GitHub Viewer
/* stringlib: codec implementations */
#if !STRINGLIB_IS_UNICODE
# error "codecs.h is specific to Unicode"
#endif
/* Mask to quickly check whether a C 'long' contains a
non-ASCII, UTF8-encoded char. */
#if (SIZEOF_LONG == 8)
# define ASCII_CHAR_MASK 0x8080808080808080UL
#elif (SIZEOF_LONG == 4)
# define ASCII_CHAR_MASK 0x80808080UL
#else
# error C 'long' size should be either 4 or 8!
#endif
/* 10xxxxxx */
#define IS_CONTINUATION_BYTE(ch) ((ch) >= 0x80 && (ch) < 0xC0)
Py_LOCAL_INLINE(Py_UCS4)
STRINGLIB(utf8_decode)(const char **inptr, const char *end,
STRINGLIB_CHAR *dest,
Py_ssize_t *outpos)
{
Py_UCS4 ch;
const char *s = *inptr;
const char *aligned_end = (const char *) _Py_ALIGN_DOWN(end, SIZEOF_LONG);
STRINGLIB_CHAR *p = dest + *outpos;
while (s < end) {
ch = (unsigned char)*s;
if (ch < 0x80) {
/* Fast path for runs of ASCII characters. Given that common UTF-8
input will consist of an overwhelming majority of ASCII
characters, we try to optimize for this case by checking
as many characters as a C 'long' can contain.
First, check if we can do an aligned read, as most CPUs have
a penalty for unaligned reads.
*/
if (_Py_IS_ALIGNED(s, SIZEOF_LONG)) {
/* Help register allocation */
const char *_s = s;
STRINGLIB_CHAR *_p = p;
while (_s < aligned_end) {
/* Read a whole long at a time (either 4 or 8 bytes),
and do a fast unrolled copy if it only contains ASCII
characters. */
unsigned long value = *(unsigned long *) _s;
if (value & ASCII_CHAR_MASK)
break;
#if PY_LITTLE_ENDIAN
_p[0] = (STRINGLIB_CHAR)(value & 0xFFu);
_p[1] = (STRINGLIB_CHAR)((value >> 8) & 0xFFu);
_p[2] = (STRINGLIB_CHAR)((value >> 16) & 0xFFu);
_p[3] = (STRINGLIB_CHAR)((value >> 24) & 0xFFu);
# if SIZEOF_LONG == 8
_p[4] = (STRINGLIB_CHAR)((value >> 32) & 0xFFu);
_p[5] = (STRINGLIB_CHAR)((value >> 40) & 0xFFu);
_p[6] = (STRINGLIB_CHAR)((value >> 48) & 0xFFu);
_p[7] = (STRINGLIB_CHAR)((value >> 56) & 0xFFu);
# endif
#else
# if SIZEOF_LONG == 8
_p[0] = (STRINGLIB_CHAR)((value >> 56) & 0xFFu);
_p[1] = (STRINGLIB_CHAR)((value >> 48) & 0xFFu);
_p[2] = (STRINGLIB_CHAR)((value >> 40) & 0xFFu);
_p[3] = (STRINGLIB_CHAR)((value >> 32) & 0xFFu);
_p[4] = (STRINGLIB_CHAR)((value >> 24) & 0xFFu);
_p[5] = (STRINGLIB_CHAR)((value >> 16) & 0xFFu);
_p[6] = (STRINGLIB_CHAR)((value >> 8) & 0xFFu);
_p[7] = (STRINGLIB_CHAR)(value & 0xFFu);
# else
_p[0] = (STRINGLIB_CHAR)((value >> 24) & 0xFFu);
_p[1] = (STRINGLIB_CHAR)((value >> 16) & 0xFFu);
_p[2] = (STRINGLIB_CHAR)((value >> 8) & 0xFFu);
_p[3] = (STRINGLIB_CHAR)(value & 0xFFu);
# endif
#endif
_s += SIZEOF_LONG;
_p += SIZEOF_LONG;
}
s = _s;
p = _p;
if (s == end)
break;
ch = (unsigned char)*s;
}
if (ch < 0x80) {
s++;
*p++ = ch;
continue;
}
}
if (ch < 0xE0) {
/* \xC2\x80-\xDF\xBF -- 0080-07FF */
Py_UCS4 ch2;
if (ch < 0xC2) {
/* invalid sequence
\x80-\xBF -- continuation byte
\xC0-\xC1 -- fake 0000-007F */
goto InvalidStart;
}
if (end - s < 2) {
/* unexpected end of data: the caller will decide whether
it's an error or not */
break;
}
ch2 = (unsigned char)s[1];
if (!IS_CONTINUATION_BYTE(ch2))
/* invalid continuation byte */
goto InvalidContinuation1;
ch = (ch = 0x90) {
/* invalid sequence
\xF4\x90\x80\80- -- 110000- overflow */
goto InvalidContinuation1;
}
if (!IS_CONTINUATION_BYTE(ch3)) {
/* invalid continuation byte */
goto InvalidContinuation2;
}
if (!IS_CONTINUATION_BYTE(ch4)) {
/* invalid continuation byte */
goto InvalidContinuation3;
}
ch = (ch 12));
*p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*p++ = (char)(0x80 | (ch & 0x3f));
}
i += (endpos - startpos - 1);
break;
case _Py_ERROR_BACKSLASHREPLACE:
/* subtract preallocated bytes */
writer.min_size -= max_char_size * (endpos - startpos);
p = backslashreplace(&writer, p,
unicode, startpos, endpos);
if (p == NULL)
goto error;
i += (endpos - startpos - 1);
break;
case _Py_ERROR_XMLCHARREFREPLACE:
/* subtract preallocated bytes */
writer.min_size -= max_char_size * (endpos - startpos);
p = xmlcharrefreplace(&writer, p,
unicode, startpos, endpos);
if (p == NULL)
goto error;
i += (endpos - startpos - 1);
break;
case _Py_ERROR_SURROGATEESCAPE:
for (k=startpos; k> 12));
*p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*p++ = (char)(0x80 | (ch & 0x3f));
}
#if STRINGLIB_SIZEOF_CHAR > 2
else /* ch >= 0x10000 */
{
assert(ch > 18));
*p++ = (char)(0x80 | ((ch >> 12) & 0x3f));
*p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
*p++ = (char)(0x80 | (ch & 0x3f));
}
#endif /* STRINGLIB_SIZEOF_CHAR > 2 */
#endif /* STRINGLIB_SIZEOF_CHAR > 1 */
}
#if STRINGLIB_SIZEOF_CHAR > 1
Py_XDECREF(error_handler_obj);
Py_XDECREF(exc);
#endif
return _PyBytesWriter_Finish(&writer, p);
#if STRINGLIB_SIZEOF_CHAR > 1
error:
Py_XDECREF(rep);
Py_XDECREF(error_handler_obj);
Py_XDECREF(exc);
_PyBytesWriter_Dealloc(&writer);
return NULL;
#endif
}
/* The pattern for constructing UCS2-repeated masks. */
#if SIZEOF_LONG == 8
# define UCS2_REPEAT_MASK 0x0001000100010001ul
#elif SIZEOF_LONG == 4
# define UCS2_REPEAT_MASK 0x00010001ul
#else
# error C 'long' size should be either 4 or 8!
#endif
/* The mask for fast checking. */
#if STRINGLIB_SIZEOF_CHAR == 1
/* The mask for fast checking of whether a C 'long' contains a
non-ASCII or non-Latin1 UTF16-encoded characters. */
# define FAST_CHAR_MASK (UCS2_REPEAT_MASK * (0xFFFFu & ~STRINGLIB_MAX_CHAR))
#else
/* The mask for fast checking of whether a C 'long' may contain
UTF16-encoded surrogate characters. This is an efficient heuristic,
assuming that non-surrogate characters with a code point >= 0x8000 are
rare in most input.
*/
# define FAST_CHAR_MASK (UCS2_REPEAT_MASK * 0x8000u)
#endif
/* The mask for fast byte-swapping. */
#define STRIPPED_MASK (UCS2_REPEAT_MASK * 0x00FFu)
/* Swap bytes. */
#define SWAB(value) ((((value) >> 8) & STRIPPED_MASK) | \
(((value) & STRIPPED_MASK) >= 8;
#else
block = SWAB(block);
#endif
}
#if PY_LITTLE_ENDIAN
# if SIZEOF_LONG == 4
p[0] = (STRINGLIB_CHAR)(block & 0xFFFFu);
p[1] = (STRINGLIB_CHAR)(block >> 16);
# elif SIZEOF_LONG == 8
p[0] = (STRINGLIB_CHAR)(block & 0xFFFFu);
p[1] = (STRINGLIB_CHAR)((block >> 16) & 0xFFFFu);
p[2] = (STRINGLIB_CHAR)((block >> 32) & 0xFFFFu);
p[3] = (STRINGLIB_CHAR)(block >> 48);
# endif
#else
# if SIZEOF_LONG == 4
p[0] = (STRINGLIB_CHAR)(block >> 16);
p[1] = (STRINGLIB_CHAR)(block & 0xFFFFu);
# elif SIZEOF_LONG == 8
p[0] = (STRINGLIB_CHAR)(block >> 48);
p[1] = (STRINGLIB_CHAR)((block >> 32) & 0xFFFFu);
p[2] = (STRINGLIB_CHAR)((block >> 16) & 0xFFFFu);
p[3] = (STRINGLIB_CHAR)(block & 0xFFFFu);
# endif
#endif
_q += SIZEOF_LONG;
p += SIZEOF_LONG / 2;
}
q = _q;
if (q >= e)
break;
}
ch = (q[ihi] STRINGLIB_MAX_CHAR)
/* Out-of-range */
goto Return;
#endif
*p++ = (STRINGLIB_CHAR)ch;
continue;
}
/* UTF-16 code pair: */
if (q >= e)
goto UnexpectedEnd;
if (!Py_UNICODE_IS_HIGH_SURROGATE(ch))
goto IllegalEncoding;
ch2 = (q[ihi] = 0x80
Py_LOCAL_INLINE(Py_ssize_t)
STRINGLIB(utf16_encode)(const STRINGLIB_CHAR *in,
Py_ssize_t len,
unsigned short **outptr,
int native_ordering)
{
unsigned short *out = *outptr;
const STRINGLIB_CHAR *end = in + len;
#if STRINGLIB_SIZEOF_CHAR == 1
if (native_ordering) {
const STRINGLIB_CHAR *unrolled_end = in + _Py_SIZE_ROUND_DOWN(len, 4);
while (in < unrolled_end) {
out[0] = in[0];
out[1] = in[1];
out[2] = in[2];
out[3] = in[3];
in += 4; out += 4;
}
while (in < end) {
*out++ = *in++;
}
} else {
# define SWAB2(CH) ((CH) = 0x10000
else if (ch >= 0x10000) {
out[0] = Py_UNICODE_HIGH_SURROGATE(ch);
out[1] = Py_UNICODE_LOW_SURROGATE(ch);
out += 2;
}
#endif
else
*out++ = ch;
}
} else {
#define SWAB2(CH) (((CH) > 8))
#if STRINGLIB_MAX_CHAR < 0x10000
const STRINGLIB_CHAR *unrolled_end = in + _Py_SIZE_ROUND_DOWN(len, 4);
while (in < unrolled_end) {
/* check if any character is a surrogate character */
if (((in[0] ^ 0xd800) &
(in[1] ^ 0xd800) &
(in[2] ^ 0xd800) &
(in[3] ^ 0xd800) & 0xf800) == 0)
break;
out[0] = SWAB2(in[0]);
out[1] = SWAB2(in[1]);
out[2] = SWAB2(in[2]);
out[3] = SWAB2(in[3]);
in += 4; out += 4;
}
#endif
while (in < end) {
Py_UCS4 ch = *in++;
if (ch < 0xd800)
*out++ = SWAB2((Py_UCS2)ch);
else if (ch < 0xe000)
/* reject surrogate characters (U+D800-U+DFFF) */
goto fail;
#if STRINGLIB_MAX_CHAR >= 0x10000
else if (ch >= 0x10000) {
Py_UCS2 ch1 = Py_UNICODE_HIGH_SURROGATE(ch);
Py_UCS2 ch2 = Py_UNICODE_LOW_SURROGATE(ch);
out[0] = SWAB2(ch1);
out[1] = SWAB2(ch2);
out += 2;
}
#endif
else
*out++ = SWAB2((Py_UCS2)ch);
}
#undef SWAB2
}
*outptr = out;
return len;
fail:
*outptr = out;
return len - (end - in + 1);
#endif
}
#if STRINGLIB_SIZEOF_CHAR == 1
# define SWAB4(CH, tmp) ((CH) 1
/* check if any character is a surrogate character */
if (((in[0] ^ 0xd800) &
(in[1] ^ 0xd800) &
(in[2] ^ 0xd800) &
(in[3] ^ 0xd800) & 0xf800) == 0)
break;
#endif
out[0] = in[0];
out[1] = in[1];
out[2] = in[2];
out[3] = in[3];
in += 4; out += 4;
}
while (in < end) {
Py_UCS4 ch;
ch = *in++;
#if STRINGLIB_SIZEOF_CHAR > 1
if (Py_UNICODE_IS_SURROGATE(ch)) {
/* reject surrogate characters (U+D800-U+DFFF) */
goto fail;
}
#endif
*out++ = ch;
}
} else {
const STRINGLIB_CHAR *unrolled_end = in + _Py_SIZE_ROUND_DOWN(len, 4);
while (in < unrolled_end) {
#if STRINGLIB_SIZEOF_CHAR > 1
Py_UCS4 ch1, ch2, ch3, ch4;
/* check if any character is a surrogate character */
if (((in[0] ^ 0xd800) &
(in[1] ^ 0xd800) &
(in[2] ^ 0xd800) &
(in[3] ^ 0xd800) & 0xf800) == 0)
break;
#endif
out[0] = SWAB4(in[0], ch1);
out[1] = SWAB4(in[1], ch2);
out[2] = SWAB4(in[2], ch3);
out[3] = SWAB4(in[3], ch4);
in += 4; out += 4;
}
while (in < end) {
Py_UCS4 ch = *in++;
#if STRINGLIB_SIZEOF_CHAR > 1
if (Py_UNICODE_IS_SURROGATE(ch)) {
/* reject surrogate characters (U+D800-U+DFFF) */
goto fail;
}
#endif
*out++ = SWAB4(ch, ch);
}
}
*outptr = out;
return len;
#if STRINGLIB_SIZEOF_CHAR > 1
fail:
*outptr = out;
return len - (end - in + 1);
#endif
}
#undef SWAB4
#endif