Source code for pyspark.context

#
# 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 os
import shutil
import signal
import sys
import threading
import warnings
import importlib
from threading import RLock
from tempfile import NamedTemporaryFile
from types import TracebackType
from typing import (
    Any,
    Callable,
    cast,
    ClassVar,
    Dict,
    Iterable,
    List,
    NoReturn,
    Optional,
    Sequence,
    Tuple,
    Type,
    TYPE_CHECKING,
    TypeVar,
    Set,
)

from py4j.java_collections import JavaMap
from py4j.protocol import Py4JError

from pyspark import accumulators
from pyspark.accumulators import Accumulator
from pyspark.broadcast import Broadcast, BroadcastPickleRegistry
from pyspark.conf import SparkConf
from pyspark.files import SparkFiles
from pyspark.java_gateway import launch_gateway, local_connect_and_auth
from pyspark.serializers import (
    CPickleSerializer,
    BatchedSerializer,
    Serializer,
    UTF8Deserializer,
    PairDeserializer,
    AutoBatchedSerializer,
    NoOpSerializer,
    ChunkedStream,
)
from pyspark.storagelevel import StorageLevel
from pyspark.resource.information import ResourceInformation
from pyspark.rdd import RDD, _load_from_socket
from pyspark.taskcontext import TaskContext
from pyspark.traceback_utils import CallSite, first_spark_call
from pyspark.status import StatusTracker
from pyspark.profiler import ProfilerCollector, BasicProfiler, UDFBasicProfiler, MemoryProfiler
from pyspark.errors import PySparkRuntimeError
from py4j.java_gateway import is_instance_of, JavaGateway, JavaObject, JVMView

if TYPE_CHECKING:
    from pyspark.accumulators import AccumulatorParam

__all__ = ["SparkContext"]


# These are special default configs for PySpark, they will overwrite
# the default ones for Spark if they are not configured by user.
DEFAULT_CONFIGS: Dict[str, Any] = {
    "spark.serializer.objectStreamReset": 100,
    "spark.rdd.compress": True,
}

T = TypeVar("T")
U = TypeVar("U")


[docs]class SparkContext: """ Main entry point for Spark functionality. A SparkContext represents the connection to a Spark cluster, and can be used to create :class:`RDD` and broadcast variables on that cluster. When you create a new SparkContext, at least the master and app name should be set, either through the named parameters here or through `conf`. Parameters ---------- master : str, optional Cluster URL to connect to (e.g. mesos://host:port, spark://host:port, local[4]). appName : str, optional A name for your job, to display on the cluster web UI. sparkHome : str, optional Location where Spark is installed on cluster nodes. pyFiles : list, optional Collection of .zip or .py files to send to the cluster and add to PYTHONPATH. These can be paths on the local file system or HDFS, HTTP, HTTPS, or FTP URLs. environment : dict, optional A dictionary of environment variables to set on worker nodes. batchSize : int, optional, default 0 The number of Python objects represented as a single Java object. Set 1 to disable batching, 0 to automatically choose the batch size based on object sizes, or -1 to use an unlimited batch size serializer : :class:`Serializer`, optional, default :class:`CPickleSerializer` The serializer for RDDs. conf : :class:`SparkConf`, optional An object setting Spark properties. gateway : class:`py4j.java_gateway.JavaGateway`, optional Use an existing gateway and JVM, otherwise a new JVM will be instantiated. This is only used internally. jsc : class:`py4j.java_gateway.JavaObject`, optional The JavaSparkContext instance. This is only used internally. profiler_cls : type, optional, default :class:`BasicProfiler` A class of custom Profiler used to do profiling udf_profiler_cls : type, optional, default :class:`UDFBasicProfiler` A class of custom Profiler used to do udf profiling Notes ----- Only one :class:`SparkContext` should be active per JVM. You must `stop()` the active :class:`SparkContext` before creating a new one. :class:`SparkContext` instance is not supported to share across multiple processes out of the box, and PySpark does not guarantee multi-processing execution. Use threads instead for concurrent processing purpose. Examples -------- >>> from pyspark.context import SparkContext >>> sc = SparkContext('local', 'test') >>> sc2 = SparkContext('local', 'test2') # doctest: +IGNORE_EXCEPTION_DETAIL Traceback (most recent call last): ... ValueError: ... """ _gateway: ClassVar[Optional[JavaGateway]] = None _jvm: ClassVar[Optional[JVMView]] = None _next_accum_id = 0 _active_spark_context: ClassVar[Optional["SparkContext"]] = None _lock = RLock() _python_includes: Optional[ List[str] ] = None # zip and egg files that need to be added to PYTHONPATH serializer: Serializer profiler_collector: ProfilerCollector PACKAGE_EXTENSIONS: Iterable[str] = (".zip", ".egg", ".jar") def __init__( self, master: Optional[str] = None, appName: Optional[str] = None, sparkHome: Optional[str] = None, pyFiles: Optional[List[str]] = None, environment: Optional[Dict[str, Any]] = None, batchSize: int = 0, serializer: "Serializer" = CPickleSerializer(), conf: Optional[SparkConf] = None, gateway: Optional[JavaGateway] = None, jsc: Optional[JavaObject] = None, profiler_cls: Type[BasicProfiler] = BasicProfiler, udf_profiler_cls: Type[UDFBasicProfiler] = UDFBasicProfiler, memory_profiler_cls: Type[MemoryProfiler] = MemoryProfiler, ): if "SPARK_CONNECT_MODE_ENABLED" in os.environ and "SPARK_LOCAL_REMOTE" not in os.environ: raise PySparkRuntimeError( error_class="CONTEXT_UNAVAILABLE_FOR_REMOTE_CLIENT", message_parameters={}, ) if conf is None or conf.get("spark.executor.allowSparkContext", "false").lower() != "true": # In order to prevent SparkContext from being created in executors. SparkContext._assert_on_driver() self._callsite = first_spark_call() or CallSite(None, None, None) if gateway is not None and gateway.gateway_parameters.auth_token is None: raise ValueError( "You are trying to pass an insecure Py4j gateway to Spark. This" " is not allowed as it is a security risk." ) SparkContext._ensure_initialized(self, gateway=gateway, conf=conf) try: self._do_init( master, appName, sparkHome, pyFiles, environment, batchSize, serializer, conf, jsc, profiler_cls, udf_profiler_cls, memory_profiler_cls, ) except BaseException: # If an error occurs, clean up in order to allow future SparkContext creation: self.stop() raise def _do_init( self, master: Optional[str], appName: Optional[str], sparkHome: Optional[str], pyFiles: Optional[List[str]], environment: Optional[Dict[str, Any]], batchSize: int, serializer: Serializer, conf: Optional[SparkConf], jsc: JavaObject, profiler_cls: Type[BasicProfiler] = BasicProfiler, udf_profiler_cls: Type[UDFBasicProfiler] = UDFBasicProfiler, memory_profiler_cls: Type[MemoryProfiler] = MemoryProfiler, ) -> None: self.environment = environment or {} # java gateway must have been launched at this point. if conf is not None and conf._jconf is not None: # conf has been initialized in JVM properly, so use conf directly. This represents the # scenario that JVM has been launched before SparkConf is created (e.g. SparkContext is # created and then stopped, and we create a new SparkConf and new SparkContext again) self._conf = conf else: self._conf = SparkConf(_jvm=SparkContext._jvm) if conf is not None: for k, v in conf.getAll(): self._conf.set(k, v) self._batchSize = batchSize # -1 represents an unlimited batch size self._unbatched_serializer = serializer if batchSize == 0: self.serializer = AutoBatchedSerializer(self._unbatched_serializer) else: self.serializer = BatchedSerializer(self._unbatched_serializer, batchSize) # Set any parameters passed directly to us on the conf if master: self._conf.setMaster(master) if appName: self._conf.setAppName(appName) if sparkHome: self._conf.setSparkHome(sparkHome) if environment: for key, value in environment.items(): self._conf.setExecutorEnv(key, value) for key, value in DEFAULT_CONFIGS.items(): self._conf.setIfMissing(key, value) # Check that we have at least the required parameters if not self._conf.contains("spark.master"): raise PySparkRuntimeError( error_class="MASTER_URL_NOT_SET", message_parameters={}, ) if not self._conf.contains("spark.app.name"): raise PySparkRuntimeError( error_class="APPLICATION_NAME_NOT_SET", message_parameters={}, ) # Read back our properties from the conf in case we loaded some of them from # the classpath or an external config file self.master = self._conf.get("spark.master") self.appName = self._conf.get("spark.app.name") self.sparkHome = self._conf.get("spark.home", None) for (k, v) in self._conf.getAll(): if k.startswith("spark.executorEnv."): varName = k[len("spark.executorEnv.") :] self.environment[varName] = v self.environment["PYTHONHASHSEED"] = os.environ.get("PYTHONHASHSEED", "0") # Create the Java SparkContext through Py4J self._jsc = jsc or self._initialize_context(self._conf._jconf) # Reset the SparkConf to the one actually used by the SparkContext in JVM. self._conf = SparkConf(_jconf=self._jsc.sc().conf()) # Create a single Accumulator in Java that we'll send all our updates through; # they will be passed back to us through a TCP server assert self._gateway is not None auth_token = self._gateway.gateway_parameters.auth_token start_update_server = accumulators._start_update_server self._accumulatorServer = start_update_server(auth_token) (host, port) = self._accumulatorServer.server_address assert self._jvm is not None self._javaAccumulator = self._jvm.PythonAccumulatorV2(host, port, auth_token) self._jsc.sc().register(self._javaAccumulator) # If encryption is enabled, we need to setup a server in the jvm to read broadcast # data via a socket. # scala's mangled names w/ $ in them require special treatment. self._encryption_enabled = self._jvm.PythonUtils.isEncryptionEnabled(self._jsc) os.environ["SPARK_AUTH_SOCKET_TIMEOUT"] = str( self._jvm.PythonUtils.getPythonAuthSocketTimeout(self._jsc) ) os.environ["SPARK_BUFFER_SIZE"] = str(self._jvm.PythonUtils.getSparkBufferSize(self._jsc)) self.pythonExec = os.environ.get("PYSPARK_PYTHON", "python3") self.pythonVer = "%d.%d" % sys.version_info[:2] # Broadcast's __reduce__ method stores Broadcast instances here. # This allows other code to determine which Broadcast instances have # been pickled, so it can determine which Java broadcast objects to # send. self._pickled_broadcast_vars = BroadcastPickleRegistry() SparkFiles._sc = self root_dir = SparkFiles.getRootDirectory() sys.path.insert(1, root_dir) # Deploy any code dependencies specified in the constructor self._python_includes = list() for path in pyFiles or []: self.addPyFile(path) # Deploy code dependencies set by spark-submit; these will already have been added # with SparkContext.addFile, so we just need to add them to the PYTHONPATH for path in self._conf.get("spark.submit.pyFiles", "").split(","): if path != "": (dirname, filename) = os.path.split(path) try: filepath = os.path.join(SparkFiles.getRootDirectory(), filename) if not os.path.exists(filepath): # In case of YARN with shell mode, 'spark.submit.pyFiles' files are # not added via SparkContext.addFile. Here we check if the file exists, # try to copy and then add it to the path. See SPARK-21945. shutil.copyfile(path, filepath) if filename[-4:].lower() in self.PACKAGE_EXTENSIONS: self._python_includes.append(filename) sys.path.insert(1, filepath) except Exception: warnings.warn( "Failed to add file [%s] specified in 'spark.submit.pyFiles' to " "Python path:\n %s" % (path, "\n ".join(sys.path)), RuntimeWarning, ) # Create a temporary directory inside spark.local.dir: assert self._jvm is not None local_dir = self._jvm.org.apache.spark.util.Utils.getLocalDir(self._jsc.sc().conf()) self._temp_dir = self._jvm.org.apache.spark.util.Utils.createTempDir( local_dir, "pyspark" ).getAbsolutePath() # profiling stats collected for each PythonRDD if ( self._conf.get("spark.python.profile", "false") == "true" or self._conf.get("spark.python.profile.memory", "false") == "true" ): dump_path = self._conf.get("spark.python.profile.dump", None) self.profiler_collector = ProfilerCollector( profiler_cls, udf_profiler_cls, memory_profiler_cls, dump_path ) else: self.profiler_collector = None # type: ignore[assignment] # create a signal handler which would be invoked on receiving SIGINT def signal_handler(signal: Any, frame: Any) -> NoReturn: self.cancelAllJobs() raise KeyboardInterrupt() # see http://stackoverflow.com/questions/23206787/ if isinstance( threading.current_thread(), threading._MainThread # type: ignore[attr-defined] ): signal.signal(signal.SIGINT, signal_handler) def __repr__(self) -> str: return "<SparkContext master={master} appName={appName}>".format( master=self.master, appName=self.appName, ) def _repr_html_(self) -> str: return """ <div> <p><b>SparkContext</b></p> <p><a href="{sc.uiWebUrl}">Spark UI</a></p> <dl> <dt>Version</dt> <dd><code>v{sc.version}</code></dd> <dt>Master</dt> <dd><code>{sc.master}</code></dd> <dt>AppName</dt> <dd><code>{sc.appName}</code></dd> </dl> </div> """.format( sc=self ) def _initialize_context(self, jconf: JavaObject) -> JavaObject: """ Initialize SparkContext in function to allow subclass specific initialization """ assert self._jvm is not None return self._jvm.JavaSparkContext(jconf) @classmethod def _ensure_initialized( cls, instance: Optional["SparkContext"] = None, gateway: Optional[JavaGateway] = None, conf: Optional[SparkConf] = None, ) -> None: """ Checks whether a SparkContext is initialized or not. Throws error if a SparkContext is already running. """ with SparkContext._lock: if not SparkContext._gateway: SparkContext._gateway = gateway or launch_gateway(conf) SparkContext._jvm = SparkContext._gateway.jvm if instance: if ( SparkContext._active_spark_context and SparkContext._active_spark_context != instance ): currentMaster = SparkContext._active_spark_context.master currentAppName = SparkContext._active_spark_context.appName callsite = SparkContext._active_spark_context._callsite # Raise error if there is already a running Spark context raise ValueError( "Cannot run multiple SparkContexts at once; " "existing SparkContext(app=%s, master=%s)" " created by %s at %s:%s " % ( currentAppName, currentMaster, callsite.function, callsite.file, callsite.linenum, ) ) else: SparkContext._active_spark_context = instance def __getnewargs__(self) -> NoReturn: # This method is called when attempting to pickle SparkContext, which is always an error: raise PySparkRuntimeError( error_class="CONTEXT_ONLY_VALID_ON_DRIVER", message_parameters={}, ) def __enter__(self) -> "SparkContext": """ Enable 'with SparkContext(...) as sc: app(sc)' syntax. """ return self def __exit__( self, type: Optional[Type[BaseException]], value: Optional[BaseException], trace: Optional[TracebackType], ) -> None: """ Enable 'with SparkContext(...) as sc: app' syntax. Specifically stop the context on exit of the with block. """ self.stop()
[docs] @classmethod def getOrCreate(cls, conf: Optional[SparkConf] = None) -> "SparkContext": """ Get or instantiate a :class:`SparkContext` and register it as a singleton object. .. versionadded:: 1.4.0 Parameters ---------- conf : :class:`SparkConf`, optional :class:`SparkConf` that will be used for initialization of the :class:`SparkContext`. Returns ------- :class:`SparkContext` current :class:`SparkContext`, or a new one if it wasn't created before the function call. Examples -------- >>> SparkContext.getOrCreate() <SparkContext ...> """ with SparkContext._lock: if SparkContext._active_spark_context is None: SparkContext(conf=conf or SparkConf()) assert SparkContext._active_spark_context is not None return SparkContext._active_spark_context
[docs] def setLogLevel(self, logLevel: str) -> None: """ Control our logLevel. This overrides any user-defined log settings. Valid log levels include: ALL, DEBUG, ERROR, FATAL, INFO, OFF, TRACE, WARN .. versionadded:: 1.4.0 Parameters ---------- logLevel : str The desired log level as a string. Examples -------- >>> sc.setLogLevel("WARN") # doctest :+SKIP """ self._jsc.setLogLevel(logLevel)
[docs] @classmethod def setSystemProperty(cls, key: str, value: str) -> None: """ Set a Java system property, such as `spark.executor.memory`. This must be invoked before instantiating :class:`SparkContext`. .. versionadded:: 0.9.0 Parameters ---------- key : str The key of a new Java system property. value : str The value of a new Java system property. """ SparkContext._ensure_initialized() assert SparkContext._jvm is not None SparkContext._jvm.java.lang.System.setProperty(key, value)
@property def version(self) -> str: """ The version of Spark on which this application is running. .. versionadded:: 1.1.0 Examples -------- >>> _ = sc.version """ return self._jsc.version() @property def applicationId(self) -> str: """ A unique identifier for the Spark application. Its format depends on the scheduler implementation. * in case of local spark app something like 'local-1433865536131' * in case of YARN something like 'application_1433865536131_34483' .. versionadded:: 1.5.0 Examples -------- >>> sc.applicationId # doctest: +ELLIPSIS 'local-...' """ return self._jsc.sc().applicationId() @property def uiWebUrl(self) -> Optional[str]: """Return the URL of the SparkUI instance started by this :class:`SparkContext` .. versionadded:: 2.1.0 Notes ----- When the web ui is disabled, e.g., by ``spark.ui.enabled`` set to ``False``, it returns ``None``. Examples -------- >>> sc.uiWebUrl 'http://...' """ jurl = self._jsc.sc().uiWebUrl() return jurl.get() if jurl.nonEmpty() else None @property def startTime(self) -> int: """Return the epoch time when the :class:`SparkContext` was started. .. versionadded:: 1.5.0 Examples -------- >>> _ = sc.startTime """ return self._jsc.startTime() @property def defaultParallelism(self) -> int: """ Default level of parallelism to use when not given by user (e.g. for reduce tasks) .. versionadded:: 0.7.0 Examples -------- >>> sc.defaultParallelism > 0 True """ return self._jsc.sc().defaultParallelism() @property def defaultMinPartitions(self) -> int: """ Default min number of partitions for Hadoop RDDs when not given by user .. versionadded:: 1.1.0 Examples -------- >>> sc.defaultMinPartitions > 0 True """ return self._jsc.sc().defaultMinPartitions()
[docs] def stop(self) -> None: """ Shut down the :class:`SparkContext`. .. versionadded:: 0.7.0 """ if getattr(self, "_jsc", None): try: self._jsc.stop() except Py4JError: # Case: SPARK-18523 warnings.warn( "Unable to cleanly shutdown Spark JVM process." " It is possible that the process has crashed," " been killed or may also be in a zombie state.", RuntimeWarning, ) finally: self._jsc = None if getattr(self, "_accumulatorServer", None): self._accumulatorServer.shutdown() self._accumulatorServer = None # type: ignore[assignment] with SparkContext._lock: SparkContext._active_spark_context = None
[docs] def emptyRDD(self) -> RDD[Any]: """ Create an :class:`RDD` that has no partitions or elements. .. versionadded:: 1.5.0 Returns ------- :class:`RDD` An empty RDD Examples -------- >>> sc.emptyRDD() EmptyRDD... >>> sc.emptyRDD().count() 0 """ return RDD(self._jsc.emptyRDD(), self, NoOpSerializer())
[docs] def range( self, start: int, end: Optional[int] = None, step: int = 1, numSlices: Optional[int] = None ) -> RDD[int]: """ Create a new RDD of int containing elements from `start` to `end` (exclusive), increased by `step` every element. Can be called the same way as python's built-in range() function. If called with a single argument, the argument is interpreted as `end`, and `start` is set to 0. .. versionadded:: 1.5.0 Parameters ---------- start : int the start value end : int, optional the end value (exclusive) step : int, optional, default 1 the incremental step numSlices : int, optional the number of partitions of the new RDD Returns ------- :class:`RDD` An RDD of int See Also -------- :meth:`pyspark.sql.SparkSession.range` Examples -------- >>> sc.range(5).collect() [0, 1, 2, 3, 4] >>> sc.range(2, 4).collect() [2, 3] >>> sc.range(1, 7, 2).collect() [1, 3, 5] Generate RDD with a negative step >>> sc.range(5, 0, -1).collect() [5, 4, 3, 2, 1] >>> sc.range(0, 5, -1).collect() [] Control the number of partitions >>> sc.range(5, numSlices=1).getNumPartitions() 1 >>> sc.range(5, numSlices=10).getNumPartitions() 10 """ if end is None: end = start start = 0 return self.parallelize(range(start, end, step), numSlices)
[docs] def parallelize(self, c: Iterable[T], numSlices: Optional[int] = None) -> RDD[T]: """ Distribute a local Python collection to form an RDD. Using range is recommended if the input represents a range for performance. .. versionadded:: 0.7.0 Parameters ---------- c : :class:`collections.abc.Iterable` iterable collection to distribute numSlices : int, optional the number of partitions of the new RDD Returns ------- :class:`RDD` RDD representing distributed collection. Examples -------- >>> sc.parallelize([0, 2, 3, 4, 6], 5).glom().collect() [[0], [2], [3], [4], [6]] >>> sc.parallelize(range(0, 6, 2), 5).glom().collect() [[], [0], [], [2], [4]] Deal with a list of strings. >>> strings = ["a", "b", "c"] >>> sc.parallelize(strings, 2).glom().collect() [['a'], ['b', 'c']] """ numSlices = int(numSlices) if numSlices is not None else self.defaultParallelism if isinstance(c, range): size = len(c) if size == 0: return self.parallelize([], numSlices) step = c[1] - c[0] if size > 1 else 1 # type: ignore[index] start0 = c[0] # type: ignore[index] def getStart(split: int) -> int: assert numSlices is not None return start0 + int((split * size / numSlices)) * step def f(split: int, iterator: Iterable[T]) -> Iterable: # it's an empty iterator here but we need this line for triggering the # logic of signal handling in FramedSerializer.load_stream, for instance, # SpecialLengths.END_OF_DATA_SECTION in _read_with_length. Since # FramedSerializer.load_stream produces a generator, the control should # at least be in that function once. Here we do it by explicitly converting # the empty iterator to a list, thus make sure worker reuse takes effect. # See more details in SPARK-26549. assert len(list(iterator)) == 0 return range(getStart(split), getStart(split + 1), step) return self.parallelize([], numSlices).mapPartitionsWithIndex(f) # Make sure we distribute data evenly if it's smaller than self.batchSize if "__len__" not in dir(c): c = list(c) # Make it a list so we can compute its length batchSize = max( 1, min(len(c) // numSlices, self._batchSize or 1024) # type: ignore[arg-type] ) serializer = BatchedSerializer(self._unbatched_serializer, batchSize) def reader_func(temp_filename: str) -> JavaObject: assert self._jvm is not None return self._jvm.PythonRDD.readRDDFromFile(self._jsc, temp_filename, numSlices) def createRDDServer() -> JavaObject: assert self._jvm is not None return self._jvm.PythonParallelizeServer(self._jsc.sc(), numSlices) jrdd = self._serialize_to_jvm(c, serializer, reader_func, createRDDServer) return RDD(jrdd, self, serializer)
def _serialize_to_jvm( self, data: Iterable[T], serializer: Serializer, reader_func: Callable, server_func: Callable, ) -> JavaObject: """ Using Py4J to send a large dataset to the jvm is slow, so we use either a file or a socket if we have encryption enabled. Examples -------- data object to be serialized serializer : class:`pyspark.serializers.Serializer` reader_func : function A function which takes a filename and reads in the data in the jvm and returns a JavaRDD. Only used when encryption is disabled. server_func : function A function which creates a SocketAuthServer in the JVM to accept the serialized data, for use when encryption is enabled. """ if self._encryption_enabled: # with encryption, we open a server in java and send the data directly server = server_func() (sock_file, _) = local_connect_and_auth(server.port(), server.secret()) chunked_out = ChunkedStream(sock_file, 8192) serializer.dump_stream(data, chunked_out) chunked_out.close() # this call will block until the server has read all the data and processed it (or # throws an exception) r = server.getResult() return r else: # without encryption, we serialize to a file, and we read the file in java and # parallelize from there. tempFile = NamedTemporaryFile(delete=False, dir=self._temp_dir) try: try: serializer.dump_stream(data, tempFile) finally: tempFile.close() return reader_func(tempFile.name) finally: # we eagerly reads the file so we can delete right after. os.unlink(tempFile.name)
[docs] def pickleFile(self, name: str, minPartitions: Optional[int] = None) -> RDD[Any]: """ Load an RDD previously saved using :meth:`RDD.saveAsPickleFile` method. .. versionadded:: 1.1.0 Parameters ---------- name : str directory to the input data files, the path can be comma separated paths as a list of inputs minPartitions : int, optional suggested minimum number of partitions for the resulting RDD Returns ------- :class:`RDD` RDD representing unpickled data from the file(s). See Also -------- :meth:`RDD.saveAsPickleFile` Examples -------- >>> import os >>> import tempfile >>> with tempfile.TemporaryDirectory() as d: ... # Write a temporary pickled file ... path1 = os.path.join(d, "pickled1") ... sc.parallelize(range(10)).saveAsPickleFile(path1, 3) ... ... # Write another temporary pickled file ... path2 = os.path.join(d, "pickled2") ... sc.parallelize(range(-10, -5)).saveAsPickleFile(path2, 3) ... ... # Load picked file ... collected1 = sorted(sc.pickleFile(path1, 3).collect()) ... collected2 = sorted(sc.pickleFile(path2, 4).collect()) ... ... # Load two picked files together ... collected3 = sorted(sc.pickleFile('{},{}'.format(path1, path2), 5).collect()) >>> collected1 [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] >>> collected2 [-10, -9, -8, -7, -6] >>> collected3 [-10, -9, -8, -7, -6, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9] """ minPartitions = minPartitions or self.defaultMinPartitions return RDD(self._jsc.objectFile(name, minPartitions), self)
[docs] def textFile( self, name: str, minPartitions: Optional[int] = None, use_unicode: bool = True ) -> RDD[str]: """ Read a text file from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI, and return it as an RDD of Strings. The text files must be encoded as UTF-8. .. versionadded:: 0.7.0 Parameters ---------- name : str directory to the input data files, the path can be comma separated paths as a list of inputs minPartitions : int, optional suggested minimum number of partitions for the resulting RDD use_unicode : bool, default True If `use_unicode` is False, the strings will be kept as `str` (encoding as `utf-8`), which is faster and smaller than unicode. .. versionadded:: 1.2.0 Returns ------- :class:`RDD` RDD representing text data from the file(s). See Also -------- :meth:`RDD.saveAsTextFile` :meth:`SparkContext.wholeTextFiles` Examples -------- >>> import os >>> import tempfile >>> with tempfile.TemporaryDirectory() as d: ... path1 = os.path.join(d, "text1") ... path2 = os.path.join(d, "text2") ... ... # Write a temporary text file ... sc.parallelize(["x", "y", "z"]).saveAsTextFile(path1) ... ... # Write another temporary text file ... sc.parallelize(["aa", "bb", "cc"]).saveAsTextFile(path2) ... ... # Load text file ... collected1 = sorted(sc.textFile(path1, 3).collect()) ... collected2 = sorted(sc.textFile(path2, 4).collect()) ... ... # Load two text files together ... collected3 = sorted(sc.textFile('{},{}'.format(path1, path2), 5).collect()) >>> collected1 ['x', 'y', 'z'] >>> collected2 ['aa', 'bb', 'cc'] >>> collected3 ['aa', 'bb', 'cc', 'x', 'y', 'z'] """ minPartitions = minPartitions or min(self.defaultParallelism, 2) return RDD(self._jsc.textFile(name, minPartitions), self, UTF8Deserializer(use_unicode))
[docs] def wholeTextFiles( self, path: str, minPartitions: Optional[int] = None, use_unicode: bool = True ) -> RDD[Tuple[str, str]]: """ Read a directory of text files from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI. Each file is read as a single record and returned in a key-value pair, where the key is the path of each file, the value is the content of each file. The text files must be encoded as UTF-8. .. versionadded:: 1.0.0 For example, if you have the following files: .. code-block:: text hdfs://a-hdfs-path/part-00000 hdfs://a-hdfs-path/part-00001 ... hdfs://a-hdfs-path/part-nnnnn Do ``rdd = sparkContext.wholeTextFiles("hdfs://a-hdfs-path")``, then ``rdd`` contains: .. code-block:: text (a-hdfs-path/part-00000, its content) (a-hdfs-path/part-00001, its content) ... (a-hdfs-path/part-nnnnn, its content) Parameters ---------- path : str directory to the input data files, the path can be comma separated paths as a list of inputs minPartitions : int, optional suggested minimum number of partitions for the resulting RDD use_unicode : bool, default True If `use_unicode` is False, the strings will be kept as `str` (encoding as `utf-8`), which is faster and smaller than unicode. .. versionadded:: 1.2.0 Returns ------- :class:`RDD` RDD representing path-content pairs from the file(s). Notes ----- Small files are preferred, as each file will be loaded fully in memory. See Also -------- :meth:`RDD.saveAsTextFile` :meth:`SparkContext.textFile` Examples -------- >>> import os >>> import tempfile >>> with tempfile.TemporaryDirectory() as d: ... # Write a temporary text file ... with open(os.path.join(d, "1.txt"), "w") as f: ... _ = f.write("123") ... ... # Write another temporary text file ... with open(os.path.join(d, "2.txt"), "w") as f: ... _ = f.write("xyz") ... ... collected = sorted(sc.wholeTextFiles(d).collect()) >>> collected [('.../1.txt', '123'), ('.../2.txt', 'xyz')] """ minPartitions = minPartitions or self.defaultMinPartitions return RDD( self._jsc.wholeTextFiles(path, minPartitions), self, PairDeserializer(UTF8Deserializer(use_unicode), UTF8Deserializer(use_unicode)), )
[docs] def binaryFiles(self, path: str, minPartitions: Optional[int] = None) -> RDD[Tuple[str, bytes]]: """ Read a directory of binary files from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI as a byte array. Each file is read as a single record and returned in a key-value pair, where the key is the path of each file, the value is the content of each file. .. versionadded:: 1.3.0 Parameters ---------- path : str directory to the input data files, the path can be comma separated paths as a list of inputs minPartitions : int, optional suggested minimum number of partitions for the resulting RDD Returns ------- :class:`RDD` RDD representing path-content pairs from the file(s). Notes ----- Small files are preferred, large file is also allowable, but may cause bad performance. See Also -------- :meth:`SparkContext.binaryRecords` Examples -------- >>> import os >>> import tempfile >>> with tempfile.TemporaryDirectory() as d: ... # Write a temporary binary file ... with open(os.path.join(d, "1.bin"), "wb") as f1: ... _ = f1.write(b"binary data I") ... ... # Write another temporary binary file ... with open(os.path.join(d, "2.bin"), "wb") as f2: ... _ = f2.write(b"binary data II") ... ... collected = sorted(sc.binaryFiles(d).collect()) >>> collected [('.../1.bin', b'binary data I'), ('.../2.bin', b'binary data II')] """ minPartitions = minPartitions or self.defaultMinPartitions return RDD( self._jsc.binaryFiles(path, minPartitions), self, PairDeserializer(UTF8Deserializer(), NoOpSerializer()), )
[docs] def binaryRecords(self, path: str, recordLength: int) -> RDD[bytes]: """ Load data from a flat binary file, assuming each record is a set of numbers with the specified numerical format (see ByteBuffer), and the number of bytes per record is constant. .. versionadded:: 1.3.0 Parameters ---------- path : str Directory to the input data files recordLength : int The length at which to split the records Returns ------- :class:`RDD` RDD of data with values, represented as byte arrays See Also -------- :meth:`SparkContext.binaryFiles` Examples -------- >>> import os >>> import tempfile >>> with tempfile.TemporaryDirectory() as d: ... # Write a temporary file ... with open(os.path.join(d, "1.bin"), "w") as f: ... for i in range(3): ... _ = f.write("%04d" % i) ... ... # Write another file ... with open(os.path.join(d, "2.bin"), "w") as f: ... for i in [-1, -2, -10]: ... _ = f.write("%04d" % i) ... ... collected = sorted(sc.binaryRecords(d, 4).collect()) >>> collected [b'-001', b'-002', b'-010', b'0000', b'0001', b'0002'] """ return RDD(self._jsc.binaryRecords(path, recordLength), self, NoOpSerializer())
def _dictToJavaMap(self, d: Optional[Dict[str, str]]) -> JavaMap: assert self._jvm is not None jm = self._jvm.java.util.HashMap() if not d: d = {} for k, v in d.items(): jm[k] = v return jm
[docs] def sequenceFile( self, path: str, keyClass: Optional[str] = None, valueClass: Optional[str] = None, keyConverter: Optional[str] = None, valueConverter: Optional[str] = None, minSplits: Optional[int] = None, batchSize: int = 0, ) -> RDD[Tuple[T, U]]: """ Read a Hadoop SequenceFile with arbitrary key and value Writable class from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI. The mechanism is as follows: 1. A Java RDD is created from the SequenceFile or other InputFormat, and the key and value Writable classes 2. Serialization is attempted via Pickle pickling 3. If this fails, the fallback is to call 'toString' on each key and value 4. :class:`CPickleSerializer` is used to deserialize pickled objects on the Python side .. versionadded:: 1.3.0 Parameters ---------- path : str path to sequencefile keyClass: str, optional fully qualified classname of key Writable class (e.g. "org.apache.hadoop.io.Text") valueClass : str, optional fully qualified classname of value Writable class (e.g. "org.apache.hadoop.io.LongWritable") keyConverter : str, optional fully qualified name of a function returning key WritableConverter valueConverter : str, optional fully qualifiedname of a function returning value WritableConverter minSplits : int, optional minimum splits in dataset (default min(2, sc.defaultParallelism)) batchSize : int, optional, default 0 The number of Python objects represented as a single Java object. (default 0, choose batchSize automatically) Returns ------- :class:`RDD` RDD of tuples of key and corresponding value See Also -------- :meth:`RDD.saveAsSequenceFile` :meth:`RDD.saveAsNewAPIHadoopFile` :meth:`RDD.saveAsHadoopFile` :meth:`SparkContext.newAPIHadoopFile` :meth:`SparkContext.hadoopFile` Examples -------- >>> import os >>> import tempfile Set the class of output format >>> output_format_class = "org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat" >>> with tempfile.TemporaryDirectory() as d: ... path = os.path.join(d, "hadoop_file") ... ... # Write a temporary Hadoop file ... rdd = sc.parallelize([(1, {3.0: "bb"}), (2, {1.0: "aa"}), (3, {2.0: "dd"})]) ... rdd.saveAsNewAPIHadoopFile(path, output_format_class) ... ... collected = sorted(sc.sequenceFile(path).collect()) >>> collected [(1, {3.0: 'bb'}), (2, {1.0: 'aa'}), (3, {2.0: 'dd'})] """ minSplits = minSplits or min(self.defaultParallelism, 2) assert self._jvm is not None jrdd = self._jvm.PythonRDD.sequenceFile( self._jsc, path, keyClass, valueClass, keyConverter, valueConverter, minSplits, batchSize, ) return RDD(jrdd, self)
[docs] def newAPIHadoopFile( self, path: str, inputFormatClass: str, keyClass: str, valueClass: str, keyConverter: Optional[str] = None, valueConverter: Optional[str] = None, conf: Optional[Dict[str, str]] = None, batchSize: int = 0, ) -> RDD[Tuple[T, U]]: """ Read a 'new API' Hadoop InputFormat with arbitrary key and value class from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI. The mechanism is the same as for meth:`SparkContext.sequenceFile`. A Hadoop configuration can be passed in as a Python dict. This will be converted into a Configuration in Java .. versionadded:: 1.1.0 Parameters ---------- path : str path to Hadoop file inputFormatClass : str fully qualified classname of Hadoop InputFormat (e.g. "org.apache.hadoop.mapreduce.lib.input.TextInputFormat") keyClass : str fully qualified classname of key Writable class (e.g. "org.apache.hadoop.io.Text") valueClass : str fully qualified classname of value Writable class (e.g. "org.apache.hadoop.io.LongWritable") keyConverter : str, optional fully qualified name of a function returning key WritableConverter None by default valueConverter : str, optional fully qualified name of a function returning value WritableConverter None by default conf : dict, optional Hadoop configuration, passed in as a dict None by default batchSize : int, optional, default 0 The number of Python objects represented as a single Java object. (default 0, choose batchSize automatically) Returns ------- :class:`RDD` RDD of tuples of key and corresponding value See Also -------- :meth:`RDD.saveAsSequenceFile` :meth:`RDD.saveAsNewAPIHadoopFile` :meth:`RDD.saveAsHadoopFile` :meth:`SparkContext.sequenceFile` :meth:`SparkContext.hadoopFile` Examples -------- >>> import os >>> import tempfile Set the related classes >>> output_format_class = "org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat" >>> input_format_class = "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat" >>> key_class = "org.apache.hadoop.io.IntWritable" >>> value_class = "org.apache.hadoop.io.Text" >>> with tempfile.TemporaryDirectory() as d: ... path = os.path.join(d, "new_hadoop_file") ... ... # Write a temporary Hadoop file ... rdd = sc.parallelize([(1, ""), (1, "a"), (3, "x")]) ... rdd.saveAsNewAPIHadoopFile(path, output_format_class, key_class, value_class) ... ... loaded = sc.newAPIHadoopFile(path, input_format_class, key_class, value_class) ... collected = sorted(loaded.collect()) >>> collected [(1, ''), (1, 'a'), (3, 'x')] """ jconf = self._dictToJavaMap(conf) assert self._jvm is not None jrdd = self._jvm.PythonRDD.newAPIHadoopFile( self._jsc, path, inputFormatClass, keyClass, valueClass, keyConverter, valueConverter, jconf, batchSize, ) return RDD(jrdd, self)
[docs] def newAPIHadoopRDD( self, inputFormatClass: str, keyClass: str, valueClass: str, keyConverter: Optional[str] = None, valueConverter: Optional[str] = None, conf: Optional[Dict[str, str]] = None, batchSize: int = 0, ) -> RDD[Tuple[T, U]]: """ Read a 'new API' Hadoop InputFormat with arbitrary key and value class, from an arbitrary Hadoop configuration, which is passed in as a Python dict. This will be converted into a Configuration in Java. The mechanism is the same as for meth:`SparkContext.sequenceFile`. .. versionadded:: 1.1.0 Parameters ---------- inputFormatClass : str fully qualified classname of Hadoop InputFormat (e.g. "org.apache.hadoop.mapreduce.lib.input.TextInputFormat") keyClass : str fully qualified classname of key Writable class (e.g. "org.apache.hadoop.io.Text") valueClass : str fully qualified classname of value Writable class (e.g. "org.apache.hadoop.io.LongWritable") keyConverter : str, optional fully qualified name of a function returning key WritableConverter (None by default) valueConverter : str, optional fully qualified name of a function returning value WritableConverter (None by default) conf : dict, optional Hadoop configuration, passed in as a dict (None by default) batchSize : int, optional, default 0 The number of Python objects represented as a single Java object. (default 0, choose batchSize automatically) Returns ------- :class:`RDD` RDD of tuples of key and corresponding value See Also -------- :meth:`RDD.saveAsNewAPIHadoopDataset` :meth:`RDD.saveAsHadoopDataset` :meth:`SparkContext.hadoopRDD` :meth:`SparkContext.hadoopFile` Examples -------- >>> import os >>> import tempfile Set the related classes >>> output_format_class = "org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat" >>> input_format_class = "org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat" >>> key_class = "org.apache.hadoop.io.IntWritable" >>> value_class = "org.apache.hadoop.io.Text" >>> with tempfile.TemporaryDirectory() as d: ... path = os.path.join(d, "new_hadoop_file") ... ... # Create the conf for writing ... write_conf = { ... "mapreduce.job.outputformat.class": (output_format_class), ... "mapreduce.job.output.key.class": key_class, ... "mapreduce.job.output.value.class": value_class, ... "mapreduce.output.fileoutputformat.outputdir": path, ... } ... ... # Write a temporary Hadoop file ... rdd = sc.parallelize([(1, ""), (1, "a"), (3, "x")]) ... rdd.saveAsNewAPIHadoopDataset(conf=write_conf) ... ... # Create the conf for reading ... read_conf = {"mapreduce.input.fileinputformat.inputdir": path} ... ... loaded = sc.newAPIHadoopRDD(input_format_class, ... key_class, value_class, conf=read_conf) ... collected = sorted(loaded.collect()) >>> collected [(1, ''), (1, 'a'), (3, 'x')] """ jconf = self._dictToJavaMap(conf) assert self._jvm is not None jrdd = self._jvm.PythonRDD.newAPIHadoopRDD( self._jsc, inputFormatClass, keyClass, valueClass, keyConverter, valueConverter, jconf, batchSize, ) return RDD(jrdd, self)
[docs] def hadoopFile( self, path: str, inputFormatClass: str, keyClass: str, valueClass: str, keyConverter: Optional[str] = None, valueConverter: Optional[str] = None, conf: Optional[Dict[str, str]] = None, batchSize: int = 0, ) -> RDD[Tuple[T, U]]: """ Read an 'old' Hadoop InputFormat with arbitrary key and value class from HDFS, a local file system (available on all nodes), or any Hadoop-supported file system URI. The mechanism is the same as for meth:`SparkContext.sequenceFile`. .. versionadded:: 1.1.0 A Hadoop configuration can be passed in as a Python dict. This will be converted into a Configuration in Java. Parameters ---------- path : str path to Hadoop file inputFormatClass : str fully qualified classname of Hadoop InputFormat (e.g. "org.apache.hadoop.mapreduce.lib.input.TextInputFormat") keyClass : str fully qualified classname of key Writable class (e.g. "org.apache.hadoop.io.Text") valueClass : str fully qualified classname of value Writable class (e.g. "org.apache.hadoop.io.LongWritable") keyConverter : str, optional fully qualified name of a function returning key WritableConverter valueConverter : str, optional fully qualified name of a function returning value WritableConverter conf : dict, optional Hadoop configuration, passed in as a dict batchSize : int, optional, default 0 The number of Python objects represented as a single Java object. (default 0, choose batchSize automatically) Returns ------- :class:`RDD` RDD of tuples of key and corresponding value See Also -------- :meth:`RDD.saveAsSequenceFile` :meth:`RDD.saveAsNewAPIHadoopFile` :meth:`RDD.saveAsHadoopFile` :meth:`SparkContext.newAPIHadoopFile` :meth:`SparkContext.hadoopRDD` Examples -------- >>> import os >>> import tempfile Set the related classes >>> output_format_class = "org.apache.hadoop.mapred.TextOutputFormat" >>> input_format_class = "org.apache.hadoop.mapred.TextInputFormat" >>> key_class = "org.apache.hadoop.io.IntWritable" >>> value_class = "org.apache.hadoop.io.Text" >>> with tempfile.TemporaryDirectory() as d: ... path = os.path.join(d, "old_hadoop_file") ... ... # Write a temporary Hadoop file ... rdd = sc.parallelize([(1, ""), (1, "a"), (3, "x")]) ... rdd.saveAsHadoopFile(path, output_format_class, key_class, value_class) ... ... loaded = sc.hadoopFile(path, input_format_class, key_class, value_class) ... collected = sorted(loaded.collect()) >>> collected [(0, '1\\t'), (0, '1\\ta'), (0, '3\\tx')] """ jconf = self._dictToJavaMap(conf) assert self._jvm is not None jrdd = self._jvm.PythonRDD.hadoopFile( self._jsc, path, inputFormatClass, keyClass, valueClass, keyConverter, valueConverter, jconf, batchSize, ) return RDD(jrdd, self)
[docs] def hadoopRDD( self, inputFormatClass: str, keyClass: str, valueClass: str, keyConverter: Optional[str] = None, valueConverter: Optional[str] = None, conf: Optional[Dict[str, str]] = None, batchSize: int = 0, ) -> RDD[Tuple[T, U]]: """ Read an 'old' Hadoop InputFormat with arbitrary key and value class, from an arbitrary Hadoop configuration, which is passed in as a Python dict. This will be converted into a Configuration in Java. The mechanism is the same as for meth:`SparkContext.sequenceFile`. .. versionadded:: 1.1.0 Parameters ---------- inputFormatClass : str fully qualified classname of Hadoop InputFormat (e.g. "org.apache.hadoop.mapreduce.lib.input.TextInputFormat") keyClass : str fully qualified classname of key Writable class (e.g. "org.apache.hadoop.io.Text") valueClass : str fully qualified classname of value Writable class (e.g. "org.apache.hadoop.io.LongWritable") keyConverter : str, optional fully qualified name of a function returning key WritableConverter valueConverter : str, optional fully qualified name of a function returning value WritableConverter conf : dict, optional Hadoop configuration, passed in as a dict batchSize : int, optional, default 0 The number of Python objects represented as a single Java object. (default 0, choose batchSize automatically) Returns ------- :class:`RDD` RDD of tuples of key and corresponding value See Also -------- :meth:`RDD.saveAsNewAPIHadoopDataset` :meth:`RDD.saveAsHadoopDataset` :meth:`SparkContext.newAPIHadoopRDD` :meth:`SparkContext.hadoopFile` Examples -------- >>> import os >>> import tempfile Set the related classes >>> output_format_class = "org.apache.hadoop.mapred.TextOutputFormat" >>> input_format_class = "org.apache.hadoop.mapred.TextInputFormat" >>> key_class = "org.apache.hadoop.io.IntWritable" >>> value_class = "org.apache.hadoop.io.Text" >>> with tempfile.TemporaryDirectory() as d: ... path = os.path.join(d, "old_hadoop_file") ... ... # Create the conf for writing ... write_conf = { ... "mapred.output.format.class": output_format_class, ... "mapreduce.job.output.key.class": key_class, ... "mapreduce.job.output.value.class": value_class, ... "mapreduce.output.fileoutputformat.outputdir": path, ... } ... ... # Write a temporary Hadoop file ... rdd = sc.parallelize([(1, ""), (1, "a"), (3, "x")]) ... rdd.saveAsHadoopDataset(conf=write_conf) ... ... # Create the conf for reading ... read_conf = {"mapreduce.input.fileinputformat.inputdir": path} ... ... loaded = sc.hadoopRDD(input_format_class, key_class, value_class, conf=read_conf) ... collected = sorted(loaded.collect()) >>> collected [(0, '1\\t'), (0, '1\\ta'), (0, '3\\tx')] """ jconf = self._dictToJavaMap(conf) assert self._jvm is not None jrdd = self._jvm.PythonRDD.hadoopRDD( self._jsc, inputFormatClass, keyClass, valueClass, keyConverter, valueConverter, jconf, batchSize, ) return RDD(jrdd, self)
def _checkpointFile(self, name: str, input_deserializer: PairDeserializer) -> RDD: jrdd = self._jsc.checkpointFile(name) return RDD(jrdd, self, input_deserializer)
[docs] def union(self, rdds: List[RDD[T]]) -> RDD[T]: """ Build the union of a list of RDDs. This supports unions() of RDDs with different serialized formats, although this forces them to be reserialized using the default serializer: .. versionadded:: 0.7.0 See Also -------- :meth:`RDD.union` Examples -------- >>> import os >>> import tempfile >>> with tempfile.TemporaryDirectory() as d: ... # generate a text RDD ... with open(os.path.join(d, "union-text.txt"), "w") as f: ... _ = f.write("Hello") ... text_rdd = sc.textFile(d) ... ... # generate another RDD ... parallelized = sc.parallelize(["World!"]) ... ... unioned = sorted(sc.union([text_rdd, parallelized]).collect()) >>> unioned ['Hello', 'World!'] """ first_jrdd_deserializer = rdds[0]._jrdd_deserializer if any(x._jrdd_deserializer != first_jrdd_deserializer for x in rdds): rdds = [x._reserialize() for x in rdds] gw = SparkContext._gateway assert gw is not None jvm = SparkContext._jvm assert jvm is not None jrdd_cls = jvm.org.apache.spark.api.java.JavaRDD jpair_rdd_cls = jvm.org.apache.spark.api.java.JavaPairRDD jdouble_rdd_cls = jvm.org.apache.spark.api.java.JavaDoubleRDD if is_instance_of(gw, rdds[0]._jrdd, jrdd_cls): cls = jrdd_cls elif is_instance_of(gw, rdds[0]._jrdd, jpair_rdd_cls): cls = jpair_rdd_cls elif is_instance_of(gw, rdds[0]._jrdd, jdouble_rdd_cls): cls = jdouble_rdd_cls else: cls_name = rdds[0]._jrdd.getClass().getCanonicalName() raise TypeError("Unsupported Java RDD class %s" % cls_name) jrdds = gw.new_array(cls, len(rdds)) for i in range(0, len(rdds)): jrdds[i] = rdds[i]._jrdd return RDD(self._jsc.union(jrdds), self, rdds[0]._jrdd_deserializer)
[docs] def broadcast(self, value: T) -> "Broadcast[T]": """ Broadcast a read-only variable to the cluster, returning a :class:`Broadcast` object for reading it in distributed functions. The variable will be sent to each cluster only once. .. versionadded:: 0.7.0 Parameters ---------- value : T value to broadcast to the Spark nodes Returns ------- :class:`Broadcast` :class:`Broadcast` object, a read-only variable cached on each machine Examples -------- >>> mapping = {1: 10001, 2: 10002} >>> bc = sc.broadcast(mapping) >>> rdd = sc.range(5) >>> rdd2 = rdd.map(lambda i: bc.value[i] if i in bc.value else -1) >>> rdd2.collect() [-1, 10001, 10002, -1, -1] >>> bc.destroy() """ return Broadcast(self, value, self._pickled_broadcast_vars)
[docs] def accumulator( self, value: T, accum_param: Optional["AccumulatorParam[T]"] = None ) -> "Accumulator[T]": """ Create an :class:`Accumulator` with the given initial value, using a given :class:`AccumulatorParam` helper object to define how to add values of the data type if provided. Default AccumulatorParams are used for integers and floating-point numbers if you do not provide one. For other types, a custom AccumulatorParam can be used. .. versionadded:: 0.7.0 Parameters ---------- value : T initialized value accum_param : :class:`pyspark.AccumulatorParam`, optional helper object to define how to add values Returns ------- :class:`Accumulator` `Accumulator` object, a shared variable that can be accumulated Examples -------- >>> acc = sc.accumulator(9) >>> acc.value 9 >>> acc += 1 >>> acc.value 10 Accumulator object can be accumulated in RDD operations: >>> rdd = sc.range(5) >>> def f(x): ... global acc ... acc += 1 ... >>> rdd.foreach(f) >>> acc.value 15 """ if accum_param is None: if isinstance(value, int): accum_param = cast("AccumulatorParam[T]", accumulators.INT_ACCUMULATOR_PARAM) elif isinstance(value, float): accum_param = cast("AccumulatorParam[T]", accumulators.FLOAT_ACCUMULATOR_PARAM) elif isinstance(value, complex): accum_param = cast("AccumulatorParam[T]", accumulators.COMPLEX_ACCUMULATOR_PARAM) else: raise TypeError("No default accumulator param for type %s" % type(value)) SparkContext._next_accum_id += 1 return Accumulator(SparkContext._next_accum_id - 1, value, accum_param)
[docs] def addFile(self, path: str, recursive: bool = False) -> None: """ Add a file to be downloaded with this Spark job on every node. The `path` passed can be either a local file, a file in HDFS (or other Hadoop-supported filesystems), or an HTTP, HTTPS or FTP URI. To access the file in Spark jobs, use :meth:`SparkFiles.get` with the filename to find its download location. A directory can be given if the recursive option is set to True. Currently directories are only supported for Hadoop-supported filesystems. .. versionadded:: 0.7.0 Parameters ---------- path : str can be either a local file, a file in HDFS (or other Hadoop-supported filesystems), or an HTTP, HTTPS or FTP URI. To access the file in Spark jobs, use :meth:`SparkFiles.get` to find its download location. recursive : bool, default False whether to recursively add files in the input directory See Also -------- :meth:`SparkContext.listFiles` :meth:`SparkContext.addPyFile` :meth:`SparkFiles.get` Notes ----- A path can be added only once. Subsequent additions of the same path are ignored. Examples -------- >>> import os >>> import tempfile >>> from pyspark import SparkFiles >>> with tempfile.TemporaryDirectory() as d: ... path1 = os.path.join(d, "test1.txt") ... with open(path1, "w") as f: ... _ = f.write("100") ... ... path2 = os.path.join(d, "test2.txt") ... with open(path2, "w") as f: ... _ = f.write("200") ... ... sc.addFile(path1) ... file_list1 = sorted(sc.listFiles) ... ... sc.addFile(path2) ... file_list2 = sorted(sc.listFiles) ... ... # add path2 twice, this addition will be ignored ... sc.addFile(path2) ... file_list3 = sorted(sc.listFiles) ... ... def func(iterator): ... with open(SparkFiles.get("test1.txt")) as f: ... mul = int(f.readline()) ... return [x * mul for x in iterator] ... ... collected = sc.parallelize([1, 2, 3, 4]).mapPartitions(func).collect() >>> file_list1 ['file:/.../test1.txt'] >>> file_list2 ['file:/.../test1.txt', 'file:/.../test2.txt'] >>> file_list3 ['file:/.../test1.txt', 'file:/.../test2.txt'] >>> collected [100, 200, 300, 400] """ self._jsc.sc().addFile(path, recursive)
@property def listFiles(self) -> List[str]: """Returns a list of file paths that are added to resources. .. versionadded:: 3.4.0 See Also -------- :meth:`SparkContext.addFile` """ return list( self._jvm.scala.collection.JavaConverters.seqAsJavaList( # type: ignore[union-attr] self._jsc.sc().listFiles() ) )
[docs] def addPyFile(self, path: str) -> None: """ Add a .py or .zip dependency for all tasks to be executed on this SparkContext in the future. The `path` passed can be either a local file, a file in HDFS (or other Hadoop-supported filesystems), or an HTTP, HTTPS or FTP URI. .. versionadded:: 0.7.0 Parameters ---------- path : str can be either a .py file or .zip dependency. See Also -------- :meth:`SparkContext.addFile` Notes ----- A path can be added only once. Subsequent additions of the same path are ignored. """ self.addFile(path) (dirname, filename) = os.path.split(path) # dirname may be directory or HDFS/S3 prefix if filename[-4:].lower() in self.PACKAGE_EXTENSIONS: assert self._python_includes is not None self._python_includes.append(filename) # for tests in local mode sys.path.insert(1, os.path.join(SparkFiles.getRootDirectory(), filename)) importlib.invalidate_caches()
[docs] def addArchive(self, path: str) -> None: """ Add an archive to be downloaded with this Spark job on every node. The `path` passed can be either a local file, a file in HDFS (or other Hadoop-supported filesystems), or an HTTP, HTTPS or FTP URI. To access the file in Spark jobs, use :meth:`SparkFiles.get` with the filename to find its download/unpacked location. The given path should be one of .zip, .tar, .tar.gz, .tgz and .jar. .. versionadded:: 3.3.0 Parameters ---------- path : str can be either a local file, a file in HDFS (or other Hadoop-supported filesystems), or an HTTP, HTTPS or FTP URI. To access the file in Spark jobs, use :meth:`SparkFiles.get` to find its download location. See Also -------- :meth:`SparkContext.listArchives` :meth:`SparkFiles.get` Notes ----- A path can be added only once. Subsequent additions of the same path are ignored. This API is experimental. Examples -------- Creates a zipped file that contains a text file written '100'. >>> import os >>> import tempfile >>> import zipfile >>> from pyspark import SparkFiles >>> with tempfile.TemporaryDirectory() as d: ... path = os.path.join(d, "test.txt") ... with open(path, "w") as f: ... _ = f.write("100") ... ... zip_path1 = os.path.join(d, "test1.zip") ... with zipfile.ZipFile(zip_path1, "w", zipfile.ZIP_DEFLATED) as z: ... z.write(path, os.path.basename(path)) ... ... zip_path2 = os.path.join(d, "test2.zip") ... with zipfile.ZipFile(zip_path2, "w", zipfile.ZIP_DEFLATED) as z: ... z.write(path, os.path.basename(path)) ... ... sc.addArchive(zip_path1) ... arch_list1 = sorted(sc.listArchives) ... ... sc.addArchive(zip_path2) ... arch_list2 = sorted(sc.listArchives) ... ... # add zip_path2 twice, this addition will be ignored ... sc.addArchive(zip_path2) ... arch_list3 = sorted(sc.listArchives) ... ... def func(iterator): ... with open("%s/test.txt" % SparkFiles.get("test1.zip")) as f: ... mul = int(f.readline()) ... return [x * mul for x in iterator] ... ... collected = sc.parallelize([1, 2, 3, 4]).mapPartitions(func).collect() >>> arch_list1 ['file:/.../test1.zip'] >>> arch_list2 ['file:/.../test1.zip', 'file:/.../test2.zip'] >>> arch_list3 ['file:/.../test1.zip', 'file:/.../test2.zip'] >>> collected [100, 200, 300, 400] """ self._jsc.sc().addArchive(path)
@property def listArchives(self) -> List[str]: """Returns a list of archive paths that are added to resources. .. versionadded:: 3.4.0 See Also -------- :meth:`SparkContext.addArchive` """ return list( self._jvm.scala.collection.JavaConverters.seqAsJavaList( # type: ignore[union-attr] self._jsc.sc().listArchives() ) )
[docs] def setCheckpointDir(self, dirName: str) -> None: """ Set the directory under which RDDs are going to be checkpointed. The directory must be an HDFS path if running on a cluster. .. versionadded:: 0.7.0 Parameters ---------- dirName : str path to the directory where checkpoint files will be stored (must be HDFS path if running in cluster) See Also -------- :meth:`SparkContext.getCheckpointDir` :meth:`RDD.checkpoint` :meth:`RDD.getCheckpointFile` """ self._jsc.sc().setCheckpointDir(dirName)
[docs] def getCheckpointDir(self) -> Optional[str]: """ Return the directory where RDDs are checkpointed. Returns None if no checkpoint directory has been set. .. versionadded:: 3.1.0 See Also -------- :meth:`SparkContext.setCheckpointDir` :meth:`RDD.checkpoint` :meth:`RDD.getCheckpointFile` """ if not self._jsc.sc().getCheckpointDir().isEmpty(): return self._jsc.sc().getCheckpointDir().get() return None
def _getJavaStorageLevel(self, storageLevel: StorageLevel) -> JavaObject: """ Returns a Java StorageLevel based on a pyspark.StorageLevel. """ if not isinstance(storageLevel, StorageLevel): raise TypeError("storageLevel must be of type pyspark.StorageLevel") assert self._jvm is not None newStorageLevel = self._jvm.org.apache.spark.storage.StorageLevel return newStorageLevel( storageLevel.useDisk, storageLevel.useMemory, storageLevel.useOffHeap, storageLevel.deserialized, storageLevel.replication, )
[docs] def setJobGroup(self, groupId: str, description: str, interruptOnCancel: bool = False) -> None: """ Assigns a group ID to all the jobs started by this thread until the group ID is set to a different value or cleared. Often, a unit of execution in an application consists of multiple Spark actions or jobs. Application programmers can use this method to group all those jobs together and give a group description. Once set, the Spark web UI will associate such jobs with this group. The application can use :meth:`SparkContext.cancelJobGroup` to cancel all running jobs in this group. .. versionadded:: 1.0.0 Parameters ---------- groupId : str The group ID to assign. description : str The description to set for the job group. interruptOnCancel : bool, optional, default False whether to interrupt jobs on job cancellation. Notes ----- If interruptOnCancel is set to true for the job group, then job cancellation will result in Thread.interrupt() being called on the job's executor threads. This is useful to help ensure that the tasks are actually stopped in a timely manner, but is off by default due to HDFS-1208, where HDFS may respond to Thread.interrupt() by marking nodes as dead. If you run jobs in parallel, use :class:`pyspark.InheritableThread` for thread local inheritance. See Also -------- :meth:`SparkContext.cancelJobGroup` Examples -------- >>> import threading >>> from time import sleep >>> from pyspark import InheritableThread >>> result = "Not Set" >>> lock = threading.Lock() >>> def map_func(x): ... sleep(100) ... raise RuntimeError("Task should have been cancelled") ... >>> def start_job(x): ... global result ... try: ... sc.setJobGroup("job_to_cancel", "some description") ... result = sc.parallelize(range(x)).map(map_func).collect() ... except Exception as e: ... result = "Cancelled" ... lock.release() ... >>> def stop_job(): ... sleep(5) ... sc.cancelJobGroup("job_to_cancel") ... >>> suppress = lock.acquire() >>> suppress = InheritableThread(target=start_job, args=(10,)).start() >>> suppress = InheritableThread(target=stop_job).start() >>> suppress = lock.acquire() >>> print(result) Cancelled """ self._jsc.setJobGroup(groupId, description, interruptOnCancel)
[docs] def setInterruptOnCancel(self, interruptOnCancel: bool) -> None: """ Set the behavior of job cancellation from jobs started in this thread. .. versionadded:: 3.5.0 Parameters ---------- interruptOnCancel : bool If true, then job cancellation will result in ``Thread.interrupt()`` being called on the job's executor threads. This is useful to help ensure that the tasks are actually stopped in a timely manner, but is off by default due to HDFS-1208, where HDFS may respond to ``Thread.interrupt()`` by marking nodes as dead. See Also -------- :meth:`SparkContext.addJobTag` :meth:`SparkContext.removeJobTag` :meth:`SparkContext.cancelAllJobs` :meth:`SparkContext.cancelJobGroup` :meth:`SparkContext.cancelJobsWithTag` """ self._jsc.setInterruptOnCancel(interruptOnCancel)
[docs] def addJobTag(self, tag: str) -> None: """ Add a tag to be assigned to all the jobs started by this thread. .. versionadded:: 3.5.0 Parameters ---------- tag : str The tag to be added. Cannot contain ',' (comma) character. See Also -------- :meth:`SparkContext.removeJobTag` :meth:`SparkContext.getJobTags` :meth:`SparkContext.clearJobTags` :meth:`SparkContext.cancelJobsWithTag` :meth:`SparkContext.setInterruptOnCancel` Examples -------- >>> import threading >>> from time import sleep >>> from pyspark import InheritableThread >>> sc.setInterruptOnCancel(interruptOnCancel=True) >>> result = "Not Set" >>> lock = threading.Lock() >>> def map_func(x): ... sleep(100) ... raise RuntimeError("Task should have been cancelled") ... >>> def start_job(x): ... global result ... try: ... sc.addJobTag("job_to_cancel") ... result = sc.parallelize(range(x)).map(map_func).collect() ... except Exception as e: ... result = "Cancelled" ... lock.release() ... >>> def stop_job(): ... sleep(5) ... sc.cancelJobsWithTag("job_to_cancel") ... >>> suppress = lock.acquire() >>> suppress = InheritableThread(target=start_job, args=(10,)).start() >>> suppress = InheritableThread(target=stop_job).start() >>> suppress = lock.acquire() >>> print(result) Cancelled >>> sc.clearJobTags() """ self._jsc.addJobTag(tag)
[docs] def removeJobTag(self, tag: str) -> None: """ Remove a tag previously added to be assigned to all the jobs started by this thread. Noop if such a tag was not added earlier. .. versionadded:: 3.5.0 Parameters ---------- tag : str The tag to be removed. Cannot contain ',' (comma) character. See Also -------- :meth:`SparkContext.addJobTag` :meth:`SparkContext.getJobTags` :meth:`SparkContext.clearJobTags` :meth:`SparkContext.cancelJobsWithTag` :meth:`SparkContext.setInterruptOnCancel` Examples -------- >>> sc.addJobTag("job_to_cancel1") >>> sc.addJobTag("job_to_cancel2") >>> sc.getJobTags() {'job_to_cancel1', 'job_to_cancel2'} >>> sc.removeJobTag("job_to_cancel1") >>> sc.getJobTags() {'job_to_cancel2'} >>> sc.clearJobTags() """ self._jsc.removeJobTag(tag)
[docs] def getJobTags(self) -> Set[str]: """ Get the tags that are currently set to be assigned to all the jobs started by this thread. .. versionadded:: 3.5.0 Returns ------- set of str the tags that are currently set to be assigned to all the jobs started by this thread. See Also -------- :meth:`SparkContext.addJobTag` :meth:`SparkContext.removeJobTag` :meth:`SparkContext.clearJobTags` :meth:`SparkContext.cancelJobsWithTag` :meth:`SparkContext.setInterruptOnCancel` Examples -------- >>> sc.addJobTag("job_to_cancel") >>> sc.getJobTags() {'job_to_cancel'} >>> sc.clearJobTags() """ return self._jsc.getJobTags()
[docs] def clearJobTags(self) -> None: """ Clear the current thread's job tags. .. versionadded:: 3.5.0 See Also -------- :meth:`SparkContext.addJobTag` :meth:`SparkContext.removeJobTag` :meth:`SparkContext.getJobTags` :meth:`SparkContext.cancelJobsWithTag` :meth:`SparkContext.setInterruptOnCancel` Examples -------- >>> sc.addJobTag("job_to_cancel") >>> sc.clearJobTags() >>> sc.getJobTags() set() """ self._jsc.clearJobTags()
[docs] def setLocalProperty(self, key: str, value: str) -> None: """ Set a local property that affects jobs submitted from this thread, such as the Spark fair scheduler pool. .. versionadded:: 1.0.0 Parameters ---------- key : str The key of the local property to set. value : str The value of the local property to set. See Also -------- :meth:`SparkContext.getLocalProperty` Notes ----- If you run jobs in parallel, use :class:`pyspark.InheritableThread` for thread local inheritance. """ self._jsc.setLocalProperty(key, value)
[docs] def getLocalProperty(self, key: str) -> Optional[str]: """ Get a local property set in this thread, or null if it is missing. See :meth:`setLocalProperty`. .. versionadded:: 1.0.0 See Also -------- :meth:`SparkContext.setLocalProperty` """ return self._jsc.getLocalProperty(key)
[docs] def setJobDescription(self, value: str) -> None: """ Set a human readable description of the current job. .. versionadded:: 2.3.0 Parameters ---------- value : str The job description to set. Notes ----- If you run jobs in parallel, use :class:`pyspark.InheritableThread` for thread local inheritance. """ self._jsc.setJobDescription(value)
[docs] def sparkUser(self) -> str: """ Get SPARK_USER for user who is running SparkContext. .. versionadded:: 1.0.0 """ return self._jsc.sc().sparkUser()
[docs] def cancelJobGroup(self, groupId: str) -> None: """ Cancel active jobs for the specified group. See :meth:`SparkContext.setJobGroup`. for more information. .. versionadded:: 1.1.0 Parameters ---------- groupId : str The group ID to cancel the job. See Also -------- :meth:`SparkContext.setJobGroup` """ self._jsc.sc().cancelJobGroup(groupId)
[docs] def cancelJobsWithTag(self, tag: str) -> None: """ Cancel active jobs that have the specified tag. See :meth:`SparkContext.addJobTag`. .. versionadded:: 3.5.0 Parameters ---------- tag : str The tag to be cancelled. Cannot contain ',' (comma) character. See Also -------- :meth:`SparkContext.addJobTag` :meth:`SparkContext.removeJobTag` :meth:`SparkContext.getJobTags` :meth:`SparkContext.clearJobTags` :meth:`SparkContext.setInterruptOnCancel` """ return self._jsc.cancelJobsWithTag(tag)
[docs] def cancelAllJobs(self) -> None: """ Cancel all jobs that have been scheduled or are running. .. versionadded:: 1.1.0 See Also -------- :meth:`SparkContext.cancelJobGroup` :meth:`SparkContext.cancelJobsWithTag` :meth:`SparkContext.runJob` """ self._jsc.sc().cancelAllJobs()
[docs] def statusTracker(self) -> StatusTracker: """ Return :class:`StatusTracker` object .. versionadded:: 1.4.0 """ return StatusTracker(self._jsc.statusTracker())
[docs] def runJob( self, rdd: RDD[T], partitionFunc: Callable[[Iterable[T]], Iterable[U]], partitions: Optional[Sequence[int]] = None, allowLocal: bool = False, ) -> List[U]: """ Executes the given partitionFunc on the specified set of partitions, returning the result as an array of elements. If 'partitions' is not specified, this will run over all partitions. .. versionadded:: 1.1.0 Parameters ---------- rdd : :class:`RDD` target RDD to run tasks on partitionFunc : function a function to run on each partition of the RDD partitions : list, optional set of partitions to run on; some jobs may not want to compute on all partitions of the target RDD, e.g. for operations like `first` allowLocal : bool, default False this parameter takes no effect Returns ------- list results of specified partitions See Also -------- :meth:`SparkContext.cancelAllJobs` Examples -------- >>> myRDD = sc.parallelize(range(6), 3) >>> sc.runJob(myRDD, lambda part: [x * x for x in part]) [0, 1, 4, 9, 16, 25] >>> myRDD = sc.parallelize(range(6), 3) >>> sc.runJob(myRDD, lambda part: [x * x for x in part], [0, 2], True) [0, 1, 16, 25] """ if partitions is None: partitions = list(range(rdd._jrdd.partitions().size())) # Implementation note: This is implemented as a mapPartitions followed # by runJob() in order to avoid having to pass a Python lambda into # SparkContext#runJob. mappedRDD = rdd.mapPartitions(partitionFunc) assert self._jvm is not None sock_info = self._jvm.PythonRDD.runJob(self._jsc.sc(), mappedRDD._jrdd, partitions) return list(_load_from_socket(sock_info, mappedRDD._jrdd_deserializer))
[docs] def show_profiles(self) -> None: """Print the profile stats to stdout .. versionadded:: 1.2.0 See Also -------- :meth:`SparkContext.dump_profiles` """ if self.profiler_collector is not None: self.profiler_collector.show_profiles() else: raise PySparkRuntimeError( error_class="INCORRECT_CONF_FOR_PROFILE", message_parameters={}, )
[docs] def dump_profiles(self, path: str) -> None: """Dump the profile stats into directory `path` .. versionadded:: 1.2.0 See Also -------- :meth:`SparkContext.show_profiles` """ if self.profiler_collector is not None: self.profiler_collector.dump_profiles(path) else: raise PySparkRuntimeError( error_class="INCORRECT_CONF_FOR_PROFILE", message_parameters={}, )
[docs] def getConf(self) -> SparkConf: """Return a copy of this SparkContext's configuration :class:`SparkConf`. .. versionadded:: 2.1.0 """ conf = SparkConf() conf.setAll(self._conf.getAll()) return conf
@property def resources(self) -> Dict[str, ResourceInformation]: """ Return the resource information of this :class:`SparkContext`. A resource could be a GPU, FPGA, etc. .. versionadded:: 3.0.0 """ resources = {} jresources = self._jsc.resources() for x in jresources: name = jresources[x].name() jaddresses = jresources[x].addresses() addrs = [addr for addr in jaddresses] resources[name] = ResourceInformation(name, addrs) return resources @staticmethod def _assert_on_driver() -> None: """ Called to ensure that SparkContext is created only on the Driver. Throws an exception if a SparkContext is about to be created in executors. """ if TaskContext.get() is not None: raise PySparkRuntimeError( error_class="CONTEXT_ONLY_VALID_ON_DRIVER", message_parameters={}, )
def _test() -> None: import doctest from pyspark import SparkConf globs = globals().copy() conf = SparkConf().set("spark.ui.enabled", "True") globs["sc"] = SparkContext("local[4]", "context tests", conf=conf) (failure_count, test_count) = doctest.testmod(globs=globs, optionflags=doctest.ELLIPSIS) globs["sc"].stop() if failure_count: sys.exit(-1) if __name__ == "__main__": _test()