pcre_compile.c

Scheme跨平台编译器

    break;

    /* Handle single-char matchers */

    case OP_PROP:
    case OP_NOTPROP:
    cc += 2;
    /* Fall through */

    case OP_NOT_DIGIT:
    case OP_DIGIT:
    case OP_NOT_WHITESPACE:
    case OP_WHITESPACE:
    case OP_NOT_WORDCHAR:
    case OP_WORDCHAR:
    case OP_ANY:
    case OP_ALLANY:
    branchlength++;
    cc++;
    break;

    /* The single-byte matcher isn't allowed */

    case OP_ANYBYTE:
    return -2;

    /* Check a class for variable quantification */

#ifdef SUPPORT_UTF8
    case OP_XCLASS:
    cc += GET(cc, 1) - 33;
    /* Fall through */
#endif

    case OP_CLASS:
    case OP_NCLASS:
    cc += 33;

    switch (*cc)
      {
      case OP_CRSTAR:
      case OP_CRMINSTAR:
      case OP_CRQUERY:
      case OP_CRMINQUERY:
      return -1;

      case OP_CRRANGE:
      case OP_CRMINRANGE:
      if (GET2(cc,1) != GET2(cc,3)) return -1;
      branchlength += GET2(cc,1);
      cc += 5;
      break;

      default:
      branchlength++;
      }
    break;

    /* Anything else is variable length */

    default:
    return -1;
    }
  }
/* Control never gets here */
}




/*************************************************
*    Scan compiled regex for numbered bracket    *
*************************************************/

/* This little function scans through a compiled pattern until it finds a
capturing bracket with the given number.

Arguments:
  code        points to start of expression
  utf8        TRUE in UTF-8 mode
  number      the required bracket number

Returns:      pointer to the opcode for the bracket, or NULL if not found
*/

static const uschar *
find_bracket(const uschar *code, BOOL utf8, int number)
{
for (;;)
  {
  register int c = *code;
  if (c == OP_END) return NULL;

  /* XCLASS is used for classes that cannot be represented just by a bit
  map. This includes negated single high-valued characters. The length in
  the table is zero; the actual length is stored in the compiled code. */

  if (c == OP_XCLASS) code += GET(code, 1);

  /* Handle capturing bracket */

  else if (c == OP_CBRA)
    {
    int n = GET2(code, 1+LINK_SIZE);
    if (n == number) return (uschar *)code;
    code += _pcre_OP_lengths[c];
    }

  /* Otherwise, we can get the item's length from the table, except that for
  repeated character types, we have to test for \p and \P, which have an extra
  two bytes of parameters. */

  else
    {
    switch(c)
      {
      case OP_TYPESTAR:
      case OP_TYPEMINSTAR:
      case OP_TYPEPLUS:
      case OP_TYPEMINPLUS:
      case OP_TYPEQUERY:
      case OP_TYPEMINQUERY:
      case OP_TYPEPOSSTAR:
      case OP_TYPEPOSPLUS:
      case OP_TYPEPOSQUERY:
      if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
      break;

      case OP_TYPEUPTO:
      case OP_TYPEMINUPTO:
      case OP_TYPEEXACT:
      case OP_TYPEPOSUPTO:
      if (code[3] == OP_PROP || code[3] == OP_NOTPROP) code += 2;
      break;
      }

    /* Add in the fixed length from the table */

    code += _pcre_OP_lengths[c];

  /* In UTF-8 mode, opcodes that are followed by a character may be followed by
  a multi-byte character. The length in the table is a minimum, so we have to
  arrange to skip the extra bytes. */

#ifdef SUPPORT_UTF8
    if (utf8) switch(c)
      {
      case OP_CHAR:
      case OP_CHARNC:
      case OP_EXACT:
      case OP_UPTO:
      case OP_MINUPTO:
      case OP_POSUPTO:
      case OP_STAR:
      case OP_MINSTAR:
      case OP_POSSTAR:
      case OP_PLUS:
      case OP_MINPLUS:
      case OP_POSPLUS:
      case OP_QUERY:
      case OP_MINQUERY:
      case OP_POSQUERY:
      if (code[-1] >= 0xc0) code += _pcre_utf8_table4[code[-1] & 0x3f];
      break;
      }
#endif
    }
  }
}



/*************************************************
*   Scan compiled regex for recursion reference  *
*************************************************/

/* This little function scans through a compiled pattern until it finds an
instance of OP_RECURSE.

Arguments:
  code        points to start of expression
  utf8        TRUE in UTF-8 mode

Returns:      pointer to the opcode for OP_RECURSE, or NULL if not found
*/

static const uschar *
find_recurse(const uschar *code, BOOL utf8)
{
for (;;)
  {
  register int c = *code;
  if (c == OP_END) return NULL;
  if (c == OP_RECURSE) return code;

  /* XCLASS is used for classes that cannot be represented just by a bit
  map. This includes negated single high-valued characters. The length in
  the table is zero; the actual length is stored in the compiled code. */

  if (c == OP_XCLASS) code += GET(code, 1);

  /* Otherwise, we can get the item's length from the table, except that for
  repeated character types, we have to test for \p and \P, which have an extra
  two bytes of parameters. */

  else
    {
    switch(c)
      {
      case OP_TYPESTAR:
      case OP_TYPEMINSTAR:
      case OP_TYPEPLUS:
      case OP_TYPEMINPLUS:
      case OP_TYPEQUERY:
      case OP_TYPEMINQUERY:
      case OP_TYPEPOSSTAR:
      case OP_TYPEPOSPLUS:
      case OP_TYPEPOSQUERY:
      if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
      break;

      case OP_TYPEPOSUPTO:
      case OP_TYPEUPTO:
      case OP_TYPEMINUPTO:
      case OP_TYPEEXACT:
      if (code[3] == OP_PROP || code[3] == OP_NOTPROP) code += 2;
      break;
      }

    /* Add in the fixed length from the table */

    code += _pcre_OP_lengths[c];

    /* In UTF-8 mode, opcodes that are followed by a character may be followed
    by a multi-byte character. The length in the table is a minimum, so we have
    to arrange to skip the extra bytes. */

#ifdef SUPPORT_UTF8
    if (utf8) switch(c)
      {
      case OP_CHAR:
      case OP_CHARNC:
      case OP_EXACT:
      case OP_UPTO:
      case OP_MINUPTO:
      case OP_POSUPTO:
      case OP_STAR:
      case OP_MINSTAR:
      case OP_POSSTAR:
      case OP_PLUS:
      case OP_MINPLUS:
      case OP_POSPLUS:
      case OP_QUERY:
      case OP_MINQUERY:
      case OP_POSQUERY:
      if (code[-1] >= 0xc0) code += _pcre_utf8_table4[code[-1] & 0x3f];
      break;
      }
#endif
    }
  }
}



/*************************************************
*    Scan compiled branch for non-emptiness      *
*************************************************/

/* This function scans through a branch of a compiled pattern to see whether it
can match the empty string or not. It is called from could_be_empty()
below and from compile_branch() when checking for an unlimited repeat of a
group that can match nothing. Note that first_significant_code() skips over
backward and negative forward assertions when its final argument is TRUE. If we
hit an unclosed bracket, we return "empty" - this means we've struck an inner
bracket whose current branch will already have been scanned.

Arguments:
  code        points to start of search
  endcode     points to where to stop
  utf8        TRUE if in UTF8 mode

Returns:      TRUE if what is matched could be empty
*/

static BOOL
could_be_empty_branch(const uschar *code, const uschar *endcode, BOOL utf8)
{
register int c;
for (code = first_significant_code(code + _pcre_OP_lengths[*code], NULL, 0, TRUE);
     code < endcode;
     code = first_significant_code(code + _pcre_OP_lengths[c], NULL, 0, TRUE))
  {
  const uschar *ccode;

  c = *code;

  /* Skip over forward assertions; the other assertions are skipped by
  first_significant_code() with a TRUE final argument. */

  if (c == OP_ASSERT)
    {
    do code += GET(code, 1); while (*code == OP_ALT);
    c = *code;
    continue;
    }

  /* Groups with zero repeats can of course be empty; skip them. */

  if (c == OP_BRAZERO || c == OP_BRAMINZERO || c == OP_SKIPZERO)
    {
    code += _pcre_OP_lengths[c];
    do code += GET(code, 1); while (*code == OP_ALT);
    c = *code;
    continue;
    }

  /* For other groups, scan the branches. */

  if (c == OP_BRA || c == OP_CBRA || c == OP_ONCE || c == OP_COND)
    {
    BOOL empty_branch;
    if (GET(code, 1) == 0) return TRUE;    /* Hit unclosed bracket */

    /* Scan a closed bracket */

    empty_branch = FALSE;
    do
      {
      if (!empty_branch && could_be_empty_branch(code, endcode, utf8))
        empty_branch = TRUE;
      code += GET(code, 1);
      }
    while (*code == OP_ALT);
    if (!empty_branch) return FALSE;   /* All branches are non-empty */
    c = *code;
    continue;
    }

  /* Handle the other opcodes */

  switch (c)
    {
    /* Check for quantifiers after a class. XCLASS is used for classes that
    cannot be represented just by a bit map. This includes negated single
    high-valued characters. The length in _pcre_OP_lengths[] is zero; the
    actual length is stored in the compiled code, so we must update "code"
    here. */

#ifdef SUPPORT_UTF8
    case OP_XCLASS:
    ccode = code += GET(code, 1);
    goto CHECK_CLASS_REPEAT;
#endif

    case OP_CLASS:
    case OP_NCLASS:
    ccode = code + 33;

#ifdef SUPPORT_UTF8
    CHECK_CLASS_REPEAT:
#endif

    switch (*ccode)
      {
      case OP_CRSTAR:            /* These could be empty; continue */
      case OP_CRMINSTAR:
      case OP_CRQUERY:
      case OP_CRMINQUERY:
      break;

      default:                   /* Non-repeat => class must match */
      case OP_CRPLUS:            /* These repeats aren't empty */
      case OP_CRMINPLUS:
      return FALSE;

      case OP_CRRANGE:
      case OP_CRMINRANGE:
      if (GET2(ccode, 1) > 0) return FALSE;  /* Minimum > 0 */
      break;
      }
    break;

    /* Opcodes that must match a character */

    case OP_PROP:
    case OP_NOTPROP:
    case OP_EXTUNI:
    case OP_NOT_DIGIT:
    case OP_DIGIT:
    case OP_NOT_WHITESPACE:
    case OP_WHITESPACE:
    case OP_NOT_WORDCHAR:
    case OP_WORDCHAR:
    case OP_ANY:
    case OP_ALLANY:
    case OP_ANYBYTE:
    case OP_CHAR:
    case OP_CHARNC:
    case OP_NOT:
    case OP_PLUS:
    case OP_MINPLUS:
    case OP_POSPLUS:
    case OP_EXACT:
    case OP_NOTPLUS:
    case OP_NOTMINPLUS:
    case OP_NOTPOSPLUS:
    case OP_NOTEXACT:
    case OP_TYPEPLUS:
    case OP_TYPEMINPLUS:
    case OP_TYPEPOSPLUS:
    case OP_TYPEEXACT:
    return FALSE;

    /* These are going to continue, as they may be empty, but we have to
    fudge the length for the \p and \P cases. */

    case OP_TYPESTAR:
    case OP_TYPEMINSTAR:
    case OP_TYPEPOSSTAR:
    case OP_TYPEQUERY:
    case OP_TYPEMINQUERY:
    case OP_TYPEPOSQUERY:
    if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
    break;

    /* Same for these */

    case OP_TYPEUPTO: