
    return NULL;

  /* Eat the leading '_'.  */
  if (d_next_char (di) != '_')
    return NULL;
  len--;

  str = d_str (di);
  i = 0;

  while (len > 0)
    {
      c = str[i];
      if (!c)
	return NULL;

      /* Each chunk is either a '$' escape...  */
      if (c == '$')
	{
	  i++;
	  switch (str[i++])
	    {
	    case 'S':
	      c = '/';
	      break;
	    case '_':
	      c = '.';
	      break;
	    case '$':
	      c = '$';
	      break;
	    default:
	      return NULL;
	    }
	  next = d_make_character (di, c);
	  d_advance (di, i);
	  str = d_str (di);
	  len -= i;
	  i = 0;
	  if (next == NULL)
	    return NULL;
	}
      /* ... or a sequence of characters.  */
      else
	{
	  while (i < len && str[i] && str[i] != '$')
	    i++;

	  next = d_make_name (di, str, i);
	  d_advance (di, i);
	  str = d_str (di);
	  len -= i;
	  i = 0;
	  if (next == NULL)
	    return NULL;
	}

      if (p == NULL)
	p = next;
      else
	{
	  p = d_make_comp (di, DEMANGLE_COMPONENT_COMPOUND_NAME, p, next);
	  if (p == NULL)
	    return NULL;
	}
    }

  p = d_make_comp (di, DEMANGLE_COMPONENT_JAVA_RESOURCE, p, NULL);

  return p;
}

/* <special-name> ::= TV <type>
                  ::= TT <type>
                  ::= TI <type>
                  ::= TS <type>
                  ::= GV <(object) name>
                  ::= T <call-offset> <(base) encoding>
                  ::= Tc <call-offset> <call-offset> <(base) encoding>
   Also g++ extensions:
                  ::= TC <type> <(offset) number> _ <(base) type>
                  ::= TF <type>
                  ::= TJ <type>
                  ::= GR <name>
		  ::= GA <encoding>
		  ::= Gr <resource name>
*/

static struct demangle_component *
d_special_name (struct d_info *di)
{
  di->expansion += 20;
  if (d_check_char (di, 'T'))
    {
      switch (d_next_char (di))
	{
	case 'V':
	  di->expansion -= 5;
	  return d_make_comp (di, DEMANGLE_COMPONENT_VTABLE,
			      cplus_demangle_type (di), NULL);
	case 'T':
	  di->expansion -= 10;
	  return d_make_comp (di, DEMANGLE_COMPONENT_VTT,
			      cplus_demangle_type (di), NULL);
	case 'I':
	  return d_make_comp (di, DEMANGLE_COMPONENT_TYPEINFO,
			      cplus_demangle_type (di), NULL);
	case 'S':
	  return d_make_comp (di, DEMANGLE_COMPONENT_TYPEINFO_NAME,
			      cplus_demangle_type (di), NULL);

	case 'h':
	  if (! d_call_offset (di, 'h'))
	    return NULL;
	  return d_make_comp (di, DEMANGLE_COMPONENT_THUNK,
			      d_encoding (di, 0), NULL);

	case 'v':
	  if (! d_call_offset (di, 'v'))
	    return NULL;
	  return d_make_comp (di, DEMANGLE_COMPONENT_VIRTUAL_THUNK,
			      d_encoding (di, 0), NULL);

	case 'c':
	  if (! d_call_offset (di, '\0'))
	    return NULL;
	  if (! d_call_offset (di, '\0'))
	    return NULL;
	  return d_make_comp (di, DEMANGLE_COMPONENT_COVARIANT_THUNK,
			      d_encoding (di, 0), NULL);

	case 'C':
	  {
	    struct demangle_component *derived_type;
	    long offset;
	    struct demangle_component *base_type;

	    derived_type = cplus_demangle_type (di);
	    offset = d_number (di);
	    if (offset < 0)
	      return NULL;
	    if (! d_check_char (di, '_'))
	      return NULL;
	    base_type = cplus_demangle_type (di);
	    /* We don't display the offset.  FIXME: We should display
	       it in verbose mode.  */
	    di->expansion += 5;
	    return d_make_comp (di, DEMANGLE_COMPONENT_CONSTRUCTION_VTABLE,
				base_type, derived_type);
	  }

	case 'F':
	  return d_make_comp (di, DEMANGLE_COMPONENT_TYPEINFO_FN,
			      cplus_demangle_type (di), NULL);
	case 'J':
	  return d_make_comp (di, DEMANGLE_COMPONENT_JAVA_CLASS,
			      cplus_demangle_type (di), NULL);

	default:
	  return NULL;
	}
    }
  else if (d_check_char (di, 'G'))
    {
      switch (d_next_char (di))
	{
	case 'V':
	  return d_make_comp (di, DEMANGLE_COMPONENT_GUARD, d_name (di), NULL);

	case 'R':
	  return d_make_comp (di, DEMANGLE_COMPONENT_REFTEMP, d_name (di),
			      NULL);

	case 'A':
	  return d_make_comp (di, DEMANGLE_COMPONENT_HIDDEN_ALIAS,
			      d_encoding (di, 0), NULL);

	case 'r':
	  return d_java_resource (di);

	default:
	  return NULL;
	}
    }
  else
    return NULL;
}

/* <call-offset> ::= h <nv-offset> _
                 ::= v <v-offset> _

   <nv-offset> ::= <(offset) number>

   <v-offset> ::= <(offset) number> _ <(virtual offset) number>

   The C parameter, if not '\0', is a character we just read which is
   the start of the <call-offset>.

   We don't display the offset information anywhere.  FIXME: We should
   display it in verbose mode.  */

static int
d_call_offset (struct d_info *di, int c)
{
  if (c == '\0')
    c = d_next_char (di);

  if (c == 'h')
    d_number (di);
  else if (c == 'v')
    {
      d_number (di);
      if (! d_check_char (di, '_'))
	return 0;
      d_number (di);
    }
  else
    return 0;

  if (! d_check_char (di, '_'))
    return 0;

  return 1;
}

/* <ctor-dtor-name> ::= C1
                    ::= C2
                    ::= C3
                    ::= D0
                    ::= D1
                    ::= D2
*/

static struct demangle_component *
d_ctor_dtor_name (struct d_info *di)
{
  if (di->last_name != NULL)
    {
      if (di->last_name->type == DEMANGLE_COMPONENT_NAME)
	di->expansion += di->last_name->u.s_name.len;
      else if (di->last_name->type == DEMANGLE_COMPONENT_SUB_STD)
	di->expansion += di->last_name->u.s_string.len;
    }
  switch (d_peek_char (di))
    {
    case 'C':
      {
	enum gnu_v3_ctor_kinds kind;

	switch (d_peek_next_char (di))
	  {
	  case '1':
	    kind = gnu_v3_complete_object_ctor;
	    break;
	  case '2':
	    kind = gnu_v3_base_object_ctor;
	    break;
	  case '3':
	    kind = gnu_v3_complete_object_allocating_ctor;
	    break;
	  default:
	    return NULL;
	  }
	d_advance (di, 2);
	return d_make_ctor (di, kind, di->last_name);
      }

    case 'D':
      {
	enum gnu_v3_dtor_kinds kind;

	switch (d_peek_next_char (di))
	  {
	  case '0':
	    kind = gnu_v3_deleting_dtor;
	    break;
	  case '1':
	    kind = gnu_v3_complete_object_dtor;
	    break;
	  case '2':
	    kind = gnu_v3_base_object_dtor;
	    break;
	  default:
	    return NULL;
	  }
	d_advance (di, 2);
	return d_make_dtor (di, kind, di->last_name);
      }

    default:
      return NULL;
    }
}

/* <type> ::= <builtin-type>
          ::= <function-type>
          ::= <class-enum-type>
          ::= <array-type>
          ::= <pointer-to-member-type>
          ::= <template-param>
          ::= <template-template-param> <template-args>
          ::= <substitution>
          ::= <CV-qualifiers> <type>
          ::= P <type>
          ::= R <type>
          ::= O <type> (C++0x)
          ::= C <type>
          ::= G <type>
          ::= U <source-name> <type>

   <builtin-type> ::= various one letter codes
                  ::= u <source-name>
*/

CP_STATIC_IF_GLIBCPP_V3
const struct demangle_builtin_type_info
cplus_demangle_builtin_types[D_BUILTIN_TYPE_COUNT] =
{
  /* a */ { NL ("signed char"),	NL ("signed char"),	D_PRINT_DEFAULT },
  /* b */ { NL ("bool"),	NL ("boolean"),		D_PRINT_BOOL },
  /* c */ { NL ("char"),	NL ("byte"),		D_PRINT_DEFAULT },
  /* d */ { NL ("double"),	NL ("double"),		D_PRINT_FLOAT },
  /* e */ { NL ("long double"),	NL ("long double"),	D_PRINT_FLOAT },
  /* f */ { NL ("float"),	NL ("float"),		D_PRINT_FLOAT },
  /* g */ { NL ("__float128"),	NL ("__float128"),	D_PRINT_FLOAT },
  /* h */ { NL ("unsigned char"), NL ("unsigned char"),	D_PRINT_DEFAULT },
  /* i */ { NL ("int"),		NL ("int"),		D_PRINT_INT },
  /* j */ { NL ("unsigned int"), NL ("unsigned"),	D_PRINT_UNSIGNED },
  /* k */ { NULL, 0,		NULL, 0,		D_PRINT_DEFAULT },
  /* l */ { NL ("long"),	NL ("long"),		D_PRINT_LONG },
  /* m */ { NL ("unsigned long"), NL ("unsigned long"),	D_PRINT_UNSIGNED_LONG },
  /* n */ { NL ("__int128"),	NL ("__int128"),	D_PRINT_DEFAULT },
  /* o */ { NL ("unsigned __int128"), NL ("unsigned __int128"),
	    D_PRINT_DEFAULT },
  /* p */ { NULL, 0,		NULL, 0,		D_PRINT_DEFAULT },
  /* q */ { NULL, 0,		NULL, 0,		D_PRINT_DEFAULT },
  /* r */ { NULL, 0,		NULL, 0,		D_PRINT_DEFAULT },
  /* s */ { NL ("short"),	NL ("short"),		D_PRINT_DEFAULT },
  /* t */ { NL ("unsigned short"), NL ("unsigned short"), D_PRINT_DEFAULT },
  /* u */ { NULL, 0,		NULL, 0,		D_PRINT_DEFAULT },
  /* v */ { NL ("void"),	NL ("void"),		D_PRINT_VOID },
  /* w */ { NL ("wchar_t"),	NL ("char"),		D_PRINT_DEFAULT },
  /* x */ { NL ("long long"),	NL ("long"),		D_PRINT_LONG_LONG },
  /* y */ { NL ("unsigned long long"), NL ("unsigned long long"),
	    D_PRINT_UNSIGNED_LONG_LONG },
  /* z */ { NL ("..."),		NL ("..."),		D_PRINT_DEFAULT },
};

CP_STATIC_IF_GLIBCPP_V3
struct demangle_component *
cplus_demangle_type (struct d_info *di)
{
  char peek;
  struct demangle_component *ret;
  int can_subst;

  /* The ABI specifies that when CV-qualifiers are used, the base type
     is substitutable, and the fully qualified type is substitutable,
     but the base type with a strict subset of the CV-qualifiers is
     not substitutable.  The natural recursive implementation of the
     CV-qualifiers would cause subsets to be substitutable, so instead
     we pull them all off now.

     FIXME: The ABI says that order-insensitive vendor qualifiers
     should be handled in the same way, but we have no way to tell
     which vendor qualifiers are order-insensitive and which are
     order-sensitive.  So we just assume that they are all
     order-sensitive.  g++ 3.4 supports only one vendor qualifier,
     __vector, and it treats it as order-sensitive when mangling
     names.  */

  peek = d_peek_char (di);
  if (peek == 'r' || peek == 'V' || peek == 'K')
    {
      struct demangle_component **pret;

      pret = d_cv_qualifiers (di, &ret, 0);
      if (pret == NULL)
	return NULL;
      *pret = cplus_demangle_type (di);
      if (! *pret || ! d_add_substitution (di, ret))
	return NULL;
      return ret;
    }

  can_subst = 1;

  switch (peek)
    {
    case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': case 'g':
    case 'h': case 'i': case 'j':           case 'l': case 'm': case 'n':
    case 'o':                               case 's': case 't':
    case 'v': case 'w': case 'x': case 'y': case 'z':
      ret = d_make_builtin_type (di,
				 &cplus_demangle_builtin_types[peek - 'a']);
      di->expansion += ret->u.s_builtin.type->len;
      can_subst = 0;
      d_advance (di, 1);
      break;

    case 'u':
      d_advance (di, 1);
      ret = d_make_comp (di, DEMANGLE_COMPONENT_VENDOR_TYPE,
			 d_source_name (di), NULL);
      break;

    case 'F':
      ret = d_function_type (di);
      break;

    case '0': case '1': case '2': case '3': case '4':
    case '5': case '6': case '7': case '8': case '9':
    case 'N':
    case 'Z':
      ret = d_class_enum_type (di);
      break;