squashme

sopel/builtins/url.py

@@ -26,6 +26,7 @@
 
 if TYPE_CHECKING:
     from collections.abc import Generator
+
     from sopel.bot import Sopel, SopelWrapper
     from sopel.config import Config
     from sopel.trigger import Trigger
@@ -426,10 +427,16 @@ def process_urls(
             continue
 
         # Call the URL to get a title, if possible
+        unsafe_urls = [
+            url
+            for url, data in bot.memory.get("safety_cache", {}).items()
+            if data.get("positives")
+        ]
         title_results = find_title(
             url,
             allow_local=bot.config.url.enable_private_resolution,
-            memory=bot.memory,
+            unsafe_urls=unsafe_urls,
+            unsafe_domains=bot.memory.get("safety_cache_local", {}).values(),
         )
         if not title_results:
             # No title found: don't handle this URL
@@ -484,13 +491,15 @@ def find_title(
     url: str,
     verify: bool = True,
     allow_local: bool = False,
-    memory: tools.SopelMemory = {},
-) -> Optional[Tuple[str, str]]:
+    unsafe_urls: list = [],
+    unsafe_domains: list = [],
+) -> Optional[tuple[str, str]]:
     """Fetch the title for the given URL.
 
     :param verify: Whether to require a valid certificate when using https
     :param allow_local: Allow requests to non-global addresses (RFC1918, etc.)
-    :param memory: The bot.memory to search for safety.py caches
+    :param unsafe_urls: A list of URLs to consider malicious and ignore
+    :param unsafe_domains: A list of domains to consider malicious and ignore
     :return: A tuple of the (title, final_hostname) that were found, or None
     """
     original_url = url
@@ -501,14 +510,14 @@ def find_title(
     while redirects_left > 0:
         redirects_left -= 1
         parsed_url = urlparse(url)
+        if not parsed_url.hostname:
+            return None
 
         # Avoid fetching known malicious links
-        safety_cache = bot.memory.get("safety_cache", {})
-        safety_cache_local = bot.memory.get("safety_cache_local", {})
-        if url in safety_cache and safety_cache[url]["positives"] > 0:
-                LOGGER.debug("Ignoring unsafe URL: %r", url)
-                return None
-        if parsed_url.hostname.lower() in safety_cache_local:
+        if url in unsafe_urls:
+            LOGGER.debug("Ignoring unsafe URL: %r", url)
+            return None
+        if parsed_url.hostname.lower() in unsafe_domains:
             LOGGER.debug("Ignoring unsafe domain: %r", url)
             return None
 
@@ -548,9 +557,9 @@ def find_title(
                 LOGGER.debug(
                     "URL %r redirected to %r", url, response.headers.get("Location")
                 )
-                url = response.headers.get("Location")
-                if url is None:
+                if "Location" not in response.headers:
                     return None
+                url = response.headers["Location"]
                 continue
 
             content_bytes = b''