[SPARK-54046][INFRA] Upgrade PyArrow to 22.0.0

### What changes were proposed in this pull request?

This PR aims to upgrade `PyArrow` to 22.0.0.

### Why are the changes needed?

To test against the latest `PyArrow` version. `PyArrow 22.0.0` is the first version to support `Python 3.14`.
- https://pypi.org/project/pyarrow/22.0.0/ (2025-10-24)

### Does this PR introduce _any_ user-facing change?

No.

### How was this patch tested?

Pass the CIs.

### Was this patch authored or co-authored using generative AI tooling?

No.

Closes #52748 from dongjoon-hyun/SPARK-54046.

Authored-by: Dongjoon Hyun <[email protected]>
Signed-off-by: Dongjoon Hyun <[email protected]>

Author: dongjoon-hyun

.github/workflows/python_hosted_runner_test.yml

@@ -147,7 +147,7 @@ jobs:
         run: |
           python${{matrix.python}} -m pip install --ignore-installed 'blinker>=1.6.2'
           python${{matrix.python}} -m pip install --ignore-installed 'six==1.16.0'
-          python${{matrix.python}} -m pip install numpy 'pyarrow>=21.0.0' 'six==1.16.0' 'pandas==2.3.3' scipy 'plotly<6.0.0' 'mlflow>=2.8.1' coverage matplotlib openpyxl 'memory-profiler>=0.61.0' 'scikit-learn>=1.3.2' unittest-xml-reporting && \
+          python${{matrix.python}} -m pip install numpy 'pyarrow>=22.0.0' 'six==1.16.0' 'pandas==2.3.3' scipy 'plotly<6.0.0' 'mlflow>=2.8.1' coverage matplotlib openpyxl 'memory-profiler>=0.61.0' 'scikit-learn>=1.3.2' unittest-xml-reporting && \
           python${{matrix.python}} -m pip install 'grpcio==1.76.0' 'grpcio-status==1.76.0' 'protobuf==6.33.0' 'googleapis-common-protos==1.71.0' 'graphviz==0.20.3' && \
           python${{matrix.python}} -m pip cache purge
       - name: List Python packages

dev/spark-test-image/lint/Dockerfile

@@ -94,7 +94,7 @@ RUN python3.11 -m pip install \
     'pandas' \
     'pandas-stubs==1.2.0.53' \
     'plotly>=4.8' \
-    'pyarrow>=21.0.0' \
+    'pyarrow>=22.0.0' \
     'pytest-mypy-plugins==1.9.3' \
     'pytest==7.1.3' \
     && python3.11 -m pip install torch torchvision --index-url https://download.pytorch.org/whl/cpu \

dev/spark-test-image/numpy-213/Dockerfile

@@ -69,7 +69,7 @@ RUN apt-get update && apt-get install -y \
 
 
 # Pin numpy==2.1.3
-ARG BASIC_PIP_PKGS="numpy==2.1.3 pyarrow>=21.0.0 six==1.16.0 pandas==2.2.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy==2.1.3 pyarrow>=22.0.0 six==1.16.0 pandas==2.2.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
 # Python deps for Spark Connect
 ARG CONNECT_PIP_PKGS="grpcio==1.76.0 grpcio-status==1.76.0 protobuf==6.33.0 googleapis-common-protos==1.71.0 graphviz==0.20.3"
 

dev/spark-test-image/python-310/Dockerfile

@@ -64,7 +64,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 
-ARG BASIC_PIP_PKGS="numpy pyarrow>=21.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy pyarrow>=22.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
 # Python deps for Spark Connect
 ARG CONNECT_PIP_PKGS="grpcio==1.76.0 grpcio-status==1.76.0 protobuf==6.33.0 googleapis-common-protos==1.71.0 graphviz==0.20.3"
 

dev/spark-test-image/python-311-classic-only/Dockerfile

@@ -68,7 +68,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 
-ARG BASIC_PIP_PKGS="numpy pyarrow>=21.0.0 pandas==2.3.3 plotly<6.0.0 matplotlib openpyxl memory-profiler>=0.61.0 mlflow>=2.8.1 scipy scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy pyarrow>=22.0.0 pandas==2.3.3 plotly<6.0.0 matplotlib openpyxl memory-profiler>=0.61.0 mlflow>=2.8.1 scipy scikit-learn>=1.3.2"
 ARG TEST_PIP_PKGS="coverage unittest-xml-reporting"
 
 # Install Python 3.11 packages

dev/spark-test-image/python-311/Dockerfile

@@ -68,7 +68,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 
-ARG BASIC_PIP_PKGS="numpy pyarrow>=21.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy pyarrow>=22.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
 # Python deps for Spark Connect
 ARG CONNECT_PIP_PKGS="grpcio==1.76.0 grpcio-status==1.76.0 protobuf==6.33.0 googleapis-common-protos==1.71.0 graphviz==0.20.3"
 

dev/spark-test-image/python-312/Dockerfile

@@ -68,7 +68,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 
-ARG BASIC_PIP_PKGS="numpy pyarrow>=21.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy pyarrow>=22.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
 # Python deps for Spark Connect
 ARG CONNECT_PIP_PKGS="grpcio==1.76.0 grpcio-status==1.76.0 protobuf==6.33.0 googleapis-common-protos==1.71.0 graphviz==0.20.3"
 

dev/spark-test-image/python-313-nogil/Dockerfile

@@ -68,7 +68,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 
-ARG BASIC_PIP_PKGS="numpy pyarrow>=21.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy pyarrow>=22.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
 ARG CONNECT_PIP_PKGS="grpcio==1.76.0 grpcio-status==1.76.0 protobuf==6.33.0 googleapis-common-protos==1.71.0 graphviz==0.20.3"
 
 

dev/spark-test-image/python-313/Dockerfile

@@ -68,7 +68,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 
-ARG BASIC_PIP_PKGS="numpy pyarrow>=21.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
+ARG BASIC_PIP_PKGS="numpy pyarrow>=22.0.0 six==1.16.0 pandas==2.3.3 scipy plotly<6.0.0 mlflow>=2.8.1 coverage matplotlib openpyxl memory-profiler>=0.61.0 scikit-learn>=1.3.2"
 # Python deps for Spark Connect
 ARG CONNECT_PIP_PKGS="grpcio==1.76.0 grpcio-status==1.76.0 protobuf==6.33.0 googleapis-common-protos==1.71.0 graphviz==0.20.3"
 

python/pyspark/pandas/tests/io/test_feather.py

@@ -20,6 +20,7 @@
 import sys
 
 from pyspark import pandas as ps
+from pyspark.loose_version import LooseVersion
 from pyspark.testing.pandasutils import PandasOnSparkTestCase, TestUtils
 
 
@@ -35,7 +36,16 @@ def pdf(self):
     def psdf(self):
         return ps.from_pandas(self.pdf)
 
-    @unittest.skipIf(sys.version_info > (3, 13), "SPARK-54068")
+    has_arrow_21_or_below = False
+    try:
+        import pyarrow as pa
+
+        if LooseVersion(pa.__version__) < LooseVersion("22.0.0"):
+            has_arrow_21_or_below = True
+    except ImportError:
+        pass
+
+    @unittest.skipIf(not has_arrow_21_or_below, "SPARK-54068")
     def test_to_feather(self):
         with self.temp_dir() as dirpath:
             path1 = f"{dirpath}/file1.feather"