maxframe 0.1.0b3__cp38-cp38-win_amd64.whl → 0.1.0b4__cp38-cp38-win_amd64.whl

maxframe/config/config.py

@@ -358,6 +358,9 @@ default_options.register_option(
 default_options.register_option(
     "show_progress", "auto", validator=any_validator(is_bool, is_string)
 )
+default_options.register_option(
+    "dag.settings", value=dict(), validator=is_dict, remote=True
+)
 
 ################
 # SPE Settings #

maxframe/dataframe/__init__.py

@@ -57,6 +57,11 @@ try:
 except ImportError:  # pragma: no cover
     pass
 
+try:
+    from . import _internal
+except ImportError:  # pragma: no cover
+    pass
+
 del (
     arithmetic,
     datasource,

maxframe/dataframe/core.py

@@ -960,7 +960,9 @@ class BaseSeriesData(HasShapeTileableData, _ToPandasMixin):
             buf = StringIO()
             max_rows = pd.get_option("display.max_rows")
             corner_max_rows = (
-                max_rows if self.shape[0] <= max_rows else corner_data.shape[0] - 1
+                max_rows
+                if self.shape[0] <= max_rows or corner_data.shape[0] == 0
+                else corner_data.shape[0] - 1
             )  # make sure max_rows < corner_data
 
             with pd.option_context("display.max_rows", corner_max_rows):
@@ -1605,7 +1607,7 @@ class DataFrameData(_BatchedFetcher, BaseDataFrameData):
             buf = StringIO()
             max_rows = pd.get_option("display.max_rows")
 
-            if self.shape[0] <= max_rows:
+            if self.shape[0] <= max_rows or corner_data.shape[0] == 0:
                 buf.write(repr(corner_data) if representation else str(corner_data))
             else:
                 # remember we cannot directly call repr(df),

maxframe/dataframe/datasource/read_odps_query.py

@@ -263,7 +263,9 @@ def read_odps_query(
     result: DataFrame
         DataFrame read from MaxCompute (ODPS) table
     """
-    odps_entry = odps_entry or ODPS.from_environments()
+    odps_entry = odps_entry or ODPS.from_global() or ODPS.from_environments()
+    if odps_entry is None:
+        raise ValueError("Missing odps_entry parameter")
     inst = odps_entry.execute_sql(f"EXPLAIN {query}")
     explain_str = list(inst.get_task_results().values())[0]
 

maxframe/dataframe/datasource/read_odps_table.py

@@ -164,6 +164,8 @@ def read_odps_table(
         DataFrame read from MaxCompute (ODPS) table
     """
     odps_entry = odps_entry or ODPS.from_global() or ODPS.from_environments()
+    if odps_entry is None:
+        raise ValueError("Missing odps_entry parameter")
     if isinstance(table_name, Table):
         table = table_name
     else:

maxframe/dataframe/datastore/core.py

@@ -0,0 +1,19 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ..operators import DataFrameOperator, DataFrameOperatorMixin
+
+
+class DataFrameDataStore(DataFrameOperator, DataFrameOperatorMixin):
+    pass

maxframe/dataframe/datastore/to_csv.py

@@ -23,11 +23,11 @@ from ...serialization.serializables import (
     ListField,
     StringField,
 )
-from ..operators import DataFrameOperator, DataFrameOperatorMixin
 from ..utils import parse_index
+from .core import DataFrameDataStore
 
 
-class DataFrameToCSV(DataFrameOperator, DataFrameOperatorMixin):
+class DataFrameToCSV(DataFrameDataStore):
     _op_type_ = opcodes.TO_CSV
 
     input = KeyField("input")

maxframe/dataframe/datastore/to_odps.py

@@ -32,13 +32,13 @@ from ...serialization.serializables import (
 )
 from ...typing_ import TileableType
 from ..core import DataFrame  # noqa: F401
-from ..operators import DataFrameOperator, DataFrameOperatorMixin
 from ..utils import parse_index
+from .core import DataFrameDataStore
 
 logger = logging.getLogger(__name__)
 
 
-class DataFrameToODPSTable(DataFrameOperator, DataFrameOperatorMixin):
+class DataFrameToODPSTable(DataFrameDataStore):
     _op_type_ = opcodes.TO_ODPS_TABLE
 
     dtypes = SeriesField("dtypes")

maxframe/dataframe/indexing/reset_index.py

@@ -107,7 +107,6 @@ def df_reset_index(
     inplace=False,
     col_level=0,
     col_fill="",
-    incremental_index=False,
 ):
     """
     Reset the index, or a level of it.
@@ -133,12 +132,6 @@ def df_reset_index(
     col_fill : object, default ''
         If the columns have multiple levels, determines how the other
         levels are named. If None then the index name is repeated.
-    incremental_index: bool, default False
-        Ensure RangeIndex incremental, when output DataFrame has multiple chunks,
-        ensuring index incremental costs more computation,
-        so by default, each chunk will have index which starts from 0,
-        setting incremental_index=True，reset_index will guarantee that
-        output DataFrame's index is from 0 to n - 1.
 
     Returns
     -------
@@ -264,7 +257,6 @@ def df_reset_index(
         drop=drop,
         col_level=col_level,
         col_fill=col_fill,
-        incremental_index=incremental_index,
         output_types=[OutputType.dataframe],
     )
     ret = op(df)
@@ -280,7 +272,6 @@ def series_reset_index(
     drop=False,
     name=no_default,
     inplace=False,
-    incremental_index=False,
 ):
     """
     Generate a new DataFrame or Series with the index reset.
@@ -303,12 +294,6 @@ def series_reset_index(
         when `drop` is True.
     inplace : bool, default False
         Modify the Series in place (do not create a new object).
-    incremental_index: bool, default False
-        Ensure RangeIndex incremental, when output Series has multiple chunks,
-        ensuring index incremental costs more computation,
-        so by default, each chunk will have index which starts from 0,
-        setting incremental_index=True，reset_index will guarantee that
-        output Series's index is from 0 to n - 1.
 
     Returns
     -------
@@ -406,8 +391,7 @@ def series_reset_index(
         level=level,
         drop=drop,
         name=name,
-        incremental_index=incremental_index,
-        output_types=[OutputType.series],
+        output_types=[OutputType.series if drop else OutputType.dataframe],
     )
     ret = op(series)
     if not inplace:

maxframe/odpsio/arrow.py

@@ -65,14 +65,19 @@ def arrow_to_pandas(
         raise ValueError(f"Does not support meta type {table_meta.type!r}")
 
 
-def pandas_to_arrow(df: Any, nthreads=1) -> Tuple[ArrowTableType, DataFrameTableMeta]:
-    table_meta = build_dataframe_table_meta(df)
+def pandas_to_arrow(
+    df: Any, nthreads=1, ignore_index=False
+) -> Tuple[ArrowTableType, DataFrameTableMeta]:
+    table_meta = build_dataframe_table_meta(df, ignore_index)
     df = df.copy() if callable(getattr(df, "copy", None)) else df
     if table_meta.type in (OutputType.dataframe, OutputType.series):
         if table_meta.type == OutputType.series:
             df = df.to_frame("_data" if df.name is None else df.name)
         df.columns = pd.Index(table_meta.table_column_names)
-        df = df.rename_axis(table_meta.table_index_column_names).reset_index()
+        if not ignore_index:
+            df = df.rename_axis(table_meta.table_index_column_names).reset_index()
+    elif ignore_index:
+        df = pd.DataFrame([], columns=[])
     elif table_meta.type == OutputType.index:
         names = [f"_idx_{idx}" for idx in range(len(df.names))]
         df = df.to_frame(name=names[0] if len(names) == 1 else names)

maxframe/odpsio/schema.py

@@ -175,7 +175,9 @@ def _scalar_as_index(df_obj: Any) -> pd.Index:
 
 
 def pandas_to_odps_schema(
-    df_obj: Any, unknown_as_string: bool = False
+    df_obj: Any,
+    unknown_as_string: bool = False,
+    ignore_index=False,
 ) -> Tuple[odps_types.OdpsSchema, DataFrameTableMeta]:
     from .. import dataframe as md
     from .arrow import pandas_to_arrow
@@ -209,7 +211,7 @@ def pandas_to_odps_schema(
     else:
         empty_df_obj = df_obj
 
-    arrow_data, table_meta = pandas_to_arrow(empty_df_obj)
+    arrow_data, table_meta = pandas_to_arrow(empty_df_obj, ignore_index=ignore_index)
     return (
         arrow_schema_to_odps_schema(
             arrow_data.schema, unknown_as_string=unknown_as_string
@@ -268,7 +270,9 @@ def build_table_column_name(
     return col_name
 
 
-def build_dataframe_table_meta(df_obj: Any) -> DataFrameTableMeta:
+def build_dataframe_table_meta(
+    df_obj: Any, ignore_index: bool = False
+) -> DataFrameTableMeta:
     from .. import dataframe as md
 
     col_to_count = defaultdict(lambda: 0)
@@ -285,6 +289,8 @@ def build_dataframe_table_meta(df_obj: Any) -> DataFrameTableMeta:
     else:  # pragma: no cover
         raise TypeError(f"Cannot accept type {type(df_obj)}")
 
+    assert not ignore_index or obj_type in (OutputType.dataframe, OutputType.series)
+
     if obj_type == OutputType.scalar:
         pd_dtypes = pd.Series([])
         column_index_names = []
@@ -340,12 +346,19 @@ def build_dataframe_table_meta(df_obj: Any) -> DataFrameTableMeta:
     else:
         index_dtypes = pd.Series([pd_index_val.dtype], index=pd_index_val.names)
 
+    if ignore_index:
+        table_index_column_names = []
+        pd_index_dtypes = pd.Series([], index=[])
+    else:
+        table_index_column_names = [f"_idx_{i}" for i in range(len(index_obj.names))]
+        pd_index_dtypes = index_dtypes
+
     return DataFrameTableMeta(
         table_name=table_name,
         type=obj_type,
         table_column_names=final_sql_columns,
-        table_index_column_names=[f"_idx_{i}" for i in range(len(index_obj.names))],
+        table_index_column_names=table_index_column_names,
         pd_column_dtypes=pd_dtypes,
         pd_column_level_names=column_index_names,
-        pd_index_dtypes=index_dtypes,
+        pd_index_dtypes=pd_index_dtypes,
     )

maxframe/odpsio/tests/test_schema.py

@@ -61,6 +61,16 @@ def test_pandas_to_odps_schema_dataframe(wrap_obj):
     assert meta.pd_column_level_names == [None]
     assert meta.pd_index_level_names == [None]
 
+    test_df = _wrap_maxframe_obj(data, wrap=wrap_obj)
+    schema, meta = pandas_to_odps_schema(test_df, ignore_index=True)
+    assert [c.name for c in schema.columns] == list(test_df.dtypes.index.str.lower())
+    assert [c.type.name for c in schema.columns] == ["double"] * len(test_df.columns)
+    assert meta.type == OutputType.dataframe
+    assert meta.table_column_names == list(test_df.dtypes.index.str.lower())
+    assert meta.table_index_column_names == []
+    assert meta.pd_column_level_names == [None]
+    assert meta.pd_index_level_names == []
+
     data.columns = pd.MultiIndex.from_tuples(
         [("A", "A"), ("A", "B"), ("A", "C"), ("B", "A"), ("B", "B")], names=["c1", "c2"]
     )
@@ -99,6 +109,15 @@ def test_pandas_to_odps_schema_series(wrap_obj):
     assert meta.pd_column_level_names == [None]
     assert meta.pd_index_level_names == [None]
 
+    schema, meta = pandas_to_odps_schema(test_s, ignore_index=True)
+    assert [c.name for c in schema.columns] == ["_data"]
+    assert [c.type.name for c in schema.columns] == ["double"]
+    assert meta.type == OutputType.series
+    assert meta.table_column_names == ["_data"]
+    assert meta.table_index_column_names == []
+    assert meta.pd_column_level_names == [None]
+    assert meta.pd_index_level_names == []
+
     data.index = pd.MultiIndex.from_arrays(
         [np.random.choice(list("ABC"), 100), np.random.randint(0, 10, 100)],
         names=["c1", "c2"],
@@ -130,6 +149,9 @@ def test_pandas_to_odps_schema_index(wrap_obj):
     assert meta.pd_column_level_names == []
     assert meta.pd_index_level_names == [None]
 
+    with pytest.raises(AssertionError):
+        pandas_to_odps_schema(test_idx, unknown_as_string=True, ignore_index=True)
+
     data = pd.MultiIndex.from_arrays(
         [np.random.choice(list("ABC"), 100), np.random.randint(0, 10, 100)],
         names=["c1", "c2"],
@@ -159,6 +181,9 @@ def test_pandas_to_odps_schema_scalar(wrap_obj):
     assert meta.pd_column_level_names == []
     assert meta.pd_index_level_names == [None]
 
+    with pytest.raises(AssertionError):
+        pandas_to_odps_schema(test_scalar, unknown_as_string=True, ignore_index=True)
+
 
 def test_odps_arrow_schema_conversion():
     odps_schema = odps_types.OdpsSchema(

maxframe/opcodes.py

@@ -564,6 +564,11 @@ CHOLESKY_FUSE = 999988
 # MaxFrame-dedicated functions
 DATAFRAME_RESHUFFLE = 10001
 
+# MaxFrame internal operators
+DATAFRAME_PROJECTION_SAME_INDEX_MERGE = 100001
+GROUPBY_AGGR_SAME_INDEX_MERGE = 100002
+DATAFRAME_ILOC_GET_AND_RENAME_ITEM = 100003
+
 # fetches
 FETCH_SHUFFLE = 999998
 FETCH = 999999

maxframe/session.py

@@ -1211,7 +1211,7 @@ def new_session(
     # load third party extensions.
     ensure_isolation_created(kwargs)
 
-    odps_entry = odps_entry or ODPS.from_environments()
+    odps_entry = odps_entry or ODPS.from_global() or ODPS.from_environments()
     if address is None:
         from maxframe_client.session.consts import ODPS_SESSION_INSECURE_SCHEME
 
@@ -1255,7 +1255,9 @@ def get_default_or_create(**kwargs):
         if session is None:
             # no session attached, try to create one
             warnings.warn(warning_msg)
-            session = new_session(ODPS.from_environments(), **kwargs)
+            session = new_session(
+                ODPS.from_global() or ODPS.from_environments(), **kwargs
+            )
             session.as_default()
     if isinstance(session, IsolatedAsyncSession):
         session = SyncSession.from_isolated_session(session)

maxframe/utils.py

@@ -381,6 +381,11 @@ def build_temp_table_name(session_id: str, tileable_key: str) -> str:
     return f"tmp_mf_{session_id}_{tileable_key}"
 
 
+def build_temp_intermediate_table_name(session_id: str, tileable_key: str) -> str:
+    temp_table = build_temp_table_name(session_id, tileable_key)
+    return f"{temp_table}_intermediate"
+
+
 def build_session_volume_name(session_id: str) -> str:
     return f"mf_vol_{session_id}"
 

{maxframe-0.1.0b3.dist-info → maxframe-0.1.0b4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: maxframe
-Version: 0.1.0b3
+Version: 0.1.0b4
 Summary: MaxFrame operator-based data analyze framework
 Requires-Dist: numpy >=1.19.0
 Requires-Dist: pandas >=1.0.0

{maxframe-0.1.0b3.dist-info → maxframe-0.1.0b4.dist-info}/RECORD

@@ -1,5 +1,5 @@
 maxframe/__init__.py,sha256=MgltwhBvnUQDKKmHDg9Y69TJkRGmQQ9m8-D9FC2pcLU,1007
-maxframe/_utils.cp38-win_amd64.pyd,sha256=v0SiHG8U69LbmO5jg_wMpsYNpTlx3bjMIsvfOeEDnxE,308736
+maxframe/_utils.cp38-win_amd64.pyd,sha256=6iQnaOs0v1br0i1SjkUtFgrbof_YHT5dTJBQF65XPag,308736
 maxframe/_utils.pxd,sha256=_qHN-lCY1FQgDFIrrqA79Ys0SBdonp9kXRMS93xKSYk,1187
 maxframe/_utils.pyx,sha256=_3p6aJEJ6WZYLcNZ6o4DxoxsxqadTlJXFlgDeFPxqUQ,17564
 maxframe/codegen.py,sha256=S23hTTu2fIJTv5PMIItE_NPUnadUyoiMmb-1YAqbaWw,16329
@@ -8,14 +8,14 @@ maxframe/env.py,sha256=xY4wjMWIJ4qLsFAQ5F-X5CrVR7dDSWiryPXni0YSK5c,1435
 maxframe/errors.py,sha256=xBnvoJjjNcHVLhwj77Dux9ut8isGVmmJXFqefmmx8Ak,711
 maxframe/extension.py,sha256=o7yiS99LWTtLF7ZX6F78UUJAqUyd-LllOXA2l69np50,2455
 maxframe/mixin.py,sha256=QfX0KqVIWDlVDSFs0lwdzLexw7lS7W_IUuK7aY1Ib8c,3624
-maxframe/opcodes.py,sha256=L-BvSFEUOMrtGJFXIH6zk2Xv_-te5VZxv5eDDCNNt0U,10566
+maxframe/opcodes.py,sha256=Gcqv1DIMPOXLHOt9yYSRisu2ihP1ez6QX0d5c_L06TA,10732
 maxframe/protocol.py,sha256=N4i0ggLY131gwnxOrCgKeZwzhLKSRB171cx1lWRvUcw,14605
-maxframe/session.py,sha256=Mme-jB2hioJC6ttrXfX4XSeHLCMYoFR0ikmmhx82-vc,36624
+maxframe/session.py,sha256=ETR-n-DDyCYVh7I1juJfNBxcOrESJSjxW7JOHaSt9To,36700
 maxframe/typing_.py,sha256=pAgOhHHSM376N7PJLtNXvS5LHNYywz5dIjnA_hHRWSM,1133
 maxframe/udf.py,sha256=EFAAV2c8SpWKcF9_8Pocpjc4bXsEASf57Qy_Q30YH4Q,2315
-maxframe/utils.py,sha256=CpA4Cqf5Lg7LMKJeJFsmybqVutcSh3sSqMhdHnIi0h4,35017
+maxframe/utils.py,sha256=fzlh5MTJcX9ZBEnIZ4tVMzf846X9YsUmVGTKG5GtZr4,35212
 maxframe/config/__init__.py,sha256=AHo3deaCm1JnbbRX_udboJEDYrYytdvivp9RFxJcumI,671
-maxframe/config/config.py,sha256=Top-9hUEkPvaTlAGrP9GzGjwvZk6YPuHtwy4TEryKqg,13475
+maxframe/config/config.py,sha256=Za9cgJfSJ4noNEcOaQ8tI1lhjT0Lvd3uCURqR7vlmbg,13578
 maxframe/config/validators.py,sha256=pKnloh2kEOBRSsT8ks-zL8XVSaMMVIEvHvwNJlideeo,1672
 maxframe/config/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe/config/tests/test_config.py,sha256=FWQZ6KBUG_jY1-KaR-GKXl7khhlTbuLlk3uaEV8koM8,2839
@@ -33,7 +33,7 @@ maxframe/core/entity/output_types.py,sha256=NnNeDBVAEhD8dtPBWzpM7n6s8neVFrahjd0z
 maxframe/core/entity/tileables.py,sha256=6jJyFscvb8sH5K_k2VaNGeUm8YrpevCtou3WSUl4Dw8,13973
 maxframe/core/entity/utils.py,sha256=454RYVbTMVW_8KnfDqUPec4kz1p98izVTC2OrzhOkao,966
 maxframe/core/graph/__init__.py,sha256=n1WiszgVu0VdXsk12oiAyggduNwu-1-9YKnfZqvmmXk,838
-maxframe/core/graph/core.cp38-win_amd64.pyd,sha256=Pxd3Gfcq9JBPGAfySG6YaTFnJiTphRtr1FNe3NZCBUo,250368
+maxframe/core/graph/core.cp38-win_amd64.pyd,sha256=_JfoM2G1nDBu9EZTOm40LYHvpP34d37SWYM1JmYHvSg,250368
 maxframe/core/graph/core.pyx,sha256=WYlYtXXSs72vfhf2ttJO-4u85exYzy2J9mlALHOMqoA,16354
 maxframe/core/graph/entity.py,sha256=RT_xbP5niUN5D6gqZ5Pg1vUegHn8bqPk8G8A30quOVA,5730
 maxframe/core/graph/builder/__init__.py,sha256=vTRY5xRPOMHUsK0jAtNIb1BjSPGqi_6lv86AroiiiL4,718
@@ -54,9 +54,9 @@ maxframe/core/operator/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH
 maxframe/core/operator/tests/test_core.py,sha256=iqZk4AWubFLO24V_VeV6SEy5xrzBFLP9qKK6tKO0SGs,1755
 maxframe/core/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe/core/tests/test_mode.py,sha256=fyRH-ksa6MogEs6kNhtXhCZyvhYqflgaXJYI3nSo-ps,2507
-maxframe/dataframe/__init__.py,sha256=t_Fnu2MNCgOr_N7mMqGU3LeZXvBF1yOHtLb5u1PoHt0,2187
+maxframe/dataframe/__init__.py,sha256=BdHZ903J7SkZDOxHDxNZvDxxXjSW8Gj-fJiHG26p-kM,2275
 maxframe/dataframe/arrays.py,sha256=rOvhxMQars9E3SOYSu0ygBuuRVY0QV6xzengnMqKs4s,29616
-maxframe/dataframe/core.py,sha256=63TqgmOCTr5wtwbILJ6bLtHZQ6xnZd7LvVrMqR27GDQ,76034
+maxframe/dataframe/core.py,sha256=mGlHzJqs9EZa0qLxxe-Qp0Kd5SnsVI3r6itDhLxVdGM,76126
 maxframe/dataframe/initializer.py,sha256=WW96yQjquofNFt6RPZvgWW4SBmH0OEDj8-BxpuyKThY,10552
 maxframe/dataframe/operators.py,sha256=jl611oPN5TGpf6UDuIwcLUsjmTcbVBNLLd6cvq8TvKo,8144
 maxframe/dataframe/utils.py,sha256=5bk441fXa9dpeT98g8OY8TtMypuoB3XSc3rLA0gjPL4,45371
@@ -119,15 +119,16 @@ maxframe/dataframe/datasource/from_records.py,sha256=ygpKOMXZnDdWzGxMxQ4KdGv-tJF
 maxframe/dataframe/datasource/from_tensor.py,sha256=mShHYi0fZcG7ZShFVgIezaphh8tSFqR9-nQMm5YKIhw,15146
 maxframe/dataframe/datasource/index.py,sha256=X_NShW67nYJGxaWp3qOrvyInNkz9L-XHjbApU4fHoes,4518
 maxframe/dataframe/datasource/read_csv.py,sha256=IvQihmpcZIdzSD7ziX92aTAHNyP5WnTgd2cZz_h43sQ,24668
-maxframe/dataframe/datasource/read_odps_query.py,sha256=QKU_7R6dd6GXzNX4IN9uwu3jB17MfhTr2nUtx0LFSNk,10116
-maxframe/dataframe/datasource/read_odps_table.py,sha256=ocSKQQ7SwIkzliFCFWNzy3e8J3CBZsf4971oRdlgdks,9294
+maxframe/dataframe/datasource/read_odps_query.py,sha256=rvlFp35g0vYIRZszhMSjHt5tAR668ir1GVBb-T80TQk,10224
+maxframe/dataframe/datasource/read_odps_table.py,sha256=r_VbiWWgBpJArBBB-NJCMRUbCTfUrGEOMJmq9a2TIC8,9380
 maxframe/dataframe/datasource/read_parquet.py,sha256=SZPrWoax2mwMBNvRk_3lkS72pZLe-_X_GwQ1JROBMs4,14952
 maxframe/dataframe/datasource/series.py,sha256=elQVupKETh-hUHI2fTu8TRxBE729Vyrmpjx17XlRV-8,1964
 maxframe/dataframe/datasource/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe/dataframe/datasource/tests/test_datasource.py,sha256=UumRBjE-bIuCi7Z4_3t8qb58ZcF8ePRZf3xF7DTvqIA,15041
 maxframe/dataframe/datastore/__init__.py,sha256=MmlHYvFacMReOHDQMXF-z2bCsLyrSHYBVwIlCsZGOK4,810
-maxframe/dataframe/datastore/to_csv.py,sha256=lheaF3ZmBPrcwcWyhK5gEVAGIaLJbvTyVAzqZFGG7eM,8026
-maxframe/dataframe/datastore/to_odps.py,sha256=Ml_iF9AspqIgGbeOAjTF3ukAwv-7SizribuqXZdxuXo,5776
+maxframe/dataframe/datastore/core.py,sha256=HCqrZN47RP-IC6zDqLX_RErDUAWkcTB58FHNU70V2b4,762
+maxframe/dataframe/datastore/to_csv.py,sha256=sns4bBgNpq7Ihb-goNqaBRdiEtrG-V6jqhNkWGZ1YaE,7974
+maxframe/dataframe/datastore/to_odps.py,sha256=NVHLccpNYbF6YPk3PKziStonkKlR0JaOFF0AxWlMhBw,5724
 maxframe/dataframe/extensions/__init__.py,sha256=x6QCVQIfpa8JP2Vu-nZwHJ1CzATnyPoKCBMqxjXwpO0,1439
 maxframe/dataframe/extensions/accessor.py,sha256=0OA8YPL3rofSvdU0z_1kMLImahrvow_vhxdQDYODki0,1497
 maxframe/dataframe/extensions/reshuffle.py,sha256=yOlJ-3R4v9CoiEKFA1zgCOvbocy00MxpFBbQuTn-uDw,2720
@@ -159,7 +160,7 @@ maxframe/dataframe/indexing/loc.py,sha256=senwgO_ijLJtbzaeqS_CMefV8nlf3guEQXKdSQ
 maxframe/dataframe/indexing/reindex.py,sha256=v4Rd85aNfh3onzcFqOhdUjiLrDv9QuNtGh-OaWpnG-4,19699
 maxframe/dataframe/indexing/rename.py,sha256=E7gI6lHGoBbMnldtErxv5StmS7jrGDdXGtpDusavihA,14009
 maxframe/dataframe/indexing/rename_axis.py,sha256=ugKcve4Kp8EuSmokQFUL-mVhGQ1cd6IDZ3UauHPiFeQ,6511
-maxframe/dataframe/indexing/reset_index.py,sha256=h77bfAHkh1mknUwOBsnTmickejKikW6IUMXmhn6rjcg,14410
+maxframe/dataframe/indexing/reset_index.py,sha256=_NFQZTjHzc_IgiqC-aqFJUfjneyJUN42-ujxGPfBVnQ,13524
 maxframe/dataframe/indexing/sample.py,sha256=cVpmTV4q0Lo5dK3RdIpP3G5Yo6A6rwCRcqQ-rBEKnPs,8393
 maxframe/dataframe/indexing/set_axis.py,sha256=ECRV5rRfbsKAQ90nEZVWCtGyu_0hN8ZPTmWNRGIJ0zo,5724
 maxframe/dataframe/indexing/set_index.py,sha256=XHX9CA0nvPd0War2GTKgr_FKuir_Tiu1bfQ5qz3vBKo,2180
@@ -283,7 +284,7 @@ maxframe/learn/contrib/pytorch/tests/test_pytorch.py,sha256=GHP-oD5uMU8LD90Jt2cH
 maxframe/lib/__init__.py,sha256=_PB28W40qku6YiT8fJYqdmEdRMQfelOwGeksCOZJfCc,657
 maxframe/lib/compression.py,sha256=QQpNK79iUC9zck74I0HKMhapSRnLBXtTRyS91taEVIc,1497
 maxframe/lib/functools_compat.py,sha256=2LTrkSw5i-z5E9XCtZzfg9-0vPrYxicKvDjnnNrAL1Q,2697
-maxframe/lib/mmh3.cp38-win_amd64.pyd,sha256=HMgH8mFYmm50QB_5WtKQ7UIOgGnFUGJoPzGKKfApJVs,17408
+maxframe/lib/mmh3.cp38-win_amd64.pyd,sha256=wkfm7AMJoM3aVDDpH2BKcmmDwkj55JLalxBr6TE0Vq8,17408
 maxframe/lib/version.py,sha256=VOVZu3KHS53YUsb_vQsT7AyHwcCWAgc-3bBqV5ANcbQ,18941
 maxframe/lib/wrapped_pickle.py,sha256=bzEaokhAZlkjXqw1xfeKO1KX2awhKIz_1RT81yPPoag,3949
 maxframe/lib/aio/__init__.py,sha256=xzIYnV42_7CYuDTTv8svscIXQeJMF0nn8AXMbpv173M,963
@@ -332,13 +333,13 @@ maxframe/lib/tblib/pickling_support.py,sha256=D9A0eX7gJeyqhXWxJJZ1GRwwcc5lj86wBR
 maxframe/lib/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe/lib/tests/test_wrapped_pickle.py,sha256=WV0EJQ1hTSp8xjuosWWtEO7PeiBqdDUYgStxp72_c94,1575
 maxframe/odpsio/__init__.py,sha256=0SesD04XxFli4Gp23ipMkefFQ2ZTB0PItwZoSHpDC-k,820
-maxframe/odpsio/arrow.py,sha256=fYNQa_n-Dy-y3l9oAu50-fOsV3C1GaNZpBsOhlYO8i4,3732
-maxframe/odpsio/schema.py,sha256=clW2pgjGY6lsHSfyKx-M6xqvPJerCsqlrGm6ommFHmY,12068
+maxframe/odpsio/arrow.py,sha256=vvi9g5IZ8Wn224obv3vOO8U1X6UlNNTxSOfDA4Dvx0M,3875
+maxframe/odpsio/schema.py,sha256=Hba-eCXnBUS6NxHRsshaohzO1eThm4HeVzzvAF7E3Vg,12479
 maxframe/odpsio/tableio.py,sha256=r_Y47V5NhvLHsK7kypdswSDmOWag1Zoh9-FEjKe5MG0,9660
 maxframe/odpsio/volumeio.py,sha256=IT_OO-RG2rJZOEx8C8xRr0oNR358RSAJQAp6WGxeXzI,3838
 maxframe/odpsio/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe/odpsio/tests/test_arrow.py,sha256=yeDWFzsm2IMS-k6jimlQ7uim5T6WW1Anuy8didaf4cs,3194
-maxframe/odpsio/tests/test_schema.py,sha256=cXQYEhETUSoJmQ7MW33bFyR2ho6WciU4ot9LnxFsARM,10839
+maxframe/odpsio/tests/test_schema.py,sha256=BI2Eq_DkyCimJEjSylAm_z3SLVNdEzKto7nqu7uyiW0,12088
 maxframe/odpsio/tests/test_tableio.py,sha256=ZyQxBAVA5GG3j_NOPTTFs5vCQqQywhRKC9OAJx9LJxM,4789
 maxframe/odpsio/tests/test_volumeio.py,sha256=xvnrPZueZ76OAWK2zW_tHHI_cDxo7gJXTHiEe0lkmjk,3112
 maxframe/remote/__init__.py,sha256=Yu1ZDLICbehNfd1ur7_2bnIn2VFIsTxH_cILCbHAeZY,747
@@ -346,7 +347,7 @@ maxframe/remote/core.py,sha256=w_eTDEs0O7iIzLn1YrMGh2gcNAzzbqV0mx2bRT7su_U,7001
 maxframe/remote/run_script.py,sha256=k93-vaFLUanWoBRai4-78DX_SLeZ8_rbbxcCtOIXZO8,3677
 maxframe/serialization/__init__.py,sha256=nxxU7CI6MRcL3sjA1KmLkpTGKA3KG30FKl-MJJ0MCdI,947
 maxframe/serialization/arrow.py,sha256=OMeDjLcPgagqzokG7g3Vhwm6Xw1j-Kph1V2QsIwi6dw,3513
-maxframe/serialization/core.cp38-win_amd64.pyd,sha256=pm3uyk95Wnhn8B_LPdI9pV_gHACS9uBywUES38pY2OI,397824
+maxframe/serialization/core.cp38-win_amd64.pyd,sha256=gBaf2kOFy6elDw4BCzayrqDqK84HpYn9vBrat_XEHdo,397824
 maxframe/serialization/core.pxd,sha256=Fymih3Wo-CrOY27_o_DRINdbRGR7mgiT-XCaXCXafxM,1347
 maxframe/serialization/core.pyx,sha256=Qmipu3LiJGIBVy_7d4tSJqcYWnG5xj2I7IaPv2PSq5E,35078
 maxframe/serialization/exception.py,sha256=e7bZyPlZ8XhSCdeOwlYreq0HazPXKOgOA6r9Q4Ecn2Y,3113
@@ -616,14 +617,14 @@ maxframe_client/clients/spe.py,sha256=uizNBejhU_FrMhsgsFgDnq7gL7Cxk803LeLYmr3nmx
 maxframe_client/session/__init__.py,sha256=9zFCd3zkSADESAFc4SPoQ2nkvRwsIhhpNNO2TtSaWbU,854
 maxframe_client/session/consts.py,sha256=nD-D0zHXumbQI8w3aUyltJS59K5ftipf3xCtHNLmtc8,1380
 maxframe_client/session/graph.py,sha256=GSZaJ-PV4DK8bTcNtoSoY5kDTyyIRAKleh4tOCSUbsI,4470
-maxframe_client/session/odps.py,sha256=RG7_28UaS_8tgJUOa4ohw2QtwX2fF4yqsGGy2MksQWI,16700
+maxframe_client/session/odps.py,sha256=AZKFz1_Pl8-zJ_RwxBoN00uUDKf4MbzhvVZ5cTywURM,16781
 maxframe_client/session/task.py,sha256=R8x8OERIb673vTq-o0ig6Zy2NT4_jvi8AbLhyMaljo8,11409
 maxframe_client/session/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe_client/session/tests/test_task.py,sha256=861usEURVXeTUzfJYZmBfwsHfZFexG23mMtT5IJOOm4,3364
 maxframe_client/tests/__init__.py,sha256=29eM5D4knhYwe3TF42naTuC5b4Ym3VeH4rK8KpdLWNY,609
 maxframe_client/tests/test_fetcher.py,sha256=7iYXLMIoCJLfgUkjB2HBkV-sqQ-xGlhtzfp9hRJz_kM,3605
-maxframe_client/tests/test_session.py,sha256=s8pxf0I6PjOd6ZZQ4IYjfOM_3F3wf6SGPRMz0tAZFmo,6514
-maxframe-0.1.0b3.dist-info/METADATA,sha256=aZIVKoKmfRSjIExk2K16jB0fJj5iApw9FhoHqv6hs_k,3147
-maxframe-0.1.0b3.dist-info/WHEEL,sha256=Wb4yjwIXVKEpht4JWFUZNCzpG7JLBNZnqtK2YNdqLkI,100
-maxframe-0.1.0b3.dist-info/top_level.txt,sha256=64x-fc2q59c_vXwNUkehyjF1vb8JWqFSdYmUqIFqoTM,31
-maxframe-0.1.0b3.dist-info/RECORD,,
+maxframe_client/tests/test_session.py,sha256=_i6ICUTg6LMs-KD3w3pYNSt5_k9gCOkMx4gTfE20CF8,7130
+maxframe-0.1.0b4.dist-info/METADATA,sha256=uH-sH39DV-QlQoIEKtD3W7Os0Z99jZgTs7uyKf-t30Q,3147
+maxframe-0.1.0b4.dist-info/WHEEL,sha256=Wb4yjwIXVKEpht4JWFUZNCzpG7JLBNZnqtK2YNdqLkI,100
+maxframe-0.1.0b4.dist-info/top_level.txt,sha256=64x-fc2q59c_vXwNUkehyjF1vb8JWqFSdYmUqIFqoTM,31
+maxframe-0.1.0b4.dist-info/RECORD,,

maxframe_client/session/odps.py

@@ -115,7 +115,7 @@ class MaxFrameSession(ToThreadMixin, IsolatedAsyncSession):
     ):
         super().__init__(address, session_id)
         self.timeout = timeout
-        self._odps_entry = odps_entry or ODPS.from_environments()
+        self._odps_entry = odps_entry or ODPS.from_global() or ODPS.from_environments()
         self._tileable_to_infos = weakref.WeakKeyDictionary()
 
         self._caller = self._create_caller(odps_entry, address, **kwargs)
@@ -147,15 +147,16 @@ class MaxFrameSession(ToThreadMixin, IsolatedAsyncSession):
         data = t.op.get_data()
         batch_size = options.session.upload_batch_size
 
-        halo_client = HaloTableIO(self._odps_entry)
-        with halo_client.open_writer(table_obj.full_table_name) as writer:
-            for batch_start in range(0, len(data), batch_size):
-                if isinstance(data, pd.Index):
-                    batch = data[batch_start : batch_start + batch_size]
-                else:
-                    batch = data.iloc[batch_start : batch_start + batch_size]
-                arrow_batch, _ = pandas_to_arrow(batch)
-                writer.write(arrow_batch)
+        if len(data):
+            halo_client = HaloTableIO(self._odps_entry)
+            with halo_client.open_writer(table_obj.full_table_name) as writer:
+                for batch_start in range(0, len(data), batch_size):
+                    if isinstance(data, pd.Index):
+                        batch = data[batch_start : batch_start + batch_size]
+                    else:
+                        batch = data.iloc[batch_start : batch_start + batch_size]
+                    arrow_batch, _ = pandas_to_arrow(batch)
+                    writer.write(arrow_batch)
 
         read_tileable = read_odps_table(
             table_obj.full_table_name,

maxframe_client/tests/test_session.py

@@ -28,6 +28,7 @@ from maxframe.lib.aio import stop_isolation
 from maxframe.protocol import ResultInfo
 from maxframe.serialization import RemoteException
 from maxframe.session import new_session
+from maxframe.tests.utils import tn
 from maxframe.utils import build_temp_table_name
 from maxframe_framedriver.app.tests.test_framedriver_webapp import (  # noqa: F401
     framedriver_app,
@@ -115,6 +116,26 @@ def test_simple_run_dataframe(start_mock_session):
     assert not odps_entry.exist_table(build_temp_table_name(start_mock_session, key))
 
 
+def test_run_empty_table(start_mock_session):
+    odps_entry = ODPS.from_environments()
+
+    table_name = tn("test_session_empty_table")
+    odps_entry.delete_table(table_name, if_exists=True)
+    empty_table = odps_entry.create_table(
+        table_name, "_idx_0 bigint, a double, b double", lifecycle=1
+    )
+    df = md.read_odps_table(table_name, index_col="_idx_0")
+    df["d"] = df["a"] + df["b"]
+
+    executed = df.execute()
+    assert "Index: []" in str(executed)
+
+    fetched = executed.fetch()
+    assert 0 == len(fetched)
+
+    empty_table.drop()
+
+
 def test_run_dataframe_with_pd_source(start_mock_session):
     odps_entry = ODPS.from_environments()