From 314be33707ce047716fcbd7453346c10f10e0640 Mon Sep 17 00:00:00 2001
From: Noah Levitt <nlevitt@archive.org>
Date: Wed, 19 Oct 2016 13:43:44 -0700
Subject: [PATCH] new test that reveals connection hang on https urls missing a
 content-length http response header (not chunked and server leaves connection
 open) -- reported by Alex Osborne

---
 setup.py               |  2 +-
 tests/test_warcprox.py | 42 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index 6eca329..e29a4f1 100755
--- a/setup.py
+++ b/setup.py
@@ -51,7 +51,7 @@ except:
 
 setuptools.setup(
         name='warcprox',
-        version='2.0b2.dev29',
+        version='2.0b2.dev30',
         description='WARC writing MITM HTTP/S proxy',
         url='https://github.com/internetarchive/warcprox',
         author='Noah Levitt',
diff --git a/tests/test_warcprox.py b/tests/test_warcprox.py
index 10f214a..6d4b986 100755
--- a/tests/test_warcprox.py
+++ b/tests/test_warcprox.py
@@ -123,6 +123,11 @@ class _TestHttpRequestHandler(http_server.BaseHTTPRequestHandler):
                     +  special_header + b'\r\n'
                     +  b'Content-Length: ' + str(len(payload)).encode('ascii') + b'\r\n'
                     +  b'\r\n')
+        elif self.path == '/missing-content-length':
+            headers = (b'HTTP/1.1 200 OK\r\n'
+                    +  b'Content-Type: text/plain\r\n'
+                    +  b'\r\n')
+            payload = b'This response is missing a Content-Length http header.'
         else:
             payload = b'404 Not Found\n'
             headers = (b'HTTP/1.1 404 Not Found\r\n'
@@ -1103,6 +1108,43 @@ def _test_tor_onion(archiving_proxies):
         proxies=archiving_proxies, verify=False, allow_redirects=False)
     assert response.status_code == 200
 
+def test_missing_content_length(archiving_proxies, http_daemon, https_daemon):
+    # double-check that our test http server is responding as expected
+    url = 'http://localhost:%s/missing-content-length' % (
+            http_daemon.server_port)
+    response = requests.get(url, verify=False, timeout=10)
+    assert response.content == (
+            b'This response is missing a Content-Length http header.')
+    assert not 'content-length' in response.headers
+
+    # double-check that our test https server is responding as expected
+    url = 'https://localhost:%s/missing-content-length' % (
+            https_daemon.server_port)
+    response = requests.get(url, verify=False, timeout=10)
+    assert response.content == (
+            b'This response is missing a Content-Length http header.')
+    assert not 'content-length' in response.headers
+
+    # now check that the proxy doesn't hang (http)
+    url = 'http://localhost:%s/missing-content-length' % (
+            http_daemon.server_port)
+    response = requests.get(
+            url, proxies=archiving_proxies, verify=False, timeout=10)
+    assert response.content == (
+            b'This response is missing a Content-Length http header.')
+    assert not 'content-length' in response.headers
+
+    # now check that the proxy doesn't hang (https)
+    url = 'https://localhost:%s/missing-content-length' % (
+            https_daemon.server_port)
+    # before fixing the issue this tests for, this would fail by raising
+    # requests.exceptions.ConnectionError: ... Read timed out
+    response = requests.get(
+            url, proxies=archiving_proxies, verify=False, timeout=10)
+    assert response.content == (
+            b'This response is missing a Content-Length http header.')
+    assert not 'content-length' in response.headers
+
 if __name__ == '__main__':
     pytest.main()