dataframe.py

#
# 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 sys
import random

if sys.version >= '3':
    basestring = unicode = str
    long = int
    from functools import reduce
else:
    from itertools import imap as map

import warnings

from pyspark import copy_func, since, _NoValue
from pyspark.rdd import RDD, _load_from_socket, ignore_unicode_prefix
from pyspark.serializers import ArrowStreamSerializer, BatchedSerializer, PickleSerializer, \
    UTF8Deserializer
from pyspark.storagelevel import StorageLevel
from pyspark.traceback_utils import SCCallSiteSync
from pyspark.sql.types import _parse_datatype_json_string
from pyspark.sql.column import Column, _to_seq, _to_list, _to_java_column
from pyspark.sql.readwriter import DataFrameWriter
from pyspark.sql.streaming import DataStreamWriter
from pyspark.sql.types import IntegralType
from pyspark.sql.types import *
from pyspark.util import _exception_message

__all__ = ["DataFrame", "DataFrameNaFunctions", "DataFrameStatFunctions"]


class DataFrame(object):
    """A distributed collection of data grouped into named columns.

    A :class:`DataFrame` is equivalent to a relational table in Spark SQL,
    and can be created using various functions in :class:`SparkSession`::

        people = spark.read.parquet("...")

    Once created, it can be manipulated using the various domain-specific-language
    (DSL) functions defined in: :class:`DataFrame`, :class:`Column`.

    To select a column from the data frame, use the apply method::

        ageCol = people.age

    A more concrete example::

        # To create DataFrame using SparkSession
        people = spark.read.parquet("...")
        department = spark.read.parquet("...")

        people.filter(people.age > 30).join(department, people.deptId == department.id) \\
          .groupBy(department.name, "gender").agg({"salary": "avg", "age": "max"})

    .. versionadded:: 1.3
    """

    def __init__(self, jdf, sql_ctx):
        self._jdf = jdf
        self.sql_ctx = sql_ctx
        self._sc = sql_ctx and sql_ctx._sc
        self.is_cached = False
        self._schema = None  # initialized lazily
        self._lazy_rdd = None
        # Check whether _repr_html is supported or not, we use it to avoid calling _jdf twice
        # by __repr__ and _repr_html_ while eager evaluation opened.
        self._support_repr_html = False

    @property
    @since(1.3)
    def rdd(self):
        """Returns the content as an :class:`pyspark.RDD` of :class:`Row`.
        """
        if self._lazy_rdd is None:
            jrdd = self._jdf.javaToPython()
            self._lazy_rdd = RDD(jrdd, self.sql_ctx._sc, BatchedSerializer(PickleSerializer()))
        return self._lazy_rdd

    @property
    @since("1.3.1")
    def na(self):
        """Returns a :class:`DataFrameNaFunctions` for handling missing values.
        """
        return DataFrameNaFunctions(self)

    @property
    @since(1.4)
    def stat(self):
        """Returns a :class:`DataFrameStatFunctions` for statistic functions.
        """
        return DataFrameStatFunctions(self)

    @ignore_unicode_prefix
    @since(1.3)
    def toJSON(self, use_unicode=True):
        """Converts a :class:`DataFrame` into a :class:`RDD` of string.

        Each row is turned into a JSON document as one element in the returned RDD.

        >>> df.toJSON().first()
        u'{"age":2,"name":"Alice"}'
        """
        rdd = self._jdf.toJSON()
        return RDD(rdd.toJavaRDD(), self._sc, UTF8Deserializer(use_unicode))

    @since(1.3)
    def registerTempTable(self, name):
        """Registers this DataFrame as a temporary table using the given name.

        The lifetime of this temporary table is tied to the :class:`SparkSession`
        that was used to create this :class:`DataFrame`.

        >>> df.registerTempTable("people")
        >>> df2 = spark.sql("select * from people")
        >>> sorted(df.collect()) == sorted(df2.collect())
        True
        >>> spark.catalog.dropTempView("people")

        .. note:: Deprecated in 2.0, use createOrReplaceTempView instead.
        """
        warnings.warn(
            "Deprecated in 2.0, use createOrReplaceTempView instead.", DeprecationWarning)
        self._jdf.createOrReplaceTempView(name)

    @since(2.0)
    def createTempView(self, name):
        """Creates a local temporary view with this DataFrame.

        The lifetime of this temporary table is tied to the :class:`SparkSession`
        that was used to create this :class:`DataFrame`.
        throws :class:`TempTableAlreadyExistsException`, if the view name already exists in the
        catalog.

        >>> df.createTempView("people")
        >>> df2 = spark.sql("select * from people")
        >>> sorted(df.collect()) == sorted(df2.collect())
        True
        >>> df.createTempView("people")  # doctest: +IGNORE_EXCEPTION_DETAIL
        Traceback (most recent call last):
        ...
        AnalysisException: u"Temporary table 'people' already exists;"
        >>> spark.catalog.dropTempView("people")

        """
        self._jdf.createTempView(name)

    @since(2.0)
    def createOrReplaceTempView(self, name):
        """Creates or replaces a local temporary view with this DataFrame.

        The lifetime of this temporary table is tied to the :class:`SparkSession`
        that was used to create this :class:`DataFrame`.

        >>> df.createOrReplaceTempView("people")
        >>> df2 = df.filter(df.age > 3)
        >>> df2.createOrReplaceTempView("people")
        >>> df3 = spark.sql("select * from people")
        >>> sorted(df3.collect()) == sorted(df2.collect())
        True
        >>> spark.catalog.dropTempView("people")

        """
        self._jdf.createOrReplaceTempView(name)

    @since(2.1)
    def createGlobalTempView(self, name):
        """Creates a global temporary view with this DataFrame.

        The lifetime of this temporary view is tied to this Spark application.
        throws :class:`TempTableAlreadyExistsException`, if the view name already exists in the
        catalog.

        >>> df.createGlobalTempView("people")
        >>> df2 = spark.sql("select * from global_temp.people")
        >>> sorted(df.collect()) == sorted(df2.collect())
        True
        >>> df.createGlobalTempView("people")  # doctest: +IGNORE_EXCEPTION_DETAIL
        Traceback (most recent call last):
        ...
        AnalysisException: u"Temporary table 'people' already exists;"
        >>> spark.catalog.dropGlobalTempView("people")

        """
        self._jdf.createGlobalTempView(name)

    @since(2.2)
    def createOrReplaceGlobalTempView(self, name):
        """Creates or replaces a global temporary view using the given name.

        The lifetime of this temporary view is tied to this Spark application.

        >>> df.createOrReplaceGlobalTempView("people")
        >>> df2 = df.filter(df.age > 3)
        >>> df2.createOrReplaceGlobalTempView("people")
        >>> df3 = spark.sql("select * from global_temp.people")
        >>> sorted(df3.collect()) == sorted(df2.collect())
        True
        >>> spark.catalog.dropGlobalTempView("people")

        """
        self._jdf.createOrReplaceGlobalTempView(name)

    @property
    @since(1.4)
    def write(self):
        """
        Interface for saving the content of the non-streaming :class:`DataFrame` out into external
        storage.

        :return: :class:`DataFrameWriter`
        """
        return DataFrameWriter(self)

    @property
    @since(2.0)
    def writeStream(self):
        """
        Interface for saving the content of the streaming :class:`DataFrame` out into external
        storage.

        .. note:: Evolving.

        :return: :class:`DataStreamWriter`
        """
        return DataStreamWriter(self)

    @property
    @since(1.3)
    def schema(self):
        """Returns the schema of this :class:`DataFrame` as a :class:`pyspark.sql.types.StructType`.

        >>> df.schema
        StructType(List(StructField(age,IntegerType,true),StructField(name,StringType,true)))
        """
        if self._schema is None:
            try:
                self._schema = _parse_datatype_json_string(self._jdf.schema().json())
            except AttributeError as e:
                raise Exception(
                    "Unable to parse datatype from schema. %s" % e)
        return self._schema

    @since(1.3)
    def printSchema(self):
        """Prints out the schema in the tree format.

        >>> df.printSchema()
        root
         |-- age: integer (nullable = true)
         |-- name: string (nullable = true)
        <BLANKLINE>
        """
        print(self._jdf.schema().treeString())

    @since(1.3)
    def explain(self, extended=False):
        """Prints the (logical and physical) plans to the console for debugging purpose.

        :param extended: boolean, default ``False``. If ``False``, prints only the physical plan.

        >>> df.explain()
        == Physical Plan ==
        Scan ExistingRDD[age#0,name#1]

        >>> df.explain(True)
        == Parsed Logical Plan ==
        ...
        == Analyzed Logical Plan ==
        ...
        == Optimized Logical Plan ==
        ...
        == Physical Plan ==
        ...
        """
        if extended:
            print(self._jdf.queryExecution().toString())
        else:
            print(self._jdf.queryExecution().simpleString())

    @since(2.4)
    def exceptAll(self, other):
        """Return a new :class:`DataFrame` containing rows in this :class:`DataFrame` but
        not in another :class:`DataFrame` while preserving duplicates.

        This is equivalent to `EXCEPT ALL` in SQL.

        >>> df1 = spark.createDataFrame(
        ...         [("a", 1), ("a", 1), ("a", 1), ("a", 2), ("b",  3), ("c", 4)], ["C1", "C2"])
        >>> df2 = spark.createDataFrame([("a", 1), ("b", 3)], ["C1", "C2"])

        >>> df1.exceptAll(df2).show()
        +---+---+
        | C1| C2|
        +---+---+
        |  a|  1|
        |  a|  1|
        |  a|  2|
        |  c|  4|
        +---+---+

        Also as standard in SQL, this function resolves columns by position (not by name).
        """
        return DataFrame(self._jdf.exceptAll(other._jdf), self.sql_ctx)

    @since(1.3)
    def isLocal(self):
        """Returns ``True`` if the :func:`collect` and :func:`take` methods can be run locally
        (without any Spark executors).
        """
        return self._jdf.isLocal()

    @property
    @since(2.0)
    def isStreaming(self):
        """Returns true if this :class:`Dataset` contains one or more sources that continuously
        return data as it arrives. A :class:`Dataset` that reads data from a streaming source
        must be executed as a :class:`StreamingQuery` using the :func:`start` method in
        :class:`DataStreamWriter`.  Methods that return a single answer, (e.g., :func:`count` or
        :func:`collect`) will throw an :class:`AnalysisException` when there is a streaming
        source present.

        .. note:: Evolving
        """
        return self._jdf.isStreaming()

    @since(1.3)
    def show(self, n=20, truncate=True, vertical=False):
        """Prints the first ``n`` rows to the console.

        :param n: Number of rows to show.
        :param truncate: If set to True, truncate strings longer than 20 chars by default.
            If set to a number greater than one, truncates long strings to length ``truncate``
            and align cells right.
        :param vertical: If set to True, print output rows vertically (one line
            per column value).

        >>> df
        DataFrame[age: int, name: string]
        >>> df.show()
        +---+-----+
        |age| name|
        +---+-----+
        |  2|Alice|
        |  5|  Bob|
        +---+-----+
        >>> df.show(truncate=3)
        +---+----+
        |age|name|
        +---+----+
        |  2| Ali|
        |  5| Bob|
        +---+----+
        >>> df.show(vertical=True)
        -RECORD 0-----
         age  | 2
         name | Alice
        -RECORD 1-----
         age  | 5
         name | Bob
        """
        if isinstance(truncate, bool) and truncate:
            print(self._jdf.showString(n, 20, vertical))
        else:
            print(self._jdf.showString(n, int(truncate), vertical))

    def __repr__(self):
        if not self._support_repr_html and self.sql_ctx._conf.isReplEagerEvalEnabled():
            vertical = False
            return self._jdf.showString(
                self.sql_ctx._conf.replEagerEvalMaxNumRows(),
                self.sql_ctx._conf.replEagerEvalTruncate(), vertical)
        else:
            return "DataFrame[%s]" % (", ".join("%s: %s" % c for c in self.dtypes))

    def _repr_html_(self):
        """Returns a dataframe with html code when you enabled eager evaluation
        by 'spark.sql.repl.eagerEval.enabled', this only called by REPL you are
        using support eager evaluation with HTML.
        """
        import cgi
        if not self._support_repr_html:
            self._support_repr_html = True
        if self.sql_ctx._conf.isReplEagerEvalEnabled():
            max_num_rows = max(self.sql_ctx._conf.replEagerEvalMaxNumRows(), 0)
            sock_info = self._jdf.getRowsToPython(
                max_num_rows, self.sql_ctx._conf.replEagerEvalTruncate())
            rows = list(_load_from_socket(sock_info, BatchedSerializer(PickleSerializer())))
            head = rows[0]
            row_data = rows[1:]
            has_more_data = len(row_data) > max_num_rows
            row_data = row_data[:max_num_rows]

            html = "<table border='1'>\n"
            # generate table head
            html += "<tr><th>%s</th></tr>\n" % "</th><th>".join(map(lambda x: cgi.escape(x), head))
            # generate table rows
            for row in row_data:
                html += "<tr><td>%s</td></tr>\n" % "</td><td>".join(
                    map(lambda x: cgi.escape(x), row))
            html += "</table>\n"
            if has_more_data:
                html += "only showing top %d %s\n" % (
                    max_num_rows, "row" if max_num_rows == 1 else "rows")
            return html
        else:
            return None

    @since(2.1)
    def checkpoint(self, eager=True):
        """Returns a checkpointed version of this Dataset. Checkpointing can be used to truncate the
        logical plan of this DataFrame, which is especially useful in iterative algorithms where the
        plan may grow exponentially. It will be saved to files inside the checkpoint
        directory set with L{SparkContext.setCheckpointDir()}.

        :param eager: Whether to checkpoint this DataFrame immediately

        .. note:: Experimental
        """
        jdf = self._jdf.checkpoint(eager)
        return DataFrame(jdf, self.sql_ctx)

    @since(2.3)
    def localCheckpoint(self, eager=True):
        """Returns a locally checkpointed version of this Dataset. Checkpointing can be used to
        truncate the logical plan of this DataFrame, which is especially useful in iterative
        algorithms where the plan may grow exponentially. Local checkpoints are stored in the
        executors using the caching subsystem and therefore they are not reliable.

        :param eager: Whether to checkpoint this DataFrame immediately

        .. note:: Experimental
        """
        jdf = self._jdf.localCheckpoint(eager)
        return DataFrame(jdf, self.sql_ctx)

    @since(2.1)
    def withWatermark(self, eventTime, delayThreshold):
        """Defines an event time watermark for this :class:`DataFrame`. A watermark tracks a point
        in time before which we assume no more late data is going to arrive.

        Spark will use this watermark for several purposes:
          - To know when a given time window aggregation can be finalized and thus can be emitted
            when using output modes that do not allow updates.

          - To minimize the amount of state that we need to keep for on-going aggregations.

        The current watermark is computed by looking at the `MAX(eventTime)` seen across
        all of the partitions in the query minus a user specified `delayThreshold`.  Due to the cost
        of coordinating this value across partitions, the actual watermark used is only guaranteed
        to be at least `delayThreshold` behind the actual event time.  In some cases we may still
        process records that arrive more than `delayThreshold` late.

        :param eventTime: the name of the column that contains the event time of the row.
        :param delayThreshold: the minimum delay to wait to data to arrive late, relative to the
            latest record that has been processed in the form of an interval
            (e.g. "1 minute" or "5 hours").

        .. note:: Evolving

        >>> sdf.select('name', sdf.time.cast('timestamp')).withWatermark('time', '10 minutes')
        DataFrame[name: string, time: timestamp]
        """
        if not eventTime or type(eventTime) is not str:
            raise TypeError("eventTime should be provided as a string")
        if not delayThreshold or type(delayThreshold) is not str:
            raise TypeError("delayThreshold should be provided as a string interval")
        jdf = self._jdf.withWatermark(eventTime, delayThreshold)
        return DataFrame(jdf, self.sql_ctx)

    @since(2.2)
    def hint(self, name, *parameters):
        """Specifies some hint on the current DataFrame.

        :param name: A name of the hint.
        :param parameters: Optional parameters.
        :return: :class:`DataFrame`

        >>> df.join(df2.hint("broadcast"), "name").show()
        +----+---+------+
        |name|age|height|
        +----+---+------+
        | Bob|  5|    85|
        +----+---+------+
        """
        if len(parameters) == 1 and isinstance(parameters[0], list):
            parameters = parameters[0]

        if not isinstance(name, str):
            raise TypeError("name should be provided as str, got {0}".format(type(name)))

        for p in parameters:
            if not isinstance(p, str):
                raise TypeError(
                    "all parameters should be str, got {0} of type {1}".format(p, type(p)))

        jdf = self._jdf.hint(name, self._jseq(parameters))
        return DataFrame(jdf, self.sql_ctx)

    @since(1.3)
    def count(self):
        """Returns the number of rows in this :class:`DataFrame`.

        >>> df.count()
        2
        """
        return int(self._jdf.count())

    @ignore_unicode_prefix
    @since(1.3)
    def collect(self):
        """Returns all the records as a list of :class:`Row`.

        >>> df.collect()
        [Row(age=2, name=u'Alice'), Row(age=5, name=u'Bob')]
        """
        with SCCallSiteSync(self._sc) as css:
            sock_info = self._jdf.collectToPython()
        return list(_load_from_socket(sock_info, BatchedSerializer(PickleSerializer())))

    @ignore_unicode_prefix
    @since(2.0)
    def toLocalIterator(self):
        """
        Returns an iterator that contains all of the rows in this :class:`DataFrame`.
        The iterator will consume as much memory as the largest partition in this DataFrame.

        >>> list(df.toLocalIterator())
        [Row(age=2, name=u'Alice'), Row(age=5, name=u'Bob')]
        """
        with SCCallSiteSync(self._sc) as css:
            sock_info = self._jdf.toPythonIterator()
        return _load_from_socket(sock_info, BatchedSerializer(PickleSerializer()))

    @ignore_unicode_prefix
    @since(1.3)
    def limit(self, num):
        """Limits the result count to the number specified.

        >>> df.limit(1).collect()
        [Row(age=2, name=u'Alice')]
        >>> df.limit(0).collect()
        []
        """
        jdf = self._jdf.limit(num)
        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def take(self, num):
        """Returns the first ``num`` rows as a :class:`list` of :class:`Row`.

        >>> df.take(2)
        [Row(age=2, name=u'Alice'), Row(age=5, name=u'Bob')]
        """
        return self.limit(num).collect()

    @since(1.3)
    def foreach(self, f):
        """Applies the ``f`` function to all :class:`Row` of this :class:`DataFrame`.

        This is a shorthand for ``df.rdd.foreach()``.

        >>> def f(person):
        ...     print(person.name)
        >>> df.foreach(f)
        """
        self.rdd.foreach(f)

    @since(1.3)
    def foreachPartition(self, f):
        """Applies the ``f`` function to each partition of this :class:`DataFrame`.

        This a shorthand for ``df.rdd.foreachPartition()``.

        >>> def f(people):
        ...     for person in people:
        ...         print(person.name)
        >>> df.foreachPartition(f)
        """
        self.rdd.foreachPartition(f)

    @since(1.3)
    def cache(self):
        """Persists the :class:`DataFrame` with the default storage level (C{MEMORY_AND_DISK}).

        .. note:: The default storage level has changed to C{MEMORY_AND_DISK} to match Scala in 2.0.
        """
        self.is_cached = True
        self._jdf.cache()
        return self

    @since(1.3)
    def persist(self, storageLevel=StorageLevel.MEMORY_AND_DISK):
        """Sets the storage level to persist the contents of the :class:`DataFrame` across
        operations after the first time it is computed. This can only be used to assign
        a new storage level if the :class:`DataFrame` does not have a storage level set yet.
        If no storage level is specified defaults to (C{MEMORY_AND_DISK}).

        .. note:: The default storage level has changed to C{MEMORY_AND_DISK} to match Scala in 2.0.
        """
        self.is_cached = True
        javaStorageLevel = self._sc._getJavaStorageLevel(storageLevel)
        self._jdf.persist(javaStorageLevel)
        return self

    @property
    @since(2.1)
    def storageLevel(self):
        """Get the :class:`DataFrame`'s current storage level.

        >>> df.storageLevel
        StorageLevel(False, False, False, False, 1)
        >>> df.cache().storageLevel
        StorageLevel(True, True, False, True, 1)
        >>> df2.persist(StorageLevel.DISK_ONLY_2).storageLevel
        StorageLevel(True, False, False, False, 2)
        """
        java_storage_level = self._jdf.storageLevel()
        storage_level = StorageLevel(java_storage_level.useDisk(),
                                     java_storage_level.useMemory(),
                                     java_storage_level.useOffHeap(),
                                     java_storage_level.deserialized(),
                                     java_storage_level.replication())
        return storage_level

    @since(1.3)
    def unpersist(self, blocking=False):
        """Marks the :class:`DataFrame` as non-persistent, and remove all blocks for it from
        memory and disk.

        .. note:: `blocking` default has changed to False to match Scala in 2.0.
        """
        self.is_cached = False
        self._jdf.unpersist(blocking)
        return self

    @since(1.4)
    def coalesce(self, numPartitions):
        """
        Returns a new :class:`DataFrame` that has exactly `numPartitions` partitions.

        :param numPartitions: int, to specify the target number of partitions

        Similar to coalesce defined on an :class:`RDD`, this operation results in a
        narrow dependency, e.g. if you go from 1000 partitions to 100 partitions,
        there will not be a shuffle, instead each of the 100 new partitions will
        claim 10 of the current partitions. If a larger number of partitions is requested,
        it will stay at the current number of partitions.

        However, if you're doing a drastic coalesce, e.g. to numPartitions = 1,
        this may result in your computation taking place on fewer nodes than
        you like (e.g. one node in the case of numPartitions = 1). To avoid this,
        you can call repartition(). This will add a shuffle step, but means the
        current upstream partitions will be executed in parallel (per whatever
        the current partitioning is).

        >>> df.coalesce(1).rdd.getNumPartitions()
        1
        """
        return DataFrame(self._jdf.coalesce(numPartitions), self.sql_ctx)

    @since(1.3)
    def repartition(self, numPartitions, *cols):
        """
        Returns a new :class:`DataFrame` partitioned by the given partitioning expressions. The
        resulting DataFrame is hash partitioned.

        :param numPartitions:
            can be an int to specify the target number of partitions or a Column.
            If it is a Column, it will be used as the first partitioning column. If not specified,
            the default number of partitions is used.

        .. versionchanged:: 1.6
           Added optional arguments to specify the partitioning columns. Also made numPartitions
           optional if partitioning columns are specified.

        >>> df.repartition(10).rdd.getNumPartitions()
        10
        >>> data = df.union(df).repartition("age")
        >>> data.show()
        +---+-----+
        |age| name|
        +---+-----+
        |  5|  Bob|
        |  5|  Bob|
        |  2|Alice|
        |  2|Alice|
        +---+-----+
        >>> data = data.repartition(7, "age")
        >>> data.show()
        +---+-----+
        |age| name|
        +---+-----+
        |  2|Alice|
        |  5|  Bob|
        |  2|Alice|
        |  5|  Bob|
        +---+-----+
        >>> data.rdd.getNumPartitions()
        7
        >>> data = data.repartition("name", "age")
        >>> data.show()
        +---+-----+
        |age| name|
        +---+-----+
        |  5|  Bob|
        |  5|  Bob|
        |  2|Alice|
        |  2|Alice|
        +---+-----+
        """
        if isinstance(numPartitions, int):
            if len(cols) == 0:
                return DataFrame(self._jdf.repartition(numPartitions), self.sql_ctx)
            else:
                return DataFrame(
                    self._jdf.repartition(numPartitions, self._jcols(*cols)), self.sql_ctx)
        elif isinstance(numPartitions, (basestring, Column)):
            cols = (numPartitions, ) + cols
            return DataFrame(self._jdf.repartition(self._jcols(*cols)), self.sql_ctx)
        else:
            raise TypeError("numPartitions should be an int or Column")

    @since("2.4.0")
    def repartitionByRange(self, numPartitions, *cols):
        """
        Returns a new :class:`DataFrame` partitioned by the given partitioning expressions. The
        resulting DataFrame is range partitioned.

        :param numPartitions:
            can be an int to specify the target number of partitions or a Column.
            If it is a Column, it will be used as the first partitioning column. If not specified,
            the default number of partitions is used.

        At least one partition-by expression must be specified.
        When no explicit sort order is specified, "ascending nulls first" is assumed.

        >>> df.repartitionByRange(2, "age").rdd.getNumPartitions()
        2
        >>> df.show()
        +---+-----+
        |age| name|
        +---+-----+
        |  2|Alice|
        |  5|  Bob|
        +---+-----+
        >>> df.repartitionByRange(1, "age").rdd.getNumPartitions()
        1
        >>> data = df.repartitionByRange("age")
        >>> df.show()
        +---+-----+
        |age| name|
        +---+-----+
        |  2|Alice|
        |  5|  Bob|
        +---+-----+
        """
        if isinstance(numPartitions, int):
            if len(cols) == 0:
                return ValueError("At least one partition-by expression must be specified.")
            else:
                return DataFrame(
                    self._jdf.repartitionByRange(numPartitions, self._jcols(*cols)), self.sql_ctx)
        elif isinstance(numPartitions, (basestring, Column)):
            cols = (numPartitions,) + cols
            return DataFrame(self._jdf.repartitionByRange(self._jcols(*cols)), self.sql_ctx)
        else:
            raise TypeError("numPartitions should be an int, string or Column")

    @since(1.3)
    def distinct(self):
        """Returns a new :class:`DataFrame` containing the distinct rows in this :class:`DataFrame`.

        >>> df.distinct().count()
        2
        """
        return DataFrame(self._jdf.distinct(), self.sql_ctx)

    @since(1.3)
    def sample(self, withReplacement=None, fraction=None, seed=None):
        """Returns a sampled subset of this :class:`DataFrame`.

        :param withReplacement: Sample with replacement or not (default False).
        :param fraction: Fraction of rows to generate, range [0.0, 1.0].
        :param seed: Seed for sampling (default a random seed).

        .. note:: This is not guaranteed to provide exactly the fraction specified of the total
            count of the given :class:`DataFrame`.

        .. note:: `fraction` is required and, `withReplacement` and `seed` are optional.

        >>> df = spark.range(10)
        >>> df.sample(0.5, 3).count()
        4
        >>> df.sample(fraction=0.5, seed=3).count()
        4
        >>> df.sample(withReplacement=True, fraction=0.5, seed=3).count()
        1
        >>> df.sample(1.0).count()
        10
        >>> df.sample(fraction=1.0).count()
        10
        >>> df.sample(False, fraction=1.0).count()
        10
        """

        # For the cases below:
        #   sample(True, 0.5 [, seed])
        #   sample(True, fraction=0.5 [, seed])
        #   sample(withReplacement=False, fraction=0.5 [, seed])
        is_withReplacement_set = \
            type(withReplacement) == bool and isinstance(fraction, float)

        # For the case below:
        #   sample(faction=0.5 [, seed])
        is_withReplacement_omitted_kwargs = \
            withReplacement is None and isinstance(fraction, float)

        # For the case below:
        #   sample(0.5 [, seed])
        is_withReplacement_omitted_args = isinstance(withReplacement, float)

        if not (is_withReplacement_set
                or is_withReplacement_omitted_kwargs
                or is_withReplacement_omitted_args):
            argtypes = [
                str(type(arg)) for arg in [withReplacement, fraction, seed] if arg is not None]
            raise TypeError(
                "withReplacement (optional), fraction (required) and seed (optional)"
                " should be a bool, float and number; however, "
                "got [%s]." % ", ".join(argtypes))

        if is_withReplacement_omitted_args:
            if fraction is not None:
                seed = fraction
            fraction = withReplacement
            withReplacement = None

        seed = long(seed) if seed is not None else None
        args = [arg for arg in [withReplacement, fraction, seed] if arg is not None]
        jdf = self._jdf.sample(*args)
        return DataFrame(jdf, self.sql_ctx)

    @since(1.5)
    def sampleBy(self, col, fractions, seed=None):
        """
        Returns a stratified sample without replacement based on the
        fraction given on each stratum.

        :param col: column that defines strata
        :param fractions:
            sampling fraction for each stratum. If a stratum is not
            specified, we treat its fraction as zero.
        :param seed: random seed
        :return: a new DataFrame that represents the stratified sample

        >>> from pyspark.sql.functions import col
        >>> dataset = sqlContext.range(0, 100).select((col("id") % 3).alias("key"))
        >>> sampled = dataset.sampleBy("key", fractions={0: 0.1, 1: 0.2}, seed=0)
        >>> sampled.groupBy("key").count().orderBy("key").show()
        +---+-----+
        |key|count|
        +---+-----+
        |  0|    5|
        |  1|    9|
        +---+-----+

        """
        if not isinstance(col, basestring):
            raise ValueError("col must be a string, but got %r" % type(col))
        if not isinstance(fractions, dict):
            raise ValueError("fractions must be a dict but got %r" % type(fractions))
        for k, v in fractions.items():
            if not isinstance(k, (float, int, long, basestring)):
                raise ValueError("key must be float, int, long, or string, but got %r" % type(k))
            fractions[k] = float(v)
        seed = seed if seed is not None else random.randint(0, sys.maxsize)
        return DataFrame(self._jdf.stat().sampleBy(col, self._jmap(fractions), seed), self.sql_ctx)

    @since(1.4)
    def randomSplit(self, weights, seed=None):
        """Randomly splits this :class:`DataFrame` with the provided weights.

        :param weights: list of doubles as weights with which to split the DataFrame. Weights will
            be normalized if they don't sum up to 1.0.
        :param seed: The seed for sampling.

        >>> splits = df4.randomSplit([1.0, 2.0], 24)
        >>> splits[0].count()
        1

        >>> splits[1].count()
        3
        """
        for w in weights:
            if w < 0.0:
                raise ValueError("Weights must be positive. Found weight value: %s" % w)
        seed = seed if seed is not None else random.randint(0, sys.maxsize)
        rdd_array = self._jdf.randomSplit(_to_list(self.sql_ctx._sc, weights), long(seed))
        return [DataFrame(rdd, self.sql_ctx) for rdd in rdd_array]

    @property
    @since(1.3)
    def dtypes(self):
        """Returns all column names and their data types as a list.

        >>> df.dtypes
        [('age', 'int'), ('name', 'string')]
        """
        return [(str(f.name), f.dataType.simpleString()) for f in self.schema.fields]

    @property
    @since(1.3)
    def columns(self):
        """Returns all column names as a list.

        >>> df.columns
        ['age', 'name']
        """
        return [f.name for f in self.schema.fields]

    @since(2.3)
    def colRegex(self, colName):
        """
        Selects column based on the column name specified as a regex and returns it
        as :class:`Column`.

        :param colName: string, column name specified as a regex.

        >>> df = spark.createDataFrame([("a", 1), ("b", 2), ("c",  3)], ["Col1", "Col2"])
        >>> df.select(df.colRegex("`(Col1)?+.+`")).show()
        +----+
        |Col2|
        +----+
        |   1|
        |   2|
        |   3|
        +----+
        """
        if not isinstance(colName, basestring):
            raise ValueError("colName should be provided as string")
        jc = self._jdf.colRegex(colName)
        return Column(jc)

    @ignore_unicode_prefix
    @since(1.3)
    def alias(self, alias):
        """Returns a new :class:`DataFrame` with an alias set.

        :param alias: string, an alias name to be set for the DataFrame.

        >>> from pyspark.sql.functions import *
        >>> df_as1 = df.alias("df_as1")
        >>> df_as2 = df.alias("df_as2")
        >>> joined_df = df_as1.join(df_as2, col("df_as1.name") == col("df_as2.name"), 'inner')
        >>> joined_df.select("df_as1.name", "df_as2.name", "df_as2.age").collect()
        [Row(name=u'Bob', name=u'Bob', age=5), Row(name=u'Alice', name=u'Alice', age=2)]
        """
        assert isinstance(alias, basestring), "alias should be a string"
        return DataFrame(getattr(self._jdf, "as")(alias), self.sql_ctx)

    @ignore_unicode_prefix
    @since(2.1)
    def crossJoin(self, other):
        """Returns the cartesian product with another :class:`DataFrame`.

        :param other: Right side of the cartesian product.

        >>> df.select("age", "name").collect()
        [Row(age=2, name=u'Alice'), Row(age=5, name=u'Bob')]
        >>> df2.select("name", "height").collect()
        [Row(name=u'Tom', height=80), Row(name=u'Bob', height=85)]
        >>> df.crossJoin(df2.select("height")).select("age", "name", "height").collect()
        [Row(age=2, name=u'Alice', height=80), Row(age=2, name=u'Alice', height=85),
         Row(age=5, name=u'Bob', height=80), Row(age=5, name=u'Bob', height=85)]
        """

        jdf = self._jdf.crossJoin(other._jdf)
        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def join(self, other, on=None, how=None):
        """Joins with another :class:`DataFrame`, using the given join expression.

        :param other: Right side of the join
        :param on: a string for the join column name, a list of column names,
            a join expression (Column), or a list of Columns.
            If `on` is a string or a list of strings indicating the name of the join column(s),
            the column(s) must exist on both sides, and this performs an equi-join.
        :param how: str, default ``inner``. Must be one of: ``inner``, ``cross``, ``outer``,
            ``full``, ``full_outer``, ``left``, ``left_outer``, ``right``, ``right_outer``,
            ``left_semi``, and ``left_anti``.

        The following performs a full outer join between ``df1`` and ``df2``.

        >>> df.join(df2, df.name == df2.name, 'outer').select(df.name, df2.height).collect()
        [Row(name=None, height=80), Row(name=u'Bob', height=85), Row(name=u'Alice', height=None)]

        >>> df.join(df2, 'name', 'outer').select('name', 'height').collect()
        [Row(name=u'Tom', height=80), Row(name=u'Bob', height=85), Row(name=u'Alice', height=None)]

        >>> cond = [df.name == df3.name, df.age == df3.age]
        >>> df.join(df3, cond, 'outer').select(df.name, df3.age).collect()
        [Row(name=u'Alice', age=2), Row(name=u'Bob', age=5)]

        >>> df.join(df2, 'name').select(df.name, df2.height).collect()
        [Row(name=u'Bob', height=85)]

        >>> df.join(df4, ['name', 'age']).select(df.name, df.age).collect()
        [Row(name=u'Bob', age=5)]
        """

        if on is not None and not isinstance(on, list):
            on = [on]

        if on is not None:
            if isinstance(on[0], basestring):
                on = self._jseq(on)
            else:
                assert isinstance(on[0], Column), "on should be Column or list of Column"
                on = reduce(lambda x, y: x.__and__(y), on)
                on = on._jc

        if on is None and how is None:
            jdf = self._jdf.join(other._jdf)
        else:
            if how is None:
                how = "inner"
            if on is None:
                on = self._jseq([])
            assert isinstance(how, basestring), "how should be basestring"
            jdf = self._jdf.join(other._jdf, on, how)
        return DataFrame(jdf, self.sql_ctx)

    @since(1.6)
    def sortWithinPartitions(self, *cols, **kwargs):
        """Returns a new :class:`DataFrame` with each partition sorted by the specified column(s).

        :param cols: list of :class:`Column` or column names to sort by.
        :param ascending: boolean or list of boolean (default True).
            Sort ascending vs. descending. Specify list for multiple sort orders.
            If a list is specified, length of the list must equal length of the `cols`.

        >>> df.sortWithinPartitions("age", ascending=False).show()
        +---+-----+
        |age| name|
        +---+-----+
        |  2|Alice|
        |  5|  Bob|
        +---+-----+
        """
        jdf = self._jdf.sortWithinPartitions(self._sort_cols(cols, kwargs))
        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def sort(self, *cols, **kwargs):
        """Returns a new :class:`DataFrame` sorted by the specified column(s).

        :param cols: list of :class:`Column` or column names to sort by.
        :param ascending: boolean or list of boolean (default True).
            Sort ascending vs. descending. Specify list for multiple sort orders.
            If a list is specified, length of the list must equal length of the `cols`.

        >>> df.sort(df.age.desc()).collect()
        [Row(age=5, name=u'Bob'), Row(age=2, name=u'Alice')]
        >>> df.sort("age", ascending=False).collect()
        [Row(age=5, name=u'Bob'), Row(age=2, name=u'Alice')]
        >>> df.orderBy(df.age.desc()).collect()
        [Row(age=5, name=u'Bob'), Row(age=2, name=u'Alice')]
        >>> from pyspark.sql.functions import *
        >>> df.sort(asc("age")).collect()
        [Row(age=2, name=u'Alice'), Row(age=5, name=u'Bob')]
        >>> df.orderBy(desc("age"), "name").collect()
        [Row(age=5, name=u'Bob'), Row(age=2, name=u'Alice')]
        >>> df.orderBy(["age", "name"], ascending=[0, 1]).collect()
        [Row(age=5, name=u'Bob'), Row(age=2, name=u'Alice')]
        """
        jdf = self._jdf.sort(self._sort_cols(cols, kwargs))
        return DataFrame(jdf, self.sql_ctx)

    orderBy = sort

    def _jseq(self, cols, converter=None):
        """Return a JVM Seq of Columns from a list of Column or names"""
        return _to_seq(self.sql_ctx._sc, cols, converter)

    def _jmap(self, jm):
        """Return a JVM Scala Map from a dict"""
        return _to_scala_map(self.sql_ctx._sc, jm)

    def _jcols(self, *cols):
        """Return a JVM Seq of Columns from a list of Column or column names

        If `cols` has only one list in it, cols[0] will be used as the list.
        """
        if len(cols) == 1 and isinstance(cols[0], list):
            cols = cols[0]
        return self._jseq(cols, _to_java_column)

    def _sort_cols(self, cols, kwargs):
        """ Return a JVM Seq of Columns that describes the sort order
        """
        if not cols:
            raise ValueError("should sort by at least one column")
        if len(cols) == 1 and isinstance(cols[0], list):
            cols = cols[0]
        jcols = [_to_java_column(c) for c in cols]
        ascending = kwargs.get('ascending', True)
        if isinstance(ascending, (bool, int)):
            if not ascending:
                jcols = [jc.desc() for jc in jcols]
        elif isinstance(ascending, list):
            jcols = [jc if asc else jc.desc()
                     for asc, jc in zip(ascending, jcols)]
        else:
            raise TypeError("ascending can only be boolean or list, but got %s" % type(ascending))
        return self._jseq(jcols)

    @since("1.3.1")
    def describe(self, *cols):
        """Computes basic statistics for numeric and string columns.

        This include count, mean, stddev, min, and max. If no columns are
        given, this function computes statistics for all numerical or string columns.

        .. note:: This function is meant for exploratory data analysis, as we make no
            guarantee about the backward compatibility of the schema of the resulting DataFrame.

        >>> df.describe(['age']).show()
        +-------+------------------+
        |summary|               age|
        +-------+------------------+
        |  count|                 2|
        |   mean|               3.5|
        | stddev|2.1213203435596424|
        |    min|                 2|
        |    max|                 5|
        +-------+------------------+
        >>> df.describe().show()
        +-------+------------------+-----+
        |summary|               age| name|
        +-------+------------------+-----+
        |  count|                 2|    2|
        |   mean|               3.5| null|
        | stddev|2.1213203435596424| null|
        |    min|                 2|Alice|
        |    max|                 5|  Bob|
        +-------+------------------+-----+

        Use summary for expanded statistics and control over which statistics to compute.
        """
        if len(cols) == 1 and isinstance(cols[0], list):
            cols = cols[0]
        jdf = self._jdf.describe(self._jseq(cols))
        return DataFrame(jdf, self.sql_ctx)

    @since("2.3.0")
    def summary(self, *statistics):
        """Computes specified statistics for numeric and string columns. Available statistics are:
        - count
        - mean
        - stddev
        - min
        - max
        - arbitrary approximate percentiles specified as a percentage (eg, 75%)

        If no statistics are given, this function computes count, mean, stddev, min,
        approximate quartiles (percentiles at 25%, 50%, and 75%), and max.

        .. note:: This function is meant for exploratory data analysis, as we make no
            guarantee about the backward compatibility of the schema of the resulting DataFrame.

        >>> df.summary().show()
        +-------+------------------+-----+
        |summary|               age| name|
        +-------+------------------+-----+
        |  count|                 2|    2|
        |   mean|               3.5| null|
        | stddev|2.1213203435596424| null|
        |    min|                 2|Alice|
        |    25%|                 2| null|
        |    50%|                 2| null|
        |    75%|                 5| null|
        |    max|                 5|  Bob|
        +-------+------------------+-----+

        >>> df.summary("count", "min", "25%", "75%", "max").show()
        +-------+---+-----+
        |summary|age| name|
        +-------+---+-----+
        |  count|  2|    2|
        |    min|  2|Alice|
        |    25%|  2| null|
        |    75%|  5| null|
        |    max|  5|  Bob|
        +-------+---+-----+

        To do a summary for specific columns first select them:

        >>> df.select("age", "name").summary("count").show()
        +-------+---+----+
        |summary|age|name|
        +-------+---+----+
        |  count|  2|   2|
        +-------+---+----+

        See also describe for basic statistics.
        """
        if len(statistics) == 1 and isinstance(statistics[0], list):
            statistics = statistics[0]
        jdf = self._jdf.summary(self._jseq(statistics))
        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def head(self, n=None):
        """Returns the first ``n`` rows.

        .. note:: This method should only be used if the resulting array is expected
            to be small, as all the data is loaded into the driver's memory.

        :param n: int, default 1. Number of rows to return.
        :return: If n is greater than 1, return a list of :class:`Row`.
            If n is 1, return a single Row.

        >>> df.head()
        Row(age=2, name=u'Alice')
        >>> df.head(1)
        [Row(age=2, name=u'Alice')]
        """
        if n is None:
            rs = self.head(1)
            return rs[0] if rs else None
        return self.take(n)

    @ignore_unicode_prefix
    @since(1.3)
    def first(self):
        """Returns the first row as a :class:`Row`.

        >>> df.first()
        Row(age=2, name=u'Alice')
        """
        return self.head()

    @ignore_unicode_prefix
    @since(1.3)
    def __getitem__(self, item):
        """Returns the column as a :class:`Column`.

        >>> df.select(df['age']).collect()
        [Row(age=2), Row(age=5)]
        >>> df[ ["name", "age"]].collect()
        [Row(name=u'Alice', age=2), Row(name=u'Bob', age=5)]
        >>> df[ df.age > 3 ].collect()
        [Row(age=5, name=u'Bob')]
        >>> df[df[0] > 3].collect()
        [Row(age=5, name=u'Bob')]
        """
        if isinstance(item, basestring):
            jc = self._jdf.apply(item)
            return Column(jc)
        elif isinstance(item, Column):
            return self.filter(item)
        elif isinstance(item, (list, tuple)):
            return self.select(*item)
        elif isinstance(item, int):
            jc = self._jdf.apply(self.columns[item])
            return Column(jc)
        else:
            raise TypeError("unexpected item type: %s" % type(item))

    @since(1.3)
    def __getattr__(self, name):
        """Returns the :class:`Column` denoted by ``name``.

        >>> df.select(df.age).collect()
        [Row(age=2), Row(age=5)]
        """
        if name not in self.columns:
            raise AttributeError(
                "'%s' object has no attribute '%s'" % (self.__class__.__name__, name))
        jc = self._jdf.apply(name)
        return Column(jc)

    @ignore_unicode_prefix
    @since(1.3)
    def select(self, *cols):
        """Projects a set of expressions and returns a new :class:`DataFrame`.

        :param cols: list of column names (string) or expressions (:class:`Column`).
            If one of the column names is '*', that column is expanded to include all columns
            in the current DataFrame.

        >>> df.select('*').collect()
        [Row(age=2, name=u'Alice'), Row(age=5, name=u'Bob')]
        >>> df.select('name', 'age').collect()
        [Row(name=u'Alice', age=2), Row(name=u'Bob', age=5)]
        >>> df.select(df.name, (df.age + 10).alias('age')).collect()
        [Row(name=u'Alice', age=12), Row(name=u'Bob', age=15)]
        """
        jdf = self._jdf.select(self._jcols(*cols))
        return DataFrame(jdf, self.sql_ctx)

    @since(1.3)
    def selectExpr(self, *expr):
        """Projects a set of SQL expressions and returns a new :class:`DataFrame`.

        This is a variant of :func:`select` that accepts SQL expressions.

        >>> df.selectExpr("age * 2", "abs(age)").collect()
        [Row((age * 2)=4, abs(age)=2), Row((age * 2)=10, abs(age)=5)]
        """
        if len(expr) == 1 and isinstance(expr[0], list):
            expr = expr[0]
        jdf = self._jdf.selectExpr(self._jseq(expr))
        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def filter(self, condition):
        """Filters rows using the given condition.

        :func:`where` is an alias for :func:`filter`.

        :param condition: a :class:`Column` of :class:`types.BooleanType`
            or a string of SQL expression.

        >>> df.filter(df.age > 3).collect()
        [Row(age=5, name=u'Bob')]
        >>> df.where(df.age == 2).collect()
        [Row(age=2, name=u'Alice')]

        >>> df.filter("age > 3").collect()
        [Row(age=5, name=u'Bob')]
        >>> df.where("age = 2").collect()
        [Row(age=2, name=u'Alice')]
        """
        if isinstance(condition, basestring):
            jdf = self._jdf.filter(condition)
        elif isinstance(condition, Column):
            jdf = self._jdf.filter(condition._jc)
        else:
            raise TypeError("condition should be string or Column")
        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def groupBy(self, *cols):
        """Groups the :class:`DataFrame` using the specified columns,
        so we can run aggregation on them. See :class:`GroupedData`
        for all the available aggregate functions.

        :func:`groupby` is an alias for :func:`groupBy`.

        :param cols: list of columns to group by.
            Each element should be a column name (string) or an expression (:class:`Column`).

        >>> df.groupBy().avg().collect()
        [Row(avg(age)=3.5)]
        >>> sorted(df.groupBy('name').agg({'age': 'mean'}).collect())
        [Row(name=u'Alice', avg(age)=2.0), Row(name=u'Bob', avg(age)=5.0)]
        >>> sorted(df.groupBy(df.name).avg().collect())
        [Row(name=u'Alice', avg(age)=2.0), Row(name=u'Bob', avg(age)=5.0)]
        >>> sorted(df.groupBy(['name', df.age]).count().collect())
        [Row(name=u'Alice', age=2, count=1), Row(name=u'Bob', age=5, count=1)]
        """
        jgd = self._jdf.groupBy(self._jcols(*cols))
        from pyspark.sql.group import GroupedData
        return GroupedData(jgd, self)

    @since(1.4)
    def rollup(self, *cols):
        """
        Create a multi-dimensional rollup for the current :class:`DataFrame` using
        the specified columns, so we can run aggregation on them.

        >>> df.rollup("name", df.age).count().orderBy("name", "age").show()
        +-----+----+-----+
        | name| age|count|
        +-----+----+-----+
        | null|null|    2|
        |Alice|null|    1|
        |Alice|   2|    1|
        |  Bob|null|    1|
        |  Bob|   5|    1|
        +-----+----+-----+
        """
        jgd = self._jdf.rollup(self._jcols(*cols))
        from pyspark.sql.group import GroupedData
        return GroupedData(jgd, self)

    @since(1.4)
    def cube(self, *cols):
        """
        Create a multi-dimensional cube for the current :class:`DataFrame` using
        the specified columns, so we can run aggregation on them.

        >>> df.cube("name", df.age).count().orderBy("name", "age").show()
        +-----+----+-----+
        | name| age|count|
        +-----+----+-----+
        | null|null|    2|
        | null|   2|    1|
        | null|   5|    1|
        |Alice|null|    1|
        |Alice|   2|    1|
        |  Bob|null|    1|
        |  Bob|   5|    1|
        +-----+----+-----+
        """
        jgd = self._jdf.cube(self._jcols(*cols))
        from pyspark.sql.group import GroupedData
        return GroupedData(jgd, self)

    @since(1.3)
    def agg(self, *exprs):
        """ Aggregate on the entire :class:`DataFrame` without groups
        (shorthand for ``df.groupBy.agg()``).

        >>> df.agg({"age": "max"}).collect()
        [Row(max(age)=5)]
        >>> from pyspark.sql import functions as F
        >>> df.agg(F.min(df.age)).collect()
        [Row(min(age)=2)]
        """
        return self.groupBy().agg(*exprs)

    @since(2.0)
    def union(self, other):
        """ Return a new :class:`DataFrame` containing union of rows in this and another frame.

        This is equivalent to `UNION ALL` in SQL. To do a SQL-style set union
        (that does deduplication of elements), use this function followed by :func:`distinct`.

        Also as standard in SQL, this function resolves columns by position (not by name).
        """
        return DataFrame(self._jdf.union(other._jdf), self.sql_ctx)

    @since(1.3)
    def unionAll(self, other):
        """ Return a new :class:`DataFrame` containing union of rows in this and another frame.

        This is equivalent to `UNION ALL` in SQL. To do a SQL-style set union
        (that does deduplication of elements), use this function followed by :func:`distinct`.

        Also as standard in SQL, this function resolves columns by position (not by name).

        .. note:: Deprecated in 2.0, use :func:`union` instead.
        """
        warnings.warn("Deprecated in 2.0, use union instead.", DeprecationWarning)
        return self.union(other)

    @since(2.3)
    def unionByName(self, other):
        """ Returns a new :class:`DataFrame` containing union of rows in this and another frame.

        This is different from both `UNION ALL` and `UNION DISTINCT` in SQL. To do a SQL-style set
        union (that does deduplication of elements), use this function followed by :func:`distinct`.

        The difference between this function and :func:`union` is that this function
        resolves columns by name (not by position):

        >>> df1 = spark.createDataFrame([[1, 2, 3]], ["col0", "col1", "col2"])
        >>> df2 = spark.createDataFrame([[4, 5, 6]], ["col1", "col2", "col0"])
        >>> df1.unionByName(df2).show()
        +----+----+----+
        |col0|col1|col2|
        +----+----+----+
        |   1|   2|   3|
        |   6|   4|   5|
        +----+----+----+
        """
        return DataFrame(self._jdf.unionByName(other._jdf), self.sql_ctx)

    @since(1.3)
    def intersect(self, other):
        """ Return a new :class:`DataFrame` containing rows only in
        both this frame and another frame.

        This is equivalent to `INTERSECT` in SQL.
        """
        return DataFrame(self._jdf.intersect(other._jdf), self.sql_ctx)

    @since(2.4)
    def intersectAll(self, other):
        """ Return a new :class:`DataFrame` containing rows in both this dataframe and other
        dataframe while preserving duplicates.

        This is equivalent to `INTERSECT ALL` in SQL.
        >>> df1 = spark.createDataFrame([("a", 1), ("a", 1), ("b", 3), ("c", 4)], ["C1", "C2"])
        >>> df2 = spark.createDataFrame([("a", 1), ("a", 1), ("b", 3)], ["C1", "C2"])

        >>> df1.intersectAll(df2).sort("C1", "C2").show()
        +---+---+
        | C1| C2|
        +---+---+
        |  a|  1|
        |  a|  1|
        |  b|  3|
        +---+---+

        Also as standard in SQL, this function resolves columns by position (not by name).
        """
        return DataFrame(self._jdf.intersectAll(other._jdf), self.sql_ctx)

    @since(1.3)
    def subtract(self, other):
        """ Return a new :class:`DataFrame` containing rows in this frame
        but not in another frame.

        This is equivalent to `EXCEPT DISTINCT` in SQL.

        """
        return DataFrame(getattr(self._jdf, "except")(other._jdf), self.sql_ctx)

    @since(1.4)
    def dropDuplicates(self, subset=None):
        """Return a new :class:`DataFrame` with duplicate rows removed,
        optionally only considering certain columns.

        For a static batch :class:`DataFrame`, it just drops duplicate rows. For a streaming
        :class:`DataFrame`, it will keep all data across triggers as intermediate state to drop
        duplicates rows. You can use :func:`withWatermark` to limit how late the duplicate data can
        be and system will accordingly limit the state. In addition, too late data older than
        watermark will be dropped to avoid any possibility of duplicates.

        :func:`drop_duplicates` is an alias for :func:`dropDuplicates`.

        >>> from pyspark.sql import Row
        >>> df = sc.parallelize([ \\
        ...     Row(name='Alice', age=5, height=80), \\
        ...     Row(name='Alice', age=5, height=80), \\
        ...     Row(name='Alice', age=10, height=80)]).toDF()
        >>> df.dropDuplicates().show()
        +---+------+-----+
        |age|height| name|
        +---+------+-----+
        |  5|    80|Alice|
        | 10|    80|Alice|
        +---+------+-----+

        >>> df.dropDuplicates(['name', 'height']).show()
        +---+------+-----+
        |age|height| name|
        +---+------+-----+
        |  5|    80|Alice|
        +---+------+-----+
        """
        if subset is None:
            jdf = self._jdf.dropDuplicates()
        else:
            jdf = self._jdf.dropDuplicates(self._jseq(subset))
        return DataFrame(jdf, self.sql_ctx)

    @since("1.3.1")
    def dropna(self, how='any', thresh=None, subset=None):
        """Returns a new :class:`DataFrame` omitting rows with null values.
        :func:`DataFrame.dropna` and :func:`DataFrameNaFunctions.drop` are aliases of each other.

        :param how: 'any' or 'all'.
            If 'any', drop a row if it contains any nulls.
            If 'all', drop a row only if all its values are null.
        :param thresh: int, default None
            If specified, drop rows that have less than `thresh` non-null values.
            This overwrites the `how` parameter.
        :param subset: optional list of column names to consider.

        >>> df4.na.drop().show()
        +---+------+-----+
        |age|height| name|
        +---+------+-----+
        | 10|    80|Alice|
        +---+------+-----+
        """
        if how is not None and how not in ['any', 'all']:
            raise ValueError("how ('" + how + "') should be 'any' or 'all'")

        if subset is None:
            subset = self.columns
        elif isinstance(subset, basestring):
            subset = [subset]
        elif not isinstance(subset, (list, tuple)):
            raise ValueError("subset should be a list or tuple of column names")

        if thresh is None:
            thresh = len(subset) if how == 'any' else 1

        return DataFrame(self._jdf.na().drop(thresh, self._jseq(subset)), self.sql_ctx)

    @since("1.3.1")
    def fillna(self, value, subset=None):
        """Replace null values, alias for ``na.fill()``.
        :func:`DataFrame.fillna` and :func:`DataFrameNaFunctions.fill` are aliases of each other.

        :param value: int, long, float, string, bool or dict.
            Value to replace null values with.
            If the value is a dict, then `subset` is ignored and `value` must be a mapping
            from column name (string) to replacement value. The replacement value must be
            an int, long, float, boolean, or string.
        :param subset: optional list of column names to consider.
            Columns specified in subset that do not have matching data type are ignored.
            For example, if `value` is a string, and subset contains a non-string column,
            then the non-string column is simply ignored.

        >>> df4.na.fill(50).show()
        +---+------+-----+
        |age|height| name|
        +---+------+-----+
        | 10|    80|Alice|
        |  5|    50|  Bob|
        | 50|    50|  Tom|
        | 50|    50| null|
        +---+------+-----+

        >>> df5.na.fill(False).show()
        +----+-------+-----+
        | age|   name|  spy|
        +----+-------+-----+
        |  10|  Alice|false|
        |   5|    Bob|false|
        |null|Mallory| true|
        +----+-------+-----+

        >>> df4.na.fill({'age': 50, 'name': 'unknown'}).show()
        +---+------+-------+
        |age|height|   name|
        +---+------+-------+
        | 10|    80|  Alice|
        |  5|  null|    Bob|
        | 50|  null|    Tom|
        | 50|  null|unknown|
        +---+------+-------+
        """
        if not isinstance(value, (float, int, long, basestring, bool, dict)):
            raise ValueError("value should be a float, int, long, string, bool or dict")

        # Note that bool validates isinstance(int), but we don't want to
        # convert bools to floats

        if not isinstance(value, bool) and isinstance(value, (int, long)):
            value = float(value)

        if isinstance(value, dict):
            return DataFrame(self._jdf.na().fill(value), self.sql_ctx)
        elif subset is None:
            return DataFrame(self._jdf.na().fill(value), self.sql_ctx)
        else:
            if isinstance(subset, basestring):
                subset = [subset]
            elif not isinstance(subset, (list, tuple)):
                raise ValueError("subset should be a list or tuple of column names")

            return DataFrame(self._jdf.na().fill(value, self._jseq(subset)), self.sql_ctx)

    @since(1.4)
    def replace(self, to_replace, value=_NoValue, subset=None):
        """Returns a new :class:`DataFrame` replacing a value with another value.
        :func:`DataFrame.replace` and :func:`DataFrameNaFunctions.replace` are
        aliases of each other.
        Values to_replace and value must have the same type and can only be numerics, booleans,
        or strings. Value can have None. When replacing, the new value will be cast
        to the type of the existing column.
        For numeric replacements all values to be replaced should have unique
        floating point representation. In case of conflicts (for example with `{42: -1, 42.0: 1}`)
        and arbitrary replacement will be used.

        :param to_replace: bool, int, long, float, string, list or dict.
            Value to be replaced.
            If the value is a dict, then `value` is ignored or can be omitted, and `to_replace`
            must be a mapping between a value and a replacement.
        :param value: bool, int, long, float, string, list or None.
            The replacement value must be a bool, int, long, float, string or None. If `value` is a
            list, `value` should be of the same length and type as `to_replace`.
            If `value` is a scalar and `to_replace` is a sequence, then `value` is
            used as a replacement for each item in `to_replace`.
        :param subset: optional list of column names to consider.
            Columns specified in subset that do not have matching data type are ignored.
            For example, if `value` is a string, and subset contains a non-string column,
            then the non-string column is simply ignored.

        >>> df4.na.replace(10, 20).show()
        +----+------+-----+
        | age|height| name|
        +----+------+-----+
        |  20|    80|Alice|
        |   5|  null|  Bob|
        |null|  null|  Tom|
        |null|  null| null|
        +----+------+-----+

        >>> df4.na.replace('Alice', None).show()
        +----+------+----+
        | age|height|name|
        +----+------+----+
        |  10|    80|null|
        |   5|  null| Bob|
        |null|  null| Tom|
        |null|  null|null|
        +----+------+----+

        >>> df4.na.replace({'Alice': None}).show()
        +----+------+----+
        | age|height|name|
        +----+------+----+
        |  10|    80|null|
        |   5|  null| Bob|
        |null|  null| Tom|
        |null|  null|null|
        +----+------+----+

        >>> df4.na.replace(['Alice', 'Bob'], ['A', 'B'], 'name').show()
        +----+------+----+
        | age|height|name|
        +----+------+----+
        |  10|    80|   A|
        |   5|  null|   B|
        |null|  null| Tom|
        |null|  null|null|
        +----+------+----+
        """
        if value is _NoValue:
            if isinstance(to_replace, dict):
                value = None
            else:
                raise TypeError("value argument is required when to_replace is not a dictionary.")

        # Helper functions
        def all_of(types):
            """Given a type or tuple of types and a sequence of xs
            check if each x is instance of type(s)

            >>> all_of(bool)([True, False])
            True
            >>> all_of(basestring)(["a", 1])
            False
            """
            def all_of_(xs):
                return all(isinstance(x, types) for x in xs)
            return all_of_

        all_of_bool = all_of(bool)
        all_of_str = all_of(basestring)
        all_of_numeric = all_of((float, int, long))

        # Validate input types
        valid_types = (bool, float, int, long, basestring, list, tuple)
        if not isinstance(to_replace, valid_types + (dict, )):
            raise ValueError(
                "to_replace should be a bool, float, int, long, string, list, tuple, or dict. "
                "Got {0}".format(type(to_replace)))

        if not isinstance(value, valid_types) and value is not None \
                and not isinstance(to_replace, dict):
            raise ValueError("If to_replace is not a dict, value should be "
                             "a bool, float, int, long, string, list, tuple or None. "
                             "Got {0}".format(type(value)))

        if isinstance(to_replace, (list, tuple)) and isinstance(value, (list, tuple)):
            if len(to_replace) != len(value):
                raise ValueError("to_replace and value lists should be of the same length. "
                                 "Got {0} and {1}".format(len(to_replace), len(value)))

        if not (subset is None or isinstance(subset, (list, tuple, basestring))):
            raise ValueError("subset should be a list or tuple of column names, "
                             "column name or None. Got {0}".format(type(subset)))

        # Reshape input arguments if necessary
        if isinstance(to_replace, (float, int, long, basestring)):
            to_replace = [to_replace]

        if isinstance(to_replace, dict):
            rep_dict = to_replace
            if value is not None:
                warnings.warn("to_replace is a dict and value is not None. value will be ignored.")
        else:
            if isinstance(value, (float, int, long, basestring)) or value is None:
                value = [value for _ in range(len(to_replace))]
            rep_dict = dict(zip(to_replace, value))

        if isinstance(subset, basestring):
            subset = [subset]

        # Verify we were not passed in mixed type generics.
        if not any(all_of_type(rep_dict.keys())
                   and all_of_type(x for x in rep_dict.values() if x is not None)
                   for all_of_type in [all_of_bool, all_of_str, all_of_numeric]):
            raise ValueError("Mixed type replacements are not supported")

        if subset is None:
            return DataFrame(self._jdf.na().replace('*', rep_dict), self.sql_ctx)
        else:
            return DataFrame(
                self._jdf.na().replace(self._jseq(subset), self._jmap(rep_dict)), self.sql_ctx)

    @since(2.0)
    def approxQuantile(self, col, probabilities, relativeError):
        """
        Calculates the approximate quantiles of numerical columns of a
        DataFrame.

        The result of this algorithm has the following deterministic bound:
        If the DataFrame has N elements and if we request the quantile at
        probability `p` up to error `err`, then the algorithm will return
        a sample `x` from the DataFrame so that the *exact* rank of `x` is
        close to (p * N). More precisely,

          floor((p - err) * N) <= rank(x) <= ceil((p + err) * N).

        This method implements a variation of the Greenwald-Khanna
        algorithm (with some speed optimizations). The algorithm was first
        present in [[http://dx.doi.org/10.1145/375663.375670
        Space-efficient Online Computation of Quantile Summaries]]
        by Greenwald and Khanna.

        Note that null values will be ignored in numerical columns before calculation.
        For columns only containing null values, an empty list is returned.

        :param col: str, list.
          Can be a single column name, or a list of names for multiple columns.
        :param probabilities: a list of quantile probabilities
          Each number must belong to [0, 1].
          For example 0 is the minimum, 0.5 is the median, 1 is the maximum.
        :param relativeError:  The relative target precision to achieve
          (>= 0). If set to zero, the exact quantiles are computed, which
          could be very expensive. Note that values greater than 1 are
          accepted but give the same result as 1.
        :return:  the approximate quantiles at the given probabilities. If
          the input `col` is a string, the output is a list of floats. If the
          input `col` is a list or tuple of strings, the output is also a
          list, but each element in it is a list of floats, i.e., the output
          is a list of list of floats.

        .. versionchanged:: 2.2
           Added support for multiple columns.
        """

        if not isinstance(col, (basestring, list, tuple)):
            raise ValueError("col should be a string, list or tuple, but got %r" % type(col))

        isStr = isinstance(col, basestring)

        if isinstance(col, tuple):
            col = list(col)
        elif isStr:
            col = [col]

        for c in col:
            if not isinstance(c, basestring):
                raise ValueError("columns should be strings, but got %r" % type(c))
        col = _to_list(self._sc, col)

        if not isinstance(probabilities, (list, tuple)):
            raise ValueError("probabilities should be a list or tuple")
        if isinstance(probabilities, tuple):
            probabilities = list(probabilities)
        for p in probabilities:
            if not isinstance(p, (float, int, long)) or p < 0 or p > 1:
                raise ValueError("probabilities should be numerical (float, int, long) in [0,1].")
        probabilities = _to_list(self._sc, probabilities)

        if not isinstance(relativeError, (float, int, long)) or relativeError < 0:
            raise ValueError("relativeError should be numerical (float, int, long) >= 0.")
        relativeError = float(relativeError)

        jaq = self._jdf.stat().approxQuantile(col, probabilities, relativeError)
        jaq_list = [list(j) for j in jaq]
        return jaq_list[0] if isStr else jaq_list

    @since(1.4)
    def corr(self, col1, col2, method=None):
        """
        Calculates the correlation of two columns of a DataFrame as a double value.
        Currently only supports the Pearson Correlation Coefficient.
        :func:`DataFrame.corr` and :func:`DataFrameStatFunctions.corr` are aliases of each other.

        :param col1: The name of the first column
        :param col2: The name of the second column
        :param method: The correlation method. Currently only supports "pearson"
        """
        if not isinstance(col1, basestring):
            raise ValueError("col1 should be a string.")
        if not isinstance(col2, basestring):
            raise ValueError("col2 should be a string.")
        if not method:
            method = "pearson"
        if not method == "pearson":
            raise ValueError("Currently only the calculation of the Pearson Correlation " +
                             "coefficient is supported.")
        return self._jdf.stat().corr(col1, col2, method)

    @since(1.4)
    def cov(self, col1, col2):
        """
        Calculate the sample covariance for the given columns, specified by their names, as a
        double value. :func:`DataFrame.cov` and :func:`DataFrameStatFunctions.cov` are aliases.

        :param col1: The name of the first column
        :param col2: The name of the second column
        """
        if not isinstance(col1, basestring):
            raise ValueError("col1 should be a string.")
        if not isinstance(col2, basestring):
            raise ValueError("col2 should be a string.")
        return self._jdf.stat().cov(col1, col2)

    @since(1.4)
    def crosstab(self, col1, col2):
        """
        Computes a pair-wise frequency table of the given columns. Also known as a contingency
        table. The number of distinct values for each column should be less than 1e4. At most 1e6
        non-zero pair frequencies will be returned.
        The first column of each row will be the distinct values of `col1` and the column names
        will be the distinct values of `col2`. The name of the first column will be `$col1_$col2`.
        Pairs that have no occurrences will have zero as their counts.
        :func:`DataFrame.crosstab` and :func:`DataFrameStatFunctions.crosstab` are aliases.

        :param col1: The name of the first column. Distinct items will make the first item of
            each row.
        :param col2: The name of the second column. Distinct items will make the column names
            of the DataFrame.
        """
        if not isinstance(col1, basestring):
            raise ValueError("col1 should be a string.")
        if not isinstance(col2, basestring):
            raise ValueError("col2 should be a string.")
        return DataFrame(self._jdf.stat().crosstab(col1, col2), self.sql_ctx)

    @since(1.4)
    def freqItems(self, cols, support=None):
        """
        Finding frequent items for columns, possibly with false positives. Using the
        frequent element count algorithm described in
        "http://dx.doi.org/10.1145/762471.762473, proposed by Karp, Schenker, and Papadimitriou".
        :func:`DataFrame.freqItems` and :func:`DataFrameStatFunctions.freqItems` are aliases.

        .. note:: This function is meant for exploratory data analysis, as we make no
            guarantee about the backward compatibility of the schema of the resulting DataFrame.

        :param cols: Names of the columns to calculate frequent items for as a list or tuple of
            strings.
        :param support: The frequency with which to consider an item 'frequent'. Default is 1%.
            The support must be greater than 1e-4.
        """
        if isinstance(cols, tuple):
            cols = list(cols)
        if not isinstance(cols, list):
            raise ValueError("cols must be a list or tuple of column names as strings.")
        if not support:
            support = 0.01
        return DataFrame(self._jdf.stat().freqItems(_to_seq(self._sc, cols), support), self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def withColumn(self, colName, col):
        """
        Returns a new :class:`DataFrame` by adding a column or replacing the
        existing column that has the same name.

        The column expression must be an expression over this DataFrame; attempting to add
        a column from some other dataframe will raise an error.

        :param colName: string, name of the new column.
        :param col: a :class:`Column` expression for the new column.

        >>> df.withColumn('age2', df.age + 2).collect()
        [Row(age=2, name=u'Alice', age2=4), Row(age=5, name=u'Bob', age2=7)]

        """
        assert isinstance(col, Column), "col should be Column"
        return DataFrame(self._jdf.withColumn(colName, col._jc), self.sql_ctx)

    @ignore_unicode_prefix
    @since(1.3)
    def withColumnRenamed(self, existing, new):
        """Returns a new :class:`DataFrame` by renaming an existing column.
        This is a no-op if schema doesn't contain the given column name.

        :param existing: string, name of the existing column to rename.
        :param new: string, new name of the column.

        >>> df.withColumnRenamed('age', 'age2').collect()
        [Row(age2=2, name=u'Alice'), Row(age2=5, name=u'Bob')]
        """
        return DataFrame(self._jdf.withColumnRenamed(existing, new), self.sql_ctx)

    @since(1.4)
    @ignore_unicode_prefix
    def drop(self, *cols):
        """Returns a new :class:`DataFrame` that drops the specified column.
        This is a no-op if schema doesn't contain the given column name(s).

        :param cols: a string name of the column to drop, or a
            :class:`Column` to drop, or a list of string name of the columns to drop.

        >>> df.drop('age').collect()
        [Row(name=u'Alice'), Row(name=u'Bob')]

        >>> df.drop(df.age).collect()
        [Row(name=u'Alice'), Row(name=u'Bob')]

        >>> df.join(df2, df.name == df2.name, 'inner').drop(df.name).collect()
        [Row(age=5, height=85, name=u'Bob')]

        >>> df.join(df2, df.name == df2.name, 'inner').drop(df2.name).collect()
        [Row(age=5, name=u'Bob', height=85)]

        >>> df.join(df2, 'name', 'inner').drop('age', 'height').collect()
        [Row(name=u'Bob')]
        """
        if len(cols) == 1:
            col = cols[0]
            if isinstance(col, basestring):
                jdf = self._jdf.drop(col)
            elif isinstance(col, Column):
                jdf = self._jdf.drop(col._jc)
            else:
                raise TypeError("col should be a string or a Column")
        else:
            for col in cols:
                if not isinstance(col, basestring):
                    raise TypeError("each col in the param list should be a string")
            jdf = self._jdf.drop(self._jseq(cols))

        return DataFrame(jdf, self.sql_ctx)

    @ignore_unicode_prefix
    def toDF(self, *cols):
        """Returns a new class:`DataFrame` that with new specified column names

        :param cols: list of new column names (string)

        >>> df.toDF('f1', 'f2').collect()
        [Row(f1=2, f2=u'Alice'), Row(f1=5, f2=u'Bob')]
        """
        jdf = self._jdf.toDF(self._jseq(cols))
        return DataFrame(jdf, self.sql_ctx)

    @since(1.3)
    def toPandas(self):
        """
        Returns the contents of this :class:`DataFrame` as Pandas ``pandas.DataFrame``.

        This is only available if Pandas is installed and available.

        .. note:: This method should only be used if the resulting Pandas's DataFrame is expected
            to be small, as all the data is loaded into the driver's memory.

        .. note:: Usage with spark.sql.execution.arrow.enabled=True is experimental.

        >>> df.toPandas()  # doctest: +SKIP
           age   name
        0    2  Alice
        1    5    Bob
        """
        from pyspark.sql.utils import require_minimum_pandas_version
        require_minimum_pandas_version()

        import pandas as pd

        if self.sql_ctx._conf.pandasRespectSessionTimeZone():
            timezone = self.sql_ctx._conf.sessionLocalTimeZone()
        else:
            timezone = None

        if self.sql_ctx._conf.arrowEnabled():
            use_arrow = True
            try:
                from pyspark.sql.types import to_arrow_schema
                from pyspark.sql.utils import require_minimum_pyarrow_version

                require_minimum_pyarrow_version()
                to_arrow_schema(self.schema)
            except Exception as e:

                if self.sql_ctx._conf.arrowFallbackEnabled():
                    msg = (
                        "toPandas attempted Arrow optimization because "
                        "'spark.sql.execution.arrow.enabled' is set to true; however, "
                        "failed by the reason below:\n  %s\n"
                        "Attempting non-optimization as "
                        "'spark.sql.execution.arrow.fallback.enabled' is set to "
                        "true." % _exception_message(e))
                    warnings.warn(msg)
                    use_arrow = False
                else:
                    msg = (
                        "toPandas attempted Arrow optimization because "
                        "'spark.sql.execution.arrow.enabled' is set to true, but has reached "
                        "the error below and will not continue because automatic fallback "
                        "with 'spark.sql.execution.arrow.fallback.enabled' has been set to "
                        "false.\n  %s" % _exception_message(e))
                    warnings.warn(msg)
                    raise

            # Try to use Arrow optimization when the schema is supported and the required version
            # of PyArrow is found, if 'spark.sql.execution.arrow.enabled' is enabled.
            if use_arrow:
                try:
                    from pyspark.sql.types import _check_dataframe_convert_date, \
                        _check_dataframe_localize_timestamps
                    import pyarrow
                    batches = self._collectAsArrow()
                    if len(batches) > 0:
                        table = pyarrow.Table.from_batches(batches)
                        pdf = table.to_pandas()
                        pdf = _check_dataframe_convert_date(pdf, self.schema)
                        return _check_dataframe_localize_timestamps(pdf, timezone)
                    else:
                        return pd.DataFrame.from_records([], columns=self.columns)
                except Exception as e:
                    # We might have to allow fallback here as well but multiple Spark jobs can
                    # be executed. So, simply fail in this case for now.
                    msg = (
                        "toPandas attempted Arrow optimization because "
                        "'spark.sql.execution.arrow.enabled' is set to true, but has reached "
                        "the error below and can not continue. Note that "
                        "'spark.sql.execution.arrow.fallback.enabled' does not have an effect "
                        "on failures in the middle of computation.\n  %s" % _exception_message(e))
                    warnings.warn(msg)
                    raise

        # Below is toPandas without Arrow optimization.
        pdf = pd.DataFrame.from_records(self.collect(), columns=self.columns)

        dtype = {}
        for field in self.schema:
            pandas_type = _to_corrected_pandas_type(field.dataType)
            # SPARK-21766: if an integer field is nullable and has null values, it can be
            # inferred by pandas as float column. Once we convert the column with NaN back
            # to integer type e.g., np.int16, we will hit exception. So we use the inferred
            # float type, not the corrected type from the schema in this case.
            if pandas_type is not None and \
                not(isinstance(field.dataType, IntegralType) and field.nullable and
                    pdf[field.name].isnull().any()):
                dtype[field.name] = pandas_type

        for f, t in dtype.items():
            pdf[f] = pdf[f].astype(t, copy=False)

        if timezone is None:
            return pdf
        else:
            from pyspark.sql.types import _check_series_convert_timestamps_local_tz
            for field in self.schema:
                # TODO: handle nested timestamps, such as ArrayType(TimestampType())?
                if isinstance(field.dataType, TimestampType):
                    pdf[field.name] = \
                        _check_series_convert_timestamps_local_tz(pdf[field.name], timezone)
            return pdf

    def _collectAsArrow(self):
        """
        Returns all records as a list of ArrowRecordBatches, pyarrow must be installed
        and available on driver and worker Python environments.

        .. note:: Experimental.
        """
        with SCCallSiteSync(self._sc) as css:
            sock_info = self._jdf.collectAsArrowToPython()
        return list(_load_from_socket(sock_info, ArrowStreamSerializer()))

    ##########################################################################################
    # Pandas compatibility
    ##########################################################################################

    groupby = copy_func(
        groupBy,
        sinceversion=1.4,
        doc=":func:`groupby` is an alias for :func:`groupBy`.")

    drop_duplicates = copy_func(
        dropDuplicates,
        sinceversion=1.4,
        doc=":func:`drop_duplicates` is an alias for :func:`dropDuplicates`.")

    where = copy_func(
        filter,
        sinceversion=1.3,
        doc=":func:`where` is an alias for :func:`filter`.")


def _to_scala_map(sc, jm):
    """
    Convert a dict into a JVM Map.
    """
    return sc._jvm.PythonUtils.toScalaMap(jm)


def _to_corrected_pandas_type(dt):
    """
    When converting Spark SQL records to Pandas DataFrame, the inferred data type may be wrong.
    This method gets the corrected data type for Pandas if that type may be inferred uncorrectly.
    """
    import numpy as np
    if type(dt) == ByteType:
        return np.int8
    elif type(dt) == ShortType:
        return np.int16
    elif type(dt) == IntegerType:
        return np.int32
    elif type(dt) == FloatType:
        return np.float32
    else:
        return None


class DataFrameNaFunctions(object):
    """Functionality for working with missing data in :class:`DataFrame`.

    .. versionadded:: 1.4
    """

    def __init__(self, df):
        self.df = df

    def drop(self, how='any', thresh=None, subset=None):
        return self.df.dropna(how=how, thresh=thresh, subset=subset)

    drop.__doc__ = DataFrame.dropna.__doc__

    def fill(self, value, subset=None):
        return self.df.fillna(value=value, subset=subset)

    fill.__doc__ = DataFrame.fillna.__doc__

    def replace(self, to_replace, value=_NoValue, subset=None):
        return self.df.replace(to_replace, value, subset)

    replace.__doc__ = DataFrame.replace.__doc__


class DataFrameStatFunctions(object):
    """Functionality for statistic functions with :class:`DataFrame`.

    .. versionadded:: 1.4
    """

    def __init__(self, df):
        self.df = df

    def approxQuantile(self, col, probabilities, relativeError):
        return self.df.approxQuantile(col, probabilities, relativeError)

    approxQuantile.__doc__ = DataFrame.approxQuantile.__doc__

    def corr(self, col1, col2, method=None):
        return self.df.corr(col1, col2, method)

    corr.__doc__ = DataFrame.corr.__doc__

    def cov(self, col1, col2):
        return self.df.cov(col1, col2)

    cov.__doc__ = DataFrame.cov.__doc__

    def crosstab(self, col1, col2):
        return self.df.crosstab(col1, col2)

    crosstab.__doc__ = DataFrame.crosstab.__doc__

    def freqItems(self, cols, support=None):
        return self.df.freqItems(cols, support)

    freqItems.__doc__ = DataFrame.freqItems.__doc__

    def sampleBy(self, col, fractions, seed=None):
        return self.df.sampleBy(col, fractions, seed)

    sampleBy.__doc__ = DataFrame.sampleBy.__doc__


def _test():
    import doctest
    from pyspark.context import SparkContext
    from pyspark.sql import Row, SQLContext, SparkSession
    import pyspark.sql.dataframe
    from pyspark.sql.functions import from_unixtime
    globs = pyspark.sql.dataframe.__dict__.copy()
    sc = SparkContext('local[4]', 'PythonTest')
    globs['sc'] = sc
    globs['sqlContext'] = SQLContext(sc)
    globs['spark'] = SparkSession(sc)
    globs['df'] = sc.parallelize([(2, 'Alice'), (5, 'Bob')])\
        .toDF(StructType([StructField('age', IntegerType()),
                          StructField('name', StringType())]))
    globs['df2'] = sc.parallelize([Row(name='Tom', height=80), Row(name='Bob', height=85)]).toDF()
    globs['df3'] = sc.parallelize([Row(name='Alice', age=2),
                                   Row(name='Bob', age=5)]).toDF()
    globs['df4'] = sc.parallelize([Row(name='Alice', age=10, height=80),
                                   Row(name='Bob', age=5, height=None),
                                   Row(name='Tom', age=None, height=None),
                                   Row(name=None, age=None, height=None)]).toDF()
    globs['df5'] = sc.parallelize([Row(name='Alice', spy=False, age=10),
                                   Row(name='Bob', spy=None, age=5),
                                   Row(name='Mallory', spy=True, age=None)]).toDF()
    globs['sdf'] = sc.parallelize([Row(name='Tom', time=1479441846),
                                   Row(name='Bob', time=1479442946)]).toDF()

    (failure_count, test_count) = doctest.testmod(
        pyspark.sql.dataframe, globs=globs,
        optionflags=doctest.ELLIPSIS | doctest.NORMALIZE_WHITESPACE | doctest.REPORT_NDIFF)
    globs['sc'].stop()
    if failure_count:
        sys.exit(-1)


if __name__ == "__main__":
    _test()