From: John Runyon <redacted>
Date: Mon, 9 Oct 2023 00:18:08 +0000 (-0600)
Subject: urls - swallow InvalidURL responses
X-Git-Url: https://jfr.im/git/erebus.git/commitdiff_plain/169ed3b58865476d4337d09a1cdfda1c066cd436

urls - swallow InvalidURL responses
---

diff --git a/modules/urls.py b/modules/urls.py
index 859f9ac..36c3a58 100644
--- a/modules/urls.py
+++ b/modules/urls.py
@@ -35,6 +35,7 @@ else:
 	import urllib.parse as urlparse
 	import html
 	from bs4 import BeautifulSoup
+import http.client
 
 import re, json, datetime
 
@@ -241,7 +242,7 @@ def _humanize_bytes(b):
 		return "%.2f%siB" % (b, table[i])
 
 def _do_request(url, try_aia=False):
-	"""Returns the HTTPResponse object, or a string on error"""
+	"""Returns the HTTPResponse object, or a string on error. Empty string -> no response."""
 	request = urllib2.Request(url, headers={'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36', 'Sec-Ch-Ua': '"Chromium";v="116", "Not)A;Brand";v="24", "Google Chrome";v="116"', 'Sec-Ch-Ua-Mobile': '?0', 'Sec-Ch-Ua-Platform': '"Linux"', 'Sec-Fetch-Dest': 'document', 'Sec-Fetch-Mode': 'navigate', 'Sec-Fetch-Site': 'same-origin', 'Sec-Fetch-User': '?1', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7', 'Accept-Language': 'en-US,en;q=0.9', 'Cache-Control': 'no-cache', 'Pragma': 'no-cache', 'Upgrade-Insecure-Requests': '1'})
 	if try_aia:
 		opener = urllib2.build_opener(urllib2.HTTPSHandler(context=aia_session.ssl_context_from_url(url)), SmartRedirectHandler())
@@ -251,6 +252,8 @@ def _do_request(url, try_aia=False):
 	# Send request and handle errors
 	try:
 		response = opener.open(request, timeout=2)
+	except http.client.InvalidURL as e: # why does a method under urllib.request raise an exception under http.client???
+		return ''
 	except urllib2.HTTPError as e:
 		return 'Request error: %s %s' % (e.code, e.reason)
 	except urllib2.URLError as e:
@@ -265,7 +268,7 @@ def _do_request(url, try_aia=False):
 	except TimeoutError as e:
 		return 'Request error: request timed out'
 	except Exception as e:
-		return 'Unknown error: %s %r' % (type(e).__name__, e.args)
+		return 'Unknown error: %s %s %r' % (e.__module__, type(e).__name__, e.args)
 
 	return response