Coverage for pyspark/sql/functions.py : 90%
Hot-keys on this page
r m x p toggle line displays
j k next/prev highlighted chunk
0 (zero) top of page
1 (one) first highlighted chunk
|
# # 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. #
A collections of builtin functions """
# Keep UserDefinedFunction import for backwards compatible import; moved in SPARK-22409 # Keep pandas_udf and PandasUDFType import for backwards compatible import; moved in SPARK-28264
# Note to developers: all of PySpark functions here take string as column names whenever possible. # Namely, if columns are referred as arguments, they can be always both Column or string, # even though there might be few exceptions for legacy or inevitable reasons. # If you are fixing other language APIs together, also please note that Scala side is not the case # since it requires to make every single overridden definition.
""" Retrieves JVM function identified by name from Java gateway associated with sc. """
""" Invokes JVM function identified by name with args and wraps the result with :class:`~pyspark.sql.Column`. """
""" Invokes unary JVM function identified by name and wraps the result with :class:`~pyspark.sql.Column`. """
""" Invokes binary JVM math function identified by name and wraps the result with :class:`~pyspark.sql.Column`. """ name, # For legacy reasons, the arguments here can be implicitly converted into floats, # if they are not columns or strings. _to_java_column(col1) if isinstance(col1, (str, Column)) else float(col1), _to_java_column(col2) if isinstance(col2, (str, Column)) else float(col2) )
""" Creates a :class:`~pyspark.sql.Column` of literal value.
.. versionadded:: 1.3.0
Examples -------- >>> df.select(lit(5).alias('height')).withColumn('spark_user', lit(True)).take(1) [Row(height=5, spark_user=True)] """
def col(col): """ Returns a :class:`~pyspark.sql.Column` based on the given column name.' Examples -------- >>> col('x') Column<'x'> >>> column('x') Column<'x'> """
def asc(col): """ Returns a sort expression based on the ascending order of the given column name. """ col.asc() if isinstance(col, Column) else _invoke_function("asc", col) )
def desc(col): """ Returns a sort expression based on the descending order of the given column name. """ col.desc() if isinstance(col, Column) else _invoke_function("desc", col) )
def sqrt(col): """ Computes the square root of the specified float value. """ return _invoke_function_over_column("sqrt", col)
def abs(col): """ Computes the absolute value. """ return _invoke_function_over_column("abs", col)
def max(col): """ Aggregate function: returns the maximum value of the expression in a group. """
def min(col): """ Aggregate function: returns the minimum value of the expression in a group. """
def count(col): """ Aggregate function: returns the number of items in a group. """
def sum(col): """ Aggregate function: returns the sum of all values in the expression. """
def avg(col): """ Aggregate function: returns the average of the values in a group. """ return _invoke_function_over_column("avg", col)
def mean(col): """ Aggregate function: returns the average of the values in a group. """
def sumDistinct(col): """ Aggregate function: returns the sum of distinct values in the expression.
.. deprecated:: 3.2.0 Use :func:`sum_distinct` instead. """
def sum_distinct(col): """ Aggregate function: returns the sum of distinct values in the expression. """
""" Aggregate function: returns the product of the values in a group.
.. versionadded:: 3.2.0
Parameters ---------- col : str, :class:`Column` column containing values to be multiplied together
Examples -------- >>> df = spark.range(1, 10).toDF('x').withColumn('mod3', col('x') % 3) >>> prods = df.groupBy('mod3').agg(product('x').alias('product')) >>> prods.orderBy('mod3').show() +----+-------+ |mod3|product| +----+-------+ | 0| 162.0| | 1| 28.0| | 2| 80.0| +----+-------+
"""
""" .. versionadded:: 1.4.0
Returns ------- :class:`~pyspark.sql.Column` inverse cosine of `col`, as if computed by `java.lang.Math.acos()` """ return _invoke_function_over_column("acos", col)
""" Computes inverse hyperbolic cosine of the input column.
.. versionadded:: 3.1.0
Returns ------- :class:`~pyspark.sql.Column` """
""" .. versionadded:: 1.3.0
Returns ------- :class:`~pyspark.sql.Column` inverse sine of `col`, as if computed by `java.lang.Math.asin()` """ return _invoke_function_over_column("asin", col)
""" Computes inverse hyperbolic sine of the input column.
.. versionadded:: 3.1.0
Returns ------- :class:`~pyspark.sql.Column` """
""" .. versionadded:: 1.4.0
Returns ------- :class:`~pyspark.sql.Column` inverse tangent of `col`, as if computed by `java.lang.Math.atan()` """ return _invoke_function_over_column("atan", col)
""" Computes inverse hyperbolic tangent of the input column.
.. versionadded:: 3.1.0
Returns ------- :class:`~pyspark.sql.Column` """
def cbrt(col): """ Computes the cube-root of the given value. """ return _invoke_function_over_column("cbrt", col)
def ceil(col): """ Computes the ceiling of the given value. """ return _invoke_function_over_column("ceil", col)
""" .. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str angle in radians
Returns ------- :class:`~pyspark.sql.Column` cosine of the angle, as if computed by `java.lang.Math.cos()`. """
""" .. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str hyperbolic angle
Returns ------- :class:`~pyspark.sql.Column` hyperbolic cosine of the angle, as if computed by `java.lang.Math.cosh()` """ return _invoke_function_over_column("cosh", col)
def exp(col): """ Computes the exponential of the given value. """ return _invoke_function_over_column("exp", col)
def expm1(col): """ Computes the exponential of the given value minus one. """ return _invoke_function_over_column("expm1", col)
def floor(col): """ Computes the floor of the given value. """ return _invoke_function_over_column("floor", col)
def log(col): """ Computes the natural logarithm of the given value. """ return _invoke_function_over_column("log", col)
def log10(col): """ Computes the logarithm of the given value in Base 10. """ return _invoke_function_over_column("log10", col)
def log1p(col): """ Computes the natural logarithm of the given value plus one. """ return _invoke_function_over_column("log1p", col)
def rint(col): """ Returns the double value that is closest in value to the argument and is equal to a mathematical integer. """ return _invoke_function_over_column("rint", col)
def signum(col): """ Computes the signum of the given value. """ return _invoke_function_over_column("signum", col)
""" .. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str
Returns ------- :class:`~pyspark.sql.Column` sine of the angle, as if computed by `java.lang.Math.sin()` """
""" .. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str hyperbolic angle
Returns ------- :class:`~pyspark.sql.Column` hyperbolic sine of the given value, as if computed by `java.lang.Math.sinh()` """ return _invoke_function_over_column("sinh", col)
""" .. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str angle in radians
Returns ------- :class:`~pyspark.sql.Column` tangent of the given value, as if computed by `java.lang.Math.tan()` """ return _invoke_function_over_column("tan", col)
""" .. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str hyperbolic angle
Returns ------- :class:`~pyspark.sql.Column` hyperbolic tangent of the given value as if computed by `java.lang.Math.tanh()` """ return _invoke_function_over_column("tanh", col)
def toDegrees(col): """ .. deprecated:: 2.1.0 Use :func:`degrees` instead. """ warnings.warn("Deprecated in 2.1, use degrees instead.", FutureWarning) return degrees(col)
def toRadians(col): """ .. deprecated:: 2.1.0 Use :func:`radians` instead. """ warnings.warn("Deprecated in 2.1, use radians instead.", FutureWarning) return radians(col)
def bitwiseNOT(col): """ Computes bitwise not.
.. deprecated:: 3.2.0 Use :func:`bitwise_not` instead. """
def bitwise_not(col): """ Computes bitwise not. """
def asc_nulls_first(col): """ Returns a sort expression based on the ascending order of the given column name, and null values return before non-null values. """ col.asc_nulls_first() if isinstance(col, Column) else _invoke_function("asc_nulls_first", col) )
def asc_nulls_last(col): """ Returns a sort expression based on the ascending order of the given column name, and null values appear after non-null values. """ col.asc_nulls_last() if isinstance(col, Column) else _invoke_function("asc_nulls_last", col) )
def desc_nulls_first(col): """ Returns a sort expression based on the descending order of the given column name, and null values appear before non-null values. """ col.desc_nulls_first() if isinstance(col, Column) else _invoke_function("desc_nulls_first", col) )
def desc_nulls_last(col): """ Returns a sort expression based on the descending order of the given column name, and null values appear after non-null values. """ col.desc_nulls_last() if isinstance(col, Column) else _invoke_function("desc_nulls_last", col) )
def stddev(col): """ Aggregate function: alias for stddev_samp. """ return _invoke_function_over_column("stddev", col)
def stddev_samp(col): """ Aggregate function: returns the unbiased sample standard deviation of the expression in a group. """ return _invoke_function_over_column("stddev_samp", col)
def stddev_pop(col): """ Aggregate function: returns population standard deviation of the expression in a group. """ return _invoke_function_over_column("stddev_pop", col)
def variance(col): """ Aggregate function: alias for var_samp """ return _invoke_function_over_column("variance", col)
def var_samp(col): """ Aggregate function: returns the unbiased sample variance of the values in a group. """ return _invoke_function_over_column("var_samp", col)
def var_pop(col): """ Aggregate function: returns the population variance of the values in a group. """ return _invoke_function_over_column("var_pop", col)
def skewness(col): """ Aggregate function: returns the skewness of the values in a group. """ return _invoke_function_over_column("skewness", col)
def kurtosis(col): """ Aggregate function: returns the kurtosis of the values in a group. """ return _invoke_function_over_column("kurtosis", col)
""" Aggregate function: returns a list of objects with duplicates.
.. versionadded:: 1.6.0
Notes ----- The function is non-deterministic because the order of collected results depends on the order of the rows which may be non-deterministic after a shuffle.
Examples -------- >>> df2 = spark.createDataFrame([(2,), (5,), (5,)], ('age',)) >>> df2.agg(collect_list('age')).collect() [Row(collect_list(age)=[2, 5, 5])] """
""" Aggregate function: returns a set of objects with duplicate elements eliminated.
.. versionadded:: 1.6.0
Notes ----- The function is non-deterministic because the order of collected results depends on the order of the rows which may be non-deterministic after a shuffle.
Examples -------- >>> df2 = spark.createDataFrame([(2,), (5,), (5,)], ('age',)) >>> df2.agg(collect_set('age')).collect() [Row(collect_set(age)=[5, 2])] """
""" Converts an angle measured in radians to an approximately equivalent angle measured in degrees.
.. versionadded:: 2.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str angle in radians
Returns ------- :class:`~pyspark.sql.Column` angle in degrees, as if computed by `java.lang.Math.toDegrees()` """ return _invoke_function_over_column("degrees", col)
""" Converts an angle measured in degrees to an approximately equivalent angle measured in radians.
.. versionadded:: 2.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str angle in degrees
Returns ------- :class:`~pyspark.sql.Column` angle in radians, as if computed by `java.lang.Math.toRadians()` """ return _invoke_function_over_column("radians", col)
""" .. versionadded:: 1.4.0
Parameters ---------- col1 : str, :class:`~pyspark.sql.Column` or float coordinate on y-axis col2 : str, :class:`~pyspark.sql.Column` or float coordinate on x-axis
Returns ------- :class:`~pyspark.sql.Column` the `theta` component of the point (`r`, `theta`) in polar coordinates that corresponds to the point (`x`, `y`) in Cartesian coordinates, as if computed by `java.lang.Math.atan2()` """ return _invoke_binary_math_function("atan2", col1, col2)
def hypot(col1, col2): """ Computes ``sqrt(a^2 + b^2)`` without intermediate overflow or underflow. """
def pow(col1, col2): """ Returns the value of the first argument raised to the power of the second argument. """
def row_number(): """ Window function: returns a sequential number starting at 1 within a window partition. """
def dense_rank(): """ Window function: returns the rank of rows within a window partition, without any gaps.
The difference between rank and dense_rank is that dense_rank leaves no gaps in ranking sequence when there are ties. That is, if you were ranking a competition using dense_rank and had three people tie for second place, you would say that all three were in second place and that the next person came in third. Rank would give me sequential numbers, making the person that came in third place (after the ties) would register as coming in fifth.
This is equivalent to the DENSE_RANK function in SQL. """
def rank(): """ Window function: returns the rank of rows within a window partition.
The difference between rank and dense_rank is that dense_rank leaves no gaps in ranking sequence when there are ties. That is, if you were ranking a competition using dense_rank and had three people tie for second place, you would say that all three were in second place and that the next person came in third. Rank would give me sequential numbers, making the person that came in third place (after the ties) would register as coming in fifth.
This is equivalent to the RANK function in SQL. """
def cume_dist(): """ Window function: returns the cumulative distribution of values within a window partition, i.e. the fraction of rows that are below the current row. """ return _invoke_function("cume_dist")
def percent_rank(): """ Window function: returns the relative rank (i.e. percentile) of rows within a window partition. """ return _invoke_function("percent_rank")
""" .. deprecated:: 2.1.0 Use :func:`approx_count_distinct` instead. """ warnings.warn("Deprecated in 2.1, use approx_count_distinct instead.", FutureWarning) return approx_count_distinct(col, rsd)
"""Aggregate function: returns a new :class:`~pyspark.sql.Column` for approximate distinct count of column `col`.
.. versionadded:: 2.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str rsd : float, optional maximum relative standard deviation allowed (default = 0.05). For rsd < 0.01, it is more efficient to use :func:`count_distinct`
Examples -------- >>> df.agg(approx_count_distinct(df.age).alias('distinct_ages')).collect() [Row(distinct_ages=2)] """ else: jc = sc._jvm.functions.approx_count_distinct(_to_java_column(col), rsd)
def broadcast(df): """Marks a DataFrame as small enough for use in broadcast joins."""
"""Returns the first column that is not null.
.. versionadded:: 1.4.0
Examples -------- >>> cDf = spark.createDataFrame([(None, None), (1, None), (None, 2)], ("a", "b")) >>> cDf.show() +----+----+ | a| b| +----+----+ |null|null| | 1|null| |null| 2| +----+----+
>>> cDf.select(coalesce(cDf["a"], cDf["b"])).show() +--------------+ |coalesce(a, b)| +--------------+ | null| | 1| | 2| +--------------+
>>> cDf.select('*', coalesce(cDf["a"], lit(0.0))).show() +----+----+----------------+ | a| b|coalesce(a, 0.0)| +----+----+----------------+ |null|null| 0.0| | 1|null| 1.0| |null| 2| 0.0| +----+----+----------------+ """
"""Returns a new :class:`~pyspark.sql.Column` for the Pearson Correlation Coefficient for ``col1`` and ``col2``.
.. versionadded:: 1.6.0
Examples -------- >>> a = range(20) >>> b = [2 * x for x in range(20)] >>> df = spark.createDataFrame(zip(a, b), ["a", "b"]) >>> df.agg(corr("a", "b").alias('c')).collect() [Row(c=1.0)] """
"""Returns a new :class:`~pyspark.sql.Column` for the population covariance of ``col1`` and ``col2``.
.. versionadded:: 2.0.0
Examples -------- >>> a = [1] * 10 >>> b = [1] * 10 >>> df = spark.createDataFrame(zip(a, b), ["a", "b"]) >>> df.agg(covar_pop("a", "b").alias('c')).collect() [Row(c=0.0)] """
"""Returns a new :class:`~pyspark.sql.Column` for the sample covariance of ``col1`` and ``col2``.
.. versionadded:: 2.0.0
Examples -------- >>> a = [1] * 10 >>> b = [1] * 10 >>> df = spark.createDataFrame(zip(a, b), ["a", "b"]) >>> df.agg(covar_samp("a", "b").alias('c')).collect() [Row(c=0.0)] """
"""Returns a new :class:`~pyspark.sql.Column` for distinct count of ``col`` or ``cols``.
An alias of :func:`count_distinct`, and it is encouraged to use :func:`count_distinct` directly.
.. versionadded:: 1.3.0 """
"""Returns a new :class:`Column` for distinct count of ``col`` or ``cols``.
.. versionadded:: 3.2.0
Examples -------- >>> df.agg(count_distinct(df.age, df.name).alias('c')).collect() [Row(c=2)]
>>> df.agg(count_distinct("age", "name").alias('c')).collect() [Row(c=2)] """
"""Aggregate function: returns the first value in a group.
The function by default returns the first values it sees. It will return the first non-null value it sees when ignoreNulls is set to true. If all values are null, then null is returned.
.. versionadded:: 1.3.0
Notes ----- The function is non-deterministic because its results depends on the order of the rows which may be non-deterministic after a shuffle. """
""" Aggregate function: indicates whether a specified column in a GROUP BY list is aggregated or not, returns 1 for aggregated or 0 for not aggregated in the result set.
.. versionadded:: 2.0.0
Examples -------- >>> df.cube("name").agg(grouping("name"), sum("age")).orderBy("name").show() +-----+--------------+--------+ | name|grouping(name)|sum(age)| +-----+--------------+--------+ | null| 1| 7| |Alice| 0| 2| | Bob| 0| 5| +-----+--------------+--------+ """
""" Aggregate function: returns the level of grouping, equals to
(grouping(c1) << (n-1)) + (grouping(c2) << (n-2)) + ... + grouping(cn)
.. versionadded:: 2.0.0
Notes ----- The list of columns should match with grouping columns exactly, or empty (means all the grouping columns).
Examples -------- >>> df.cube("name").agg(grouping_id(), sum("age")).orderBy("name").show() +-----+-------------+--------+ | name|grouping_id()|sum(age)| +-----+-------------+--------+ | null| 1| 7| |Alice| 0| 2| | Bob| 0| 5| +-----+-------------+--------+ """
def input_file_name(): """Creates a string column for the file name of the current Spark task. """
"""An expression that returns true iff the column is NaN.
.. versionadded:: 1.6.0
Examples -------- >>> df = spark.createDataFrame([(1.0, float('nan')), (float('nan'), 2.0)], ("a", "b")) >>> df.select(isnan("a").alias("r1"), isnan(df.a).alias("r2")).collect() [Row(r1=False, r2=False), Row(r1=True, r2=True)] """
"""An expression that returns true iff the column is null.
.. versionadded:: 1.6.0
Examples -------- >>> df = spark.createDataFrame([(1, None), (None, 2)], ("a", "b")) >>> df.select(isnull("a").alias("r1"), isnull(df.a).alias("r2")).collect() [Row(r1=False, r2=False), Row(r1=True, r2=True)] """
"""Aggregate function: returns the last value in a group.
The function by default returns the last values it sees. It will return the last non-null value it sees when ignoreNulls is set to true. If all values are null, then null is returned.
.. versionadded:: 1.3.0
Notes ----- The function is non-deterministic because its results depends on the order of the rows which may be non-deterministic after a shuffle. """
"""A column that generates monotonically increasing 64-bit integers.
The generated ID is guaranteed to be monotonically increasing and unique, but not consecutive. The current implementation puts the partition ID in the upper 31 bits, and the record number within each partition in the lower 33 bits. The assumption is that the data frame has less than 1 billion partitions, and each partition has less than 8 billion records.
.. versionadded:: 1.6.0
Notes ----- The function is non-deterministic because its result depends on partition IDs.
As an example, consider a :class:`DataFrame` with two partitions, each with 3 records. This expression would return the following IDs: 0, 1, 2, 8589934592 (1L << 33), 8589934593, 8589934594.
>>> df0 = sc.parallelize(range(2), 2).mapPartitions(lambda x: [(1,), (2,), (3,)]).toDF(['col1']) >>> df0.select(monotonically_increasing_id().alias('id')).collect() [Row(id=0), Row(id=1), Row(id=2), Row(id=8589934592), Row(id=8589934593), Row(id=8589934594)] """
"""Returns col1 if it is not NaN, or col2 if col1 is NaN.
Both inputs should be floating point columns (:class:`DoubleType` or :class:`FloatType`).
.. versionadded:: 1.6.0
Examples -------- >>> df = spark.createDataFrame([(1.0, float('nan')), (float('nan'), 2.0)], ("a", "b")) >>> df.select(nanvl("a", "b").alias("r1"), nanvl(df.a, df.b).alias("r2")).collect() [Row(r1=1.0, r2=1.0), Row(r1=2.0, r2=2.0)] """
"""Returns the approximate `percentile` of the numeric column `col` which is the smallest value in the ordered `col` values (sorted from least to greatest) such that no more than `percentage` of `col` values is less than the value or equal to that value. The value of percentage must be between 0.0 and 1.0.
The accuracy parameter (default: 10000) is a positive numeric literal which controls approximation accuracy at the cost of memory. Higher value of accuracy yields better accuracy, 1.0/accuracy is the relative error of the approximation.
When percentage is an array, each value of the percentage array must be between 0.0 and 1.0. In this case, returns the approximate percentile array of column col at the given percentage array.
.. versionadded:: 3.1.0
Examples -------- >>> key = (col("id") % 3).alias("key") >>> value = (randn(42) + key * 10).alias("value") >>> df = spark.range(0, 1000, 1, 1).select(key, value) >>> df.select( ... percentile_approx("value", [0.25, 0.5, 0.75], 1000000).alias("quantiles") ... ).printSchema() root |-- quantiles: array (nullable = true) | |-- element: double (containsNull = false)
>>> df.groupBy("key").agg( ... percentile_approx("value", 0.5, lit(1000000)).alias("median") ... ).printSchema() root |-- key: long (nullable = true) |-- median: double (nullable = true) """
# A local list _create_column_from_literal(x) for x in percentage ])) # Already a Column else: # Probably scalar
_to_java_column(accuracy) if isinstance(accuracy, Column) else _create_column_from_literal(accuracy) )
"""Generates a random column with independent and identically distributed (i.i.d.) samples uniformly distributed in [0.0, 1.0).
.. versionadded:: 1.4.0
Notes ----- The function is non-deterministic in general case.
Examples -------- >>> df.withColumn('rand', rand(seed=42) * 3).collect() [Row(age=2, name='Alice', rand=2.4052597283576684), Row(age=5, name='Bob', rand=2.3913904055683974)] """ else:
"""Generates a column with independent and identically distributed (i.i.d.) samples from the standard normal distribution.
.. versionadded:: 1.4.0
Notes ----- The function is non-deterministic in general case.
Examples -------- >>> df.withColumn('randn', randn(seed=42)).collect() [Row(age=2, name='Alice', randn=1.1027054481455365), Row(age=5, name='Bob', randn=0.7400395449950132)] """ else: jc = sc._jvm.functions.randn()
""" Round the given value to `scale` decimal places using HALF_UP rounding mode if `scale` >= 0 or at integral part when `scale` < 0.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([(2.5,)], ['a']).select(round('a', 0).alias('r')).collect() [Row(r=3.0)] """
""" Round the given value to `scale` decimal places using HALF_EVEN rounding mode if `scale` >= 0 or at integral part when `scale` < 0.
.. versionadded:: 2.0.0
Examples -------- >>> spark.createDataFrame([(2.5,)], ['a']).select(bround('a', 0).alias('r')).collect() [Row(r=2.0)] """
"""Shift the given value numBits left.
.. versionadded:: 1.5.0
.. deprecated:: 3.2.0 Use :func:`shiftleft` instead. """
"""Shift the given value numBits left.
.. versionadded:: 3.2.0
Examples -------- >>> spark.createDataFrame([(21,)], ['a']).select(shiftleft('a', 1).alias('r')).collect() [Row(r=42)] """
"""(Signed) shift the given value numBits right.
.. versionadded:: 1.5.0
.. deprecated:: 3.2.0 Use :func:`shiftright` instead. """
"""(Signed) shift the given value numBits right.
.. versionadded:: 3.2.0
Examples -------- >>> spark.createDataFrame([(42,)], ['a']).select(shiftright('a', 1).alias('r')).collect() [Row(r=21)] """
"""Unsigned shift the given value numBits right.
.. versionadded:: 1.5.0
.. deprecated:: 3.2.0 Use :func:`shiftrightunsigned` instead. """
"""Unsigned shift the given value numBits right.
.. versionadded:: 3.2.0
Examples -------- >>> df = spark.createDataFrame([(-42,)], ['a']) >>> df.select(shiftrightunsigned('a', 1).alias('r')).collect() [Row(r=9223372036854775787)] """
"""A column for partition ID.
.. versionadded:: 1.6.0
Notes ----- This is non deterministic because it depends on data partitioning and task scheduling.
Examples -------- >>> df.repartition(1).select(spark_partition_id().alias("pid")).collect() [Row(pid=0), Row(pid=0)] """
"""Parses the expression string into the column that it represents
.. versionadded:: 1.5.0
Examples -------- >>> df.select(expr("length(name)")).collect() [Row(length(name)=5), Row(length(name)=3)] """
"""Creates a new struct column.
.. versionadded:: 1.4.0
Parameters ---------- cols : list, set, str or :class:`~pyspark.sql.Column` column names or :class:`~pyspark.sql.Column`\\s to contain in the output struct.
Examples -------- >>> df.select(struct('age', 'name').alias("struct")).collect() [Row(struct=Row(age=2, name='Alice')), Row(struct=Row(age=5, name='Bob'))] >>> df.select(struct([df.age, df.name]).alias("struct")).collect() [Row(struct=Row(age=2, name='Alice')), Row(struct=Row(age=5, name='Bob'))] """
""" Returns the greatest value of the list of column names, skipping null values. This function takes at least 2 parameters. It will return null iff all parameters are null.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([(1, 4, 3)], ['a', 'b', 'c']) >>> df.select(greatest(df.a, df.b, df.c).alias("greatest")).collect() [Row(greatest=4)] """ raise ValueError("greatest should take at least two columns")
""" Returns the least value of the list of column names, skipping null values. This function takes at least 2 parameters. It will return null iff all parameters are null.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([(1, 4, 3)], ['a', 'b', 'c']) >>> df.select(least(df.a, df.b, df.c).alias("least")).collect() [Row(least=1)] """ raise ValueError("least should take at least two columns")
"""Evaluates a list of conditions and returns one of multiple possible result expressions. If :func:`pyspark.sql.Column.otherwise` is not invoked, None is returned for unmatched conditions.
.. versionadded:: 1.4.0
Parameters ---------- condition : :class:`~pyspark.sql.Column` a boolean :class:`~pyspark.sql.Column` expression. value : a literal value, or a :class:`~pyspark.sql.Column` expression.
>>> df.select(when(df['age'] == 2, 3).otherwise(4).alias("age")).collect() [Row(age=3), Row(age=4)]
>>> df.select(when(df.age == 2, df.age + 1).alias("age")).collect() [Row(age=3), Row(age=None)] """ raise TypeError("condition should be a Column")
"""Returns the first argument-based logarithm of the second argument.
If there is only one argument, then this takes the natural logarithm of the argument.
.. versionadded:: 1.5.0
Examples -------- >>> df.select(log(10.0, df.age).alias('ten')).rdd.map(lambda l: str(l.ten)[:7]).collect() ['0.30102', '0.69897']
>>> df.select(log(df.age).alias('e')).rdd.map(lambda l: str(l.e)[:7]).collect() ['0.69314', '1.60943'] """ else:
"""Returns the base-2 logarithm of the argument.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([(4,)], ['a']).select(log2('a').alias('log2')).collect() [Row(log2=2.0)] """
""" Convert a number in a string column from one base to another.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([("010101",)], ['n']) >>> df.select(conv(df.n, 2, 16).alias('hex')).collect() [Row(hex='15')] """
""" Computes the factorial of the given value.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([(5,)], ['n']) >>> df.select(factorial(df.n).alias('f')).collect() [Row(f=120)] """
# --------------- Window functions ------------------------
""" Window function: returns the value that is `offset` rows before the current row, and `default` if there is less than `offset` rows before the current row. For example, an `offset` of one will return the previous row at any given point in the window partition.
This is equivalent to the LAG function in SQL.
.. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression offset : int, optional number of row to extend default : optional default value """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.lag(_to_java_column(col), offset, default))
""" Window function: returns the value that is `offset` rows after the current row, and `default` if there is less than `offset` rows after the current row. For example, an `offset` of one will return the next row at any given point in the window partition.
This is equivalent to the LEAD function in SQL.
.. versionadded:: 1.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression offset : int, optional number of row to extend default : optional default value """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.lead(_to_java_column(col), offset, default))
""" Window function: returns the value that is the `offset`\\th row of the window frame (counting from 1), and `null` if the size of window frame is less than `offset` rows.
It will return the `offset`\\th non-null value it sees when `ignoreNulls` is set to true. If all values are null, then null is returned.
This is equivalent to the nth_value function in SQL.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression offset : int, optional number of row to use as the value ignoreNulls : bool, optional indicates the Nth value should skip null in the determination of which row to use """
""" Window function: returns the ntile group id (from 1 to `n` inclusive) in an ordered window partition. For example, if `n` is 4, the first quarter of the rows will get value 1, the second quarter will get 2, the third quarter will get 3, and the last quarter will get 4.
This is equivalent to the NTILE function in SQL.
.. versionadded:: 1.4.0
Parameters ---------- n : int an integer """
# ---------------------- Date/Timestamp functions ------------------------------
def current_date(): """ Returns the current date at the start of query evaluation as a :class:`DateType` column. All calls of current_date within the same query return the same value. """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.current_date())
""" Returns the current timestamp at the start of query evaluation as a :class:`TimestampType` column. All calls of current_timestamp within the same query return the same value. """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.current_timestamp())
""" Converts a date/timestamp/string to a value of string in the format specified by the date format given by the second argument.
A pattern could be for instance `dd.MM.yyyy` and could return a string like '18.03.1993'. All pattern letters of `datetime pattern`_. can be used.
.. _datetime pattern: https://spark.apache.org/docs/latest/sql-ref-datetime-pattern.html
.. versionadded:: 1.5.0
Notes ----- Whenever possible, use specialized functions like `year`.
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(date_format('dt', 'MM/dd/yyy').alias('date')).collect() [Row(date='04/08/2015')] """
""" Extract the year of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(year('dt').alias('year')).collect() [Row(year=2015)] """
""" Extract the quarter of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(quarter('dt').alias('quarter')).collect() [Row(quarter=2)] """
""" Extract the month of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(month('dt').alias('month')).collect() [Row(month=4)] """
""" Extract the day of the week of a given date as integer. Ranges from 1 for a Sunday through to 7 for a Saturday
.. versionadded:: 2.3.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(dayofweek('dt').alias('day')).collect() [Row(day=4)] """
""" Extract the day of the month of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(dayofmonth('dt').alias('day')).collect() [Row(day=8)] """
""" Extract the day of the year of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(dayofyear('dt').alias('day')).collect() [Row(day=98)] """
""" Extract the hours of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08 13:08:15',)], ['ts']) >>> df.select(hour('ts').alias('hour')).collect() [Row(hour=13)] """
""" Extract the minutes of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08 13:08:15',)], ['ts']) >>> df.select(minute('ts').alias('minute')).collect() [Row(minute=8)] """
""" Extract the seconds of a given date as integer.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08 13:08:15',)], ['ts']) >>> df.select(second('ts').alias('second')).collect() [Row(second=15)] """
""" Extract the week number of a given date as integer. A week is considered to start on a Monday and week 1 is the first week with more than 3 days, as defined by ISO 8601
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(weekofyear(df.dt).alias('week')).collect() [Row(week=15)] """
""" Returns the date that is `days` days after `start`
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(date_add(df.dt, 1).alias('next_date')).collect() [Row(next_date=datetime.date(2015, 4, 9))] """
""" Returns the date that is `days` days before `start`
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(date_sub(df.dt, 1).alias('prev_date')).collect() [Row(prev_date=datetime.date(2015, 4, 7))] """
""" Returns the number of days from `start` to `end`.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08','2015-05-10')], ['d1', 'd2']) >>> df.select(datediff(df.d2, df.d1).alias('diff')).collect() [Row(diff=32)] """
""" Returns the date that is `months` months after `start`
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> df.select(add_months(df.dt, 1).alias('next_month')).collect() [Row(next_month=datetime.date(2015, 5, 8))] """
""" Returns number of months between dates date1 and date2. If date1 is later than date2, then the result is positive. A whole number is returned if both inputs have the same day of month or both are the last day of their respective months. Otherwise, the difference is calculated assuming 31 days per month. The result is rounded off to 8 digits unless `roundOff` is set to `False`.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('1997-02-28 10:30:00', '1996-10-30')], ['date1', 'date2']) >>> df.select(months_between(df.date1, df.date2).alias('months')).collect() [Row(months=3.94959677)] >>> df.select(months_between(df.date1, df.date2, False).alias('months')).collect() [Row(months=3.9495967741935485)] """ _to_java_column(date1), _to_java_column(date2), roundOff))
"""Converts a :class:`~pyspark.sql.Column` into :class:`pyspark.sql.types.DateType` using the optionally specified format. Specify formats according to `datetime pattern`_. By default, it follows casting rules to :class:`pyspark.sql.types.DateType` if the format is omitted. Equivalent to ``col.cast("date")``.
.. _datetime pattern: https://spark.apache.org/docs/latest/sql-ref-datetime-pattern.html
.. versionadded:: 2.2.0
Examples -------- >>> df = spark.createDataFrame([('1997-02-28 10:30:00',)], ['t']) >>> df.select(to_date(df.t).alias('date')).collect() [Row(date=datetime.date(1997, 2, 28))]
>>> df = spark.createDataFrame([('1997-02-28 10:30:00',)], ['t']) >>> df.select(to_date(df.t, 'yyyy-MM-dd HH:mm:ss').alias('date')).collect() [Row(date=datetime.date(1997, 2, 28))] """ else:
"""Converts a :class:`~pyspark.sql.Column` into :class:`pyspark.sql.types.TimestampType` using the optionally specified format. Specify formats according to `datetime pattern`_. By default, it follows casting rules to :class:`pyspark.sql.types.TimestampType` if the format is omitted. Equivalent to ``col.cast("timestamp")``.
.. _datetime pattern: https://spark.apache.org/docs/latest/sql-ref-datetime-pattern.html
.. versionadded:: 2.2.0
Examples -------- >>> df = spark.createDataFrame([('1997-02-28 10:30:00',)], ['t']) >>> df.select(to_timestamp(df.t).alias('dt')).collect() [Row(dt=datetime.datetime(1997, 2, 28, 10, 30))]
>>> df = spark.createDataFrame([('1997-02-28 10:30:00',)], ['t']) >>> df.select(to_timestamp(df.t, 'yyyy-MM-dd HH:mm:ss').alias('dt')).collect() [Row(dt=datetime.datetime(1997, 2, 28, 10, 30))] """ else:
""" Returns date truncated to the unit specified by the format.
.. versionadded:: 1.5.0
Parameters ---------- date : :class:`~pyspark.sql.Column` or str format : str 'year', 'yyyy', 'yy' to truncate by year, or 'month', 'mon', 'mm' to truncate by month Other options are: 'week', 'quarter'
Examples -------- >>> df = spark.createDataFrame([('1997-02-28',)], ['d']) >>> df.select(trunc(df.d, 'year').alias('year')).collect() [Row(year=datetime.date(1997, 1, 1))] >>> df.select(trunc(df.d, 'mon').alias('month')).collect() [Row(month=datetime.date(1997, 2, 1))] """
""" Returns timestamp truncated to the unit specified by the format.
.. versionadded:: 2.3.0
Parameters ---------- format : str 'year', 'yyyy', 'yy' to truncate by year, 'month', 'mon', 'mm' to truncate by month, 'day', 'dd' to truncate by day, Other options are: 'microsecond', 'millisecond', 'second', 'minute', 'hour', 'week', 'quarter' timestamp : :class:`~pyspark.sql.Column` or str
Examples -------- >>> df = spark.createDataFrame([('1997-02-28 05:02:11',)], ['t']) >>> df.select(date_trunc('year', df.t).alias('year')).collect() [Row(year=datetime.datetime(1997, 1, 1, 0, 0))] >>> df.select(date_trunc('mon', df.t).alias('month')).collect() [Row(month=datetime.datetime(1997, 2, 1, 0, 0))] """
""" Returns the first date which is later than the value of the date column.
Day of the week parameter is case insensitive, and accepts: "Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun".
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('2015-07-27',)], ['d']) >>> df.select(next_day(df.d, 'Sun').alias('date')).collect() [Row(date=datetime.date(2015, 8, 2))] """
""" Returns the last day of the month which the given date belongs to.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('1997-02-10',)], ['d']) >>> df.select(last_day(df.d).alias('date')).collect() [Row(date=datetime.date(1997, 2, 28))] """
""" Converts the number of seconds from unix epoch (1970-01-01 00:00:00 UTC) to a string representing the timestamp of that moment in the current system time zone in the given format.
.. versionadded:: 1.5.0
Examples -------- >>> spark.conf.set("spark.sql.session.timeZone", "America/Los_Angeles") >>> time_df = spark.createDataFrame([(1428476400,)], ['unix_time']) >>> time_df.select(from_unixtime('unix_time').alias('ts')).collect() [Row(ts='2015-04-08 00:00:00')] >>> spark.conf.unset("spark.sql.session.timeZone") """
""" Convert time string with given pattern ('yyyy-MM-dd HH:mm:ss', by default) to Unix time stamp (in seconds), using the default timezone and the default locale, return null if fail.
if `timestamp` is None, then it returns current timestamp.
.. versionadded:: 1.5.0
Examples -------- >>> spark.conf.set("spark.sql.session.timeZone", "America/Los_Angeles") >>> time_df = spark.createDataFrame([('2015-04-08',)], ['dt']) >>> time_df.select(unix_timestamp('dt', 'yyyy-MM-dd').alias('unix_time')).collect() [Row(unix_time=1428476400)] >>> spark.conf.unset("spark.sql.session.timeZone") """ return Column(sc._jvm.functions.unix_timestamp())
""" This is a common function for databases supporting TIMESTAMP WITHOUT TIMEZONE. This function takes a timestamp which is timezone-agnostic, and interprets it as a timestamp in UTC, and renders that timestamp as a timestamp in the given time zone.
However, timestamp in Spark represents number of microseconds from the Unix epoch, which is not timezone-agnostic. So in Spark this function just shift the timestamp value from UTC timezone to the given timezone.
This function may return confusing result if the input is a string with timezone, e.g. '2018-03-13T06:18:23+00:00'. The reason is that, Spark firstly cast the string to timestamp according to the timezone in the string, and finally display the result by converting the timestamp to string according to the session local timezone.
.. versionadded:: 1.5.0
Parameters ---------- timestamp : :class:`~pyspark.sql.Column` or str the column that contains timestamps tz : :class:`~pyspark.sql.Column` or str A string detailing the time zone ID that the input should be adjusted to. It should be in the format of either region-based zone IDs or zone offsets. Region IDs must have the form 'area/city', such as 'America/Los_Angeles'. Zone offsets must be in the format '(+|-)HH:mm', for example '-08:00' or '+01:00'. Also 'UTC' and 'Z' are supported as aliases of '+00:00'. Other short names are not recommended to use because they can be ambiguous.
.. versionchanged:: 2.4 `tz` can take a :class:`~pyspark.sql.Column` containing timezone ID strings.
Examples -------- >>> df = spark.createDataFrame([('1997-02-28 10:30:00', 'JST')], ['ts', 'tz']) >>> df.select(from_utc_timestamp(df.ts, "PST").alias('local_time')).collect() [Row(local_time=datetime.datetime(1997, 2, 28, 2, 30))] >>> df.select(from_utc_timestamp(df.ts, df.tz).alias('local_time')).collect() [Row(local_time=datetime.datetime(1997, 2, 28, 19, 30))] """
""" This is a common function for databases supporting TIMESTAMP WITHOUT TIMEZONE. This function takes a timestamp which is timezone-agnostic, and interprets it as a timestamp in the given timezone, and renders that timestamp as a timestamp in UTC.
However, timestamp in Spark represents number of microseconds from the Unix epoch, which is not timezone-agnostic. So in Spark this function just shift the timestamp value from the given timezone to UTC timezone.
This function may return confusing result if the input is a string with timezone, e.g. '2018-03-13T06:18:23+00:00'. The reason is that, Spark firstly cast the string to timestamp according to the timezone in the string, and finally display the result by converting the timestamp to string according to the session local timezone.
.. versionadded:: 1.5.0
Parameters ---------- timestamp : :class:`~pyspark.sql.Column` or str the column that contains timestamps tz : :class:`~pyspark.sql.Column` or str A string detailing the time zone ID that the input should be adjusted to. It should be in the format of either region-based zone IDs or zone offsets. Region IDs must have the form 'area/city', such as 'America/Los_Angeles'. Zone offsets must be in the format '(+|-)HH:mm', for example '-08:00' or '+01:00'. Also 'UTC' and 'Z' are upported as aliases of '+00:00'. Other short names are not recommended to use because they can be ambiguous.
.. versionchanged:: 2.4.0 `tz` can take a :class:`~pyspark.sql.Column` containing timezone ID strings.
Examples -------- >>> df = spark.createDataFrame([('1997-02-28 10:30:00', 'JST')], ['ts', 'tz']) >>> df.select(to_utc_timestamp(df.ts, "PST").alias('utc_time')).collect() [Row(utc_time=datetime.datetime(1997, 2, 28, 18, 30))] >>> df.select(to_utc_timestamp(df.ts, df.tz).alias('utc_time')).collect() [Row(utc_time=datetime.datetime(1997, 2, 28, 1, 30))] """
""" .. versionadded:: 3.1.0
Examples -------- >>> from pyspark.sql.functions import timestamp_seconds >>> spark.conf.set("spark.sql.session.timeZone", "America/Los_Angeles") >>> time_df = spark.createDataFrame([(1230219000,)], ['unix_time']) >>> time_df.select(timestamp_seconds(time_df.unix_time).alias('ts')).show() +-------------------+ | ts| +-------------------+ |2008-12-25 07:30:00| +-------------------+ >>> spark.conf.unset("spark.sql.session.timeZone") """
"""Bucketize rows into one or more time windows given a timestamp specifying column. Window starts are inclusive but the window ends are exclusive, e.g. 12:05 will be in the window [12:05,12:10) but not in [12:00,12:05). Windows can support microsecond precision. Windows in the order of months are not supported.
The time column must be of :class:`pyspark.sql.types.TimestampType`.
Durations are provided as strings, e.g. '1 second', '1 day 12 hours', '2 minutes'. Valid interval strings are 'week', 'day', 'hour', 'minute', 'second', 'millisecond', 'microsecond'. If the ``slideDuration`` is not provided, the windows will be tumbling windows.
The startTime is the offset with respect to 1970-01-01 00:00:00 UTC with which to start window intervals. For example, in order to have hourly tumbling windows that start 15 minutes past the hour, e.g. 12:15-13:15, 13:15-14:15... provide `startTime` as `15 minutes`.
The output column will be a struct called 'window' by default with the nested columns 'start' and 'end', where 'start' and 'end' will be of :class:`pyspark.sql.types.TimestampType`.
.. versionadded:: 2.0.0
Examples -------- >>> df = spark.createDataFrame([("2016-03-11 09:00:07", 1)]).toDF("date", "val") >>> w = df.groupBy(window("date", "5 seconds")).agg(sum("val").alias("sum")) >>> w.select(w.window.start.cast("string").alias("start"), ... w.window.end.cast("string").alias("end"), "sum").collect() [Row(start='2016-03-11 09:00:05', end='2016-03-11 09:00:10', sum=1)] """ raise TypeError("%s should be provided as a string" % fieldName)
check_string_field(slideDuration, "slideDuration") check_string_field(startTime, "startTime") res = sc._jvm.functions.window(time_col, windowDuration, slideDuration, startTime) check_string_field(slideDuration, "slideDuration") res = sc._jvm.functions.window(time_col, windowDuration, slideDuration) check_string_field(startTime, "startTime") res = sc._jvm.functions.window(time_col, windowDuration, windowDuration, startTime) else:
""" Generates session window given a timestamp specifying column. Session window is one of dynamic windows, which means the length of window is varying according to the given inputs. The length of session window is defined as "the timestamp of latest input of the session + gap duration", so when the new inputs are bound to the current session window, the end time of session window can be expanded according to the new inputs. Windows can support microsecond precision. Windows in the order of months are not supported. For a streaming query, you may use the function `current_timestamp` to generate windows on processing time. gapDuration is provided as strings, e.g. '1 second', '1 day 12 hours', '2 minutes'. Valid interval strings are 'week', 'day', 'hour', 'minute', 'second', 'millisecond', 'microsecond'. The output column will be a struct called 'session_window' by default with the nested columns 'start' and 'end', where 'start' and 'end' will be of :class:`pyspark.sql.types.TimestampType`. .. versionadded:: 3.2.0 Examples -------- >>> df = spark.createDataFrame([("2016-03-11 09:00:07", 1)]).toDF("date", "val") >>> w = df.groupBy(session_window("date", "5 seconds")).agg(sum("val").alias("sum")) >>> w.select(w.session_window.start.cast("string").alias("start"), ... w.session_window.end.cast("string").alias("end"), "sum").collect() [Row(start='2016-03-11 09:00:07', end='2016-03-11 09:00:12', sum=1)] """ raise TypeError("%s should be provided as a string" % fieldName)
# ---------------------------- misc functions ----------------------------------
""" Calculates the cyclic redundancy check value (CRC32) of a binary column and returns the value as a bigint.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('ABC',)], ['a']).select(crc32('a').alias('crc32')).collect() [Row(crc32=2743272264)] """
"""Calculates the MD5 digest and returns the value as a 32 character hex string.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('ABC',)], ['a']).select(md5('a').alias('hash')).collect() [Row(hash='902fbdd2b1df0c4f70b4a5d23525e932')] """
"""Returns the hex string result of SHA-1.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('ABC',)], ['a']).select(sha1('a').alias('hash')).collect() [Row(hash='3c01bdbb26f358bab27f267924aa2c9a03fcfdb8')] """
"""Returns the hex string result of SHA-2 family of hash functions (SHA-224, SHA-256, SHA-384, and SHA-512). The numBits indicates the desired bit length of the result, which must have a value of 224, 256, 384, 512, or 0 (which is equivalent to 256).
.. versionadded:: 1.5.0
Examples -------- >>> digests = df.select(sha2(df.name, 256).alias('s')).collect() >>> digests[0] Row(s='3bc51062973c458d5a6f2d8d64a023246354ad7e064b1e4e009ec8a0699a3043') >>> digests[1] Row(s='cd9fb1e148ccd8442e5aa74904cc73bf6fb54d1d54d333bd596aa9bb4bb4e961') """
"""Calculates the hash code of given columns, and returns the result as an int column.
.. versionadded:: 2.0.0
Examples -------- >>> spark.createDataFrame([('ABC',)], ['a']).select(hash('a').alias('hash')).collect() [Row(hash=-757602832)] """
"""Calculates the hash code of given columns using the 64-bit variant of the xxHash algorithm, and returns the result as a long column.
.. versionadded:: 3.0.0
Examples -------- >>> spark.createDataFrame([('ABC',)], ['a']).select(xxhash64('a').alias('hash')).collect() [Row(hash=4105715581806190027)] """
""" Returns null if the input column is true; throws an exception with the provided error message otherwise.
.. versionadded:: 3.1.0
Examples -------- >>> df = spark.createDataFrame([(0,1)], ['a', 'b']) >>> df.select(assert_true(df.a < df.b).alias('r')).collect() [Row(r=None)] >>> df = spark.createDataFrame([(0,1)], ['a', 'b']) >>> df.select(assert_true(df.a < df.b, df.a).alias('r')).collect() [Row(r=None)] >>> df = spark.createDataFrame([(0,1)], ['a', 'b']) >>> df.select(assert_true(df.a < df.b, 'error').alias('r')).collect() [Row(r=None)] """ "errMsg should be a Column or a str, got {}".format(type(errMsg)) )
_create_column_from_literal(errMsg) if isinstance(errMsg, str) else _to_java_column(errMsg) )
def raise_error(errMsg): """ Throws an exception with the provided error message. """ "errMsg should be a Column or a str, got {}".format(type(errMsg)) )
_create_column_from_literal(errMsg) if isinstance(errMsg, str) else _to_java_column(errMsg) )
# ---------------------- String/Binary functions ------------------------------
def upper(col): """ Converts a string expression to upper case. """
def lower(col): """ Converts a string expression to lower case. """
def ascii(col): """ Computes the numeric value of the first character of the string column. """
def base64(col): """ Computes the BASE64 encoding of a binary column and returns it as a string column. """
def unbase64(col): """ Decodes a BASE64 encoded string column and returns it as a binary column. """
def ltrim(col): """ Trim the spaces from left end for the specified string value. """
def rtrim(col): """ Trim the spaces from right end for the specified string value. """
def trim(col): """ Trim the spaces from both ends for the specified string column. """
""" Concatenates multiple input string columns together into a single string column, using the given separator.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('abcd','123')], ['s', 'd']) >>> df.select(concat_ws('-', df.s, df.d).alias('s')).collect() [Row(s='abcd-123')] """
def decode(col, charset): """ Computes the first argument into a string from a binary using the provided character set (one of 'US-ASCII', 'ISO-8859-1', 'UTF-8', 'UTF-16BE', 'UTF-16LE', 'UTF-16'). """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.decode(_to_java_column(col), charset))
def encode(col, charset): """ Computes the first argument into a binary from a string using the provided character set (one of 'US-ASCII', 'ISO-8859-1', 'UTF-8', 'UTF-16BE', 'UTF-16LE', 'UTF-16'). """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.encode(_to_java_column(col), charset))
""" Formats the number X to a format like '#,--#,--#.--', rounded to d decimal places with HALF_EVEN round mode, and returns the result as a string.
.. versionadded:: 1.5.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str the column name of the numeric value to be formatted d : int the N decimal places
>>> spark.createDataFrame([(5,)], ['a']).select(format_number('a', 4).alias('v')).collect() [Row(v='5.0000')] """
""" Formats the arguments in printf-style and returns the result as a string column.
.. versionadded:: 1.5.0
Parameters ---------- format : str string that can contain embedded format tags and used as result column's value cols : :class:`~pyspark.sql.Column` or str column names or :class:`~pyspark.sql.Column`\\s to be used in formatting
Examples -------- >>> df = spark.createDataFrame([(5, "hello")], ['a', 'b']) >>> df.select(format_string('%d %s', df.a, df.b).alias('v')).collect() [Row(v='5 hello')] """
""" Locate the position of the first occurrence of substr column in the given string. Returns null if either of the arguments are null.
.. versionadded:: 1.5.0
Notes ----- The position is not zero based, but 1 based index. Returns 0 if substr could not be found in str.
>>> df = spark.createDataFrame([('abcd',)], ['s',]) >>> df.select(instr(df.s, 'b').alias('s')).collect() [Row(s=2)] """
""" Overlay the specified portion of `src` with `replace`, starting from byte position `pos` of `src` and proceeding for `len` bytes.
.. versionadded:: 3.0.0
Examples -------- >>> df = spark.createDataFrame([("SPARK_SQL", "CORE")], ("x", "y")) >>> df.select(overlay("x", "y", 7).alias("overlayed")).show() +----------+ | overlayed| +----------+ |SPARK_CORE| +----------+ """ raise TypeError( "pos should be an integer or a Column / column name, got {}".format(type(pos))) raise TypeError( "len should be an integer or a Column / column name, got {}".format(type(len)))
_to_java_column(src), _to_java_column(replace), pos, len ))
""" Splits a string into arrays of sentences, where each sentence is an array of words. The 'language' and 'country' arguments are optional, and if omitted, the default locale is used.
.. versionadded:: 3.2.0
Parameters ---------- string : :class:`~pyspark.sql.Column` or str a string to be split language : :class:`~pyspark.sql.Column` or str, optional a language of the locale country : :class:`~pyspark.sql.Column` or str, optional a country of the locale
Examples -------- >>> df = spark.createDataFrame([["This is an example sentence."]], ["string"]) >>> df.select(sentences(df.string, lit("en"), lit("US"))).show(truncate=False) +-----------------------------------+ |sentences(string, en, US) | +-----------------------------------+ |[[This, is, an, example, sentence]]| +-----------------------------------+ """ language = lit("") country = lit("")
_to_java_column(string), _to_java_column(language), _to_java_column(country) ))
""" Substring starts at `pos` and is of length `len` when str is String type or returns the slice of byte array that starts at `pos` in byte and is of length `len` when str is Binary type.
.. versionadded:: 1.5.0
Notes ----- The position is not zero based, but 1 based index.
Examples -------- >>> df = spark.createDataFrame([('abcd',)], ['s',]) >>> df.select(substring(df.s, 1, 2).alias('s')).collect() [Row(s='ab')] """
""" Returns the substring from string str before count occurrences of the delimiter delim. If count is positive, everything the left of the final delimiter (counting from left) is returned. If count is negative, every to the right of the final delimiter (counting from the right) is returned. substring_index performs a case-sensitive match when searching for delim.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('a.b.c.d',)], ['s']) >>> df.select(substring_index(df.s, '.', 2).alias('s')).collect() [Row(s='a.b')] >>> df.select(substring_index(df.s, '.', -3).alias('s')).collect() [Row(s='b.c.d')] """
"""Computes the Levenshtein distance of the two given strings.
.. versionadded:: 1.5.0
Examples -------- >>> df0 = spark.createDataFrame([('kitten', 'sitting',)], ['l', 'r']) >>> df0.select(levenshtein('l', 'r').alias('d')).collect() [Row(d=3)] """
""" Locate the position of the first occurrence of substr in a string column, after position pos.
.. versionadded:: 1.5.0
Parameters ---------- substr : str a string str : :class:`~pyspark.sql.Column` or str a Column of :class:`pyspark.sql.types.StringType` pos : int, optional start position (zero based)
Notes ----- The position is not zero based, but 1 based index. Returns 0 if substr could not be found in str.
Examples -------- >>> df = spark.createDataFrame([('abcd',)], ['s',]) >>> df.select(locate('b', df.s, 1).alias('s')).collect() [Row(s=2)] """
""" Left-pad the string column to width `len` with `pad`.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('abcd',)], ['s',]) >>> df.select(lpad(df.s, 6, '#').alias('s')).collect() [Row(s='##abcd')] """
""" Right-pad the string column to width `len` with `pad`.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('abcd',)], ['s',]) >>> df.select(rpad(df.s, 6, '#').alias('s')).collect() [Row(s='abcd##')] """
""" Repeats a string column n times, and returns it as a new string column.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('ab',)], ['s',]) >>> df.select(repeat(df.s, 3).alias('s')).collect() [Row(s='ababab')] """
""" Splits str around matches of the given pattern.
.. versionadded:: 1.5.0
Parameters ---------- str : :class:`~pyspark.sql.Column` or str a string expression to split pattern : str a string representing a regular expression. The regex string should be a Java regular expression. limit : int, optional an integer which controls the number of times `pattern` is applied.
* ``limit > 0``: The resulting array's length will not be more than `limit`, and the resulting array's last entry will contain all input beyond the last matched pattern. * ``limit <= 0``: `pattern` will be applied as many times as possible, and the resulting array can be of any size.
.. versionchanged:: 3.0 `split` now takes an optional `limit` field. If not provided, default limit value is -1.
Examples -------- >>> df = spark.createDataFrame([('oneAtwoBthreeC',)], ['s',]) >>> df.select(split(df.s, '[ABC]', 2).alias('s')).collect() [Row(s=['one', 'twoBthreeC'])] >>> df.select(split(df.s, '[ABC]', -1).alias('s')).collect() [Row(s=['one', 'two', 'three', ''])] """
r"""Extract a specific group matched by a Java regex, from the specified string column. If the regex did not match, or the specified group did not match, an empty string is returned.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('100-200',)], ['str']) >>> df.select(regexp_extract('str', r'(\d+)-(\d+)', 1).alias('d')).collect() [Row(d='100')] >>> df = spark.createDataFrame([('foo',)], ['str']) >>> df.select(regexp_extract('str', r'(\d+)', 1).alias('d')).collect() [Row(d='')] >>> df = spark.createDataFrame([('aaaac',)], ['str']) >>> df.select(regexp_extract('str', '(a+)(b)?(c)', 2).alias('d')).collect() [Row(d='')] """
r"""Replace all substrings of the specified string value that match regexp with rep.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('100-200',)], ['str']) >>> df.select(regexp_replace('str', r'(\d+)', '--').alias('d')).collect() [Row(d='-----')] """
"""Translate the first letter of each word to upper case in the sentence.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('ab cd',)], ['a']).select(initcap("a").alias('v')).collect() [Row(v='Ab Cd')] """
""" Returns the SoundEx encoding for a string
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([("Peters",),("Uhrbach",)], ['name']) >>> df.select(soundex(df.name).alias("soundex")).collect() [Row(soundex='P362'), Row(soundex='U612')] """
"""Returns the string representation of the binary value of the given column.
.. versionadded:: 1.5.0
Examples -------- >>> df.select(bin(df.age).alias('c')).collect() [Row(c='10'), Row(c='101')] """
"""Computes hex value of the given column, which could be :class:`pyspark.sql.types.StringType`, :class:`pyspark.sql.types.BinaryType`, :class:`pyspark.sql.types.IntegerType` or :class:`pyspark.sql.types.LongType`.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('ABC', 3)], ['a', 'b']).select(hex('a'), hex('b')).collect() [Row(hex(a)='414243', hex(b)='3')] """
"""Inverse of hex. Interprets each pair of characters as a hexadecimal number and converts to the byte representation of number.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('414243',)], ['a']).select(unhex('a')).collect() [Row(unhex(a)=bytearray(b'ABC'))] """
"""Computes the character length of string data or number of bytes of binary data. The length of character data includes the trailing spaces. The length of binary data includes binary zeros.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('ABC ',)], ['a']).select(length('a').alias('length')).collect() [Row(length=4)] """
"""A function translate any character in the `srcCol` by a character in `matching`. The characters in `replace` is corresponding to the characters in `matching`. The translate will happen when any character in the string matching with the character in the `matching`.
.. versionadded:: 1.5.0
Examples -------- >>> spark.createDataFrame([('translate',)], ['a']).select(translate('a', "rnlt", "123") \\ ... .alias('r')).collect() [Row(r='1a2s3ae')] """
# ---------------------- Collection functions ------------------------------
"""Creates a new map column.
.. versionadded:: 2.0.0
Parameters ---------- cols : :class:`~pyspark.sql.Column` or str column names or :class:`~pyspark.sql.Column`\\s that are grouped as key-value pairs, e.g. (key1, value1, key2, value2, ...).
Examples -------- >>> df.select(create_map('name', 'age').alias("map")).collect() [Row(map={'Alice': 2}), Row(map={'Bob': 5})] >>> df.select(create_map([df.name, df.age]).alias("map")).collect() [Row(map={'Alice': 2}), Row(map={'Bob': 5})] """
"""Creates a new map from two arrays.
.. versionadded:: 2.4.0
Parameters ---------- col1 : :class:`~pyspark.sql.Column` or str name of column containing a set of keys. All elements should not be null col2 : :class:`~pyspark.sql.Column` or str name of column containing a set of values
Examples -------- >>> df = spark.createDataFrame([([2, 5], ['a', 'b'])], ['k', 'v']) >>> df.select(map_from_arrays(df.k, df.v).alias("map")).show() +----------------+ | map| +----------------+ |{2 -> a, 5 -> b}| +----------------+ """
"""Creates a new array column.
.. versionadded:: 1.4.0
Parameters ---------- cols : :class:`~pyspark.sql.Column` or str column names or :class:`~pyspark.sql.Column`\\s that have the same data type.
Examples -------- >>> df.select(array('age', 'age').alias("arr")).collect() [Row(arr=[2, 2]), Row(arr=[5, 5])] >>> df.select(array([df.age, df.age]).alias("arr")).collect() [Row(arr=[2, 2]), Row(arr=[5, 5])] """
""" Collection function: returns null if the array is null, true if the array contains the given value, and false otherwise.
.. versionadded:: 1.5.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column containing array value : value or column to check for in array
Examples -------- >>> df = spark.createDataFrame([(["a", "b", "c"],), ([],)], ['data']) >>> df.select(array_contains(df.data, "a")).collect() [Row(array_contains(data, a)=True), Row(array_contains(data, a)=False)] >>> df.select(array_contains(df.data, lit("a"))).collect() [Row(array_contains(data, a)=True), Row(array_contains(data, a)=False)] """
""" Collection function: returns true if the arrays contain any common non-null element; if not, returns null if both the arrays are non-empty and any of them contains a null element; returns false otherwise.
.. versionadded:: 2.4.0
Examples -------- >>> df = spark.createDataFrame([(["a", "b"], ["b", "c"]), (["a"], ["b", "c"])], ['x', 'y']) >>> df.select(arrays_overlap(df.x, df.y).alias("overlap")).collect() [Row(overlap=True), Row(overlap=False)] """
""" Collection function: returns an array containing all the elements in `x` from index `start` (array indices start at 1, or from the end if `start` is negative) with the specified `length`.
.. versionadded:: 2.4.0
Parameters ---------- x : :class:`~pyspark.sql.Column` or str the array to be sliced start : :class:`~pyspark.sql.Column` or int the starting index length : :class:`~pyspark.sql.Column` or int the length of the slice
Examples -------- >>> df = spark.createDataFrame([([1, 2, 3],), ([4, 5],)], ['x']) >>> df.select(slice(df.x, 2, 2).alias("sliced")).collect() [Row(sliced=[2, 3]), Row(sliced=[5])] """ _to_java_column(x), start._jc if isinstance(start, Column) else start, length._jc if isinstance(length, Column) else length ))
""" Concatenates the elements of `column` using the `delimiter`. Null values are replaced with `null_replacement` if set, otherwise they are ignored.
.. versionadded:: 2.4.0
Examples -------- >>> df = spark.createDataFrame([(["a", "b", "c"],), (["a", None],)], ['data']) >>> df.select(array_join(df.data, ",").alias("joined")).collect() [Row(joined='a,b,c'), Row(joined='a')] >>> df.select(array_join(df.data, ",", "NULL").alias("joined")).collect() [Row(joined='a,b,c'), Row(joined='a,NULL')] """ else: _to_java_column(col), delimiter, null_replacement))
""" Concatenates multiple input columns together into a single column. The function works with strings, binary and compatible array columns.
.. versionadded:: 1.5.0
Examples -------- >>> df = spark.createDataFrame([('abcd','123')], ['s', 'd']) >>> df.select(concat(df.s, df.d).alias('s')).collect() [Row(s='abcd123')]
>>> df = spark.createDataFrame([([1, 2], [3, 4], [5]), ([1, 2], None, [3])], ['a', 'b', 'c']) >>> df.select(concat(df.a, df.b, df.c).alias("arr")).collect() [Row(arr=[1, 2, 3, 4, 5]), Row(arr=None)] """
""" Collection function: Locates the position of the first occurrence of the given value in the given array. Returns null if either of the arguments are null.
.. versionadded:: 2.4.0
Notes ----- The position is not zero based, but 1 based index. Returns 0 if the given value could not be found in the array.
Examples -------- >>> df = spark.createDataFrame([(["c", "b", "a"],), ([],)], ['data']) >>> df.select(array_position(df.data, "a")).collect() [Row(array_position(data, a)=3), Row(array_position(data, a)=0)] """
""" Collection function: Returns element of array at given index in extraction if col is array. Returns value for the given key in extraction if col is map.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column containing array or map extraction : index to check for in array or key to check for in map
Notes ----- The position is not zero based, but 1 based index.
Examples -------- >>> df = spark.createDataFrame([(["a", "b", "c"],), ([],)], ['data']) >>> df.select(element_at(df.data, 1)).collect() [Row(element_at(data, 1)='a'), Row(element_at(data, 1)=None)]
>>> df = spark.createDataFrame([({"a": 1.0, "b": 2.0},), ({},)], ['data']) >>> df.select(element_at(df.data, lit("a"))).collect() [Row(element_at(data, a)=1.0), Row(element_at(data, a)=None)] """ _to_java_column(col), lit(extraction)._jc))
""" Collection function: Remove all elements that equal to element from the given array.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column containing array element : element to be removed from the array
Examples -------- >>> df = spark.createDataFrame([([1, 2, 3, 1, 1],), ([],)], ['data']) >>> df.select(array_remove(df.data, 1)).collect() [Row(array_remove(data, 1)=[2, 3]), Row(array_remove(data, 1)=[])] """
""" Collection function: removes duplicate values from the array.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([([1, 2, 3, 2],), ([4, 5, 5, 4],)], ['data']) >>> df.select(array_distinct(df.data)).collect() [Row(array_distinct(data)=[1, 2, 3]), Row(array_distinct(data)=[4, 5])] """
""" Collection function: returns an array of the elements in the intersection of col1 and col2, without duplicates.
.. versionadded:: 2.4.0
Parameters ---------- col1 : :class:`~pyspark.sql.Column` or str name of column containing array col2 : :class:`~pyspark.sql.Column` or str name of column containing array
Examples -------- >>> from pyspark.sql import Row >>> df = spark.createDataFrame([Row(c1=["b", "a", "c"], c2=["c", "d", "a", "f"])]) >>> df.select(array_intersect(df.c1, df.c2)).collect() [Row(array_intersect(c1, c2)=['a', 'c'])] """
""" Collection function: returns an array of the elements in the union of col1 and col2, without duplicates.
.. versionadded:: 2.4.0
Parameters ---------- col1 : :class:`~pyspark.sql.Column` or str name of column containing array col2 : :class:`~pyspark.sql.Column` or str name of column containing array
Examples -------- >>> from pyspark.sql import Row >>> df = spark.createDataFrame([Row(c1=["b", "a", "c"], c2=["c", "d", "a", "f"])]) >>> df.select(array_union(df.c1, df.c2)).collect() [Row(array_union(c1, c2)=['b', 'a', 'c', 'd', 'f'])] """
""" Collection function: returns an array of the elements in col1 but not in col2, without duplicates.
.. versionadded:: 2.4.0
Parameters ---------- col1 : :class:`~pyspark.sql.Column` or str name of column containing array col2 : :class:`~pyspark.sql.Column` or str name of column containing array
Examples -------- >>> from pyspark.sql import Row >>> df = spark.createDataFrame([Row(c1=["b", "a", "c"], c2=["c", "d", "a", "f"])]) >>> df.select(array_except(df.c1, df.c2)).collect() [Row(array_except(c1, c2)=['b'])] """
""" Returns a new row for each element in the given array or map. Uses the default column name `col` for elements in the array and `key` and `value` for elements in the map unless specified otherwise.
.. versionadded:: 1.4.0
Examples -------- >>> from pyspark.sql import Row >>> eDF = spark.createDataFrame([Row(a=1, intlist=[1,2,3], mapfield={"a": "b"})]) >>> eDF.select(explode(eDF.intlist).alias("anInt")).collect() [Row(anInt=1), Row(anInt=2), Row(anInt=3)]
>>> eDF.select(explode(eDF.mapfield).alias("key", "value")).show() +---+-----+ |key|value| +---+-----+ | a| b| +---+-----+ """
""" Returns a new row for each element with position in the given array or map. Uses the default column name `pos` for position, and `col` for elements in the array and `key` and `value` for elements in the map unless specified otherwise.
.. versionadded:: 2.1.0
Examples -------- >>> from pyspark.sql import Row >>> eDF = spark.createDataFrame([Row(a=1, intlist=[1,2,3], mapfield={"a": "b"})]) >>> eDF.select(posexplode(eDF.intlist)).collect() [Row(pos=0, col=1), Row(pos=1, col=2), Row(pos=2, col=3)]
>>> eDF.select(posexplode(eDF.mapfield)).show() +---+---+-----+ |pos|key|value| +---+---+-----+ | 0| a| b| +---+---+-----+ """
""" Returns a new row for each element in the given array or map. Unlike explode, if the array/map is null or empty then null is produced. Uses the default column name `col` for elements in the array and `key` and `value` for elements in the map unless specified otherwise.
.. versionadded:: 2.3.0
Examples -------- >>> df = spark.createDataFrame( ... [(1, ["foo", "bar"], {"x": 1.0}), (2, [], {}), (3, None, None)], ... ("id", "an_array", "a_map") ... ) >>> df.select("id", "an_array", explode_outer("a_map")).show() +---+----------+----+-----+ | id| an_array| key|value| +---+----------+----+-----+ | 1|[foo, bar]| x| 1.0| | 2| []|null| null| | 3| null|null| null| +---+----------+----+-----+
>>> df.select("id", "a_map", explode_outer("an_array")).show() +---+----------+----+ | id| a_map| col| +---+----------+----+ | 1|{x -> 1.0}| foo| | 1|{x -> 1.0}| bar| | 2| {}|null| | 3| null|null| +---+----------+----+ """
""" Returns a new row for each element with position in the given array or map. Unlike posexplode, if the array/map is null or empty then the row (null, null) is produced. Uses the default column name `pos` for position, and `col` for elements in the array and `key` and `value` for elements in the map unless specified otherwise.
.. versionadded:: 2.3.0
Examples -------- >>> df = spark.createDataFrame( ... [(1, ["foo", "bar"], {"x": 1.0}), (2, [], {}), (3, None, None)], ... ("id", "an_array", "a_map") ... ) >>> df.select("id", "an_array", posexplode_outer("a_map")).show() +---+----------+----+----+-----+ | id| an_array| pos| key|value| +---+----------+----+----+-----+ | 1|[foo, bar]| 0| x| 1.0| | 2| []|null|null| null| | 3| null|null|null| null| +---+----------+----+----+-----+ >>> df.select("id", "a_map", posexplode_outer("an_array")).show() +---+----------+----+----+ | id| a_map| pos| col| +---+----------+----+----+ | 1|{x -> 1.0}| 0| foo| | 1|{x -> 1.0}| 1| bar| | 2| {}|null|null| | 3| null|null|null| +---+----------+----+----+ """
""" Extracts json object from a json string based on json path specified, and returns json string of the extracted json object. It will return null if the input json string is invalid.
.. versionadded:: 1.6.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str string column in json format path : str path to the json object to extract
Examples -------- >>> data = [("1", '''{"f1": "value1", "f2": "value2"}'''), ("2", '''{"f1": "value12"}''')] >>> df = spark.createDataFrame(data, ("key", "jstring")) >>> df.select(df.key, get_json_object(df.jstring, '$.f1').alias("c0"), \\ ... get_json_object(df.jstring, '$.f2').alias("c1") ).collect() [Row(key='1', c0='value1', c1='value2'), Row(key='2', c0='value12', c1=None)] """
"""Creates a new row for a json column according to the given field names.
.. versionadded:: 1.6.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str string column in json format fields : str fields to extract
Examples -------- >>> data = [("1", '''{"f1": "value1", "f2": "value2"}'''), ("2", '''{"f1": "value12"}''')] >>> df = spark.createDataFrame(data, ("key", "jstring")) >>> df.select(df.key, json_tuple(df.jstring, 'f1', 'f2')).collect() [Row(key='1', c0='value1', c1='value2'), Row(key='2', c0='value12', c1=None)] """
""" Parses a column containing a JSON string into a :class:`MapType` with :class:`StringType` as keys type, :class:`StructType` or :class:`ArrayType` with the specified schema. Returns `null`, in the case of an unparseable string.
.. versionadded:: 2.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str string column in json format schema : :class:`DataType` or str a StructType or ArrayType of StructType to use when parsing the json column.
.. versionchanged:: 2.3 the DDL-formatted string is also supported for ``schema``. options : dict, optional options to control parsing. accepts the same options as the json datasource. See `Data Source Option <https://spark.apache.org/docs/latest/sql-data-sources-json.html#data-source-option>`_ in the version you use.
.. # noqa
Examples -------- >>> from pyspark.sql.types import * >>> data = [(1, '''{"a": 1}''')] >>> schema = StructType([StructField("a", IntegerType())]) >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(from_json(df.value, schema).alias("json")).collect() [Row(json=Row(a=1))] >>> df.select(from_json(df.value, "a INT").alias("json")).collect() [Row(json=Row(a=1))] >>> df.select(from_json(df.value, "MAP<STRING,INT>").alias("json")).collect() [Row(json={'a': 1})] >>> data = [(1, '''[{"a": 1}]''')] >>> schema = ArrayType(StructType([StructField("a", IntegerType())])) >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(from_json(df.value, schema).alias("json")).collect() [Row(json=[Row(a=1)])] >>> schema = schema_of_json(lit('''{"a": 0}''')) >>> df.select(from_json(df.value, schema).alias("json")).collect() [Row(json=Row(a=None))] >>> data = [(1, '''[1, 2, 3]''')] >>> schema = ArrayType(IntegerType()) >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(from_json(df.value, schema).alias("json")).collect() [Row(json=[1, 2, 3])] """
""" Converts a column containing a :class:`StructType`, :class:`ArrayType` or a :class:`MapType` into a JSON string. Throws an exception, in the case of an unsupported type.
.. versionadded:: 2.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column containing a struct, an array or a map. options : dict, optional options to control converting. accepts the same options as the JSON datasource. See `Data Source Option <https://spark.apache.org/docs/latest/sql-data-sources-json.html#data-source-option>`_ in the version you use. Additionally the function supports the `pretty` option which enables pretty JSON generation.
.. # noqa
Examples -------- >>> from pyspark.sql import Row >>> from pyspark.sql.types import * >>> data = [(1, Row(age=2, name='Alice'))] >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(to_json(df.value).alias("json")).collect() [Row(json='{"age":2,"name":"Alice"}')] >>> data = [(1, [Row(age=2, name='Alice'), Row(age=3, name='Bob')])] >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(to_json(df.value).alias("json")).collect() [Row(json='[{"age":2,"name":"Alice"},{"age":3,"name":"Bob"}]')] >>> data = [(1, {"name": "Alice"})] >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(to_json(df.value).alias("json")).collect() [Row(json='{"name":"Alice"}')] >>> data = [(1, [{"name": "Alice"}, {"name": "Bob"}])] >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(to_json(df.value).alias("json")).collect() [Row(json='[{"name":"Alice"},{"name":"Bob"}]')] >>> data = [(1, ["Alice", "Bob"])] >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(to_json(df.value).alias("json")).collect() [Row(json='["Alice","Bob"]')] """
""" Parses a JSON string and infers its schema in DDL format.
.. versionadded:: 2.4.0
Parameters ---------- json : :class:`~pyspark.sql.Column` or str a JSON string or a foldable string column containing a JSON string. options : dict, optional options to control parsing. accepts the same options as the JSON datasource. See `Data Source Option <https://spark.apache.org/docs/latest/sql-data-sources-json.html#data-source-option>`_ in the version you use.
.. # noqa
.. versionchanged:: 3.0 It accepts `options` parameter to control schema inferring.
Examples -------- >>> df = spark.range(1) >>> df.select(schema_of_json(lit('{"a": 0}')).alias("json")).collect() [Row(json='STRUCT<`a`: BIGINT>')] >>> schema = schema_of_json('{a: 1}', {'allowUnquotedFieldNames':'true'}) >>> df.select(schema.alias("json")).collect() [Row(json='STRUCT<`a`: BIGINT>')] """ else: raise TypeError("schema argument should be a column or string")
""" Parses a CSV string and infers its schema in DDL format.
.. versionadded:: 3.0.0
Parameters ---------- csv : :class:`~pyspark.sql.Column` or str a CSV string or a foldable string column containing a CSV string. options : dict, optional options to control parsing. accepts the same options as the CSV datasource. See `Data Source Option <https://spark.apache.org/docs/latest/sql-data-sources-csv.html#data-source-option>`_ in the version you use.
.. # noqa
Examples -------- >>> df = spark.range(1) >>> df.select(schema_of_csv(lit('1|a'), {'sep':'|'}).alias("csv")).collect() [Row(csv='STRUCT<`_c0`: INT, `_c1`: STRING>')] >>> df.select(schema_of_csv('1|a', {'sep':'|'}).alias("csv")).collect() [Row(csv='STRUCT<`_c0`: INT, `_c1`: STRING>')] """ else: raise TypeError("schema argument should be a column or string")
""" Converts a column containing a :class:`StructType` into a CSV string. Throws an exception, in the case of an unsupported type.
.. versionadded:: 3.0.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column containing a struct. options: dict, optional options to control converting. accepts the same options as the CSV datasource. See `Data Source Option <https://spark.apache.org/docs/latest/sql-data-sources-csv.html#data-source-option>`_ in the version you use.
.. # noqa
Examples -------- >>> from pyspark.sql import Row >>> data = [(1, Row(age=2, name='Alice'))] >>> df = spark.createDataFrame(data, ("key", "value")) >>> df.select(to_csv(df.value).alias("csv")).collect() [Row(csv='2,Alice')] """
""" Collection function: returns the length of the array or map stored in the column.
.. versionadded:: 1.5.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([([1, 2, 3],),([1],),([],)], ['data']) >>> df.select(size(df.data)).collect() [Row(size(data)=3), Row(size(data)=1), Row(size(data)=0)] """
""" Collection function: returns the minimum value of the array.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([([2, 1, 3],), ([None, 10, -1],)], ['data']) >>> df.select(array_min(df.data).alias('min')).collect() [Row(min=1), Row(min=-1)] """
""" Collection function: returns the maximum value of the array.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([([2, 1, 3],), ([None, 10, -1],)], ['data']) >>> df.select(array_max(df.data).alias('max')).collect() [Row(max=3), Row(max=10)] """
""" Collection function: sorts the input array in ascending or descending order according to the natural ordering of the array elements. Null elements will be placed at the beginning of the returned array in ascending order or at the end of the returned array in descending order.
.. versionadded:: 1.5.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression asc : bool, optional
Examples -------- >>> df = spark.createDataFrame([([2, 1, None, 3],),([1],),([],)], ['data']) >>> df.select(sort_array(df.data).alias('r')).collect() [Row(r=[None, 1, 2, 3]), Row(r=[1]), Row(r=[])] >>> df.select(sort_array(df.data, asc=False).alias('r')).collect() [Row(r=[3, 2, 1, None]), Row(r=[1]), Row(r=[])] """
""" Collection function: sorts the input array in ascending order. The elements of the input array must be orderable. Null elements will be placed at the end of the returned array.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([([2, 1, None, 3],),([1],),([],)], ['data']) >>> df.select(array_sort(df.data).alias('r')).collect() [Row(r=[1, 2, 3, None]), Row(r=[1]), Row(r=[])] """
""" Collection function: Generates a random permutation of the given array.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Notes ----- The function is non-deterministic.
Examples -------- >>> df = spark.createDataFrame([([1, 20, 3, 5],), ([1, 20, None, 3],)], ['data']) >>> df.select(shuffle(df.data).alias('s')).collect() # doctest: +SKIP [Row(s=[3, 1, 5, 20]), Row(s=[20, None, 3, 1])] """ sc = SparkContext._active_spark_context return Column(sc._jvm.functions.shuffle(_to_java_column(col)))
""" Collection function: returns a reversed string or an array with reverse order of elements.
.. versionadded:: 1.5.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([('Spark SQL',)], ['data']) >>> df.select(reverse(df.data).alias('s')).collect() [Row(s='LQS krapS')] >>> df = spark.createDataFrame([([2, 1, 3],) ,([1],) ,([],)], ['data']) >>> df.select(reverse(df.data).alias('r')).collect() [Row(r=[3, 1, 2]), Row(r=[1]), Row(r=[])] """
""" Collection function: creates a single array from an array of arrays. If a structure of nested arrays is deeper than two levels, only one level of nesting is removed.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> df = spark.createDataFrame([([[1, 2, 3], [4, 5], [6]],), ([None, [4, 5]],)], ['data']) >>> df.select(flatten(df.data).alias('r')).collect() [Row(r=[1, 2, 3, 4, 5, 6]), Row(r=None)] """
""" Collection function: Returns an unordered array containing the keys of the map.
.. versionadded:: 2.3.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> from pyspark.sql.functions import map_keys >>> df = spark.sql("SELECT map(1, 'a', 2, 'b') as data") >>> df.select(map_keys("data").alias("keys")).show() +------+ | keys| +------+ |[1, 2]| +------+ """
""" Collection function: Returns an unordered array containing the values of the map.
.. versionadded:: 2.3.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> from pyspark.sql.functions import map_values >>> df = spark.sql("SELECT map(1, 'a', 2, 'b') as data") >>> df.select(map_values("data").alias("values")).show() +------+ |values| +------+ |[a, b]| +------+ """
""" Collection function: Returns an unordered array of all entries in the given map.
.. versionadded:: 3.0.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> from pyspark.sql.functions import map_entries >>> df = spark.sql("SELECT map(1, 'a', 2, 'b') as data") >>> df.select(map_entries("data").alias("entries")).show() +----------------+ | entries| +----------------+ |[{1, a}, {2, b}]| +----------------+ """
""" Collection function: Returns a map created from the given array of entries.
.. versionadded:: 2.4.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression
Examples -------- >>> from pyspark.sql.functions import map_from_entries >>> df = spark.sql("SELECT array(struct(1, 'a'), struct(2, 'b')) as data") >>> df.select(map_from_entries("data").alias("map")).show() +----------------+ | map| +----------------+ |{1 -> a, 2 -> b}| +----------------+ """
""" Collection function: creates an array containing a column repeated count times.
.. versionadded:: 2.4.0
Examples -------- >>> df = spark.createDataFrame([('ab',)], ['data']) >>> df.select(array_repeat(df.data, 3).alias('r')).collect() [Row(r=['ab', 'ab', 'ab'])] """ _to_java_column(col), _to_java_column(count) if isinstance(count, Column) else count ))
""" Collection function: Returns a merged array of structs in which the N-th struct contains all N-th values of input arrays.
.. versionadded:: 2.4.0
Parameters ---------- cols : :class:`~pyspark.sql.Column` or str columns of arrays to be merged.
Examples -------- >>> from pyspark.sql.functions import arrays_zip >>> df = spark.createDataFrame([(([1, 2, 3], [2, 3, 4]))], ['vals1', 'vals2']) >>> df.select(arrays_zip(df.vals1, df.vals2).alias('zipped')).collect() [Row(zipped=[Row(vals1=1, vals2=2), Row(vals1=2, vals2=3), Row(vals1=3, vals2=4)])] """
"""Returns the union of all the given maps.
.. versionadded:: 2.4.0
Parameters ---------- cols : :class:`~pyspark.sql.Column` or str column names or :class:`~pyspark.sql.Column`\\s
Examples -------- >>> from pyspark.sql.functions import map_concat >>> df = spark.sql("SELECT map(1, 'a', 2, 'b') as map1, map(3, 'c') as map2") >>> df.select(map_concat("map1", "map2").alias("map3")).show(truncate=False) +------------------------+ |map3 | +------------------------+ |{1 -> a, 2 -> b, 3 -> c}| +------------------------+ """ cols = cols[0]
""" Generate a sequence of integers from `start` to `stop`, incrementing by `step`. If `step` is not set, incrementing by 1 if `start` is less than or equal to `stop`, otherwise -1.
.. versionadded:: 2.4.0
Examples -------- >>> df1 = spark.createDataFrame([(-2, 2)], ('C1', 'C2')) >>> df1.select(sequence('C1', 'C2').alias('r')).collect() [Row(r=[-2, -1, 0, 1, 2])] >>> df2 = spark.createDataFrame([(4, -4, -2)], ('C1', 'C2', 'C3')) >>> df2.select(sequence('C1', 'C2', 'C3').alias('r')).collect() [Row(r=[4, 2, 0, -2, -4])] """ else: _to_java_column(start), _to_java_column(stop), _to_java_column(step)))
""" Parses a column containing a CSV string to a row with the specified schema. Returns `null`, in the case of an unparseable string.
.. versionadded:: 3.0.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str string column in CSV format schema :class:`~pyspark.sql.Column` or str a string with schema in DDL format to use when parsing the CSV column. options : dict, optional options to control parsing. accepts the same options as the CSV datasource. See `Data Source Option <https://spark.apache.org/docs/latest/sql-data-sources-csv.html#data-source-option>`_ in the version you use.
.. # noqa
Examples -------- >>> data = [("1,2,3",)] >>> df = spark.createDataFrame(data, ("value",)) >>> df.select(from_csv(df.value, "a INT, b INT, c INT").alias("csv")).collect() [Row(csv=Row(a=1, b=2, c=3))] >>> value = data[0][0] >>> df.select(from_csv(df.value, schema_of_csv(value)).alias("csv")).collect() [Row(csv=Row(_c0=1, _c1=2, _c2=3))] >>> data = [(" abc",)] >>> df = spark.createDataFrame(data, ("value",)) >>> options = {'ignoreLeadingWhiteSpace': True} >>> df.select(from_csv(df.value, "s string", options).alias("csv")).collect() [Row(csv=Row(s='abc'))] """
else: raise TypeError("schema argument should be a column or string")
""" Create `o.a.s.sql.expressions.UnresolvedNamedLambdaVariable`, convert it to o.s.sql.Column and wrap in Python `Column`
Parameters ---------- name_parts : str """ sc._jvm.Column( expressions.UnresolvedNamedLambdaVariable(name_parts_seq) ) )
# We should exclude functions that use # variable args and keyword argnames # as well as keyword only args inspect.Parameter.POSITIONAL_OR_KEYWORD, inspect.Parameter.POSITIONAL_ONLY, }
# Validate that # function arity is between 1 and 3 "f should take between 1 and 3 arguments, but provided function takes {}".format( len(parameters) ) )
# and all arguments can be used as positional "f should use only POSITIONAL or POSITIONAL OR KEYWORD arguments" )
""" Create `o.a.s.sql.expressions.LambdaFunction` corresponding to transformation described by f
:param f: A Python of one of the following forms: - (Column) -> Column: ... - (Column, Column) -> Column: ... - (Column, Column, Column) -> Column: ... """
_unresolved_named_lambda_variable( expressions.UnresolvedNamedLambdaVariable.freshVarName(arg) ) for arg in argnames[: len(parameters)] ]
""" Invokes expression identified by name, (relative to ```org.apache.spark.sql.catalyst.expressions``) and wraps the result with Column (first Scala one, then Python).
:param name: Name of the expression :param cols: a list of columns :param funs: a list of((*Column) -> Column functions.
:return: a Column """
""" Returns an array of elements after applying a transformation to each element in the input array.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function a function that is applied to each element of the input array. Can take one of the following forms:
- Unary ``(x: Column) -> Column: ...`` - Binary ``(x: Column, i: Column) -> Column...``, where the second argument is a 0-based index of the element.
and can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, [1, 2, 3, 4])], ("key", "values")) >>> df.select(transform("values", lambda x: x * 2).alias("doubled")).show() +------------+ | doubled| +------------+ |[2, 4, 6, 8]| +------------+
>>> def alternate(x, i): ... return when(i % 2 == 0, x).otherwise(-x) >>> df.select(transform("values", alternate).alias("alternated")).show() +--------------+ | alternated| +--------------+ |[1, -2, 3, -4]| +--------------+ """
""" Returns whether a predicate holds for one or more elements in the array.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function ``(x: Column) -> Column: ...`` returning the Boolean expression. Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__). :return: a :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, [1, 2, 3, 4]), (2, [3, -1, 0])],("key", "values")) >>> df.select(exists("values", lambda x: x < 0).alias("any_negative")).show() +------------+ |any_negative| +------------+ | false| | true| +------------+ """
""" Returns whether a predicate holds for every element in the array.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function ``(x: Column) -> Column: ...`` returning the Boolean expression. Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame( ... [(1, ["bar"]), (2, ["foo", "bar"]), (3, ["foobar", "foo"])], ... ("key", "values") ... ) >>> df.select(forall("values", lambda x: x.rlike("foo")).alias("all_foo")).show() +-------+ |all_foo| +-------+ | false| | false| | true| +-------+ """
""" Returns an array of elements for which a predicate holds in a given array.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function A function that returns the Boolean expression. Can take one of the following forms:
- Unary ``(x: Column) -> Column: ...`` - Binary ``(x: Column, i: Column) -> Column...``, where the second argument is a 0-based index of the element.
and can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame( ... [(1, ["2018-09-20", "2019-02-03", "2019-07-01", "2020-06-01"])], ... ("key", "values") ... ) >>> def after_second_quarter(x): ... return month(to_date(x)) > 6 >>> df.select( ... filter("values", after_second_quarter).alias("after_second_quarter") ... ).show(truncate=False) +------------------------+ |after_second_quarter | +------------------------+ |[2018-09-20, 2019-07-01]| +------------------------+ """
""" Applies a binary operator to an initial state and all elements in the array, and reduces this to a single state. The final state is converted into the final result by applying a finish function.
Both functions can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression initialValue : :class:`~pyspark.sql.Column` or str initial value. Name of column or expression merge : function a binary function ``(acc: Column, x: Column) -> Column...`` returning expression of the same type as ``zero`` finish : function an optional unary function ``(x: Column) -> Column: ...`` used to convert accumulated value.
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, [20.0, 4.0, 2.0, 6.0, 10.0])], ("id", "values")) >>> df.select(aggregate("values", lit(0.0), lambda acc, x: acc + x).alias("sum")).show() +----+ | sum| +----+ |42.0| +----+
>>> def merge(acc, x): ... count = acc.count + 1 ... sum = acc.sum + x ... return struct(count.alias("count"), sum.alias("sum")) >>> df.select( ... aggregate( ... "values", ... struct(lit(0).alias("count"), lit(0.0).alias("sum")), ... merge, ... lambda acc: acc.sum / acc.count, ... ).alias("mean") ... ).show() +----+ |mean| +----+ | 8.4| +----+ """ "ArrayAggregate", [col, initialValue], [merge, finish] )
else: "ArrayAggregate", [col, initialValue], [merge] )
""" Merge two given arrays, element-wise, into a single array using a function. If one array is shorter, nulls are appended at the end to match the length of the longer array, before applying the function.
.. versionadded:: 3.1.0
Parameters ---------- left : :class:`~pyspark.sql.Column` or str name of the first column or expression right : :class:`~pyspark.sql.Column` or str name of the second column or expression f : function a binary function ``(x1: Column, x2: Column) -> Column...`` Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, [1, 3, 5, 8], [0, 2, 4, 6])], ("id", "xs", "ys")) >>> df.select(zip_with("xs", "ys", lambda x, y: x ** y).alias("powers")).show(truncate=False) +---------------------------+ |powers | +---------------------------+ |[1.0, 9.0, 625.0, 262144.0]| +---------------------------+
>>> df = spark.createDataFrame([(1, ["foo", "bar"], [1, 2, 3])], ("id", "xs", "ys")) >>> df.select(zip_with("xs", "ys", lambda x, y: concat_ws("_", x, y)).alias("xs_ys")).show() +-----------------+ | xs_ys| +-----------------+ |[foo_1, bar_2, 3]| +-----------------+ """
""" Applies a function to every key-value pair in a map and returns a map with the results of those applications as the new keys for the pairs.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function a binary function ``(k: Column, v: Column) -> Column...`` Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, {"foo": -2.0, "bar": 2.0})], ("id", "data")) >>> df.select(transform_keys( ... "data", lambda k, _: upper(k)).alias("data_upper") ... ).show(truncate=False) +-------------------------+ |data_upper | +-------------------------+ |{BAR -> 2.0, FOO -> -2.0}| +-------------------------+ """
""" Applies a function to every key-value pair in a map and returns a map with the results of those applications as the new values for the pairs.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function a binary function ``(k: Column, v: Column) -> Column...`` Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, {"IT": 10.0, "SALES": 2.0, "OPS": 24.0})], ("id", "data")) >>> df.select(transform_values( ... "data", lambda k, v: when(k.isin("IT", "OPS"), v + 10.0).otherwise(v) ... ).alias("new_data")).show(truncate=False) +---------------------------------------+ |new_data | +---------------------------------------+ |{OPS -> 34.0, IT -> 20.0, SALES -> 2.0}| +---------------------------------------+ """
""" Returns a map whose key-value pairs satisfy a predicate.
.. versionadded:: 3.1.0
Parameters ---------- col : :class:`~pyspark.sql.Column` or str name of column or expression f : function a binary function ``(k: Column, v: Column) -> Column...`` Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([(1, {"foo": 42.0, "bar": 1.0, "baz": 32.0})], ("id", "data")) >>> df.select(map_filter( ... "data", lambda _, v: v > 30.0).alias("data_filtered") ... ).show(truncate=False) +--------------------------+ |data_filtered | +--------------------------+ |{baz -> 32.0, foo -> 42.0}| +--------------------------+ """
""" Merge two given maps, key-wise into a single map using a function.
.. versionadded:: 3.1.0
Parameters ---------- col1 : :class:`~pyspark.sql.Column` or str name of the first column or expression col2 : :class:`~pyspark.sql.Column` or str name of the second column or expression f : function a ternary function ``(k: Column, v1: Column, v2: Column) -> Column...`` Can use methods of :class:`~pyspark.sql.Column`, functions defined in :py:mod:`pyspark.sql.functions` and Scala ``UserDefinedFunctions``. Python ``UserDefinedFunctions`` are not supported (`SPARK-27052 <https://issues.apache.org/jira/browse/SPARK-27052>`__).
Returns ------- :class:`~pyspark.sql.Column`
Examples -------- >>> df = spark.createDataFrame([ ... (1, {"IT": 24.0, "SALES": 12.00}, {"IT": 2.0, "SALES": 1.4})], ... ("id", "base", "ratio") ... ) >>> df.select(map_zip_with( ... "base", "ratio", lambda k, v1, v2: round(v1 * v2, 2)).alias("updated_data") ... ).show(truncate=False) +---------------------------+ |updated_data | +---------------------------+ |{SALES -> 16.8, IT -> 48.0}| +---------------------------+ """
# ---------------------- Partition transform functions --------------------------------
""" Partition transform function: A transform for timestamps and dates to partition data into years.
.. versionadded:: 3.1.0
Examples -------- >>> df.writeTo("catalog.db.table").partitionedBy( # doctest: +SKIP ... years("ts") ... ).createOrReplace()
Notes ----- This function can be used only in combination with :py:meth:`~pyspark.sql.readwriter.DataFrameWriterV2.partitionedBy` method of the `DataFrameWriterV2`.
"""
""" Partition transform function: A transform for timestamps and dates to partition data into months.
.. versionadded:: 3.1.0
Examples -------- >>> df.writeTo("catalog.db.table").partitionedBy( ... months("ts") ... ).createOrReplace() # doctest: +SKIP
Notes ----- This function can be used only in combination with :py:meth:`~pyspark.sql.readwriter.DataFrameWriterV2.partitionedBy` method of the `DataFrameWriterV2`.
"""
""" Partition transform function: A transform for timestamps and dates to partition data into days.
.. versionadded:: 3.1.0
Examples -------- >>> df.writeTo("catalog.db.table").partitionedBy( # doctest: +SKIP ... days("ts") ... ).createOrReplace()
Notes ----- This function can be used only in combination with :py:meth:`~pyspark.sql.readwriter.DataFrameWriterV2.partitionedBy` method of the `DataFrameWriterV2`.
"""
""" Partition transform function: A transform for timestamps to partition data into hours.
.. versionadded:: 3.1.0
Examples -------- >>> df.writeTo("catalog.db.table").partitionedBy( # doctest: +SKIP ... hours("ts") ... ).createOrReplace()
Notes ----- This function can be used only in combination with :py:meth:`~pyspark.sql.readwriter.DataFrameWriterV2.partitionedBy` method of the `DataFrameWriterV2`.
"""
""" Partition transform function: A transform for any type that partitions by a hash of the input column.
.. versionadded:: 3.1.0
Examples -------- >>> df.writeTo("catalog.db.table").partitionedBy( # doctest: +SKIP ... bucket(42, "ts") ... ).createOrReplace()
Notes ----- This function can be used only in combination with :py:meth:`~pyspark.sql.readwriter.DataFrameWriterV2.partitionedBy` method of the `DataFrameWriterV2`.
""" raise TypeError( "numBuckets should be a Column or an int, got {}".format(type(numBuckets)) )
_create_column_from_literal(numBuckets) if isinstance(numBuckets, int) else _to_java_column(numBuckets) )
# ---------------------------- User Defined Function ----------------------------------
"""Creates a user defined function (UDF).
.. versionadded:: 1.3.0
Parameters ---------- f : function python function if used as a standalone function returnType : :class:`pyspark.sql.types.DataType` or str the return type of the user-defined function. The value can be either a :class:`pyspark.sql.types.DataType` object or a DDL-formatted type string.
Examples -------- >>> from pyspark.sql.types import IntegerType >>> slen = udf(lambda s: len(s), IntegerType()) >>> @udf ... def to_upper(s): ... if s is not None: ... return s.upper() ... >>> @udf(returnType=IntegerType()) ... def add_one(x): ... if x is not None: ... return x + 1 ... >>> df = spark.createDataFrame([(1, "John Doe", 21)], ("id", "name", "age")) >>> df.select(slen("name").alias("slen(name)"), to_upper("name"), add_one("age")).show() +----------+--------------+------------+ |slen(name)|to_upper(name)|add_one(age)| +----------+--------------+------------+ | 8| JOHN DOE| 22| +----------+--------------+------------+
Notes ----- The user-defined functions are considered deterministic by default. Due to optimization, duplicate invocations may be eliminated or the function may even be invoked more times than it is present in the query. If your function is not deterministic, call `asNondeterministic` on the user defined function. E.g.:
>>> from pyspark.sql.types import IntegerType >>> import random >>> random_udf = udf(lambda: int(random.random() * 100), IntegerType()).asNondeterministic()
The user-defined functions do not support conditional expressions or short circuiting in boolean expressions and it ends up with being executed all internally. If the functions can fail on special rows, the workaround is to incorporate the condition into the functions.
The user-defined functions do not take keyword arguments on the calling side. """
# The following table shows most of Python data and SQL type conversions in normal UDFs that # are not yet visible to the user. Some of behaviors are buggy and might be changed in the near # future. The table might have to be eventually documented externally. # Please see SPARK-28131's PR to see the codes in order to generate the table below. # # +-----------------------------+--------------+----------+------+---------------+--------------------+-----------------------------+----------+----------------------+---------+--------------------+----------------------------+------------+--------------+------------------+----------------------+ # noqa # |SQL Type \ Python Value(Type)|None(NoneType)|True(bool)|1(int)| a(str)| 1970-01-01(date)|1970-01-01 00:00:00(datetime)|1.0(float)|array('i', [1])(array)|[1](list)| (1,)(tuple)|bytearray(b'ABC')(bytearray)| 1(Decimal)|{'a': 1}(dict)|Row(kwargs=1)(Row)|Row(namedtuple=1)(Row)| # noqa # +-----------------------------+--------------+----------+------+---------------+--------------------+-----------------------------+----------+----------------------+---------+--------------------+----------------------------+------------+--------------+------------------+----------------------+ # noqa # | boolean| None| True| None| None| None| None| None| None| None| None| None| None| None| X| X| # noqa # | tinyint| None| None| 1| None| None| None| None| None| None| None| None| None| None| X| X| # noqa # | smallint| None| None| 1| None| None| None| None| None| None| None| None| None| None| X| X| # noqa # | int| None| None| 1| None| None| None| None| None| None| None| None| None| None| X| X| # noqa # | bigint| None| None| 1| None| None| None| None| None| None| None| None| None| None| X| X| # noqa # | string| None| 'true'| '1'| 'a'|'java.util.Gregor...| 'java.util.Gregor...| '1.0'| '[I@66cbb73a'| '[1]'|'[Ljava.lang.Obje...| '[B@5a51eb1a'| '1'| '{a=1}'| X| X| # noqa # | date| None| X| X| X|datetime.date(197...| datetime.date(197...| X| X| X| X| X| X| X| X| X| # noqa # | timestamp| None| X| X| X| X| datetime.datetime...| X| X| X| X| X| X| X| X| X| # noqa # | float| None| None| None| None| None| None| 1.0| None| None| None| None| None| None| X| X| # noqa # | double| None| None| None| None| None| None| 1.0| None| None| None| None| None| None| X| X| # noqa # | array<int>| None| None| None| None| None| None| None| [1]| [1]| [1]| [65, 66, 67]| None| None| X| X| # noqa # | binary| None| None| None|bytearray(b'a')| None| None| None| None| None| None| bytearray(b'ABC')| None| None| X| X| # noqa # | decimal(10,0)| None| None| None| None| None| None| None| None| None| None| None|Decimal('1')| None| X| X| # noqa # | map<string,int>| None| None| None| None| None| None| None| None| None| None| None| None| {'a': 1}| X| X| # noqa # | struct<_1:int>| None| X| X| X| X| X| X| X|Row(_1=1)| Row(_1=1)| X| X| Row(_1=None)| Row(_1=1)| Row(_1=1)| # noqa # +-----------------------------+--------------+----------+------+---------------+--------------------+-----------------------------+----------+----------------------+---------+--------------------+----------------------------+------------+--------------+------------------+----------------------+ # noqa # # Note: DDL formatted string is used for 'SQL Type' for simplicity. This string can be # used in `returnType`. # Note: The values inside of the table are generated by `repr`. # Note: 'X' means it throws an exception during the conversion. # Note: Python 3.7.3 is used.
# decorator @udf, @udf(), @udf(dataType()) # If DataType has been passed as a positional argument # for decorator use it as a returnType evalType=PythonEvalType.SQL_BATCHED_UDF) else: evalType=PythonEvalType.SQL_BATCHED_UDF)
.master("local[4]")\ .appName("sql.functions tests")\ .getOrCreate() pyspark.sql.functions, globs=globs, optionflags=doctest.ELLIPSIS | doctest.NORMALIZE_WHITESPACE) sys.exit(-1)
|