merging pull containing revert-radius with 2nd version of regex scanner w/ unit tests

[mediawiki_dump_tools.git] / test / Wikiq_Unit_Test.py

diff --git a/test/Wikiq_Unit_Test.py b/test/Wikiq_Unit_Test.py
index 9011d799831548c1def240c823aa7a4ba3945d8d..159fd1088af1f3fbfeb9d10fb0810a099dd6ccc5 100644 (file)
--- a/test/Wikiq_Unit_Test.py
+++ b/test/Wikiq_Unit_Test.py
@@ -282,6 +282,119 @@ class Test_Stdout(unittest.TestCase):
         test = pd.read_table(StringIO(outs))
         baseline = pd.read_table(baseline_file)
         assert_frame_equal(test,baseline)
-        
+
+class Test_Regex(unittest.TestCase):
+
+    def setUp(self):
+        self.wiki = 'regextest'
+        self.wikiq_out_name = self.wiki + '.tsv'
+        self.infile = "{0}.xml.bz2".format(self.wiki)
+
+        self.input_dir = "dumps"
+        self.input_file = os.path.join(".", self.input_dir,self.infile)
+
+        if not os.path.exists("test_output"):
+            os.mkdir("test_output")
+
+        self.test_output_dir = os.path.join(".", "test_output")
+        self.call_output = os.path.join(self.test_output_dir, self.wikiq_out_name)
+        # we have two base calls, one for checking inputs and the other for checking outputs
+        self.base_call = "../wikiq {0}"
+        self.base_call_outs = "../wikiq {0} -o {1}"
+
+        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]
+
+        # 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 = [self.good_input1,self.good_input2,self.good_input3, self.good_input4]
+
+        # and with capture group(s) / test_capturegroup_regex
+        self.cap_input1 = "-RP 'Li Chevalier' -RPl li_cheval -CP '(?P<letter>\\b[a-zA-Z]{3}\\b)|(?P<number>\\b\\d+\\b)|(?P<cat>\\bcat\\b)' -CPl three"
+        self.cap_input2 = "-CP '(?P<a>\\bTestCaseA\\b)|(?P<b>\\bTestCaseB\\b)|(?P<c>\\bTestCaseC\\b)|(?P<d>\\bTestCaseD\\b)' -CPl testcase -RP '(?P<npov>npov|NPOV)|(?P<neutral>neutral 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:
+            call = self.base_call.format(self.input_file)
+            call = call + " --stdout " + input
+            print(call)
+            proc = subprocess.Popen(call,stdout=subprocess.PIPE,stderr=subprocess.PIPE,shell=True)
+            stdout,stderr = proc.communicate()
+            #print(proc.returncode)
+            
+            # we want to check that the bad inputs were caught and sys.exit is stopping the code
+            print(stderr.decode("utf-8"))
+            self.assertNotEqual(proc.returncode,0)
+
+    def test_basic_regex(self):
+        i = 1
+        for input in self.good_inputs_list:
+
+            test_filename = "basic_{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)
+            if os.path.exists(test_file):
+                os.remove(test_file)
+
+            call = self.base_call_outs.format(self.input_file, self.test_output_dir)
+            call = call + " " + input
+            print(call)
+
+            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
+
+
+    def test_capturegroup_regex(self):
+        i = 1
+        for input in 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)
+            if os.path.exists(test_file):
+                os.remove(test_file)
+
+            call = self.base_call_outs.format(self.input_file, self.test_output_dir)
+            call = call + " " + input
+            print(call)
+
+            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
+
 if __name__ == '__main__':
     unittest.main()