maxframe 0.1.0b3__cp39-cp39-macosx_10_9_x86_64.whl → 0.1.0b4__cp39-cp39-macosx_10_9_x86_64.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

@@ -1,42 +1,42 @@
-maxframe-0.1.0b3.dist-info/RECORD,,
-maxframe-0.1.0b3.dist-info/WHEEL,sha256=wEvD9HODgqXG21zAA07x3Gh0qpNEHOQ7qzjMqXvopac,109
-maxframe-0.1.0b3.dist-info/top_level.txt,sha256=64x-fc2q59c_vXwNUkehyjF1vb8JWqFSdYmUqIFqoTM,31
-maxframe-0.1.0b3.dist-info/METADATA,sha256=056zhE5bxFHo_1CAOnpi1A2Ov1JI8hg435HsygF02lA,3043
+maxframe-0.1.0b4.dist-info/RECORD,,
+maxframe-0.1.0b4.dist-info/WHEEL,sha256=wEvD9HODgqXG21zAA07x3Gh0qpNEHOQ7qzjMqXvopac,109
+maxframe-0.1.0b4.dist-info/top_level.txt,sha256=64x-fc2q59c_vXwNUkehyjF1vb8JWqFSdYmUqIFqoTM,31
+maxframe-0.1.0b4.dist-info/METADATA,sha256=M5MjCW0gBbtdN7bOgXnAyVv3__iRHgIGuqRb_qHAHi4,3043
 maxframe_client/conftest.py,sha256=7cwy2sFy5snEaxvtMvxfYFUnG6WtYC_9XxVrwJxOpcU,643
 maxframe_client/__init__.py,sha256=3b-z0oFVVwtIzVFBxOb9pw7gz4IhTSh4FiHtVgnxS4Q,724
 maxframe_client/fetcher.py,sha256=Ys_qu2qtniXuj9YSfeHvevdrAAEgm8k4YjyoZusdVmg,6813
 maxframe_client/clients/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
 maxframe_client/clients/spe.py,sha256=ArZMNQ7olicI4O1JO7CyRP7-hb60DF71ZKCTO0N39uE,3593
 maxframe_client/clients/framedriver.py,sha256=Rn09529D2qBTgNGc0oCY0l7b3FgzT87TqS1nujGQaHw,4463
-maxframe_client/tests/test_session.py,sha256=75mxU5UTWjb1loQLo9BAkg7BzKuj4vYX49j4vP1-8fA,6307
+maxframe_client/tests/test_session.py,sha256=ZaiQTsnn-qr3VzdyuSei05AeM3FYhVpXQizg50mcSRc,6902
 maxframe_client/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
 maxframe_client/tests/test_fetcher.py,sha256=q7kYCznM6WSxx9TCbHrxs7Zy1L2a5zu9D-Pi1XNgQzg,3516
 maxframe_client/session/task.py,sha256=z5j8qtBM6cs_lZrvfy4Ji3F3sVOhPOCr5r1RsNe7rN4,11102
 maxframe_client/session/graph.py,sha256=nwILNOIVaIf4E3xWffTAAlRsKRYU_zGW3oVO10du8Xw,4351
 maxframe_client/session/__init__.py,sha256=KPqhSlAJiuUz8TC-z5o7mHDVXzLSqWwrZ33zNni7piY,832
 maxframe_client/session/consts.py,sha256=R37BxDF3kgCy0qmDdwLaH5jB7mb7SzfYV6g9yHBKAwk,1344
-maxframe_client/session/odps.py,sha256=uuPk5rc1OpFk9PNFU1R6r4HQXOHAOU7ZVvbcdGk_N_s,16248
+maxframe_client/session/odps.py,sha256=4O4lpoEICC7PwDu6koGZ1XOXuCt_zSblF6xtPARl7n8,16328
 maxframe_client/session/tests/test_task.py,sha256=lDdw3gToaM3xSaRXEmHUoAo2h0id7t4v_VvpdKxQAao,3279
 maxframe_client/session/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
 maxframe/_utils.pyx,sha256=I4kmfhNr-xXK2ak22dr4Vwahzn-JmTaYctbL3y9_UBQ,17017
 maxframe/conftest.py,sha256=ZzwKhGp7LAVpzQYJkniwIUQqIegcaDQAhyzDyU2qld4,4264
-maxframe/opcodes.py,sha256=L1_GGenexNnD4dk1ueelgpAebHCC-cY7d--1QPhw8l4,9984
+maxframe/opcodes.py,sha256=FE9MUlENCKaO2gVIaSweF5Rsti6nKfYoD-vSRkoPlgM,10145
 maxframe/env.py,sha256=_K499f7giN7Iu9f39iI9p_naaEDoJ0rx8dInbzqFOVI,1402
 maxframe/mixin.py,sha256=HBAeWYGb7N6ZIgkA-YpkKiSY1GetcEVNTuMb0ieznBs,3524
 maxframe/protocol.py,sha256=LjjE6iw0ZVx82tBMbff4izkGuiJxRG0MTOaPYYpRL10,14190
-maxframe/session.py,sha256=TspAq0EUhi-7VQb1kAuQyjhpX8vMYqgdk4vzbGCkJcY,35344
+maxframe/session.py,sha256=y1XSn2G5PAQenuAkz2hb54CuWStbb_S8IFdMcJiNyoQ,35418
 maxframe/__init__.py,sha256=YGnga-nYEppPoDZoZ5s64PZ0RYLaWtcYtmYSLTjKUBE,976
-maxframe/utils.py,sha256=BrMJLO0-iPKABkNztXWpqiQWHoe-X75HGiCJr5aeles,33914
+maxframe/utils.py,sha256=0cAfRm_M5vfSGzaLgLqEu8M75nYK9BOd_lQowejpdHY,34104
 maxframe/extension.py,sha256=4IzF9sPyaRoAzLud0iDLooOgcyq4QunXH0ki3q9Hn8I,2374
 maxframe/errors.py,sha256=nQZoLGdLJz-Uf9c2vUvQl08QMvRqtkBOhKfdPYRZA4o,690
 maxframe/udf.py,sha256=tWOMTkNqGWw3ZpNC9wEU0GGNSBV8sV7E-Ye80DI28eg,2241
-maxframe/_utils.cpython-39-darwin.so,sha256=-0bVNg7VTmXNgn_QM6AG8QBuR80wx2tSP8OuzLUiQPs,472632
+maxframe/_utils.cpython-39-darwin.so,sha256=clk1nu3odUr-0DHv5Yr6gYGONMfH0ASYm22lGE9ZDUo,472632
 maxframe/typing_.py,sha256=fzHETru3IOZAJwU9I7n_ib3wHuQRJ9XFVmAk7WpqkMo,1096
 maxframe/codegen.py,sha256=pSiGHoEo9YajFPHbFHvi7fGkiJmAQdBCe0mMXNOG6-U,15846
 maxframe/_utils.pxd,sha256=AhJ4vA_UqZqPshi5nvIZq1xgr80fhIVQ9dm5-UdkYJ8,1154
 maxframe/dataframe/arrays.py,sha256=RWzimUcrds5CsIlPausfJAkLUjcktBSSXwdXyUNKEtU,28752
-maxframe/dataframe/__init__.py,sha256=tpbt4OgW4hoU_1OsQu1WSxbpZD4YPXW_oT4xOxNPNqE,2112
-maxframe/dataframe/core.py,sha256=mJ5I_qBP8HDAzlwPnkj8RN4EyBHC8d5unbe2LC6HKXg,73641
+maxframe/dataframe/__init__.py,sha256=h4SQPQQBpVPxTNGP5Rtx5uO7RI8gO7_FpXjeh_k9PN8,2195
+maxframe/dataframe/core.py,sha256=aNkexKwBuaQv28R0HYTLXcD64U4nEcBsBfLfGCFSSGs,73731
 maxframe/dataframe/initializer.py,sha256=4BpZJB8bbyFnABUYWBrk_qzzrogEsWgFuU21Ma9IsjY,10264
 maxframe/dataframe/utils.py,sha256=qWRo51rcMTlo4mvZ8ZZq1zIF9CiAgU1qRtoCAaYrR34,44111
 maxframe/dataframe/operators.py,sha256=T7Ik1shfoyrMZ1x0wHXG26bsod1_YjMGQgGAFNpH6k0,7871
@@ -83,10 +83,10 @@ maxframe/dataframe/datasource/from_index.py,sha256=2061zsQn-BhyHTT0X9tE0JK8vLxQU
 maxframe/dataframe/datasource/dataframe.py,sha256=LxAKF4gBIHhnJQPuaAUdIEyMAq7HTfiEeNVls5n4I4A,2023
 maxframe/dataframe/datasource/series.py,sha256=QcYiBNcR8jjH6vdO6l6H9F46KHmlBqVCTI2tv9eyZ9w,1909
 maxframe/dataframe/datasource/__init__.py,sha256=C8EKsHTJi-1jvJUKIpZtMtsK-ZID3dtxL1voXnaltTs,640
-maxframe/dataframe/datasource/read_odps_query.py,sha256=nj8l38S0iVAXgNXgzDFERO-HNp6lJ1GahImwDpEzXXw,9821
+maxframe/dataframe/datasource/read_odps_query.py,sha256=PkDtM1lG2odWcgjtKOdr5A95rqlkiFUKQy03v5-4pqU,9927
 maxframe/dataframe/datasource/core.py,sha256=ozFmDgw1og7nK9_jU-u3tLEq9pNbitN-8w8XWdbKkJ0,2687
 maxframe/dataframe/datasource/date_range.py,sha256=CDGpxDyjLwnb66j-MIiiTfXGXHGh5MLhEmj6x2riIlU,17244
-maxframe/dataframe/datasource/read_odps_table.py,sha256=VCqWxJswcjujVoUNXb2kbTkOZroMkFCg5n272Yn7ME4,9045
+maxframe/dataframe/datasource/read_odps_table.py,sha256=u3TNBlSMvQzKMQXFincqWKcUu8a4jeXX_RW3gctPXF4,9129
 maxframe/dataframe/datasource/read_parquet.py,sha256=9auOcy8snTxCOohgXZCUXfT_O39irdkBngZH5svgx0E,14531
 maxframe/dataframe/datasource/from_tensor.py,sha256=4viuN5SLLye7Xeb8kouOpm-osoQ2yEovWTDNPQuW8gE,14727
 maxframe/dataframe/datasource/from_records.py,sha256=WBYouYyg7m_8NJdN-yUWSfJlIpm6DVP3IMfLXZFugyI,3442
@@ -138,8 +138,9 @@ maxframe/dataframe/groupby/apply.py,sha256=gZVHN8nMXoy7BEHTBAVLQKtGicQ_jB3dsny8j
 maxframe/dataframe/groupby/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
 maxframe/dataframe/groupby/tests/test_groupby.py,sha256=VsHDG-UaZUXlZ1l498y44ecfzJJ9D2oY4QmPCjS6wBs,12142
 maxframe/dataframe/datastore/__init__.py,sha256=Ujd4ix4UJ7Zq-sV2pXTRDvZkKUvtt8sr-FfiUoi6Kh4,784
-maxframe/dataframe/datastore/to_csv.py,sha256=VUTocKlrDVyOs3Cvve9NwHtiKNJ2sQTnYnG11ML3kxE,7799
-maxframe/dataframe/datastore/to_odps.py,sha256=h8WOZ759Q9aHw8HdCAws0T979_SYXuSP5NcUgT2cTPw,5614
+maxframe/dataframe/datastore/core.py,sha256=hLGhqYxX73tq9sOFxMyYBRMawTnsVQqluW5FcE3YHfE,743
+maxframe/dataframe/datastore/to_csv.py,sha256=xVfpEsicD5A5uXLSdCN08K8uGyWB4QxnRcAbgBVUzbs,7747
+maxframe/dataframe/datastore/to_odps.py,sha256=wZzD3yc6YQAGYVwEzYY-qxscn8Sj9I43lR70tFHe3m0,5562
 maxframe/dataframe/fetch/__init__.py,sha256=E3VjxLRKjtr-D__K-c0aRETvBG7CQRG1iEtH0Qghq0s,653
 maxframe/dataframe/fetch/core.py,sha256=z61_orBtOIrKFpFdJTNqyPhOw5ZCGe6QkXnDrGdMy6U,3338
 maxframe/dataframe/reduction/max.py,sha256=CqaueIN3cuF7vlE2rt2MAcXAD3Syd_R8W2dzcFhTmV0,1660
@@ -233,7 +234,7 @@ maxframe/dataframe/arithmetic/radians.py,sha256=pSq9S0KlV0Xx0vGSdxdol3348dSyYuxG
 maxframe/dataframe/arithmetic/sin.py,sha256=zgWoijVpASlkZ7cnZZxKSSHUiLqQ0pKUAIOHEfCBubc,915
 maxframe/dataframe/arithmetic/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
 maxframe/dataframe/arithmetic/tests/test_arithmetic.py,sha256=Zm0hwTo_NqFAa3iDmYF2Q10-884d8ltOsby2WjcioBA,24372
-maxframe/dataframe/indexing/reset_index.py,sha256=pdLv7YwyY-lMGg4qegGmRzu-MLNjg--xESZ4jeDcl1I,13992
+maxframe/dataframe/indexing/reset_index.py,sha256=uzGszHfkhwZgZtEaygy5UPrvHPNPXgWZaTOXdNxGBT8,13122
 maxframe/dataframe/indexing/iat.py,sha256=ANURJ8qn_UitgM01gDPeaQOlSBxbk0pmN4Yr-iPRXM8,1127
 maxframe/dataframe/indexing/loc.py,sha256=ZmiK3a2f-krkXFvNLSlzRRa6GWGiAAXk_3mWZC_MqdQ,14845
 maxframe/dataframe/indexing/align.py,sha256=BdrfIrf6A7v8W0GIuv1mdx87CzR15ARnqIH_aWUjWqY,12089
@@ -273,7 +274,7 @@ maxframe/core/entity/objects.py,sha256=RMHLTGbIHZNxxX59lAuQydAKcR32qKleIYUqdElGS
 maxframe/core/entity/fuse.py,sha256=47U6MHRzA2ZvUi-kJb7b3mC_gN07x3yebBgX2Jj7VZo,2277
 maxframe/core/entity/chunks.py,sha256=yNSLCWOpA_Z6aGr6ZI32dIJf3xPdRBWbvdsl8sTM3BE,2134
 maxframe/core/graph/__init__.py,sha256=rnsXwW0ouh1f7SVtq73-PzLE-MBM6Op_0l6J7b7wGRE,821
-maxframe/core/graph/core.cpython-39-darwin.so,sha256=JyoianB3w6hvC1hEN5m-sSMxSzMgVuHZEqlxiCqkPWY,361136
+maxframe/core/graph/core.cpython-39-darwin.so,sha256=x0X9hBLzRMexkwWkUgCWPaKpCtAZSk8usunXor2vlsU,361136
 maxframe/core/graph/entity.py,sha256=56gjXyDXN-TTPm3AQOxuRVQbb_fguKFDL_Xm7i95XEk,5559
 maxframe/core/graph/core.pyx,sha256=ZJPx_MTOBMaX-6mns6tAiu-wrIBvRAKN44YAGTypJ1Y,15887
 maxframe/core/graph/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
@@ -294,7 +295,7 @@ maxframe/core/operator/fuse.py,sha256=0RGemF99gQCwV4aEk-K6T5KAGToO-487dFk8LyYDIZ
 maxframe/core/operator/base.py,sha256=nxuSKjbBzDrItM9PGmFo8RLwParazu525jMLWj0kXkM,15251
 maxframe/core/operator/tests/test_core.py,sha256=57aICnc5VLqdVK7icAORTWC81bSjBxeeVWIJcha9J_0,1691
 maxframe/core/operator/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
-maxframe/config/config.py,sha256=3lQj99eMGg9MBW1gMJAdGFD88UEcdZf71sHgAHXASAk,13045
+maxframe/config/config.py,sha256=zWw0kzlkNbsKY5_oee2e5mdWqC_qrxor42SRkA-hSz0,13145
 maxframe/config/validators.py,sha256=2m9MrkjDUFiU4PPaWIw8tjwMaOy8AYmuJFqVnnY8IMY,1615
 maxframe/config/__init__.py,sha256=g5lN3nP2HTAXa6ExGxU1NwU1M9ulYPmAcsV-gU7nIW8,656
 maxframe/config/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
@@ -307,7 +308,7 @@ maxframe/serialization/arrow.py,sha256=VnGxNLU9UV_cUPTze43bEFCIbYLAOZnp2pAwVJbAI
 maxframe/serialization/__init__.py,sha256=9eSnoDww1uw2DAXEBBTB2atJQHzd-38XVxrCkoaypxA,921
 maxframe/serialization/maxframe_objects.py,sha256=R9WEjbHL0Kr56OGkYDU9fcGi7gII6fGlXhi6IyihTsM,1365
 maxframe/serialization/numpy.py,sha256=8_GSo45l_eNoMn4NAGEb9NLXY_9i4tf9KK4EzG0mKpA,3213
-maxframe/serialization/core.cpython-39-darwin.so,sha256=73Gy4lTdTxSp10w4Wn2HlIDyWtS5NSIxM72I9PUSiUs,695248
+maxframe/serialization/core.cpython-39-darwin.so,sha256=Yk95GGq96MQvY8nK32haE24gtgsVl-o-sdM-sViW3ko,695248
 maxframe/serialization/scipy.py,sha256=hP0fAW0di9UgJrGtANB2S8hLDbFBtR8p5NDqAMt5rDI,2427
 maxframe/serialization/core.pyx,sha256=AATN47RdBTq2zg7--3xX2VHyAZSvoAuYRt7B7gEgKPE,33984
 maxframe/serialization/tests/test_serial.py,sha256=Wj_I6CBQMaOtE8WtqdUaBoU8FhBOihht6SfeHOJV-zU,12511
@@ -320,13 +321,13 @@ maxframe/serialization/serializables/tests/test_field_type.py,sha256=T3ebXbUkKve
 maxframe/serialization/serializables/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
 maxframe/serialization/serializables/tests/test_serializable.py,sha256=nwdN7B2xnI_Bh-s90TyjPvyFFVWOE9JVBqm4bdwYZ6o,8243
 maxframe/odpsio/tableio.py,sha256=QjjcNxb8h0OXvif3xDarsjq-52aqsWdPezpSn46__g0,9360
-maxframe/odpsio/arrow.py,sha256=S9wgULzuUeqQNhBuEXp2yaAA5E9EZM0PTteqM345tRA,3645
+maxframe/odpsio/arrow.py,sha256=jNbfb36TDzpMjloDyXkwCfGDiCe-8z3IGMNlF_s1fIs,3783
 maxframe/odpsio/__init__.py,sha256=HcxZsE4hRwbhtE-ZXhDWZMmQlv-2dOTvQq2NajhGEo4,799
 maxframe/odpsio/volumeio.py,sha256=b2SQqusgrtkJJ6uMjnFv5s56XjchF8F4lLTTSHynRMc,3743
-maxframe/odpsio/schema.py,sha256=WsJ0bi1-78g6adyPKQFqFJRH3YPXMqGl_qmOe5z7me4,11717
+maxframe/odpsio/schema.py,sha256=aXvK_1BSwttuUyROyIa_HNHohLZBp7LrW9VXzHPGXyY,12115
 maxframe/odpsio/tests/test_tableio.py,sha256=5nKq8I8Pzrfl89BjIIGyrvqPRiXdejTcYCtd-R2vTAo,4653
 maxframe/odpsio/tests/__init__.py,sha256=FEFOVLi3o2GpZoedTtLYvbie0eQBehJIjtCrWca2ZHw,596
-maxframe/odpsio/tests/test_schema.py,sha256=yss1ly55ErYse95XMFq2s_GWL8UnwZga5RyNIhtcx70,10572
+maxframe/odpsio/tests/test_schema.py,sha256=pfBGHdgnCOcvuvEe66WCZDt_ijG2jaoA_lCKC9-ejYo,11796
 maxframe/odpsio/tests/test_arrow.py,sha256=SQ9EmI9_VOOC8u6Rg6nh3IPC2fPbLvJ9HwtpMNDRhL8,3106
 maxframe/odpsio/tests/test_volumeio.py,sha256=UEqFANuPKyFtlIh2JNi-LoixH52bxsgHdxu3himnEvs,3022
 maxframe/tests/test_utils.py,sha256=xaAoURr5NOJUTY0XVa2H8qOStcEH5UQSXItkatHFxFE,11977
@@ -336,7 +337,7 @@ maxframe/tests/utils.py,sha256=wJtSFXt3BD4i5zdO4JBQk_kNAxrtyGLro0jodCA4xuY,4568
 maxframe/tests/test_codegen.py,sha256=GMrnpSb2eyB_nmuv8-_p47Kw877ElKS3BP52SpqZNIQ,2208
 maxframe/lib/wrapped_pickle.py,sha256=xJa0wI-GsBZFKQpVnlh_hZBlQ2u1D8VO2aBIW7VOdP4,3810
 maxframe/lib/version.py,sha256=yQ6HkDOvU9X1rpI49auh-qku2g7gIiztgEH6v1urOrk,18321
-maxframe/lib/mmh3.cpython-39-darwin.so,sha256=h0knHWqb6EkQMYeDR3S4njVKljScfp8z4PjRh84-odI,38064
+maxframe/lib/mmh3.cpython-39-darwin.so,sha256=k1LxN-M5KCoPVF9uLkXcmQDNjieQSQ7DTrmA7wO5fvU,38064
 maxframe/lib/compression.py,sha256=k9DSrl_dNBsn5azLjBdL5B4WZ6eNvmCrdMbcF1G7JSc,1442
 maxframe/lib/__init__.py,sha256=CzfbLNqqm1yR1i6fDwCd4h1ptuKVDbURFVCb0ra7QNc,642
 maxframe/lib/functools_compat.py,sha256=PMSkct9GIbzq-aBwTnggrOLNfLh4xQnYTIFMPblzCUA,2616

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()