[SPARK-56219][PS][FOLLOW-UP] Keep legacy groupby idxmax and idxmin skipna=False behavior for pandas 2
### What changes were proposed in this pull request?
This is a follow-up of apache/spark#55021.
This PR updates pandas-on-Spark `GroupBy.idxmax` and `GroupBy.idxmin` for `skipna=False` to keep the legacy behavior for all pandas 2 versions.
With this change:
- pandas `< 3.0.0` keeps the legacy `idxmax` and `idxmin` result for `skipna=False`
- pandas `>= 3.0.0` keeps the existing error behavior for NA-containing input
This PR also updates the related test in `python/pyspark/pandas/tests/groupby/test_index.py` to validate the pandas 2 behavior directly instead of relying on pandas 2.2 and 2.3 having the same result.
### Why are the changes needed?
The previous fix split pandas 2.2 and pandas 2.3 behavior for `GroupBy.idxmax(skipna=False)` and `GroupBy.idxmin(skipna=False)` on NA-containing input.
For example:
```python
pdf = pd.DataFrame({"a": [1, 1, 2, 2], "b": [1, None, 3, 4], "c": [4, 3, 2, 1]})
pdf.groupby(["a"]).idxmax(skipna=False).sort_index()
```
In pandas 2.2, this returns:
```python
b c
a
1 0 0
2 3 2
```
In pandas 2.3, this returns:
```python
b c
a
1 NaN 0
2 3.0 2
```
In pandas 3, this raises `ValueError`.
Instead of matching the pandas 2.2 / 2.3 difference, this PR keeps the legacy pandas 2 behavior across all pandas 2 environments and continues to follow the pandas 3 behavior in pandas 3 environments.
### Does this PR introduce _any_ user-facing change?
Yes.
In pandas-on-Spark with pandas 2.x, `GroupBy.idxmax(skipna=False)` and `GroupBy.idxmin(skipna=False)` on NA-containing groups now consistently keep the legacy result behavior instead of varying with the installed pandas 2 version.
For pandas 3, behavior is unchanged from the current implementation.
### How was this patch tested?
Ran the related pandas-on-Spark regression test in three environments:
- pandas 2.2: `GroupbyIndexTests.test_idxmax_idxmin_skipna_false_with_na`
- pandas 2.3: `GroupbyIndexTests.test_idxmax_idxmin_skipna_false_with_na`
- pandas 3.0: `GroupbyIndexTests.test_idxmax_idxmin_skipna_false_with_na`
### Was this patch authored or co-authored using generative AI tooling?
Generated-by: Codex (GPT-5)
Closes #55121 from ueshin/issues/SPARK-56219/pd2.2.
Authored-by: Takuya Ueshin <ueshin@databricks.com>
Signed-off-by: Takuya Ueshin <ueshin@databricks.com>
Spark is a unified analytics engine for large-scale data processing. It provides high-level APIs in Scala, Java, Python, and R (Deprecated), and an optimized engine that supports general computation graphs for data analysis. It also supports a rich set of higher-level tools including Spark SQL for SQL and DataFrames, pandas API on Spark for pandas workloads, MLlib for machine learning, GraphX for graph processing, and Structured Streaming for stream processing.
You can find the latest Spark documentation, including a programming guide, on the project web page. This README file only contains basic setup instructions.
| Branch | Status |
|---|---|
| master |  |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
 | |
| branch-4.1 |  |
 | |
 | |
 | |
 | |
 | |
 | |
| branch-4.0 |  |
 | |
 | |
 | |
 | |
 | |
 | |
| branch-3.5 |  |
 |
Spark is built using Apache Maven. To build Spark and its example programs, run:
./build/mvn -DskipTests clean package
(You do not need to do this if you downloaded a pre-built package.)
More detailed documentation is available from the project site, at “Building Spark”.
For general development tips, including info on developing Spark using an IDE, see “Useful Developer Tools”.
The easiest way to start using Spark is through the Scala shell:
./bin/spark-shell
Try the following command, which should return 1,000,000,000:
scala> spark.range(1000 * 1000 * 1000).count()
Alternatively, if you prefer Python, you can use the Python shell:
./bin/pyspark
And run the following command, which should also return 1,000,000,000:
>>> spark.range(1000 * 1000 * 1000).count()
Spark also comes with several sample programs in the examples directory. To run one of them, use ./bin/run-example <class> [params]. For example:
./bin/run-example SparkPi
will run the Pi example locally.
You can set the MASTER environment variable when running examples to submit examples to a cluster. This can be spark:// URL, “yarn” to run on YARN, and “local” to run locally with one thread, or “local[N]” to run locally with N threads. You can also use an abbreviated class name if the class is in the examples package. For instance:
MASTER=spark://host:7077 ./bin/run-example SparkPi
Many of the example programs print usage help if no params are given.
Testing first requires building Spark. Once Spark is built, tests can be run using:
./dev/run-tests
Please see the guidance on how to run tests for a module, or individual tests.
There is also a Kubernetes integration test, see resource-managers/kubernetes/integration-tests/README.md
Spark uses the Hadoop core library to talk to HDFS and other Hadoop-supported storage systems. Because the protocols have changed in different versions of Hadoop, you must build Spark against the same version that your cluster runs.
Please refer to the build documentation at “Specifying the Hadoop Version and Enabling YARN” for detailed guidance on building for a particular distribution of Hadoop, including building for particular Hive and Hive Thriftserver distributions.
Please refer to the Configuration Guide in the online documentation for an overview on how to configure Spark.
Please review the Contribution to Spark guide for information on how to get started contributing to the project.
