From b1efe75eee5660571fccd361c17d132746316bf7 Mon Sep 17 00:00:00 2001
From: "Christopher K. Hoadley" <chris.hoadley@gmail.com>
Date: Sat, 7 Dec 2019 17:24:09 -0600
Subject: [PATCH] Remove timeout global and make it available on the command
 line.  Change default to no timeout.

---
 sherlock.py   | 50 ++++++++++++++++++++++++++++++++++++++++++++------
 tests/base.py |  4 +++-
 2 files changed, 47 insertions(+), 7 deletions(-)

diff --git a/sherlock.py b/sherlock.py
index d329789..aec4425 100755
--- a/sherlock.py
+++ b/sherlock.py
@@ -30,10 +30,9 @@ __version__ = "0.9.12"
 amount = 0
 
 
-global proxy_list, timeout_limit
+global proxy_list
 
 proxy_list = []
-timeout_limit = 10
 
 class ElapsedFuturesSession(FuturesSession):
     """
@@ -135,7 +134,8 @@ def get_response(request_future, error_type, social_network, verbose=False, retr
     return None, "", -1
 
 
-def sherlock(username, site_data, verbose=False, tor=False, unique_tor=False, proxy=None, print_found_only=False):
+def sherlock(username, site_data, verbose=False, tor=False, unique_tor=False,
+             proxy=None, print_found_only=False, timeout=None):
     """Run Sherlock Analysis.
 
     Checks for existence of username on various social media sites.
@@ -148,6 +148,8 @@ def sherlock(username, site_data, verbose=False, tor=False, unique_tor=False, pr
     tor                    -- Boolean indicating whether to use a tor circuit for the requests.
     unique_tor             -- Boolean indicating whether to use a new tor circuit for each request.
     proxy                  -- String indicating the proxy URL
+    timeout                -- Time in seconds to wait before timing out request.
+                              Default is no timeout.
 
     Return Value:
     Dictionary containing results from report. Key of dictionary is the name
@@ -248,12 +250,12 @@ def sherlock(username, site_data, verbose=False, tor=False, unique_tor=False, pr
                 future = request_method(url=url_probe, headers=headers,
                                         proxies=proxies,
                                         allow_redirects=allow_redirects,
-                                        timeout=timeout_limit
+                                        timeout=timeout
                                         )
             else:
                 future = request_method(url=url_probe, headers=headers,
                                         allow_redirects=allow_redirects,
-                                        timeout=timeout_limit
+                                        timeout=timeout
                                         )
 
             # Store future in data for access later
@@ -362,6 +364,31 @@ def sherlock(username, site_data, verbose=False, tor=False, unique_tor=False, pr
     return results_total
 
 
+def timeout_check(value):
+    """Check Timeout Argument.
+
+    Checks timeout for validity.
+
+    Keyword Arguments:
+    value                  -- Time in seconds to wait before timing out request.
+
+    Return Value:
+    Floating point number representing the time (in seconds) that should be
+    used for the timeout.
+
+    NOTE:  Will raise an exception if the timeout in invalid.
+    """
+    from argparse import ArgumentTypeError
+
+    try:
+        timeout = float(value)
+    except:
+        raise ArgumentTypeError(f"Timeout '{value}' must be a number.")
+    if timeout <= 0:
+        raise ArgumentTypeError(f"Timeout '{value}' must be greater than 0.0s.")
+    return timeout
+
+
 def main():
     # Colorama module's initialization.
     init(autoreset=True)
@@ -422,6 +449,14 @@ def main():
                              "The script will check if the proxies supplied in the .csv file are working and anonymous."
                              "Put 0 for no limit on successfully checked proxies, or another number to institute a limit."
                         )
+    parser.add_argument("--timeout",
+                        action="store", metavar='TIMEOUT',
+                        dest="timeout", type=timeout_check, default=None,
+                        help="Time (in seconds) to wait for response to requests. "
+                             "Default timeout of 60.0s."
+                             "A longer timeout will be more likely to get results from slow sites."
+                             "On the other hand, this may cause a long delay to gather all results."
+                        )
     parser.add_argument("--print-found",
                         action="store_true", dest="print_found_only", default=False,
                         help="Do not output sites where the username was not found."
@@ -576,7 +611,10 @@ def main():
 
         results = {}
         results = sherlock(username, site_data, verbose=args.verbose,
-                           tor=args.tor, unique_tor=args.unique_tor, proxy=args.proxy, print_found_only=args.print_found_only)
+                           tor=args.tor, unique_tor=args.unique_tor,
+                           proxy=args.proxy,
+                           print_found_only=args.print_found_only,
+                           timeout=args.timeout)
 
         exists_counter = 0
         for website_name in results:
diff --git a/tests/base.py b/tests/base.py
index 4f1bee7..da34f95 100644
--- a/tests/base.py
+++ b/tests/base.py
@@ -43,6 +43,7 @@ class SherlockBaseTest(unittest.TestCase):
         self.verbose=False
         self.tor=False
         self.unique_tor=False
+        self.timeout=None
 
         return
 
@@ -104,7 +105,8 @@ class SherlockBaseTest(unittest.TestCase):
                                         site_data,
                                         verbose=self.verbose,
                                         tor=self.tor,
-                                        unique_tor=self.unique_tor
+                                        unique_tor=self.unique_tor,
+                                        timeout=self.timeout
                                        )
             for site, result in results.items():
                 with self.subTest(f"Checking Username '{username}' "