mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-09-26 18:29:57 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 10

Implement the changes proposed in patch #413333. unicode(obj) now

Browse source 
works just like str(obj) in that it tries __str__/tp_str on the object
in case it finds that the object is not a string or buffer.
This commit is contained in:
Marc-André Lemburg 2001-09-20 12:53:16 +00:00
parent c60e6f7771
commit 6871f6ac57
2 changed files with 61 additions and 43 deletions
Download patch file Download diff file Expand all files Collapse all files

5
Lib/test/test_unicode.py
View file

@ -429,6 +429,7 @@ verify(unicode('hello','utf-8') == u'hello')
verify(unicode('hello','utf8') == u'hello') verify(unicode('hello','utf8') == u'hello')
verify(unicode('hello','latin-1') == u'hello') verify(unicode('hello','latin-1') == u'hello')
# Compatibility to str():
class String: class String:
x = '' x = ''
def __str__(self): def __str__(self):
@ -444,6 +445,10 @@ o.x = u'abc'
verify(unicode(o) == u'abc') verify(unicode(o) == u'abc')
verify(str(o) == 'abc') verify(str(o) == 'abc')
for obj in (123, 123.45, 123L):
verify(unicode(obj) == unicode(str(obj)))
# Error handling
try: try:
u'Andr\202 x'.encode('ascii') u'Andr\202 x'.encode('ascii')
u'Andr\202 x'.encode('ascii','strict') u'Andr\202 x'.encode('ascii','strict')

57
Objects/unicodeobject.c
View file

@ -398,10 +398,11 @@ PyObject *PyUnicode_FromEncodedObject(register PyObject *obj,
const char *encoding, const char *encoding,
const char *errors) const char *errors)
{ {
const char *s; const char *s = NULL;
int len; int len;
int owned = 0; int owned = 0;
PyObject *v; PyObject *v;
int reclevel;
if (obj == NULL) { if (obj == NULL) {
PyErr_BadInternalCall(); PyErr_BadInternalCall();
@ -409,25 +410,13 @@ PyObject *PyUnicode_FromEncodedObject(register PyObject *obj,
} }
/* Coerce object */ /* Coerce object */
if (PyInstance_Check(obj)) { for (reclevel = 0; reclevel < 2; reclevel++) {
PyObject *func;
func = PyObject_GetAttrString(obj, "__str__");
if (func == NULL) {
PyErr_SetString(PyExc_TypeError,
"coercing to Unicode: instance doesn't define __str__");
return NULL;
}
obj = PyEval_CallObject(func, NULL);
Py_DECREF(func);
if (obj == NULL)
return NULL;
owned = 1;
}
if (PyUnicode_Check(obj)) { if (PyUnicode_Check(obj)) {
if (encoding) { if (encoding) {
PyErr_SetString(PyExc_TypeError, PyErr_SetString(PyExc_TypeError,
"decoding Unicode is not supported"); "decoding Unicode is not supported");
return NULL; goto onError;
} }
if (PyUnicode_CheckExact(obj)) { if (PyUnicode_CheckExact(obj)) {
Py_INCREF(obj); Py_INCREF(obj);
@ -444,14 +433,38 @@ PyObject *PyUnicode_FromEncodedObject(register PyObject *obj,
else if (PyString_Check(obj)) { else if (PyString_Check(obj)) {
s = PyString_AS_STRING(obj); s = PyString_AS_STRING(obj);
len = PyString_GET_SIZE(obj); len = PyString_GET_SIZE(obj);
break;
} }
else if (PyObject_AsCharBuffer(obj, &s, &len)) { else {
/* Overwrite the error message with something more useful in PyObject *w;
case of a TypeError. */
if (PyErr_ExceptionMatches(PyExc_TypeError)) /* Try char buffer interface */
if (PyObject_AsCharBuffer(obj, &s, &len))
PyErr_Clear();
else
break;
/* Mimic the behaviour of str(object) if everything else
fails (see PyObject_Str()); this also covers instances
which implement __str__. */
if (obj->ob_type->tp_str == NULL)
w = PyObject_Repr(obj);
else
w = (*obj->ob_type->tp_str)(obj);
if (w == NULL)
goto onError;
if (owned) {
Py_DECREF(obj);
}
obj = w;
owned = 1;
}
}
if (s == NULL) {
PyErr_Format(PyExc_TypeError, PyErr_Format(PyExc_TypeError,
"coercing to Unicode: need string or buffer, " "coercing to Unicode: __str__ recursion limit exceeded "
"%.80s found", "(last type: %.80s)",
obj->ob_type->tp_name); obj->ob_type->tp_name);
goto onError; goto onError;
} }