Create and use default timeouts for the requests package

With the release of version 1.7.5 the bandit package now flags requests
calls (get(), post(), etc.) as medium severity security risks if they
do not have a timeout value provided. The default values we create are
the same as the defaults used by the github3.py package. This provides
as much parity as possible with other uses of the requests package in
this project.

scraper/doecode/__init__.py

@@ -3,6 +3,8 @@
 
 import requests
 
+from scraper.util import DEFAULT_REQUESTS_TIMEOUTS
+
 logger = logging.getLogger(__name__)
 
 
@@ -32,7 +34,11 @@ def process_url(url, key):
     if key is None:
         raise ValueError("DOE CODE API Key value is missing!")
 
-    response = requests.get(url, headers={"Authorization": "Basic " + key})
+    response = requests.get(
+        url,
+        headers={"Authorization": "Basic " + key},
+        timeout=DEFAULT_REQUESTS_TIMEOUTS,
+    )
     doecode_json = response.json()
 
     for record in doecode_json["records"]:

scraper/github/__init__.py

@@ -8,6 +8,8 @@
 import github3
 import requests
 
+from scraper.util import DEFAULT_REQUESTS_TIMEOUTS
+
 logger = logging.getLogger(__name__)
 
 
@@ -23,7 +25,8 @@ def gov_orgs():
     us_gov_github_orgs = set()
 
     gov_orgs_json = requests.get(
-        "https://government.github.com/organizations.json"
+        "https://government.github.com/organizations.json",
+        timeout=DEFAULT_REQUESTS_TIMEOUTS,
     ).json()
 
     us_gov_github_orgs.update(gov_orgs_json["governments"]["U.S. Federal"])

scraper/github/queryManager.py

@@ -14,6 +14,8 @@
 import pytz
 import requests
 
+from scraper.util import DEFAULT_REQUESTS_TIMEOUTS
+
 
 def _vPrint(verbose, *args, **kwargs):
     """Easy verbosity-control print method.
@@ -490,10 +492,13 @@ def _submitQuery(
                 "https://api.github.com/graphql",
                 data=gitqueryJSON,
                 headers={**authhead, **headers},
+                timeout=DEFAULT_REQUESTS_TIMEOUTS,
             )
         else:
             fullResponse = requests.get(
-                "https://api.github.com" + gitquery, headers={**authhead, **headers}
+                "https://api.github.com" + gitquery,
+                headers={**authhead, **headers},
+                timeout=DEFAULT_REQUESTS_TIMEOUTS,
             )
         _vPrint(
             verbose,

scraper/util.py

@@ -8,6 +8,10 @@
 
 logger = logging.getLogger(__name__)
 
+# These mirror the defaults in github3.py sessions per:
+# https://github.com/sigmavirus24/github3.py/blob/ce43e6e5fdef6555f5a6b6602e2cc4b66c428aef/src/github3/session.py#L98
+DEFAULT_REQUESTS_TIMEOUTS = (4, 10)
+
 
 def execute(command, cwd=None):
     logger.debug("Forking command: %s", command)