BLD: Shrink sdist/wheel sizes

.gitattributes

@@ -14,3 +14,71 @@
 *.xls binary
 *.xlsx binary
 pandas/_version.py export-subst
+
+
+*.bz2 export-ignore
+*.csv export-ignore
+*.data export-ignore
+*.dta export-ignore
+*.feather export-ignore
+*.tar export-ignore
+*.gz export-ignore
+*.h5 export-ignore
+*.html export-ignore
+*.json export-ignore
+*.jsonl export-ignore
+*.kml export-ignore
+*.msgpack export-ignore
+*.pdf export-ignore
+*.parquet export-ignore
+*.pickle export-ignore
+*.pkl export-ignore
+*.png export-ignore
+*.pptx export-ignore
+*.ods export-ignore
+*.odt export-ignore
+*.orc export-ignore
+*.sas7bdat export-ignore
+*.sav export-ignore
+*.so export-ignore
+*.txt export-ignore
+*.xls export-ignore
+*.xlsb export-ignore
+*.xlsm export-ignore
+*.xlsx export-ignore
+*.xpt export-ignore
+*.cpt export-ignore
+*.xml export-ignore
+*.xsl export-ignore
+*.xz export-ignore
+*.zip export-ignore
+*.zst export-ignore
+*~ export-ignore
+.DS_Store export-ignore
+.git* export-ignore
+
+*.py[ocd] export-ignore
+*.pxi export-ignore
+
+# Ignoring stuff from the top level
+.circleci export-ignore
+.github export-ignore
+asv_bench export-ignore
+ci export-ignore
+doc export-ignore
+gitpod export-ignore
+MANIFEST.in export-ignore
+scripts export-ignore
+typings export-ignore
+web export-ignore
+CITATION.cff export-ignore
+codecov.yml export-ignore
+Dockerfile export-ignore
+environment.yml export-ignore
+setup.py export-ignore
+
+
+# GH 39321
+# csv_dir_path fixture checks the existence of the directory
+# exclude the whole directory to avoid running related tests in sdist
+pandas/tests/io/parser/data export-ignore

.github/workflows/wheels.yml

@@ -104,6 +104,8 @@ jobs:
         with:
           fetch-depth: 0
 
+      # We need to build wheels from the sdist since the sdist
+      # removes unnecessary files from the release
       - name: Download sdist
         uses: actions/download-artifact@v3
         with:
@@ -115,8 +117,8 @@ jobs:
         # TODO: Build wheels from sdist again
         # There's some sort of weird race condition?
         # within Github that makes the sdist be missing files
-        #with:
-        #  package-dir: ./dist/${{ needs.build_sdist.outputs.sdist_file }}
+        with:
+         package-dir: ./dist/${{ needs.build_sdist.outputs.sdist_file }}
         env:
           CIBW_PRERELEASE_PYTHONS: True
           CIBW_BUILD: ${{ matrix.python[0] }}-${{ matrix.buildplat[1] }}
@@ -144,7 +146,7 @@ jobs:
           $TST_CMD = @"
           python -m pip install pytz six numpy python-dateutil tzdata>=2022.1 hypothesis>=6.46.1 pytest>=7.3.2 pytest-xdist>=2.2.0 pytest-asyncio>=0.17;
           python -m pip install --find-links=pandas\wheelhouse --no-index pandas;
-          python -c `'import pandas as pd; pd.test()`';
+          python -c `'import pandas as pd; pd.test(extra_args=[\"`\"--no-strict-data-files`\"\", \"`\"-m not clipboard and not single_cpu and not slow and not network and not db`\"\"])`';
           "@
           docker pull python:${{ matrix.python[1] }}-windowsservercore
           docker run --env PANDAS_CI='1' -v ${PWD}:C:\pandas python:${{ matrix.python[1] }}-windowsservercore powershell -Command $TST_CMD

pandas/conftest.py

@@ -102,9 +102,9 @@
 
 def pytest_addoption(parser) -> None:
     parser.addoption(
-        "--strict-data-files",
-        action="store_true",
-        help="Fail if a test is skipped for missing data file.",
+        "--no-strict-data-files",
+        action="store_false",
+        help="Don't fail if a test is skipped for missing data file.",
     )
 
 
@@ -1172,9 +1172,9 @@ def all_numeric_accumulations(request):
 @pytest.fixture
 def strict_data_files(pytestconfig):
     """
-    Returns the configuration for the test setting `--strict-data-files`.
+    Returns the configuration for the test setting `--no-strict-data-files`.
     """
-    return pytestconfig.getoption("--strict-data-files")
+    return pytestconfig.getoption("--no-strict-data-files")
 
 
 @pytest.fixture
@@ -1204,7 +1204,7 @@ def datapath(strict_data_files: str) -> Callable[..., str]:
     Raises
     ------
     ValueError
-        If the path doesn't exist and the --strict-data-files option is set.
+        If the path doesn't exist and the --no-strict-data-files option is not set.
     """
     BASE_PATH = os.path.join(os.path.dirname(__file__), "tests")
 
@@ -1213,7 +1213,7 @@ def deco(*args):
         if not os.path.exists(path):
             if strict_data_files:
                 raise ValueError(
-                    f"Could not find file {path} and --strict-data-files is set."
+                    f"Could not find file {path} and --no-strict-data-files is not set."
                 )
             pytest.skip(f"Could not find {path}.")
         return path

pandas/tests/io/xml/conftest.py

@@ -2,35 +2,35 @@
 
 
 @pytest.fixture
-def xml_data_path(tests_io_data_path):
+def xml_data_path(tests_io_data_path, datapath):
     return tests_io_data_path / "xml"
 
 
 @pytest.fixture
-def xml_books(xml_data_path):
-    return xml_data_path / "books.xml"
+def xml_books(xml_data_path, datapath):
+    return datapath(xml_data_path / "books.xml")
 
 
 @pytest.fixture
-def xml_doc_ch_utf(xml_data_path):
-    return xml_data_path / "doc_ch_utf.xml"
+def xml_doc_ch_utf(xml_data_path, datapath):
+    return datapath(xml_data_path / "doc_ch_utf.xml")
 
 
 @pytest.fixture
-def xml_baby_names(xml_data_path):
-    return xml_data_path / "baby_names.xml"
+def xml_baby_names(xml_data_path, datapath):
+    return datapath(xml_data_path / "baby_names.xml")
 
 
 @pytest.fixture
-def kml_cta_rail_lines(xml_data_path):
-    return xml_data_path / "cta_rail_lines.kml"
+def kml_cta_rail_lines(xml_data_path, datapath):
+    return datapath(xml_data_path / "cta_rail_lines.kml")
 
 
 @pytest.fixture
-def xsl_flatten_doc(xml_data_path):
-    return xml_data_path / "flatten_doc.xsl"
+def xsl_flatten_doc(xml_data_path, datapath):
+    return datapath(xml_data_path / "flatten_doc.xsl")
 
 
 @pytest.fixture
-def xsl_row_field_output(xml_data_path):
-    return xml_data_path / "row_field_output.xsl"
+def xsl_row_field_output(xml_data_path, datapath):
+    return datapath(xml_data_path / "row_field_output.xsl")

pyproject.toml

@@ -153,8 +153,8 @@ environment = {LDFLAGS="-Wl,--strip-all"}
 test-requires = "hypothesis>=6.46.1 pytest>=7.3.2 pytest-xdist>=2.2.0 pytest-asyncio>=0.17"
 test-command = """
   PANDAS_CI='1' python -c 'import pandas as pd; \
-  pd.test(extra_args=["-m not clipboard and not single_cpu and not slow and not network and not db", "-n 2"]); \
-  pd.test(extra_args=["-m not clipboard and single_cpu and not slow and not network and not db"]);' \
+  pd.test(extra_args=["-m not clipboard and not single_cpu and not slow and not network and not db", "-n 2", "--no-strict-data-files"]); \
+  pd.test(extra_args=["-m not clipboard and single_cpu and not slow and not network and not db", "--no-strict-data-files"]);' \
   """
 
 [tool.cibuildwheel.macos]
@@ -471,7 +471,7 @@ disable = [
 [tool.pytest.ini_options]
 # sync minversion with pyproject.toml & install.rst
 minversion = "7.3.2"
-addopts = "--strict-data-files --strict-markers --strict-config --capture=no --durations=30 --junitxml=test-data.xml"
+addopts = "--strict-markers --strict-config --capture=no --durations=30 --junitxml=test-data.xml"
 empty_parameter_set_mark = "fail_at_collect"
 xfail_strict = true
 testpaths = "pandas"