[SPARK-32714][PYTHON] Initial pyspark-stubs port.

dev/.rat-excludes

@@ -124,3 +124,4 @@ GangliaReporter.java
 application_1578436911597_0052
 config.properties
 app-20200706201101-0003
+py.typed

dev/tox.ini

@@ -20,5 +20,5 @@ exclude=python/pyspark/cloudpickle/*.py,shared.py,python/docs/source/conf.py,wor
 
 [flake8]
 select = E901,E999,F821,F822,F823,F401,F405
-exclude = python/pyspark/cloudpickle/*.py,shared.py,python/docs/source/conf.py,work/*/*.py,python/.eggs/*,dist/*,.git/*
+exclude = python/pyspark/cloudpickle/*.py,shared.py*,python/docs/source/conf.py,work/*/*.py,python/.eggs/*,dist/*,.git/*,python/out,python/pyspark/sql/pandas/functions.pyi,python/pyspark/sql/column.pyi,python/pyspark/worker.pyi,python/pyspark/java_gateway.pyi
 max-line-length = 100

examples/src/main/python/ml/estimator_transformer_param_example.py

@@ -56,12 +56,14 @@
     # We may alternatively specify parameters using a Python dictionary as a paramMap
     paramMap = {lr.maxIter: 20}
     paramMap[lr.maxIter] = 30  # Specify 1 Param, overwriting the original maxIter.
-    paramMap.update({lr.regParam: 0.1, lr.threshold: 0.55})  # Specify multiple Params.
+    # Specify multiple Params.
+    paramMap.update({lr.regParam: 0.1, lr.threshold: 0.55})  # type: ignore
 
     # You can combine paramMaps, which are python dictionaries.
-    paramMap2 = {lr.probabilityCol: "myProbability"}  # Change output column name
+    # Change output column name
+    paramMap2 = {lr.probabilityCol: "myProbability"}  # type: ignore
     paramMapCombined = paramMap.copy()
-    paramMapCombined.update(paramMap2)
+    paramMapCombined.update(paramMap2)  # type: ignore
 
     # Now learn a new model using the paramMapCombined parameters.
     # paramMapCombined overrides all parameters set earlier via lr.set* methods.

examples/src/main/python/ml/fm_classifier_example.py

@@ -67,9 +67,9 @@
     print("Test set accuracy = %g" % accuracy)
 
     fmModel = model.stages[2]
-    print("Factors: " + str(fmModel.factors))
-    print("Linear: " + str(fmModel.linear))
-    print("Intercept: " + str(fmModel.intercept))
+    print("Factors: " + str(fmModel.factors))  # type: ignore
+    print("Linear: " + str(fmModel.linear))  # type: ignore
+    print("Intercept: " + str(fmModel.intercept))  # type: ignore
     # $example off$
 
     spark.stop()

examples/src/main/python/ml/fm_regressor_example.py

@@ -64,9 +64,9 @@
     print("Root Mean Squared Error (RMSE) on test data = %g" % rmse)
 
     fmModel = model.stages[1]
-    print("Factors: " + str(fmModel.factors))
-    print("Linear: " + str(fmModel.linear))
-    print("Intercept: " + str(fmModel.intercept))
+    print("Factors: " + str(fmModel.factors))  # type: ignore
+    print("Linear: " + str(fmModel.linear))  # type: ignore
+    print("Intercept: " + str(fmModel.intercept))  # type: ignore
     # $example off$
 
     spark.stop()

examples/src/main/python/ml/pipeline_example.py

@@ -62,8 +62,12 @@
     prediction = model.transform(test)
     selected = prediction.select("id", "text", "probability", "prediction")
     for row in selected.collect():
-        rid, text, prob, prediction = row
-        print("(%d, %s) --> prob=%s, prediction=%f" % (rid, text, str(prob), prediction))
+        rid, text, prob, prediction = row  # type: ignore
+        print(
+            "(%d, %s) --> prob=%s, prediction=%f" % (
+                rid, text, str(prob), prediction   # type: ignore
+            )
+        )
     # $example off$
 
     spark.stop()

examples/src/main/python/sql/arrow.py

@@ -32,8 +32,8 @@
 
 
 def dataframe_with_arrow_example(spark):
-    import numpy as np
-    import pandas as pd
+    import numpy as np  # type: ignore[import]
+    import pandas as pd  # type: ignore[import]
 
     # Enable Arrow-based columnar data transfers
     spark.conf.set("spark.sql.execution.arrow.pyspark.enabled", "true")

python/MANIFEST.in

@@ -22,4 +22,5 @@ recursive-include deps/data *.data *.txt
 recursive-include deps/licenses *.txt
 recursive-include deps/examples *.py
 recursive-include lib *.zip
+recursive-include pyspark *.pyi py.typed
 include README.md

python/mypy.ini

@@ -0,0 +1,36 @@
+;
+; Licensed to the Apache Software Foundation (ASF) under one or more
+; contributor license agreements.  See the NOTICE file distributed with
+; this work for additional information regarding copyright ownership.
+; The ASF licenses this file to You 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.
+;
+
+[mypy]
+
+[mypy-pyspark.cloudpickle.*]
+ignore_errors = True
+
+[mypy-py4j.*]
+ignore_missing_imports = True
+
+[mypy-numpy]
+ignore_missing_imports = True
+
+[mypy-scipy.*]
+ignore_missing_imports = True
+
+[mypy-pandas.*]
+ignore_missing_imports = True
+
+[mypy-pyarrow]
+ignore_missing_imports = True

python/pyspark/__init__.pyi

@@ -0,0 +1,73 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you 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 typing import Callable, Optional, TypeVar
+
+from pyspark.accumulators import (  # noqa: F401
+    Accumulator as Accumulator,
+    AccumulatorParam as AccumulatorParam,
+)
+from pyspark.broadcast import Broadcast as Broadcast  # noqa: F401
+from pyspark.conf import SparkConf as SparkConf  # noqa: F401
+from pyspark.context import SparkContext as SparkContext  # noqa: F401
+from pyspark.files import SparkFiles as SparkFiles  # noqa: F401
+from pyspark.status import (
+    StatusTracker as StatusTracker,
+    SparkJobInfo as SparkJobInfo,
+    SparkStageInfo as SparkStageInfo,
+)  # noqa: F401
+from pyspark.profiler import (  # noqa: F401
+    BasicProfiler as BasicProfiler,
+    Profiler as Profiler,
+)
+from pyspark.rdd import RDD as RDD, RDDBarrier as RDDBarrier  # noqa: F401
+from pyspark.serializers import (  # noqa: F401
+    MarshalSerializer as MarshalSerializer,
+    PickleSerializer as PickleSerializer,
+)
+from pyspark.status import (  # noqa: F401
+    SparkJobInfo as SparkJobInfo,
+    SparkStageInfo as SparkStageInfo,
+    StatusTracker as StatusTracker,
+)
+from pyspark.storagelevel import StorageLevel as StorageLevel  # noqa: F401
+from pyspark.taskcontext import (  # noqa: F401
+    BarrierTaskContext as BarrierTaskContext,
+    BarrierTaskInfo as BarrierTaskInfo,
+    TaskContext as TaskContext,
+)
+from pyspark.util import InheritableThread as InheritableThread  # noqa: F401
+
+# Compatiblity imports
+from pyspark.sql import (  # noqa: F401
+    SQLContext as SQLContext,
+    HiveContext as HiveContext,
+    Row as Row,
+)
+
+T = TypeVar("T")
+F = TypeVar("F", bound=Callable)
+
+def since(version: str) -> Callable[[T], T]: ...
+def copy_func(
+    f: F,
+    name: Optional[str] = ...,
+    sinceversion: Optional[str] = ...,
+    doc: Optional[str] = ...,
+) -> F: ...
+def keyword_only(func: F) -> F: ...

python/pyspark/_globals.pyi

@@ -0,0 +1,27 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you 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.
+
+# NOTE: This dynamically typed stub was automatically generated by stubgen.
+
+from typing import Any
+
+__ALL__: Any
+
+class _NoValueType:
+    def __new__(cls): ...
+    def __reduce__(self): ...

python/pyspark/_typing.pyi

@@ -0,0 +1,33 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you 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 typing import Callable, Iterable, Sized, TypeVar, Union
+from typing_extensions import Protocol
+
+F = TypeVar("F", bound=Callable)
+T = TypeVar("T", covariant=True)
+
+PrimitiveType = Union[bool, float, int, str]
+
+class SupportsIAdd(Protocol):
+    def __iadd__(self, other: SupportsIAdd) -> SupportsIAdd: ...
+
+class SupportsOrdering(Protocol):
+    def __le__(self, other: SupportsOrdering) -> bool: ...
+
+class SizedIterable(Protocol, Sized, Iterable[T]): ...

python/pyspark/accumulators.pyi

@@ -0,0 +1,71 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you 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 typing import Callable, Generic, Tuple, Type, TypeVar
+
+import socketserver.BaseRequestHandler  # type: ignore
+
+from pyspark._typing import SupportsIAdd
+
+T = TypeVar("T")
+U = TypeVar("U", bound=SupportsIAdd)
+
+import socketserver as SocketServer
+
+class Accumulator(Generic[T]):
+    aid: int
+    accum_param: AccumulatorParam[T]
+    def __init__(
+        self, aid: int, value: T, accum_param: AccumulatorParam[T]
+    ) -> None: ...
+    def __reduce__(
+        self,
+    ) -> Tuple[
+        Callable[[int, int, AccumulatorParam[T]], Accumulator[T]],
+        Tuple[int, int, AccumulatorParam[T]],
+    ]: ...
+    @property
+    def value(self) -> T: ...
+    @value.setter
+    def value(self, value: T) -> None: ...
+    def add(self, term: T) -> None: ...
+    def __iadd__(self, term: T) -> Accumulator[T]: ...
+
+class AccumulatorParam(Generic[T]):
+    def zero(self, value: T) -> T: ...
+    def addInPlace(self, value1: T, value2: T) -> T: ...
+
+class AddingAccumulatorParam(AccumulatorParam[U]):
+    zero_value: U
+    def __init__(self, zero_value: U) -> None: ...
+    def zero(self, value: U) -> U: ...
+    def addInPlace(self, value1: U, value2: U) -> U: ...
+
+class _UpdateRequestHandler(SocketServer.StreamRequestHandler):
+    def handle(self) -> None: ...
+
+class AccumulatorServer(SocketServer.TCPServer):
+    auth_token: str
+    def __init__(
+        self,
+        server_address: Tuple[str, int],
+        RequestHandlerClass: Type[socketserver.BaseRequestHandler],
+        auth_token: str,
+    ) -> None: ...
+    server_shutdown: bool
+    def shutdown(self) -> None: ...

python/pyspark/broadcast.pyi

@@ -0,0 +1,46 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you 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 threading
+from typing import Any, Generic, Optional, TypeVar
+
+T = TypeVar("T")
+
+class Broadcast(Generic[T]):
+    def __init__(
+        self,
+        sc: Optional[Any] = ...,
+        value: Optional[T] = ...,
+        pickle_registry: Optional[Any] = ...,
+        path: Optional[Any] = ...,
+        sock_file: Optional[Any] = ...,
+    ) -> None: ...
+    def dump(self, value: Any, f: Any) -> None: ...
+    def load_from_path(self, path: Any): ...
+    def load(self, file: Any): ...
+    @property
+    def value(self) -> T: ...
+    def unpersist(self, blocking: bool = ...) -> None: ...
+    def destroy(self, blocking: bool = ...) -> None: ...
+    def __reduce__(self): ...
+
+class BroadcastPickleRegistry(threading.local):
+    def __init__(self) -> None: ...
+    def __iter__(self) -> None: ...
+    def add(self, bcast: Any) -> None: ...
+    def clear(self) -> None: ...