Skip to content

Commit

Permalink
add change notes, reduce spurious diffs
Browse files Browse the repository at this point in the history
  • Loading branch information
@mcdonc
mcdonc committed Jan 28, 2012
1 parent 3517083 commit 32764a0
Show file tree
Hide file tree
Showing 2 changed files with 75 additions and 30 deletions.
47 changes: 47 additions & 0 deletions docs/news.txt
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,53 @@ Next release
* Fix corner-case ``response.status_int`` and ``response.status`` mutation
bug on py3 (use explicit floor division).

* Backwards incompatibility: Request and BaseRequest objects now return
Unicode for ``request.path_info`` and ``request.script_name`` under Python
2. Rationale: the legacy behavior of returning the respective raw environ
values was nonsensical on Python 3. Working with non-ascii encoded environ
variables as raw WSGI values under Python 3 makes no sense, as PEP 3333
specifies that environ variables are bytes-tunneled-as-latin-1 strings, and
therefore their values are not sensible to use if there are high-order
characters in them.

If you don't care about Python 3, and you need backwards compatibility, to
get legacy behavior of returning bytes on Python 2 for these attributes,
use ``webob.LegacyRequest`` instead of ``webob.Request``. Although it's
possible to use ``webob.LegacyRequest`` under Python 3, it makes no sense
at all to do so, so don't.

* The above backwards incompatibility fixed nonsensical behavior of
``request.host_url``, ``request.application_url``, ``request.path_url``,
``request.path``, ``request.path_qs``, ``request.url``,
``request.relative_url``, ``request.path_info_peek``,
``request.path_info_pop`` under Python 3.

* The WebOb Request object now respects two WSGI environment variables:
``webob.url_encoding`` and ``webob.remote_user_encoding``.
``webob.url_encoding`` will be used to decode the raw WSGI PATH_INFO and
SCRIPT_NAME variables when the ``request.path_info`` API is used.
``webob.remote_user_encoding`` will be used to decode the raw WSGI
REMOTE_USER variable when the ``request.remote_user`` API is used.

* Request objects now accept two additional constructor parameters:
``url_encoding`` and ``remote_user_encoding``. These will be used to
decode PATH_INFO/SCRIPT_NAME and REMOTE_USER respectively from their
WSGI-encoded values.

Note that passing ``url_encoding`` will cause the WSGI environment variable
``webob.url_encoding`` to be set, and passing ``remote_user_encoding`` will
cause the WSGI environment variable ``webob.remote_user_encoding`` to be
set.

* Fix ``webob.response._request_uri`` internal function to generate sensible
request URI under Python 3. This fixed a problem under Python 3 if you
were using non-absolute Location headers in responses.

* Internal: a cached version of EnvironHeaders is no longer kept as a
_headers attribute of the request when ``request.headers`` is used, because
changes to EnvironHeaders meant the caching would have implied a circular
reference.

1.2b2
------

Expand Down
58 changes: 28 additions & 30 deletions webob/request.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -208,7 +208,7 @@ def decode(self, charset=None, errors='strict'):

new_content_type = CHARSET_RE.sub('; charset="UTF-8"',
self._content_type_raw)
content_type = native_(self.content_type)
content_type = self.content_type
r = self.__class__(
self.environ.copy(),
query_string=t.transcode_query(self.query_string),
Expand All @@ -228,7 +228,7 @@ def decode(self, charset=None, errors='strict'):
fs = cgi.FieldStorage(fp=self.body_file,
environ=fs_environ,
keep_blank_values=True,
encoding=native_(charset),
encoding=charset,
errors=errors)
else:
fs = cgi.FieldStorage(fp=self.body_file,
Expand Down Expand Up @@ -308,7 +308,7 @@ def body_file_seekable(self):
return self.body_file_raw

url_encoding = environ_getter('webob.url_encoding', 'UTF-8')
remote_user_encoding = environ_getter('webob.remote_user_encoding', 'UTF-8')
remote_user_encoding = environ_getter('webob.remote_user_encoding', 'ascii')
scheme = environ_getter('wsgi.url_scheme')
method = environ_getter('REQUEST_METHOD', 'GET')
http_version = environ_getter('SERVER_PROTOCOL')
Expand Down Expand Up @@ -345,7 +345,7 @@ def _content_type__get(self):
return self._content_type_raw.split(';', 1)[0]
def _content_type__set(self, value=None):
if value is not None:
value = native_(value)
value = str(value)
if ';' not in value:
content_type = self._content_type_raw
if ';' in content_type:
Expand Down Expand Up @@ -393,11 +393,12 @@ def client_addr(self):
``HTTP_X_FORWARDED_FOR`` has the correct values. The WSGI server
must be behind a trusted proxy for this to be true.
"""
xff = self.environ.get('HTTP_X_FORWARDED_FOR', None)
e = self.environ
xff = e.get('HTTP_X_FORWARDED_FOR', None)
if xff is not None:
addr = xff.split(',')[0].strip()
else:
addr = self.environ.get('REMOTE_ADDR', None)
addr = e.get('REMOTE_ADDR', None)
return addr

@property
Expand All @@ -412,36 +413,38 @@ def host_port(self):
the environ at all, this attribute will return the value of the
``SERVER_PORT`` header (which is guaranteed to be present).
"""
host = self.environ.get('HTTP_HOST', None)
e = self.environ
host = e.get('HTTP_HOST', None)
if host is not None:
if ':' in host:
host, port = host.split(':', 1)
else:
url_scheme = self.environ['wsgi.url_scheme']
url_scheme = e['wsgi.url_scheme']
if url_scheme == 'https':
port = '443'
else:
port = '80'
else:
port = self.environ['SERVER_PORT']
port = e['SERVER_PORT']
return port

@property
def host_url(self):
"""
The URL through the host (no path)
"""
scheme = self.environ.get('wsgi.url_scheme')
e = self.environ
scheme = e.get('wsgi.url_scheme')
url = scheme + '://'
host = self.environ.get('HTTP_HOST', None)
host = e.get('HTTP_HOST', None)
if host is not None:
if ':' in host:
host, port = host.split(':', 1)
else:
port = None
else:
host = self.environ.get('SERVER_NAME')
port = self.environ.get('SERVER_PORT')
host = e.get('SERVER_NAME')
port = e.get('SERVER_PORT')
if scheme == 'https':
if port == '443':
port = None
Expand Down Expand Up @@ -484,7 +487,7 @@ def path_qs(self):
The path of the request, without host but with query string
"""
path = self.path
qs = self.encget('QUERY_STRING', None)
qs = self.environ.get('QUERY_STRING', None)
if qs:
path += '?' + qs
return path
Expand All @@ -495,7 +498,7 @@ def url(self):
The full request URL, including QUERY_STRING
"""
url = self.path_url
qs = self.encget('QUERY_STRING', None)
qs = self.environ.get('QUERY_STRING', None)
if qs:
url += '?' + qs
return url
Expand Down Expand Up @@ -649,12 +652,10 @@ def is_xhr(self):

def _host__get(self):
"""Host name provided in HTTP_HOST, with fall-back to SERVER_NAME"""
host = self.environ.get('HTTP_HOST')
if host is None:
h = self.environ.get('SERVER_NAME')
p = self.environ.get('SERVER_PORT')
host = h + ':' + p
return host
if 'HTTP_HOST' in self.environ:
return self.environ['HTTP_HOST']
else:
return '%(SERVER_NAME)s:%(SERVER_PORT)s' % self.environ
def _host__set(self, value):
self.environ['HTTP_HOST'] = value
def _host__del(self):
Expand Down Expand Up @@ -987,7 +988,7 @@ def _cache_control__get(self):
<http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.9>`_)
"""
env = self.environ
value = self.environ.get('HTTP_CACHE_CONTROL', '')
value = env.get('HTTP_CACHE_CONTROL', '')
cache_header, cache_obj = env.get('webob._cache_control', (None, None))
if cache_obj is not None and cache_header == value:
return cache_obj
Expand All @@ -1004,10 +1005,10 @@ def _cache_control__set(self, value):
value = CacheControl(value, type='request')
if isinstance(value, CacheControl):
str_value = str(value)
self.environ['HTTP_CACHE_CONTROL'] = str_value
env['HTTP_CACHE_CONTROL'] = str_value
env['webob._cache_control'] = (str_value, value)
else:
self.environ['HTTP_CACHE_CONTROL'] = value
env['HTTP_CACHE_CONTROL'] = value
env['webob._cache_control'] = (None, None)

def _cache_control__del(self):
Expand Down Expand Up @@ -1076,15 +1077,13 @@ def as_bytes(self, skip_body=False):
host = self.host_url
assert url.startswith(host)
url = url[len(host):]
method = self.method
http_version = self.http_version
parts = [bytes_('%s %s %s' % (method, url, http_version))]
parts = [bytes_('%s %s %s' % (self.method, url, self.http_version))]
#self.headers.setdefault('Host', self.host)

# acquire body before we handle headers so that
# content-length will be set
body = None
if self.method in ('POST', 'PUT'):
if self.method in ('PUT', 'POST'):
if skip_body > 1:
if len(self.body) > skip_body:
body = bytes_('<body skipped (len=%s)>' % len(self.body))
Expand Down Expand Up @@ -1186,7 +1185,7 @@ def from_file(cls, fp):
if hname in r.headers:
hval = r.headers[hname] + ', ' + hval
r.headers[hname] = hval
if r.method in ('POST', 'PUT'):
if r.method in ('PUT', 'POST'):
clen = r.content_length
if clen is None:
body = fp.read()
Expand Down Expand Up @@ -1590,7 +1589,6 @@ def _encode_multipart(vars, content_type, fout=None):
return content_type, f.getvalue()

def detect_charset(ctype):
ctype = native_(ctype)
m = CHARSET_RE.search(ctype)
if m:
return m.group(1).strip('"').strip()
Expand Down

0 comments on commit 32764a0

Please sign in to comment.