Option to override User-Agent with an environment variable

CHANGES.rst

@@ -1,10 +1,12 @@
 Changelog
 =========
 
-2.1.1 (unreleased)
+2.2.0 (unreleased)
 ------------------
 
-- Nothing changed yet.
+- Allow configuring the User-Agent for the requests to get feeds,
+  via the REQUESTS_USER_AGENT environment variable.
+  [davisagli]
 
 
 2.1.0 (2023-03-10)

README.rst

@@ -95,6 +95,11 @@ There is a 5s default timeout for retrieve RSS feeds.
 
 You can override it with an environment variable: **RSS_SERVICE_TIMEOUT**
 
+Set User-Agent
+--------------
+
+You can override the default ``User-Agent`` for the requests to get feeds by setting an environment variable: **RSS_USER_AGENT**
+
 Installation
 ============
 

setup.py

@@ -16,7 +16,7 @@
 
 setup(
     name="redturtle.rssservice",
-    version="2.1.1.dev0",
+    version="2.2.0.dev0",
     description="An add-on for Plone",
     long_description=long_description,
     # Get more from https://pypi.org/classifiers/

src/redturtle/rssservice/rss_mixer.py

@@ -33,6 +33,7 @@
 FEED_DATA = {}  # url: ({date, title, url, itemlist})
 
 REQUESTS_TIMEOUT = int(environ.get("RSS_SERVICE_TIMEOUT", "5")) or 5
+REQUESTS_USER_AGENT = environ.get("RSS_USER_AGENT")
 
 
 class RSSMixerService(Service):
@@ -226,8 +227,15 @@ def _getFeedFromUrl(self, url):
         Use urllib to retrieve an rss feed.
         In this way, we can manage timeouts.
         """
+        headers = {}
+        if REQUESTS_USER_AGENT:
+            headers["User-Agent"] = REQUESTS_USER_AGENT
         try:
-            response = requests.get(url, timeout=REQUESTS_TIMEOUT)
+            response = requests.get(
+                url,
+                headers=headers,
+                timeout=REQUESTS_TIMEOUT,
+            )
         except (Timeout, RequestException) as e:
             logger.exception(e)
             return None