From b0b1feaa674c37e023121da301c6202d6b248cc1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Wed, 10 Jun 2020 21:04:33 +0200
Subject: [PATCH] request 'transparent.gif' when solving Cloudflare challenges

This currently also works without, but they might be using these to
detect potential bots in the future.
---
 gallery_dl/cloudflare.py | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/gallery_dl/cloudflare.py b/gallery_dl/cloudflare.py
index 0cf5a57b..857b4f31 100644
--- a/gallery_dl/cloudflare.py
+++ b/gallery_dl/cloudflare.py
@@ -33,11 +33,21 @@ def solve_challenge(session, response, kwargs):
     parsed = urllib.parse.urlsplit(response.url)
     root = parsed.scheme + "://" + parsed.netloc
 
+    page = response.text
+    try:
+        params = {"ray": text.extract(page, '?ray=', '"')[0]}
+
+        url = root + "/cdn-cgi/images/trace/jschal/nojs/transparent.gif"
+        session.request("GET", url, params=params)
+
+        url = root + "/cdn-cgi/images/trace/jschal/js/nocookie/transparent.gif"
+        session.request("GET", url, params=params)
+    except Exception:
+        pass
+
     cf_kwargs = {}
     headers = cf_kwargs["headers"] = collections.OrderedDict()
     params = cf_kwargs["data"] = collections.OrderedDict()
-
-    page = response.text
     url = root + text.unescape(text.extract(page, 'action="', '"')[0])
     headers["Referer"] = response.url