1.. highlightlang:: c
2
3.. _number:
4
5Number Protocol
6===============
7
8
9.. c:function:: int PyNumber_Check(PyObject *o)
10
11   Returns ``1`` if the object *o* provides numeric protocols, and false otherwise.
12   This function always succeeds.
13
14
15.. c:function:: PyObject* PyNumber_Add(PyObject *o1, PyObject *o2)
16
17   Returns the result of adding *o1* and *o2*, or *NULL* on failure.  This is the
18   equivalent of the Python expression ``o1 + o2``.
19
20
21.. c:function:: PyObject* PyNumber_Subtract(PyObject *o1, PyObject *o2)
22
23   Returns the result of subtracting *o2* from *o1*, or *NULL* on failure.  This is
24   the equivalent of the Python expression ``o1 - o2``.
25
26
27.. c:function:: PyObject* PyNumber_Multiply(PyObject *o1, PyObject *o2)
28
29   Returns the result of multiplying *o1* and *o2*, or *NULL* on failure.  This is
30   the equivalent of the Python expression ``o1 * o2``.
31
32
33.. c:function:: PyObject* PyNumber_Divide(PyObject *o1, PyObject *o2)
34
35   Returns the result of dividing *o1* by *o2*, or *NULL* on failure.  This is the
36   equivalent of the Python expression ``o1 / o2``.
37
38
39.. c:function:: PyObject* PyNumber_FloorDivide(PyObject *o1, PyObject *o2)
40
41   Return the floor of *o1* divided by *o2*, or *NULL* on failure.  This is
42   equivalent to the "classic" division of integers.
43
44   .. versionadded:: 2.2
45
46
47.. c:function:: PyObject* PyNumber_TrueDivide(PyObject *o1, PyObject *o2)
48
49   Return a reasonable approximation for the mathematical value of *o1* divided by
50   *o2*, or *NULL* on failure.  The return value is "approximate" because binary
51   floating point numbers are approximate; it is not possible to represent all real
52   numbers in base two.  This function can return a floating point value when
53   passed two integers.
54
55   .. versionadded:: 2.2
56
57
58.. c:function:: PyObject* PyNumber_Remainder(PyObject *o1, PyObject *o2)
59
60   Returns the remainder of dividing *o1* by *o2*, or *NULL* on failure.  This is
61   the equivalent of the Python expression ``o1 % o2``.
62
63
64.. c:function:: PyObject* PyNumber_Divmod(PyObject *o1, PyObject *o2)
65
66   .. index:: builtin: divmod
67
68   See the built-in function :func:`divmod`. Returns *NULL* on failure.  This is
69   the equivalent of the Python expression ``divmod(o1, o2)``.
70
71
72.. c:function:: PyObject* PyNumber_Power(PyObject *o1, PyObject *o2, PyObject *o3)
73
74   .. index:: builtin: pow
75
76   See the built-in function :func:`pow`. Returns *NULL* on failure.  This is the
77   equivalent of the Python expression ``pow(o1, o2, o3)``, where *o3* is optional.
78   If *o3* is to be ignored, pass :c:data:`Py_None` in its place (passing *NULL* for
79   *o3* would cause an illegal memory access).
80
81
82.. c:function:: PyObject* PyNumber_Negative(PyObject *o)
83
84   Returns the negation of *o* on success, or *NULL* on failure. This is the
85   equivalent of the Python expression ``-o``.
86
87
88.. c:function:: PyObject* PyNumber_Positive(PyObject *o)
89
90   Returns *o* on success, or *NULL* on failure.  This is the equivalent of the
91   Python expression ``+o``.
92
93
94.. c:function:: PyObject* PyNumber_Absolute(PyObject *o)
95
96   .. index:: builtin: abs
97
98   Returns the absolute value of *o*, or *NULL* on failure.  This is the equivalent
99   of the Python expression ``abs(o)``.
100
101
102.. c:function:: PyObject* PyNumber_Invert(PyObject *o)
103
104   Returns the bitwise negation of *o* on success, or *NULL* on failure.  This is
105   the equivalent of the Python expression ``~o``.
106
107
108.. c:function:: PyObject* PyNumber_Lshift(PyObject *o1, PyObject *o2)
109
110   Returns the result of left shifting *o1* by *o2* on success, or *NULL* on
111   failure.  This is the equivalent of the Python expression ``o1 << o2``.
112
113
114.. c:function:: PyObject* PyNumber_Rshift(PyObject *o1, PyObject *o2)
115
116   Returns the result of right shifting *o1* by *o2* on success, or *NULL* on
117   failure.  This is the equivalent of the Python expression ``o1 >> o2``.
118
119
120.. c:function:: PyObject* PyNumber_And(PyObject *o1, PyObject *o2)
121
122   Returns the "bitwise and" of *o1* and *o2* on success and *NULL* on failure.
123   This is the equivalent of the Python expression ``o1 & o2``.
124
125
126.. c:function:: PyObject* PyNumber_Xor(PyObject *o1, PyObject *o2)
127
128   Returns the "bitwise exclusive or" of *o1* by *o2* on success, or *NULL* on
129   failure.  This is the equivalent of the Python expression ``o1 ^ o2``.
130
131
132.. c:function:: PyObject* PyNumber_Or(PyObject *o1, PyObject *o2)
133
134   Returns the "bitwise or" of *o1* and *o2* on success, or *NULL* on failure.
135   This is the equivalent of the Python expression ``o1 | o2``.
136
137
138.. c:function:: PyObject* PyNumber_InPlaceAdd(PyObject *o1, PyObject *o2)
139
140   Returns the result of adding *o1* and *o2*, or *NULL* on failure.  The operation
141   is done *in-place* when *o1* supports it.  This is the equivalent of the Python
142   statement ``o1 += o2``.
143
144
145.. c:function:: PyObject* PyNumber_InPlaceSubtract(PyObject *o1, PyObject *o2)
146
147   Returns the result of subtracting *o2* from *o1*, or *NULL* on failure.  The
148   operation is done *in-place* when *o1* supports it.  This is the equivalent of
149   the Python statement ``o1 -= o2``.
150
151
152.. c:function:: PyObject* PyNumber_InPlaceMultiply(PyObject *o1, PyObject *o2)
153
154   Returns the result of multiplying *o1* and *o2*, or *NULL* on failure.  The
155   operation is done *in-place* when *o1* supports it.  This is the equivalent of
156   the Python statement ``o1 *= o2``.
157
158
159.. c:function:: PyObject* PyNumber_InPlaceDivide(PyObject *o1, PyObject *o2)
160
161   Returns the result of dividing *o1* by *o2*, or *NULL* on failure.  The
162   operation is done *in-place* when *o1* supports it. This is the equivalent of
163   the Python statement ``o1 /= o2``.
164
165
166.. c:function:: PyObject* PyNumber_InPlaceFloorDivide(PyObject *o1, PyObject *o2)
167
168   Returns the mathematical floor of dividing *o1* by *o2*, or *NULL* on failure.
169   The operation is done *in-place* when *o1* supports it.  This is the equivalent
170   of the Python statement ``o1 //= o2``.
171
172   .. versionadded:: 2.2
173
174
175.. c:function:: PyObject* PyNumber_InPlaceTrueDivide(PyObject *o1, PyObject *o2)
176
177   Return a reasonable approximation for the mathematical value of *o1* divided by
178   *o2*, or *NULL* on failure.  The return value is "approximate" because binary
179   floating point numbers are approximate; it is not possible to represent all real
180   numbers in base two.  This function can return a floating point value when
181   passed two integers.  The operation is done *in-place* when *o1* supports it.
182
183   .. versionadded:: 2.2
184
185
186.. c:function:: PyObject* PyNumber_InPlaceRemainder(PyObject *o1, PyObject *o2)
187
188   Returns the remainder of dividing *o1* by *o2*, or *NULL* on failure.  The
189   operation is done *in-place* when *o1* supports it.  This is the equivalent of
190   the Python statement ``o1 %= o2``.
191
192
193.. c:function:: PyObject* PyNumber_InPlacePower(PyObject *o1, PyObject *o2, PyObject *o3)
194
195   .. index:: builtin: pow
196
197   See the built-in function :func:`pow`. Returns *NULL* on failure.  The operation
198   is done *in-place* when *o1* supports it.  This is the equivalent of the Python
199   statement ``o1 **= o2`` when o3 is :c:data:`Py_None`, or an in-place variant of
200   ``pow(o1, o2, o3)`` otherwise. If *o3* is to be ignored, pass :c:data:`Py_None`
201   in its place (passing *NULL* for *o3* would cause an illegal memory access).
202
203
204.. c:function:: PyObject* PyNumber_InPlaceLshift(PyObject *o1, PyObject *o2)
205
206   Returns the result of left shifting *o1* by *o2* on success, or *NULL* on
207   failure.  The operation is done *in-place* when *o1* supports it.  This is the
208   equivalent of the Python statement ``o1 <<= o2``.
209
210
211.. c:function:: PyObject* PyNumber_InPlaceRshift(PyObject *o1, PyObject *o2)
212
213   Returns the result of right shifting *o1* by *o2* on success, or *NULL* on
214   failure.  The operation is done *in-place* when *o1* supports it.  This is the
215   equivalent of the Python statement ``o1 >>= o2``.
216
217
218.. c:function:: PyObject* PyNumber_InPlaceAnd(PyObject *o1, PyObject *o2)
219
220   Returns the "bitwise and" of *o1* and *o2* on success and *NULL* on failure. The
221   operation is done *in-place* when *o1* supports it.  This is the equivalent of
222   the Python statement ``o1 &= o2``.
223
224
225.. c:function:: PyObject* PyNumber_InPlaceXor(PyObject *o1, PyObject *o2)
226
227   Returns the "bitwise exclusive or" of *o1* by *o2* on success, or *NULL* on
228   failure.  The operation is done *in-place* when *o1* supports it.  This is the
229   equivalent of the Python statement ``o1 ^= o2``.
230
231
232.. c:function:: PyObject* PyNumber_InPlaceOr(PyObject *o1, PyObject *o2)
233
234   Returns the "bitwise or" of *o1* and *o2* on success, or *NULL* on failure.  The
235   operation is done *in-place* when *o1* supports it.  This is the equivalent of
236   the Python statement ``o1 |= o2``.
237
238
239.. c:function:: int PyNumber_Coerce(PyObject **p1, PyObject **p2)
240
241   .. index:: builtin: coerce
242
243   This function takes the addresses of two variables of type :c:type:`PyObject\*`.
244   If the objects pointed to by ``*p1`` and ``*p2`` have the same type, increment
245   their reference count and return ``0`` (success). If the objects can be
246   converted to a common numeric type, replace ``*p1`` and ``*p2`` by their
247   converted value (with 'new' reference counts), and return ``0``. If no
248   conversion is possible, or if some other error occurs, return ``-1`` (failure)
249   and don't increment the reference counts.  The call ``PyNumber_Coerce(&o1,
250   &o2)`` is equivalent to the Python statement ``o1, o2 = coerce(o1, o2)``.
251
252
253.. c:function:: int PyNumber_CoerceEx(PyObject **p1, PyObject **p2)
254
255   This function is similar to :c:func:`PyNumber_Coerce`, except that it returns
256   ``1`` when the conversion is not possible and when no error is raised.
257   Reference counts are still not increased in this case.
258
259
260.. c:function:: PyObject* PyNumber_Int(PyObject *o)
261
262   .. index:: builtin: int
263
264   Returns the *o* converted to an integer object on success, or *NULL* on failure.
265   If the argument is outside the integer range a long object will be returned
266   instead. This is the equivalent of the Python expression ``int(o)``.
267
268
269.. c:function:: PyObject* PyNumber_Long(PyObject *o)
270
271   .. index:: builtin: long
272
273   Returns the *o* converted to a long integer object on success, or *NULL* on
274   failure.  This is the equivalent of the Python expression ``long(o)``.
275
276
277.. c:function:: PyObject* PyNumber_Float(PyObject *o)
278
279   .. index:: builtin: float
280
281   Returns the *o* converted to a float object on success, or *NULL* on failure.
282   This is the equivalent of the Python expression ``float(o)``.
283
284
285.. c:function:: PyObject* PyNumber_Index(PyObject *o)
286
287   Returns the *o* converted to a Python int or long on success or *NULL* with a
288   :exc:`TypeError` exception raised on failure.
289
290   .. versionadded:: 2.5
291
292
293.. c:function:: PyObject* PyNumber_ToBase(PyObject *n, int base)
294
295   Returns the integer *n* converted to *base* as a string with a base
296   marker of ``'0b'``, ``'0o'``, or ``'0x'`` if applicable.  When
297   *base* is not 2, 8, 10, or 16, the format is ``'x#num'`` where x is the
298   base. If *n* is not an int object, it is converted with
299   :c:func:`PyNumber_Index` first.
300
301   .. versionadded:: 2.6
302
303
304.. c:function:: Py_ssize_t PyNumber_AsSsize_t(PyObject *o, PyObject *exc)
305
306   Returns *o* converted to a Py_ssize_t value if *o* can be interpreted as an
307   integer. If *o* can be converted to a Python int or long but the attempt to
308   convert to a Py_ssize_t value would raise an :exc:`OverflowError`, then the
309   *exc* argument is the type of exception that will be raised (usually
310   :exc:`IndexError` or :exc:`OverflowError`).  If *exc* is *NULL*, then the
311   exception is cleared and the value is clipped to *PY_SSIZE_T_MIN* for a negative
312   integer or *PY_SSIZE_T_MAX* for a positive integer.
313
314   .. versionadded:: 2.5
315
316
317.. c:function:: int PyIndex_Check(PyObject *o)
318
319   Returns ``1`` if *o* is an index integer (has the nb_index slot of  the
320   tp_as_number structure filled in), and ``0`` otherwise.
321
322   .. versionadded:: 2.5
323