maxframe 0.1.0b5__cp38-cp38-macosx_10_9_universal2.whl → 1.0.0__cp38-cp38-macosx_10_9_universal2.whl

maxframe/dataframe/arithmetic/docstring.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# FIXME：https://github.com/aliyun/alibabacloud-odps-maxframe-client/issues/17
 _flex_doc_FRAME = """
 Get {desc} of dataframe and other, element-wise (binary operator `{op_name}`).
 Equivalent to ``{equiv}``, but with support to substitute a fill_value
@@ -127,44 +128,15 @@ circle          0
 triangle        3
 rectangle       4
 
->>> (df * other).execute()
-           angles  degrees
-circle          0      NaN
-triangle        9      NaN
-rectangle      16      NaN
-
 >>> df.mul(other, fill_value=0).execute()
            angles  degrees
 circle          0      0.0
 triangle        9      0.0
 rectangle      16      0.0
 
-Divide by a MultiIndex by level.
-
->>> df_multindex = md.DataFrame({{'angles': [0, 3, 4, 4, 5, 6],
-...                              'degrees': [360, 180, 360, 360, 540, 720]}},
-...                             index=[['A', 'A', 'A', 'B', 'B', 'B'],
-...                                    ['circle', 'triangle', 'rectangle',
-...                                     'square', 'pentagon', 'hexagon']])
->>> df_multindex.execute()
-             angles  degrees
-A circle          0      360
-  triangle        3      180
-  rectangle       4      360
-B square          4      360
-  pentagon        5      540
-  hexagon         6      720
-
->>> df.div(df_multindex, level=1, fill_value=0).execute()
-             angles  degrees
-A circle        NaN      1.0
-  triangle      1.0      1.0
-  rectangle     1.0      1.0
-B square        0.0      0.0
-  pentagon      0.0      0.0
-  hexagon       0.0      0.0
 """
 
+# FIXME：https://github.com/aliyun/alibabacloud-odps-maxframe-client/issues/28
 _flex_doc_SERIES = """
 Return {desc} of series and other, element-wise (binary operator `{op_name}`).
 
@@ -257,7 +229,8 @@ Mismatched indices will be unioned together.
 
 Examples
 --------
->>> df = pd.DataFrame({{'cost': [250, 150, 100],
+>>> import maxframe.dataframe as md
+>>> df = md.DataFrame({{'cost': [250, 150, 100],
 ...                    'revenue': [100, 250, 300]}},
 ...                   index=['A', 'B', 'C'])
 >>> df.execute()
@@ -317,7 +290,7 @@ C   True    False
 
 Compare to a DataFrame of different shape.
 
->>> other = pd.DataFrame({{'revenue': [300, 250, 100, 150]}},
+>>> other = md.DataFrame({{'revenue': [300, 250, 100, 150]}},
 ...                      index=['A', 'B', 'C', 'D'])
 >>> other.execute()
    revenue
@@ -335,7 +308,7 @@ D  False    False
 
 Compare to a MultiIndex by level.
 
->>> df_multindex = pd.DataFrame({{'cost': [250, 150, 100, 150, 300, 220],
+>>> df_multindex = md.DataFrame({{'cost': [250, 150, 100, 150, 300, 220],
 ...                              'revenue': [100, 250, 300, 200, 175, 225]}},
 ...                             index=[['Q1', 'Q1', 'Q1', 'Q2', 'Q2', 'Q2'],
 ...                                    ['A', 'B', 'C', 'A', 'B', 'C']])
@@ -356,6 +329,7 @@ Q1 A   True     True
 Q2 A  False     True
    B   True    False
    C   True    False
+
 """
 
 

maxframe/dataframe/arithmetic/equal.py

@@ -51,6 +51,8 @@ dtype: bool
 
 
 @bin_compare_doc("Equal to", equiv="==", series_example=_eq_example)
-def eq(df, other, axis="columns", level=None):
-    op = DataFrameEqual(axis=axis, level=level, lhs=df, rhs=other)
+def eq(df, other, axis="columns", level=None, fill_value=None):
+    op = DataFrameEqual(
+        axis=axis, level=level, lhs=df, rhs=other, fill_value=fill_value
+    )
     return op(df, other)

maxframe/dataframe/arithmetic/greater.py

@@ -52,6 +52,8 @@ dtype: bool
 
 
 @bin_compare_doc("Greater than", equiv=">", series_example=_gt_example)
-def gt(df, other, axis="columns", level=None):
-    op = DataFrameGreater(axis=axis, level=level, lhs=df, rhs=other)
+def gt(df, other, axis="columns", level=None, fill_value=None):
+    op = DataFrameGreater(
+        axis=axis, level=level, lhs=df, rhs=other, fill_value=fill_value
+    )
     return op(df, other)

maxframe/dataframe/arithmetic/greater_equal.py

@@ -52,6 +52,8 @@ dtype: bool
 
 
 @bin_compare_doc("Greater than or equal to", equiv=">=", series_example=_ge_example)
-def ge(df, other, axis="columns", level=None):
-    op = DataFrameGreaterEqual(axis=axis, level=level, lhs=df, rhs=other)
+def ge(df, other, axis="columns", level=None, fill_value=None):
+    op = DataFrameGreaterEqual(
+        axis=axis, level=level, lhs=df, rhs=other, fill_value=fill_value
+    )
     return op(df, other)

maxframe/dataframe/arithmetic/less.py

@@ -52,6 +52,6 @@ dtype: bool
 
 
 @bin_compare_doc("Less than", equiv="<", series_example=_lt_example)
-def lt(df, other, axis="columns", level=None):
-    op = DataFrameLess(axis=axis, level=level, lhs=df, rhs=other)
+def lt(df, other, axis="columns", level=None, fill_value=None):
+    op = DataFrameLess(axis=axis, level=level, lhs=df, rhs=other, fill_value=fill_value)
     return op(df, other)

maxframe/dataframe/arithmetic/less_equal.py

@@ -52,6 +52,8 @@ dtype: bool
 
 
 @bin_compare_doc("Less than or equal to", equiv="<=", series_example=_le_example)
-def le(df, other, axis="columns", level=None):
-    op = DataFrameLessEqual(axis=axis, level=level, lhs=df, rhs=other)
+def le(df, other, axis="columns", level=None, fill_value=None):
+    op = DataFrameLessEqual(
+        axis=axis, level=level, lhs=df, rhs=other, fill_value=fill_value
+    )
     return op(df, other)

maxframe/dataframe/arithmetic/not_equal.py

@@ -51,6 +51,8 @@ dtype: bool
 
 
 @bin_compare_doc("Not equal to", equiv="!=", series_example=_ne_example)
-def ne(df, other, axis="columns", level=None):
-    op = DataFrameNotEqual(axis=axis, level=level, lhs=df, rhs=other)
+def ne(df, other, axis="columns", level=None, fill_value=None):
+    op = DataFrameNotEqual(
+        axis=axis, level=level, lhs=df, rhs=other, fill_value=fill_value
+    )
     return op(df, other)

maxframe/dataframe/arithmetic/tests/test_arithmetic.py

@@ -22,6 +22,7 @@ import pandas as pd
 import pytest
 
 from ....core import OperatorType
+from ....tests.utils import assert_mf_index_dtype
 from ....utils import dataslots
 from ...core import IndexValue
 from ...datasource.dataframe import from_pandas
@@ -164,7 +165,7 @@ def test_without_shuffle(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -176,7 +177,7 @@ def test_without_shuffle(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -239,6 +240,28 @@ def test_dataframe_and_series_with_shuffle(func_name, func_opts):
     assert df2.columns_value.key != df1.columns_value.key
 
 
+@pytest.mark.parametrize("func_name, func_opts", binary_functions.items())
+def test_dataframe_and_series_with_multiindex(func_name, func_opts):
+    data1 = pd.DataFrame(
+        np.random.rand(10, 10),
+        index=pd.MultiIndex.from_arrays(
+            [list("AAAAABBBBB"), [4, 9, 3, 2, 1, 5, 8, 6, 7, 10]]
+        ),
+        columns=[4, 1, 3, 2, 10, 5, 9, 8, 6, 7],
+    )
+    data1 = to_boolean_if_needed(func_opts.func_name, data1)
+    df1 = from_pandas(data1, chunk_size=5)
+    s1 = from_pandas_series(data1[10].reset_index(level=0, drop=True), chunk_size=6)
+
+    df2 = getattr(df1, func_opts.func_name)(s1, level=1, axis=0)
+
+    # test df2's index and columns
+    assert df2.shape == (np.nan, df1.shape[1])
+    assert df2.index_value.key != df1.index_value.key
+    assert df2.index_value.names == df1.index_value.names
+    assert df2.columns_value.key == df1.columns_value.key
+
+
 @pytest.mark.parametrize("func_name, func_opts", binary_functions.items())
 def test_series_and_series_with_align_map(func_name, func_opts):
     data1 = pd.DataFrame(
@@ -348,7 +371,7 @@ def test_with_one_shuffle(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -381,7 +404,7 @@ def test_with_all_shuffle(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -411,7 +434,7 @@ def test_with_all_shuffle(func_name, func_opts):
     pd.testing.assert_index_equal(
         df6.columns_value.to_pandas(), func_opts.func(data4, data5).columns
     )
-    assert isinstance(df6.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df6.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df6.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -446,7 +469,7 @@ def test_without_shuffle_and_with_one_chunk(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -479,7 +502,7 @@ def test_both_one_chunk(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -512,7 +535,7 @@ def test_with_shuffle_and_one_chunk(func_name, func_opts):
     pd.testing.assert_index_equal(
         df3.columns_value.to_pandas(), func_opts.func(data1, data2).columns
     )
-    assert isinstance(df3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df3.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df3.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -536,7 +559,7 @@ def test_on_same_dataframe(func_name, func_opts):
     pd.testing.assert_index_equal(
         df2.columns_value.to_pandas(), func_opts.func(data, data).columns
     )
-    assert isinstance(df2.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df2.index_value.value, np.int64)
     pd.testing.assert_index_equal(
         df2.index_value.to_pandas(), pd.Index([], dtype=np.int64)
     )
@@ -568,19 +591,19 @@ def test_dataframe_and_scalar(func_name, func_opts):
     pd.testing.assert_series_equal(result.dtypes, expected.dtypes)
 
     pd.testing.assert_index_equal(result.columns_value.to_pandas(), data.columns)
-    assert isinstance(result.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(result.index_value.value, np.int64)
 
     pd.testing.assert_index_equal(result2.columns_value.to_pandas(), data.columns)
-    assert isinstance(result2.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(result2.index_value.value, np.int64)
 
     pd.testing.assert_index_equal(result3.columns_value.to_pandas(), data.columns)
-    assert isinstance(result3.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(result3.index_value.value, np.int64)
 
     pd.testing.assert_index_equal(result4.columns_value.to_pandas(), data.columns)
-    assert isinstance(result4.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(result4.index_value.value, np.int64)
 
     pd.testing.assert_index_equal(result5.columns_value.to_pandas(), data.columns)
-    assert isinstance(result5.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(result5.index_value.value, np.int64)
 
     if "builtin_function_or_method" not in str(type(func_opts.func)):
         # skip NotImplemented test for comparison function
@@ -657,7 +680,7 @@ def test_abs():
     pd.testing.assert_index_equal(
         df2.columns_value.to_pandas(), df1.columns_value.to_pandas()
     )
-    assert isinstance(df2.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df2.index_value.value, np.int64)
     assert df2.shape == (10, 10)
 
 
@@ -675,7 +698,7 @@ def test_not():
     pd.testing.assert_index_equal(
         df2.columns_value.to_pandas(), df1.columns_value.to_pandas()
     )
-    assert isinstance(df2.index_value.value, IndexValue.Int64Index)
+    assert_mf_index_dtype(df2.index_value.value, np.int64)
     assert df2.shape == (10, 10)
 
 

maxframe/dataframe/core.py

@@ -142,6 +142,14 @@ class IndexValue(Serializable):
         _data = NDArrayField("data")
         _dtype = DataTypeField("dtype")
 
+        @property
+        def dtype(self):
+            return getattr(self, "_dtype", None)
+
+        @property
+        def inferred_type(self):
+            return "floating" if self.dtype.kind == "f" else "integer"
+
     class RangeIndex(IndexBase):
         _name = AnyField("name")
         _slice = SliceField("slice")
@@ -243,6 +251,10 @@ class IndexValue(Serializable):
         _data = NDArrayField("data")
         _dtype = DataTypeField("dtype")
 
+        @property
+        def dtype(self):
+            return getattr(self, "_dtype", None)
+
         @property
         def inferred_type(self):
             return "integer"
@@ -254,6 +266,10 @@ class IndexValue(Serializable):
         _data = NDArrayField("data")
         _dtype = DataTypeField("dtype")
 
+        @property
+        def dtype(self):
+            return getattr(self, "_dtype", None)
+
         @property
         def inferred_type(self):
             return "integer"
@@ -265,6 +281,10 @@ class IndexValue(Serializable):
         _data = NDArrayField("data")
         _dtype = DataTypeField("dtype")
 
+        @property
+        def dtype(self):
+            return getattr(self, "_dtype", None)
+
         @property
         def inferred_type(self):
             return "floating"
@@ -1086,11 +1106,11 @@ class Series(HasShapeTileable, _ToPandasMixin):
         --------
         >>> import maxframe.dataframe as md
         >>> s = md.Series({'a': 1, 'b': 2, 'c': 3})
-        >>> s.ndim.execute()
+        >>> s.ndim
         1
 
         >>> df = md.DataFrame({'col1': [1, 2], 'col2': [3, 4]})
-        >>> df.ndim.execute()
+        >>> df.ndim
         2
         """
         return super().ndim
@@ -1514,15 +1534,17 @@ class BaseDataFrameData(HasShapeTileableData, _ToPandasMixin):
         refresh_index_value(self)
         refresh_dtypes(self)
 
-    def refresh_from_table_meta(self, table_meta: DataFrameTableMeta) -> None:
-        dtypes = table_meta.pd_column_dtypes
+    def refresh_from_dtypes(self, dtypes: pd.Series) -> None:
         self._dtypes = dtypes
         self._columns_value = parse_index(dtypes.index, store_data=True)
         self._dtypes_value = DtypesValue(key=tokenize(dtypes), value=dtypes)
         new_shape = list(self._shape)
-        new_shape[0] = len(dtypes)
+        new_shape[-1] = len(dtypes)
         self._shape = tuple(new_shape)
 
+    def refresh_from_table_meta(self, table_meta: DataFrameTableMeta) -> None:
+        self.refresh_from_dtypes(table_meta.pd_column_dtypes)
+
     @property
     def dtypes(self):
         dt = getattr(self, "_dtypes", None)
@@ -1666,6 +1688,8 @@ class DataFrameData(_BatchedFetcher, BaseDataFrameData):
             raise NotImplementedError
 
         corner_data = fetch_corner_data(self, session=self._executed_sessions[-1])
+        if corner_data is None:
+            return
 
         buf = StringIO()
         max_rows = pd.get_option("display.max_rows")
@@ -1761,11 +1785,11 @@ class DataFrame(HasShapeTileable, _ToPandasMixin):
         --------
         >>> import maxframe.dataframe as md
         >>> s = md.Series({'a': 1, 'b': 2, 'c': 3})
-        >>> s.ndim.execute()
+        >>> s.ndim
         1
 
         >>> df = md.DataFrame({'col1': [1, 2], 'col2': [3, 4]})
-        >>> df.ndim.execute()
+        >>> df.ndim
         2
         """
         return super().ndim

maxframe/dataframe/datasource/date_range.py

@@ -22,7 +22,7 @@ from pandas._libs.tslibs import timezones
 from pandas.tseries.frequencies import to_offset
 from pandas.tseries.offsets import Tick
 
-from ... import opcodes as OperandDef
+from ... import opcodes
 from ...core import OutputType
 from ...serialization.serializables import AnyField, BoolField, Int64Field, StringField
 from ...utils import no_default, pd_release_version
@@ -117,7 +117,7 @@ def generate_range_count(
 
 
 class DataFrameDateRange(DataFrameOperator, DataFrameOperatorMixin):
-    _op_type_ = OperandDef.DATE_RANGE
+    _op_type_ = opcodes.DATE_RANGE
 
     start = AnyField("start")
     end = AnyField("end")