Buffer Protocol¶
Certain objects available in Python wrap access to an underlying memory
array orbuffer.Such objects include the built-inbytes
and
bytearray
,and some extension types likearray.array
.
Third-party libraries may define their own types for special purposes, such
as image processing or numeric analysis.
While each of these types have their own semantics, they share the common characteristic of being backed by a possibly large memory buffer. It is then desirable, in some situations, to access that buffer directly and without intermediate copying.
Python provides such a facility at the C level in the form of thebuffer protocol.This protocol has two sides:
on the producer side, a type can export a “buffer interface” which allows objects of that type to expose information about their underlying buffer. This interface is described in the sectionBuffer Object Structures;
on the consumer side, several means are available to obtain a pointer to the raw underlying data of an object (for example a method parameter).
Simple objects such asbytes
andbytearray
expose their
underlying buffer in byte-oriented form. Other forms are possible; for example,
the elements exposed by anarray.array
can be multi-byte values.
An example consumer of the buffer interface is thewrite()
method of file objects: any object that can export a series of bytes through
the buffer interface can be written to a file. Whilewrite()
only
needs read-only access to the internal contents of the object passed to it,
other methods such asreadinto()
need write access
to the contents of their argument. The buffer interface allows objects to
selectively allow or reject exporting of read-write and read-only buffers.
There are two ways for a consumer of the buffer interface to acquire a buffer over a target object:
call
PyObject_GetBuffer()
with the right parameters;call
PyArg_ParseTuple()
(or one of its siblings) with one of they*
,w*
ors*
format codes.
In both cases,PyBuffer_Release()
must be called when the buffer
isn’t needed anymore. Failure to do so could lead to various issues such as
resource leaks.
Buffer structure¶
Buffer structures (or simply “buffers” ) are useful as a way to expose the binary data from another object to the Python programmer. They can also be used as a zero-copy slicing mechanism. Using their ability to reference a block of memory, it is possible to expose any data to the Python programmer quite easily. The memory could be a large, constant array in a C extension, it could be a raw block of memory for manipulation before passing to an operating system library, or it could be used to pass around structured data in its native, in-memory format.
Contrary to most data types exposed by the Python interpreter, buffers
are notPyObject
pointers but rather simple C structures. This
allows them to be created and copied very simply. When a generic wrapper
around a buffer is needed, amemoryviewobject
can be created.
For short instructions how to write an exporting object, see
Buffer Object Structures.For obtaining
a buffer, seePyObject_GetBuffer()
.
-
typePy_buffer¶
- Part of theStable ABI(including all members) since version 3.11.
-
void*buf¶
A pointer to the start of the logical structure described by the buffer fields. This can be any location within the underlying physical memory block of the exporter. For example, with negative
strides
the value may point to the end of the memory block.Forcontiguousarrays, the value points to the beginning of the memory block.
-
PyObject*obj¶
A new reference to the exporting object. The reference is owned by the consumer and automatically released (i.e. reference count decremented) and set to
NULL
byPyBuffer_Release()
.The field is the equivalent of the return value of any standard C-API function.As a special case, fortemporarybuffers that are wrapped by
PyMemoryView_FromBuffer()
orPyBuffer_FillInfo()
this field isNULL
.In general, exporting objects MUST NOT use this scheme.
-
Py_ssize_tlen¶
product(shape)*itemsize
.For contiguous arrays, this is the length of the underlying memory block. For non-contiguous arrays, it is the length that the logical structure would have if it were copied to a contiguous representation.Accessing
((char*)buf)[0]upto((char*)buf)[len-1]
is only valid if the buffer has been obtained by a request that guarantees contiguity. In most cases such a request will bePyBUF_SIMPLE
orPyBUF_WRITABLE
.
-
intreadonly¶
An indicator of whether the buffer is read-only. This field is controlled by the
PyBUF_WRITABLE
flag.
-
Py_ssize_titemsize¶
Item size in bytes of a single element. Same as the value of
struct.calcsize()
called on non-NULL
format
values.Important exception: If a consumer requests a buffer without the
PyBUF_FORMAT
flag,format
will be set toNULL
,butitemsize
still has the value for the original format.If
shape
is present, the equalityproduct(shape)*itemsize==len
still holds and the consumer can useitemsize
to navigate the buffer.If
shape
isNULL
as a result of aPyBUF_SIMPLE
or aPyBUF_WRITABLE
request, the consumer must disregarditemsize
and assumeitemsize==1
.
-
char*format¶
ANULLterminated string in
struct
module style syntax describing the contents of a single item. If this isNULL
,"B"
(unsigned bytes) is assumed.This field is controlled by the
PyBUF_FORMAT
flag.
-
intndim¶
The number of dimensions the memory represents as an n-dimensional array. If it is
0
,buf
points to a single item representing a scalar. In this case,shape
,strides
andsuboffsets
MUST beNULL
. The maximum number of dimensions is given byPyBUF_MAX_NDIM
.
-
Py_ssize_t*shape¶
An array of
Py_ssize_t
of lengthndim
indicating the shape of the memory as an n-dimensional array. Note thatshape[0]*...*shape[ndim-1]*itemsize
MUST be equal tolen
.Shape values are restricted to
shape[n]>=0
.The caseshape[n]==0
requires special attention. Seecomplex arrays for further information.The shape array is read-only for the consumer.
-
Py_ssize_t*strides¶
An array of
Py_ssize_t
of lengthndim
giving the number of bytes to skip to get to a new element in each dimension.Stride values can be any integer. For regular arrays, strides are usually positive, but a consumer MUST be able to handle the case
strides[n]<=0
.Seecomplex arraysfor further information.The strides array is read-only for the consumer.
-
Py_ssize_t*suboffsets¶
An array of
Py_ssize_t
of lengthndim
. Ifsuboffsets[n]>=0
,the values stored along the nth dimension are pointers and the suboffset value dictates how many bytes to add to each pointer after de-referencing. A suboffset value that is negative indicates that no de-referencing should occur (striding in a contiguous memory block).If all suboffsets are negative (i.e. no de-referencing is needed), then this field must be
NULL
(the default value).This type of array representation is used by the Python Imaging Library (PIL). Seecomplex arraysfor further information how to access elements of such an array.
The suboffsets array is read-only for the consumer.
-
void*internal¶
This is for use internally by the exporting object. For example, this might be re-cast as an integer by the exporter and used to store flags about whether or not the shape, strides, and suboffsets arrays must be freed when the buffer is released. The consumer MUST NOT alter this value.
-
void*buf¶
Constants:
-
PyBUF_MAX_NDIM¶
The maximum number of dimensions the memory represents. Exporters MUST respect this limit, consumers of multi-dimensional buffers SHOULD be able to handle up to
PyBUF_MAX_NDIM
dimensions. Currently set to 64.
Buffer request types¶
Buffers are usually obtained by sending a buffer request to an exporting
object viaPyObject_GetBuffer()
.Since the complexity of the logical
structure of the memory can vary drastically, the consumer uses theflags
argument to specify the exact buffer type it can handle.
AllPy_buffer
fields are unambiguously defined by the request
type.
request-independent fields¶
The following fields are not influenced byflagsand must always be filled in
with the correct values:obj
,buf
,
len
,itemsize
,ndim
.
readonly, format¶
PyBUF_WRITABLE
can be |’d to any of the flags in the next section.
SincePyBUF_SIMPLE
is defined as 0,PyBUF_WRITABLE
can be used as a stand-alone flag to request a simple writable buffer.
PyBUF_FORMAT
can be |’d to any of the flags exceptPyBUF_SIMPLE
.
The latter already implies formatB
(unsigned bytes).
shape, strides, suboffsets¶
The flags that control the logical structure of the memory are listed in decreasing order of complexity. Note that each flag contains all bits of the flags below it.
Request |
shape |
strides |
suboffsets |
---|---|---|---|
|
yes |
yes |
if needed |
|
yes |
yes |
NULL |
|
yes |
NULL |
NULL |
|
NULL |
NULL |
NULL |
contiguity requests¶
C or Fortrancontiguitycan be explicitly requested, with and without stride information. Without stride information, the buffer must be C-contiguous.
Request |
shape |
strides |
suboffsets |
contig |
---|---|---|---|---|
|
yes |
yes |
NULL |
C |
|
yes |
yes |
NULL |
F |
|
yes |
yes |
NULL |
C or F |
yes |
NULL |
NULL |
C |
compound requests¶
All possible requests are fully defined by some combination of the flags in the previous section. For convenience, the buffer protocol provides frequently used combinations as single flags.
In the following tableUstands for undefined contiguity. The consumer would
have to callPyBuffer_IsContiguous()
to determine contiguity.
Request |
shape |
strides |
suboffsets |
contig |
readonly |
format |
---|---|---|---|---|---|---|
|
yes |
yes |
if needed |
U |
0 |
yes |
|
yes |
yes |
if needed |
U |
1 or 0 |
yes |
|
yes |
yes |
NULL |
U |
0 |
yes |
|
yes |
yes |
NULL |
U |
1 or 0 |
yes |
|
yes |
yes |
NULL |
U |
0 |
NULL |
|
yes |
yes |
NULL |
U |
1 or 0 |
NULL |
|
yes |
NULL |
NULL |
C |
0 |
NULL |
|
yes |
NULL |
NULL |
C |
1 or 0 |
NULL |
Complex arrays¶
NumPy-style: shape and strides¶
The logical structure of NumPy-style arrays is defined byitemsize
,
ndim
,shape
andstrides
.
Ifndim==0
,the memory location pointed to bybuf
is
interpreted as a scalar of sizeitemsize
.In that case,
bothshape
andstrides
areNULL
.
Ifstrides
isNULL
,the array is interpreted as
a standard n-dimensional C-array. Otherwise, the consumer must access an
n-dimensional array as follows:
ptr=(char*)buf+indices[0]*strides[0]+...+indices[n-1]*strides[n-1];
item=*((typeof(item)*)ptr);
As noted above,buf
can point to any location within
the actual memory block. An exporter can check the validity of a buffer with
this function:
defverify_structure(memlen,itemsize,ndim,shape,strides,offset):
"""Verify that the parameters represent a valid array within
the bounds of the allocated memory:
char *mem: start of the physical memory block
memlen: length of the physical memory block
offset: (char *)buf - mem
"""
ifoffset%itemsize:
returnFalse
ifoffset<0oroffset+itemsize>memlen:
returnFalse
ifany(v%itemsizeforvinstrides):
returnFalse
ifndim<=0:
returnndim==0andnotshapeandnotstrides
if0inshape:
returnTrue
imin=sum(strides[j]*(shape[j]-1)forjinrange(ndim)
ifstrides[j]<=0)
imax=sum(strides[j]*(shape[j]-1)forjinrange(ndim)
ifstrides[j]>0)
return0<=offset+iminandoffset+imax+itemsize<=memlen
PIL-style: shape, strides and suboffsets¶
In addition to the regular items, PIL-style arrays can contain pointers
that must be followed in order to get to the next element in a dimension.
For example, the regular three-dimensional C-arraycharv[2][2][3]
can
also be viewed as an array of 2 pointers to 2 two-dimensional arrays:
char(*v[2])[2][3]
.In suboffsets representation, those two pointers
can be embedded at the start ofbuf
,pointing
to twocharx[2][3]
arrays that can be located anywhere in memory.
Here is a function that returns a pointer to the element in an N-D array
pointed to by an N-dimensional index when there are both non-NULL
strides
and suboffsets:
void*get_item_pointer(intndim,void*buf,Py_ssize_t*strides,
Py_ssize_t*suboffsets,Py_ssize_t*indices){
char*pointer=(char*)buf;
inti;
for(i=0;i<ndim;i++){
pointer+=strides[i]*indices[i];
if(suboffsets[i]>=0){
pointer=*((char**)pointer)+suboffsets[i];
}
}
return(void*)pointer;
}