Backport PR #45732: TST: Ensure tm.network has pytest.mark.network (#45741)

Author: mroeschke

pandas/tests/io/conftest.py

@@ -114,7 +114,7 @@ def s3_base(worker_id):
                 proc.terminate()
 
 
-@pytest.fixture()
+@pytest.fixture
 def s3_resource(s3_base, tips_file, jsonl_file, feather_file):
     """
     Sets up S3 bucket with contents

pandas/tests/io/excel/test_readers.py

@@ -763,6 +763,7 @@ def test_corrupt_bytes_raises(self, read_ext, engine):
         with pytest.raises(error, match=msg):
             pd.read_excel(bad_stream)
 
+    @pytest.mark.network
     @tm.network
     def test_read_from_http_url(self, read_ext):
         url = (

pandas/tests/io/json/data/teams.csv

@@ -987,18 +987,16 @@ def test_misc_example(self):
         expected = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
         tm.assert_frame_equal(result, expected)
 
-    @tm.network
-    @pytest.mark.single
-    def test_round_trip_exception_(self):
+    def test_round_trip_exception_(self, datapath):
         # GH 3867
-        csv = "https://raw.github.com/hayd/lahman2012/master/csvs/Teams.csv"
-        df = pd.read_csv(csv)
+        path = datapath("io", "json", "data", "teams.csv")
+        df = pd.read_csv(path)
         s = df.to_json()
         result = read_json(s)
         tm.assert_frame_equal(result.reindex(index=df.index, columns=df.columns), df)
 
+    @pytest.mark.network
     @tm.network
-    @pytest.mark.single
     @pytest.mark.parametrize(
         "field,dtype",
         [

pandas/tests/io/json/test_pandas.py

@@ -26,6 +26,7 @@
 pytestmark = pytest.mark.usefixtures("pyarrow_skip")
 
 
+@pytest.mark.network
 @tm.network
 def test_url(all_parsers, csv_dir_path):
     parser = all_parsers

pandas/tests/io/parser/common/test_file_buffer_url.py

@@ -22,17 +22,13 @@
 
 
 @pytest.mark.network
+@tm.network
 @pytest.mark.parametrize("mode", ["explicit", "infer"])
 @pytest.mark.parametrize("engine", ["python", "c"])
 def test_compressed_urls(salaries_table, mode, engine, compression_only):
-    extension = icom._compression_to_extension[compression_only]
-    check_compressed_urls(salaries_table, compression_only, extension, mode, engine)
-
-
-@tm.network
-def check_compressed_urls(salaries_table, compression, extension, mode, engine):
     # test reading compressed urls with various engines and
     # extension inference
+    extension = icom._compression_to_extension[compression_only]
     base_url = (
         "https://github.com/pandas-dev/pandas/raw/main/"
         "pandas/tests/io/parser/data/salaries.csv"
@@ -41,13 +37,14 @@ def check_compressed_urls(salaries_table, compression, extension, mode, engine):
     url = base_url + extension
 
     if mode != "explicit":
-        compression = mode
+        compression_only = mode
 
-    url_table = read_csv(url, sep="\t", compression=compression, engine=engine)
+    url_table = read_csv(url, sep="\t", compression=compression_only, engine=engine)
     tm.assert_frame_equal(url_table, salaries_table)
 
 
-@tm.network("https://raw.githubusercontent.com/", check_before_test=True)
+@pytest.mark.network
+@tm.network
 def test_url_encoding_csv():
     """
     read_csv should honor the requested encoding for URLs.

pandas/tests/io/parser/test_network.py

@@ -181,6 +181,7 @@ def test_passthrough_keywords(self):
         df = tm.makeDataFrame().reset_index()
         self.check_round_trip(df, write_kwargs={"version": 1})
 
+    @pytest.mark.network
     @tm.network
     def test_http_path(self, feather_file):
         # GH 29055

pandas/tests/io/test_feather.py

@@ -133,6 +133,7 @@ def test_to_html_compat(self):
         res = self.read_html(out, attrs={"class": "dataframe"}, index_col=0)[0]
         tm.assert_frame_equal(res, df)
 
+    @pytest.mark.network
     @tm.network
     def test_banklist_url_positional_match(self):
         url = "http://www.fdic.gov/resources/resolutions/bank-failures/failed-bank-list/index.html"  # noqa E501
@@ -152,6 +153,7 @@ def test_banklist_url_positional_match(self):
 
         assert_framelist_equal(df1, df2)
 
+    @pytest.mark.network
     @tm.network
     def test_banklist_url(self):
         url = "http://www.fdic.gov/resources/resolutions/bank-failures/failed-bank-list/index.html"  # noqa E501
@@ -168,6 +170,7 @@ def test_banklist_url(self):
 
         assert_framelist_equal(df1, df2)
 
+    @pytest.mark.network
     @tm.network
     def test_spam_url(self):
         url = (
@@ -316,13 +319,15 @@ def test_file_like(self):
 
         assert_framelist_equal(df1, df2)
 
+    @pytest.mark.network
     @tm.network
     def test_bad_url_protocol(self):
         with pytest.raises(URLError, match="urlopen error unknown url type: git"):
             self.read_html("git://github.com", match=".*Water.*")
 
-    @tm.network
     @pytest.mark.slow
+    @pytest.mark.network
+    @tm.network
     def test_invalid_url(self):
         msg = (
             "Name or service not known|Temporary failure in name resolution|"
@@ -403,12 +408,14 @@ def test_negative_skiprows(self):
         with pytest.raises(ValueError, match=msg):
             self.read_html(self.spam_data, match="Water", skiprows=-1)
 
+    @pytest.mark.network
     @tm.network
     def test_multiple_matches(self):
         url = "https://docs.python.org/2/"
         dfs = self.read_html(url, match="Python")
         assert len(dfs) > 1
 
+    @pytest.mark.network
     @tm.network
     def test_python_docs_table(self):
         url = "https://docs.python.org/2/"

pandas/tests/io/test_html.py

@@ -379,6 +379,7 @@ def check_external_error_on_write(self, df, engine, exc):
             with tm.external_error_raised(exc):
                 to_parquet(df, path, engine, compression=None)
 
+    @pytest.mark.network
     @tm.network
     def test_parquet_read_from_url(self, df_compat, engine):
         if engine != "auto":

pandas/tests/io/test_parquet.py

@@ -21,17 +21,19 @@ def test_streaming_s3_objects():
         read_csv(body)
 
 
-@tm.network
 @td.skip_if_no("s3fs")
+@pytest.mark.network
+@tm.network
 def test_read_without_creds_from_pub_bucket():
     # GH 34626
     # Use Amazon Open Data Registry - https://registry.opendata.aws/gdelt
     result = read_csv("s3://gdelt-open-data/events/1981.csv", nrows=3)
     assert len(result) == 3
 
 
-@tm.network
 @td.skip_if_no("s3fs")
+@pytest.mark.network
+@tm.network
 def test_read_with_creds_from_pub_bucket():
     # Ensure we can read from a public bucket with credentials
     # GH 34626

pandas/tests/io/test_s3.py

@@ -1331,10 +1331,10 @@ def test_unsuported_compression(datapath, parser):
 # STORAGE OPTIONS
 
 
-@tm.network
 @td.skip_if_no("s3fs")
 @td.skip_if_no("lxml")
-def test_s3_permission_output(parser):
+def test_s3_permission_output(parser, s3_resource):
+    # s3_resource hosts pandas-test
     import s3fs
 
     with pytest.raises(PermissionError, match="Access Denied"):

pandas/tests/io/xml/test_to_xml.py

@@ -254,10 +254,11 @@ def test_parser_consistency_file(datapath):
     tm.assert_frame_equal(df_file_lxml, df_file_etree)
 
 
-@tm.network
+@pytest.mark.network
 @pytest.mark.slow
 @td.skip_if_no("lxml")
-def test_parser_consistency_url(datapath):
+@tm.network
+def test_parser_consistency_url():
     url = (
         "https://data.cityofchicago.org/api/views/"
         "8pix-ypme/rows.xml?accessType=DOWNLOAD"
@@ -401,6 +402,7 @@ def test_wrong_file_path_etree():
         read_xml(filename, parser="etree")
 
 
+@pytest.mark.network
 @tm.network
 @td.skip_if_no("lxml")
 def test_url():
@@ -421,6 +423,7 @@ def test_url():
     tm.assert_frame_equal(df_url, df_expected)
 
 
+@pytest.mark.network
 @tm.network
 def test_wrong_url(parser):
     with pytest.raises(HTTPError, match=("HTTP Error 404: Not Found")):
@@ -1016,8 +1019,9 @@ def test_empty_stylesheet(val):
         read_xml(kml, stylesheet=val)
 
 
-@tm.network
+@pytest.mark.network
 @td.skip_if_no("lxml")
+@tm.network
 def test_online_stylesheet():
     xml = "https://www.w3schools.com/xml/cdcatalog_with_xsl.xml"
     xsl = "https://www.w3schools.com/xml/cdcatalog.xsl"
@@ -1099,9 +1103,14 @@ def test_unsuported_compression(datapath, parser):
 # STORAGE OPTIONS
 
 
-@tm.network
+@pytest.mark.network
 @td.skip_if_no("s3fs")
 @td.skip_if_no("lxml")
+@pytest.mark.skipif(
+    os.environ.get("PANDAS_CI", "0") == "1",
+    reason="2022.1.17: Hanging on the CI min versions build.",
+)
+@tm.network
 def test_s3_parser_consistency():
     # Python Software Foundation (2019 IRS-990 RETURN)
     s3 = "s3://irs-form-990/201923199349319487_public.xml"

pandas/tests/io/xml/test_xml.py

@@ -118,6 +118,7 @@ def test_oo_optimized_datetime_index_unpickle():
     )
 
 
+@pytest.mark.network
 @tm.network
 # Cython import warning
 @pytest.mark.filterwarnings("ignore:pandas.util.testing is deprecated")
@@ -158,6 +159,7 @@ def test_scikit_learn(df):
 
 
 # Cython import warning and traitlets
+@pytest.mark.network
 @tm.network
 @pytest.mark.filterwarnings("ignore")
 def test_seaborn():
@@ -172,12 +174,13 @@ def test_pandas_gbq(df):
     pandas_gbq = import_module("pandas_gbq")  # noqa:F841
 
 
+@pytest.mark.network
+@tm.network
 @pytest.mark.xfail(
     raises=ValueError,
     reason="The Quandl API key must be provided either through the api_key "
     "variable or through the environmental variable QUANDL_API_KEY",
 )
-@tm.network
 def test_pandas_datareader():
 
     pandas_datareader = import_module("pandas_datareader")