gconv_simple.c

KPIT GNU Tools is a set of GNU development tools for Renesas microcontrollers.

  }
#define LOOP_NEED_FLAGS
#include <iconv/loop.c>
#include <iconv/skeleton.c>


/* Convert from UTF-8 to the internal (UCS4-like) format.  */
#define DEFINE_INIT		0
#define DEFINE_FINI		0
#define MIN_NEEDED_FROM		1
#define MAX_NEEDED_FROM		6
#define MIN_NEEDED_TO		4
#define FROM_DIRECTION		1
#define FROM_LOOP		utf8_internal_loop
#define TO_LOOP			utf8_internal_loop /* This is not used.  */
#define FUNCTION_NAME		__gconv_transform_utf8_internal
#define ONE_DIRECTION		1

#define MIN_NEEDED_INPUT	MIN_NEEDED_FROM
#define MAX_NEEDED_INPUT	MAX_NEEDED_FROM
#define MIN_NEEDED_OUTPUT	MIN_NEEDED_TO
#define LOOPFCT			FROM_LOOP
#define BODY \
  {									      \
    uint32_t ch;							      \
    uint32_t cnt;							      \
    uint32_t i;							      \
									      \
    /* Next input byte.  */						      \
    ch = *inptr;							      \
									      \
    if (ch < 0x80)							      \
      {									      \
	/* One byte sequence.  */					      \
	cnt = 1;							      \
	++inptr;							      \
      }									      \
    else								      \
      {									      \
 	if (ch >= 0xc2 && ch < 0xe0)					      \
	  {								      \
 	    /* We expect two bytes.  The first byte cannot be 0xc0 or 0xc1,   \
 	       otherwise the wide character could have been represented	      \
 	       using a single byte.  */					      \
	    cnt = 2;							      \
	    ch &= 0x1f;							      \
	  }								      \
        else if (__builtin_expect ((ch & 0xf0) == 0xe0, 1))		      \
	  {								      \
	    /* We expect three bytes.  */				      \
	    cnt = 3;							      \
	    ch &= 0x0f;							      \
	  }								      \
	else if (__builtin_expect ((ch & 0xf8) == 0xf0, 1))		      \
	  {								      \
	    /* We expect four bytes.  */				      \
	    cnt = 4;							      \
	    ch &= 0x07;							      \
	  }								      \
	else if (__builtin_expect ((ch & 0xfc) == 0xf8, 1))		      \
	  {								      \
	    /* We expect five bytes.  */				      \
	    cnt = 5;							      \
	    ch &= 0x03;							      \
	  }								      \
	else if (__builtin_expect ((ch & 0xfe) == 0xfc, 1))		      \
	  {								      \
	    /* We expect six bytes.  */					      \
	    cnt = 6;							      \
	    ch &= 0x01;							      \
	  }								      \
	else								      \
	  {								      \
	    int skipped;						      \
									      \
	    if (! ignore_errors_p ())					      \
	      {								      \
		/* This is an illegal encoding.  */			      \
		result = __GCONV_ILLEGAL_INPUT;				      \
		break;							      \
	      }								      \
									      \
	    /* Search the end of this ill-formed UTF-8 character.  This	      \
	       is the next byte with (x & 0xc0) != 0x80.  */		      \
	     skipped = 0;						      \
	     do								      \
	       {							      \
		 ++inptr;						      \
		 ++skipped;						      \
	       }							      \
	     while (inptr < inend && (*inptr & 0xc0) == 0x80 && skipped < 5); \
									      \
	     continue;							      \
	  }								      \
									      \
	if (__builtin_expect (inptr + cnt > inend, 0))			      \
	  {								      \
	    /* We don't have enough input.  But before we report that check   \
	       that all the bytes are correct.  */			      \
	    for (i = 1; inptr + i < inend; ++i)				      \
	      if ((inptr[i] & 0xc0) != 0x80)				      \
		break;							      \
									      \
	    if (__builtin_expect (inptr + i == inend, 1))		      \
	      {								      \
		result = __GCONV_INCOMPLETE_INPUT;			      \
		break;							      \
	      }								      \
									      \
	    if (ignore_errors_p ())					      \
	      {								      \
		/* Ignore it.  */					      \
		inptr += i;						      \
		*irreversible = *irreversible + 1;			      \
		continue;						      \
	      }								      \
									      \
	    result = __GCONV_ILLEGAL_INPUT;				      \
	    break;							      \
	  }								      \
									      \
	/* Read the possible remaining bytes.  */			      \
	for (i = 1; i < cnt; ++i)					      \
	  {								      \
	    uint32_t byte = inptr[i];					      \
									      \
	    if ((byte & 0xc0) != 0x80)					      \
	      /* This is an illegal encoding.  */			      \
	      break;							      \
									      \
	    ch <<= 6;							      \
	    ch |= byte & 0x3f;						      \
	  }								      \
									      \
	/* If i < cnt, some trail byte was not >= 0x80, < 0xc0.		      \
	   If cnt > 2 and ch < 2^(5*cnt-4), the wide character ch could	      \
	   have been represented with fewer than cnt bytes.  */		      \
	if (i < cnt || (cnt > 2 && (ch >> (5 * cnt - 4)) == 0))		      \
	  {								      \
	    /* This is an illegal encoding.  */				      \
	    if (ignore_errors_p ())					      \
	      {								      \
		inptr += i;						      \
		*irreversible = *irreversible + 1;			      \
		continue;						      \
	      }								      \
									      \
	    result = __GCONV_ILLEGAL_INPUT;				      \
	    break;							      \
	  }								      \
									      \
	inptr += cnt;							      \
      }									      \
									      \
    /* Now adjust the pointers and store the result.  */		      \
    *((uint32_t *) outptr) = ch;					      \
    outptr = (unsigned char *)((uint32_t *) outptr + 1);		      \
  }
#define LOOP_NEED_FLAGS

#define STORE_REST \
  {									      \
    /* We store the remaining bytes while converting them into the UCS4	      \
       format.  We can assume that the first byte in the buffer is	      \
       correct and that it requires a larger number of bytes than there	      \
       are in the input buffer.  */					      \
    wint_t ch = **inptrp;						      \
    size_t cnt;								      \
									      \
    state->__count = inend - *inptrp;					      \
									      \
    if (ch >= 0xc2 && ch < 0xe0)					      \
      {									      \
	/* We expect two bytes.  The first byte cannot be 0xc0 or	      \
	   0xc1, otherwise the wide character could have been		      \
	   represented using a single byte.  */				      \
	cnt = 2;							      \
	ch &= 0x1f;							      \
      }									      \
    else if (__builtin_expect ((ch & 0xf0) == 0xe0, 1))			      \
      {									      \
	/* We expect three bytes.  */					      \
	cnt = 3;							      \
	ch &= 0x0f;							      \
      }									      \
    else if (__builtin_expect ((ch & 0xf8) == 0xf0, 1))			      \
      {									      \
	/* We expect four bytes.  */					      \
	cnt = 4;							      \
	ch &= 0x07;							      \
      }									      \
    else if (__builtin_expect ((ch & 0xfc) == 0xf8, 1))			      \
      {									      \
	/* We expect five bytes.  */					      \
	cnt = 5;							      \
	ch &= 0x03;							      \
      }									      \
    else								      \
      {									      \
	/* We expect six bytes.  */					      \
	cnt = 6;							      \
	ch &= 0x01;							      \
      }									      \
									      \
    /* The first byte is already consumed.  */				      \
    --cnt;								      \
    while (++(*inptrp) < inend)						      \
      {									      \
	ch <<= 6;							      \
	ch |= **inptrp & 0x3f;						      \
	--cnt;								      \
      }									      \
									      \
    /* Shift for the so far missing bytes.  */				      \
    ch <<= cnt * 6;							      \
									      \
    /* Store the value.  */						      \
    state->__value.__wch = ch;						      \
  }

#define UNPACK_BYTES \
  {									      \
    wint_t wch = state->__value.__wch;					      \
    size_t ntotal;							      \
    inlen = state->__count;						      \
									      \
    if (state->__value.__wch <= 0x7ff)					      \
      {									      \
	bytebuf[0] = 0xc0;						      \
	ntotal = 2;							      \
      }									      \
    else if (__builtin_expect (state->__value.__wch <= 0xffff, 1))	      \
      {									      \
	bytebuf[0] = 0xe0;						      \
	ntotal = 3;							      \
      }									      \
    else if (__builtin_expect (state->__value.__wch < 0x1fffff, 1))	      \
      {									      \
	bytebuf[0] = 0xf0;						      \
	ntotal = 4;							      \
      }									      \
    else if (__builtin_expect (state->__value.__wch < 0x3ffffff, 1))	      \
      {									      \
	bytebuf[0] = 0xf8;						      \
	ntotal = 5;							      \
      }									      \
    else								      \
      {									      \
	bytebuf[0] = 0xfc;						      \
	ntotal = 6;							      \
      }									      \
									      \
    do									      \
      {									      \
	if (--ntotal < inlen)						      \
	  bytebuf[ntotal] = 0x80 | (wch & 0x3f);			      \
	wch >>= 6;							      \
      }									      \
    while (ntotal > 1);							      \
									      \
    bytebuf[0] |= wch;							      \
  }

#include <iconv/loop.c>
#include <iconv/skeleton.c>


/* Convert from UCS2 to the internal (UCS4-like) format.  */
#define DEFINE_INIT		0
#define DEFINE_FINI		0
#define MIN_NEEDED_FROM		2
#define MIN_NEEDED_TO		4
#define FROM_DIRECTION		1
#define FROM_LOOP		ucs2_internal_loop
#define TO_LOOP			ucs2_internal_loop /* This is not used.  */
#define FUNCTION_NAME		__gconv_transform_ucs2_internal
#define ONE_DIRECTION		1

#define MIN_NEEDED_INPUT	MIN_NEEDED_FROM
#define MIN_NEEDED_OUTPUT	MIN_NEEDED_TO
#define LOOPFCT			FROM_LOOP
#define BODY \
  {									      \
    uint16_t u1 = *((const uint16_t *) inptr);				      \
									      \
    if (__builtin_expect (u1 >= 0xd800 && u1 < 0xe000, 0))		      \
      {									      \
	/* Surrogate characters in UCS-2 input are not valid.  Reject	      \
	   them.  (Catching this here is not security relevant.)  */	      \
	if (! ignore_errors_p ())					      \
	  {								      \
	    result = __GCONV_ILLEGAL_INPUT;				      \
	    break;							      \
	  }								      \
	inptr += 2;							      \
	*irreversible = *irreversible + 1;				      \
	continue;							      \
      }									      \
									      \
    *((uint32_t *) outptr) = u1;					      \
    outptr = (unsigned char *)((uint32_t *) outptr + 1);		      \
    inptr += 2;								      \
  }
#define LOOP_NEED_FLAGS
#include <iconv/loop.c>
#include <iconv/skeleton.c>


/* Convert from the internal (UCS4-like) format to UCS2.  */
#define DEFINE_INIT		0
#define DEFINE_FINI		0
#define MIN_NEEDED_FROM		4
#define MIN_NEEDED_TO		2
#define FROM_DIRECTION		1
#define FROM_LOOP		internal_ucs2_loop
#define TO_LOOP			internal_ucs2_loop /* This is not used.  */
#define FUNCTION_NAME		__gconv_transform_internal_ucs2
#define ONE_DIRECTION		1

#define MIN_NEEDED_INPUT	MIN_NEEDED_FROM
#define MIN_NEEDED_OUTPUT	MIN_NEEDED_TO
#define LOOPFCT			FROM_LOOP
#define BODY \
  {									      \
    uint32_t val = *((const uint32_t *) inptr);				      \
									      \
    if (__builtin_expect (val >= 0x10000, 0))				      \
      {									      \
	UNICODE_TAG_HANDLER (val, 4);					      \
	STANDARD_ERR_HANDLER (4);					      \
      }									      \
    else if (__builtin_expect (val >= 0xd800 && val < 0xe000, 0))	      \
      {									      \
	/* Surrogate characters in UCS-4 input are not valid.		      \
	   We must catch this, because the UCS-2 output might be	      \
	   interpreted as UTF-16 by other programs.  If we let		      \
	   surrogates pass through, attackers could make a security	      \
	   hole exploit by synthesizing any desired plane 1-16		      \
	   character.  */						      \
	if (! ignore_errors_p ())					      \
	  {								      \
	    result = __GCONV_ILLEGAL_INPUT;				      \
	    break;							      \
	  }								      \
	inptr += 4;							      \
	*irreversible = *irreversible + 1;				      \
	continue;							      \
      }									      \
    else 								      \
      {									      \
	*((uint16_t *) outptr) = val;					      \
	outptr = (unsigned char *)((uint16_t *) outptr + 1);		      \
	inptr += 4;							      \
      }									      \
  }
#define LOOP_NEED_FLAGS
#include <iconv/loop.c>
#include <iconv/skeleton.c>


/* Convert from UCS2 in other endianness to the internal (UCS4-like) format. */
#define DEFINE_INIT		0
#define DEFINE_FINI		0
#define MIN_NEEDED_FROM		2
#define MIN_NEEDED_TO		4
#define FROM_DIRECTION		1
#define FROM_LOOP		ucs2reverse_internal_loop
#define TO_LOOP			ucs2reverse_internal_loop/* This is not used.*/
#define FUNCTION_NAME		__gconv_transform_ucs2reverse_internal
#define ONE_DIRECTION		1

#define MIN_NEEDED_INPUT	MIN_NEEDED_FROM
#define MIN_NEEDED_OUTPUT	MIN_NEEDED_TO
#define LOOPFCT			FROM_LOOP
#define BODY \
  {									      \
    uint16_t u1 = bswap_16 (*((const uint16_t *) inptr));		      \
									      \
    if (__builtin_expect (u1 >= 0xd800 && u1 < 0xe000, 0))		      \
      {									      \
	/* Surrogate characters in UCS-2 input are not valid.  Reject	      \
	   them.  (Catching this here is not security relevant.)  */	      \
	if (! ignore_errors_p ())					      \
	  {								      \
	    result = __GCONV_ILLEGAL_INPUT;				      \
	    break;							      \
	  }								      \
	inptr += 2;							      \
	*irreversible = *irreversible + 1;				      \
	continue;							      \
      }									      \
									      \
    *((uint32_t *) outptr) = u1;					      \
    outptr = (unsigned char *)((uint32_t *) outptr + 1);		      \
    inptr += 2;								      \
  }
#define LOOP_NEED_FLAGS
#include <iconv/loop.c>
#include <iconv/skeleton.c>


/* Convert from the internal (UCS4-like) format to UCS2 in other endianness. */
#define DEFINE_INIT		0
#define DEFINE_FINI		0
#define MIN_NEEDED_FROM		4
#define MIN_NEEDED_TO		2
#define FROM_DIRECTION		1
#define FROM_LOOP		internal_ucs2reverse_loop
#define TO_LOOP			internal_ucs2reverse_loop/* This is not used.*/
#define FUNCTION_NAME		__gconv_transform_internal_ucs2reverse
#define ONE_DIRECTION		1

#define MIN_NEEDED_INPUT	MIN_NEEDED_FROM
#define MIN_NEEDED_OUTPUT	MIN_NEEDED_TO
#define LOOPFCT			FROM_LOOP
#define BODY \
  {									      \
    uint32_t val = *((const uint32_t *) inptr);				      \
    if (__builtin_expect (val >= 0x10000, 0))				      \
      {									      \
	UNICODE_TAG_HANDLER (val, 4);					      \
	STANDARD_ERR_HANDLER (4);					      \
      }									      \
    else if (__builtin_expect (val >= 0xd800 && val < 0xe000, 0))	      \
      {									      \
	/* Surrogate characters in UCS-4 input are not valid.		      \
	   We must catch this, because the UCS-2 output might be	      \
	   interpreted as UTF-16 by other programs.  If we let		      \
	   surrogates pass through, attackers could make a security	      \
	   hole exploit by synthesizing any desired plane 1-16		      \
	   character.  */						      \
	if (! ignore_errors_p ())					      \
	  {								      \
	    result = __GCONV_ILLEGAL_INPUT;				      \
	    break;							      \
	  }								      \
	inptr += 4;							      \
	*irreversible = *irreversible + 1;				      \
	continue;							      \
      }									      \
    else 								      \
      {									      \
	*((uint16_t *) outptr) = bswap_16 (val);			      \
	outptr = (unsigned char *)((uint16_t *) outptr + 1);		      \
	inptr += 4;							      \
      }									      \
  }
#define LOOP_NEED_FLAGS
#include <iconv/loop.c>
#include <iconv/skeleton.c>