fugue 0.8.7.dev5__py3-none-any.whl → 0.8.7.dev6__py3-none-any.whl

fugue/api.py

@@ -6,6 +6,7 @@ from .dataframe.api import (
     as_array_iterable,
     as_arrow,
     as_dict_iterable,
+    as_dicts,
     as_fugue_df,
     as_pandas,
     drop_columns,

fugue/dataframe/api.py

@@ -116,15 +116,32 @@ def as_array_iterable(
     return as_fugue_df(df).as_array_iterable(columns=columns, type_safe=type_safe)
 
 
+@fugue_plugin
+def as_dicts(
+    df: AnyDataFrame, columns: Optional[List[str]] = None
+) -> List[Dict[str, Any]]:
+    """Convert any dataframe to a list of python dicts
+
+    :param df: the object that can be recognized as a dataframe by Fugue
+    :param columns: columns to extract, defaults to None
+    :return: a list of python dicts
+
+    .. note::
+
+        The default implementation enforces ``type_safe`` True
+    """
+    return as_fugue_df(df).as_dicts(columns=columns)
+
+
 @fugue_plugin
 def as_dict_iterable(
     df: AnyDataFrame, columns: Optional[List[str]] = None
 ) -> Iterable[Dict[str, Any]]:
-    """Convert any dataframe to iterable of native python dicts
+    """Convert any dataframe to iterable of python dicts
 
     :param df: the object that can be recognized as a dataframe by Fugue
     :param columns: columns to extract, defaults to None
-    :return: iterable of native python dicts
+    :return: iterable of python dicts
 
     .. note::
 

fugue/dataframe/arrow_dataframe.py

@@ -21,6 +21,10 @@ from fugue.exceptions import FugueDataFrameOperationError
 
 from .api import (
     alter_columns,
+    as_array,
+    as_array_iterable,
+    as_dict_iterable,
+    as_dicts,
     as_pandas,
     drop_columns,
     get_column_names,
@@ -30,6 +34,12 @@ from .api import (
     select_columns,
 )
 from .dataframe import DataFrame, LocalBoundedDataFrame, _input_schema
+from .utils import (
+    pa_table_as_array,
+    pa_table_as_array_iterable,
+    pa_table_as_dict_iterable,
+    pa_table_as_dicts,
+)
 
 
 class ArrowDataFrame(LocalBoundedDataFrame):
@@ -174,21 +184,20 @@ class ArrowDataFrame(LocalBoundedDataFrame):
     def as_array(
         self, columns: Optional[List[str]] = None, type_safe: bool = False
     ) -> List[Any]:
-        return list(self.as_array_iterable(columns, type_safe=type_safe))
+        return pa_table_as_array(self.native, columns=columns)
+
+    def as_dicts(self, columns: Optional[List[str]] = None) -> List[Dict[str, Any]]:
+        return pa_table_as_dicts(self.native, columns=columns)
 
     def as_array_iterable(
         self, columns: Optional[List[str]] = None, type_safe: bool = False
     ) -> Iterable[Any]:
-        if self.empty:
-            return
-        if columns is not None:
-            for x in self[columns].as_array_iterable(type_safe=type_safe):
-                yield x
-        else:
-            d = self.native.to_pydict()
-            cols = [d[n] for n in self.columns]
-            for arr in zip(*cols):
-                yield list(arr)
+        yield from pa_table_as_array_iterable(self.native, columns=columns)
+
+    def as_dict_iterable(
+        self, columns: Optional[List[str]] = None
+    ) -> Iterable[Dict[str, Any]]:
+        yield from pa_table_as_dict_iterable(self.native, columns=columns)
 
 
 @as_local.candidate(lambda df: isinstance(df, pa.Table))
@@ -212,6 +221,34 @@ def _pa_table_as_pandas(df: pa.Table) -> pd.DataFrame:
     )
 
 
+@as_array.candidate(lambda df, *args, **kwargs: isinstance(df, pa.Table))
+def _pa_table_as_array(
+    df: pa.Table, columns: Optional[List[str]] = None, type_safe: bool = False
+) -> List[Any]:
+    return pa_table_as_array(df, columns=columns)
+
+
+@as_array_iterable.candidate(lambda df, *args, **kwargs: isinstance(df, pa.Table))
+def _pa_table_as_array_iterable(
+    df: pa.Table, columns: Optional[List[str]] = None, type_safe: bool = False
+) -> Iterable[Any]:
+    yield from pa_table_as_array_iterable(df, columns=columns)
+
+
+@as_dicts.candidate(lambda df, *args, **kwargs: isinstance(df, pa.Table))
+def _pa_table_as_dicts(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> List[Dict[str, Any]]:
+    return pa_table_as_dicts(df, columns=columns)
+
+
+@as_dict_iterable.candidate(lambda df, *args, **kwargs: isinstance(df, pa.Table))
+def _pa_table_as_dict_iterable(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> Iterable[Dict[str, Any]]:
+    yield from pa_table_as_dict_iterable(df, columns=columns)
+
+
 @alter_columns.candidate(lambda df, *args, **kwargs: isinstance(df, pa.Table))
 def _pa_table_alter_columns(
     df: pa.Table, columns: Any, as_fugue: bool = False

fugue/dataframe/dataframe.py

@@ -237,13 +237,31 @@ class DataFrame(Dataset):
         """
         raise NotImplementedError
 
+    def as_dicts(self, columns: Optional[List[str]] = None) -> List[Dict[str, Any]]:
+        """Convert to a list of python dicts
+
+        :param columns: columns to extract, defaults to None
+        :return: a list of python dicts
+
+        .. note::
+
+            The default implementation enforces ``type_safe`` True
+        """
+        if columns is None:
+            columns = self.columns
+        idx = range(len(columns))
+        return [
+            {columns[i]: x[i] for i in idx}
+            for x in self.as_array(columns, type_safe=True)
+        ]
+
     def as_dict_iterable(
         self, columns: Optional[List[str]] = None
     ) -> Iterable[Dict[str, Any]]:
-        """Convert to iterable of native python dicts
+        """Convert to iterable of python dicts
 
         :param columns: columns to extract, defaults to None
-        :return: iterable of native python dicts
+        :return: iterable of python dicts
 
         .. note::
 

fugue/dataframe/function_wrapper.py

@@ -269,7 +269,7 @@ class _EmptyAwareIterableListParam(_LocalNoSchemaDataFrameParam):
 class _ListDictParam(_LocalNoSchemaDataFrameParam):
     @no_type_check
     def to_input_data(self, df: DataFrame, ctx: Any) -> List[Dict[str, Any]]:
-        return list(df.as_local().as_dict_iterable())
+        return df.as_local().as_dicts()
 
     @no_type_check
     def to_output_df(

fugue/dataframe/iterable_dataframe.py

@@ -105,6 +105,9 @@ class IterableDataFrame(LocalUnboundedDataFrame):
     ) -> List[Any]:
         return list(self.as_array_iterable(columns, type_safe=type_safe))
 
+    def as_dicts(self, columns: Optional[List[str]] = None) -> List[Dict[str, Any]]:
+        return list(self.as_dict_iterable(columns))
+
     def as_array_iterable(
         self, columns: Optional[List[str]] = None, type_safe: bool = False
     ) -> Iterable[Any]:

fugue/dataframe/pandas_dataframe.py

@@ -1,8 +1,11 @@
 from typing import Any, Dict, Iterable, List, Optional, Tuple
 
 import pandas as pd
+import pyarrow as pa
+from triad import assert_or_throw
 from triad.collections.schema import Schema
 from triad.utils.pandas_like import PD_UTILS
+from triad.utils.pyarrow import pa_batch_to_dicts
 
 from fugue.dataset.api import (
     as_fugue_dataset,
@@ -17,6 +20,10 @@ from fugue.dataset.api import (
 from fugue.exceptions import FugueDataFrameOperationError
 
 from .api import (
+    as_array,
+    as_array_iterable,
+    as_dict_iterable,
+    as_dicts,
     drop_columns,
     get_column_names,
     get_schema,
@@ -134,6 +141,9 @@ class PandasDataFrame(LocalBoundedDataFrame):
             return self
         return PandasDataFrame(self.native, new_schema)
 
+    def as_arrow(self, type_safe: bool = False) -> pa.Table:
+        return PD_UTILS.as_arrow(self.native, schema=self.schema.pa_schema)
+
     def as_array(
         self, columns: Optional[List[str]] = None, type_safe: bool = False
     ) -> List[Any]:
@@ -150,6 +160,18 @@ class PandasDataFrame(LocalBoundedDataFrame):
         ):
             yield row
 
+    def as_dicts(self, columns: Optional[List[str]] = None) -> List[Dict[str, Any]]:
+        res: List[Dict[str, Any]] = []
+        for block in _to_dicts(self.native, columns, self.schema):
+            res += block
+        return res
+
+    def as_dict_iterable(
+        self, columns: Optional[List[str]] = None
+    ) -> Iterable[Dict[str, Any]]:
+        for block in _to_dicts(self.native, columns, self.schema):
+            yield from block
+
     def head(
         self, n: int, columns: Optional[List[str]] = None
     ) -> LocalBoundedDataFrame:
@@ -272,6 +294,43 @@ def _pd_head(
     return _adjust_df(df.head(n), as_fugue=as_fugue)
 
 
+@as_array.candidate(lambda df, *args, **kwargs: isinstance(df, pd.DataFrame))
+def _pd_as_array(
+    df: pd.DataFrame, columns: Optional[List[str]] = None, type_safe: bool = False
+) -> List[Any]:
+    return list(_pd_as_array_iterable(df, columns, type_safe=type_safe))
+
+
+@as_array_iterable.candidate(lambda df, *args, **kwargs: isinstance(df, pd.DataFrame))
+def _pd_as_array_iterable(
+    df: pd.DataFrame, columns: Optional[List[str]] = None, type_safe: bool = False
+) -> Iterable[Any]:
+    for row in PD_UTILS.as_array_iterable(
+        df,
+        columns=columns,
+        type_safe=type_safe,
+    ):
+        yield row
+
+
+@as_dicts.candidate(lambda df, *args, **kwargs: isinstance(df, pd.DataFrame))
+def _pd_as_dicts(
+    df: pd.DataFrame, columns: Optional[List[str]] = None
+) -> List[Dict[str, Any]]:
+    res: List[Dict[str, Any]] = []
+    for block in _to_dicts(df, columns):
+        res += block
+    return res
+
+
+@as_dict_iterable.candidate(lambda df, *args, **kwargs: isinstance(df, pd.DataFrame))
+def _pd_as_dict_iterable(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> Iterable[Dict[str, Any]]:
+    for block in _to_dicts(df, columns):
+        yield from block
+
+
 def _adjust_df(res: pd.DataFrame, as_fugue: bool):
     return res if not as_fugue else PandasDataFrame(res)
 
@@ -280,3 +339,17 @@ def _assert_no_missing(df: pd.DataFrame, columns: Iterable[Any]) -> None:
     missing = [x for x in columns if x not in df.columns]
     if len(missing) > 0:
         raise FugueDataFrameOperationError("found nonexistent columns: {missing}")
+
+
+def _to_dicts(
+    df: pd.DataFrame,
+    columns: Optional[List[str]] = None,
+    schema: Optional[Schema] = None,
+) -> Iterable[List[Dict[str, Any]]]:
+    cols = list(df.columns) if columns is None else columns
+    assert_or_throw(len(cols) > 0, ValueError("columns cannot be empty"))
+    pa_schema = schema.extract(cols).pa_schema if schema is not None else None
+    adf = PD_UTILS.as_arrow(df[cols], schema=pa_schema)
+    for batch in adf.to_batches():
+        if batch.num_rows > 0:
+            yield pa_batch_to_dicts(batch)

fugue/dataframe/utils.py

@@ -1,15 +1,16 @@
 import os
 import pickle
-from typing import Any, Iterable, Optional, Tuple
+from typing import Any, Iterable, Optional, Tuple, List, Dict
 
 import pandas as pd
 import pyarrow as pa
 from fs import open_fs
-from triad import FileSystem, Schema
+from triad import FileSystem, Schema, assert_or_throw
 from triad.collections.schema import SchemaError
 from triad.exceptions import InvalidOperationError
 from triad.utils.assertion import assert_arg_not_none
 from triad.utils.assertion import assert_or_throw as aot
+from triad.utils.pyarrow import pa_batch_to_dicts
 
 from .api import as_fugue_df, get_column_names, normalize_column_names, rename
 from .dataframe import DataFrame, LocalBoundedDataFrame
@@ -250,3 +251,68 @@ def get_join_schemas(
     else:
         aot(len(on) > 0, SchemaError("join on columns must be specified"))
     return cm, (df1.schema.union(schema2))
+
+
+def pa_table_as_array_iterable(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> Iterable[List[List[Any]]]:
+    """Convert a pyarrow table to an iterable of list
+
+    :param df: pyarrow table
+    :param columns: if not None, only these columns will be returned, defaults to None
+    :return: an iterable of list
+    """
+    assert_or_throw(columns is None or len(columns) > 0, ValueError("empty columns"))
+    _df = df if columns is None or len(columns) == 0 else df.select(columns)
+    for batch in _df.to_batches():
+        for x in zip(*batch.to_pydict().values()):
+            yield list(x)
+
+
+def pa_table_as_array(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> List[List[List[Any]]]:
+    """Convert a pyarrow table to a list of list
+
+    :param df: pyarrow table
+    :param columns: if not None, only these columns will be returned, defaults to None
+    :return: a list of list
+    """
+    return list(pa_table_as_array_iterable(df, columns=columns))
+
+
+def pa_table_as_dict_iterable(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> Iterable[Dict[str, Any]]:
+    """Convert a pyarrow table to an iterable of dict
+
+    :param df: pyarrow table
+    :param columns: if not None, only these columns will be returned, defaults to None
+    :return: an iterable of dict
+    """
+    for ck in _pa_table_as_dicts_chunks(df, columns=columns):
+        yield from ck
+
+
+def pa_table_as_dicts(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> List[Dict[str, Any]]:
+    """Convert a pyarrow table to a list of dict
+
+    :param df: pyarrow table
+    :param columns: if not None, only these columns will be returned, defaults to None
+    :return: a list of dict
+    """
+    res: List[Dict[str, Any]] = []
+    for ck in _pa_table_as_dicts_chunks(df, columns=columns):
+        res += ck
+    return res
+
+
+def _pa_table_as_dicts_chunks(
+    df: pa.Table, columns: Optional[List[str]] = None
+) -> Iterable[List[Dict[str, Any]]]:
+    assert_or_throw(columns is None or len(columns) > 0, ValueError("empty columns"))
+    _df = df if columns is None or len(columns) == 0 else df.select(columns)
+    for batch in _df.to_batches():
+        yield pa_batch_to_dicts(batch)

fugue/execution/execution_engine.py

@@ -1323,7 +1323,7 @@ class _Comap:
         self._on_init(partition_no, empty_dfs)
 
     def run(self, cursor: PartitionCursor, df: LocalDataFrame) -> LocalDataFrame:
-        data = list(df.as_dict_iterable())
+        data = df.as_dicts()
         if self.how == "inner":
             if len(data) < self.dfs_count:
                 return ArrayDataFrame([], self.output_schema)

fugue/plugins.py

@@ -7,6 +7,7 @@ from fugue.dataframe import (
     as_array_iterable,
     as_arrow,
     as_dict_iterable,
+    as_dicts,
     as_pandas,
     drop_columns,
     fugue_annotated_param,

{fugue-0.8.7.dev5.dist-info → fugue-0.8.7.dev6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fugue
-Version: 0.8.7.dev5
+Version: 0.8.7.dev6
 Summary: An abstraction layer for distributed computation
 Home-page: http://github.com/fugue-project/fugue
 Author: The Fugue Development Team
@@ -19,7 +19,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3 :: Only
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
-Requires-Dist: triad ==0.9.2.dev3
+Requires-Dist: triad ==0.9.2.dev5
 Requires-Dist: adagio >=0.2.4
 Requires-Dist: qpd >=0.4.4
 Requires-Dist: fugue-sql-antlr >=0.1.6
@@ -32,7 +32,7 @@ Requires-Dist: fugue-sql-antlr[cpp] >=0.1.6 ; extra == 'all'
 Requires-Dist: pyspark >=3.1.1 ; extra == 'all'
 Requires-Dist: dask[dataframe,distributed] >=2023.5.0 ; extra == 'all'
 Requires-Dist: dask-sql ; extra == 'all'
-Requires-Dist: ray[data] >=2.1.0 ; extra == 'all'
+Requires-Dist: ray[data] >=2.4.0 ; extra == 'all'
 Requires-Dist: notebook ; extra == 'all'
 Requires-Dist: jupyterlab ; extra == 'all'
 Requires-Dist: ipython >=7.10.0 ; extra == 'all'
@@ -59,7 +59,7 @@ Requires-Dist: ipython >=7.10.0 ; extra == 'notebook'
 Provides-Extra: polars
 Requires-Dist: polars ; extra == 'polars'
 Provides-Extra: ray
-Requires-Dist: ray[data] >=2.1.0 ; extra == 'ray'
+Requires-Dist: ray[data] >=2.4.0 ; extra == 'ray'
 Requires-Dist: duckdb >=0.5.0 ; extra == 'ray'
 Requires-Dist: pyarrow >=6.0.1 ; extra == 'ray'
 Provides-Extra: spark

{fugue-0.8.7.dev5.dist-info → fugue-0.8.7.dev6.dist-info}/RECORD

@@ -1,9 +1,9 @@
 fugue/__init__.py,sha256=xT5zuNZfRkjbA8a-uTT5oLK6hLGuezGZLWYBl6eS5J4,2749
-fugue/api.py,sha256=6_d3vYwJGAX7tW7NMhHB_NAX4aPsfzK2L06Zr2V78Ks,1240
+fugue/api.py,sha256=dLUrigFhDMB5x7cvlWSK8EyaY2o0AmhgPr0VRtfzSz0,1254
 fugue/constants.py,sha256=crd0VqX8WtBcjSUNwZDi2LDIEkhUMWOlSn73H8JI9ds,3385
 fugue/dev.py,sha256=GQCkezBBl4V0lVDWhGtUQKqomiCxgR9dMhfqj9C8cS8,1369
 fugue/exceptions.py,sha256=ylP8gkZL8ao_ZLinNYKv16FPyO_n7c29dN-4QChUxi0,1544
-fugue/plugins.py,sha256=SJ-jqs04StHIHJ65lgdGP0IDopVIGBDpmzHHllNK8p0,998
+fugue/plugins.py,sha256=kao-H5z-cRbujBKW1QC9IHUOBKxXMhpVQ6saIE7cXm8,1012
 fugue/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 fugue/registry.py,sha256=SNULGv08f37fRO-cIxFDmnVcod7ref2fNLSK6G7nVnI,868
 fugue/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -25,22 +25,22 @@ fugue/column/expressions.py,sha256=fdGX9oPCqJBuROFZqrOYVcwkjghdXT9ngaSTG5tW_i8,2
 fugue/column/functions.py,sha256=ygLyn2gp5lTdGbYqJXeGeMmRNhbm4-vfJvAY_Zt0pb0,9774
 fugue/column/sql.py,sha256=s_qTtHgnvRFqjhCWr7s595PTrHM-Pr9zHUQfU5xcTVA,17391
 fugue/dataframe/__init__.py,sha256=zm7TbsaJLIvfm7zymWm2LGcuJd3nxfGsFnQiyrSnenM,678
-fugue/dataframe/api.py,sha256=KEbZbXCnaUgfwGF7iZODZsCtJTL4-reQS4qVbigFrps,10554
+fugue/dataframe/api.py,sha256=aWBvMaiSUxOvdQMfe79zHShWuPfLcgiWggC9HvVxvSE,11017
 fugue/dataframe/array_dataframe.py,sha256=4scWnmQ6sjy1A6o7IYdRc0VVutBEfcJrA1f9wkph4Kg,4440
-fugue/dataframe/arrow_dataframe.py,sha256=mJzrYBGs9mEMsHgxmnhDdiLUiOkcOs3YBAzHs75KFsI,10202
-fugue/dataframe/dataframe.py,sha256=a7jhYUaovN7w8vcJ-OU2AMfkfqxpvFF06cYWFqIJWqM,16418
+fugue/dataframe/arrow_dataframe.py,sha256=r5zcZBX_N6XO5dmixBkTCPgLcMmgDF022piZvrwRp_c,11485
+fugue/dataframe/dataframe.py,sha256=xmyG85i14A6LDRkNmPt29oYq7PJsq668s1QvFHK8PV4,16964
 fugue/dataframe/dataframe_iterable_dataframe.py,sha256=lx71KfaI4lsVKI-79buc-idaeT20JEMBOq21SQcAiY8,7259
 fugue/dataframe/dataframes.py,sha256=tBSpHsENgbcdOJ0Jgst6PTKbjG7_uoFJch96oTlaQIs,4160
-fugue/dataframe/function_wrapper.py,sha256=r6H1SQWaag2eSbJ50327t_bt7MZunbOMOl9OcOcQW2E,14827
-fugue/dataframe/iterable_dataframe.py,sha256=9g2BAF9A6QPbo63Si-trFq_9nPVqAD9vSePRCV71AfY,4620
-fugue/dataframe/pandas_dataframe.py,sha256=JNkr24h5gir1Msttx3lNfzFjwMqjHbjDswNynpCiizo,9158
-fugue/dataframe/utils.py,sha256=Oid7L9-V-NjKnwnkN8Jg85E2OPMWkjkjNI5OoeKTnbs,9132
+fugue/dataframe/function_wrapper.py,sha256=V1eQMOn27UroEYT7_YiwoEF0RjZYIM0zkD3vfaMAQFs,14813
+fugue/dataframe/iterable_dataframe.py,sha256=TcOoNKa4jNbHbvAZ0XAhtMmGcioygIHPxI9budDtenQ,4758
+fugue/dataframe/pandas_dataframe.py,sha256=0L0wYCGhD2BpQbruoT07Ox9iQM5YLHLNrcgzudc-yKs,11633
+fugue/dataframe/utils.py,sha256=VS1qLCr-9NEcEjaK-219rADJadDf6EfzYZCGRUpn1fY,11405
 fugue/dataset/__init__.py,sha256=5f2CAJ4xst6Z2o9Q2e2twfDOGUw8ZJoE2ild4JEU2pg,112
 fugue/dataset/api.py,sha256=DacI4L2w5NJ-eZ6nFxNMqmReEnb0WUXswbjVp7BeErk,2794
 fugue/dataset/dataset.py,sha256=jWXZqy3msMPFFkhas2PYJEX55ZAI3gk3Txq5f4-Qya4,4759
 fugue/execution/__init__.py,sha256=iZGxAznZz9piM3k4gp0tln97MDIBxdliLyNbD-0Zc48,427
 fugue/execution/api.py,sha256=KsFOLGdWQMlXmlQ5JRgRsbUeB64qzTVHxSEaunjiojo,39818
-fugue/execution/execution_engine.py,sha256=n-mw0k0QtK8FQgP4w4_NrWJbg0XvrR4sFn70tSaOi0I,47735
+fugue/execution/execution_engine.py,sha256=G_SsTmcuDcy6_azi_88lGzsOodiizu0JdWxebxgbqRg,47721
 fugue/execution/factory.py,sha256=5ICzfNh2QqqABuVyYLijY5-7LZgfRqczlaZN32p78bE,21003
 fugue/execution/native_execution_engine.py,sha256=Mm9BVC3dEMS3IWRZe4YvGKp6_mmW7dLmoLMK5HgAPcs,14408
 fugue/extensions/__init__.py,sha256=y-uLKd6mZ8sZ_8-OdW6ELoBO_9IfC0gDmEbE_rMCvOA,599
@@ -87,8 +87,8 @@ fugue_contrib/viz/_ext.py,sha256=Lu_DlS5DcmrFz27fHcKTCkhKyknVWcfS5kzZVVuO9xM,134
 fugue_dask/__init__.py,sha256=2CcJ0AsN-k_f7dZ-yAyYpaICfUMPfH3l0FvUJSBzTr0,161
 fugue_dask/_constants.py,sha256=35UmTVITk21GhRyRlbJOwPPdQsytM_p_2NytOXEay18,510
 fugue_dask/_io.py,sha256=9G516yM6zQvSC5_JA6qHb3LwBDmhWcxK5sjFHrQ81zo,6012
-fugue_dask/_utils.py,sha256=uFoJAL95rmnBgieU2hPyqxFZGvR6ZJgPRMq5TAJqIBI,8520
-fugue_dask/dataframe.py,sha256=TdKjxhoQpsU5CvBTgO2c5Zo_4LfyelR0IK8bPgjAxcg,10218
+fugue_dask/_utils.py,sha256=n70N3wPPMz13Jh0GWJM3Je-TCYpU36yGP_YCwIHqUrc,8908
+fugue_dask/dataframe.py,sha256=MuG9TqCND7qI66lPvxzuomfE7yA4sW7DjrvbyvE6XEU,13471
 fugue_dask/execution_engine.py,sha256=XJp6wrdkaNh5pOpwt-Hjoa2sxgCOgusFRWrcqoCcaNM,21153
 fugue_dask/ibis_engine.py,sha256=kQdaG_KlZZ2AjtYETNCdTJOgtwI_eH0aGzLaAiIBbRI,2120
 fugue_dask/registry.py,sha256=7UTg_eie7zKlHYKMCyOo0TNn5y2TiIjE8kiS2PruHFc,2200
@@ -96,14 +96,14 @@ fugue_duckdb/__init__.py,sha256=nSNv-fxBAKD6W23EbMeV4dVRIaSTqr9DzQUWuVOES8s,379
 fugue_duckdb/_io.py,sha256=Sq228unVnroYTq4GX-Wnv22SLHC9Ji-aWgiqrfdu81w,8880
 fugue_duckdb/_utils.py,sha256=ElKbHUyn5fWSPGXsK57iqMzcqKtCf0c8pBVBYGe5Ql4,5020
 fugue_duckdb/dask.py,sha256=agoLzeB7Swxj2kVWfmXFbWD1NS2lbbTlnrjSkR8kKWY,5014
-fugue_duckdb/dataframe.py,sha256=vNZF2BC1sJpW3P5TVFTpU6C1Ddam81jPC_4i8kBuEpo,6512
+fugue_duckdb/dataframe.py,sha256=LRfTv7Y46wMM_IDYSP1R-5OXuHuBg8GHjPGFFt8u7l0,8444
 fugue_duckdb/execution_engine.py,sha256=fkkQb4Eh0m7SwKrTplVk2oQalLkNoj3CW0R12g01ofk,20536
 fugue_duckdb/ibis_engine.py,sha256=MrypeABozqwetKOpqtrmWvCJX2QPfBXhbSEhvK9vqmI,1990
 fugue_duckdb/registry.py,sha256=Dj0Tng1cXVT6Q7t-KxOky2k1dD9xSBjYGQmI26UgZPo,3095
 fugue_ibis/__init__.py,sha256=PcUt66KlLyGGicad7asq5j2U567_fhR0HzvWQBhV1VM,362
 fugue_ibis/_compat.py,sha256=zKdTaTfuC02eUIzZPkcd7oObnVBi_X5mQjQf7SDme3Y,246
 fugue_ibis/_utils.py,sha256=BUL5swA5FE4eQu0t5Z17hZVu9a2MFfxlFH6Ymy9xifg,6607
-fugue_ibis/dataframe.py,sha256=olGfVYY9n5wwPOZojS30Fs3XEOMlenCzX8fuR2WPaq4,7295
+fugue_ibis/dataframe.py,sha256=0Fb1vJjwEeffgoUCDfDGIMuSFaPgUJqcB-JqJOAALfs,7789
 fugue_ibis/execution_engine.py,sha256=p5zy0IBXiJgLi67RBHCRcHgZsaJMANdNSpUxz0k_6C0,18453
 fugue_ibis/extensions.py,sha256=H8l-SPfoqLuUoILtOuL2nccOpoL83zHeSoIhoqjtWQM,6905
 fugue_ibis/execution/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -117,11 +117,11 @@ fugue_notebook/nbextension/description.yaml,sha256=CsXgx9CSLbAlO4Z1kvX9ejYA_TImP
 fugue_notebook/nbextension/main.js,sha256=Px2tQuBCNGEZOEBKsnfVruFEg-AxK7Tj0dY84ktub_U,3709
 fugue_polars/__init__.py,sha256=NDkjlbLhHPTjUaCAw6mAwIqeK3HSeh-z88s9dqmwheQ,61
 fugue_polars/_utils.py,sha256=7rGGWgB1-VqFwh4PcBLYk_5VNjd8FNOS4TDFyDVz2sg,159
-fugue_polars/polars_dataframe.py,sha256=Ll4ZUuRhAETWtmSf87KsdUCqZPiexFqy4FiPkvWQkN0,7348
+fugue_polars/polars_dataframe.py,sha256=8LQ0IB-JFFdjW2ltDzq8DfIbUC_jjjDr1YM29usJag0,8831
 fugue_polars/registry.py,sha256=gd6qQ-OxYtTAQFyvYbLDPXmSvCR-LW6n5K5ylgMY_7A,2950
 fugue_ray/__init__.py,sha256=HzEHfG2mpc0ugf3nf1Pdy15Bhg35K6maZpYejn1aoyI,119
 fugue_ray/_constants.py,sha256=vu5l1w-Wi-2V_nm0HLXKOYhh5HdWRCc5yQktO2XzhOg,569
-fugue_ray/dataframe.py,sha256=vyVShPnNtMef_KBsVP3iTHcssA_fm33-Y077c7S3J-A,10612
+fugue_ray/dataframe.py,sha256=7asw2qf9vm6vLBSzqghm9pUcNAppJOz5CkT7XyR0S5g,12514
 fugue_ray/execution_engine.py,sha256=NT_mnacijp1zskFbtganUwA3JNRPU-FNNvJswA6U_Yg,12607
 fugue_ray/registry.py,sha256=xJRAhbwNrg695EwghQDnVtTKi4YkqZ0_61BD4OAblSA,1685
 fugue_ray/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -130,7 +130,7 @@ fugue_ray/_utils/dataframe.py,sha256=_EadzS4rPom1A_cF0pqoPlwrNYZTfTwcyyu86_fFsqU
 fugue_ray/_utils/io.py,sha256=SFTU4qXubGEmO5IGZA5yHy8Hu4b9aFZ9-eTU4Qs-NsQ,8757
 fugue_spark/__init__.py,sha256=rvrMpFs9socMgyH_58gLbnAqmirBf5oidXoO4cekW6U,165
 fugue_spark/_constants.py,sha256=K2uLQfjvMxXk75K-7_Wn47Alpwq5rW57BtECAUrOeqA,177
-fugue_spark/dataframe.py,sha256=HJHMDVLaT-7QZ8mhMcvpLDRiKuFjtw4XtLm1N2QskKs,9704
+fugue_spark/dataframe.py,sha256=lYa8FizM3p_lsKYFR49FazkVZMJKyi2LABKTpP5YBLo,12006
 fugue_spark/execution_engine.py,sha256=rqgY9U1bpjh0GFNyNkuPcI7iV0xeipadURhNIir4w08,33147
 fugue_spark/ibis_engine.py,sha256=Yl5xxwROo1idcD2hFaylaI1IpmBUgbvOZRWtcrE0Zjo,1697
 fugue_spark/registry.py,sha256=kyIMk6dAiKRSKCHawQKyXu9DhZ24T6j3gL57TiOAZ8c,4162
@@ -144,7 +144,7 @@ fugue_sql/exceptions.py,sha256=ltS0MC8gMnVVrJbQiOZ0kRUWvVQ2LTx33dCW3ugqtb0,260
 fugue_test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 fugue_test/bag_suite.py,sha256=WbDCFjuAHYoJh4GXSPiSJxOoOwE1VMtYpJ3lQrsUK-Y,2483
 fugue_test/builtin_suite.py,sha256=o8aMZTKa74nKBmcUTTBbliTJMtNbsXE9SPKZopS504o,78400
-fugue_test/dataframe_suite.py,sha256=mOr_x94H-Ylp0lJ-KBwHXJu-Q-qesqY3PzJxR9LI_Ko,18323
+fugue_test/dataframe_suite.py,sha256=LgB931CkASbGOrRQ9j92DGk9wPb__FoNusOk-HeqU9E,19165
 fugue_test/execution_suite.py,sha256=FI6UmwBvdoT1jkJRBqJT_Q0IDehFryvv00UL6jjxyAk,47689
 fugue_test/ibis_suite.py,sha256=Dk4AHVD00RcFsNm9VvJ4_4LOyFdGX30OnAtpO2SPruE,3529
 fugue_test/plugins/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -155,9 +155,9 @@ fugue_test/plugins/duckdb/fixtures.py,sha256=UxQbIMRbSrTZ3pgCmKZgd5wd1YvnVrqLSUP
 fugue_test/plugins/ray/__init__.py,sha256=nyKGW6xgTXtMhSs7yjgFNKO7mVboCNg63Bvdf39fO_I,55
 fugue_test/plugins/ray/fixtures.py,sha256=hZkvuo0AcD63XJl5JUroc9tm2LWHUPszg2zzY6FCSao,141
 fugue_version/__init__.py,sha256=vTwvdJOZi8jZb9U-Em7-d50qNDNPS2z51IXqRoojeNM,22
-fugue-0.8.7.dev5.dist-info/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
-fugue-0.8.7.dev5.dist-info/METADATA,sha256=yQYxW_TTsinAtSyHUUfLVxrFlyc9x5FU-lfRH-77wfA,17860
-fugue-0.8.7.dev5.dist-info/WHEEL,sha256=yQN5g4mg4AybRjkgi-9yy4iQEFibGQmlz78Pik5Or-A,92
-fugue-0.8.7.dev5.dist-info/entry_points.txt,sha256=F4V76epxLiTYZgyacpmxJzNgfGqy2mUnIIG-PMlvBo8,536
-fugue-0.8.7.dev5.dist-info/top_level.txt,sha256=y1eCfzGdQ1_RkgcShcfbvXs-bopD3DwJcIOxP9EFXno,140
-fugue-0.8.7.dev5.dist-info/RECORD,,
+fugue-0.8.7.dev6.dist-info/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
+fugue-0.8.7.dev6.dist-info/METADATA,sha256=0i4ibczIy_wEMtZ6vFvaCw40x5KmuQa6OsuBVWUTQyk,17860
+fugue-0.8.7.dev6.dist-info/WHEEL,sha256=yQN5g4mg4AybRjkgi-9yy4iQEFibGQmlz78Pik5Or-A,92
+fugue-0.8.7.dev6.dist-info/entry_points.txt,sha256=N_BIIy3lSvF6Z32QE0yXTucgdHrPbUrOwH1zj7bZ0ow,536
+fugue-0.8.7.dev6.dist-info/top_level.txt,sha256=y1eCfzGdQ1_RkgcShcfbvXs-bopD3DwJcIOxP9EFXno,140
+fugue-0.8.7.dev6.dist-info/RECORD,,

{fugue-0.8.7.dev5.dist-info → fugue-0.8.7.dev6.dist-info}/entry_points.txt

@@ -7,7 +7,7 @@ ibis = fugue_ibis [ibis]
 polars = fugue_polars.registry [polars]
 ray = fugue_ray.registry [ray]
 spark = fugue_spark.registry [spark]
-spark_ibis = fugue_spark.ibis_engine [ibis,spark]
+spark_ibis = fugue_spark.ibis_engine [spark,ibis]
 
 [pytest11]
 fugue_test_dask = fugue_test.plugins.dask [dask]

fugue_dask/_utils.py

@@ -1,13 +1,14 @@
 import math
-from typing import Any, List, Optional, Tuple
+from typing import Any, Callable, List, Optional, Tuple, TypeVar
 
 import dask.dataframe as dd
 import numpy as np
 import pandas as pd
 import pyarrow as pa
 from dask.dataframe.core import DataFrame
+from dask.delayed import delayed
 from dask.distributed import Client, get_client
-from triad.utils.pandas_like import PandasLikeUtils, PD_UTILS
+from triad.utils.pandas_like import PD_UTILS, PandasLikeUtils
 from triad.utils.pyarrow import to_pandas_dtype
 
 import fugue.api as fa
@@ -16,6 +17,7 @@ from fugue.constants import FUGUE_CONF_DEFAULT_PARTITIONS
 from ._constants import FUGUE_DASK_CONF_DEFAULT_PARTITIONS
 
 _FUGUE_DASK_TEMP_IDX_COLUMN = "_fugue_dask_temp_index"
+T = TypeVar("T")
 
 
 def get_default_partitions() -> int:
@@ -28,6 +30,17 @@ def get_default_partitions() -> int:
     return n if n > 0 else fa.get_current_parallelism() * 2
 
 
+def collect(df: dd.DataFrame, func: Callable[[pd.DataFrame], T]) -> Tuple[T]:
+    """Compute each partition in parallel and collect the results
+
+    :param df: dask dataframe
+    :return: the collected result
+    """
+    dfs = df.to_delayed()
+    objs = [delayed(func)(df) for df in dfs]
+    return dd.compute(*objs)
+
+
 def hash_repartition(df: dd.DataFrame, num: int, cols: List[Any]) -> dd.DataFrame:
     """Repartition the dataframe by hashing the given columns