[SPARK-45900][SQL][PYTHON] Add xxh3_64 and xxh3_128 functions

python/docs/source/reference/pyspark.sql/functions.rst

@@ -349,6 +349,8 @@ Hash Functions
     sha
     sha1
     sha2
+    xxh3_128
+    xxh3_64
     xxhash64
 
 

python/pyspark/sql/connect/functions/builtin.py

@@ -4674,6 +4674,20 @@ def md5(col: "ColumnOrName") -> Column:
 md5.__doc__ = pysparkfuncs.md5.__doc__
 
 
+def xxh3_64(col: "ColumnOrName") -> Column:
+    return _invoke_function_over_columns("xxh3_64", col)
+
+
+xxh3_64.__doc__ = pysparkfuncs.xxh3_64.__doc__
+
+
+def xxh3_128(col: "ColumnOrName") -> Column:
+    return _invoke_function_over_columns("xxh3_128", col)
+
+
+xxh3_128.__doc__ = pysparkfuncs.xxh3_128.__doc__
+
+
 def sha1(col: "ColumnOrName") -> Column:
     return _invoke_function_over_columns("sha1", col)
 

python/pyspark/sql/functions/__init__.py

@@ -292,6 +292,8 @@
     "sha",
     "sha1",
     "sha2",
+    "xxh3_128",
+    "xxh3_64",
     "xxhash64",
     # Collection Functions
     "aggregate",

python/pyspark/sql/functions/builtin.py

@@ -14697,6 +14697,71 @@ def md5(col: "ColumnOrName") -> Column:
     return _invoke_function_over_columns("md5", col)
 
 
+@_try_remote_functions
+def xxh3_64(col: "ColumnOrName") -> Column:
+    """Returns a 64-bit hash value of the argument using the XXH3 algorithm.
+
+    Unlike :func:`xxhash64`, which hashes one or more columns structurally, this hashes the raw
+    bytes of a single value with seed 0, so its result is byte compatible with the reference XXH3.
+
+    .. versionadded:: 4.3.0
+
+    Parameters
+    ----------
+    col : :class:`~pyspark.sql.Column` or column name
+        target column to compute on. A column of string or binary type.
+
+    Returns
+    -------
+    :class:`~pyspark.sql.Column`
+        Returns a column that evaluates to a long.
+
+    See Also
+    --------
+    :meth:`pyspark.sql.functions.xxh3_128`
+    :meth:`pyspark.sql.functions.xxhash64`
+
+    Examples
+    --------
+    >>> import pyspark.sql.functions as sf
+    >>> df = spark.createDataFrame([('Spark',)], ['a'])
+    >>> df.select(sf.xxh3_64('a').alias('h')).collect()
+    [Row(h=80997306238743657)]
+    """
+    return _invoke_function_over_columns("xxh3_64", col)
+
+
+@_try_remote_functions
+def xxh3_128(col: "ColumnOrName") -> Column:
+    """Returns a 128-bit XXH3 hash of the argument as a 32-character hex string.
+
+    .. versionadded:: 4.3.0
+
+    Parameters
+    ----------
+    col : :class:`~pyspark.sql.Column` or column name
+        target column to compute on. A column of string or binary type.
+
+    Returns
+    -------
+    :class:`~pyspark.sql.Column`
+        Returns a column that evaluates to a string.
+
+    See Also
+    --------
+    :meth:`pyspark.sql.functions.xxh3_64`
+    :meth:`pyspark.sql.functions.md5`
+
+    Examples
+    --------
+    >>> import pyspark.sql.functions as sf
+    >>> df = spark.createDataFrame([('Spark',)], ['a'])
+    >>> df.select(sf.xxh3_128('a').alias('h')).collect()
+    [Row(h='7d57dd84c60c86ca1f4e82ab91a12b5e')]
+    """
+    return _invoke_function_over_columns("xxh3_128", col)
+
+
 @_try_remote_functions
 def sha1(col: "ColumnOrName") -> Column:
     """Returns the hex string result of SHA-1.

sql/api/src/main/scala/org/apache/spark/sql/functions.scala

@@ -6551,6 +6551,30 @@ object functions {
   @scala.annotation.varargs
   def xxhash64(cols: Column*): Column = Column.fn("xxhash64", cols: _*)
 
+  /**
+   * Returns a 64-bit hash value of the argument using the XXH3 algorithm.
+   *
+   * @param col
+   *   the column to hash. A column of string or binary type.
+   * @group hash_funcs
+   * @since 4.3.0
+   * @return
+   *   Returns a column that evaluates to a long.
+   */
+  def xxh3_64(col: Column): Column = Column.fn("xxh3_64", col)
+
+  /**
+   * Returns a 128-bit XXH3 hash of the argument as a 32-character hex string.
+   *
+   * @param col
+   *   the column to hash. A column of string or binary type.
+   * @group hash_funcs
+   * @since 4.3.0
+   * @return
+   *   Returns a column that evaluates to a string.
+   */
+  def xxh3_128(col: Column): Column = Column.fn("xxh3_128", col)
+
   /**
    * Returns null if the condition is true, and throws an exception otherwise.
    *