[SPARK-57579][PYTHON] Add PySpark support for unix_nanos function

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

@@ -327,6 +327,7 @@ Date and Timestamp Functions
     unix_date
     unix_micros
     unix_millis
+    unix_nanos
     unix_seconds
     unix_timestamp
     weekday

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

@@ -3531,6 +3531,13 @@ def unix_millis(col: "ColumnOrName") -> Column:
 unix_millis.__doc__ = pysparkfuncs.unix_millis.__doc__
 
 
+def unix_nanos(col: "ColumnOrName") -> Column:
+    return _invoke_function_over_columns("unix_nanos", col)
+
+
+unix_nanos.__doc__ = pysparkfuncs.unix_nanos.__doc__
+
+
 def unix_seconds(col: "ColumnOrName") -> Column:
     return _invoke_function_over_columns("unix_seconds", col)
 

python/pyspark/sql/functions/__init__.py

@@ -276,6 +276,7 @@
     "unix_date",
     "unix_micros",
     "unix_millis",
+    "unix_nanos",
     "unix_seconds",
     "unix_timestamp",
     "weekday",

python/pyspark/sql/functions/builtin.py

@@ -11792,6 +11792,57 @@ def unix_millis(col: "ColumnOrName") -> Column:
     return _invoke_function_over_columns("unix_millis", col)
 
 
+@_try_remote_functions
+def unix_nanos(col: "ColumnOrName") -> Column:
+    """Returns the number of nanoseconds since 1970-01-01 00:00:00 UTC as ``DECIMAL(21, 0)``.
+    Only supported for ``TIMESTAMP_LTZ(p)`` and ``TIMESTAMP_NTZ(p)`` with precision ``p``
+    in ``[7, 9]``.
+
+    .. versionadded:: 4.3.0
+
+    Parameters
+    ----------
+    col : :class:`~pyspark.sql.Column` or column name
+        input column of nanosecond-precision timestamp values to convert.
+
+    Returns
+    -------
+    :class:`~pyspark.sql.Column`
+        the number of nanoseconds since 1970-01-01 00:00:00 UTC as ``DECIMAL(21, 0)``.
+
+    See Also
+    --------
+    :meth:`pyspark.sql.functions.unix_date`
+    :meth:`pyspark.sql.functions.unix_seconds`
+    :meth:`pyspark.sql.functions.unix_millis`
+    :meth:`pyspark.sql.functions.unix_micros`
+
+    Examples
+    --------
+    >>> import pyspark.sql.functions as sf
+    >>> spark.conf.set("spark.sql.timestampNanosTypes.enabled", True)
+    >>> df = spark.sql(
+    ...     "SELECT TIMESTAMP_NTZ '2020-01-01 13:24:35.123456789' AS ts"
+    ... )
+    >>> df.select('*', sf.unix_nanos('ts')).show(truncate=False)
+    +-----------------------------+-------------------+
+    |ts                           |unix_nanos(ts)     |
+    +-----------------------------+-------------------+
+    |2020-01-01 13:24:35.123456789|1577885075123456789|
+    +-----------------------------+-------------------+
+
+    >>> df.select(sf.unix_nanos(sf.lit(None).cast('timestamp_ntz(9)'))).show()
+    +------------------------------------------+
+    |unix_nanos(CAST(NULL AS TIMESTAMP_NTZ(9)))|
+    +------------------------------------------+
+    |                                      NULL|
+    +------------------------------------------+
+
+    >>> spark.conf.unset("spark.sql.timestampNanosTypes.enabled")
+    """
+    return _invoke_function_over_columns("unix_nanos", col)
+
+
 @_try_remote_functions
 def unix_seconds(col: "ColumnOrName") -> Column:
     """Returns the number of seconds since 1970-01-01 00:00:00 UTC.

python/pyspark/sql/tests/test_functions.py

@@ -82,9 +82,7 @@ def test_function_parity(self):
         missing_in_py = jvm_fn_set.difference(py_fn_set)
 
         # Functions that we expect to be missing in python until they are added to pyspark
-        expected_missing_in_py = {
-            "unix_nanos",  # SPARK-57527: PySpark support tracked as a follow-up
-        }
+        expected_missing_in_py = set()
 
         self.assertEqual(
             expected_missing_in_py, missing_in_py, "Missing functions in pyspark not as expected"