maxframe 0.1.0b1__cp311-cp311-win_amd64.whl → 0.1.0b3__cp311-cp311-win_amd64.whl

maxframe/dataframe/datasource/tests/test_datasource.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import os
 from collections import OrderedDict
 
 import numpy as np
@@ -22,7 +23,7 @@ from odps import ODPS
 from .... import tensor as mt
 from ....tests.utils import tn
 from ....utils import lazy_import
-from ... import read_odps_table
+from ... import read_odps_query, read_odps_table
 from ...core import DatetimeIndex, Float64Index, IndexValue, Int64Index, MultiIndex
 from ..dataframe import from_pandas as from_pandas_df
 from ..date_range import date_range
@@ -33,6 +34,7 @@ from ..from_tensor import (
 )
 from ..index import from_pandas as from_pandas_index
 from ..index import from_tileable
+from ..read_odps_query import ColumnSchema, _resolve_task_sector
 from ..series import from_pandas as from_pandas_series
 
 ray = lazy_import("ray")
@@ -228,6 +230,7 @@ def test_from_odps_table():
     assert df.op.table_name == test_table.full_table_name
     assert df.index_value.name is None
     assert isinstance(df.index_value.value, IndexValue.RangeIndex)
+    assert df.op.get_columns() == ["col1", "col2", "col3"]
     pd.testing.assert_series_equal(
         df.dtypes,
         pd.Series(
@@ -247,6 +250,7 @@ def test_from_odps_table():
     assert df.op.table_name == test_table.full_table_name
     assert df.index_value.name is None
     assert isinstance(df.index_value.value, IndexValue.RangeIndex)
+    assert df.op.get_columns() == ["col1", "col2"]
     pd.testing.assert_series_equal(
         df.dtypes,
         pd.Series([np.dtype("O"), np.dtype("int64")], index=["col1", "col2"]),
@@ -257,6 +261,7 @@ def test_from_odps_table():
     assert df.index_value.name == "col1"
     assert isinstance(df.index_value.value, IndexValue.Index)
     assert df.index.dtype == np.dtype("O")
+    assert df.op.get_columns() == ["col2", "col3"]
     pd.testing.assert_series_equal(
         df.dtypes,
         pd.Series([np.dtype("int64"), np.dtype("float64")], index=["col2", "col3"]),
@@ -267,6 +272,7 @@ def test_from_odps_table():
 
     df = read_odps_table(test_parted_table, append_partitions=True)
     assert df.op.append_partitions is True
+    assert df.op.get_columns() == ["col1", "col2", "col3", "pt"]
     pd.testing.assert_series_equal(
         df.dtypes,
         pd.Series(
@@ -280,6 +286,7 @@ def test_from_odps_table():
     )
     assert df.op.append_partitions is True
     assert df.op.partitions == ["pt=20240103"]
+    assert df.op.get_columns() == ["col1", "col2", "pt"]
     pd.testing.assert_series_equal(
         df.dtypes,
         pd.Series(
@@ -292,6 +299,67 @@ def test_from_odps_table():
     test_parted_table.drop()
 
 
+def test_from_odps_query():
+    odps_entry = ODPS.from_environments()
+    table1_name = tn("test_from_odps_query_src1")
+    table2_name = tn("test_from_odps_query_src2")
+    odps_entry.delete_table(table1_name, if_exists=True)
+    odps_entry.delete_table(table2_name, if_exists=True)
+    test_table = odps_entry.create_table(
+        table1_name, "col1 string, col2 bigint, col3 double", lifecycle=1
+    )
+    # need some data to produce complicated plans
+    odps_entry.write_table(test_table, [["A", 10, 3.5]])
+    test_table2 = odps_entry.create_table(
+        table2_name, "col1 string, col2 bigint, col3 double", lifecycle=1
+    )
+    odps_entry.write_table(test_table2, [["A", 10, 4.5]])
+
+    with pytest.raises(ValueError) as err_info:
+        read_odps_query(f"CREATE TABLE dummy_table AS SELECT * FROM {table1_name}")
+    assert "instant query" in err_info.value.args[0]
+
+    query1 = f"SELECT * FROM {table1_name} WHERE col1 > 10"
+    df = read_odps_query(query1)
+    assert df.op.query == query1
+    assert df.index_value.name is None
+    assert isinstance(df.index_value.value, IndexValue.RangeIndex)
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series(
+            [np.dtype("O"), np.dtype("int64"), np.dtype("float64")],
+            index=["col1", "col2", "col3"],
+        ),
+    )
+
+    df = read_odps_query(query1, index_col="col1")
+    assert df.op.query == query1
+    assert df.index_value.name == "col1"
+    assert isinstance(df.index_value.value, IndexValue.Index)
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series([np.dtype("int64"), np.dtype("float64")], index=["col2", "col3"]),
+    )
+
+    query2 = (
+        f"SELECT t1.col1, t1.col2, t1.col3 as c31, t2.col3 as c32 "
+        f"FROM {table1_name} t1 "
+        f"INNER JOIN {table2_name} t2 "
+        f"ON t1.col1 = t2.col1 AND t1.col2 = t2.col2"
+    )
+    df = read_odps_query(query2, index_col=["col1", "col2"])
+    assert df.op.query == query2
+    assert df.index_value.names == ["col1", "col2"]
+    assert isinstance(df.index_value.value, IndexValue.MultiIndex)
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series([np.dtype("float64"), np.dtype("float64")], index=["c31", "c32"]),
+    )
+
+    test_table.drop()
+    test_table2.drop()
+
+
 def test_date_range():
     with pytest.raises(TypeError):
         _ = date_range("2020-1-1", periods="2")
@@ -316,3 +384,18 @@ def test_date_range():
     assert dr.index_value.is_unique == expected.is_unique
     assert dr.index_value.is_monotonic_increasing == expected.is_monotonic_increasing
     assert dr.name == expected.name
+
+
+def test_resolve_task_sector():
+    input_path = os.path.join(os.path.dirname(__file__), "test-data", "task-input.txt")
+    with open(input_path, "r") as f:
+        sector = f.read()
+    actual_sector = _resolve_task_sector("job0", sector)
+
+    assert actual_sector.job_name == "job0"
+    assert actual_sector.task_name == "M1"
+    assert actual_sector.output_target == "Screen"
+    assert len(actual_sector.schema) == 78
+    assert actual_sector.schema[0] == ColumnSchema("unnamed: 0", "bigint", "")
+    assert actual_sector.schema[1] == ColumnSchema("id", "bigint", "id_alias")
+    assert actual_sector.schema[2] == ColumnSchema("listing_url", "string", "")

maxframe/dataframe/groupby/__init__.py

@@ -14,6 +14,7 @@
 
 # noinspection PyUnresolvedReferences
 from ..core import DataFrameGroupBy, GroupBy, SeriesGroupBy
+from .core import NamedAgg
 
 
 def _install():
@@ -25,6 +26,7 @@ def _install():
     from .fill import bfill, ffill, fillna
     from .getitem import df_groupby_getitem
     from .head import head
+    from .sample import groupby_sample
     from .transform import groupby_transform
 
     for cls in DATAFRAME_TYPE:
@@ -65,6 +67,8 @@ def _install():
 
         setattr(cls, "head", head)
 
+        setattr(cls, "sample", groupby_sample)
+
         setattr(cls, "ffill", ffill)
         setattr(cls, "bfill", bfill)
         setattr(cls, "backfill", bfill)

maxframe/dataframe/groupby/core.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from collections import namedtuple
+
 import pandas as pd
 
 from ... import opcodes
@@ -30,6 +32,9 @@ _GROUP_KEYS_NO_DEFAULT = pd_release_version >= (1, 5, 0)
 _default_group_keys = no_default if _GROUP_KEYS_NO_DEFAULT else True
 
 
+NamedAgg = namedtuple("NamedAgg", ["column", "aggfunc"])
+
+
 class DataFrameGroupByOperator(MapReduceOperator, DataFrameOperatorMixin):
     _op_type_ = opcodes.GROUPBY
 

maxframe/dataframe/misc/to_numeric.py

@@ -29,6 +29,10 @@ class DataFrameToNumeric(DataFrameOperator, DataFrameOperatorMixin):
     def __init__(self, errors="raise", downcast=None, **kw):
         super().__init__(errors=errors, downcast=downcast, **kw)
 
+    @property
+    def input(self):
+        return self.inputs[0]
+
     def __call__(self, arg):
         if isinstance(arg, pd.Series):
             arg = asseries(arg)

maxframe/dataframe/window/aggregation.py

@@ -18,14 +18,7 @@ from collections.abc import Iterable
 import numpy as np
 import pandas as pd
 
-from ...serialization.serializables import (
-    AnyField,
-    BoolField,
-    DictField,
-    Int32Field,
-    Int64Field,
-    StringField,
-)
+from ...serialization.serializables import AnyField, BoolField, Int32Field, Int64Field
 from ..core import DATAFRAME_TYPE
 from ..operators import DataFrameOperator, DataFrameOperatorMixin
 from ..utils import build_df, build_empty_series, parse_index
@@ -41,22 +34,6 @@ class BaseDataFrameExpandingAgg(DataFrameOperator, DataFrameOperatorMixin):
     # True if function name is treated as new index
     append_index = BoolField("append_index", default=None)
 
-    # chunk params
-    output_agg = BoolField("output_agg", default=None)
-
-    map_groups = DictField("map_groups", default=None)
-    map_sources = DictField("map_sources", default=None)
-    combine_sources = DictField("combine_sources", default=None)
-    combine_columns = DictField("combine_columns", default=None)
-    combine_funcs = DictField("combine_funcs", default=None)
-    key_to_funcs = DictField("keys_to_funcs", default=None)
-
-    min_periods_func_name = StringField("min_periods_func_name", default=None)
-
-    @property
-    def output_limit(self):
-        return 2 if self.output_agg else 1
-
     def __call__(self, expanding):
         inp = expanding.input
         raw_func = self.func

maxframe/dataframe/window/ewm.py

@@ -233,13 +233,6 @@ def ewm(
     if alpha <= 0 or alpha > 1:
         raise ValueError("alpha must satisfy: 0 < alpha <= 1")
 
-    if not adjust and not ignore_na:
-        raise NotImplementedError(
-            "adjust == False when ignore_na == False not implemented"
-        )
-    if axis == 1:
-        raise NotImplementedError("axis other than 0 is not supported")
-
     if alpha == 1:
         return obj.expanding(min_periods=min_periods, axis=axis)
 

maxframe/dataframe/window/tests/test_ewm.py

@@ -23,9 +23,6 @@ def test_ewm():
     df = pd.DataFrame(np.random.rand(4, 3), columns=list("abc"))
     df2 = md.DataFrame(df)
 
-    with pytest.raises(NotImplementedError):
-        _ = df2.ewm(2, adjust=False, ignore_na=False)
-
     with pytest.raises(ValueError):
         _ = df2.ewm()
 
@@ -59,9 +56,6 @@ def test_ewm_agg():
     df = pd.DataFrame(np.random.rand(4, 3), columns=list("abc"))
     df2 = md.DataFrame(df, chunk_size=3)
 
-    with pytest.raises(NotImplementedError):
-        _ = df2.ewm(span=3, axis=1).agg("mean")
-
     r = df2.ewm(span=3).agg("mean")
     expected = df.ewm(span=3).agg("mean")
 

maxframe/errors.py

@@ -0,0 +1,21 @@
+# 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.
+
+
+class MaxFrameError(Exception):
+    pass
+
+
+class MaxFrameUserError(MaxFrameError):
+    pass

maxframe/lib/aio/isolation.py

@@ -14,11 +14,14 @@
 
 import asyncio
 import atexit
+import itertools
 import threading
 from typing import Dict, Optional
 
 
 class Isolation:
+    _counter = itertools.count().__next__
+
     loop: asyncio.AbstractEventLoop
     _stopped: Optional[asyncio.Event]
     _thread: Optional[threading.Thread]
@@ -38,7 +41,9 @@ class Isolation:
 
     def start(self):
         if self._threaded:
-            self._thread = thread = threading.Thread(target=self._run)
+            self._thread = thread = threading.Thread(
+                name=f"IsolationThread-{self._counter()}", target=self._run
+            )
             thread.daemon = True
             thread.start()
             self._thread_ident = thread.ident

maxframe/opcodes.py

@@ -462,6 +462,7 @@ READ_ODPS_TABLE = 20111
 TO_ODPS_TABLE = 20112
 READ_ODPS_VOLUME = 20113
 TO_ODPS_VOLUME = 20114
+READ_ODPS_QUERY = 20115
 
 TO_CSV_STAT = 2102
 

maxframe/protocol.py

@@ -46,6 +46,8 @@ BodyType = TypeVar("BodyType", bound="Serializable")
 
 
 class JsonSerializable(Serializable):
+    _ignore_non_existing_keys = True
+
     @classmethod
     def from_json(cls, serialized: dict) -> "JsonSerializable":
         raise NotImplementedError
@@ -209,7 +211,10 @@ class ErrorInfo(JsonSerializable):
         kw["raw_error_source"] = ErrorSource(serialized["raw_error_source"])
         if kw.get("raw_error_data"):
             bufs = [base64.b64decode(s) for s in kw["raw_error_data"]]
-            kw["raw_error_data"] = pickle.loads(bufs[0], buffers=bufs[1:])
+            try:
+                kw["raw_error_data"] = pickle.loads(bufs[0], buffers=bufs[1:])
+            except:
+                kw["raw_error_data"] = None
         return cls(**kw)
 
     def to_json(self) -> dict:
@@ -242,6 +247,8 @@ class DagInfo(JsonSerializable):
         default_factory=dict,
     )
     error_info: Optional[ErrorInfo] = ReferenceField("error_info", default=None)
+    start_timestamp: Optional[float] = Float64Field("start_timestamp", default=None)
+    end_timestamp: Optional[float] = Float64Field("end_timestamp", default=None)
 
     @classmethod
     def from_json(cls, serialized: dict) -> "DagInfo":
@@ -262,7 +269,10 @@ class DagInfo(JsonSerializable):
             "dag_id": self.dag_id,
             "status": self.status.value,
             "progress": self.progress,
+            "start_timestamp": self.start_timestamp,
+            "end_timestamp": self.end_timestamp,
         }
+        ret = {k: v for k, v in ret.items() if v is not None}
         if self.tileable_to_result_infos:
             ret["tileable_to_result_infos"] = {
                 k: v.to_json() for k, v in self.tileable_to_result_infos.items()
@@ -278,12 +288,18 @@ class CreateSessionRequest(Serializable):
 
 class SessionInfo(JsonSerializable):
     session_id: str = StringField("session_id")
-    settings: Dict[str, Any] = DictField("settings", key_type=FieldTypes.string)
-    start_timestamp: float = Float64Field("start_timestamp")
-    idle_timestamp: float = Float64Field("idle_timestamp")
+    settings: Dict[str, Any] = DictField(
+        "settings", key_type=FieldTypes.string, default=None
+    )
+    start_timestamp: float = Float64Field("start_timestamp", default=None)
+    idle_timestamp: float = Float64Field("idle_timestamp", default=None)
     dag_infos: Dict[str, Optional[DagInfo]] = DictField(
-        "dag_infos", key_type=FieldTypes.string, value_type=FieldTypes.reference
+        "dag_infos",
+        key_type=FieldTypes.string,
+        value_type=FieldTypes.reference,
+        default=None,
     )
+    error_info: Optional[ErrorInfo] = ReferenceField("error_info", default=None)
 
     @classmethod
     def from_json(cls, serialized: dict) -> "SessionInfo":
@@ -292,6 +308,8 @@ class SessionInfo(JsonSerializable):
             kw["dag_infos"] = {
                 k: DagInfo.from_json(v) for k, v in kw["dag_infos"].items()
             }
+        if kw.get("error_info"):
+            kw["error_info"] = ErrorInfo.from_json(kw["error_info"])
         return SessionInfo(**kw)
 
     def to_json(self) -> dict:
@@ -303,6 +321,8 @@ class SessionInfo(JsonSerializable):
         }
         if self.dag_infos:
             ret["dag_infos"] = {k: v.to_json() for k, v in self.dag_infos.items()}
+        if self.error_info:
+            ret["error_info"] = self.error_info.to_json()
         return ret
 
 

maxframe/serialization/exception.py

@@ -16,13 +16,14 @@ import logging
 import traceback
 from typing import Dict, List
 
+from ..errors import MaxFrameError
 from ..lib import wrapped_pickle as pickle
 from .core import Serializer, buffered, pickle_buffers, unpickle_buffers
 
 logger = logging.getLogger(__name__)
 
 
-class RemoteException(Exception):
+class RemoteException(MaxFrameError):
     def __init__(
         self, messages: List[str], tracebacks: List[List[str]], buffers: List[bytes]
     ):

maxframe/serialization/serializables/core.py

@@ -112,6 +112,7 @@ class Serializable(metaclass=SerializableMeta):
     __slots__ = ("__weakref__",)
 
     _cache_primitive_serial = False
+    _ignore_non_existing_keys = False
 
     _FIELDS: Dict[str, Field]
     _FIELD_ORDER: List[str]
@@ -128,7 +129,11 @@ class Serializable(metaclass=SerializableMeta):
         else:
             values = kwargs
         for k, v in values.items():
-            fields[k].set(self, v)
+            try:
+                fields[k].set(self, v)
+            except KeyError:
+                if not self._ignore_non_existing_keys:
+                    raise
 
     def __on_deserialize__(self):
         pass

maxframe/serialization/serializables/field.py

@@ -507,12 +507,14 @@ class ReferenceField(Field):
         tag: str,
         reference_type: Union[str, Type] = None,
         default: Any = no_default,
+        default_factory: Optional[Callable] = None,
         on_serialize: Callable[[Any], Any] = None,
         on_deserialize: Callable[[Any], Any] = None,
     ):
         super().__init__(
             tag,
             default=default,
+            default_factory=default_factory,
             on_serialize=on_serialize,
             on_deserialize=on_deserialize,
         )

maxframe/tensor/core.py

@@ -43,7 +43,7 @@ from ..serialization.serializables import (
     StringField,
     TupleField,
 )
-from ..utils import on_deserialize_shape, on_serialize_shape
+from ..utils import on_deserialize_shape, on_serialize_shape, skip_na_call
 from .utils import fetch_corner_data, get_chunk_slices
 
 logger = logging.getLogger(__name__)
@@ -181,8 +181,8 @@ class TensorData(HasShapeTileableData, _ExecuteAndFetchMixin):
     _chunks = ListField(
         "chunks",
         FieldTypes.reference(TensorChunkData),
-        on_serialize=lambda x: [it.data for it in x] if x is not None else x,
-        on_deserialize=lambda x: [TensorChunk(it) for it in x] if x is not None else x,
+        on_serialize=skip_na_call(lambda x: [it.data for it in x]),
+        on_deserialize=skip_na_call(lambda x: [TensorChunk(it) for it in x]),
     )
 
     def __init__(

maxframe/tests/test_codegen.py

@@ -0,0 +1,69 @@
+# -*- coding: utf-8 -*-
+# 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.
+
+import base64
+from typing import List, Tuple
+
+# 使用pytest生成单元测试
+import pytest
+
+from maxframe.codegen import UserCodeMixin
+from maxframe.lib import wrapped_pickle
+from maxframe.serialization.core import PickleContainer
+
+
+@pytest.mark.parametrize(
+    "input_obj, expected_output",
+    [
+        (None, "None"),
+        (10, "10"),
+        (3.14, "3.14"),
+        (True, "True"),
+        (False, "False"),
+        (b"hello", "base64.b64decode(b'aGVsbG8=')"),
+        ("hello", "'hello'"),
+        ([1, 2, 3], "[1, 2, 3]"),
+        ({"a": 1, "b": 2}, "{'a': 1, 'b': 2}"),
+        ((1, 2, 3), "(1, 2, 3)"),
+        ((1,), "(1,)"),
+        ((), "()"),
+        ({1, 2, 3}, "{1, 2, 3}"),
+        (set(), "set()"),
+    ],
+)
+def test_obj_to_python_expr(input_obj, expected_output):
+    assert UserCodeMixin.obj_to_python_expr(input_obj) == expected_output
+
+
+def test_obj_to_python_expr_custom_object():
+    class CustomClass:
+        def __init__(self, a: int, b: List[int], c: Tuple[int, int]):
+            self.a = a
+            self.b = b
+            self.c = c
+
+    custom_obj = CustomClass(1, [2, 3], (4, 5))
+    pickle_data = wrapped_pickle.dumps(custom_obj)
+    pickle_str = base64.b64encode(pickle_data)
+    custom_obj_pickle_container = PickleContainer([pickle_data])
+
+    # with class obj will not support currently
+    with pytest.raises(ValueError):
+        UserCodeMixin.obj_to_python_expr(custom_obj)
+
+    assert (
+        UserCodeMixin.obj_to_python_expr(custom_obj_pickle_container)
+        == f"cloudpickle.loads(base64.b64decode({pickle_str}), buffers=[])"
+    )

maxframe/tests/test_protocol.py

@@ -11,6 +11,8 @@
 # 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.
+
+import json
 import time
 
 import pytest
@@ -29,28 +31,32 @@ from ..serialization import RemoteException
 from ..utils import deserialize_serializable, serialize_serializable
 
 
+def _json_round_trip(json_data: dict) -> dict:
+    return json.loads(json.dumps(json_data))
+
+
 def test_result_info_json_serialize():
-    ri = ResultInfo.from_json(ResultInfo().to_json())
+    ri = ResultInfo.from_json(_json_round_trip(ResultInfo().to_json()))
     assert type(ri) is ResultInfo
 
     ri = ODPSTableResultInfo(
         full_table_name="table_name", partition_specs=["pt=partition"]
     )
-    deserial_ri = ResultInfo.from_json(ri.to_json())
+    deserial_ri = ResultInfo.from_json(_json_round_trip(ri.to_json()))
     assert type(ri) is ODPSTableResultInfo
     assert ri.result_type == deserial_ri.result_type
     assert ri.full_table_name == deserial_ri.full_table_name
     assert ri.partition_specs == deserial_ri.partition_specs
 
     ri = ODPSTableResultInfo(full_table_name="table_name")
-    deserial_ri = ResultInfo.from_json(ri.to_json())
+    deserial_ri = ResultInfo.from_json(_json_round_trip(ri.to_json()))
     assert type(ri) is ODPSTableResultInfo
     assert ri.result_type == deserial_ri.result_type
     assert ri.full_table_name == deserial_ri.full_table_name
     assert ri.partition_specs == deserial_ri.partition_specs
 
     ri = ODPSVolumeResultInfo(volume_name="vol_name", volume_path="vol_path")
-    deserial_ri = ResultInfo.from_json(ri.to_json())
+    deserial_ri = ResultInfo.from_json(_json_round_trip(ri.to_json()))
     assert type(ri) is ODPSVolumeResultInfo
     assert ri.result_type == deserial_ri.result_type
     assert ri.volume_name == deserial_ri.volume_name
@@ -63,7 +69,7 @@ def test_error_info_json_serialize():
     except ValueError as ex:
         err_info = ErrorInfo.from_exception(ex)
 
-    deserial_err_info = ErrorInfo.from_json(err_info.to_json())
+    deserial_err_info = ErrorInfo.from_json(_json_round_trip(err_info.to_json()))
     assert deserial_err_info.error_messages == err_info.error_messages
     assert isinstance(deserial_err_info.raw_error_data, ValueError)
 
@@ -73,7 +79,7 @@ def test_error_info_json_serialize():
     with pytest.raises(RemoteException):
         mf_err_info.reraise()
 
-    deserial_err_info = ErrorInfo.from_json(mf_err_info.to_json())
+    deserial_err_info = ErrorInfo.from_json(_json_round_trip(mf_err_info.to_json()))
     assert isinstance(deserial_err_info.raw_error_data, ValueError)
     with pytest.raises(ValueError):
         deserial_err_info.reraise()
@@ -94,7 +100,9 @@ def test_dag_info_json_serialize():
         },
         error_info=err_info,
     )
-    deserial_info = DagInfo.from_json(info.to_json())
+    json_info = info.to_json()
+    json_info["non_existing_field"] = "non_existing"
+    deserial_info = DagInfo.from_json(_json_round_trip(json_info))
     assert deserial_info.session_id == info.session_id
     assert deserial_info.dag_id == info.dag_id
     assert deserial_info.status == info.status
@@ -121,7 +129,7 @@ def test_session_info_json_serialize():
         idle_timestamp=None,
         dag_infos={"test_dag_id": dag_info},
     )
-    deserial_info = SessionInfo.from_json(info.to_json())
+    deserial_info = SessionInfo.from_json(_json_round_trip(info.to_json()))
     assert deserial_info.session_id == info.session_id
     assert deserial_info.settings == info.settings
     assert deserial_info.start_timestamp == info.start_timestamp

maxframe/tests/utils.py

@@ -104,6 +104,7 @@ def run_app_in_thread(app_func):
         q = queue.Queue()
         exit_event = asyncio.Event(loop=app_loop)
         app_thread = Thread(
+            name="TestAppThread",
             target=app_thread_func,
             args=(app_loop, q, exit_event, args, kwargs),
         )