X-Git-Url: https://code.communitydata.science/mediawiki_dump_tools.git/blobdiff_plain/f147e1d899ea269dad715aea393153683353e946..c2854026831c54085ed254c97bbdff0779c9e3e0:/test/Wikiq_Unit_Test.py diff --git a/test/Wikiq_Unit_Test.py b/test/Wikiq_Unit_Test.py index 159fd10..0a90c6c 100644 --- a/test/Wikiq_Unit_Test.py +++ b/test/Wikiq_Unit_Test.py @@ -3,7 +3,7 @@ import os import subprocess from shutil import copyfile import pandas as pd -from pandas.util.testing import assert_frame_equal +from pandas.testing import assert_frame_equal from io import StringIO # with / without pwr DONE @@ -49,7 +49,7 @@ class Test_Wikipedia(unittest.TestCase): # as a test let's make sure that we get equal data frames test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) def test_WP_namespaces(self): print(os.path.abspath('.')) @@ -71,7 +71,7 @@ class Test_Wikipedia(unittest.TestCase): num_wrong_ns = sum(~ test.namespace.isin({0,1})) self.assertEqual(num_wrong_ns, 0) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) def test_WP_revert_radius(self): print(os.path.abspath('.')) @@ -93,7 +93,7 @@ class Test_Wikipedia(unittest.TestCase): num_wrong_ns = sum(~ test.namespace.isin({0,1})) self.assertEqual(num_wrong_ns, 0) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) @@ -131,7 +131,7 @@ class Test_Basic(unittest.TestCase): test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) def test_collapse_user(self): @@ -151,7 +151,7 @@ class Test_Basic(unittest.TestCase): baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) def test_pwr_segment(self): test_filename = "persistence_segment_" + self.wikiq_out_name @@ -171,7 +171,7 @@ class Test_Basic(unittest.TestCase): test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) def test_pwr_legacy(self): test_filename = "persistence_legacy_" + self.wikiq_out_name @@ -191,7 +191,7 @@ class Test_Basic(unittest.TestCase): test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) def test_pwr(self): test_filename = "persistence_" + self.wikiq_out_name @@ -211,7 +211,9 @@ class Test_Basic(unittest.TestCase): test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + + test = test.reindex(columns=sorted(test.columns)) + assert_frame_equal(test,baseline, check_like=True) def test_url_encode(self): @@ -230,7 +232,9 @@ class Test_Basic(unittest.TestCase): baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) test = pd.read_table(test_file) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + + test = test.reindex(columns=sorted(test.columns)) + assert_frame_equal(test,baseline, check_like=True) class Test_Malformed(unittest.TestCase): @@ -241,7 +245,7 @@ class Test_Malformed(unittest.TestCase): self.wiki = 'twinpeaks' self.wikiq_out_name = self.wiki + ".tsv" self.test_output_dir = os.path.join(".", "test_output") - self.call_output = os.path.join(self.test_output_dir, self.wikiq_out_name) + self.call_output = os.path.join(self.test_output_dir, self.wikiq_out_name) self.infile = "{0}.xml.7z".format(self.wiki) self.base_call = "../wikiq {0} -o {1}" @@ -273,6 +277,7 @@ class Test_Stdout(unittest.TestCase): def test_noargs(self): call = self.base_call.format(self.input_file) + print(call) proc = subprocess.run(call,stdout=subprocess.PIPE,shell=True) outs = proc.stdout.decode("utf8") @@ -281,7 +286,7 @@ class Test_Stdout(unittest.TestCase): print(baseline_file) test = pd.read_table(StringIO(outs)) baseline = pd.read_table(baseline_file) - assert_frame_equal(test,baseline) + assert_frame_equal(test,baseline, check_like=True) class Test_Regex(unittest.TestCase): @@ -305,27 +310,33 @@ class Test_Regex(unittest.TestCase): self.baseline_output_dir = "baseline_output" # sample inputs for checking that bad inputs get terminated / test_regex_inputs - self.bad_input1 = "-RP '\\b\\d+\\b'" #label is missing - self.bad_input2 = "-RP 'NPO V' -RP THE -RPl testlabel" #number of reg and number of labels do not match - self.bad_input3 = "-CP '(Tamil|Li)' -RPl testlabel" #cp but rp label - self.bad_input4 = "-CPl testlabel" #regex is missing - self.bad_input5 = "-RP '\\b\\w{3}\\b' -RPl threeletters -CP '\\b\\w{3}\\b'" - - self.bad_inputs_list = [self.bad_input1,self.bad_input2,self.bad_input3,self.bad_input4,self.bad_input5] + self.bad_inputs_list = [ + #label is missing + "-RP '\\b\\d+\\b'", + #number of reg and number of labels do not match + "-RP 'NPO V' -RP THE -RPl testlabel", + #cp but rp label + "-CP '(Tamil|Li)' -RPl testlabel", + #regex is missing + "-CPl testlabel", + "-RP '\\b\\w{3}\\b' -RPl threeletters -CP '\\b\\w{3}\\b'" + ] # sample inputs for checking the outcomes of good inputs / test_basic_regex - self.good_input1 = "-RP '\\b\\d{3}\\b' -RPl threedigits" - self.good_input2 = "-RP 'TestCase' -RP 'page' -RPl testcases -RPl page_word" - self.good_input3 = "-CP 'Chevalier' -CPl chev_com -RP 'welcome to Wikipedia' -RPl wiki_welcome -CP 'Warning' -CPl warning" - self.good_input4 = "-CP 'WP:EVADE' -CPl wp_evade" + self.good_inputs_list = [ + "-RP '\\b\\d{3}\\b' -RPl threedigits", + "-RP 'TestCase' -RP 'page' -RPl testcases -RPl page_word", + "-CP 'Chevalier' -CPl chev_com -RP 'welcome to Wikipedia' -RPl wiki_welcome -CP 'Warning' -CPl warning", + "-CP 'WP:EVADE' -CPl wp_evade" + ] - self.good_inputs_list = [self.good_input1,self.good_input2,self.good_input3, self.good_input4] + + self.cap_inputs_list = [ + "-RP 'Li Chevalier' -RPl li_cheval -CP '(?P\\b[a-zA-Z]{3}\\b)|(?P\\b\\d+\\b)|(?P\\bcat\\b)' -CPl three", + "-CP '(?P\\bTestCaseA\\b)|(?P\\bTestCaseB\\b)|(?P\\bTestCaseC\\b)|(?P\\bTestCaseD\\b)' -CPl testcase -RP '(?Pnpov|NPOV)|(?Pneutral point of view)' -RPl npov" + ] - # and with capture group(s) / test_capturegroup_regex - self.cap_input1 = "-RP 'Li Chevalier' -RPl li_cheval -CP '(?P\\b[a-zA-Z]{3}\\b)|(?P\\b\\d+\\b)|(?P\\bcat\\b)' -CPl three" - self.cap_input2 = "-CP '(?P\\bTestCaseA\\b)|(?P\\bTestCaseB\\b)|(?P\\bTestCaseC\\b)|(?P\\bTestCaseD\\b)' -CPl testcase -RP '(?Pnpov|NPOV)|(?Pneutral point of view)' -RPl npov" - self.cap_inputs_list = [self.cap_input1,self.cap_input2] def test_regex_inputs(self): for input in self.bad_inputs_list: @@ -341,8 +352,7 @@ class Test_Regex(unittest.TestCase): self.assertNotEqual(proc.returncode,0) def test_basic_regex(self): - i = 1 - for input in self.good_inputs_list: + for i, input in enumerate(self.good_inputs_list): test_filename = "basic_{0}_{1}.tsv".format(self.wikiq_out_name[:-4], str(i)) #print(test_filename) @@ -356,22 +366,18 @@ class Test_Regex(unittest.TestCase): proc = subprocess.Popen(call,stdout=subprocess.PIPE,stderr=subprocess.PIPE,shell=True) proc.wait() - copyfile(self.call_output, test_file) - f = open(self.call_output, 'w') - f.close() - - # don't have a baseline file to compare a test to?? - # baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) - i += 1 - #TODO - #a proper assert statement still needs to go here, but I checked out the generated files for now and it functions + test = pd.read_table(test_file) + + baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline = pd.read_table(baseline_file) + assert_frame_equal(test, baseline, check_like=True) + print(i) def test_capturegroup_regex(self): - i = 1 - for input in self.cap_inputs_list: + for i, input in enumerate(self.cap_inputs_list): test_filename = "capturegroup_{0}_{1}.tsv".format(self.wikiq_out_name[:-4], str(i)) print(test_filename) test_file = os.path.join(self.test_output_dir, test_filename) @@ -386,15 +392,13 @@ class Test_Regex(unittest.TestCase): proc.wait() copyfile(self.call_output, test_file) - f = open(self.call_output, 'w') - f.close() - # don't have a baseline file to compare a test to?? - # baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) - i += 1 + test = pd.read_table(test_file) + + baseline_file = os.path.join(".", self.baseline_output_dir, test_filename) + baseline = pd.read_table(baseline_file) + assert_frame_equal(test, baseline, check_like=True) - #TODO - #a proper assert statement still needs to go here, but I checked out the generated files for now and it functions if __name__ == '__main__': unittest.main()