def decode(cls, data, charset=None, mimetype=None):
boundary = mimetype.boundary.encode('ISO8859-1')
parts = data.split(b'--%s' % (boundary,))
part = parts.pop(0)
if part:
raise DecodeError(_(u'Data before boundary: %r'), part.decode('ISO8859-1'))
try:
part = parts.pop()
except IndexError:
raise DecodeError(u'No end of boundary')
if part not in (b'--', b'--\r\n'):
raise DecodeError(_(u'Invalid multipart end: %r'), part.decode('ISO8859-1'))
from httoop.messages.body import Body
multiparts = []
for part in parts:
if not part.startswith('\r\n'):
raise DecodeError(_(u'Invalid boundary end: %r'), part[:2].decode('ISO8859-1'))
part = part[2:]
headers, separator, content = part.partition(b'\r\n\r\n')
if not separator:
raise DecodeError(_(u'Multipart does not contain CRLF header separator'))
if not content.endswith('\r\n'):
raise DecodeError(_(u'Multipart does not end with CRLF: %r'), content[-2].decode('ISO8859-1'))
content = content[:-2]
body = Body()
body.headers.clear()
body.headers.parse(headers)
body.headers.setdefault('Content-Type', cls.default_content_type)
body.parse(content)
multiparts.append(body)
return multiparts
def _unquote_host(self, host):
# IPv6 / IPvFuture
if host.startswith(b'[') and host.endswith(b']'):
host = host[1:-1]
try:
return u'[%s]' % inet_ntop(AF_INET6, inet_pton(AF_INET6, host)).decode('ascii')
except SocketError:
# IPvFuture
if host.startswith(b'v') and b'.' in host and host[1:].split(b'.', 1)[0].isdigit():
try:
return u'[%s]' % host.decode('ascii')
except UnicodeDecodeError:
raise InvalidURI(_('Invalid IPvFuture address: must be ASCII.'))
raise InvalidURI(_('Invalid IP address in URI.'))
# IPv4
if all(x.isdigit() for x in host.split(b'.')):
try:
return inet_ntop(AF_INET, inet_pton(AF_INET, host)).decode('ascii')
except SocketError:
raise InvalidURI(_('Invalid IPv4 address in URI.'))
if host.strip(Percent.UNRESERVED + Percent.SUB_DELIMS + b'%'):
raise InvalidURI(_('Invalid URI host.'))
# DNS hostname
host = self.unquote(host)
try:
return host.encode('ascii').decode('idna').lower()
except UnicodeError:
raise InvalidURI(_('Invalid host.'))
def parse(self, data):
r"""parses HTTP headers
:param data:
the header string containing headers separated by "\r\n"
without trailing "\r\n"
:type data: bytes
"""
lines = data.split(b'\r\n')
while lines:
curr = lines.pop(0)
name, __, value = curr.partition(b':')
if __ != b':':
raise InvalidHeader(_(u"Invalid header line: %r"), curr.decode('ISO8859-1'))
if self.HEADER_RE.search(name):
raise InvalidHeader(_(u"Invalid header name: %r"), name.decode('ISO8859-1'))
name, value = name.strip(), [value.lstrip()]
# continuation lines
while lines and lines[0].startswith((' ', '\t')):
value.append(lines.pop(0)[1:])
value = b''.join(value).rstrip()
Element = HEADER.get(name, HeaderElement)
value = Element.decode(value)
self.append(name, value)
def parse(self, data):
r"""parses HTTP headers
:param data:
the header string containing headers separated by "\r\n"
without trailing "\r\n"
:type data: bytes
"""
lines = data.split(b'\r\n')
while lines:
curr = lines.pop(0)
name, __, value = curr.partition(b':')
if __ != b':':
raise InvalidHeader(_(u"Invalid header line: %r"), curr.decode('ISO8859-1'))
if self.HEADER_RE.search(name):
raise InvalidHeader(_(u"Invalid header name: %r"), name.decode('ISO8859-1'))
name, value = name.strip(), [value.lstrip()]
# continuation lines
while lines and lines[0].startswith((b' ', b'\t')):
value.append(lines.pop(0)[1:])
value = b''.join(value).rstrip()
Element = HEADER.get(name, HeaderElement)
value = Element.decode(value)
self.append(name, value)
def _unquote_host(self, host):
# IPv6 / IPvFuture
if host.startswith(b'[') and host.endswith(b']'):
host = host[1:-1]
try:
return u'[%s]' % inet_ntop(AF_INET6, inet_pton(AF_INET6, host)).decode('ascii')
except SocketError:
# IPvFuture
if host.startswith(b'v') and b'.' in host and host[1:].split('.', 1)[0].isdigit():
try:
return u'[%s]' % host.decode('ascii')
except UnicodeDecodeError:
raise InvalidURI(_('Invalid IPvFuture address: must be ASCII.'))
raise InvalidURI(_('Invalid IP address in URI.'))
# IPv4
if all(x.isdigit() for x in host.split(b'.')):
try:
return inet_ntop(AF_INET, inet_pton(AF_INET, host)).decode('ascii')
except SocketError:
raise InvalidURI(_('Invalid IPv4 address in URI.'))
if host.strip(Percent.UNRESERVED + Percent.SUB_DELIMS + b'%'):
raise InvalidURI(_('Invalid URI host.'))
# DNS hostname
host = self.unquote(host)
try:
return host.encode('ascii').decode('idna').lower()
except UnicodeError:
raise InvalidURI(_('Invalid host.'))
def parse(self, line):
"""parses the request line and sets method, uri and protocol version
:param line: the request line
:type line: bytes
"""
bits = line.strip().split(None, 2)
try:
method, uri, version = bits
except ValueError:
raise InvalidLine(_(u'Invalid request line: %r'),
line.decode('ISO8859-1'))
# protocol version
super(Request, self).parse(version)
# method
self.method.parse(method)
# URI
if uri.startswith(b'//'):
raise InvalidURI(
_(u'The request URI must be an absolute path or contain a scheme.'
))
if self.method == u'CONNECT':
uri = b'//%s' % (uri, )
self.uri.parse(uri)
self.validate_request_uri()
def prevent_denial_of_service(self): if len([x for x in self.ranges if x[0] is None]) > 1 or len([x for x in self.ranges if x[1] is None]) > 1: raise InvalidHeader(_(u'too many overlapping ranges.')) byterange = set() for start, stop in ((x, y) for x, y in self.ranges if x is not None and y is not None): range_ = set(range(start, stop)) if any(x in byterange for x in range_): raise InvalidHeader(_(u'duplicated range.')) byterange.update(range_)
def prevent_denial_of_service(self):
if len([x for x in self.ranges if x[0] is None]) > 1 or len(
[x for x in self.ranges if x[1] is None]) > 1:
raise InvalidHeader(_(u'too many overlapping ranges.'))
byterange = set()
for start, stop in ((x, y) for x, y in self.ranges
if x is not None and y is not None):
range_ = set(range(start, stop))
if any(x in byterange for x in range_):
raise InvalidHeader(_(u'duplicated range.'))
byterange.update(range_)
def compose(self): try: scheme = self.schemes[self.value.lower()] except KeyError: raise InvalidHeader(_(u'Unsupported authentication scheme: %r'), self.value) try: authinfo = scheme.compose(self.params) except KeyError as key: raise InvalidHeader(_(u'Missing parameter %r for authentication scheme %r'), str(key), self.value) return b'%s %s' % (self.value.title(), authinfo)
def validate_request_uri(self): uri = self.uri if not isinstance(uri, (uri.SCHEMES['http'], uri.SCHEMES['https'])): raise InvalidURI(_(u'The request URI scheme must be HTTP based.')) if uri.fragment or uri.username or uri.password: raise InvalidURI(_(u'The request URI must not contain fragments or user information.')) if uri.path.startswith(b'//'): raise InvalidURI(_(u'The request URI path must not start with //.')) if uri.path and uri.path != u'*' and uri.path[0] != u'/': raise InvalidURI(_(u'The request URI path must start with /.')) if self.method == u'CONNECT' and (uri.scheme or uri.path or uri.query_string or not uri.host): raise InvalidURI(_(u'The request URI of an CONNECT request must be a authority.'))
def sanitize(self): self.value = self.value.lower() if self.attachment: if b'inline' in self.params: raise InvalidHeader(_(u'Mixed Content-Disposition')) elif self.inline: if b'attachment' in self.params: raise InvalidHeader(_(u'Mixed Content-Disposition')) elif self.form_data: if b'form-data' in self.params: raise InvalidHeader(_(u'Mixed Content-Disposition')) else: raise InvalidHeader(_(u'Unknown Content-Disposition: %r'), self.value,)
def sanitize(self): self.value = self.value.lower() if self.attachment: if b'inline' in self.params: raise InvalidHeader(_(u'Mixed Content-Disposition')) elif self.inline: if b'attachment' in self.params: raise InvalidHeader(_(u'Mixed Content-Disposition')) elif self.form_data: if b'form-data' in self.params: raise InvalidHeader(_(u'Mixed Content-Disposition')) else: raise InvalidHeader(_(u'Unknown Content-Disposition: %r'), self.value,)
def _rfc2231_and_continuation_params(cls, params): # TODO: complexity
count = set()
continuations = dict()
for key, value, quoted in params:
if key in count:
raise InvalidHeader(_(u'Parameter given twice: %r'),
key.decode('ISO8859-1'))
count.add(key)
if '*' in key:
if key.endswith('*') and not quoted:
charset, language, value_ = decode_rfc2231(
value.encode('ISO8859-1'))
if not charset:
yield key, value
continue
encoding = sanitize_encoding(charset)
if encoding is None:
raise InvalidHeader(_(u'Unknown encoding: %r'),
charset)
try:
key, value = key[:-1], Percent.unquote(value_).decode(
encoding)
except UnicodeDecodeError as exc:
raise InvalidHeader(_(u'%s') % (exc, ))
key_, asterisk, num = key.rpartition('*')
if asterisk:
try:
if num != '0' and num.startswith('0'):
raise ValueError
num = int(num)
except ValueError:
yield key, value
continue
continuations.setdefault(key_, {})[num] = value
continue
yield key, value
for key, lines in iteritems(continuations):
value = b''
for i in xrange(len(lines)):
try:
value += lines.pop(i)
except KeyError:
break
if not key:
raise InvalidHeader(_(u'...'))
if value:
yield key, value
for k, v in iteritems(lines):
yield '%s*%d' % (key, k), v
def parseparams(cls, elementstr): try: scheme, authinfo = elementstr.split(b' ', 1) except ValueError: raise InvalidHeader(_(u'Authorization headers must contain authentication scheme')) try: parser = cls.schemes[scheme.lower()] except KeyError: raise InvalidHeader(_(u'Unsupported authentication scheme: %r'), scheme) try: authinfo = parser.parse(authinfo) except KeyError as key: raise InvalidHeader(_(u'Missing parameter %r for authentication scheme %r'), str(key), scheme) return scheme.title(), authinfo
def parseparam(cls, atom): key, __, val = atom.partition(b'=') try: val, quoted = cls.unescape_param(val.strip()) except InvalidHeader: raise InvalidHeader(_(u'Unquoted parameter %r in %r containing TSPECIALS: %r'), key, cls.__name__, val) return cls.unescape_key(key), val, quoted
def _rfc2231_and_continuation_params(cls, params): # TODO: complexity
count = set()
continuations = dict()
for key, value, quoted in params:
if key in count:
raise InvalidHeader(_(u'Parameter given twice: %r'), key.decode('ISO8859-1'))
count.add(key)
if '*' in key:
if key.endswith('*') and not quoted:
charset, language, value_ = decode_rfc2231(value.encode('ISO8859-1'))
if not charset:
yield key, value
continue
encoding = sanitize_encoding(charset)
if encoding is None:
raise InvalidHeader(_(u'Unknown encoding: %r'), charset)
try:
key, value = key[:-1], Percent.unquote(value_).decode(encoding)
except UnicodeDecodeError as exc:
raise InvalidHeader(_(u'%s') % (exc,))
key_, asterisk, num = key.rpartition('*')
if asterisk:
try:
if num != '0' and num.startswith('0'):
raise ValueError
num = int(num)
except ValueError:
yield key, value
continue
continuations.setdefault(key_, {})[num] = value
continue
yield key, value
for key, lines in iteritems(continuations):
value = b''
for i in xrange(len(lines)):
try:
value += lines.pop(i)
except KeyError:
break
if not key:
raise InvalidHeader(_(u'...'))
if value:
yield key, value
for k, v in iteritems(lines):
yield '%s*%d' % (key, k), v
def unescape_param(cls, value): quoted = value.startswith(b'"') and value.endswith(b'"') if quoted: value = re.sub(r'\\(?!\\)', '', value.strip(b'"')) else: if cls.RE_TSPECIALS.search(value): raise InvalidHeader(_(u'Unquoted parameter in %r containing TSPECIALS: %r'), cls.__name__, value) return value, quoted
def decode(cls, data, charset=None, mimetype=None): try: with gzip.GzipFile(fileobj=io.BytesIO(data)) as fd: return fd.read() # data = zlib.decompress(data, 16 + zlib.MAX_WBITS) except (zlib.error, IOError, EOFError): raise DecodeError(_(u'Invalid gzip data.')) return Codec.decode(data, charset)
def sanitize(self):
self.value = self.value.lower()
self.host, self.port = self.HOSTPORT.match(self.value).groups()
if self.host.endswith(']') and self.host.startswith('['):
self.host = self.host[1:-1]
if self.port:
self.port = int(self.port)
if not self.hostname:
raise InvalidHeader(_(u'Invalid Host header'))
def encode(cls, data, charset=None, mimetype=None): try: out = io.BytesIO() with gzip.GzipFile(fileobj=out, mode="w") as fd: fd.write(Codec.encode(data, charset)) return out.getvalue() # return zlib.compress(Codec.encode(data, charset), cls.compression_level) except zlib.error: raise EncodeError(_(u'Invalid gzip data.'))
def decode(cls, data, charset=None):
if any(
in_table(x)
for x in cls.unquote(data.encode('ISO8859-1'), 'ISO8859-1')
for in_table in cls.INVALID):
raise DecodeError(
_(u'Invalid query string: contains invalid token'))
data = super(QueryString, cls).decode(data, charset)
return data
def validate_request_uri_scheme(self):
if self.message.uri.scheme:
if self.message.uri.scheme not in ('http', 'https'):
exc = InvalidURI(_(u'Invalid URL: wrong scheme'))
raise BAD_REQUEST(Unicode(exc))
else:
self.message.uri.scheme = self._default_scheme
self.message.uri.host = self._default_host
self.message.uri.port = self._default_port
def sanitize(self):
self.value = self.value.lower()
self.host, self.port = self.HOSTPORT.match(self.value).groups()
if self.host.endswith(']') and self.host.startswith('['):
self.host = self.host[1:-1]
if self.port:
self.port = int(self.port)
if not self.hostname:
raise InvalidHeader(_(u'Invalid Host header'))
def parse(authinfo):
try:
authinfo = authinfo.encode('ascii')
except ValueError:
raise InvalidHeader(_(u'Invalid base64 in basic authentication'))
try:
username, password = decode_base64(authinfo.strip()).split(':')
except Base64Error:
raise InvalidHeader(_(u'Basic authentication contains invalid base64'))
except ValueError:
raise InvalidHeader(_(u'No username:password provided'))
authinfo = {
'username': username.decode('ISO8859-1'),
'password': password.decode('ISO8859-1')
}
return authinfo
def validate_request_uri(self):
uri = self.uri
if not isinstance(uri, (uri.SCHEMES['http'], uri.SCHEMES['https'])):
raise InvalidURI(_(u'The request URI scheme must be HTTP based.'))
if uri.fragment or uri.username or uri.password:
raise InvalidURI(
_(u'The request URI must not contain fragments or user information.'
))
if uri.path.startswith(b'//'):
raise InvalidURI(
_(u'The request URI path must not start with //.'))
if uri.path and uri.path != u'*' and uri.path[0] != u'/':
raise InvalidURI(_(u'The request URI path must start with /.'))
if self.method == u'CONNECT' and (uri.scheme or uri.path
or uri.query_string or not uri.host):
raise InvalidURI(
_(u'The request URI of an CONNECT request must be a authority.'
))
def parseparam(cls, atom):
key, __, val = atom.partition(b'=')
try:
val, quoted = cls.unescape_param(val.strip())
except InvalidHeader:
raise InvalidHeader(
_(u'Unquoted parameter %r in %r containing TSPECIALS: %r'),
key, cls.__name__, val)
return cls.unescape_key(key), val, quoted
def validate_request_uri_scheme(self):
if self.message.uri.scheme:
if self.message.uri.scheme not in ('http', 'https'):
exc = InvalidURI(_(u'Invalid URL: wrong scheme'))
raise BAD_REQUEST(Unicode(exc))
else:
self.message.uri.scheme = self._default_scheme
self.message.uri.host = self._default_host
self.message.uri.port = self._default_port
def port(self, port): port = port or self.PORT if port: try: port = int(port) if not 0 < int(port) <= 65535: raise ValueError except ValueError: raise InvalidURI(_(u'Invalid port: %r'), port) # TODO: TypeError self._port = port
def parse(cls, elementstr): bytesunit, __, byteranges = elementstr.partition(b'=') byteranges = cls.split(byteranges) ranges = set() for brange in byteranges: start, __, stop = (x.strip() for x in brange.partition(b'-')) if (not start and not stop) or not __: raise InvalidHeader(_(u'no range start/stop.')) try: start = int(start) if start else None stop = int(stop) if stop else None if start and start < 0 or stop and stop < 0: raise ValueError except ValueError: raise InvalidHeader(_(u'no range number.')) if start is not None and stop is not None and stop < start: raise InvalidHeader(_(u'range start must be smaller than end.')) ranges.add((start, stop)) return cls(bytesunit, list(sorted(ranges)))
def port(self, port): port = port or self.PORT if port: try: port = int(port) if not 0 < int(port) <= 65535: raise ValueError except ValueError: raise InvalidURI(_(u'Invalid port: %r'), port) # TODO: TypeError self._port = port
def unescape_param(cls, value):
quoted = value.startswith(b'"') and value.endswith(b'"')
if quoted:
value = re.sub(r'\\(?!\\)', '', value.strip(b'"'))
else:
if cls.RE_TSPECIALS.search(value):
raise InvalidHeader(
_(u'Unquoted parameter in %r containing TSPECIALS: %r'),
cls.__name__, value)
return value, quoted
def parse(authinfo):
try:
authinfo = authinfo.encode('ascii')
except ValueError:
raise InvalidHeader(_(u'Invalid base64 in basic authentication'))
try:
username, password = decode_base64(authinfo.strip()).split(':')
except Base64Error:
raise InvalidHeader(
_(u'Basic authentication contains invalid base64'))
except ValueError:
raise InvalidHeader(_(u'No username:password provided'))
authinfo = {
'username': username.decode('ISO8859-1'),
'password': password.decode('ISO8859-1')
}
return authinfo
def parse(self, status):
"""parse a Statuscode and Reason-Phrase
:param status: the code and reason
:type status: bytes
"""
match = self.STATUS_RE.match(status)
if match is None:
raise InvalidLine(_(u"Invalid status %r"), status.decode('ISO8859-1'))
self.set((int(match.group(1)), match.group(2).decode('ascii'),))
def parse(cls, elementstr):
bytesunit, __, byteranges = elementstr.partition(b'=')
byteranges = cls.split(byteranges)
ranges = set()
for brange in byteranges:
start, __, stop = (x.strip() for x in brange.partition(b'-'))
if (not start and not stop) or not __:
raise InvalidHeader(_(u'no range start/stop.'))
try:
start = int(start) if start else None
stop = int(stop) if stop else None
if start and start < 0 or stop and stop < 0:
raise ValueError
except ValueError:
raise InvalidHeader(_(u'no range number.'))
if start is not None and stop is not None and stop < start:
raise InvalidHeader(
_(u'range start must be smaller than end.'))
ranges.add((start, stop))
return cls(bytesunit, list(sorted(ranges)))
def __parse_chunk_size(self):
line, rest_chunk = self.buffer.split(self.line_end, 1)
chunk_size = line.split(b";", 1)[0].strip()
try:
chunk_size = int(bytes(chunk_size), 16)
if chunk_size < 0:
raise ValueError
except (ValueError, OverflowError):
exc = InvalidHeader(_(u'Invalid chunk size: %r'), chunk_size.decode('ISO8859-1'))
raise BAD_REQUEST(Unicode(exc))
else:
return chunk_size, rest_chunk
def __parse_chunk_size(self):
line, rest_chunk = self.buffer.split(self.line_end, 1)
chunk_size = line.split(b";", 1)[0].strip()
try:
chunk_size = int(bytes(chunk_size), 16)
if chunk_size < 0:
raise ValueError
except (ValueError, OverflowError):
exc = InvalidHeader(_(u'Invalid chunk size: %r'), chunk_size.decode('ISO8859-1'))
raise BAD_REQUEST(Unicode(exc))
else:
return chunk_size, rest_chunk
def parse(self, line):
u"""parses the response line"""
bits = line.strip().split(None, 1)
try:
version, status = bits
except ValueError:
raise InvalidLine(_(u'Invalid response line: %r'), line.decode('ISO8859-1'))
# version
super(Response, self).parse(version)
# status
self.status.parse(status)
def parse(cls, elementstr):
value, start, end, complete_length = None, None, None, None
try:
value, content_range = elementstr.split(None, 1)
if value != b'bytes':
raise InvalidHeader(_(u'Only "bytes" Content-Range supported'))
byte_range, complete_length = content_range.split(b'/')
if complete_length != b'*':
complete_length = int(complete_length)
if complete_length < 0:
raise ValueError
else:
complete_length = None
if byte_range != b'*':
start, end = byte_range.split(b'-', 1)
start, end = int(start), int(end)
if start >= end or start < 0 or end < 0:
raise ValueError
if complete_length is None and start is None:
raise ValueError
except ValueError:
raise InvalidHeader(_(u'Content-Range: %r'), elementstr)
return cls(value, (start, end), complete_length)
def parse(cls, elementstr): value, start, end, complete_length = None, None, None, None try: value, content_range = elementstr.split(None, 1) if value != b'bytes': raise InvalidHeader(_(u'Only "bytes" Content-Range supported')) byte_range, complete_length = content_range.split(b'/') if complete_length != b'*': complete_length = int(complete_length) if complete_length < 0: raise ValueError else: complete_length = None if byte_range != b'*': start, end = byte_range.split(b'-', 1) start, end = int(start), int(end) if start >= end or start < 0 or end < 0: raise ValueError if complete_length is None and start is None: raise ValueError except ValueError: raise InvalidHeader(_(u'Content-Range: %r'), elementstr) return cls(value, (start, end), complete_length)
def decode(cls, data, charset=None, mimetype=None):
boundary = mimetype.boundary.encode('ISO8859-1')
parts = data.split(b'--%s' % (boundary, ))
part = parts.pop(0)
if part:
raise DecodeError(_(u'Data before boundary: %r'),
part.decode('ISO8859-1'))
try:
part = parts.pop()
except IndexError:
raise DecodeError(u'No end of boundary')
if part not in (b'--', b'--\r\n'):
raise DecodeError(_(u'Invalid multipart end: %r'),
part.decode('ISO8859-1'))
from httoop.messages.body import Body
multiparts = []
for part in parts:
if not part.startswith('\r\n'):
raise DecodeError(_(u'Invalid boundary end: %r'),
part[:2].decode('ISO8859-1'))
part = part[2:]
headers, separator, content = part.partition(b'\r\n\r\n')
if not separator:
raise DecodeError(
_(u'Multipart does not contain CRLF header separator'))
if not content.endswith('\r\n'):
raise DecodeError(_(u'Multipart does not end with CRLF: %r'),
content[-2].decode('ISO8859-1'))
content = content[:-2]
body = Body()
body.headers.clear()
body.headers.parse(headers)
body.headers.setdefault('Content-Type', cls.default_content_type)
body.parse(content)
multiparts.append(body)
return multiparts
def parse(self, line):
"""parses the request line and sets method, uri and protocol version
:param line: the request line
:type line: bytes
"""
bits = line.strip().split(None, 2)
try:
method, uri, version = bits
except ValueError:
raise InvalidLine(_(u'Invalid request line: %r'), line.decode('ISO8859-1'))
# protocol version
super(Request, self).parse(version)
# method
self.method.parse(method)
# URI
if uri.startswith(b'//'):
raise InvalidURI(_(u'The request URI must be an absolute path or contain a scheme.'))
if self.method == u'CONNECT':
uri = b'//%s' % (uri,)
self.uri.parse(uri)
self.validate_request_uri()
def parse(self, line):
u"""parses the response line"""
bits = line.strip().split(None, 1)
try:
version, status = bits
except ValueError:
raise InvalidLine(_(u'Invalid response line: %r'),
line.decode('ISO8859-1'))
# version
super(Response, self).parse(version)
# status
self.status.parse(status)
def parse(self, status):
"""parse a Statuscode and Reason-Phrase
:param status: the code and reason
:type status: bytes
"""
match = self.STATUS_RE.match(status)
if match is None:
raise InvalidLine(_(u"Invalid status %r"),
status.decode('ISO8859-1'))
self.set((
int(match.group(1)),
match.group(2).decode('ascii'),
))
def _compose_authority_iter(self):
if not self.host:
return
username, password, host, port, quote = self.username, self.password, self.host, self.port, self.quote
if username:
yield quote(username, Percent.USERINFO)
if password:
yield b':'
yield quote(password, Percent.USERINFO)
yield b'@'
try:
yield host.encode('idna')
except UnicodeError: # u'..'.encode('idna')
raise InvalidURI(_(u'Invalid URI: cannot encode host as IDNA.'))
if port and int(port) != self.PORT:
yield b':%d' % int(port)
def _compose_authority_iter(self):
if not self.host:
return
username, password, host, port, quote = self.username, self.password, self.host, self.port, self.quote
if username:
yield quote(username, Percent.USERINFO)
if password:
yield b':'
yield quote(password, Percent.USERINFO)
yield b'@'
try:
yield host.encode('idna')
except UnicodeError: # u'..'.encode('idna')
raise InvalidURI(_(u'Invalid URI: cannot encode host as IDNA.'))
if port and int(port) != self.PORT:
yield b':%d' % int(port)
def parse(cls, authinfo):
params = super(cls, cls).parse(authinfo)
if '"' in params['nonce']:
raise InvalidHeader(_(u'Nonce must not contain double quote'))
stale = params.get('stale')
if stale:
stale = {'false': False, 'true': True}.get(stale.lower())
params = [
('realm', params['realm']),
('domain', params.get('domain', '').split()),
('nonce', params['nonce']),
('opaque', params.get('opaque')),
('stale', stale),
('algorithm', params.get('algorithm')),
('qop', [p.strip() for p in params.get('qop', '').split(',')]),
]
return dict([(k, v) for k, v in params if v is not None])
def parse(cls, authinfo):
params = super(cls, cls).parse(authinfo)
if '"' in params['nonce']:
raise InvalidHeader(_(u'Nonce must not contain double quote'))
stale = params.get('stale')
if stale:
stale = {'false': False, 'true': True}.get(stale.lower())
params = [
('realm', params['realm']),
('domain', params.get('domain', '').split()),
('nonce', params['nonce']),
('opaque', params.get('opaque')),
('stale', stale),
('algorithm', params.get('algorithm')),
('qop', [p.strip() for p in params.get('qop', '').split(',')]),
]
return dict([(k, v) for k, v in params if v is not None])
def determine_message_length(self):
# RFC 2616 Section 4.4
# get message length
# TODO: check if both is set
message = self.message
if 'Transfer-Encoding' in message.headers and message.protocol >= (1, 1):
# chunked transfer in HTTP/1.1
te = message.headers['Transfer-Encoding'].lower()
self.chunked = 'chunked' == te
if not self.chunked:
raise NOT_IMPLEMENTED(u'Unknown HTTP/1.1 Transfer-Encoding: %r' % te)
else:
# Content-Length header defines the length of the message body
try:
self.message_length = int(message.headers.get("Content-Length", "0"))
if self.message_length < 0:
self.message_length = None
raise ValueError
except ValueError:
raise BAD_REQUEST(_(u'Invalid Content-Length header.'))
def determine_message_length(self):
# RFC 2616 Section 4.4
# get message length
# TODO: check if both is set
message = self.message
if 'Transfer-Encoding' in message.headers and message.protocol >= (1, 1):
# chunked transfer in HTTP/1.1
te = message.headers['Transfer-Encoding'].lower()
self.chunked = 'chunked' == te
if not self.chunked:
raise NOT_IMPLEMENTED(u'Unknown HTTP/1.1 Transfer-Encoding: %r' % te)
else:
# Content-Length header defines the length of the message body
try:
self.message_length = int(message.headers.get("Content-Length", "0"))
if self.message_length < 0:
self.message_length = None
raise ValueError
except ValueError:
raise BAD_REQUEST(_(u'Invalid Content-Length header.'))
def parse_trailers(self): # TODO: the code is exactly the same as parse_headers but # we have to make sure no invalid header fields are send (only values told in Trailer header allowed) if self.buffer.startswith(self.line_end): self.buffer = self.buffer[len(self.line_end):] return False # no trailers trailer_end = self.line_end + self.line_end if trailer_end not in self.buffer: # not received yet return NOT_RECEIVED_YET trailers, self.buffer = self.buffer.split(trailer_end, 1) self.trailers = Headers() try: self.trailers.parse(bytes(trailers)) except InvalidHeader as exc: exc = InvalidHeader(_(u'Invalid trailers: %r'), Unicode(exc)) raise BAD_REQUEST(Unicode(exc)) self.merge_trailer_into_header() return False
def parse_trailers(self): # TODO: the code is exactly the same as parse_headers but # we have to make sure no invalid header fields are send (only values told in Trailer header allowed) if self.buffer.startswith(self.line_end): self.buffer = self.buffer[len(self.line_end):] return False # no trailers trailer_end = self.line_end + self.line_end if trailer_end not in self.buffer: # not received yet return NOT_RECEIVED_YET trailers, self.buffer = self.buffer.split(trailer_end, 1) self.trailers = Headers() try: self.trailers.parse(bytes(trailers)) except InvalidHeader as exc: exc = InvalidHeader(_(u'Invalid trailers: %r'), Unicode(exc)) raise BAD_REQUEST(Unicode(exc)) self.merge_trailer_into_header() return False
def parse(cls, timestr=None):
u"""parses a HTTP date string and returns a :class:`Date` object
:param timestr: the time string in one of the http formats
:type timestr: str
:returns: the HTTP Date object
:rtype : :class:`Date`
"""
# parse the most common HTTP Date format (RFC 2822)
date = parsedate(timestr)
if date is not None:
return cls(date[:9])
old = locale.getlocale(locale.LC_TIME)
locale.setlocale(locale.LC_TIME, (None, None))
try:
# parse RFC 1036 date format
try:
date = time.strptime(timestr, '%A, %d-%b-%y %H:%M:%S GMT')
except ValueError:
pass
else:
return cls(date)
# parse C's asctime format
try:
date = time.strptime(timestr, '%a %b %d %H:%M:%S %Y')
except ValueError:
pass
else:
return cls(date)
finally:
locale.setlocale(locale.LC_TIME, old)
raise InvalidDate(_(u'Invalid date: %r'), date)
def parse(cls, timestr=None): u"""parses a HTTP date string and returns a :class:`Date` object :param timestr: the time string in one of the http formats :type timestr: str :returns: the HTTP Date object :rtype : :class:`Date` """ # parse the most common HTTP Date format (RFC 2822) date = parsedate(timestr) if date is not None: return cls(date[:9]) old = locale.getlocale(locale.LC_TIME) locale.setlocale(locale.LC_TIME, (None, None)) try: # parse RFC 1036 date format try: date = time.strptime(timestr, '%A, %d-%b-%y %H:%M:%S GMT') except ValueError: pass else: return cls(date) # parse C's asctime format try: date = time.strptime(timestr, '%a %b %d %H:%M:%S %Y') except ValueError: pass else: return cls(date) finally: locale.setlocale(locale.LC_TIME, old) raise InvalidDate(_(u'Invalid date: %r'), date)
def parse_chunked_body(self): if self.line_end not in self.buffer: # chunk size info not received yet return NOT_RECEIVED_YET chunk_size, rest_chunk = self.__parse_chunk_size() if len(rest_chunk) < (len(self.line_end) + chunk_size): # chunk not received completely return NOT_RECEIVED_YET body_part, rest_chunk = rest_chunk[:chunk_size], rest_chunk[chunk_size:] self.message.body.parse(bytes(body_part)) self.buffer = rest_chunk if chunk_size == 0: return self.parse_trailers() if not rest_chunk.startswith(self.line_end): raise InvalidBody(_(u'Invalid chunk terminator: %r'), repr(rest_chunk[:2])) self.buffer = self.buffer[len(self.line_end):] # next chunk return self.parse_chunked_body()
def parse_chunked_body(self): if self.line_end not in self.buffer: # chunk size info not received yet return NOT_RECEIVED_YET chunk_size, rest_chunk = self.__parse_chunk_size() if len(rest_chunk) < (len(self.line_end) + chunk_size): # chunk not received completely return NOT_RECEIVED_YET body_part, rest_chunk = rest_chunk[:chunk_size], rest_chunk[chunk_size:] self.message.body.parse(bytes(body_part)) self.buffer = rest_chunk if chunk_size == 0: return self.parse_trailers() if not rest_chunk.startswith(self.line_end): raise InvalidBody(_(u'Invalid chunk terminator: %r'), repr(rest_chunk[:2])) self.buffer = self.buffer[len(self.line_end):] # next chunk return self.parse_chunked_body()
def parse(self, protocol):
match = self.PROTOCOL_RE.match(protocol)
if match is None:
raise InvalidLine(_(u"Invalid HTTP protocol: %r"), protocol.decode('ISO8859-1'))
self.__protocol = (int(match.group(2)), int(match.group(3)))
self.name = match.group(1)
def parse(self, protocol):
match = self.PROTOCOL_RE.match(protocol)
if match is None:
raise InvalidLine(_(u"Invalid HTTP protocol: %r"), protocol.decode('ISO8859-1'))
self.__protocol = (int(match.group(2)), int(match.group(3)))
self.name = match.group(1)
def decode(cls, data, charset=None, mimetype=None): try: data = zlib.decompress(data, 16 + zlib.MAX_WBITS) except zlib.error: raise DecodeError(_(u'Invalid gzip data.')) return Codec.decode(data, charset)