Add an .excluded_sites list to allow pass tests

.gitignore

@@ -22,5 +22,8 @@ src/
 # Comma-Separated Values (CSV) Reports
 *.csv
 
+# Excluded sites list
+tests/.excluded_sites
+
 # MacOS Folder Metadata File
 .DS_Store

README.md

@@ -197,6 +197,10 @@ Note that we do currently have 100% test coverage.  Unfortunately, some of
 the sites that Sherlock checks are not always reliable, so it is common
 to get response errors.
 
+If some sites are failing due to conection problems (site is down, in maintainence, etc)
+you can exclude them from tests by creating a `tests/.excluded_sites` file with a
+list of sites to ignore (one site name per line).
+
 ## Stargazers over time
 
 [![Stargazers over time](https://starcharts.herokuapp.com/TheYahya/sherlock.svg)](https://starcharts.herokuapp.com/TheYahya/sherlock)

tests/base.py

@@ -32,6 +32,14 @@ class SherlockBaseTest(unittest.TestCase):
         with open(data_file_path, "r", encoding="utf-8") as raw:
             self.site_data_all = json.load(raw)
 
+        # 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()
+        except FileNotFoundError:
+          self.excluded_sites = []
+
         self.verbose=False
         self.tor=False
         self.unique_tor=False
@@ -134,6 +142,7 @@ class SherlockBaseTest(unittest.TestCase):
 
         for site, site_data in self.site_data_all.items():
             if (
+                 (site in self.excluded_sites)                 or
                  (site_data["errorType"] != detect_type)       or
                  (site_data.get("username_claimed")   is None) or
                  (site_data.get("username_unclaimed") is None)