regex_internal.h

gnu tar 源码包。 tar 软件是 Unix 系统下的一个打包软件

#endif

static reg_errcode_t re_string_realloc_buffers (re_string_t *pstr,
						Idx new_buf_len)
     internal_function;
#ifdef RE_ENABLE_I18N
static void build_wcs_buffer (re_string_t *pstr) internal_function;
static reg_errcode_t build_wcs_upper_buffer (re_string_t *pstr)
     internal_function;
#endif /* RE_ENABLE_I18N */
static void build_upper_buffer (re_string_t *pstr) internal_function;
static void re_string_translate_buffer (re_string_t *pstr) internal_function;
static unsigned int re_string_context_at (const re_string_t *input, Idx idx,
					  int eflags)
     internal_function __attribute ((pure));
#define re_string_peek_byte(pstr, offset) \
  ((pstr)->mbs[(pstr)->cur_idx + offset])
#define re_string_fetch_byte(pstr) \
  ((pstr)->mbs[(pstr)->cur_idx++])
#define re_string_first_byte(pstr, idx) \
  ((idx) == (pstr)->valid_len || (pstr)->wcs[idx] != WEOF)
#define re_string_is_single_byte_char(pstr, idx) \
  ((pstr)->wcs[idx] != WEOF && ((pstr)->valid_len == (idx) + 1 \
				|| (pstr)->wcs[(idx) + 1] != WEOF))
#define re_string_eoi(pstr) ((pstr)->stop <= (pstr)->cur_idx)
#define re_string_cur_idx(pstr) ((pstr)->cur_idx)
#define re_string_get_buffer(pstr) ((pstr)->mbs)
#define re_string_length(pstr) ((pstr)->len)
#define re_string_byte_at(pstr,idx) ((pstr)->mbs[idx])
#define re_string_skip_bytes(pstr,idx) ((pstr)->cur_idx += (idx))
#define re_string_set_index(pstr,idx) ((pstr)->cur_idx = (idx))

#include <alloca.h>

#ifndef _LIBC
# if HAVE_ALLOCA
/* The OS usually guarantees only one guard page at the bottom of the stack,
   and a page size can be as small as 4096 bytes.  So we cannot safely
   allocate anything larger than 4096 bytes.  Also care for the possibility
   of a few compiler-allocated temporary stack slots.  */
#  define __libc_use_alloca(n) ((n) < 4032)
# else
/* alloca is implemented with malloc, so just use malloc.  */
#  define __libc_use_alloca(n) 0
# endif
#endif

#ifndef MAX
# define MAX(a,b) ((a) < (b) ? (b) : (a))
#endif

#define re_malloc(t,n) ((t *) malloc ((n) * sizeof (t)))
#define re_realloc(p,t,n) ((t *) realloc (p, (n) * sizeof (t)))
#define re_free(p) free (p)

struct bin_tree_t
{
  struct bin_tree_t *parent;
  struct bin_tree_t *left;
  struct bin_tree_t *right;
  struct bin_tree_t *first;
  struct bin_tree_t *next;

  re_token_t token;

  /* `node_idx' is the index in dfa->nodes, if `type' == 0.
     Otherwise `type' indicate the type of this node.  */
  Idx node_idx;
};
typedef struct bin_tree_t bin_tree_t;

#define BIN_TREE_STORAGE_SIZE \
  ((1024 - sizeof (void *)) / sizeof (bin_tree_t))

struct bin_tree_storage_t
{
  struct bin_tree_storage_t *next;
  bin_tree_t data[BIN_TREE_STORAGE_SIZE];
};
typedef struct bin_tree_storage_t bin_tree_storage_t;

#define CONTEXT_WORD 1
#define CONTEXT_NEWLINE (CONTEXT_WORD << 1)
#define CONTEXT_BEGBUF (CONTEXT_NEWLINE << 1)
#define CONTEXT_ENDBUF (CONTEXT_BEGBUF << 1)

#define IS_WORD_CONTEXT(c) ((c) & CONTEXT_WORD)
#define IS_NEWLINE_CONTEXT(c) ((c) & CONTEXT_NEWLINE)
#define IS_BEGBUF_CONTEXT(c) ((c) & CONTEXT_BEGBUF)
#define IS_ENDBUF_CONTEXT(c) ((c) & CONTEXT_ENDBUF)
#define IS_ORDINARY_CONTEXT(c) ((c) == 0)

#define IS_WORD_CHAR(ch) (isalnum (ch) || (ch) == '_')
#define IS_NEWLINE(ch) ((ch) == NEWLINE_CHAR)
#define IS_WIDE_WORD_CHAR(ch) (iswalnum (ch) || (ch) == L'_')
#define IS_WIDE_NEWLINE(ch) ((ch) == WIDE_NEWLINE_CHAR)

#define NOT_SATISFY_PREV_CONSTRAINT(constraint,context) \
 ((((constraint) & PREV_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
  || ((constraint & PREV_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
  || ((constraint & PREV_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context))\
  || ((constraint & PREV_BEGBUF_CONSTRAINT) && !IS_BEGBUF_CONTEXT (context)))

#define NOT_SATISFY_NEXT_CONSTRAINT(constraint,context) \
 ((((constraint) & NEXT_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
  || (((constraint) & NEXT_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
  || (((constraint) & NEXT_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context)) \
  || (((constraint) & NEXT_ENDBUF_CONSTRAINT) && !IS_ENDBUF_CONTEXT (context)))

struct re_dfastate_t
{
  re_hashval_t hash;
  re_node_set nodes;
  re_node_set non_eps_nodes;
  re_node_set inveclosure;
  re_node_set *entrance_nodes;
  struct re_dfastate_t **trtable, **word_trtable;
  unsigned int context : 4;
  unsigned int halt : 1;
  /* If this state can accept `multi byte'.
     Note that we refer to multibyte characters, and multi character
     collating elements as `multi byte'.  */
  unsigned int accept_mb : 1;
  /* If this state has backreference node(s).  */
  unsigned int has_backref : 1;
  unsigned int has_constraint : 1;
};
typedef struct re_dfastate_t re_dfastate_t;

struct re_state_table_entry
{
  Idx num;
  Idx alloc;
  re_dfastate_t **array;
};

/* Array type used in re_sub_match_last_t and re_sub_match_top_t.  */

typedef struct
{
  Idx next_idx;
  Idx alloc;
  re_dfastate_t **array;
} state_array_t;

/* Store information about the node NODE whose type is OP_CLOSE_SUBEXP.  */

typedef struct
{
  Idx node;
  Idx str_idx; /* The position NODE match at.  */
  state_array_t path;
} re_sub_match_last_t;

/* Store information about the node NODE whose type is OP_OPEN_SUBEXP.
   And information about the node, whose type is OP_CLOSE_SUBEXP,
   corresponding to NODE is stored in LASTS.  */

typedef struct
{
  Idx str_idx;
  Idx node;
  state_array_t *path;
  Idx alasts; /* Allocation size of LASTS.  */
  Idx nlasts; /* The number of LASTS.  */
  re_sub_match_last_t **lasts;
} re_sub_match_top_t;

struct re_backref_cache_entry
{
  Idx node;
  Idx str_idx;
  Idx subexp_from;
  Idx subexp_to;
  char more;
  char unused;
  unsigned short int eps_reachable_subexps_map;
};

typedef struct
{
  /* The string object corresponding to the input string.  */
  re_string_t input;
#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)
  const re_dfa_t *const dfa;
#else
  const re_dfa_t *dfa;
#endif
  /* EFLAGS of the argument of regexec.  */
  int eflags;
  /* Where the matching ends.  */
  Idx match_last;
  Idx last_node;
  /* The state log used by the matcher.  */
  re_dfastate_t **state_log;
  Idx state_log_top;
  /* Back reference cache.  */
  Idx nbkref_ents;
  Idx abkref_ents;
  struct re_backref_cache_entry *bkref_ents;
  int max_mb_elem_len;
  Idx nsub_tops;
  Idx asub_tops;
  re_sub_match_top_t **sub_tops;
} re_match_context_t;

typedef struct
{
  re_dfastate_t **sifted_states;
  re_dfastate_t **limited_states;
  Idx last_node;
  Idx last_str_idx;
  re_node_set limits;
} re_sift_context_t;

struct re_fail_stack_ent_t
{
  Idx idx;
  Idx node;
  regmatch_t *regs;
  re_node_set eps_via_nodes;
};

struct re_fail_stack_t
{
  Idx num;
  Idx alloc;
  struct re_fail_stack_ent_t *stack;
};

struct re_dfa_t
{
  re_token_t *nodes;
  size_t nodes_alloc;
  size_t nodes_len;
  Idx *nexts;
  Idx *org_indices;
  re_node_set *edests;
  re_node_set *eclosures;
  re_node_set *inveclosures;
  struct re_state_table_entry *state_table;
  re_dfastate_t *init_state;
  re_dfastate_t *init_state_word;
  re_dfastate_t *init_state_nl;
  re_dfastate_t *init_state_begbuf;
  bin_tree_t *str_tree;
  bin_tree_storage_t *str_tree_storage;
  re_bitset_ptr_t sb_char;
  int str_tree_storage_idx;

  /* number of subexpressions `re_nsub' is in regex_t.  */
  re_hashval_t state_hash_mask;
  Idx init_node;
  Idx nbackref; /* The number of backreference in this dfa.  */

  /* Bitmap expressing which backreference is used.  */
  bitset_word_t used_bkref_map;
  bitset_word_t completed_bkref_map;

  unsigned int has_plural_match : 1;
  /* If this dfa has "multibyte node", which is a backreference or
     a node which can accept multibyte character or multi character
     collating element.  */
  unsigned int has_mb_node : 1;
  unsigned int is_utf8 : 1;
  unsigned int map_notascii : 1;
  unsigned int word_ops_used : 1;
  int mb_cur_max;
  bitset_t word_char;
  reg_syntax_t syntax;
  Idx *subexp_map;
#ifdef DEBUG
  char* re_str;
#endif
#ifdef _LIBC
  __libc_lock_define (, lock)
#endif
};

#define re_node_set_init_empty(set) memset (set, '\0', sizeof (re_node_set))
#define re_node_set_remove(set,id) \
  (re_node_set_remove_at (set, re_node_set_contains (set, id) - 1))
#define re_node_set_empty(p) ((p)->nelem = 0)
#define re_node_set_free(set) re_free ((set)->elems)


typedef enum
{
  SB_CHAR,
  MB_CHAR,
  EQUIV_CLASS,
  COLL_SYM,
  CHAR_CLASS
} bracket_elem_type;

typedef struct
{
  bracket_elem_type type;
  union
  {
    unsigned char ch;
    unsigned char *name;
    wchar_t wch;
  } opr;
} bracket_elem_t;


/* Inline functions for bitset_t operation.  */

static inline void
bitset_set (bitset_t set, Idx i)
{
  set[i / BITSET_WORD_BITS] |= (bitset_word_t) 1 << i % BITSET_WORD_BITS;
}

static inline void
bitset_clear (bitset_t set, Idx i)
{
  set[i / BITSET_WORD_BITS] &= ~ ((bitset_word_t) 1 << i % BITSET_WORD_BITS);
}

static inline bool
bitset_contain (const bitset_t set, Idx i)
{
  return (set[i / BITSET_WORD_BITS] >> i % BITSET_WORD_BITS) & 1;
}

static inline void
bitset_empty (bitset_t set)
{
  memset (set, '\0', sizeof (bitset_t));
}

static inline void
bitset_set_all (bitset_t set)
{
  memset (set, -1, sizeof (bitset_word_t) * (SBC_MAX / BITSET_WORD_BITS));
  if (SBC_MAX % BITSET_WORD_BITS != 0)
    set[BITSET_WORDS - 1] =
      ((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1;
}

static inline void
bitset_copy (bitset_t dest, const bitset_t src)
{
  memcpy (dest, src, sizeof (bitset_t));
}

static inline void
bitset_not (bitset_t set)
{
  int bitset_i;
  for (bitset_i = 0; bitset_i < SBC_MAX / BITSET_WORD_BITS; ++bitset_i)
    set[bitset_i] = ~set[bitset_i];
  if (SBC_MAX % BITSET_WORD_BITS != 0)
    set[BITSET_WORDS - 1] =
      ((((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1)
       & ~set[BITSET_WORDS - 1]);
}

static inline void
bitset_merge (bitset_t dest, const bitset_t src)
{
  int bitset_i;
  for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
    dest[bitset_i] |= src[bitset_i];
}

static inline void
bitset_mask (bitset_t dest, const bitset_t src)
{
  int bitset_i;
  for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
    dest[bitset_i] &= src[bitset_i];
}

#ifdef RE_ENABLE_I18N
/* Inline functions for re_string.  */
static inline int
internal_function __attribute ((pure))
re_string_char_size_at (const re_string_t *pstr, Idx idx)
{
  int byte_idx;
  if (pstr->mb_cur_max == 1)
    return 1;
  for (byte_idx = 1; idx + byte_idx < pstr->valid_len; ++byte_idx)
    if (pstr->wcs[idx + byte_idx] != WEOF)
      break;
  return byte_idx;
}

static inline wint_t
internal_function __attribute ((pure))
re_string_wchar_at (const re_string_t *pstr, Idx idx)
{
  if (pstr->mb_cur_max == 1)
    return (wint_t) pstr->mbs[idx];
  return (wint_t) pstr->wcs[idx];
}

static int
internal_function __attribute ((pure))
re_string_elem_size_at (const re_string_t *pstr, Idx idx)
{
# ifdef _LIBC
  const unsigned char *p, *extra;
  const int32_t *table, *indirect;
  int32_t tmp;
#  include <locale/weight.h>
  uint_fast32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);

  if (nrules != 0)
    {
      table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
      extra = (const unsigned char *)
	_NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB);
      indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE,
						_NL_COLLATE_INDIRECTMB);
      p = pstr->mbs + idx;
      tmp = findidx (&p);
      return p - pstr->mbs - idx;
    }
  else
# endif /* _LIBC */
    return 1;
}
#endif /* RE_ENABLE_I18N */

#endif /*  _REGEX_INTERNAL_H */