fixes

Author: pcmoritz

python/pyarrow/lib.pxd

@@ -330,15 +330,6 @@ cdef class NativeFile:
 cdef get_reader(object source, shared_ptr[RandomAccessFile]* reader)
 cdef get_writer(object source, shared_ptr[OutputStream]* writer)
 
-cdef class SerializedPyObject:
-    cdef:
-        CSerializedPyObject data
-
-    cdef readonly:
-        object base
-
-    cdef _write_to(self, OutputStream* stream)
-
 cdef public object pyarrow_wrap_buffer(const shared_ptr[CBuffer]& buf)
 cdef public object pyarrow_wrap_data_type(const shared_ptr[CDataType]& type)
 cdef public object pyarrow_wrap_field(const shared_ptr[CField]& field)

python/pyarrow/plasma.pyx

@@ -29,7 +29,7 @@ from cpython.pycapsule cimport *
 import collections
 import pyarrow
 
-from pyarrow.lib cimport Buffer, NativeFile, check_status, SerializedPyObject
+from pyarrow.lib cimport Buffer, NativeFile, check_status
 from pyarrow.includes.libarrow cimport (CMutableBuffer, CBuffer,
                                         CFixedSizeBufferWriter, CStatus)
 
@@ -616,7 +616,7 @@ def put(PlasmaClient client, value, object_id=None):
     client : PlasmaClient
         The client connected to the object store we put the value in.
     value : object
-        A Python object to store.
+        A Python object to store. Currently only lists are supported.
     object_id : ObjectID, default None
         If this is provided, the specified object ID will be used to refer
         to the object.
@@ -626,7 +626,7 @@ def put(PlasmaClient client, value, object_id=None):
     The object ID associated to the Python object.
     """
     cdef ObjectID id = object_id if object_id else ObjectID.from_random()
-    cdef SerializedPyObject serialized = pyarrow.serialize(value)
+    serialized = pyarrow.serialize(value)
     buffer = client.create(id, serialized.total_bytes)
     stream = pyarrow.FixedSizeBufferOutputStream(buffer)
     stream.set_memcopy_threads(4)
@@ -642,18 +642,19 @@ def get(PlasmaClient client, object_ids, timeout_ms=-1):
     ----------
     client : PlasmaClient
         The client connected to the object store we get objects from.
-    object_ids : list
-        The list of object IDs associated to the values we get from the store.
+    object_ids : list or ObjectID
+        Object ID or list of object IDs associated to the values we get from
+        the store.
     timeout_ms : int, default -1
         The number of milliseconds that the get call should block before
         timing out and returning. Pass -1 if the call should block and 0
         if the call should return immediately.
 
     Returns
     -------
-    list
-        List of Python values for the data associated with the object_ids
-        and ObjectNotAvailable if the object was not available.
+    list or object
+        Python value or list of Python values for the data associated with
+        the object_ids and ObjectNotAvailable if the object was not available.
     """
     if isinstance(object_ids, collections.Sequence):
         results = []

python/pyarrow/serialization.pxi

@@ -135,6 +135,11 @@ cdef class SerializedPyObject:
     """
     Arrow-serialized representation of Python object
     """
+    cdef:
+        CSerializedPyObject data
+
+    cdef readonly:
+        object base
 
     property total_bytes:
 

python/pyarrow/tests/test_plasma.py

@@ -119,7 +119,7 @@ def start_plasma_store(plasma_store_memory=DEFAULT_PLASMA_STORE_MEMORY,
     """
     if use_valgrind and use_profiler:
         raise Exception("Cannot use valgrind and profiler at the same time.")
-    plasma_store_executable = "plasma_store"
+    plasma_store_executable = os.path.join(pa.__path__[0], "plasma_store")
     plasma_store_name = "/tmp/plasma_store{}".format(random_name())
     command = [plasma_store_executable,
                "-s", plasma_store_name,
@@ -282,7 +282,7 @@ def test_put_and_get(self):
         result = pa.plasma.get(self.plasma_client, object_id)
         assert result == value
 
-        object_id = pa.plasma.ObjectID(np.bytes.random(20))
+        object_id = pa.plasma.ObjectID.from_random()
         [result] = pa.plasma.get(self.plasma_client, [object_id], timeout_ms=0)
         assert result == pa.plasma.ObjectNotAvailable