vfscanf.c

绝对正真的stdio.h的实现

			{
			  /* Add all the characters.  */
			  for (cmpp = thousands; *cmpp != '\0'; ++cmpp)
			    ADDW ((unsigned char) *cmpp);
			  if (width > 0)
			    width = avail;
			}
		      else
			{
			  /* The last read character is not part of the number
			     anymore.  */
			  ungetc (c, s);
			  break;
			}
		    }
#endif
		}
	      if (width > 0)
		--width;
	    }
	  while (width != 0 && inchar () != EOF);

	  /* Have we read any character?  If we try to read a number
	     in hexadecimal notation and we have read only the `0x'
	     prefix or no exponent this is an error.  */
	  if (wpsize == 0 || (is_hexa && (wpsize == 2 || ! got_e)))
	    conv_error ();

	scan_float:
	  /* Convert the number.  */
	  ADDW (L_('\0'));
	  if (flags & LONGDBL)
	    {
	      long double d = __strtold_internal (wp, &tw, flags & GROUP);
	      if (!(flags & SUPPRESS) && tw != wp)
		*ARG (long double *) = negative ? -d : d;
	    }
	  else if (flags & LONG)
	    {
	      double d = __strtod_internal (wp, &tw, flags & GROUP);
	      if (!(flags & SUPPRESS) && tw != wp)
		*ARG (double *) = negative ? -d : d;
	    }
	  else
	    {
	      float d = __strtof_internal (wp, &tw, flags & GROUP);
	      if (!(flags & SUPPRESS) && tw != wp)
		*ARG (float *) = negative ? -d : d;
	    }

	  if (tw == wp)
	    conv_error ();

	  if (!(flags & SUPPRESS))
	    ++done;
	  break;

	case L_('['):	/* Character class.  */
	  if (flags & LONG)
	    STRING_ARG (wstr, wchar_t);
	  else
	    STRING_ARG (str, char);

	  if (*f == L_('^'))
	    {
	      ++f;
	      not_in = 1;
	    }
	  else
	    not_in = 0;

	  if (width < 0)
	    /* There is no width given so there is also no limit on the
	       number of characters we read.  Therefore we set width to
	       a very high value to make the algorithm easier.  */
	    width = INT_MAX;

#ifdef COMPILE_WSCANF
	  /* Find the beginning and the end of the scanlist.  We are not
	     creating a lookup table since it would have to be too large.
	     Instead we search each time through the string.  This is not
	     a constant lookup time but who uses this feature deserves to
	     be punished.  */
	  tw = (wchar_t *) f;	/* Marks the beginning.  */

	  if (*f == L']')
	    ++f;

	  while ((fc = *f++) != L'\0' && fc != L']');

	  if (fc == L'\0')
	    conv_error ();
	  wp = (wchar_t *) f - 1;
#else
	  /* Fill WP with byte flags indexed by character.
	     We will use this flag map for matching input characters.  */
	  if (wpmax < UCHAR_MAX + 1)
	    {
	      wpmax = UCHAR_MAX + 1;
	      wp = (char *) alloca (wpmax);
	    }
	  memset (wp, '\0', UCHAR_MAX + 1);

	  fc = *f;
	  if (fc == ']' || fc == '-')
	    {
	      /* If ] or - appears before any char in the set, it is not
		 the terminator or separator, but the first char in the
		 set.  */
	      wp[fc] = 1;
	      ++f;
	    }

	  while ((fc = *f++) != '\0' && fc != ']')
	    if (fc == '-' && *f != '\0' && *f != ']'
		&& (unsigned char) f[-2] <= (unsigned char) *f)
	      {
		/* Add all characters from the one before the '-'
		   up to (but not including) the next format char.  */
		for (fc = (unsigned char) f[-2]; fc < (unsigned char) *f; ++fc)
		  wp[fc] = 1;
	      }
	    else
	      /* Add the character to the flag map.  */
	      wp[fc] = 1;

	  if (fc == '\0')
	    conv_error();
#endif

	  if (flags & LONG)
	    {
	      size_t now = read_in;
#ifdef COMPILE_WSCANF
	      if (inchar () == WEOF)
		input_error ();

	      do
		{
		  wchar_t *runp;

		  /* Test whether it's in the scanlist.  */
		  runp = tw;
		  while (runp < wp)
		    {
		      if (runp[0] == L'-' && runp[1] != '\0' && runp + 1 != wp
			  && runp != tw
			  && (unsigned int) runp[-1] <= (unsigned int) runp[1])
			{
			  /* Match against all characters in between the
			     first and last character of the sequence.  */
			  wchar_t wc;

			  for (wc = runp[-1] + 1; wc <= runp[1]; ++wc)
			    if ((wint_t) wc == c)
			      break;

			  if (wc <= runp[1] && !not_in)
			    break;
			  if (wc <= runp[1] && not_in)
			    {
			      /* The current character is not in the
                                 scanset.  */
			      ungetc (c, s);
			      goto out;
			    }

			  runp += 2;
			}
		      else
			{
			  if ((wint_t) *runp == c && !not_in)
			    break;
			  if ((wint_t) *runp == c && not_in)
			    {
			      ungetc (c, s);
			      goto out;
			    }

			  ++runp;
			}
		    }

		  if (runp == wp && !not_in)
		    {
		      ungetc (c, s);
		      goto out;
		    }

		  if (!(flags & SUPPRESS))
		    {
		      *wstr++ = c;

		      if ((flags & MALLOC)
			  && wstr == (wchar_t *) *strptr + strsize)
			{
			  /* Enlarge the buffer.  */
			  wstr = (wchar_t *) realloc (*strptr,
						      (2 * strsize)
						      * sizeof (wchar_t));
			  if (wstr == NULL)
			    {
			      /* Can't allocate that much.  Last-ditch
				 effort.  */
			      wstr = (wchar_t *)
				realloc (*strptr, (strsize + 1)
						  * sizeof (wchar_t));
			      if (wstr == NULL)
				{
				  /* We lose.  Oh well.  Terminate the string
				     and stop converting, so at least we don't
				     skip any input.  */
				  ((wchar_t *) (*strptr))[strsize - 1] = L'\0';
				  ++done;
				  conv_error ();
				}
			      else
				{
				  *strptr = (char *) wstr;
				  wstr += strsize;
				  ++strsize;
				}
			    }
			  else
			    {
			      *strptr = (char *) wstr;
			      wstr += strsize;
			      strsize *= 2;
			    }
			}
		    }
		}
	      while (--width > 0 && inchar () != WEOF);
	    out:
#else
	      char buf[MB_LEN_MAX];
	      size_t cnt = 0;
	      mbstate_t cstate;

	      if (inchar () == EOF)
		input_error ();

	      memset (&cstate, '\0', sizeof (cstate));

	      do
		{
		  if (wp[c] == not_in)
		    {
		      ungetc_not_eof (c, s);
		      break;
		    }

		  /* This is easy.  */
		  if (!(flags & SUPPRESS))
		    {
		      size_t n;

		      /* Convert it into a wide character.  */
		      buf[0] = c;
		      n = __mbrtowc (wstr, buf, 1, &cstate);

		      if (n == (size_t) -2)
			{
			  /* Possibly correct character, just not enough
			     input.  */
			  ++cnt;
			  assert (cnt < MB_CUR_MAX);
			  continue;
			}
		      cnt = 0;

		      ++wstr;
		      if ((flags & MALLOC)
			  && wstr == (wchar_t *) *strptr + strsize)
			{
			  /* Enlarge the buffer.  */
			  wstr = (wchar_t *) realloc (*strptr,
						      (2 * strsize
						       * sizeof (wchar_t)));
			  if (wstr == NULL)
			    {
			      /* Can't allocate that much.  Last-ditch
				 effort.  */
			      wstr = (wchar_t *)
				realloc (*strptr, ((strsize + 1)
						   * sizeof (wchar_t)));
			      if (wstr == NULL)
				{
				  /* We lose.  Oh well.  Terminate the
				     string and stop converting,
				     so at least we don't skip any input.  */
				  ((wchar_t *) (*strptr))[strsize - 1] = L'\0';
				  ++done;
				  conv_error ();
				}
			      else
				{
				  *strptr = (char *) wstr;
				  wstr += strsize;
				  ++strsize;
				}
			    }
			  else
			    {
			      *strptr = (char *) wstr;
			      wstr += strsize;
			      strsize *= 2;
			    }
			}
		    }

		  if (--width <= 0)
		    break;
		}
	      while (inchar () != EOF);

	      if (cnt != 0)
		/* We stopped in the middle of recognizing another
		   character.  That's a problem.  */
		encode_error ();
#endif

	      if (now == read_in)
		/* We haven't succesfully read any character.  */
		conv_error ();

	      if (!(flags & SUPPRESS))
		{
		  *wstr++ = L'\0';

		  if ((flags & MALLOC)
		      && wstr - (wchar_t *) *strptr != strsize)
		    {
		      wchar_t *cp = (wchar_t *)
			realloc (*strptr, ((wstr - (wchar_t *) *strptr)
					   * sizeof(wchar_t)));
		      if (cp != NULL)
			*strptr = (char *) cp;
		    }

		  ++done;
		}
	    }
	  else
	    {
	      size_t now = read_in;

	      if (inchar () == EOF)
		input_error ();

#ifdef COMPILE_WSCANF

	      memset (&state, '\0', sizeof (state));

	      do
		{
		  wchar_t *runp;
		  size_t n;

		  /* Test whether it's in the scanlist.  */
		  runp = tw;
		  while (runp < wp)
		    {
		      if (runp[0] == L'-' && runp[1] != '\0' && runp + 1 != wp
			  && runp != tw
			  && (unsigned int) runp[-1] <= (unsigned int) runp[1])
			{
			  /* Match against all characters in between the
			     first and last character of the sequence.  */
			  wchar_t wc;

			  for (wc = runp[-1] + 1; wc <= runp[1]; ++wc)
			    if ((wint_t) wc == c)
			      break;

			  if (wc <= runp[1] && !not_in)
			    break;
			  if (wc <= runp[1] && not_in)
			    {
			      /* The current character is not in the
                                 scanset.  */
			      ungetc (c, s);
			      goto out2;
			    }

			  runp += 2;
			}
		      else
			{
			  if ((wint_t) *runp == c && !not_in)
			    break;
			  if ((wint_t) *runp == c && not_in)
			    {
			      ungetc (c, s);
			      goto out2;
			    }

			  ++runp;
			}
		    }

		  if (runp == wp && !not_in)
		    {
		      ungetc (c, s);
		      goto out2;
		    }

		  if (!(flags & SUPPRESS))
		    {
		      if ((flags & MALLOC)
			  && str + MB_CUR_MAX >= *strptr + strsize)
			{
			  /* Enlarge the buffer.  */
			  size_t strleng = str - *strptr;
			  char *newstr;

			  newstr = (char *) realloc (*strptr, 2 * strsize);
			  if (newstr == NULL)
			    {
			      /* Can't allocate that much.  Last-ditch
				 effort.  */
			      newstr = (char *) realloc (*strptr,
							 strleng + MB_CUR_MAX);
			      if (newstr == NULL)
				{
				  /* We lose.  Oh well.  Terminate the string
				     and stop converting, so at least we don't
				     skip any input.  */
				  ((char *) (*strptr))[strleng] = '\0';
				  ++done;
				  conv_error ();
				}
			      else
				{
				  *strptr = newstr;
				  str = newstr + strleng;
				  strsize = strleng + MB_CUR_MAX;
				}
			    }
			  else
			    {
			      *strptr = newstr;
			      str = newstr + strleng;
			      strsize *= 2;
			    }
			}
		    }

		  n = __wcrtomb (!(flags & SUPPRESS) ? str : NULL, c, &state);
		  if (n == (size_t) -1)
		    encode_error ();

		  assert (n <= MB_CUR_MAX);
		  str += n;
		}
	      while (--width > 0 && inchar () != WEOF);
	    out2:
#else
	      do
		{
		  if (wp[c] == not_in)
		    {
		      ungetc_not_eof (c, s);
		      break;
		    }

		  /* This is easy.  */
		  if (!(flags & SUPPRESS))
		    {
		      *str++ = c;
		      if ((flags & MALLOC)
			  && (char *) str == *strptr + strsize)
			{
			  /* Enlarge the buffer.  */
			  size_t newsize = 2 * strsize;

			allocagain:
			  str = (char *) realloc (*strptr, newsize);
			  if (str == NULL)
			    {
			      /* Can't allocate that much.  Last-ditch
				 effort.  */
			      if (newsize > strsize + 1)
				{
				  newsize = strsize + 1;
				  goto allocagain;
				}
			      /* We lose.  Oh well.  Terminate the
				 string and stop converting,
				 so at least we don't skip any input.  */
			      ((char *) (*strptr))[strsize - 1] = '\0';
			      ++done;
			      conv_error ();
			    }
			  else
			    {
			      *strptr = (char *) str;
			      str += strsize;
			      strsize = newsize;
			    }
			}
		    }
		}
	      while (--width > 0 && inchar () != EOF);
#endif

	      if (now == read_in)
		/* We haven't succesfully read any character.  */
		conv_error ();

	      if (!(flags & SUPPRESS))
		{
#ifdef COMPILE_WSCANF
		  /* We have to emit the code to get into the initial
		     state.  */
		  char buf[MB_LEN_MAX];
		  size_t n = __wcrtomb (buf, L'\0', &state);
		  if (n > 0 && (flags & MALLOC)
		      && str + n >= *strptr + strsize)
		    {
		      /* Enlarge the buffer.  */
		      size_t strleng = str - *strptr;
		      char *newstr;

		      newstr = (char *) realloc (*strptr, strleng + n + 1);
		      if (newstr == NULL)
			{
			  /* We lose.  Oh well.  Terminate the string
			     and stop converting, so at least we don't
			     skip any input.  */
			  ((char *) (*strptr))[strleng] = '\0';
			  ++done;
			  conv_error ();
			}
		      else
			{
			  *strptr = newstr;
			  str = newstr + strleng;
			  strsize = strleng + n + 1;
			}
		    }

		  str = __mempcpy (str, buf, n);
#endif
		  *str++ = '\0';

		  if ((flags & MALLOC) && str - *strptr != strsize)
		    {
		      char *cp = (char *) realloc (*strptr, str - *strptr);
		      if (cp != NULL)
			*strptr = cp;
		    }

		  ++done;
		}
	    }
	  break;

	case L_('p'):	/* Generic pointer.  */
	  base = 16;
	  /* A PTR must be the same size as a `long int'.  */
	  flags &= ~(SHORT|LONGDBL);
	  if (need_long)
	    flags |= LONG;
	  number_signed = 0;
	  read_pointer = 1;
	  goto number;

	default:
	  /* If this is an unknown format character punt.  */
	  conv_error ();
	}
    }

  /* The last thing we saw int the format string was a white space.
     Consume the last white spaces.  */
  if (skip_space)
    {
      do
	c = inchar ();
      while (ISSPACE (c));
      ungetc (c, s);
    }

 errout:
  /* Unlock stream.  */
  UNLOCK_STREAM (s);

  if (errp != NULL)
    *errp |= errval;

  return done;
}

#ifdef COMPILE_WSCANF
int
__vfwscanf (FILE *s, const wchar_t *format, va_list argptr)
{
  return _IO_vfwscanf (s, format, argptr, NULL);
}
#else
int
__vfscanf (FILE *s, const char *format, va_list argptr)
{
  return INTUSE(_IO_vfscanf) (s, format, argptr, NULL);
}
libc_hidden_def (__vfscanf)
#endif

#ifdef COMPILE_WSCANF
weak_alias (__vfwscanf, vfwscanf)
#else
weak_alias (__vfscanf, vfscanf)
INTDEF(_IO_vfscanf)
#endif