1
0
mirror of https://github.com/django/django.git synced 2025-03-29 02:30:48 +00:00
django/django/template/context.py
Aymeric Augustin f2c104ada6 Split DTL context creation into its own function.
This reduces the length of rope RequestContext gives users to hang
themselves with.

Thanks Alex Hill for the report and Tim Graham for the review.
2015-02-10 14:34:05 +01:00

253 lines
8.3 KiB
Python

import warnings
from copy import copy
from django.utils.deprecation import RemovedInDjango20Warning
# Hard-coded processor for easier use of CSRF protection.
_builtin_context_processors = ('django.template.context_processors.csrf',)
_current_app_undefined = object()
class ContextPopException(Exception):
"pop() has been called more times than push()"
pass
class ContextDict(dict):
def __init__(self, context, *args, **kwargs):
super(ContextDict, self).__init__(*args, **kwargs)
context.dicts.append(self)
self.context = context
def __enter__(self):
return self
def __exit__(self, *args, **kwargs):
self.context.pop()
class BaseContext(object):
def __init__(self, dict_=None):
self._reset_dicts(dict_)
def _reset_dicts(self, value=None):
builtins = {'True': True, 'False': False, 'None': None}
self.dicts = [builtins]
if value is not None:
self.dicts.append(value)
def __copy__(self):
duplicate = copy(super(BaseContext, self))
duplicate.dicts = self.dicts[:]
return duplicate
def __repr__(self):
return repr(self.dicts)
def __iter__(self):
for d in reversed(self.dicts):
yield d
def push(self, *args, **kwargs):
return ContextDict(self, *args, **kwargs)
def pop(self):
if len(self.dicts) == 1:
raise ContextPopException
return self.dicts.pop()
def __setitem__(self, key, value):
"Set a variable in the current context"
self.dicts[-1][key] = value
def __getitem__(self, key):
"Get a variable's value, starting at the current context and going upward"
for d in reversed(self.dicts):
if key in d:
return d[key]
raise KeyError(key)
def __delitem__(self, key):
"Delete a variable from the current context"
del self.dicts[-1][key]
def has_key(self, key):
for d in self.dicts:
if key in d:
return True
return False
def __contains__(self, key):
return self.has_key(key)
def get(self, key, otherwise=None):
for d in reversed(self.dicts):
if key in d:
return d[key]
return otherwise
def new(self, values=None):
"""
Returns a new context with the same properties, but with only the
values given in 'values' stored.
"""
new_context = copy(self)
new_context._reset_dicts(values)
return new_context
def flatten(self):
"""
Returns self.dicts as one dictionary
"""
flat = {}
for d in self.dicts:
flat.update(d)
return flat
def __eq__(self, other):
"""
Compares two contexts by comparing theirs 'dicts' attributes.
"""
if isinstance(other, BaseContext):
# because dictionaries can be put in different order
# we have to flatten them like in templates
return self.flatten() == other.flatten()
# if it's not comparable return false
return False
class Context(BaseContext):
"A stack container for variable context"
def __init__(self, dict_=None, autoescape=True,
current_app=_current_app_undefined,
use_l10n=None, use_tz=None, engine=None):
if current_app is not _current_app_undefined:
warnings.warn(
"The current_app argument of Context is deprecated. Use "
"RequestContext and set the current_app attribute of its "
"request instead.", RemovedInDjango20Warning, stacklevel=2)
self.autoescape = autoescape
self._current_app = current_app
self.use_l10n = use_l10n
self.use_tz = use_tz
self.engine = engine
self.render_context = RenderContext()
super(Context, self).__init__(dict_)
@property
def current_app(self):
return None if self._current_app is _current_app_undefined else self._current_app
def __copy__(self):
duplicate = super(Context, self).__copy__()
duplicate.render_context = copy(self.render_context)
return duplicate
def update(self, other_dict):
"Pushes other_dict to the stack of dictionaries in the Context"
if not hasattr(other_dict, '__getitem__'):
raise TypeError('other_dict must be a mapping (dictionary-like) object.')
self.dicts.append(other_dict)
return other_dict
class RenderContext(BaseContext):
"""
A stack container for storing Template state.
RenderContext simplifies the implementation of template Nodes by providing a
safe place to store state between invocations of a node's `render` method.
The RenderContext also provides scoping rules that are more sensible for
'template local' variables. The render context stack is pushed before each
template is rendered, creating a fresh scope with nothing in it. Name
resolution fails if a variable is not found at the top of the RequestContext
stack. Thus, variables are local to a specific template and don't affect the
rendering of other templates as they would if they were stored in the normal
template context.
"""
def __iter__(self):
for d in self.dicts[-1]:
yield d
def has_key(self, key):
return key in self.dicts[-1]
def get(self, key, otherwise=None):
return self.dicts[-1].get(key, otherwise)
def __getitem__(self, key):
return self.dicts[-1][key]
class RequestContext(Context):
"""
This subclass of template.Context automatically populates itself using
the processors defined in the engine's configuration.
Additional processors can be specified as a list of callables
using the "processors" keyword argument.
"""
def __init__(self, request, dict_=None, processors=None,
current_app=_current_app_undefined,
use_l10n=None, use_tz=None, engine=None):
# current_app isn't passed here to avoid triggering the deprecation
# warning in Context.__init__.
super(RequestContext, self).__init__(
dict_, use_l10n=use_l10n, use_tz=use_tz, engine=engine)
if current_app is not _current_app_undefined:
warnings.warn(
"The current_app argument of RequestContext is deprecated. "
"Set the current_app attribute of its request instead.",
RemovedInDjango20Warning, stacklevel=2)
self._current_app = current_app
self.request = request
self._processors = () if processors is None else tuple(processors)
self._processors_index = len(self.dicts)
self.update({}) # placeholder for context processors output
self.engine = engine # re-run the setter in case engine is not None
@property
def engine(self):
return self._engine
@engine.setter
def engine(self, engine):
self._engine = engine
if hasattr(self, '_processors_index'):
if engine is None:
# Unset context processors.
self.dicts[self._processors_index] = {}
else:
# Set context processors for this engine.
updates = {}
for processor in engine.template_context_processors + self._processors:
updates.update(processor(self.request))
self.dicts[self._processors_index] = updates
def new(self, values=None):
new_context = super(RequestContext, self).new(values)
# This is for backwards-compatibility: RequestContexts created via
# Context.new don't include values from context processors.
if hasattr(new_context, '_processors_index'):
del new_context._processors_index
return new_context
def make_context(context, request=None):
"""
Create a suitable Context from a plain dict and optionally an HttpRequest.
"""
if request is None:
context = Context(context)
else:
# The following pattern is required to ensure values from
# context override those from template context processors.
original_context = context
context = RequestContext(request)
if original_context:
context.push(original_context)
return context