Merge pull request #1730 from sherlock-project/enhanced-unit-tests

generate unclaimed username based on regex

requirements.txt

@@ -7,3 +7,4 @@ stem>=1.8.0
 torrequest>=0.1.0
 pandas>=1.0.0
 openpyxl<=3.0.10
+exrex>=0.11.0

sherlock/resources/data.json

@@ -94,6 +94,7 @@
     "username_claimed": "pink"
   },
   "AllMyLinks": {
+    "regexCheck": "^[a-z0-9][a-z0-9-]{2,32}$",
     "errorMsg": "Not Found",
     "errorType": "message",
     "url": "https://allmylinks.com/{}",

sherlock/sherlock.py

@@ -28,7 +28,7 @@ from sites import SitesInformation
 from colorama import init
 
 module_name = "Sherlock: Find Usernames Across Social Networks"
-__version__ = "0.14.2"
+__version__ = "0.14.3"
 
 
 class SherlockFuturesSession(FuturesSession):

sherlock/tests/all.py

@@ -3,7 +3,7 @@
 This module contains various tests.
 """
 from tests.base import SherlockBaseTest
-import secrets
+import exrex
 
 
 class SherlockDetectTests(SherlockBaseTest):
@@ -27,10 +27,7 @@ class SherlockDetectTests(SherlockBaseTest):
         # Ensure that the site's detection method has not changed.
         self.assertEqual("message", site_data["errorType"])
 
-        self.username_check([site_data["username_claimed"]],
-                            [site],
-                            exist_check=True
-                            )
+        self.username_check([site_data["username_claimed"]], [site], exist_check=True)
 
         return
 
@@ -54,10 +51,16 @@ class SherlockDetectTests(SherlockBaseTest):
         # Ensure that the site's detection method has not changed.
         self.assertEqual("message", site_data["errorType"])
 
-        self.username_check([secrets.token_urlsafe(10)],
-                            [site],
-                            exist_check=False
-                            )
+        # Generate a valid username based on the regex for a username that the
+        # site supports that is *most likely* not taken. The regex is slighlty
+        # modified version of site_data["regexCheck"] as we want a username
+        # that has the maximum length that is supported by the site. This way,
+        # we wont generate a random username that might actually exist. This
+        # method is very hacky, but it does the job as having hardcoded
+        # usernames that dont exists will lead to people with ill intent to
+        # create an account with that username which will break the tests
+        valid_username = exrex.getone(r"^[a-z0-9][a-z0-9-]{32}$")
+        self.username_check([valid_username], [site], exist_check=False)
 
         return
 
@@ -75,16 +78,13 @@ class SherlockDetectTests(SherlockBaseTest):
         Will trigger an assert if detection mechanism did not work as expected.
         """
 
-        site = "9GAG"
+        site = "BitBucket"
         site_data = self.site_data_all[site]
 
         # Ensure that the site's detection method has not changed.
         self.assertEqual("status_code", site_data["errorType"])
 
-        self.username_check([site_data["username_claimed"]],
-                            [site],
-                            exist_check=True
-                            )
+        self.username_check([site_data["username_claimed"]], [site], exist_check=True)
 
         return
 
@@ -102,57 +102,27 @@ class SherlockDetectTests(SherlockBaseTest):
         Will trigger an assert if detection mechanism did not work as expected.
         """
 
-        site = "9GAG"
+        site = "BitBucket"
         site_data = self.site_data_all[site]
 
         # Ensure that the site's detection method has not changed.
         self.assertEqual("status_code", site_data["errorType"])
 
-        self.username_check([secrets.token_urlsafe(10)],
-                            [site],
-                            exist_check=False
-                            )
+        # Generate a valid username based on the regex for a username that the
+        # site supports that is *most likely* not taken. The regex is slighlty
+        # modified version of site_data["regexCheck"] as we want a username
+        # that has the maximum length that is supported by the site. This way,
+        # we wont generate a random username that might actually exist. This
+        # method is very hacky, but it does the job as having hardcoded
+        # usernames that dont exists will lead to people with ill intent to
+        # create an account with that username which will break the tests
+        valid_username = exrex.getone(r"^[a-zA-Z0-9-_]{30}") 
+        self.username_check([valid_username], [site], exist_check=False)
 
         return
 
 
 class SherlockSiteCoverageTests(SherlockBaseTest):
-    def test_coverage_false_via_response_url(self):
-        """Test Username Does Not Exist Site Coverage (Via Response URL).
-
-        This test checks all sites with the "response URL" detection mechanism
-        to ensure that a Username that does not exist is reported that way.
-
-        Keyword Arguments:
-        self                   -- This object.
-
-        Return Value:
-        Nothing.
-        Will trigger an assert if detection mechanism did not work as expected.
-        """
-
-        self.detect_type_check("response_url", exist_check=False)
-
-        return
-
-    def test_coverage_true_via_response_url(self):
-        """Test Username Does Exist Site Coverage (Via Response URL).
-
-        This test checks all sites with the "response URL" detection mechanism
-        to ensure that a Username that does exist is reported that way.
-
-        Keyword Arguments:
-        self                   -- This object.
-
-        Return Value:
-        Nothing.
-        Will trigger an assert if detection mechanism did not work as expected.
-        """
-
-        self.detect_type_check("response_url", exist_check=True)
-
-        return
-
     def test_coverage_false_via_status(self):
         """Test Username Does Not Exist Site Coverage (Via HTTP Status).
 

sherlock/tests/base.py

@@ -7,9 +7,8 @@ import os.path
 import unittest
 import sherlock
 from result import QueryStatus
-from result import QueryResult
 from notify import QueryNotify
-from sites  import SitesInformation
+from sites import SitesInformation
 import warnings
 
 
@@ -26,16 +25,16 @@ class SherlockBaseTest(unittest.TestCase):
         Nothing.
         """
 
-        #This ignores the ResourceWarning from an unclosed SSLSocket.
-        #TODO: Figure out how to fix the code so this is not needed.
+        # This ignores the ResourceWarning from an unclosed SSLSocket.
+        # TODO: Figure out how to fix the code so this is not needed.
         warnings.simplefilter("ignore", ResourceWarning)
 
-        #Create object with all information about sites we are aware of.
-        sites = SitesInformation()
+        # Create object with all information about sites we are aware of.
+        sites = SitesInformation(data_file_path=os.path.join(os.path.dirname(__file__), "../resources/data.json"))
 
-        #Create original dictionary from SitesInformation() object.
-        #Eventually, the rest of the code will be updated to use the new object
-        #directly, but this will glue the two pieces together.
+        # Create original dictionary from SitesInformation() object.
+        # Eventually, the rest of the code will be updated to use the new object
+        # directly, but this will glue the two pieces together.
         site_data_all = {}
         for site in sites:
             site_data_all[site.name] = site.information
@@ -44,18 +43,18 @@ class SherlockBaseTest(unittest.TestCase):
         # Load excluded sites list, if any
         excluded_sites_path = os.path.join(os.path.dirname(os.path.realpath(sherlock.__file__)), "tests/.excluded_sites")
         try:
-          with open(excluded_sites_path, "r", encoding="utf-8") as excluded_sites_file:
-            self.excluded_sites = excluded_sites_file.read().splitlines()
+            with open(excluded_sites_path, "r", encoding="utf-8") as excluded_sites_file:
+                self.excluded_sites = excluded_sites_file.read().splitlines()
         except FileNotFoundError:
-          self.excluded_sites = []
+            self.excluded_sites = []
 
-        #Create notify object for query results.
+        # Create notify object for query results.
         self.query_notify = QueryNotify()
 
-        self.tor=False
-        self.unique_tor=False
-        self.timeout=None
-        self.skip_error_sites=True
+        self.tor = False
+        self.unique_tor = False
+        self.timeout = None
+        self.skip_error_sites = True
 
         return
 
@@ -102,7 +101,7 @@ class SherlockBaseTest(unittest.TestCase):
         existence state.
         """
 
-        #Filter all site data down to just what is needed for this test.
+        # Filter all site data down to just what is needed for this test.
         site_data = self.site_data_filter(site_list)
 
         if exist_check:
@@ -161,8 +160,8 @@ class SherlockBaseTest(unittest.TestCase):
         existence state.
         """
 
-        #Dictionary of sites that should be tested for having a username.
-        #This will allow us to test sites with a common username in parallel.
+        # Dictionary of sites that should be tested for having a username.
+        # This will allow us to test sites with a common username in parallel.
         sites_by_username = {}
 
         for site, site_data in self.site_data_all.items():
@@ -181,9 +180,9 @@ class SherlockBaseTest(unittest.TestCase):
 
                 # Figure out which type of user
                 if exist_check:
-                     username = site_data.get("username_claimed")
+                    username = site_data.get("username_claimed")
                 else:
-                     username = site_data.get("username_unclaimed")
+                    username = site_data.get("username_unclaimed")
 
                 # Add this site to the list of sites corresponding to this
                 # username.