feat: add support for proxying feed requests

Introduced optional proxy configuration for fetching RSS feeds,
including support for using Tor SOCKS5 proxy for enhanced privacy.
Added properties to manage proxy settings and updated fetch
functionality to utilize the proxy when necessary. This helps route
requests through specified proxies, supporting privacy use cases
like accessing .onion URLs.

config.dist.ini

@@ -45,6 +45,17 @@ Operator = Contact details not set
 #
 LogLevel = info
 
+# The URL to a proxy server to use for requests to feeds, if any
+# For example, if you want to use the Tor network, you can set this to 
+# "socks5://127.0.0.1:9050"
+#
+# Proxy = socks5://yourproxy:port
+
+# Set to 1 if the proxy should only be used for .onion addresses
+# Defaults to 0
+#
+# ProxyOnionOnly = 1
+
 ###############################################################################
 
 [Matrix]

src/matrix_rssbot/classes/bot.py

@@ -22,11 +22,14 @@ from typing import Optional, List
 from configparser import ConfigParser
 from datetime import datetime
 from io import BytesIO
+from urllib.parse import urlparse
 
 import uuid
 import traceback
 import json
 
+import aiohttp
+from aiohttp_socks import ProxyConnector
 import markdown2
 import feedparser
 
@@ -108,6 +111,24 @@ class RSSBot:
         "matrix-rssbot/dev (+https://git.private.coffee/PrivateCoffee/matrix-rssbot)"
     )
 
+    @property
+    def proxy(self) -> Optional[str]:
+        """Proxy to use for HTTP requests.
+
+        Returns:
+            Optional[str]: The proxy to use for HTTP requests. Defaults to None.
+        """
+        return self.config["RSSBot"].get("Proxy")
+
+    @property
+    def proxy_onion_only(self) -> bool:
+        """Whether to use the proxy only for .onion URLs.
+
+        Returns:
+            bool: Whether to use the proxy only for .onion URLs. Defaults to False.
+        """
+        return self.config["RSSBot"].getboolean("ProxyOnionOnly", False)
+
     @classmethod
     def from_config(cls, config: ConfigParser):
         """Create a new RSSBot instance from a config file.
@@ -458,6 +479,28 @@ class RSSBot:
             room, "rssbot.event_type", {"event_type": event_type}
         )
 
+    async def fetch_feed(self, url: str) -> feedparser.FeedParserDict:
+        """Fetch the RSS feed, using Tor SOCKS5 proxy for .onion URLs.
+
+        Args:
+            url (str): The URL of the RSS feed.
+
+        Returns:
+            feedparser.FeedParserDict: The parsed RSS feed.
+        """
+        parsed = urlparse(url)
+        if self.proxy and (
+            not self.proxy_onion_only or parsed.hostname.endswith(".onion")
+        ):
+            connector = ProxyConnector.from_url(self.proxy)
+        else:
+            connector = aiohttp.TCPConnector()
+
+        async with aiohttp.ClientSession(connector=connector) as session:
+            async with session.get(url) as response:
+                content = await response.text()
+                return feedparser.parse(content)
+
     async def process_room(self, room):
         self.logger.log(f"Processing room {room}", "debug")
 
@@ -483,7 +526,7 @@ class RSSBot:
                 timestamp = 0
 
             try:
-                feed_content = feedparser.parse(feed)
+                feed_content = await self.fetch_feed(feed)
                 new_timestamp = timestamp
                 for entry in feed_content.entries:
                     try: