mirror of
https://github.com/python/cpython
synced 2024-09-05 00:05:39 +00:00
Support using UCS-4 as the Py_UNICODE type:
Add configure option --enable-unicode. Add config.h macros Py_USING_UNICODE, PY_UNICODE_TYPE, Py_UNICODE_SIZE, SIZEOF_WCHAR_T. Define Py_UCS2. Encode and decode large UTF-8 characters into single Py_UNICODE values for wide Unicode types; likewise for UTF-16. Remove test whether sizeof Py_UNICODE is two.
This commit is contained in:
parent
ff1cc902fe
commit
0ba70cc3c8
|
@ -60,16 +60,9 @@ Copyright (c) Corporation for National Research Initiatives.
|
|||
|
||||
/* experimental UCS-4 support. enable at your own risk! */
|
||||
#undef USE_UCS4_STORAGE
|
||||
|
||||
/*
|
||||
* Use this typedef when you need to represent a UTF-16 surrogate pair
|
||||
* as single unsigned integer.
|
||||
*/
|
||||
#if SIZEOF_INT >= 4
|
||||
typedef unsigned int Py_UCS4;
|
||||
#elif SIZEOF_LONG >= 4
|
||||
typedef unsigned long Py_UCS4;
|
||||
#endif
|
||||
#if Py_UNICODE_SIZE == 4
|
||||
#define USE_UCS4_STORAGE
|
||||
#endif
|
||||
|
||||
/* Set these flags if the platform has "wchar.h", "wctype.h" and the
|
||||
wchar_t type is a 16-bit unsigned type */
|
||||
|
@ -77,11 +70,16 @@ typedef unsigned long Py_UCS4;
|
|||
/* #define HAVE_USABLE_WCHAR_T */
|
||||
|
||||
/* Defaults for various platforms */
|
||||
#ifndef HAVE_USABLE_WCHAR_T
|
||||
#ifndef PY_UNICODE_TYPE
|
||||
|
||||
/* Windows has a usable wchar_t type (unless we're using UCS-4) */
|
||||
# if defined(MS_WIN32) && !defined(USE_UCS4_STORAGE)
|
||||
# define HAVE_USABLE_WCHAR_T
|
||||
# define PY_UNICODE_TYPE wchar_t
|
||||
# endif
|
||||
|
||||
# if defined(USE_UCS4_STORAGE)
|
||||
# define PY_UNICODE_TYPE Py_UCS4
|
||||
# endif
|
||||
|
||||
#endif
|
||||
|
@ -104,28 +102,23 @@ typedef unsigned long Py_UCS4;
|
|||
# include "wchar.h"
|
||||
#endif
|
||||
|
||||
#ifdef HAVE_USABLE_WCHAR_T
|
||||
|
||||
/* If the compiler defines whcar_t as a 16-bit unsigned type we can
|
||||
use the compiler type directly. Works fine with all modern Windows
|
||||
platforms. */
|
||||
|
||||
typedef wchar_t Py_UNICODE;
|
||||
|
||||
#else
|
||||
|
||||
/* Use if you have a standard ANSI compiler, without wchar_t support.
|
||||
If a short is not 16 bits on your platform, you have to fix the
|
||||
typedef below, or the module initialization code will complain. */
|
||||
|
||||
#ifdef USE_UCS4_STORAGE
|
||||
typedef Py_UCS4 Py_UNICODE;
|
||||
#else
|
||||
typedef unsigned short Py_UNICODE;
|
||||
/*
|
||||
* Use this typedef when you need to represent a UTF-16 surrogate pair
|
||||
* as single unsigned integer.
|
||||
*/
|
||||
#if SIZEOF_INT >= 4
|
||||
typedef unsigned int Py_UCS4;
|
||||
#elif SIZEOF_LONG >= 4
|
||||
typedef unsigned long Py_UCS4;
|
||||
#endif
|
||||
|
||||
#endif
|
||||
#if SIZEOF_SHORT == 2
|
||||
typedef unsigned short Py_UCS2;
|
||||
#else
|
||||
#error Cannot find a two-byte type
|
||||
#endif
|
||||
|
||||
typedef PY_UNICODE_TYPE Py_UNICODE;
|
||||
|
||||
/* --- Internal Unicode Operations ---------------------------------------- */
|
||||
|
||||
|
|
|
@ -771,13 +771,17 @@ PyObject *PyUnicode_DecodeUTF8(const char *s,
|
|||
ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
|
||||
((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
|
||||
/* validate and convert to UTF-16 */
|
||||
if ((ch < 0x10000) || /* minimum value allowed for 4
|
||||
if ((ch < 0x10000) /* minimum value allowed for 4
|
||||
byte encoding */
|
||||
(ch > 0x10ffff)) { /* maximum value allowed for
|
||||
|| (ch > 0x10ffff)) /* maximum value allowed for
|
||||
UTF-16 */
|
||||
{
|
||||
errmsg = "illegal encoding";
|
||||
goto utf8Error;
|
||||
}
|
||||
#if Py_UNICODE_SIZE == 4
|
||||
*p++ = (Py_UNICODE)ch;
|
||||
#else
|
||||
/* compute and append the two surrogates: */
|
||||
|
||||
/* translate from 10000..10FFFF to 0..FFFF */
|
||||
|
@ -788,6 +792,7 @@ PyObject *PyUnicode_DecodeUTF8(const char *s,
|
|||
|
||||
/* low surrogate = bottom 10 bits added to DC00 */
|
||||
*p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
|
||||
#endif
|
||||
break;
|
||||
|
||||
default:
|
||||
|
@ -878,7 +883,13 @@ PyObject *PyUnicode_EncodeUTF8(const Py_UNICODE *s,
|
|||
*p++ = 0x80 | (ch & 0x3f);
|
||||
cbWritten += 2;
|
||||
}
|
||||
else {
|
||||
else if (ch < 0x10000) {
|
||||
#if Py_UNICODE_SIZE == 4
|
||||
*p++ = 0xe0 | (ch>>12);
|
||||
*p++ = 0x80 | ((ch>>6) & 0x3f);
|
||||
*p++ = 0x80 | (ch & 0x3f);
|
||||
cbWritten += 3;
|
||||
#else
|
||||
/* Check for high surrogate */
|
||||
if (0xD800 <= ch && ch <= 0xDBFF) {
|
||||
if (i != size) {
|
||||
|
@ -909,7 +920,14 @@ PyObject *PyUnicode_EncodeUTF8(const Py_UNICODE *s,
|
|||
}
|
||||
*p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
|
||||
*p++ = (char)(0x80 | (ch & 0x3f));
|
||||
}
|
||||
#endif
|
||||
} else {
|
||||
*p++ = 0xf0 | (ch>>18);
|
||||
*p++ = 0x80 | ((ch>>12) & 0x3f);
|
||||
*p++ = 0x80 | ((ch>>6) & 0x3f);
|
||||
*p++ = 0x80 | (ch & 0x3f);
|
||||
cbWritten += 4;
|
||||
}
|
||||
}
|
||||
*p = '\0';
|
||||
if (_PyString_Resize(&v, p - q))
|
||||
|
@ -935,7 +953,7 @@ PyObject *PyUnicode_AsUTF8String(PyObject *unicode)
|
|||
/* --- UTF-16 Codec ------------------------------------------------------- */
|
||||
|
||||
static
|
||||
int utf16_decoding_error(const Py_UNICODE **source,
|
||||
int utf16_decoding_error(const Py_UCS2 **source,
|
||||
Py_UNICODE **dest,
|
||||
const char *errors,
|
||||
const char *details)
|
||||
|
@ -973,12 +991,12 @@ PyObject *PyUnicode_DecodeUTF16(const char *s,
|
|||
{
|
||||
PyUnicodeObject *unicode;
|
||||
Py_UNICODE *p;
|
||||
const Py_UNICODE *q, *e;
|
||||
const Py_UCS2 *q, *e;
|
||||
int bo = 0;
|
||||
const char *errmsg = "";
|
||||
|
||||
/* size should be an even number */
|
||||
if (size % sizeof(Py_UNICODE) != 0) {
|
||||
if (size % sizeof(Py_UCS2) != 0) {
|
||||
if (utf16_decoding_error(NULL, NULL, errors, "truncated data"))
|
||||
return NULL;
|
||||
/* The remaining input chars are ignored if we fall through
|
||||
|
@ -995,8 +1013,8 @@ PyObject *PyUnicode_DecodeUTF16(const char *s,
|
|||
|
||||
/* Unpack UTF-16 encoded data */
|
||||
p = unicode->str;
|
||||
q = (Py_UNICODE *)s;
|
||||
e = q + (size / sizeof(Py_UNICODE));
|
||||
q = (Py_UCS2 *)s;
|
||||
e = q + (size / sizeof(Py_UCS2));
|
||||
|
||||
if (byteorder)
|
||||
bo = *byteorder;
|
||||
|
@ -1026,7 +1044,7 @@ PyObject *PyUnicode_DecodeUTF16(const char *s,
|
|||
}
|
||||
|
||||
while (q < e) {
|
||||
register Py_UNICODE ch = *q++;
|
||||
register Py_UCS2 ch = *q++;
|
||||
|
||||
/* Swap input bytes if needed. (This assumes
|
||||
sizeof(Py_UNICODE) == 2 !) */
|
||||
|
@ -1048,17 +1066,33 @@ PyObject *PyUnicode_DecodeUTF16(const char *s,
|
|||
goto utf16Error;
|
||||
}
|
||||
if (0xDC00 <= *q && *q <= 0xDFFF) {
|
||||
q++;
|
||||
if (0xD800 <= *q && *q <= 0xDBFF) {
|
||||
Py_UCS2 ch2 = *q++;
|
||||
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
|
||||
if (bo == 1)
|
||||
ch = (ch >> 8) | (ch << 8);
|
||||
#else
|
||||
if (bo == -1)
|
||||
ch = (ch >> 8) | (ch << 8);
|
||||
#endif
|
||||
if (0xD800 <= ch && ch <= 0xDBFF) {
|
||||
#if Py_UNICODE_SIZE == 2
|
||||
/* This is valid data (a UTF-16 surrogate pair), but
|
||||
we are not able to store this information since our
|
||||
Py_UNICODE type only has 16 bits... this might
|
||||
change someday, even though it's unlikely. */
|
||||
errmsg = "code pairs are not supported";
|
||||
goto utf16Error;
|
||||
}
|
||||
else
|
||||
#else
|
||||
*p++ = (((ch & 0x3FF)<<10) | (ch2 & 0x3FF)) + 0x10000;
|
||||
continue;
|
||||
#endif
|
||||
|
||||
}
|
||||
else {
|
||||
errmsg = "illegal UTF-16 surrogate";
|
||||
goto utf16Error;
|
||||
}
|
||||
|
||||
}
|
||||
errmsg = "illegal encoding";
|
||||
/* Fall through to report the error */
|
||||
|
@ -1090,17 +1124,20 @@ PyObject *PyUnicode_EncodeUTF16(const Py_UNICODE *s,
|
|||
int byteorder)
|
||||
{
|
||||
PyObject *v;
|
||||
Py_UNICODE *p;
|
||||
Py_UCS2 *p;
|
||||
char *q;
|
||||
int i, pairs, doswap = 1;
|
||||
|
||||
/* We don't create UTF-16 pairs... */
|
||||
for (i = pairs = 0; i < size; i++)
|
||||
if (s[i] >= 0x10000)
|
||||
pairs++;
|
||||
v = PyString_FromStringAndSize(NULL,
|
||||
sizeof(Py_UNICODE) * (size + (byteorder == 0)));
|
||||
sizeof(Py_UCS2) * (size + pairs + (byteorder == 0)));
|
||||
if (v == NULL)
|
||||
return NULL;
|
||||
|
||||
q = PyString_AS_STRING(v);
|
||||
p = (Py_UNICODE *)q;
|
||||
p = (Py_UCS2 *)q;
|
||||
if (byteorder == 0)
|
||||
*p++ = 0xFEFF;
|
||||
if (size == 0)
|
||||
|
@ -1112,12 +1149,24 @@ PyObject *PyUnicode_EncodeUTF16(const Py_UNICODE *s,
|
|||
byteorder == 1
|
||||
#endif
|
||||
)
|
||||
Py_UNICODE_COPY(p, s, size);
|
||||
else
|
||||
while (size-- > 0) {
|
||||
Py_UNICODE ch = *s++;
|
||||
*p++ = (ch >> 8) | (ch << 8);
|
||||
doswap = 0;
|
||||
while (size-- > 0) {
|
||||
Py_UNICODE ch = *s++;
|
||||
Py_UNICODE ch2 = 0;
|
||||
if (ch >= 0x10000) {
|
||||
ch2 = 0xDC00|((ch-0x10000) & 0x3FF);
|
||||
ch = 0xD800|((ch-0x10000)>>10);
|
||||
}
|
||||
if (doswap){
|
||||
*p++ = (ch >> 8) | (ch << 8);
|
||||
if (ch2)
|
||||
*p++ = (ch2 >> 8) | (ch2 << 8);
|
||||
}else{
|
||||
*p++ = ch;
|
||||
if(ch2)
|
||||
*p++ = ch2;
|
||||
}
|
||||
}
|
||||
return v;
|
||||
}
|
||||
|
||||
|
@ -1271,10 +1320,14 @@ PyObject *PyUnicode_DecodeUnicodeEscape(const char *s,
|
|||
/* UCS-2 character */
|
||||
*p++ = (Py_UNICODE) chr;
|
||||
else if (chr <= 0x10ffff) {
|
||||
/* UCS-4 character. store as two surrogate characters */
|
||||
/* UCS-4 character. Either store directly, or as surrogate pair. */
|
||||
#if Py_UNICODE_SIZE == 4
|
||||
*p++ = chr;
|
||||
#else
|
||||
chr -= 0x10000L;
|
||||
*p++ = 0xD800 + (Py_UNICODE) (chr >> 10);
|
||||
*p++ = 0xDC00 + (Py_UNICODE) (chr & 0x03FF);
|
||||
#endif
|
||||
} else {
|
||||
if (unicodeescape_decoding_error(
|
||||
&s, &x, errors,
|
||||
|
@ -1383,6 +1436,19 @@ PyObject *unicodeescape_string(const Py_UNICODE *s,
|
|||
*p++ = '\\';
|
||||
*p++ = (char) ch;
|
||||
}
|
||||
/* Map 21-bit characters to '\U00xxxxxx' */
|
||||
else if (ch >= 0x10000) {
|
||||
*p++ = '\\';
|
||||
*p++ = 'U';
|
||||
*p++ = hexdigit[(ch >> 28) & 0xf];
|
||||
*p++ = hexdigit[(ch >> 24) & 0xf];
|
||||
*p++ = hexdigit[(ch >> 20) & 0xf];
|
||||
*p++ = hexdigit[(ch >> 16) & 0xf];
|
||||
*p++ = hexdigit[(ch >> 12) & 0xf];
|
||||
*p++ = hexdigit[(ch >> 8) & 0xf];
|
||||
*p++ = hexdigit[(ch >> 4) & 0xf];
|
||||
*p++ = hexdigit[ch & 15];
|
||||
}
|
||||
/* Map 16-bit characters to '\uxxxx' */
|
||||
else if (ch >= 256) {
|
||||
*p++ = '\\';
|
||||
|
@ -5281,13 +5347,6 @@ void _PyUnicode_Init(void)
|
|||
{
|
||||
int i;
|
||||
|
||||
/* Doublecheck the configuration... */
|
||||
#ifndef USE_UCS4_STORAGE
|
||||
if (sizeof(Py_UNICODE) != 2)
|
||||
Py_FatalError("Unicode configuration error: "
|
||||
"sizeof(Py_UNICODE) != 2 bytes");
|
||||
#endif
|
||||
|
||||
/* Init the implementation */
|
||||
unicode_freelist = NULL;
|
||||
unicode_freelist_size = 0;
|
||||
|
|
|
@ -324,12 +324,16 @@ builtin_unichr(PyObject *self, PyObject *args)
|
|||
s[0] = (Py_UNICODE) x;
|
||||
return PyUnicode_FromUnicode(s, 1);
|
||||
} else {
|
||||
#if Py_UNICODE_SIZE == 2
|
||||
/* UCS-4 character. store as two surrogate characters */
|
||||
x -= 0x10000L;
|
||||
s[0] = 0xD800 + (Py_UNICODE) (x >> 10);
|
||||
s[1] = 0xDC00 + (Py_UNICODE) (x & 0x03FF);
|
||||
return PyUnicode_FromUnicode(s, 2);
|
||||
#endif
|
||||
}
|
||||
s[0] = (Py_UNICODE)x;
|
||||
return PyUnicode_FromUnicode(s, 1);
|
||||
}
|
||||
|
||||
static char unichr_doc[] =
|
||||
|
|
|
@ -104,6 +104,15 @@
|
|||
/* Define if the compiler provides a wchar.h header file. */
|
||||
#undef HAVE_WCHAR_H
|
||||
|
||||
/* Define if you want to have a Unicode type. */
|
||||
#undef Py_USING_UNICODE
|
||||
|
||||
/* Define as the integral type used for Unicode representation. */
|
||||
#undef PY_UNICODE_TYPE
|
||||
|
||||
/* Define as the size of the unicode type. */
|
||||
#undef Py_UNICODE_SIZE
|
||||
|
||||
/* Define if malloc(0) returns a NULL pointer */
|
||||
#undef MALLOC_ZERO_RETURNS_NULL
|
||||
|
||||
|
|
12
config.h.in
12
config.h.in
|
@ -163,6 +163,15 @@
|
|||
/* Define if the compiler provides a wchar.h header file. */
|
||||
#undef HAVE_WCHAR_H
|
||||
|
||||
/* Define if you want to have a Unicode type. */
|
||||
#undef Py_USING_UNICODE
|
||||
|
||||
/* Define as the integral type used for Unicode representation. */
|
||||
#undef PY_UNICODE_TYPE
|
||||
|
||||
/* Define as the size of the unicode type. */
|
||||
#undef Py_UNICODE_SIZE
|
||||
|
||||
/* Define if malloc(0) returns a NULL pointer */
|
||||
#undef MALLOC_ZERO_RETURNS_NULL
|
||||
|
||||
|
@ -284,6 +293,9 @@
|
|||
/* The number of bytes in a void *. */
|
||||
#undef SIZEOF_VOID_P
|
||||
|
||||
/* The number of bytes in a wchar_t. */
|
||||
#undef SIZEOF_WCHAR_T
|
||||
|
||||
/* Define if you have the _getpty function. */
|
||||
#undef HAVE__GETPTY
|
||||
|
||||
|
|
73
configure.in
73
configure.in
|
@ -372,8 +372,8 @@ fi
|
|||
OPT="$OPT -Dss_family=__ss_family -Dss_len=__ss_len"
|
||||
AC_MSG_CHECKING([whether to enable ipv6])
|
||||
AC_ARG_ENABLE(ipv6,
|
||||
[ --enable-ipv6 Enable ipv6 (with ipv4) support
|
||||
--disable-ipv6 Disable ipv6 support],
|
||||
[ --enable-ipv6 Enable ipv6 (with ipv4) support
|
||||
--disable-ipv6 Disable ipv6 support],
|
||||
[ case "$enableval" in
|
||||
no)
|
||||
AC_MSG_RESULT(no)
|
||||
|
@ -1578,23 +1578,58 @@ AC_DEFINE(HAVE_WCHAR_H) wchar_h="yes",
|
|||
wchar_h="no"
|
||||
)
|
||||
|
||||
# check for usable wchar_t
|
||||
usable_wchar_t="unkown"
|
||||
AC_MSG_CHECKING(for usable wchar_t)
|
||||
AC_TRY_RUN([
|
||||
#include "wchar.h"
|
||||
#include "wctype.h"
|
||||
main() {
|
||||
wchar_t s;
|
||||
if (sizeof(s) == 2)
|
||||
exit(0);
|
||||
else
|
||||
exit(1);
|
||||
}
|
||||
],
|
||||
AC_DEFINE(HAVE_USABLE_WCHAR_T) usable_wchar_t="yes",
|
||||
usable_wchar_t="no")
|
||||
AC_MSG_RESULT($usable_wchar_t)
|
||||
# determine wchar_t size
|
||||
if test "$wchar_h" = yes
|
||||
then
|
||||
AC_CHECK_SIZEOF(wchar_t)
|
||||
fi
|
||||
|
||||
AC_MSG_CHECKING(what type to use for unicode)
|
||||
AC_ARG_ENABLE(unicode,
|
||||
[ --enable-unicode[=ucs2,ucs4] Enable Unicode strings (default is yes)],,enable_unicode=yes)
|
||||
|
||||
if test $enable_unicode = yes
|
||||
then
|
||||
# Let Py_UNICODE size depend on wchar_t size
|
||||
case "$ac_cv_sizeof_wchar_t" in
|
||||
2) enable_unicode="ucs2";;
|
||||
4) enable_unicode="ucs4";;
|
||||
*) enable_unicode="ucs4";; # default to UCS-4
|
||||
esac
|
||||
fi
|
||||
|
||||
case "$enable_unicode" in
|
||||
ucs2) unicode_size="2"
|
||||
AC_DEFINE(Py_UNICODE_SIZE,2)
|
||||
;;
|
||||
ucs4) unicode_size="4"
|
||||
AC_DEFINE(Py_UNICODE_SIZE,4)
|
||||
;;
|
||||
esac
|
||||
|
||||
if test "$enable_unicode" = "no"
|
||||
then
|
||||
AC_MSG_RESULT(not used)
|
||||
else
|
||||
AC_DEFINE(Py_USING_UNICODE)
|
||||
if test "$unicode_size" = "$ac_cv_sizeof_wchar_t"
|
||||
then
|
||||
PY_UNICODE_TYPE="wchar_t"
|
||||
AC_DEFINE(HAVE_USABLE_WCHAR_T)
|
||||
AC_DEFINE(PY_UNICODE_TYPE,wchar_t)
|
||||
elif test "$ac_cv_sizeof_short" = "$unicode_size"
|
||||
then
|
||||
PY_UNICODE_TYPE="unsigned short"
|
||||
AC_DEFINE(PY_UNICODE_TYPE,unsigned short)
|
||||
elif test "$ac_cv_sizeof_long" = "$unicode_size"
|
||||
then
|
||||
PY_UNICODE_TYPE="unsigned long"
|
||||
AC_DEFINE(PY_UNICODE_TYPE,unsigned long)
|
||||
else
|
||||
PY_UNICODE_TYPE="no type found"
|
||||
fi
|
||||
AC_MSG_RESULT($PY_UNICODE_TYPE)
|
||||
fi
|
||||
|
||||
# check for endianness
|
||||
AC_C_BIGENDIAN
|
||||
|
|
Loading…
Reference in a new issue