diff options
Diffstat (limited to 'netlib/http/headers.py')
| -rw-r--r-- | netlib/http/headers.py | 205 |
1 files changed, 205 insertions, 0 deletions
diff --git a/netlib/http/headers.py b/netlib/http/headers.py new file mode 100644 index 00000000..1511ea2d --- /dev/null +++ b/netlib/http/headers.py @@ -0,0 +1,205 @@ +""" + +Unicode Handling +---------------- +See also: http://lucumr.pocoo.org/2013/7/2/the-updated-guide-to-unicode/ +""" +from __future__ import absolute_import, print_function, division +import copy +try: + from collections.abc import MutableMapping +except ImportError: # Workaround for Python < 3.3 + from collections import MutableMapping + + +import six + +from netlib.utils import always_byte_args + +if six.PY2: + _native = lambda x: x + _asbytes = lambda x: x + _always_byte_args = lambda x: x +else: + # While headers _should_ be ASCII, it's not uncommon for certain headers to be utf-8 encoded. + _native = lambda x: x.decode("utf-8", "surrogateescape") + _asbytes = lambda x: x.encode("utf-8", "surrogateescape") + _always_byte_args = always_byte_args("utf-8", "surrogateescape") + + +class Headers(MutableMapping, object): + """ + Header class which allows both convenient access to individual headers as well as + direct access to the underlying raw data. Provides a full dictionary interface. + + Example: + + .. code-block:: python + + # Create header from a list of (header_name, header_value) tuples + >>> h = Headers([ + ["Host","example.com"], + ["Accept","text/html"], + ["accept","application/xml"] + ]) + + # Headers mostly behave like a normal dict. + >>> h["Host"] + "example.com" + + # HTTP Headers are case insensitive + >>> h["host"] + "example.com" + + # Multiple headers are folded into a single header as per RFC7230 + >>> h["Accept"] + "text/html, application/xml" + + # Setting a header removes all existing headers with the same name. + >>> h["Accept"] = "application/text" + >>> h["Accept"] + "application/text" + + # str(h) returns a HTTP1 header block. + >>> print(h) + Host: example.com + Accept: application/text + + # For full control, the raw header fields can be accessed + >>> h.fields + + # Headers can also be crated from keyword arguments + >>> h = Headers(host="example.com", content_type="application/xml") + + Caveats: + For use with the "Set-Cookie" header, see :py:meth:`get_all`. + """ + + @_always_byte_args + def __init__(self, fields=None, **headers): + """ + Args: + fields: (optional) list of ``(name, value)`` header tuples, + e.g. ``[("Host","example.com")]``. All names and values must be bytes. + **headers: Additional headers to set. Will overwrite existing values from `fields`. + For convenience, underscores in header names will be transformed to dashes - + this behaviour does not extend to other methods. + If ``**headers`` contains multiple keys that have equal ``.lower()`` s, + the behavior is undefined. + """ + self.fields = fields or [] + + for name, value in self.fields: + if not isinstance(name, bytes) or not isinstance(value, bytes): + raise ValueError("Headers passed as fields must be bytes.") + + # content_type -> content-type + headers = { + _asbytes(name).replace(b"_", b"-"): value + for name, value in six.iteritems(headers) + } + self.update(headers) + + def __bytes__(self): + if self.fields: + return b"\r\n".join(b": ".join(field) for field in self.fields) + b"\r\n" + else: + return b"" + + if six.PY2: + __str__ = __bytes__ + + @_always_byte_args + def __getitem__(self, name): + values = self.get_all(name) + if not values: + raise KeyError(name) + return ", ".join(values) + + @_always_byte_args + def __setitem__(self, name, value): + idx = self._index(name) + + # To please the human eye, we insert at the same position the first existing header occured. + if idx is not None: + del self[name] + self.fields.insert(idx, [name, value]) + else: + self.fields.append([name, value]) + + @_always_byte_args + def __delitem__(self, name): + if name not in self: + raise KeyError(name) + name = name.lower() + self.fields = [ + field for field in self.fields + if name != field[0].lower() + ] + + def __iter__(self): + seen = set() + for name, _ in self.fields: + name_lower = name.lower() + if name_lower not in seen: + seen.add(name_lower) + yield _native(name) + + def __len__(self): + return len(set(name.lower() for name, _ in self.fields)) + + # __hash__ = object.__hash__ + + def _index(self, name): + name = name.lower() + for i, field in enumerate(self.fields): + if field[0].lower() == name: + return i + return None + + def __eq__(self, other): + if isinstance(other, Headers): + return self.fields == other.fields + return False + + def __ne__(self, other): + return not self.__eq__(other) + + @_always_byte_args + def get_all(self, name): + """ + Like :py:meth:`get`, but does not fold multiple headers into a single one. + This is useful for Set-Cookie headers, which do not support folding. + + See also: https://tools.ietf.org/html/rfc7230#section-3.2.2 + """ + name_lower = name.lower() + values = [_native(value) for n, value in self.fields if n.lower() == name_lower] + return values + + @_always_byte_args + def set_all(self, name, values): + """ + Explicitly set multiple headers for the given key. + See: :py:meth:`get_all` + """ + values = map(_asbytes, values) # _always_byte_args does not fix lists + if name in self: + del self[name] + self.fields.extend( + [name, value] for value in values + ) + + def copy(self): + return Headers(copy.copy(self.fields)) + + # Implement the StateObject protocol from mitmproxy + def get_state(self, short=False): + return tuple(tuple(field) for field in self.fields) + + def load_state(self, state): + self.fields = [list(field) for field in state] + + @classmethod + def from_state(cls, state): + return cls([list(field) for field in state])
\ No newline at end of file |