ceph/src/arrow/cpp/src/arrow/python/deserialize.h

   1 // Licensed to the Apache Software Foundation (ASF) under one
   2 // or more contributor license agreements.  See the NOTICE file
   3 // distributed with this work for additional information
   4 // regarding copyright ownership.  The ASF licenses this file
   5 // to you under the Apache License, Version 2.0 (the
   6 // "License"); you may not use this file except in compliance
   7 // with the License.  You may obtain a copy of the License at
   8 //
   9 //   http://www.apache.org/licenses/LICENSE-2.0
  10 //
  11 // Unless required by applicable law or agreed to in writing,
  12 // software distributed under the License is distributed on an
  13 // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
  14 // KIND, either express or implied.  See the License for the
  15 // specific language governing permissions and limitations
  16 // under the License.
  17
  18 #pragma once
  19
  20 #include <cstdint>
  21 #include <memory>
  22 #include <vector>
  23
  24 #include "arrow/python/serialize.h"
  25 #include "arrow/python/visibility.h"
  26 #include "arrow/status.h"
  27
  28 namespace arrow {
  29
  30 class RecordBatch;
  31 class Tensor;
  32
  33 namespace io {
  34
  35 class RandomAccessFile;
  36
  37 }  // namespace io
  38
  39 namespace py {
  40
  41 struct ARROW_PYTHON_EXPORT SparseTensorCounts {
  42   int coo;
  43   int csr;
  44   int csc;
  45   int csf;
  46   int ndim_csf;
  47
  48   int num_total_tensors() const { return coo + csr + csc + csf; }
  49   int num_total_buffers() const {
  50     return coo * 3 + csr * 4 + csc * 4 + 2 * ndim_csf + csf;
  51   }
  52 };
  53
  54 /// \brief Read serialized Python sequence from file interface using Arrow IPC
  55 /// \param[in] src a RandomAccessFile
  56 /// \param[out] out the reconstructed data
  57 /// \return Status
  58 ARROW_PYTHON_EXPORT
  59 Status ReadSerializedObject(io::RandomAccessFile* src, SerializedPyObject* out);
  60
  61 /// \brief Reconstruct SerializedPyObject from representation produced by
  62 /// SerializedPyObject::GetComponents.
  63 ///
  64 /// \param[in] num_tensors number of tensors in the object
  65 /// \param[in] num_sparse_tensors number of sparse tensors in the object
  66 /// \param[in] num_ndarrays number of numpy Ndarrays in the object
  67 /// \param[in] num_buffers number of buffers in the object
  68 /// \param[in] data a list containing pyarrow.Buffer instances. It must be 1 +
  69 /// num_tensors * 2 + num_coo_tensors * 3 + num_csr_tensors * 4 + num_csc_tensors * 4 +
  70 /// num_csf_tensors * (2 * ndim_csf + 3) + num_buffers in length
  71 /// \param[out] out the reconstructed object
  72 /// \return Status
  73 ARROW_PYTHON_EXPORT
  74 Status GetSerializedFromComponents(int num_tensors,
  75                                    const SparseTensorCounts& num_sparse_tensors,
  76                                    int num_ndarrays, int num_buffers, PyObject* data,
  77                                    SerializedPyObject* out);
  78
  79 /// \brief Reconstruct Python object from Arrow-serialized representation
  80 /// \param[in] context Serialization context which contains custom serialization
  81 /// and deserialization callbacks. Can be any Python object with a
  82 /// _serialize_callback method for serialization and a _deserialize_callback
  83 /// method for deserialization. If context is None, no custom serialization
  84 /// will be attempted.
  85 /// \param[in] object Object to deserialize
  86 /// \param[in] base a Python object holding the underlying data that any NumPy
  87 /// arrays will reference, to avoid premature deallocation
  88 /// \param[out] out The returned object
  89 /// \return Status
  90 /// This acquires the GIL
  91 ARROW_PYTHON_EXPORT
  92 Status DeserializeObject(PyObject* context, const SerializedPyObject& object,
  93                          PyObject* base, PyObject** out);
  94
  95 /// \brief Reconstruct Ndarray from Arrow-serialized representation
  96 /// \param[in] object Object to deserialize
  97 /// \param[out] out The deserialized tensor
  98 /// \return Status
  99 ARROW_PYTHON_EXPORT
 100 Status DeserializeNdarray(const SerializedPyObject& object, std::shared_ptr<Tensor>* out);
 101
 102 ARROW_PYTHON_EXPORT
 103 Status NdarrayFromBuffer(std::shared_ptr<Buffer> src, std::shared_ptr<Tensor>* out);
 104
 105 }  // namespace py
 106 }  // namespace arrow