From 4d3900b54122303158028eb566b99d0dff970a10 Mon Sep 17 00:00:00 2001 From: Will Beason Date: Tue, 27 May 2025 14:27:49 -0500 Subject: [PATCH] Standardize calling for wikiq in tests This way failures show the output of stderr/etc. Also create path constant strings for use in tests to avoid repetition and make changes easier. Signed-off-by: Will Beason --- test/Wikiq_Unit_Test.py | 103 +++++++++++++++++++--------------------- 1 file changed, 50 insertions(+), 53 deletions(-) diff --git a/test/Wikiq_Unit_Test.py b/test/Wikiq_Unit_Test.py index 75cda8d..08fdfc4 100644 --- a/test/Wikiq_Unit_Test.py +++ b/test/Wikiq_Unit_Test.py @@ -13,6 +13,27 @@ from typing import Final TEST_DIR: Final[str] = os.path.dirname(os.path.realpath(__file__)) WIKIQ: Final[str] = os.path.join(os.path.dirname(TEST_DIR), "wikiq") TEST_OUTPUT_DIR: Final[str] = os.path.join(TEST_DIR, "test_output") +BASELINE_OUTPUT_DIR: Final[str] = os.path.join(TEST_DIR, "baseline_output") + +def setup(): + tracemalloc.start() + + # Perform directory check and reset here as this is a one-time setup step as opposed to per-test setup. + if not os.path.exists(TEST_OUTPUT_DIR): + os.mkdir(TEST_OUTPUT_DIR) + else: + # Avoid subsequent calls to tests interfering with each other. + # Otherwise, a test may erroneously pass if the program has no output + # but a previous run output what was expected. + for f in os.listdir(TEST_OUTPUT_DIR): + os.remove(os.path.join(TEST_OUTPUT_DIR, f)) + +setup() + +def call_wikiq(*args: str): + call = ' '.join([WIKIQ, *args]) + print(call) + subprocess.check_output(call, stderr=subprocess.PIPE, shell=True) # with / without pwr DONE # with / without url encode DONE @@ -26,16 +47,14 @@ TEST_OUTPUT_DIR: Final[str] = os.path.join(TEST_DIR, "test_output") class Test_Wikipedia(unittest.TestCase): def setUp(self): - self.wiki = 'ikwiki-20180301-pages-meta-history' - self.wikiq_out_name = self.wiki + ".tsv" + wiki = 'ikwiki-20180301-pages-meta-history' + self.wikiq_out_name = wiki + ".tsv" self.call_output = os.path.join(TEST_OUTPUT_DIR, self.wikiq_out_name) - self.infile = "{0}.xml.bz2".format(self.wiki) + infile = "{0}.xml.bz2".format(wiki) - self.base_call = WIKIQ + " {0} -o {1}" - self.input_dir = os.path.join(TEST_DIR, "dumps") - self.input_file = os.path.join(TEST_DIR, self.input_dir, self.infile) - self.baseline_output_dir = os.path.join(TEST_DIR, "baseline_output") + input_dir = os.path.join(TEST_DIR, "dumps") + self.input_file = os.path.join(TEST_DIR, input_dir, infile) def test_WP_url_encode(self): test_filename = "url-encode_" + self.wikiq_out_name @@ -43,17 +62,13 @@ class Test_Wikipedia(unittest.TestCase): if os.path.exists(test_file): os.remove(test_file) - call = self.base_call.format(self.input_file, TEST_OUTPUT_DIR) - call = call + " --url-encode" - print(call) try: - subprocess.check_output(call, stderr=subprocess.PIPE, shell=True) + call_wikiq(self.input_file, "-o", TEST_OUTPUT_DIR, "--url-encode") except subprocess.CalledProcessError as exc: - print(exc.stderr.decode("utf8")) - self.fail() + self.fail(exc.stderr.decode("utf8")) copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) # as a test let's make sure that we get equal data frames test = pd.read_table(test_file) @@ -67,15 +82,14 @@ class Test_Wikipedia(unittest.TestCase): if os.path.exists(test_file): os.remove(test_file) - call = self.base_call.format(self.input_file, TEST_OUTPUT_DIR) - call = call + " -n 0 -n 1" - print(call) - with subprocess.Popen(call, stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True) as proc: - proc.wait() - self.assertEqual(proc.returncode, 0) + try: + call_wikiq(self.input_file, "-o", TEST_OUTPUT_DIR, + "-n 0", "-n 1") + except subprocess.CalledProcessError as exc: + self.fail(exc.stderr.decode("utf8")) copyfile(self.call_output, test_file) - baseline_file = os.path.join(os.path.abspath("."), self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) # as a test let's make sure that we get equal data frames test = pd.read_table(test_file) @@ -91,15 +105,14 @@ class Test_Wikipedia(unittest.TestCase): if os.path.exists(test_file): os.remove(test_file) - call = self.base_call.format(self.input_file, TEST_OUTPUT_DIR) - call = call + " -n 0 -n 1 -rr 1" - print(call) - with subprocess.Popen(call, stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True) as proc: - proc.wait() - self.assertEqual(proc.returncode, 0) + try: + call_wikiq(self.input_file, "-o", TEST_OUTPUT_DIR, + "-n 0", "-n 1", "-rr 1") + except subprocess.CalledProcessError as exc: + self.fail(exc.stderr.decode("utf8")) copyfile(self.call_output, test_file) - baseline_file = os.path.join(os.path.abspath("."), self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) # as a test let's make sure that we get equal data frames test = pd.read_table(test_file) @@ -120,7 +133,6 @@ class Test_Basic(unittest.TestCase): self.base_call = WIKIQ + " {0} -o {1}" self.input_dir = os.path.join(TEST_DIR, "dumps") self.input_file = os.path.join(TEST_DIR, self.input_dir, self.infile) - self.baseline_output_dir = os.path.join(TEST_DIR, "baseline_output") def test_noargs(self): @@ -137,7 +149,7 @@ class Test_Basic(unittest.TestCase): copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) @@ -158,7 +170,7 @@ class Test_Basic(unittest.TestCase): copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) assert_frame_equal(test, baseline, check_like=True) @@ -178,7 +190,7 @@ class Test_Basic(unittest.TestCase): copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) @@ -199,7 +211,7 @@ class Test_Basic(unittest.TestCase): copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) @@ -220,7 +232,7 @@ class Test_Basic(unittest.TestCase): copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) @@ -242,7 +254,7 @@ class Test_Basic(unittest.TestCase): self.assertEqual(proc.returncode, 0) copyfile(self.call_output, test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) @@ -282,7 +294,6 @@ class Test_Stdout(unittest.TestCase): self.base_call = WIKIQ + " {0} --stdout" self.input_dir = os.path.join(TEST_DIR, "dumps") self.input_file = os.path.join(TEST_DIR, self.input_dir, self.infile) - self.baseline_output_dir = os.path.join(TEST_DIR, "baseline_output") def test_noargs(self): call = self.base_call.format(self.input_file) @@ -291,7 +302,7 @@ class Test_Stdout(unittest.TestCase): outs = proc.stdout.decode("utf8") test_file = "noargs_" + self.wikiq_out_name - baseline_file = os.path.join(".", self.baseline_output_dir, test_file) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_file) print(baseline_file) test = pd.read_table(StringIO(outs)) baseline = pd.read_table(baseline_file) @@ -312,8 +323,6 @@ class Test_Regex(unittest.TestCase): self.base_call = WIKIQ + " {0}" self.base_call_outs = WIKIQ + " {0} -o {1}" - self.baseline_output_dir = os.path.join(TEST_DIR, "baseline_output") - # sample arguments for checking that bad arguments get terminated / test_regex_arguments self.bad_arguments_list = [ # label is missing @@ -372,7 +381,7 @@ class Test_Regex(unittest.TestCase): test = pd.read_table(test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) baseline = pd.read_table(baseline_file) assert_frame_equal(test, baseline, check_like=True) print(i) @@ -397,22 +406,10 @@ class Test_Regex(unittest.TestCase): test = pd.read_table(test_file) - baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline_file = os.path.join(BASELINE_OUTPUT_DIR, test_filename) baseline = pd.read_table(baseline_file) assert_frame_equal(test, baseline, check_like=True) if __name__ == '__main__': - tracemalloc.start() - - # Perform directory check and reset here as this is a one-time setup step as opposed to per-test setup. - if not os.path.exists(TEST_OUTPUT_DIR): - os.mkdir(TEST_OUTPUT_DIR) - else: - # Avoid subsequent calls to tests interfering with each other. - # Otherwise, a test may erroneously pass if the program has no output - # but a previous run output what was expected. - for f in os.listdir(TEST_OUTPUT_DIR): - os.remove(os.path.join(TEST_OUTPUT_DIR, f)) - unittest.main()