Add C API PyUnicode_FromOrdinal() which exposes unichr() at C level.

u'%c' will now raise a ValueError in case the argument is an
integer outside the valid range of Unicode code point ordinals.

Closes SF bug #593581.
This commit is contained in:
Marc-André Lemburg 2002-08-11 12:23:04 +00:00
parent 078151da90
commit cc8764ca9d
5 changed files with 82 additions and 35 deletions

View File

@ -517,6 +517,18 @@ extern DL_IMPORT(int) PyUnicode_AsWideChar(
#endif
/* --- Unicode ordinals --------------------------------------------------- */
/* Create a Unicode Object from the given Unicode code point ordinal.
The ordinal must be in range(0x10000) on narrow Python builds
(UCS2), and range(0x110000) on wide builds (UCS4). A ValueError is
raised in case it is not.
*/
extern DL_IMPORT(PyObject*) PyUnicode_FromOrdinal(int ordinal);
/* === Builtin Codecs =====================================================
Many of these APIs take two arguments encoding and errors. These

View File

@ -453,6 +453,14 @@ except KeyError:
else:
verify(value == u'abc, def')
for ordinal in (-100, 0x20000):
try:
u"%c" % ordinal
except ValueError:
pass
else:
print '*** formatting u"%%c" % %i should give a ValueError' % ordinal
# formatting jobs delegated from the string implementation:
verify('...%(foo)s...' % {'foo':u"abc"} == u'...abc...')
verify('...%(foo)s...' % {'foo':"abc"} == '...abc...')

View File

@ -6,6 +6,9 @@ Type/class unification and new-style classes
Core and builtins
- u'%c' will now raise a ValueError in case the argument is an
integer outside the valid range of Unicode code point ordinals.
- The tempfile module has been overhauled for enhanced security. The
mktemp() function is now deprecated; new, safe replacements are
mkstemp() (for files) and mkdtemp() (for directories), and the
@ -437,6 +440,9 @@ Build
C API
- New C API PyUnicode_FromOrdinal() which exposes unichr() at C
level.
- New functions PyErr_SetExcFromWindowsErr() and
PyErr_SetExcFromWindowsErrWithFilename(). Similar to
PyErr_SetFromWindowsErrWithFilename() and

View File

@ -390,6 +390,45 @@ int PyUnicode_AsWideChar(PyUnicodeObject *unicode,
#endif
PyObject *PyUnicode_FromOrdinal(int ordinal)
{
Py_UNICODE s[2];
#ifdef Py_UNICODE_WIDE
if (ordinal < 0 || ordinal > 0x10ffff) {
PyErr_SetString(PyExc_ValueError,
"unichr() arg not in range(0x110000) "
"(wide Python build)");
return NULL;
}
#else
if (ordinal < 0 || ordinal > 0xffff) {
PyErr_SetString(PyExc_ValueError,
"unichr() arg not in range(0x10000) "
"(narrow Python build)");
return NULL;
}
#endif
if (ordinal <= 0xffff) {
/* UCS-2 character */
s[0] = (Py_UNICODE) ordinal;
return PyUnicode_FromUnicode(s, 1);
}
else {
#ifndef Py_UNICODE_WIDE
/* UCS-4 character. store as two surrogate characters */
ordinal -= 0x10000L;
s[0] = 0xD800 + (Py_UNICODE) (ordinal >> 10);
s[1] = 0xDC00 + (Py_UNICODE) (ordinal & 0x03FF);
return PyUnicode_FromUnicode(s, 2);
#else
s[0] = (Py_UNICODE)ordinal;
return PyUnicode_FromUnicode(s, 1);
#endif
}
}
PyObject *PyUnicode_FromObject(register PyObject *obj)
{
/* XXX Perhaps we should make this API an alias of
@ -5373,7 +5412,22 @@ formatchar(Py_UNICODE *buf,
x = PyInt_AsLong(v);
if (x == -1 && PyErr_Occurred())
goto onError;
buf[0] = (char) x;
#ifdef Py_UNICODE_WIDE
if (x < 0 || x > 0x10ffff) {
PyErr_SetString(PyExc_ValueError,
"%c arg not in range(0x110000) "
"(wide Python build)");
return -1;
}
#else
if (x < 0 || x > 0xffff) {
PyErr_SetString(PyExc_ValueError,
"%c arg not in range(0x10000) "
"(narrow Python build)");
return -1;
}
#endif
buf[0] = (Py_UNICODE) x;
}
buf[1] = '\0';
return 1;

View File

@ -260,44 +260,11 @@ static PyObject *
builtin_unichr(PyObject *self, PyObject *args)
{
long x;
Py_UNICODE s[2];
if (!PyArg_ParseTuple(args, "l:unichr", &x))
return NULL;
#ifdef Py_UNICODE_WIDE
if (x < 0 || x > 0x10ffff) {
PyErr_SetString(PyExc_ValueError,
"unichr() arg not in range(0x110000) "
"(wide Python build)");
return NULL;
}
#else
if (x < 0 || x > 0xffff) {
PyErr_SetString(PyExc_ValueError,
"unichr() arg not in range(0x10000) "
"(narrow Python build)");
return NULL;
}
#endif
if (x <= 0xffff) {
/* UCS-2 character */
s[0] = (Py_UNICODE) x;
return PyUnicode_FromUnicode(s, 1);
}
else {
#ifndef Py_UNICODE_WIDE
/* 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);
#else
s[0] = (Py_UNICODE)x;
return PyUnicode_FromUnicode(s, 1);
#endif
}
return PyUnicode_FromOrdinal(x);
}
PyDoc_STRVAR(unichr_doc,