diff options
Diffstat (limited to 'netlib')
| -rw-r--r-- | netlib/http/http1/protocol.py | 14 | ||||
| -rw-r--r-- | netlib/http/semantics.py | 43 | ||||
| -rw-r--r-- | netlib/odict.py | 3 | ||||
| -rw-r--r-- | netlib/tutils.py | 4 | ||||
| -rw-r--r-- | netlib/utils.py | 56 |
5 files changed, 87 insertions, 33 deletions
diff --git a/netlib/http/http1/protocol.py b/netlib/http/http1/protocol.py index 2e85a762..31e9cc85 100644 --- a/netlib/http/http1/protocol.py +++ b/netlib/http/http1/protocol.py @@ -360,20 +360,6 @@ class HTTP1Protocol(semantics.ProtocolMixin): @classmethod - def request_preamble(self, method, resource, http_major="1", http_minor="1"): - return '%s %s HTTP/%s.%s' % ( - method, resource, http_major, http_minor - ) - - - @classmethod - def response_preamble(self, code, message=None, http_major="1", http_minor="1"): - if message is None: - message = status_codes.RESPONSES.get(code) - return 'HTTP/%s.%s %s %s' % (http_major, http_minor, code, message) - - - @classmethod def has_chunked_encoding(self, headers): return "chunked" in [ i.lower() for i in utils.get_header_tokens(headers, "transfer-encoding") diff --git a/netlib/http/semantics.py b/netlib/http/semantics.py index e7ae2b5f..974fe6e6 100644 --- a/netlib/http/semantics.py +++ b/netlib/http/semantics.py @@ -7,7 +7,7 @@ import urllib import urlparse from .. import utils, odict -from . import cookies +from . import cookies, exceptions from netlib import utils, encoding HDR_FORM_URLENCODED = "application/x-www-form-urlencoded" @@ -18,10 +18,10 @@ CONTENT_MISSING = 0 class ProtocolMixin(object): - def read_request(self): + def read_request(self, *args, **kwargs): # pragma: no cover raise NotImplemented - def read_response(self): + def read_response(self, *args, **kwargs): # pragma: no cover raise NotImplemented def assemble(self, message): @@ -32,14 +32,23 @@ class ProtocolMixin(object): else: raise ValueError("HTTP message not supported.") - def assemble_request(self, request): + def assemble_request(self, request): # pragma: no cover raise NotImplemented - def assemble_response(self, response): + def assemble_response(self, response): # pragma: no cover raise NotImplemented class Request(object): + # This list is adopted legacy code. + # We probably don't need to strip off keep-alive. + _headers_to_strip_off = [ + 'Proxy-Connection', + 'Keep-Alive', + 'Connection', + 'Transfer-Encoding', + 'Upgrade', + ] def __init__( self, @@ -71,7 +80,6 @@ class Request(object): self.timestamp_start = timestamp_start self.timestamp_end = timestamp_end - def __eq__(self, other): try: self_d = [self.__dict__[k] for k in self.__dict__ if k not in ('timestamp_start', 'timestamp_end')] @@ -114,7 +122,7 @@ class Request(object): self.httpversion[1], ) else: - raise http.HttpError(400, "Invalid request form") + raise exceptions.HttpError(400, "Invalid request form") def anticache(self): """ @@ -143,7 +151,7 @@ class Request(object): if self.headers["accept-encoding"]: self.headers["accept-encoding"] = [ ', '.join( - e for e in encoding.ENCODINGS if e in self.headers["accept-encoding"][0])] + e for e in encoding.ENCODINGS if e in self.headers.get_first("accept-encoding"))] def update_host_header(self): """ @@ -317,12 +325,12 @@ class Request(object): self.scheme, self.host, self.port, self.path = parts @property - def content(self): + def content(self): # pragma: no cover # TODO: remove deprecated getter return self.body @content.setter - def content(self, content): + def content(self, content): # pragma: no cover # TODO: remove deprecated setter self.body = content @@ -343,6 +351,11 @@ class EmptyRequest(Request): class Response(object): + _headers_to_strip_off = [ + 'Proxy-Connection', + 'Alternate-Protocol', + 'Alt-Svc', + ] def __init__( self, @@ -368,7 +381,6 @@ class Response(object): self.timestamp_start = timestamp_start self.timestamp_end = timestamp_end - def __eq__(self, other): try: self_d = [self.__dict__[k] for k in self.__dict__ if k not in ('timestamp_start', 'timestamp_end')] @@ -393,7 +405,6 @@ class Response(object): size=size ) - def get_cookies(self): """ Get the contents of all Set-Cookie headers. @@ -430,21 +441,21 @@ class Response(object): self.headers["Set-Cookie"] = values @property - def content(self): + def content(self): # pragma: no cover # TODO: remove deprecated getter return self.body @content.setter - def content(self, content): + def content(self, content): # pragma: no cover # TODO: remove deprecated setter self.body = content @property - def code(self): + def code(self): # pragma: no cover # TODO: remove deprecated getter return self.status_code @code.setter - def code(self, code): + def code(self, code): # pragma: no cover # TODO: remove deprecated setter self.status_code = code diff --git a/netlib/odict.py b/netlib/odict.py index d02de08d..11d5d52a 100644 --- a/netlib/odict.py +++ b/netlib/odict.py @@ -91,8 +91,9 @@ class ODict(object): self.lst = self._filter_lst(k, self.lst) def __contains__(self, k): + k = self._kconv(k) for i in self.lst: - if self._kconv(i[0]) == self._kconv(k): + if self._kconv(i[0]) == k: return True return False diff --git a/netlib/tutils.py b/netlib/tutils.py index 5018b9e8..3c471d0d 100644 --- a/netlib/tutils.py +++ b/netlib/tutils.py @@ -119,7 +119,7 @@ def tresp(content="message"): "OK", headers, content, - time.time(), - time.time(), + timestamp_start=time.time(), + timestamp_end=time.time(), ) return resp diff --git a/netlib/utils.py b/netlib/utils.py index 35ea0ec7..2dfcafc6 100644 --- a/netlib/utils.py +++ b/netlib/utils.py @@ -4,6 +4,7 @@ import cgi import urllib import urlparse import string +import re def isascii(s): @@ -239,3 +240,58 @@ def urldecode(s): Takes a urlencoded string and returns a list of (key, value) tuples. """ return cgi.parse_qsl(s, keep_blank_values=True) + + +def parse_content_type(c): + """ + A simple parser for content-type values. Returns a (type, subtype, + parameters) tuple, where type and subtype are strings, and parameters + is a dict. If the string could not be parsed, return None. + + E.g. the following string: + + text/html; charset=UTF-8 + + Returns: + + ("text", "html", {"charset": "UTF-8"}) + """ + parts = c.split(";", 1) + ts = parts[0].split("/", 1) + if len(ts) != 2: + return None + d = {} + if len(parts) == 2: + for i in parts[1].split(";"): + clause = i.split("=", 1) + if len(clause) == 2: + d[clause[0].strip()] = clause[1].strip() + return ts[0].lower(), ts[1].lower(), d + + +def multipartdecode(hdrs, content): + """ + Takes a multipart boundary encoded string and returns list of (key, value) tuples. + """ + v = hdrs.get_first("content-type") + if v: + v = parse_content_type(v) + if not v: + return [] + boundary = v[2].get("boundary") + if not boundary: + return [] + + rx = re.compile(r'\bname="([^"]+)"') + r = [] + + for i in content.split("--" + boundary): + parts = i.splitlines() + if len(parts) > 1 and parts[0][0:2] != "--": + match = rx.search(parts[1]) + if match: + key = match.group(1) + value = "".join(parts[3 + parts[2:].index(""):]) + r.append((key, value)) + return r + return [] |