wsgi: latin-1 encoding dance for environ[PATH_INFO]468-wsgi-latin1

https://www.python.org/dev/peps/pep-0333/#unicode-issues
https://github.com/eventlet/eventlet/issues/468

2 files changed, 31 insertions, 2 deletions

diff --git a/eventlet/wsgi.py b/eventlet/wsgi.py
index 7576f0e..9f49090 100644
--- a/eventlet/wsgi.py
+++ b/eventlet/wsgi.py
@@ -59,6 +59,12 @@ def addr_to_host_port(addr):
     return (host, port)
 
 
+def encode_dance(s):
+    if not isinstance(s, bytes):
+        s = s.encode('utf-8', 'replace')
+    return s.decode('latin1')
+
+
 # Collections of error codes to compare against.  Not all attributes are set
 # on errno module on all platforms, so some are literals :(
 BAD_SOCK = set((errno.EBADF, 10053))
@@ -631,7 +637,7 @@ class HttpProtocol(BaseHTTPServer.BaseHTTPRequestHandler):
 
         pq = self.path.split('?', 1)
         env['RAW_PATH_INFO'] = pq[0]
-        env['PATH_INFO'] = urllib.parse.unquote(pq[0])
+        env['PATH_INFO'] = encode_dance(urllib.parse.unquote(pq[0]))
         if len(pq) > 1:
             env['QUERY_STRING'] = pq[1]
 
diff --git a/tests/wsgi_test.py b/tests/wsgi_test.py
index 48da085..d680154 100644
--- a/tests/wsgi_test.py
+++ b/tests/wsgi_test.py
@@ -1,3 +1,4 @@
+# coding: utf-8
 import cgi
 import collections
 import errno
@@ -1398,10 +1399,32 @@ class TestHttpd(_TestBase):
         sock = eventlet.connect(self.server_addr)
         sock.sendall(b'GET /a*b@%40%233 HTTP/1.1\r\nHost: localhost\r\nConnection: close\r\n\r\n')
         result = read_http(sock)
-        self.assertEqual(result.status, 'HTTP/1.1 200 OK')
+        assert result.status == 'HTTP/1.1 200 OK'
         assert b'decoded: /a*b@@#3' in result.body
         assert b'raw: /a*b@%40%233' in result.body
 
+    def test_path_info_latin1(self):
+        # https://github.com/eventlet/eventlet/issues/468
+        g = []
+
+        def wsgi_app(environ, start_response):
+            g.append(environ['PATH_INFO'])
+            start_response("200 OK", [])
+            return b''
+
+        self.site.application = wsgi_app
+        sock = eventlet.connect(self.server_addr)
+        sock.sendall(b'GET /%E4%BD%A0%E5%A5%BD HTTP/1.1\r\nHost: localhost\r\nConnection: close\r\n\r\n')
+        result = read_http(sock)
+        assert result.status == 'HTTP/1.1 200 OK'
+        # that was only preparation, actual test below
+        # Per PEP-0333 https://www.python.org/dev/peps/pep-0333/#unicode-issues
+        # in all WSGI environment strings application must observe either bytes in latin-1 (ISO-8859-1)
+        # or unicode code points \u0000..\u00ff
+        # wsgi_decoding_dance from Werkzeug to emulate concerned application
+        decoded = g[0].encode('latin1').decode('utf-8', 'replace')
+        assert decoded == u'/你好'
+
     @tests.skip_if_no_ipv6
     def test_ipv6(self):
         try: