This is an automated email from the ASF dual-hosted git repository.
gurwls223 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/master by this push:
new 48c5843ad4b [SPARK-41655][CONNECT] Enable doctests in
pyspark.sql.connect.column
48c5843ad4b is described below
commit 48c5843ad4b06e85e07d1db5b308a460209f6126
Author: Sandeep Singh <sand...@techaddict.me>
AuthorDate: Thu Dec 29 20:52:35 2022 +0900
[SPARK-41655][CONNECT] Enable doctests in pyspark.sql.connect.column
### What changes were proposed in this pull request?
This PR proposes to enable doctests in pyspark.sql.connect.column that is
virtually the same as pyspark.sql.column.
### Why are the changes needed?
To make sure on the PySpark compatibility and test coverage.
### Does this PR introduce any user-facing change?
No, doctest's only.
### How was this patch tested?
New Doctests Added
Closes #39249 from techaddict/SPARK-41655-pyspark.sql.connect.column.
Lead-authored-by: Sandeep Singh <sand...@techaddict.me>
Co-authored-by: Hyukjin Kwon <gurwls...@gmail.com>
Signed-off-by: Hyukjin Kwon <gurwls...@apache.org>
---
dev/sparktestsupport/modules.py | 1 +
python/pyspark/sql/column.py | 15 ++++-----
python/pyspark/sql/connect/column.py | 60 ++++++++++++++++++++++++++++++++++++
3 files changed, 69 insertions(+), 7 deletions(-)
diff --git a/dev/sparktestsupport/modules.py b/dev/sparktestsupport/modules.py
index 558d058f3e5..df3a1f180fc 100644
--- a/dev/sparktestsupport/modules.py
+++ b/dev/sparktestsupport/modules.py
@@ -507,6 +507,7 @@ pyspark_connect = Module(
"pyspark.sql.connect.catalog",
"pyspark.sql.connect.group",
"pyspark.sql.connect.window",
+ "pyspark.sql.connect.column",
# unittests
"pyspark.sql.tests.connect.test_connect_column_expressions",
"pyspark.sql.tests.connect.test_connect_plan_only",
diff --git a/python/pyspark/sql/column.py b/python/pyspark/sql/column.py
index 3bc49ef8031..5a0987b4cfe 100644
--- a/python/pyspark/sql/column.py
+++ b/python/pyspark/sql/column.py
@@ -182,6 +182,9 @@ def _reverse_op(
return _
+# TODO(SPARK-41757): Compatibility of string representation for Column
+
+
class Column:
"""
@@ -200,17 +203,16 @@ class Column:
... [(2, "Alice"), (5, "Bob")], ["age", "name"])
Select a column out of a DataFrame
-
- >>> df.name
+ >>> df.name # doctest: +SKIP
Column<'name'>
- >>> df["name"]
+ >>> df["name"] # doctest: +SKIP
Column<'name'>
Create from an expression
- >>> df.age + 1
+ >>> df.age + 1 # doctest: +SKIP
Column<'(age + 1)'>
- >>> 1 / df.age
+ >>> 1 / df.age # doctest: +SKIP
Column<'(1 / age)'>
"""
@@ -1258,8 +1260,7 @@ class Column:
>>> from pyspark.sql import Window
>>> window = Window.partitionBy("name").orderBy("age") \
.rowsBetween(Window.unboundedPreceding, Window.currentRow)
- >>> from pyspark.sql.functions import rank, min
- >>> from pyspark.sql.functions import desc
+ >>> from pyspark.sql.functions import rank, min, desc
>>> df = spark.createDataFrame(
... [(2, "Alice"), (5, "Bob")], ["age", "name"])
>>> df.withColumn("rank", rank().over(window)) \
diff --git a/python/pyspark/sql/connect/column.py
b/python/pyspark/sql/connect/column.py
index b873a757e41..58d86a3d389 100644
--- a/python/pyspark/sql/connect/column.py
+++ b/python/pyspark/sql/connect/column.py
@@ -28,6 +28,7 @@ from typing import (
Optional,
)
+from pyspark import SparkContext, SparkConf
from pyspark.sql.types import DataType
from pyspark.sql.column import Column as PySparkColumn
@@ -390,3 +391,62 @@ class Column:
Column.__doc__ = PySparkColumn.__doc__
+
+
+def _test() -> None:
+ import os
+ import sys
+ import doctest
+ from pyspark.sql import SparkSession as PySparkSession
+ from pyspark.testing.connectutils import should_test_connect,
connect_requirement_message
+
+ os.chdir(os.environ["SPARK_HOME"])
+
+ if should_test_connect:
+ import pyspark.sql.connect.column
+
+ globs = pyspark.sql.connect.column.__dict__.copy()
+ # Works around to create a regular Spark session
+ sc = SparkContext("local[4]", "sql.connect.column tests",
conf=SparkConf())
+ globs["_spark"] = PySparkSession(sc, options={"spark.app.name":
"sql.connect.column tests"})
+
+ # Creates a remote Spark session.
+ os.environ["SPARK_REMOTE"] = "sc://localhost"
+ globs["spark"] =
PySparkSession.builder.remote("sc://localhost").getOrCreate()
+
+ # TODO(SPARK-41751): Support
Column.bitwiseAND,bitwiseOR,bitwiseXOR,eqNullSafe,isNotNull,
+ # isNull,isin
+ del pyspark.sql.connect.column.Column.bitwiseAND.__doc__
+ del pyspark.sql.connect.column.Column.bitwiseOR.__doc__
+ del pyspark.sql.connect.column.Column.bitwiseXOR.__doc__
+ del pyspark.sql.connect.column.Column.eqNullSafe.__doc__
+ del pyspark.sql.connect.column.Column.isNotNull.__doc__
+ del pyspark.sql.connect.column.Column.isNull.__doc__
+ del pyspark.sql.connect.column.Column.isin.__doc__
+ # TODO(SPARK-41756): Fix createDataFrame
+ del pyspark.sql.connect.column.Column.getField.__doc__
+ del pyspark.sql.connect.column.Column.getItem.__doc__
+ # TODO(SPARK-41758): Support Window functions
+ del pyspark.sql.connect.column.Column.over.__doc__
+
+ (failure_count, test_count) = doctest.testmod(
+ pyspark.sql.connect.column,
+ globs=globs,
+ optionflags=doctest.ELLIPSIS
+ | doctest.NORMALIZE_WHITESPACE
+ | doctest.IGNORE_EXCEPTION_DETAIL,
+ )
+
+ globs["spark"].stop()
+ globs["_spark"].stop()
+ if failure_count:
+ sys.exit(-1)
+ else:
+ print(
+ f"Skipping pyspark.sql.connect.column doctests:
{connect_requirement_message}",
+ file=sys.stderr,
+ )
+
+
+if __name__ == "__main__":
+ _test()
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org
