From 9d138ba626f4489958c04f9eaeb35a2299e4341b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Cumplido?= <raulcumplido@gmail.com>
Date: Fri, 13 May 2022 11:55:27 +0200
Subject: [PATCH 1/2] ARROW-16548: [Python] Add pytest.mark.parquet to all
 tests under tests/parquet package

---
 python/pyarrow/tests/parquet/common.py   | 4 ----
 python/pyarrow/tests/parquet/conftest.py | 8 ++++++++
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/python/pyarrow/tests/parquet/common.py b/python/pyarrow/tests/parquet/common.py
index b11e412e63d..cbff41c7b10 100644
--- a/python/pyarrow/tests/parquet/common.py
+++ b/python/pyarrow/tests/parquet/common.py
@@ -43,10 +43,6 @@
      pytest.param(False, marks=pytest.mark.dataset)]
 )
 
-# Marks all of the tests in this module
-# Ignore these with pytest ... -m 'not parquet'
-pytestmark = pytest.mark.parquet
-
 
 def _write_table(table, path, **kwargs):
     # So we see the ImportError somewhere
diff --git a/python/pyarrow/tests/parquet/conftest.py b/python/pyarrow/tests/parquet/conftest.py
index 1e75493cdae..338fe7110b2 100644
--- a/python/pyarrow/tests/parquet/conftest.py
+++ b/python/pyarrow/tests/parquet/conftest.py
@@ -25,6 +25,14 @@ def datadir(base_datadir):
     return base_datadir / 'parquet'
 
 
+def pytest_collection_modifyitems(items):
+    for item in items:
+        # Marks all tests in the parquet package
+        # Ignore these with pytest ... -m 'not parquet'c
+        if item.parent.parent.name == "parquet":
+            item.add_marker(pytest.mark.parquet)
+
+
 @pytest.fixture
 def s3_bucket(s3_server):
     boto3 = pytest.importorskip('boto3')

From a2276f4b3300f65e3d28ad6707ab02492c97f6a2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ra=C3=BAl=20Cumplido?= <raulcumplido@gmail.com>
Date: Fri, 13 May 2022 14:53:48 +0200
Subject: [PATCH 2/2] Remove manual test marking via
 pytest_collection_modifyitems and add the pytest mark on each parquet python
 file

---
 python/pyarrow/tests/parquet/conftest.py                  | 8 --------
 python/pyarrow/tests/parquet/test_basic.py                | 5 +++++
 .../pyarrow/tests/parquet/test_compliant_nested_type.py   | 5 +++++
 python/pyarrow/tests/parquet/test_data_types.py           | 5 +++++
 python/pyarrow/tests/parquet/test_dataset.py              | 5 +++++
 python/pyarrow/tests/parquet/test_datetime.py             | 5 +++++
 python/pyarrow/tests/parquet/test_encryption.py           | 2 ++
 python/pyarrow/tests/parquet/test_metadata.py             | 5 +++++
 python/pyarrow/tests/parquet/test_pandas.py               | 5 +++++
 python/pyarrow/tests/parquet/test_parquet_file.py         | 5 +++++
 python/pyarrow/tests/parquet/test_parquet_writer.py       | 5 +++++
 11 files changed, 47 insertions(+), 8 deletions(-)

diff --git a/python/pyarrow/tests/parquet/conftest.py b/python/pyarrow/tests/parquet/conftest.py
index 338fe7110b2..1e75493cdae 100644
--- a/python/pyarrow/tests/parquet/conftest.py
+++ b/python/pyarrow/tests/parquet/conftest.py
@@ -25,14 +25,6 @@ def datadir(base_datadir):
     return base_datadir / 'parquet'
 
 
-def pytest_collection_modifyitems(items):
-    for item in items:
-        # Marks all tests in the parquet package
-        # Ignore these with pytest ... -m 'not parquet'c
-        if item.parent.parent.name == "parquet":
-            item.add_marker(pytest.mark.parquet)
-
-
 @pytest.fixture
 def s3_bucket(s3_server):
     boto3 = pytest.importorskip('boto3')
diff --git a/python/pyarrow/tests/parquet/test_basic.py b/python/pyarrow/tests/parquet/test_basic.py
index 6712b7fc459..9d31bdeb603 100644
--- a/python/pyarrow/tests/parquet/test_basic.py
+++ b/python/pyarrow/tests/parquet/test_basic.py
@@ -46,6 +46,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 def test_parquet_invalid_version(tempdir):
     table = pa.table({'a': [1, 2, 3]})
     with pytest.raises(ValueError, match="Unsupported Parquet format version"):
diff --git a/python/pyarrow/tests/parquet/test_compliant_nested_type.py b/python/pyarrow/tests/parquet/test_compliant_nested_type.py
index 31c62a288db..b60b0273b21 100644
--- a/python/pyarrow/tests/parquet/test_compliant_nested_type.py
+++ b/python/pyarrow/tests/parquet/test_compliant_nested_type.py
@@ -36,6 +36,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 # Tests for ARROW-11497
 _test_data_simple = [
     {'items': [1, 2]},
diff --git a/python/pyarrow/tests/parquet/test_data_types.py b/python/pyarrow/tests/parquet/test_data_types.py
index 7148af944f9..13d94b6a256 100644
--- a/python/pyarrow/tests/parquet/test_data_types.py
+++ b/python/pyarrow/tests/parquet/test_data_types.py
@@ -44,6 +44,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 # General roundtrip of data types
 # -----------------------------------------------------------------------------
 
diff --git a/python/pyarrow/tests/parquet/test_dataset.py b/python/pyarrow/tests/parquet/test_dataset.py
index 2c660a3f6e3..99d2f6d1def 100644
--- a/python/pyarrow/tests/parquet/test_dataset.py
+++ b/python/pyarrow/tests/parquet/test_dataset.py
@@ -47,6 +47,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 @pytest.mark.pandas
 def test_parquet_piece_read(tempdir):
     df = _test_dataframe(1000)
diff --git a/python/pyarrow/tests/parquet/test_datetime.py b/python/pyarrow/tests/parquet/test_datetime.py
index 18f78d28892..e10d4fd776d 100644
--- a/python/pyarrow/tests/parquet/test_datetime.py
+++ b/python/pyarrow/tests/parquet/test_datetime.py
@@ -41,6 +41,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 @pytest.mark.pandas
 @parametrize_legacy_dataset
 def test_pandas_parquet_datetime_tz(use_legacy_dataset):
diff --git a/python/pyarrow/tests/parquet/test_encryption.py b/python/pyarrow/tests/parquet/test_encryption.py
index fe28b1e0f42..f5e9e76d443 100644
--- a/python/pyarrow/tests/parquet/test_encryption.py
+++ b/python/pyarrow/tests/parquet/test_encryption.py
@@ -38,7 +38,9 @@
 
 # Marks all of the tests in this module
 # Ignore these with pytest ... -m 'not parquet_encryption'
+# Ignore these with pytest ... -m 'not parquet'
 pytestmark = pytest.mark.parquet_encryption
+pytestmark = pytest.mark.parquet
 
 
 @pytest.fixture(scope='module')
diff --git a/python/pyarrow/tests/parquet/test_metadata.py b/python/pyarrow/tests/parquet/test_metadata.py
index 35d37e47ca4..bad49847968 100644
--- a/python/pyarrow/tests/parquet/test_metadata.py
+++ b/python/pyarrow/tests/parquet/test_metadata.py
@@ -41,6 +41,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 @pytest.mark.pandas
 def test_parquet_metadata_api():
     df = alltypes_sample(size=10000)
diff --git a/python/pyarrow/tests/parquet/test_pandas.py b/python/pyarrow/tests/parquet/test_pandas.py
index 68253ae0b2b..2c37ac486e2 100644
--- a/python/pyarrow/tests/parquet/test_pandas.py
+++ b/python/pyarrow/tests/parquet/test_pandas.py
@@ -46,6 +46,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 @pytest.mark.pandas
 def test_pandas_parquet_custom_metadata(tempdir):
     df = alltypes_sample(size=10000)
diff --git a/python/pyarrow/tests/parquet/test_parquet_file.py b/python/pyarrow/tests/parquet/test_parquet_file.py
index 621cfa6d749..9b528b1859a 100644
--- a/python/pyarrow/tests/parquet/test_parquet_file.py
+++ b/python/pyarrow/tests/parquet/test_parquet_file.py
@@ -37,6 +37,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 @pytest.mark.pandas
 def test_pass_separate_metadata():
     # ARROW-471
diff --git a/python/pyarrow/tests/parquet/test_parquet_writer.py b/python/pyarrow/tests/parquet/test_parquet_writer.py
index 43fa1f94dbe..6ee3f3b606b 100644
--- a/python/pyarrow/tests/parquet/test_parquet_writer.py
+++ b/python/pyarrow/tests/parquet/test_parquet_writer.py
@@ -37,6 +37,11 @@
     pd = tm = None
 
 
+# Marks all of the tests in this module
+# Ignore these with pytest ... -m 'not parquet'
+pytestmark = pytest.mark.parquet
+
+
 @pytest.mark.pandas
 @parametrize_legacy_dataset
 def test_parquet_incremental_file_build(tempdir, use_legacy_dataset):