Many changes for Unicode, by Marc-Andre Lemburg.

Objects/abstract.c

@@ -199,6 +199,114 @@ PyObject_DelItem(o, key)
 	return -1;
 }
 
+int PyObject_AsCharBuffer(PyObject *obj,
+			  const char **buffer,
+			  int *buffer_len)
+{
+	PyBufferProcs *pb;
+	const char *pp;
+	int len;
+
+	if (obj == NULL || buffer == NULL || buffer_len == NULL) {
+		null_error();
+		return -1;
+	}
+	pb = obj->ob_type->tp_as_buffer;
+	if ( pb == NULL ||
+	     pb->bf_getcharbuffer == NULL ||
+	     pb->bf_getsegcount == NULL ) {
+		PyErr_SetString(PyExc_TypeError,
+				"expected a character buffer object");
+		goto onError;
+	}
+	if ( (*pb->bf_getsegcount)(obj,NULL) != 1 ) {
+		PyErr_SetString(PyExc_TypeError,
+				"expected a single-segment buffer object");
+		goto onError;
+	}
+	len = (*pb->bf_getcharbuffer)(obj,0,&pp);
+	if (len < 0)
+		goto onError;
+	*buffer = pp;
+	*buffer_len = len;
+	return 0;
+
+ onError:
+	return -1;
+}
+
+int PyObject_AsReadBuffer(PyObject *obj,
+			  const void **buffer,
+			  int *buffer_len)
+{
+	PyBufferProcs *pb;
+	void *pp;
+	int len;
+
+	if (obj == NULL || buffer == NULL || buffer_len == NULL) {
+		null_error();
+		return -1;
+	}
+	pb = obj->ob_type->tp_as_buffer;
+	if ( pb == NULL ||
+	     pb->bf_getreadbuffer == NULL ||
+	     pb->bf_getsegcount == NULL ) {
+		PyErr_SetString(PyExc_TypeError,
+				"expected a readable buffer object");
+		goto onError;
+	}
+	if ( (*pb->bf_getsegcount)(obj,NULL) != 1 ) {
+		PyErr_SetString(PyExc_TypeError,
+				"expected a single-segment buffer object");
+		goto onError;
+	}
+	len = (*pb->bf_getreadbuffer)(obj,0,&pp);
+	if (len < 0)
+		goto onError;
+	*buffer = pp;
+	*buffer_len = len;
+	return 0;
+
+ onError:
+	return -1;
+}
+
+int PyObject_AsWriteBuffer(PyObject *obj,
+			   void **buffer,
+			   int *buffer_len)
+{
+	PyBufferProcs *pb;
+	void*pp;
+	int len;
+
+	if (obj == NULL || buffer == NULL || buffer_len == NULL) {
+		null_error();
+		return -1;
+	}
+	pb = obj->ob_type->tp_as_buffer;
+	if ( pb == NULL ||
+	     pb->bf_getwritebuffer == NULL ||
+	     pb->bf_getsegcount == NULL ) {
+		PyErr_SetString(PyExc_TypeError,
+				"expected a writeable buffer object");
+		goto onError;
+	}
+	if ( (*pb->bf_getsegcount)(obj,NULL) != 1 ) {
+		PyErr_SetString(PyExc_TypeError,
+				"expected a single-segment buffer object");
+		goto onError;
+	}
+	len = (*pb->bf_getwritebuffer)(obj,0,&pp);
+	if (len < 0)
+		goto onError;
+	*buffer = pp;
+	*buffer_len = len;
+	return 0;
+
+ onError:
+	return -1;
+}
+
 /* Operations on numbers */
 
 int
@@ -447,6 +555,8 @@ PyNumber_Remainder(v, w)
 {
 	if (PyString_Check(v))
 		return PyString_Format(v, w);
+	else if (PyUnicode_Check(v))
+		return PyUnicode_Format(v, w);
 	BINOP(v, w, "__mod__", "__rmod__", PyNumber_Remainder);
 	if (v->ob_type->tp_as_number != NULL) {
 		PyObject *x = NULL;
@@ -621,6 +731,8 @@ PyNumber_Int(o)
 	PyObject *o;
 {
 	PyNumberMethods *m;
+	const char *buffer;
+	int buffer_len;
 
 	if (o == NULL)
 		return null_error();
@@ -629,6 +741,8 @@ PyNumber_Int(o)
 	m = o->ob_type->tp_as_number;
 	if (m && m->nb_int)
 		return m->nb_int(o);
+	if (!PyObject_AsCharBuffer(o, &buffer, &buffer_len))
+		return PyInt_FromString((char*)buffer, NULL, 10);
 
 	return type_error("object can't be converted to int");
 }
@@ -655,17 +769,19 @@ PyNumber_Int(o)
  * calls PyLong_FromString().
  */
 static PyObject *
-long_from_string(v)
-	PyObject *v;
+long_from_string(s, len)
+	const char *s;
+	int len;
 {
-	char *s, *end;
+	const char *start;
+	char *end;
 	PyObject *x;
 	char buffer[256]; /* For errors */
 
-	s = PyString_AS_STRING(v);
+	start = s;
 	while (*s && isspace(Py_CHARMASK(*s)))
 		s++;
-	x = PyLong_FromString(s, &end, 10);
+	x = PyLong_FromString((char*)s, &end, 10);
 	if (x == NULL) {
 		if (PyErr_ExceptionMatches(PyExc_ValueError))
 			goto bad;
@@ -680,7 +796,7 @@ long_from_string(v)
 		Py_XDECREF(x);
 		return NULL;
 	}
-	else if (end != PyString_AS_STRING(v) + PyString_GET_SIZE(v)) {
+	else if (end != start + len) {
 		PyErr_SetString(PyExc_ValueError,
 				"null byte in argument for long()");
 		return NULL;
@@ -693,6 +809,8 @@ PyNumber_Long(o)
 	PyObject *o;
 {
 	PyNumberMethods *m;
+	const char *buffer;
+	int buffer_len;
 
 	if (o == NULL)
 		return null_error();
@@ -701,10 +819,13 @@ PyNumber_Long(o)
 		 * doesn't do.  In particular long('9.5') must raise an
 		 * exception, not truncate the float.
 		 */
-		return long_from_string(o);
+		return long_from_string(PyString_AS_STRING(o),
+					PyString_GET_SIZE(o));
 	m = o->ob_type->tp_as_number;
 	if (m && m->nb_long)
 		return m->nb_long(o);
+	if (!PyObject_AsCharBuffer(o, &buffer, &buffer_len))
+		return long_from_string(buffer, buffer_len);
 
 	return type_error("object can't be converted to long");
 }
@@ -717,13 +838,12 @@ PyNumber_Float(o)
 
 	if (o == NULL)
 		return null_error();
-	if (PyString_Check(o))
-		return PyFloat_FromString(o, NULL);
-	m = o->ob_type->tp_as_number;
-	if (m && m->nb_float)
-		return m->nb_float(o);
-
-	return type_error("object can't be converted to float");
+	if (!PyString_Check(o)) {
+		m = o->ob_type->tp_as_number;
+		if (m && m->nb_float)
+			return m->nb_float(o);
+	}
+	return PyFloat_FromString(o, NULL);
 }
 
 /* Operations on sequences */

Objects/fileobject.c

@@ -75,6 +75,8 @@ typedef struct {
 	PyObject *f_mode;
 	int (*f_close) Py_PROTO((FILE *));
 	int f_softspace; /* Flag used by 'print' command */
+	int f_binary; /* Flag which indicates whether the file is open
+			 open in binary (1) or test (0) mode */
 } PyFileObject;
 
 FILE *
@@ -112,6 +114,10 @@ PyFile_FromFile(fp, name, mode, close)
 	f->f_mode = PyString_FromString(mode);
 	f->f_close = close;
 	f->f_softspace = 0;
+	if (strchr(mode,'b') != NULL)
+	    f->f_binary = 1;
+	else
+	    f->f_binary = 0;
 	if (f->f_name == NULL || f->f_mode == NULL) {
 		Py_DECREF(f);
 		return NULL;
@@ -863,7 +869,7 @@ file_write(f, args)
 	int n, n2;
 	if (f->f_fp == NULL)
 		return err_closed();
-	if (!PyArg_Parse(args, "s#", &s, &n))
+	if (!PyArg_Parse(args, f->f_binary ? "s#" : "t#", &s, &n))
 		return NULL;
 	f->f_softspace = 0;
 	Py_BEGIN_ALLOW_THREADS

Objects/floatobject.c

@@ -155,15 +155,22 @@ PyFloat_FromString(v, pend)
 	char **pend;
 {
 	extern double strtod Py_PROTO((const char *, char **));
-	char *s, *last, *end;
+	const char *s, *last, *end;
 	double x;
 	char buffer[256]; /* For errors */
+	int len;
 
-	if (!PyString_Check(v))
+	if (PyString_Check(v)) {
+		s = PyString_AS_STRING(v);
+		len = PyString_GET_SIZE(v);
+	}
+	else if (PyObject_AsCharBuffer(v, &s, &len)) {
+		PyErr_SetString(PyExc_TypeError,
+				"float() needs a string argument");
 		return NULL;
-	s = PyString_AS_STRING(v);
+	}
 
-	last = s + PyString_GET_SIZE(v);
+	last = s + len;
 	while (*s && isspace(Py_CHARMASK(*s)))
 		s++;
 	if (s[0] == '\0') {
@@ -172,7 +179,7 @@ PyFloat_FromString(v, pend)
 	}
 	errno = 0;
 	PyFPE_START_PROTECT("PyFloat_FromString", return 0)
-	x = strtod(s, &end);
+	x = strtod((char *)s, (char **)&end);
 	PyFPE_END_PROTECT(x)
 	/* Believe it or not, Solaris 2.6 can move end *beyond* the null
 	   byte at the end of the string, when the input is inf(inity) */
@@ -185,7 +192,7 @@ PyFloat_FromString(v, pend)
 		PyErr_SetString(PyExc_ValueError, buffer);
 		return NULL;
 	}
-	else if (end != PyString_AS_STRING(v) + PyString_GET_SIZE(v)) {
+	else if (end != last) {
 		PyErr_SetString(PyExc_ValueError,
 				"null byte in argument for float()");
 		return NULL;
@@ -196,7 +203,7 @@ PyFloat_FromString(v, pend)
 		return NULL;
 	}
 	if (pend)
-		*pend = end;
+		*pend = (char *)end;
 	return PyFloat_FromDouble(x);
 }
 
@@ -785,7 +792,7 @@ PyFloat_Fini()
 					PyFloat_AsString(buf, p);
 					fprintf(stderr,
 			     "#   <float at %lx, refcnt=%d, val=%s>\n",
-						p, p->ob_refcnt, buf);
+						(long)p, p->ob_refcnt, buf);
 				}
 			}
 			list = list->next;

Objects/intobject.c

@@ -942,7 +942,7 @@ PyInt_Fini()
 				if (PyInt_Check(p) && p->ob_refcnt != 0)
 					fprintf(stderr,
 				"#   <int at %lx, refcnt=%d, val=%ld>\n",
-						p, p->ob_refcnt, p->ob_ival);
+					  (long)p, p->ob_refcnt, p->ob_ival);
 			}
 			list = list->next;
 		}

Objects/object.c

@@ -189,11 +189,6 @@ PyObject_Print(op, fp, flags)
 					s = PyObject_Repr(op);
 				if (s == NULL)
 					ret = -1;
-				else if (!PyString_Check(s)) {
-					PyErr_SetString(PyExc_TypeError,
-						   "repr not string");
-					ret = -1;
-				}
 				else {
 					ret = PyObject_Print(s, fp,
 							     Py_PRINT_RAW);
@@ -234,14 +229,28 @@ PyObject_Repr(v)
 			v->ob_type->tp_name, (long)v);
 		return PyString_FromString(buf);
 	}
-	else
-		return (*v->ob_type->tp_repr)(v);
+	else {
+		PyObject *res;
+		res = (*v->ob_type->tp_repr)(v);
+		if (res == NULL)
+			return NULL;
+		if (!PyString_Check(res)) {
+			PyErr_Format(PyExc_TypeError,
+				     "__repr__ returned non-string (type %s)",
+				     res->ob_type->tp_name);
+			Py_DECREF(res);
+			return NULL;
+		}
+		return res;
+	}
 }
 
 PyObject *
 PyObject_Str(v)
 	PyObject *v;
 {
+	PyObject *res;
+	
 	if (v == NULL)
 		return PyString_FromString("<NULL>");
 	else if (PyString_Check(v)) {
@@ -249,10 +258,9 @@ PyObject_Str(v)
 		return v;
 	}
 	else if (v->ob_type->tp_str != NULL)
-		return (*v->ob_type->tp_str)(v);
+		res = (*v->ob_type->tp_str)(v);
 	else {
 		PyObject *func;
-		PyObject *res;
 		if (!PyInstance_Check(v) ||
 		    (func = PyObject_GetAttrString(v, "__str__")) == NULL) {
 			PyErr_Clear();
@@ -260,8 +268,17 @@ PyObject_Str(v)
 		}
 		res = PyEval_CallObject(func, (PyObject *)NULL);
 		Py_DECREF(func);
-		return res;
 	}
+	if (res == NULL)
+		return NULL;
+	if (!PyString_Check(res)) {
+		PyErr_Format(PyExc_TypeError,
+			     "__str__ returned non-string (type %s)",
+			     res->ob_type->tp_name);
+		Py_DECREF(res);
+		return NULL;
+	}
+	return res;
 }
 
 static PyObject *
@@ -330,6 +347,8 @@ PyObject_Compare(v, w)
 				return cmp;
 			}
 		}
+		else if (PyUnicode_Check(v) || PyUnicode_Check(w))
+			return PyUnicode_Compare(v, w);
 		else if (vtp->tp_as_number != NULL)
 			vname = "";
 		else if (wtp->tp_as_number != NULL)
@@ -652,7 +671,7 @@ _Py_ForgetReference(op)
 	register PyObject *op;
 {
 #ifdef SLOW_UNREF_CHECK
-	register PyObject *p;
+        register PyObject *p;
 #endif
 	if (op->ob_refcnt < 0)
 		Py_FatalError("UNREF negative refcnt");