resolve redirects if siteinfo is provided

[mediawiki_dump_tools.git] / test / Wikiq_Unit_Test.py

diff --git a/test/Wikiq_Unit_Test.py b/test/Wikiq_Unit_Test.py
index 3a1e24beb65795d8b904f3c56fcb36284379a3a1..1b5ca3f70b1095358d3d7dc95294a1070c741483 100644 (file)
--- a/test/Wikiq_Unit_Test.py
+++ b/test/Wikiq_Unit_Test.py
@@ -1,17 +1,101 @@
 import unittest
 import os
 import unittest
 import os

-import sys
+import subprocess

 from shutil import copyfile
 from shutil import copyfile

+import pandas as pd
+from pandas.testing import assert_frame_equal
+from io import StringIO

 
 # with / without pwr DONE
 # with / without url encode DONE
 # with / without collapse user DONE
 
 # with / without pwr DONE
 # with / without url encode DONE
 # with / without collapse user DONE

-# with output to sdtout
+# with output to sdtout DONE

 # note that the persistence radius is 7 by default
 # reading various file formats including
 # note that the persistence radius is 7 by default
 # reading various file formats including

-#        7z, gz, bz2, xml 
-# wikia and wikipedia data
-# malformed xmls
+#        7z, gz, bz2, xml  DONE
+# wikia and wikipedia data DONE
+# malformed xmls DONE
+
+class Test_Wikipedia(unittest.TestCase):
+    def setUp(self):
+        if not os.path.exists("test_output"):
+            os.mkdir("test_output")
+
+        self.wiki = 'ikwiki-20180301-pages-meta-history'
+        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.infile = "{0}.xml.bz2".format(self.wiki)    
+        self.base_call = "../wikiq {0} -o {1}"
+        self.input_dir = "dumps"
+        self.input_file = os.path.join(".", self.input_dir,self.infile)
+        self.baseline_output_dir = "baseline_output"
+
+    def test_WP_url_encode(self):
+        test_filename =  "url-encode_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
+        call = call + " --url-encode"
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+
+        copyfile(self.call_output, test_file)
+        baseline_file = os.path.join(".", self.baseline_output_dir, test_filename)
+
+        # 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, check_like=True)
+
+    def test_WP_namespaces(self):
+        print(os.path.abspath('.'))
+        test_filename =  "namespaces_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
+        call = call + " -n 0 -n 1"
+        print(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+        copyfile(self.call_output, test_file)
+        baseline_file = os.path.join(os.path.abspath("."), self.baseline_output_dir, test_filename)
+
+        # as a test let's make sure that we get equal data frames
+        test = pd.read_table(test_file)
+        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, check_like=True)
+
+    def test_WP_revert_radius(self):
+        print(os.path.abspath('.'))
+        test_filename =  "revert_radius_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
+        call = call + " -n 0 -n 1 -rr 1"
+        print(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+        copyfile(self.call_output, test_file)
+        baseline_file = os.path.join(os.path.abspath("."), self.baseline_output_dir, test_filename)
+
+        # as a test let's make sure that we get equal data frames
+        test = pd.read_table(test_file)
+        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, check_like=True)
+
+

 
 class Test_Basic(unittest.TestCase):
 
 
 class Test_Basic(unittest.TestCase):
 


@@ -32,77 +116,365 @@ class Test_Basic(unittest.TestCase):
 
     def test_noargs(self):
 
 
     def test_noargs(self):
 

+        test_filename =  "noargs_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
         call = self.base_call.format(self.input_file, self.test_output_dir)

-        os.system(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()

 
 

-        test_file =  "noargs_" + self.wikiq_out_name
-        copyfile(self.call_output, os.path.join(self.test_output_dir, test_file))
+        copyfile(self.call_output, test_file)

 
 

-        baseline_file = os.path.join(".", self.baseline_output_dir, test_file)
+        baseline_file = os.path.join(".", self.baseline_output_dir, test_filename)

 
 

-        test_lines = open(os.path.join(self.test_output_dir,test_file))
-        baseline_lines = open(baseline_file)
-        for test, baseline in zip(test_lines, baseline_lines):
-            self.assertEqual(test,baseline)
+        test = pd.read_table(test_file)
+        baseline = pd.read_table(baseline_file)
+        assert_frame_equal(test,baseline, check_like=True)

 
 

-        test_lines.close()
-        baseline_lines.close()

 
     def test_collapse_user(self):
 
     def test_collapse_user(self):

+        test_filename =  "collapse-user_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
         call = self.base_call.format(self.input_file, self.test_output_dir)

-        os.system(call)

         call = call + " --collapse-user"
 
         call = call + " --collapse-user"
 

-        os.system(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()

 
 

-        test_file =  "collapse-user_" + self.wikiq_out_name
-        copyfile(self.call_output, os.path.join(self.test_output_dir, test_file))
+        copyfile(self.call_output, test_file)
+
+        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, check_like=True)
+
+    def test_pwr_segment(self):
+        test_filename =  "persistence_segment_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
+        call = call + " --persistence segment"
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()

 
 

-        baseline_file = os.path.join(".", self.baseline_output_dir, test_file)

 
 

-        test_lines = open(os.path.join(self.test_output_dir,test_file))
-        baseline_lines = open(baseline_file)
-        for test, baseline in zip(test_lines, baseline_lines):
-            self.assertEqual(test,baseline)
+        copyfile(self.call_output, test_file)

 
 

-        test_lines.close()
-        baseline_lines.close()
+        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, check_like=True)
+
+    def test_pwr_legacy(self):
+        test_filename =  "persistence_legacy_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
+        call = call + " --persistence legacy"
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+
+
+        copyfile(self.call_output, test_file)
+
+        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, check_like=True)

 
     def test_pwr(self):
 
     def test_pwr(self):

+        test_filename =  "persistence_" + self.wikiq_out_name
+        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.format(self.input_file, self.test_output_dir)
         call = call + " --persistence"
         call = self.base_call.format(self.input_file, self.test_output_dir)
         call = call + " --persistence"

-        os.system(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()

 
 

-        test_file =  "persistence_" + self.wikiq_out_name
-        copyfile(self.call_output, os.path.join(self.test_output_dir, test_file))

 
 

-        baseline_file = os.path.join(".", self.baseline_output_dir, test_file)
+        copyfile(self.call_output, test_file)

 
 

-        test_lines = open(os.path.join(self.test_output_dir,test_file))
-        baseline_lines = open(baseline_file)
-        for test, baseline in zip(test_lines, baseline_lines):
-            self.assertEqual(test,baseline)
+        baseline_file = os.path.join(".", self.baseline_output_dir, test_filename)
+
+        test = pd.read_table(test_file)
+        baseline = pd.read_table(baseline_file)
+
+        test = test.reindex(columns=sorted(test.columns))
+        assert_frame_equal(test,baseline, check_like=True)

 
 

-        test_lines.close()
-        baseline_lines.close()

 
     def test_url_encode(self):
 
     def test_url_encode(self):

+        test_filename =  "url-encode_" + self.wikiq_out_name
+
+        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.format(self.input_file, self.test_output_dir)
         call = call + " --url-encode"
         call = self.base_call.format(self.input_file, self.test_output_dir)
         call = call + " --url-encode"

-        os.system(call)
-        test_file =  "url-encode_" + self.wikiq_out_name
-        copyfile(self.call_output, os.path.join(self.test_output_dir, test_file))
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+
+        copyfile(self.call_output, test_file)
+        baseline_file = os.path.join(".", self.baseline_output_dir, test_filename)
+        test = pd.read_table(test_file)
+        baseline = pd.read_table(baseline_file)
+
+        test = test.reindex(columns=sorted(test.columns))
+        assert_frame_equal(test,baseline, check_like=True)
+
+
+class Test_Malformed(unittest.TestCase):
+    def setUp(self):
+        if not os.path.exists("test_output"):
+            os.mkdir("test_output")
+
+        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.infile = "{0}.xml.7z".format(self.wiki)
+        self.base_call = "../wikiq {0} -o {1}"
+        self.input_dir = "dumps"
+        self.input_file = os.path.join(".", self.input_dir,self.infile)
+
+
+    def test_malformed_noargs(self):
+
+        call = self.base_call.format(self.input_file, self.test_output_dir)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,stderr=subprocess.PIPE, shell=True)
+        proc.wait()
+        outs, errs = proc.communicate()
+        errlines = str(errs).split("\\n")
+        self.assertEqual(errlines[-2],'xml.etree.ElementTree.ParseError: no element found: line 1369, column 0')
+
+
+class Test_Redirect_Ikwiki(unittest.TestCase):
+    def setUp(self):
+        self.wiki = 'ikwiki-20180301-pages-meta-history'
+        wiki_siteinfo = "ikwiki-20191101-siteinfo-namespaces.json.gz"
+        self.input_dir = "dumps"
+        self.wiki_siteinfo = os.path.join(".", self.input_dir, wiki_siteinfo)
+        self.wikiq_out_name =  self.wiki + ".tsv"
+        self.infile = "{0}.xml.bz2".format(self.wiki)    
+
+        self.base_call = "../wikiq {0} --siteinfo {1} -o {2}"
+        self.input_dir = "dumps"
+        self.input_file = os.path.join(".", self.input_dir, self.infile)
+        self.baseline_output_dir = "baseline_output"
+
+        self.test_output_dir = os.path.join(".", "test_output")
+        self.call_output = os.path.join(self.test_output_dir, self.wikiq_out_name)
+
+        self.infile = "{0}.xml.bz2".format(self.wiki)    
+
+    def test_redirect(self):
+
+        call = self.base_call.format(self.input_file, self.wiki_siteinfo, self.test_output_dir)
+        print(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+
+        test_file = os.path.join(self.test_output_dir, "redirect_" + self.wikiq_out_name)
+        copyfile(self.call_output, test_file)
+        baseline_file = os.path.join(".", self.baseline_output_dir, test_file)
+
+        # 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, check_like=True)
+
+
+
+class Test_Redirect_Wikia(unittest.TestCase):
+    def setUp(self):
+        self.wiki = 'pokemonfandomcom_fr-20200215-history'
+        wiki_siteinfo = "pokemonfandomcom_fr-20200215-history.xml.7z"
+        self.input_dir = "dumps"
+        self.wiki_siteinfo = os.path.join(".", self.input_dir, wiki_siteinfo)
+        self.wikiq_out_name =  self.wiki + ".tsv"
+        self.infile = "{0}.xml.7z".format(self.wiki)    
+
+        self.base_call = "../wikiq {0} --siteinfo {1} -o {2}"
+        self.input_dir = "dumps"
+        self.input_file = os.path.join(".", self.input_dir, self.infile)
+        self.baseline_output_dir = "baseline_output"
+
+        self.test_output_dir = os.path.join(".", "test_output")
+        self.call_output = os.path.join(self.test_output_dir, self.wikiq_out_name)
+
+        self.infile = "{0}.xml.7z".format(self.wiki)    
+
+    def test_redirect(self):
+
+        call = self.base_call.format(self.input_file, self.wiki_siteinfo, self.test_output_dir)
+        print(call)
+        proc = subprocess.Popen(call,stdout=subprocess.PIPE,shell=True)
+        proc.wait()
+
+        test_file = os.path.join(self.test_output_dir, "redirect_" + self.wikiq_out_name)
+        copyfile(self.call_output, test_file)
+        baseline_file = os.path.join(".", self.baseline_output_dir, test_file)
+
+        # 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, check_like=True)
+
+
+    
+class Test_Stdout(unittest.TestCase):
+
+    def setUp(self):
+        self.wiki = 'itwiki'
+        self.wikiq_out_name =  self.wiki + ".tsv"
+
+        self.infile = "{0}.xml.7z".format(self.wiki)
+        self.base_call = "../wikiq {0} --stdout"
+        self.input_dir = "dumps"
+        self.input_file = os.path.join(".", self.input_dir,self.infile)
+        self.baseline_output_dir = "baseline_output"
+
+    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")
+
+        test_file = "noargs_" + self.wikiq_out_name

         baseline_file = os.path.join(".", self.baseline_output_dir, test_file)
         baseline_file = os.path.join(".", self.baseline_output_dir, test_file)

+        print(baseline_file)
+        test = pd.read_table(StringIO(outs))
+        baseline = pd.read_table(baseline_file)
+        assert_frame_equal(test,baseline, check_like=True)
+        

 
 

-        test_lines = open(os.path.join(self.test_output_dir,test_file))
-        baseline_lines = open(baseline_file)
-        for test, baseline in zip(test_lines, baseline_lines):
-            self.assertEqual(test,baseline)
+class Test_Regex(unittest.TestCase):

 
 

-        test_lines.close()
-        baseline_lines.close()
+    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_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_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.cap_inputs_list = [
+            "-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",
+            "-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"
+        ]
+
+
+
+    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):
+        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)
+            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)
+
+            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):
+        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)
+            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)
+            
+            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)
+
+

 if __name__ == '__main__':
     unittest.main()
 if __name__ == '__main__':
     unittest.main()