[1.5.x] Fixed #19468 -- Decoded request.path correctly on Python 3.

Thanks aliva for the report and claudep for the feedback.

Backport of 1e4a27d from master.

django/contrib/staticfiles/handlers.py

@@ -6,6 +6,7 @@
     from urlparse import urlparse
 
 from django.conf import settings
+from django.core.handlers.base import get_path_info
 from django.core.handlers.wsgi import WSGIHandler
 
 from django.contrib.staticfiles import utils
@@ -67,6 +68,6 @@ def get_response(self, request):
         return super(StaticFilesHandler, self).get_response(request)
 
     def __call__(self, environ, start_response):
-        if not self._should_handle(environ['PATH_INFO']):
+        if not self._should_handle(get_path_info(environ)):
             return self.application(environ, start_response)
         return super(StaticFilesHandler, self).__call__(environ, start_response)

django/core/handlers/base.py

@@ -5,10 +5,14 @@
 import types
 
 from django import http
+from django.conf import settings
+from django.core import exceptions
+from django.core import urlresolvers
 from django.core import signals
 from django.utils.encoding import force_text
 from django.utils.importlib import import_module
 from django.utils import six
+from django.views import debug
 
 logger = logging.getLogger('django.request')
 
@@ -32,8 +36,6 @@ def load_middleware(self):
 
         Must be called after the environment is fixed (see __call__ in subclasses).
         """
-        from django.conf import settings
-        from django.core import exceptions
         self._view_middleware = []
         self._template_response_middleware = []
         self._response_middleware = []
@@ -75,9 +77,6 @@ def load_middleware(self):
 
     def get_response(self, request):
         "Returns an HttpResponse object for the given HttpRequest"
-        from django.core import exceptions, urlresolvers
-        from django.conf import settings
-
         try:
             # Setup default url resolver for this thread, this code is outside
             # the try/except so we don't get a spurious "unbound local
@@ -147,7 +146,6 @@ def get_response(self, request):
                                 'request': request
                             })
                 if settings.DEBUG:
-                    from django.views import debug
                     response = debug.technical_404_response(request, e)
                 else:
                     try:
@@ -204,8 +202,6 @@ def handle_uncaught_exception(self, request, resolver, exc_info):
         caused by anything, so assuming something like the database is always
         available would be an error.
         """
-        from django.conf import settings
-
         if settings.DEBUG_PROPAGATE_EXCEPTIONS:
             raise
 
@@ -218,7 +214,6 @@ def handle_uncaught_exception(self, request, resolver, exc_info):
         )
 
         if settings.DEBUG:
-            from django.views import debug
             return debug.technical_500_response(request, *exc_info)
 
         # If Http500 handler is not installed, re-raise last exception
@@ -238,6 +233,20 @@ def apply_response_fixes(self, request, response):
             response = func(request, response)
         return response
 
+
+def get_path_info(environ):
+    """
+    Returns the HTTP request's PATH_INFO as a unicode string.
+    """
+    path_info = environ.get('PATH_INFO', str('/'))
+    # Under Python 3, strings in environ are decoded with ISO-8859-1;
+    # re-encode to recover the original bytestring provided by the webserver.
+    if six.PY3:
+        path_info = path_info.encode('iso-8859-1')
+    # It'd be better to implement URI-to-IRI decoding, see #19508.
+    return path_info.decode('utf-8')
+
+
 def get_script_name(environ):
     """
     Returns the equivalent of the HTTP request's SCRIPT_NAME environment
@@ -246,7 +255,6 @@ def get_script_name(environ):
     from the client's perspective), unless the FORCE_SCRIPT_NAME setting is
     set (to anything).
     """
-    from django.conf import settings
     if settings.FORCE_SCRIPT_NAME is not None:
         return force_text(settings.FORCE_SCRIPT_NAME)
 
@@ -255,9 +263,14 @@ def get_script_name(environ):
     # rewrites. Unfortunately not every Web server (lighttpd!) passes this
     # information through all the time, so FORCE_SCRIPT_NAME, above, is still
     # needed.
-    script_url = environ.get('SCRIPT_URL', '')
-    if not script_url:
-        script_url = environ.get('REDIRECT_URL', '')
+    script_url = environ.get('SCRIPT_URL', environ.get('REDIRECT_URL', str('')))
     if script_url:
-        return force_text(script_url[:-len(environ.get('PATH_INFO', ''))])
-    return force_text(environ.get('SCRIPT_NAME', ''))
+        script_name = script_url[:-len(environ.get('PATH_INFO', str('')))]
+    else:
+        script_name = environ.get('SCRIPT_NAME', str(''))
+    # Under Python 3, strings in environ are decoded with ISO-8859-1;
+    # re-encode to recover the original bytestring provided by the webserver.
+    if six.PY3:
+        script_name = script_name.encode('iso-8859-1')
+    # It'd be better to implement URI-to-IRI decoding, see #19508.
+    return script_name.decode('utf-8')

django/core/handlers/wsgi.py

@@ -128,7 +128,7 @@ def readline(self, size=None):
 class WSGIRequest(http.HttpRequest):
     def __init__(self, environ):
         script_name = base.get_script_name(environ)
-        path_info = force_text(environ.get('PATH_INFO', '/'))
+        path_info = base.get_path_info(environ)
         if not path_info or path_info == script_name:
             # Sometimes PATH_INFO exists, but is empty (e.g. accessing
             # the SCRIPT_NAME URL without a trailing slash). We really need to

django/test/client.py

@@ -245,7 +245,11 @@ def _get_path(self, parsed):
         # If there are parameters, add them
         if parsed[3]:
             path += str(";") + force_str(parsed[3])
-        return unquote(path)
+        path = unquote(path)
+        # WSGI requires latin-1 encoded strings. See get_path_info().
+        if six.PY3:
+            path = path.encode('utf-8').decode('iso-8859-1')
+        return path
 
     def get(self, path, data={}, **extra):
         "Construct a GET request."

tests/regressiontests/handlers/tests.py

@@ -1,6 +1,7 @@
 from django.core.handlers.wsgi import WSGIHandler
 from django.test import RequestFactory
 from django.test.utils import override_settings
+from django.utils import six
 from django.utils import unittest
 
 class HandlerTests(unittest.TestCase):
@@ -22,7 +23,7 @@ def test_lock_safety(self):
     def test_bad_path_info(self):
         """Tests for bug #15672 ('request' referenced before assignment)"""
         environ = RequestFactory().get('/').environ
-        environ['PATH_INFO'] = b'\xed'
+        environ['PATH_INFO'] = '\xed'
         handler = WSGIHandler()
         response = handler(environ, lambda *a, **k: None)
         self.assertEqual(response.status_code, 400)

tests/regressiontests/requests/tests.py

@@ -11,6 +11,7 @@
 from django.http import HttpRequest, HttpResponse, parse_cookie, build_request_repr, UnreadablePostError
 from django.test.client import FakePayload
 from django.test.utils import override_settings, str_prefix
+from django.utils import six
 from django.utils import unittest
 from django.utils.http import cookie_date, urlencode
 from django.utils.timezone import utc
@@ -57,6 +58,16 @@ def test_wsgirequest_repr(self):
         self.assertEqual(build_request_repr(request, path_override='/otherpath/', GET_override={'a': 'b'}, POST_override={'c': 'd'}, COOKIES_override={'e': 'f'}, META_override={'g': 'h'}),
                          str_prefix("<WSGIRequest\npath:/otherpath/,\nGET:{%(_)s'a': %(_)s'b'},\nPOST:{%(_)s'c': %(_)s'd'},\nCOOKIES:{%(_)s'e': %(_)s'f'},\nMETA:{%(_)s'g': %(_)s'h'}>"))
 
+    def test_wsgirequest_path_info(self):
+        def wsgi_str(path_info):
+            path_info = path_info.encode('utf-8')           # Actual URL sent by the browser (bytestring)
+            if six.PY3:
+                path_info = path_info.decode('iso-8859-1')  # Value in the WSGI environ dict (native string)
+            return path_info
+        # Regression for #19468
+        request = WSGIRequest({'PATH_INFO': wsgi_str("/سلام/"), 'REQUEST_METHOD': 'get', 'wsgi.input': BytesIO(b'')})
+        self.assertEqual(request.path, "/سلام/")
+
     def test_parse_cookie(self):
         self.assertEqual(parse_cookie('invalid@key=true'), {})