From 6212db897aa5cb61615f9fdefc08adf4cb7f7476 Mon Sep 17 00:00:00 2001
From: Chris McDonough <chrism@plope.com>
Date: Wed, 4 Jan 2012 17:53:42 -0500
Subject: more speculative changes regarding matching

---
 pyramid/traversal.py   | 13 +++++++++----
 pyramid/urldispatch.py | 19 +++++++------------
 2 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/pyramid/traversal.py b/pyramid/traversal.py
index 3489dade0..2443e2d1c 100644
--- a/pyramid/traversal.py
+++ b/pyramid/traversal.py
@@ -444,11 +444,12 @@ def traversal_path(path):
     path = unquote_bytes_to_wsgi(path) # result will be a native string
     return traversal_path_info(path)
 
+@lru_cache(1000)
 def traversal_path_info(path):
     """ Given``path``, return a tuple representing that path which can be
     used to traverse a resource tree.  ``path`` is assumed to be an
     already-URL-decoded ``str`` type as if it had come to us from an upstream
-    WSGI server as the ``PATH_INFO`` environment variable.
+    WSGI server as the ``PATH_INFO`` environ variable.
 
     The ``path`` is first decoded to from its WSGI representation to Unicode;
     it is decoded differently depending on platform:
@@ -457,9 +458,9 @@ def traversal_path_info(path):
       decoding directly; a :exc:`pyramid.exc.URLDecodeError` is raised if a the
       URL cannot be decoded.
 
-    - On Python 3, as per the WSGI spec, ``path`` is first encoded to bytes
-      using the Latin-1 encoding; the resulting set of bytes is subsequently
-      decoded to text using the UTF-8 encoding; a
+    - On Python 3, as per the PEP 3333 spec, ``path`` is first encoded to
+      bytes using the Latin-1 encoding; the resulting set of bytes is
+      subsequently decoded to text using the UTF-8 encoding; a
       :exc:`pyramid.exc.URLDecodeError` is raised if a the URL cannot be
       decoded.
 
@@ -654,9 +655,13 @@ class ResourceTreeTraverser(object):
             # this request did not match a route
             subpath = ()
             try:
+                # empty if mounted under a path in mod_wsgi, for example
                 path = decode_path_info(environ['PATH_INFO'] or '/')
             except KeyError:
                 path = '/'
+            except UnicodeDecodeError as e:
+                raise URLDecodeError(e.encoding, e.object, e.start, e.end,
+                                     e.reason)
 
         if VH_ROOT_KEY in environ:
             # HTTP_X_VHM_ROOT
diff --git a/pyramid/urldispatch.py b/pyramid/urldispatch.py
index c7520b8d2..73875b675 100644
--- a/pyramid/urldispatch.py
+++ b/pyramid/urldispatch.py
@@ -18,8 +18,9 @@ from pyramid.compat import (
 from pyramid.exceptions import URLDecodeError
 
 from pyramid.traversal import (
-    traversal_path_info,
     quote_path_segment,
+    decode_path_info,
+    split_path_info,
     )
 
 _marker = object()
@@ -70,9 +71,11 @@ class RoutesMapper(object):
         environ = request.environ
         try:
             # empty if mounted under a path in mod_wsgi, for example
-            path = environ['PATH_INFO'] or '/' 
+            path = decode_path_info(environ['PATH_INFO'] or '/')
         except KeyError:
             path = '/'
+        except UnicodeDecodeError as e:
+            raise URLDecodeError(e.encoding, e.object, e.start, e.end, e.reason)
 
         for route in self.routelist:
             match = route.match(path)
@@ -147,17 +150,9 @@ def _compile_route(route):
         d = {}
         for k, v in m.groupdict().items():
             if k == star:
-                d[k] = traversal_path_info(v)
+                d[k] = split_path_info(v)
             else:
-                try:
-                    val = bytes_(v).decode('utf-8', 'strict')
-                    d[k] = val
-                except UnicodeDecodeError as e:
-                    raise URLDecodeError(
-                        e.encoding, e.object, e.start, e.end, e.reason
-                        )
-                        
-                        
+                d[k] = v
         return d
                     
 
-- 
cgit v1.2.3