ARROW-825: [Python] Rename pyarrow.from_pylist to pyarrow.array, test on tuples

python/doc/source/api.rst

@@ -83,12 +83,13 @@ Scalar Value Types
    StringValue
    FixedSizeBinaryValue
 
-Array Types
------------
+Array Types and Constructors
+----------------------------
 
 .. autosummary::
    :toctree: generated/
 
+   array
    Array
    NumericArray
    IntegerArray

python/doc/source/install.rst

@@ -90,26 +90,26 @@ using the default system install location will work, but for now we are being
 explicit:
 
 .. code-block:: bash
-    
+
     export ARROW_HOME=$HOME/local
 
 Now, we build Arrow:
 
 .. code-block:: bash
 
     cd arrow/cpp
-    
+
     mkdir dev-build
     cd dev-build
-    
+
     cmake -DCMAKE_INSTALL_PREFIX=$ARROW_HOME ..
-    
+
     make
-    
+
     # Use sudo here if $ARROW_HOME requires it
     make install
 
-To get the optional Parquet support, you should also build and install 
+To get the optional Parquet support, you should also build and install
 `parquet-cpp <https://github.com/apache/parquet-cpp/blob/master/README.md>`_.
 
 Install `pyarrow`
@@ -138,15 +138,14 @@ Install `pyarrow`
 
 
 .. code-block:: python
-    
+
     In [1]: import pyarrow
 
-    In [2]: pyarrow.from_pylist([1,2,3])
+    In [2]: pyarrow.array([1,2,3])
     Out[2]:
     <pyarrow.array.Int64Array object at 0x7f899f3e60e8>
     [
       1,
       2,
       3
     ]
-

python/pyarrow/__init__.py

@@ -38,7 +38,7 @@
                             DataType, FixedSizeBinaryType,
                             Field, Schema, schema,
                             Array, Tensor,
-                            from_pylist,
+                            array,
                             from_numpy_dtype,
                             NumericArray, IntegerArray, FloatingPointArray,
                             BooleanArray,

python/pyarrow/_array.pyx

@@ -790,6 +790,42 @@ cdef maybe_coerce_datetime64(values, dtype, DataType type,
     return values, type
 
 
+
+def array(object sequence, DataType type=None, MemoryPool memory_pool=None):
+    """
+    Create pyarrow.Array instance from a Python sequence
+
+    Parameters
+    ----------
+    sequence : sequence-like object of Python objects
+    type : pyarrow.DataType, optional
+        If not passed, will be inferred from the data
+    memory_pool : pyarrow.MemoryPool, optional
+        If not passed, will allocate memory from the currently-set default
+        memory pool
+
+    Returns
+    -------
+    array : pyarrow.Array
+    """
+    cdef:
+       shared_ptr[CArray] sp_array
+       CMemoryPool* pool
+
+    pool = maybe_unbox_memory_pool(memory_pool)
+    if type is None:
+        check_status(pyarrow.ConvertPySequence(sequence, pool, &sp_array))
+    else:
+        check_status(
+            pyarrow.ConvertPySequence(
+                sequence, pool, &sp_array, type.sp_type
+            )
+        )
+
+    return box_array(sp_array)
+
+
+
 cdef class Array:
 
     cdef init(self, const shared_ptr[CArray]& sp_array):
@@ -891,36 +927,6 @@ cdef class Array:
 
         return box_array(out)
 
-    @staticmethod
-    def from_list(object list_obj, DataType type=None,
-                  MemoryPool memory_pool=None):
-        """
-        Convert Python list to Arrow array
-
-        Parameters
-        ----------
-        list_obj : array_like
-
-        Returns
-        -------
-        pyarrow.array.Array
-        """
-        cdef:
-           shared_ptr[CArray] sp_array
-           CMemoryPool* pool
-
-        pool = maybe_unbox_memory_pool(memory_pool)
-        if type is None:
-            check_status(pyarrow.ConvertPySequence(list_obj, pool, &sp_array))
-        else:
-            check_status(
-                pyarrow.ConvertPySequence(
-                    list_obj, pool, &sp_array, type.sp_type
-                )
-            )
-
-        return box_array(sp_array)
-
     property null_count:
 
         def __get__(self):
@@ -1363,6 +1369,3 @@ cdef object get_series_values(object obj):
         result = PandasSeries(obj).values
 
     return result
-
-
-from_pylist = Array.from_list

python/pyarrow/parquet.py

@@ -293,9 +293,9 @@ def dictionary(self):
         # Only integer and string partition types are supported right now
         try:
             integer_keys = [int(x) for x in self.keys]
-            dictionary = _array.from_pylist(integer_keys)
+            dictionary = _array.array(integer_keys)
         except ValueError:
-            dictionary = _array.from_pylist(self.keys)
+            dictionary = _array.array(self.keys)
 
         self._dictionary = dictionary
         return dictionary

python/pyarrow/tests/test_array.py

@@ -36,12 +36,12 @@ def test_repr_on_pre_init_array():
 
 
 def test_getitem_NA():
-    arr = pa.from_pylist([1, None, 2])
+    arr = pa.array([1, None, 2])
     assert arr[1] is pa.NA
 
 
 def test_list_format():
-    arr = pa.from_pylist([[1], None, [2, 3, None]])
+    arr = pa.array([[1], None, [2, 3, None]])
     result = fmt.array_format(arr)
     expected = """\
 [
@@ -55,7 +55,7 @@ def test_list_format():
 
 
 def test_string_format():
-    arr = pa.from_pylist(['', None, 'foo'])
+    arr = pa.array(['', None, 'foo'])
     result = fmt.array_format(arr)
     expected = """\
 [
@@ -67,7 +67,7 @@ def test_string_format():
 
 
 def test_long_array_format():
-    arr = pa.from_pylist(range(100))
+    arr = pa.array(range(100))
     result = fmt.array_format(arr, window=2)
     expected = """\
 [
@@ -83,7 +83,7 @@ def test_long_array_format():
 def test_to_pandas_zero_copy():
     import gc
 
-    arr = pa.from_pylist(range(10))
+    arr = pa.array(range(10))
 
     for i in range(10):
         np_arr = arr.to_pandas()
@@ -93,7 +93,7 @@ def test_to_pandas_zero_copy():
     assert sys.getrefcount(arr) == 2
 
     for i in range(10):
-        arr = pa.from_pylist(range(10))
+        arr = pa.array(range(10))
         np_arr = arr.to_pandas()
         arr = None
         gc.collect()
@@ -108,14 +108,14 @@ def test_to_pandas_zero_copy():
 
 
 def test_array_slice():
-    arr = pa.from_pylist(range(10))
+    arr = pa.array(range(10))
 
     sliced = arr.slice(2)
-    expected = pa.from_pylist(range(2, 10))
+    expected = pa.array(range(2, 10))
     assert sliced.equals(expected)
 
     sliced2 = arr.slice(2, 4)
-    expected2 = pa.from_pylist(range(2, 6))
+    expected2 = pa.array(range(2, 6))
     assert sliced2.equals(expected2)
 
     # 0 offset