Skip to content

Commit 66fb225

Browse files
zhengruifengHyukjinKwon
zhengruifeng
authored and
HyukjinKwon
committed
[SPARK-45049][CONNECT][DOCS][TESTS] Refine docstrings of coalesce/repartition/repartitionByRange
### What changes were proposed in this pull request? Enable doctests for `coalesce/repartition/repartitionByRange`, by using `explain` instead of `rdd.getNumPartitions` ### Why are the changes needed? test coverage ### Does this PR introduce _any_ user-facing change? yes ### How was this patch tested? updated doctests ### Was this patch authored or co-authored using generative AI tooling? NO Closes #42770 from zhengruifeng/enable_doctest_partition. Authored-by: Ruifeng Zheng <ruifengz@apache.org> Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>
1 parent 8d358cd commit 66fb225

File tree

2 files changed

+99
-22
lines changed

2 files changed

+99
-22
lines changed

‎python/pyspark/sql/connect/dataframe.py‎

Lines changed: 0 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -2191,10 +2191,6 @@ def _test() -> None:
21912191
os.chdir(os.environ["SPARK_HOME"])
21922192

21932193
globs = pyspark.sql.connect.dataframe.__dict__.copy()
2194-
# Spark Connect does not support RDD but the tests depend on them.
2195-
del pyspark.sql.connect.dataframe.DataFrame.coalesce.__doc__
2196-
del pyspark.sql.connect.dataframe.DataFrame.repartition.__doc__
2197-
del pyspark.sql.connect.dataframe.DataFrame.repartitionByRange.__doc__
21982194

21992195
# TODO(SPARK-41625): Support Structured Streaming
22002196
del pyspark.sql.connect.dataframe.DataFrame.isStreaming.__doc__

‎python/pyspark/sql/dataframe.py‎

Lines changed: 99 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -1763,9 +1763,27 @@ def coalesce(self, numPartitions: int) -> "DataFrame":
17631763
17641764
Examples
17651765
--------
1766-
>>> df = spark.range(10)
1767-
>>> df.coalesce(1).rdd.getNumPartitions()
1768-
1
1766+
>>> from pyspark.sql import functions as sf
1767+
>>> spark.range(0, 10, 1, 3).select(
1768+
... sf.spark_partition_id().alias("partition")
1769+
... ).distinct().sort("partition").show()
1770+
+---------+
1771+
|partition|
1772+
+---------+
1773+
| 0|
1774+
| 1|
1775+
| 2|
1776+
+---------+
1777+
1778+
>>> from pyspark.sql import functions as sf
1779+
>>> spark.range(0, 10, 1, 3).coalesce(1).select(
1780+
... sf.spark_partition_id().alias("partition")
1781+
... ).distinct().sort("partition").show()
1782+
+---------+
1783+
|partition|
1784+
+---------+
1785+
| 0|
1786+
+---------+
17691787
"""
17701788
return DataFrame(self._jdf.coalesce(numPartitions), self.sparkSession)
17711789

@@ -1809,23 +1827,78 @@ def repartition( # type: ignore[misc]
18091827
18101828
Examples
18111829
--------
1812-
>>> df = spark.createDataFrame(
1813-
... [(14, "Tom"), (23, "Alice"), (16, "Bob")], ["age", "name"])
1830+
>>> from pyspark.sql import functions as sf
1831+
>>> df = spark.range(0, 64, 1, 9).withColumn(
1832+
... "name", sf.concat(sf.lit("name_"), sf.col("id").cast("string"))
1833+
... ).withColumn(
1834+
... "age", sf.col("id") - 32
1835+
... )
1836+
>>> df.select(
1837+
... sf.spark_partition_id().alias("partition")
1838+
... ).distinct().sort("partition").show()
1839+
+---------+
1840+
|partition|
1841+
+---------+
1842+
| 0|
1843+
| 1|
1844+
| 2|
1845+
| 3|
1846+
| 4|
1847+
| 5|
1848+
| 6|
1849+
| 7|
1850+
| 8|
1851+
+---------+
18141852
18151853
Repartition the data into 10 partitions.
18161854
1817-
>>> df.repartition(10).rdd.getNumPartitions()
1818-
10
1855+
>>> df.repartition(10).select(
1856+
... sf.spark_partition_id().alias("partition")
1857+
... ).distinct().sort("partition").show()
1858+
+---------+
1859+
|partition|
1860+
+---------+
1861+
| 0|
1862+
| 1|
1863+
| 2|
1864+
| 3|
1865+
| 4|
1866+
| 5|
1867+
| 6|
1868+
| 7|
1869+
| 8|
1870+
| 9|
1871+
+---------+
18191872
18201873
Repartition the data into 7 partitions by 'age' column.
18211874
1822-
>>> df.repartition(7, "age").rdd.getNumPartitions()
1823-
7
1875+
>>> df.repartition(7, "age").select(
1876+
... sf.spark_partition_id().alias("partition")
1877+
... ).distinct().sort("partition").show()
1878+
+---------+
1879+
|partition|
1880+
+---------+
1881+
| 0|
1882+
| 1|
1883+
| 2|
1884+
| 3|
1885+
| 4|
1886+
| 5|
1887+
| 6|
1888+
+---------+
18241889
18251890
Repartition the data into 7 partitions by 'age' and 'name columns.
18261891
1827-
>>> df.repartition(3, "name", "age").rdd.getNumPartitions()
1828-
3
1892+
>>> df.repartition(3, "name", "age").select(
1893+
... sf.spark_partition_id().alias("partition")
1894+
... ).distinct().sort("partition").show()
1895+
+---------+
1896+
|partition|
1897+
+---------+
1898+
| 0|
1899+
| 1|
1900+
| 2|
1901+
+---------+
18291902
"""
18301903
if isinstance(numPartitions, int):
18311904
if len(cols) == 0:
@@ -1893,15 +1966,23 @@ def repartitionByRange( # type: ignore[misc]
18931966
18941967
Examples
18951968
--------
1896-
>>> df = spark.createDataFrame(
1897-
... [(14, "Tom"), (23, "Alice"), (16, "Bob")], ["age", "name"])
1898-
18991969
Repartition the data into 2 partitions by range in 'age' column.
1900-
For example, the first partition can have ``(14, "Tom")``, and the second
1901-
partition would have ``(16, "Bob")`` and ``(23, "Alice")``.
1970+
For example, the first partition can have ``(14, "Tom")`` and ``(16, "Bob")``,
1971+
and the second partition would have ``(23, "Alice")``.
19021972
1903-
>>> df.repartitionByRange(2, "age").rdd.getNumPartitions()
1904-
2
1973+
>>> from pyspark.sql import functions as sf
1974+
>>> spark.createDataFrame(
1975+
... [(14, "Tom"), (23, "Alice"), (16, "Bob")], ["age", "name"]
1976+
... ).repartitionByRange(2, "age").select(
1977+
... "age", "name", sf.spark_partition_id()
1978+
... ).show()
1979+
+---+-----+--------------------+
1980+
|age| name|SPARK_PARTITION_ID()|
1981+
+---+-----+--------------------+
1982+
| 14| Tom| 0|
1983+
| 16| Bob| 0|
1984+
| 23|Alice| 1|
1985+
+---+-----+--------------------+
19051986
"""
19061987
if isinstance(numPartitions, int):
19071988
if len(cols) == 0:

0 commit comments

Comments
 (0)