lucasvl.nl/lib/python3.11/site-packages/werkzeug/datastructures/structures.py
2024-03-04 21:31:20 -06:00

1007 lines
31 KiB
Python

from __future__ import annotations
from collections.abc import MutableSet
from copy import deepcopy
from .. import exceptions
from .._internal import _missing
from .mixins import ImmutableDictMixin
from .mixins import ImmutableListMixin
from .mixins import ImmutableMultiDictMixin
from .mixins import UpdateDictMixin
def is_immutable(self):
raise TypeError(f"{type(self).__name__!r} objects are immutable")
def iter_multi_items(mapping):
"""Iterates over the items of a mapping yielding keys and values
without dropping any from more complex structures.
"""
if isinstance(mapping, MultiDict):
yield from mapping.items(multi=True)
elif isinstance(mapping, dict):
for key, value in mapping.items():
if isinstance(value, (tuple, list)):
for v in value:
yield key, v
else:
yield key, value
else:
yield from mapping
class ImmutableList(ImmutableListMixin, list):
"""An immutable :class:`list`.
.. versionadded:: 0.5
:private:
"""
def __repr__(self):
return f"{type(self).__name__}({list.__repr__(self)})"
class TypeConversionDict(dict):
"""Works like a regular dict but the :meth:`get` method can perform
type conversions. :class:`MultiDict` and :class:`CombinedMultiDict`
are subclasses of this class and provide the same feature.
.. versionadded:: 0.5
"""
def get(self, key, default=None, type=None):
"""Return the default value if the requested data doesn't exist.
If `type` is provided and is a callable it should convert the value,
return it or raise a :exc:`ValueError` if that is not possible. In
this case the function will return the default as if the value was not
found:
>>> d = TypeConversionDict(foo='42', bar='blub')
>>> d.get('foo', type=int)
42
>>> d.get('bar', -1, type=int)
-1
:param key: The key to be looked up.
:param default: The default value to be returned if the key can't
be looked up. If not further specified `None` is
returned.
:param type: A callable that is used to cast the value in the
:class:`MultiDict`. If a :exc:`ValueError` is raised
by this callable the default value is returned.
"""
try:
rv = self[key]
except KeyError:
return default
if type is not None:
try:
rv = type(rv)
except ValueError:
rv = default
return rv
class ImmutableTypeConversionDict(ImmutableDictMixin, TypeConversionDict):
"""Works like a :class:`TypeConversionDict` but does not support
modifications.
.. versionadded:: 0.5
"""
def copy(self):
"""Return a shallow mutable copy of this object. Keep in mind that
the standard library's :func:`copy` function is a no-op for this class
like for any other python immutable type (eg: :class:`tuple`).
"""
return TypeConversionDict(self)
def __copy__(self):
return self
class MultiDict(TypeConversionDict):
"""A :class:`MultiDict` is a dictionary subclass customized to deal with
multiple values for the same key which is for example used by the parsing
functions in the wrappers. This is necessary because some HTML form
elements pass multiple values for the same key.
:class:`MultiDict` implements all standard dictionary methods.
Internally, it saves all values for a key as a list, but the standard dict
access methods will only return the first value for a key. If you want to
gain access to the other values, too, you have to use the `list` methods as
explained below.
Basic Usage:
>>> d = MultiDict([('a', 'b'), ('a', 'c')])
>>> d
MultiDict([('a', 'b'), ('a', 'c')])
>>> d['a']
'b'
>>> d.getlist('a')
['b', 'c']
>>> 'a' in d
True
It behaves like a normal dict thus all dict functions will only return the
first value when multiple values for one key are found.
From Werkzeug 0.3 onwards, the `KeyError` raised by this class is also a
subclass of the :exc:`~exceptions.BadRequest` HTTP exception and will
render a page for a ``400 BAD REQUEST`` if caught in a catch-all for HTTP
exceptions.
A :class:`MultiDict` can be constructed from an iterable of
``(key, value)`` tuples, a dict, a :class:`MultiDict` or from Werkzeug 0.2
onwards some keyword parameters.
:param mapping: the initial value for the :class:`MultiDict`. Either a
regular dict, an iterable of ``(key, value)`` tuples
or `None`.
"""
def __init__(self, mapping=None):
if isinstance(mapping, MultiDict):
dict.__init__(self, ((k, l[:]) for k, l in mapping.lists()))
elif isinstance(mapping, dict):
tmp = {}
for key, value in mapping.items():
if isinstance(value, (tuple, list)):
if len(value) == 0:
continue
value = list(value)
else:
value = [value]
tmp[key] = value
dict.__init__(self, tmp)
else:
tmp = {}
for key, value in mapping or ():
tmp.setdefault(key, []).append(value)
dict.__init__(self, tmp)
def __getstate__(self):
return dict(self.lists())
def __setstate__(self, value):
dict.clear(self)
dict.update(self, value)
def __iter__(self):
# Work around https://bugs.python.org/issue43246.
# (`return super().__iter__()` also works here, which makes this look
# even more like it should be a no-op, yet it isn't.)
return dict.__iter__(self)
def __getitem__(self, key):
"""Return the first data value for this key;
raises KeyError if not found.
:param key: The key to be looked up.
:raise KeyError: if the key does not exist.
"""
if key in self:
lst = dict.__getitem__(self, key)
if len(lst) > 0:
return lst[0]
raise exceptions.BadRequestKeyError(key)
def __setitem__(self, key, value):
"""Like :meth:`add` but removes an existing key first.
:param key: the key for the value.
:param value: the value to set.
"""
dict.__setitem__(self, key, [value])
def add(self, key, value):
"""Adds a new value for the key.
.. versionadded:: 0.6
:param key: the key for the value.
:param value: the value to add.
"""
dict.setdefault(self, key, []).append(value)
def getlist(self, key, type=None):
"""Return the list of items for a given key. If that key is not in the
`MultiDict`, the return value will be an empty list. Just like `get`,
`getlist` accepts a `type` parameter. All items will be converted
with the callable defined there.
:param key: The key to be looked up.
:param type: A callable that is used to cast the value in the
:class:`MultiDict`. If a :exc:`ValueError` is raised
by this callable the value will be removed from the list.
:return: a :class:`list` of all the values for the key.
"""
try:
rv = dict.__getitem__(self, key)
except KeyError:
return []
if type is None:
return list(rv)
result = []
for item in rv:
try:
result.append(type(item))
except ValueError:
pass
return result
def setlist(self, key, new_list):
"""Remove the old values for a key and add new ones. Note that the list
you pass the values in will be shallow-copied before it is inserted in
the dictionary.
>>> d = MultiDict()
>>> d.setlist('foo', ['1', '2'])
>>> d['foo']
'1'
>>> d.getlist('foo')
['1', '2']
:param key: The key for which the values are set.
:param new_list: An iterable with the new values for the key. Old values
are removed first.
"""
dict.__setitem__(self, key, list(new_list))
def setdefault(self, key, default=None):
"""Returns the value for the key if it is in the dict, otherwise it
returns `default` and sets that value for `key`.
:param key: The key to be looked up.
:param default: The default value to be returned if the key is not
in the dict. If not further specified it's `None`.
"""
if key not in self:
self[key] = default
else:
default = self[key]
return default
def setlistdefault(self, key, default_list=None):
"""Like `setdefault` but sets multiple values. The list returned
is not a copy, but the list that is actually used internally. This
means that you can put new values into the dict by appending items
to the list:
>>> d = MultiDict({"foo": 1})
>>> d.setlistdefault("foo").extend([2, 3])
>>> d.getlist("foo")
[1, 2, 3]
:param key: The key to be looked up.
:param default_list: An iterable of default values. It is either copied
(in case it was a list) or converted into a list
before returned.
:return: a :class:`list`
"""
if key not in self:
default_list = list(default_list or ())
dict.__setitem__(self, key, default_list)
else:
default_list = dict.__getitem__(self, key)
return default_list
def items(self, multi=False):
"""Return an iterator of ``(key, value)`` pairs.
:param multi: If set to `True` the iterator returned will have a pair
for each value of each key. Otherwise it will only
contain pairs for the first value of each key.
"""
for key, values in dict.items(self):
if multi:
for value in values:
yield key, value
else:
yield key, values[0]
def lists(self):
"""Return a iterator of ``(key, values)`` pairs, where values is the list
of all values associated with the key."""
for key, values in dict.items(self):
yield key, list(values)
def values(self):
"""Returns an iterator of the first value on every key's value list."""
for values in dict.values(self):
yield values[0]
def listvalues(self):
"""Return an iterator of all values associated with a key. Zipping
:meth:`keys` and this is the same as calling :meth:`lists`:
>>> d = MultiDict({"foo": [1, 2, 3]})
>>> zip(d.keys(), d.listvalues()) == d.lists()
True
"""
return dict.values(self)
def copy(self):
"""Return a shallow copy of this object."""
return self.__class__(self)
def deepcopy(self, memo=None):
"""Return a deep copy of this object."""
return self.__class__(deepcopy(self.to_dict(flat=False), memo))
def to_dict(self, flat=True):
"""Return the contents as regular dict. If `flat` is `True` the
returned dict will only have the first item present, if `flat` is
`False` all values will be returned as lists.
:param flat: If set to `False` the dict returned will have lists
with all the values in it. Otherwise it will only
contain the first value for each key.
:return: a :class:`dict`
"""
if flat:
return dict(self.items())
return dict(self.lists())
def update(self, mapping):
"""update() extends rather than replaces existing key lists:
>>> a = MultiDict({'x': 1})
>>> b = MultiDict({'x': 2, 'y': 3})
>>> a.update(b)
>>> a
MultiDict([('y', 3), ('x', 1), ('x', 2)])
If the value list for a key in ``other_dict`` is empty, no new values
will be added to the dict and the key will not be created:
>>> x = {'empty_list': []}
>>> y = MultiDict()
>>> y.update(x)
>>> y
MultiDict([])
"""
for key, value in iter_multi_items(mapping):
MultiDict.add(self, key, value)
def pop(self, key, default=_missing):
"""Pop the first item for a list on the dict. Afterwards the
key is removed from the dict, so additional values are discarded:
>>> d = MultiDict({"foo": [1, 2, 3]})
>>> d.pop("foo")
1
>>> "foo" in d
False
:param key: the key to pop.
:param default: if provided the value to return if the key was
not in the dictionary.
"""
try:
lst = dict.pop(self, key)
if len(lst) == 0:
raise exceptions.BadRequestKeyError(key)
return lst[0]
except KeyError:
if default is not _missing:
return default
raise exceptions.BadRequestKeyError(key) from None
def popitem(self):
"""Pop an item from the dict."""
try:
item = dict.popitem(self)
if len(item[1]) == 0:
raise exceptions.BadRequestKeyError(item[0])
return (item[0], item[1][0])
except KeyError as e:
raise exceptions.BadRequestKeyError(e.args[0]) from None
def poplist(self, key):
"""Pop the list for a key from the dict. If the key is not in the dict
an empty list is returned.
.. versionchanged:: 0.5
If the key does no longer exist a list is returned instead of
raising an error.
"""
return dict.pop(self, key, [])
def popitemlist(self):
"""Pop a ``(key, list)`` tuple from the dict."""
try:
return dict.popitem(self)
except KeyError as e:
raise exceptions.BadRequestKeyError(e.args[0]) from None
def __copy__(self):
return self.copy()
def __deepcopy__(self, memo):
return self.deepcopy(memo=memo)
def __repr__(self):
return f"{type(self).__name__}({list(self.items(multi=True))!r})"
class _omd_bucket:
"""Wraps values in the :class:`OrderedMultiDict`. This makes it
possible to keep an order over multiple different keys. It requires
a lot of extra memory and slows down access a lot, but makes it
possible to access elements in O(1) and iterate in O(n).
"""
__slots__ = ("prev", "key", "value", "next")
def __init__(self, omd, key, value):
self.prev = omd._last_bucket
self.key = key
self.value = value
self.next = None
if omd._first_bucket is None:
omd._first_bucket = self
if omd._last_bucket is not None:
omd._last_bucket.next = self
omd._last_bucket = self
def unlink(self, omd):
if self.prev:
self.prev.next = self.next
if self.next:
self.next.prev = self.prev
if omd._first_bucket is self:
omd._first_bucket = self.next
if omd._last_bucket is self:
omd._last_bucket = self.prev
class OrderedMultiDict(MultiDict):
"""Works like a regular :class:`MultiDict` but preserves the
order of the fields. To convert the ordered multi dict into a
list you can use the :meth:`items` method and pass it ``multi=True``.
In general an :class:`OrderedMultiDict` is an order of magnitude
slower than a :class:`MultiDict`.
.. admonition:: note
Due to a limitation in Python you cannot convert an ordered
multi dict into a regular dict by using ``dict(multidict)``.
Instead you have to use the :meth:`to_dict` method, otherwise
the internal bucket objects are exposed.
"""
def __init__(self, mapping=None):
dict.__init__(self)
self._first_bucket = self._last_bucket = None
if mapping is not None:
OrderedMultiDict.update(self, mapping)
def __eq__(self, other):
if not isinstance(other, MultiDict):
return NotImplemented
if isinstance(other, OrderedMultiDict):
iter1 = iter(self.items(multi=True))
iter2 = iter(other.items(multi=True))
try:
for k1, v1 in iter1:
k2, v2 = next(iter2)
if k1 != k2 or v1 != v2:
return False
except StopIteration:
return False
try:
next(iter2)
except StopIteration:
return True
return False
if len(self) != len(other):
return False
for key, values in self.lists():
if other.getlist(key) != values:
return False
return True
__hash__ = None
def __reduce_ex__(self, protocol):
return type(self), (list(self.items(multi=True)),)
def __getstate__(self):
return list(self.items(multi=True))
def __setstate__(self, values):
dict.clear(self)
for key, value in values:
self.add(key, value)
def __getitem__(self, key):
if key in self:
return dict.__getitem__(self, key)[0].value
raise exceptions.BadRequestKeyError(key)
def __setitem__(self, key, value):
self.poplist(key)
self.add(key, value)
def __delitem__(self, key):
self.pop(key)
def keys(self):
return (key for key, value in self.items())
def __iter__(self):
return iter(self.keys())
def values(self):
return (value for key, value in self.items())
def items(self, multi=False):
ptr = self._first_bucket
if multi:
while ptr is not None:
yield ptr.key, ptr.value
ptr = ptr.next
else:
returned_keys = set()
while ptr is not None:
if ptr.key not in returned_keys:
returned_keys.add(ptr.key)
yield ptr.key, ptr.value
ptr = ptr.next
def lists(self):
returned_keys = set()
ptr = self._first_bucket
while ptr is not None:
if ptr.key not in returned_keys:
yield ptr.key, self.getlist(ptr.key)
returned_keys.add(ptr.key)
ptr = ptr.next
def listvalues(self):
for _key, values in self.lists():
yield values
def add(self, key, value):
dict.setdefault(self, key, []).append(_omd_bucket(self, key, value))
def getlist(self, key, type=None):
try:
rv = dict.__getitem__(self, key)
except KeyError:
return []
if type is None:
return [x.value for x in rv]
result = []
for item in rv:
try:
result.append(type(item.value))
except ValueError:
pass
return result
def setlist(self, key, new_list):
self.poplist(key)
for value in new_list:
self.add(key, value)
def setlistdefault(self, key, default_list=None):
raise TypeError("setlistdefault is unsupported for ordered multi dicts")
def update(self, mapping):
for key, value in iter_multi_items(mapping):
OrderedMultiDict.add(self, key, value)
def poplist(self, key):
buckets = dict.pop(self, key, ())
for bucket in buckets:
bucket.unlink(self)
return [x.value for x in buckets]
def pop(self, key, default=_missing):
try:
buckets = dict.pop(self, key)
except KeyError:
if default is not _missing:
return default
raise exceptions.BadRequestKeyError(key) from None
for bucket in buckets:
bucket.unlink(self)
return buckets[0].value
def popitem(self):
try:
key, buckets = dict.popitem(self)
except KeyError as e:
raise exceptions.BadRequestKeyError(e.args[0]) from None
for bucket in buckets:
bucket.unlink(self)
return key, buckets[0].value
def popitemlist(self):
try:
key, buckets = dict.popitem(self)
except KeyError as e:
raise exceptions.BadRequestKeyError(e.args[0]) from None
for bucket in buckets:
bucket.unlink(self)
return key, [x.value for x in buckets]
class CombinedMultiDict(ImmutableMultiDictMixin, MultiDict):
"""A read only :class:`MultiDict` that you can pass multiple :class:`MultiDict`
instances as sequence and it will combine the return values of all wrapped
dicts:
>>> from werkzeug.datastructures import CombinedMultiDict, MultiDict
>>> post = MultiDict([('foo', 'bar')])
>>> get = MultiDict([('blub', 'blah')])
>>> combined = CombinedMultiDict([get, post])
>>> combined['foo']
'bar'
>>> combined['blub']
'blah'
This works for all read operations and will raise a `TypeError` for
methods that usually change data which isn't possible.
From Werkzeug 0.3 onwards, the `KeyError` raised by this class is also a
subclass of the :exc:`~exceptions.BadRequest` HTTP exception and will
render a page for a ``400 BAD REQUEST`` if caught in a catch-all for HTTP
exceptions.
"""
def __reduce_ex__(self, protocol):
return type(self), (self.dicts,)
def __init__(self, dicts=None):
self.dicts = list(dicts) or []
@classmethod
def fromkeys(cls, keys, value=None):
raise TypeError(f"cannot create {cls.__name__!r} instances by fromkeys")
def __getitem__(self, key):
for d in self.dicts:
if key in d:
return d[key]
raise exceptions.BadRequestKeyError(key)
def get(self, key, default=None, type=None):
for d in self.dicts:
if key in d:
if type is not None:
try:
return type(d[key])
except ValueError:
continue
return d[key]
return default
def getlist(self, key, type=None):
rv = []
for d in self.dicts:
rv.extend(d.getlist(key, type))
return rv
def _keys_impl(self):
"""This function exists so __len__ can be implemented more efficiently,
saving one list creation from an iterator.
"""
rv = set()
rv.update(*self.dicts)
return rv
def keys(self):
return self._keys_impl()
def __iter__(self):
return iter(self.keys())
def items(self, multi=False):
found = set()
for d in self.dicts:
for key, value in d.items(multi):
if multi:
yield key, value
elif key not in found:
found.add(key)
yield key, value
def values(self):
for _key, value in self.items():
yield value
def lists(self):
rv = {}
for d in self.dicts:
for key, values in d.lists():
rv.setdefault(key, []).extend(values)
return list(rv.items())
def listvalues(self):
return (x[1] for x in self.lists())
def copy(self):
"""Return a shallow mutable copy of this object.
This returns a :class:`MultiDict` representing the data at the
time of copying. The copy will no longer reflect changes to the
wrapped dicts.
.. versionchanged:: 0.15
Return a mutable :class:`MultiDict`.
"""
return MultiDict(self)
def to_dict(self, flat=True):
"""Return the contents as regular dict. If `flat` is `True` the
returned dict will only have the first item present, if `flat` is
`False` all values will be returned as lists.
:param flat: If set to `False` the dict returned will have lists
with all the values in it. Otherwise it will only
contain the first item for each key.
:return: a :class:`dict`
"""
if flat:
return dict(self.items())
return dict(self.lists())
def __len__(self):
return len(self._keys_impl())
def __contains__(self, key):
for d in self.dicts:
if key in d:
return True
return False
def __repr__(self):
return f"{type(self).__name__}({self.dicts!r})"
class ImmutableDict(ImmutableDictMixin, dict):
"""An immutable :class:`dict`.
.. versionadded:: 0.5
"""
def __repr__(self):
return f"{type(self).__name__}({dict.__repr__(self)})"
def copy(self):
"""Return a shallow mutable copy of this object. Keep in mind that
the standard library's :func:`copy` function is a no-op for this class
like for any other python immutable type (eg: :class:`tuple`).
"""
return dict(self)
def __copy__(self):
return self
class ImmutableMultiDict(ImmutableMultiDictMixin, MultiDict):
"""An immutable :class:`MultiDict`.
.. versionadded:: 0.5
"""
def copy(self):
"""Return a shallow mutable copy of this object. Keep in mind that
the standard library's :func:`copy` function is a no-op for this class
like for any other python immutable type (eg: :class:`tuple`).
"""
return MultiDict(self)
def __copy__(self):
return self
class ImmutableOrderedMultiDict(ImmutableMultiDictMixin, OrderedMultiDict):
"""An immutable :class:`OrderedMultiDict`.
.. versionadded:: 0.6
"""
def _iter_hashitems(self):
return enumerate(self.items(multi=True))
def copy(self):
"""Return a shallow mutable copy of this object. Keep in mind that
the standard library's :func:`copy` function is a no-op for this class
like for any other python immutable type (eg: :class:`tuple`).
"""
return OrderedMultiDict(self)
def __copy__(self):
return self
class CallbackDict(UpdateDictMixin, dict):
"""A dict that calls a function passed every time something is changed.
The function is passed the dict instance.
"""
def __init__(self, initial=None, on_update=None):
dict.__init__(self, initial or ())
self.on_update = on_update
def __repr__(self):
return f"<{type(self).__name__} {dict.__repr__(self)}>"
class HeaderSet(MutableSet):
"""Similar to the :class:`ETags` class this implements a set-like structure.
Unlike :class:`ETags` this is case insensitive and used for vary, allow, and
content-language headers.
If not constructed using the :func:`parse_set_header` function the
instantiation works like this:
>>> hs = HeaderSet(['foo', 'bar', 'baz'])
>>> hs
HeaderSet(['foo', 'bar', 'baz'])
"""
def __init__(self, headers=None, on_update=None):
self._headers = list(headers or ())
self._set = {x.lower() for x in self._headers}
self.on_update = on_update
def add(self, header):
"""Add a new header to the set."""
self.update((header,))
def remove(self, header):
"""Remove a header from the set. This raises an :exc:`KeyError` if the
header is not in the set.
.. versionchanged:: 0.5
In older versions a :exc:`IndexError` was raised instead of a
:exc:`KeyError` if the object was missing.
:param header: the header to be removed.
"""
key = header.lower()
if key not in self._set:
raise KeyError(header)
self._set.remove(key)
for idx, key in enumerate(self._headers):
if key.lower() == header:
del self._headers[idx]
break
if self.on_update is not None:
self.on_update(self)
def update(self, iterable):
"""Add all the headers from the iterable to the set.
:param iterable: updates the set with the items from the iterable.
"""
inserted_any = False
for header in iterable:
key = header.lower()
if key not in self._set:
self._headers.append(header)
self._set.add(key)
inserted_any = True
if inserted_any and self.on_update is not None:
self.on_update(self)
def discard(self, header):
"""Like :meth:`remove` but ignores errors.
:param header: the header to be discarded.
"""
try:
self.remove(header)
except KeyError:
pass
def find(self, header):
"""Return the index of the header in the set or return -1 if not found.
:param header: the header to be looked up.
"""
header = header.lower()
for idx, item in enumerate(self._headers):
if item.lower() == header:
return idx
return -1
def index(self, header):
"""Return the index of the header in the set or raise an
:exc:`IndexError`.
:param header: the header to be looked up.
"""
rv = self.find(header)
if rv < 0:
raise IndexError(header)
return rv
def clear(self):
"""Clear the set."""
self._set.clear()
del self._headers[:]
if self.on_update is not None:
self.on_update(self)
def as_set(self, preserve_casing=False):
"""Return the set as real python set type. When calling this, all
the items are converted to lowercase and the ordering is lost.
:param preserve_casing: if set to `True` the items in the set returned
will have the original case like in the
:class:`HeaderSet`, otherwise they will
be lowercase.
"""
if preserve_casing:
return set(self._headers)
return set(self._set)
def to_header(self):
"""Convert the header set into an HTTP header string."""
return ", ".join(map(http.quote_header_value, self._headers))
def __getitem__(self, idx):
return self._headers[idx]
def __delitem__(self, idx):
rv = self._headers.pop(idx)
self._set.remove(rv.lower())
if self.on_update is not None:
self.on_update(self)
def __setitem__(self, idx, value):
old = self._headers[idx]
self._set.remove(old.lower())
self._headers[idx] = value
self._set.add(value.lower())
if self.on_update is not None:
self.on_update(self)
def __contains__(self, header):
return header.lower() in self._set
def __len__(self):
return len(self._set)
def __iter__(self):
return iter(self._headers)
def __bool__(self):
return bool(self._set)
def __str__(self):
return self.to_header()
def __repr__(self):
return f"{type(self).__name__}({self._headers!r})"
# circular dependencies
from .. import http