我正在将Cython内存View转换为numpy数组(以便能够在纯Python代码中使用它):fromlibc.stdlibcimportrealloccimportnumpyasnpDTYPE=np.float64ctypedefnp.float64_tDTYPE_tcpdefnp.ndarray[DTYPE_t]compute(DTYPE_t[:,::1]data):cdefunsignedintNchannels=data.shape[0]cdefunsignedintNdata=data.shape[1]cdefDTYPE_t*output=NULLcdefDTYPE_t[::

我正在将Cython内存View转换为numpy数组(以便能够在纯Python代码中使用它):fromlibc.stdlibcimportrealloccimportnumpyasnpDTYPE=np.float64ctypedefnp.float64_tDTYPE_tcpdefnp.ndarray[DTYPE_t]compute(DTYPE_t[:,::1]data):cdefunsignedintNchannels=data.shape[0]cdefunsignedintNdata=data.shape[1]cdefDTYPE_t*output=NULLcdefDTYPE_t[::

问题:固定大小记录的二进制数据想使用struct.unpack_from和struct.pack_into来操作二进制数据不想要数据的副本想要内存中的多个View以简单地抵消计算等。数据可以在array.arraybytearray或ctypes字符串缓冲区中我尝试做的事情:part1=buffer(binary_data,0,size1)part2=buffer(binary_data,size1,size2)part3=buffer(binary_data,size1+size2)#nosizeisgivenforthisoneasitshouldconsumetherestoft

考虑这个虚拟的Cython代码:#!python#cython:boundscheck=False#cython:wraparound=False#cython:initializedcheck=False#cython:cdivision=True#cython:nonecheck=Falseimportnumpyasnp#iteratorfunctioncdefdouble[:]f(double[:]data):data[0]*=1.01data[1]*=1.02returndata#loopingfunctioncdefdouble[:]_call_me(intbignumber

查看documentation在内存View上:memoryviewobjectsallowPythoncodetoaccesstheinternaldataofanobjectthatsupportsthebufferprotocolwithoutcopying.classmemoryview(obj)Createamemoryviewthatreferencesobj.objmustsupportthebufferprotocol.Built-inobjectsthatsupportthebufferprotocolincludebytesandbytearray.然后我们给出示例