regex.c

它通过提供glibc兼容使得应用程序移植到较小的c 库时相当得容易. 它能够应用到带虚拟存储的Linux和uClinux上.在大多数带MMU部件的平台上为使它更加紧凑,它也能够编译成共享库.uClib

	{
        case no_op:
          printf ("/no_op");
          break;

	case exactn:
	  mcnt = *p++;
          printf ("/exactn/%d", mcnt);
          do
	    {
              putchar ('/');
	      PUT_CHAR (*p++);
            }
          while (--mcnt);
          break;

#  ifdef MBS_SUPPORT
	case exactn_bin:
	  mcnt = *p++;
	  printf ("/exactn_bin/%d", mcnt);
          do
	    {
	      printf("/%lx", (long int) *p++);
            }
          while (--mcnt);
          break;
#  endif /* MBS_SUPPORT */

	case start_memory:
          mcnt = *p++;
          printf ("/start_memory/%d/%ld", mcnt, (long int) *p++);
          break;

	case stop_memory:
          mcnt = *p++;
	  printf ("/stop_memory/%d/%ld", mcnt, (long int) *p++);
          break;

	case duplicate:
	  printf ("/duplicate/%ld", (long int) *p++);
	  break;

	case anychar:
	  printf ("/anychar");
	  break;

	case charset:
        case charset_not:
          {
#  ifdef WCHAR
	    int i, length;
	    wchar_t *workp = p;
	    printf ("/charset [%s",
	            (re_opcode_t) *(workp - 1) == charset_not ? "^" : "");
	    p += 5;
	    length = *workp++; /* the length of char_classes */
	    for (i=0 ; i<length ; i++)
	      printf("[:%lx:]", (long int) *p++);
	    length = *workp++; /* the length of collating_symbol */
	    for (i=0 ; i<length ;)
	      {
		printf("[.");
		while(*p != 0)
		  PUT_CHAR((i++,*p++));
		i++,p++;
		printf(".]");
	      }
	    length = *workp++; /* the length of equivalence_class */
	    for (i=0 ; i<length ;)
	      {
		printf("[=");
		while(*p != 0)
		  PUT_CHAR((i++,*p++));
		i++,p++;
		printf("=]");
	      }
	    length = *workp++; /* the length of char_range */
	    for (i=0 ; i<length ; i++)
	      {
		wchar_t range_start = *p++;
		wchar_t range_end = *p++;
		printf("%C-%C", range_start, range_end);
	      }
	    length = *workp++; /* the length of char */
	    for (i=0 ; i<length ; i++)
	      printf("%C", *p++);
	    putchar (']');
#  else
            register int c, last = -100;
	    register int in_range = 0;

	    printf ("/charset [%s",
	            (re_opcode_t) *(p - 1) == charset_not ? "^" : "");

            assert (p + *p < pend);

            for (c = 0; c < 256; c++)
	      if (c / 8 < *p
		  && (p[1 + (c/8)] & (1 << (c % 8))))
		{
		  /* Are we starting a range?  */
		  if (last + 1 == c && ! in_range)
		    {
		      putchar ('-');
		      in_range = 1;
		    }
		  /* Have we broken a range?  */
		  else if (last + 1 != c && in_range)
              {
		      putchar (last);
		      in_range = 0;
		    }

		  if (! in_range)
		    putchar (c);

		  last = c;
              }

	    if (in_range)
	      putchar (last);

	    putchar (']');

	    p += 1 + *p;
#  endif /* WCHAR */
	  }
	  break;

	case begline:
	  printf ("/begline");
          break;

	case endline:
          printf ("/endline");
          break;

	case on_failure_jump:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/on_failure_jump to %td", p + mcnt - start);
#  else
  	  printf ("/on_failure_jump to %ld", (long int) (p + mcnt - start));
#  endif
          break;

	case on_failure_keep_string_jump:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/on_failure_keep_string_jump to %td", p + mcnt - start);
#  else
  	  printf ("/on_failure_keep_string_jump to %ld",
		  (long int) (p + mcnt - start));
#  endif
          break;

	case dummy_failure_jump:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/dummy_failure_jump to %td", p + mcnt - start);
#  else
  	  printf ("/dummy_failure_jump to %ld", (long int) (p + mcnt - start));
#  endif
          break;

	case push_dummy_failure:
          printf ("/push_dummy_failure");
          break;

        case maybe_pop_jump:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/maybe_pop_jump to %td", p + mcnt - start);
#  else
  	  printf ("/maybe_pop_jump to %ld", (long int) (p + mcnt - start));
#  endif
	  break;

        case pop_failure_jump:
	  PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/pop_failure_jump to %td", p + mcnt - start);
#  else
  	  printf ("/pop_failure_jump to %ld", (long int) (p + mcnt - start));
#  endif
	  break;

        case jump_past_alt:
	  PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/jump_past_alt to %td", p + mcnt - start);
#  else
  	  printf ("/jump_past_alt to %ld", (long int) (p + mcnt - start));
#  endif
	  break;

        case jump:
	  PREFIX(extract_number_and_incr) (&mcnt, &p);
#  ifdef _LIBC
  	  printf ("/jump to %td", p + mcnt - start);
#  else
  	  printf ("/jump to %ld", (long int) (p + mcnt - start));
#  endif
	  break;

        case succeed_n:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
	  p1 = p + mcnt;
          PREFIX(extract_number_and_incr) (&mcnt2, &p);
#  ifdef _LIBC
	  printf ("/succeed_n to %td, %d times", p1 - start, mcnt2);
#  else
	  printf ("/succeed_n to %ld, %d times",
		  (long int) (p1 - start), mcnt2);
#  endif
          break;

        case jump_n:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
	  p1 = p + mcnt;
          PREFIX(extract_number_and_incr) (&mcnt2, &p);
	  printf ("/jump_n to %d, %d times", p1 - start, mcnt2);
          break;

        case set_number_at:
          PREFIX(extract_number_and_incr) (&mcnt, &p);
	  p1 = p + mcnt;
          PREFIX(extract_number_and_incr) (&mcnt2, &p);
#  ifdef _LIBC
	  printf ("/set_number_at location %td to %d", p1 - start, mcnt2);
#  else
	  printf ("/set_number_at location %ld to %d",
		  (long int) (p1 - start), mcnt2);
#  endif
          break;

        case wordbound:
	  printf ("/wordbound");
	  break;

	case notwordbound:
	  printf ("/notwordbound");
          break;

	case wordbeg:
	  printf ("/wordbeg");
	  break;

	case wordend:
	  printf ("/wordend");
	  break;

#  ifdef emacs
	case before_dot:
	  printf ("/before_dot");
          break;

	case at_dot:
	  printf ("/at_dot");
          break;

	case after_dot:
	  printf ("/after_dot");
          break;

	case syntaxspec:
          printf ("/syntaxspec");
	  mcnt = *p++;
	  printf ("/%d", mcnt);
          break;

	case notsyntaxspec:
          printf ("/notsyntaxspec");
	  mcnt = *p++;
	  printf ("/%d", mcnt);
	  break;
#  endif /* emacs */

	case wordchar:
	  printf ("/wordchar");
          break;

	case notwordchar:
	  printf ("/notwordchar");
          break;

	case begbuf:
	  printf ("/begbuf");
          break;

	case endbuf:
	  printf ("/endbuf");
          break;

        default:
          printf ("?%ld", (long int) *(p-1));
	}

      putchar ('\n');
    }

#  ifdef _LIBC
  printf ("%td:\tend of pattern.\n", p - start);
#  else
  printf ("%ld:\tend of pattern.\n", (long int) (p - start));
#  endif
}


void
PREFIX(print_compiled_pattern) (bufp)
    struct re_pattern_buffer *bufp;
{
  UCHAR_T *buffer = (UCHAR_T*) bufp->buffer;

  PREFIX(print_partial_compiled_pattern) (buffer, buffer
				  + bufp->used / sizeof(UCHAR_T));
  printf ("%ld bytes used/%ld bytes allocated.\n",
	  bufp->used, bufp->allocated);

  if (bufp->fastmap_accurate && bufp->fastmap)
    {
      printf ("fastmap: ");
      print_fastmap (bufp->fastmap);
    }

#  ifdef _LIBC
  printf ("re_nsub: %Zd\t", bufp->re_nsub);
#  else
  printf ("re_nsub: %ld\t", (long int) bufp->re_nsub);
#  endif
  printf ("regs_alloc: %d\t", bufp->regs_allocated);
  printf ("can_be_null: %d\t", bufp->can_be_null);
  printf ("newline_anchor: %d\n", bufp->newline_anchor);
  printf ("no_sub: %d\t", bufp->no_sub);
  printf ("not_bol: %d\t", bufp->not_bol);
  printf ("not_eol: %d\t", bufp->not_eol);
  printf ("syntax: %lx\n", bufp->syntax);
  /* Perhaps we should print the translate table?  */
}


void
PREFIX(print_double_string) (where, string1, size1, string2, size2)
    const CHAR_T *where;
    const CHAR_T *string1;
    const CHAR_T *string2;
    int size1;
    int size2;
{
  int this_char;

  if (where == NULL)
    printf ("(null)");
  else
    {
      int cnt;

      if (FIRST_STRING_P (where))
        {
          for (this_char = where - string1; this_char < size1; this_char++)
	    PUT_CHAR (string1[this_char]);

          where = string2;
        }

      cnt = 0;
      for (this_char = where - string2; this_char < size2; this_char++)
	{
	  PUT_CHAR (string2[this_char]);
	  if (++cnt > 100)
	    {
	      fputs ("...", stdout);
	      break;
	    }
	}
    }
}

#  ifndef DEFINED_ONCE
void
printchar (c)
     int c;
{
  putc (c, stderr);
}
#  endif

# else /* not DEBUG */

#  ifndef DEFINED_ONCE
#   undef assert
#   define assert(e)

#   define DEBUG_STATEMENT(e)
#   define DEBUG_PRINT1(x)
#   define DEBUG_PRINT2(x1, x2)
#   define DEBUG_PRINT3(x1, x2, x3)
#   define DEBUG_PRINT4(x1, x2, x3, x4)
#  endif /* not DEFINED_ONCE */
#  define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
#  define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)

# endif /* not DEBUG */



# ifdef WCHAR
/* This  convert a multibyte string to a wide character string.
   And write their correspondances to offset_buffer(see below)
   and write whether each wchar_t is binary data to is_binary.
   This assume invalid multibyte sequences as binary data.
   We assume offset_buffer and is_binary is already allocated
   enough space.  */

static size_t convert_mbs_to_wcs (CHAR_T *dest, const unsigned char* src,
				  size_t len, int *offset_buffer,
				  char *is_binary);
static size_t
convert_mbs_to_wcs (dest, src, len, offset_buffer, is_binary)
     CHAR_T *dest;
     const unsigned char* src;