django1/django/urls/resolvers.py

515 lines
20 KiB
Python
Raw Normal View History

"""
This module converts requested URLs to callback view functions.
RegexURLResolver is the main class here. Its resolve() method takes a URL (as
a string) and returns a ResolverMatch object which provides access to all
attributes of the resolved URL match.
"""
import functools
import re
import threading
from importlib import import_module
from urllib.parse import quote
from django.conf import settings
from django.core.checks import Warning
from django.core.checks.urls import check_resolver
from django.core.exceptions import ImproperlyConfigured
from django.utils.datastructures import MultiValueDict
from django.utils.functional import cached_property
from django.utils.http import RFC3986_SUBDELIMS
from django.utils.regex_helper import normalize
from django.utils.translation import get_language
from .exceptions import NoReverseMatch, Resolver404
from .utils import get_callable
class ResolverMatch:
def __init__(self, func, args, kwargs, url_name=None, app_names=None, namespaces=None):
self.func = func
self.args = args
self.kwargs = kwargs
self.url_name = url_name
# If a URLRegexResolver doesn't have a namespace or app_name, it passes
# in an empty value.
self.app_names = [x for x in app_names if x] if app_names else []
self.app_name = ':'.join(self.app_names)
self.namespaces = [x for x in namespaces if x] if namespaces else []
self.namespace = ':'.join(self.namespaces)
if not hasattr(func, '__name__'):
# A class-based view
self._func_path = func.__class__.__module__ + '.' + func.__class__.__name__
else:
# A function-based view
self._func_path = func.__module__ + '.' + func.__name__
view_path = url_name or self._func_path
self.view_name = ':'.join(self.namespaces + [view_path])
def __getitem__(self, index):
return (self.func, self.args, self.kwargs)[index]
def __repr__(self):
return "ResolverMatch(func=%s, args=%s, kwargs=%s, url_name=%s, app_names=%s, namespaces=%s)" % (
self._func_path, self.args, self.kwargs, self.url_name,
self.app_names, self.namespaces,
)
@functools.lru_cache(maxsize=None)
def get_resolver(urlconf=None):
if urlconf is None:
from django.conf import settings
urlconf = settings.ROOT_URLCONF
return RegexURLResolver(r'^/', urlconf)
@functools.lru_cache(maxsize=None)
def get_ns_resolver(ns_pattern, resolver):
# Build a namespaced resolver for the given parent URLconf pattern.
# This makes it possible to have captured parameters in the parent
# URLconf pattern.
ns_resolver = RegexURLResolver(ns_pattern, resolver.url_patterns)
return RegexURLResolver(r'^/', [ns_resolver])
class LocaleRegexDescriptor:
def __get__(self, instance, cls=None):
"""
Return a compiled regular expression based on the active language.
"""
if instance is None:
return self
# As a performance optimization, if the given regex string is a regular
# string (not a lazily-translated string proxy), compile it once and
# avoid per-language compilation.
if isinstance(instance._regex, str):
instance.__dict__['regex'] = self._compile(instance._regex)
return instance.__dict__['regex']
language_code = get_language()
if language_code not in instance._regex_dict:
instance._regex_dict[language_code] = self._compile(str(instance._regex))
return instance._regex_dict[language_code]
def _compile(self, regex):
"""
Compile and return the given regular expression.
"""
try:
return re.compile(regex)
except re.error as e:
raise ImproperlyConfigured(
'"%s" is not a valid regular expression: %s' % (regex, e)
)
class LocaleRegexProvider:
"""
A mixin to provide a default regex property which can vary by active
language.
"""
def __init__(self, regex):
# regex is either a string representing a regular expression, or a
# translatable string (using gettext_lazy) representing a regular
# expression.
self._regex = regex
self._regex_dict = {}
regex = LocaleRegexDescriptor()
def describe(self):
"""
Format the URL pattern for display in warning messages.
"""
description = "'{}'".format(self.regex.pattern)
if getattr(self, 'name', False):
description += " [name='{}']".format(self.name)
return description
def _check_pattern_startswith_slash(self):
"""
Check that the pattern does not begin with a forward slash.
"""
regex_pattern = self.regex.pattern
if not settings.APPEND_SLASH:
# Skip check as it can be useful to start a URL pattern with a slash
# when APPEND_SLASH=False.
return []
if (regex_pattern.startswith('/') or regex_pattern.startswith('^/')) and not regex_pattern.endswith('/'):
warning = Warning(
"Your URL pattern {} has a regex beginning with a '/'. Remove this "
"slash as it is unnecessary. If this pattern is targeted in an "
"include(), ensure the include() pattern has a trailing '/'.".format(
self.describe()
),
id="urls.W002",
)
return [warning]
else:
return []
class RegexURLPattern(LocaleRegexProvider):
def __init__(self, regex, callback, default_args=None, name=None):
LocaleRegexProvider.__init__(self, regex)
self.callback = callback # the view
self.default_args = default_args or {}
self.name = name
def __repr__(self):
return '<%s %s %s>' % (self.__class__.__name__, self.name, self.regex.pattern)
def check(self):
warnings = self._check_pattern_name()
if not warnings:
warnings = self._check_pattern_startswith_slash()
return warnings
def _check_pattern_name(self):
"""
Check that the pattern name does not contain a colon.
"""
if self.name is not None and ":" in self.name:
warning = Warning(
"Your URL pattern {} has a name including a ':'. Remove the colon, to "
"avoid ambiguous namespace references.".format(self.describe()),
id="urls.W003",
)
return [warning]
else:
return []
def resolve(self, path):
match = self.regex.search(path)
if match:
# If there are any named groups, use those as kwargs, ignoring
# non-named groups. Otherwise, pass all non-named arguments as
# positional arguments.
kwargs = match.groupdict()
args = () if kwargs else match.groups()
# In both cases, pass any extra_kwargs as **kwargs.
kwargs.update(self.default_args)
return ResolverMatch(self.callback, args, kwargs, self.name)
@cached_property
def lookup_str(self):
"""
A string that identifies the view (e.g. 'path.to.view_function' or
'path.to.ClassBasedView').
"""
callback = self.callback
# Python 3.5 collapses nested partials, so can change "while" to "if"
# when it's the minimum supported version.
while isinstance(callback, functools.partial):
callback = callback.func
if not hasattr(callback, '__name__'):
return callback.__module__ + "." + callback.__class__.__name__
return callback.__module__ + "." + callback.__qualname__
class RegexURLResolver(LocaleRegexProvider):
def __init__(self, regex, urlconf_name, default_kwargs=None, app_name=None, namespace=None):
LocaleRegexProvider.__init__(self, regex)
# urlconf_name is the dotted Python path to the module defining
# urlpatterns. It may also be an object with an urlpatterns attribute
# or urlpatterns itself.
self.urlconf_name = urlconf_name
self.callback = None
self.default_kwargs = default_kwargs or {}
self.namespace = namespace
self.app_name = app_name
self._reverse_dict = {}
self._namespace_dict = {}
self._app_dict = {}
# set of dotted paths to all functions and classes that are used in
# urlpatterns
self._callback_strs = set()
self._populated = False
self._local = threading.local()
def __repr__(self):
if isinstance(self.urlconf_name, list) and len(self.urlconf_name):
# Don't bother to output the whole list, it can be huge
urlconf_repr = '<%s list>' % self.urlconf_name[0].__class__.__name__
else:
urlconf_repr = repr(self.urlconf_name)
return '<%s %s (%s:%s) %s>' % (
self.__class__.__name__, urlconf_repr, self.app_name,
self.namespace, self.regex.pattern,
)
def check(self):
warnings = self._check_include_trailing_dollar()
for pattern in self.url_patterns:
warnings.extend(check_resolver(pattern))
if not warnings:
warnings = self._check_pattern_startswith_slash()
return warnings
def _check_include_trailing_dollar(self):
"""
Check that include is not used with a regex ending with a dollar.
"""
regex_pattern = self.regex.pattern
if regex_pattern.endswith('$') and not regex_pattern.endswith(r'\$'):
warning = Warning(
"Your URL pattern {} uses include with a regex ending with a '$'. "
"Remove the dollar from the regex to avoid problems including "
"URLs.".format(self.describe()),
id="urls.W001",
)
return [warning]
else:
return []
def _populate(self):
# Short-circuit if called recursively in this thread to prevent
# infinite recursion. Concurrent threads may call this at the same
# time and will need to continue, so set 'populating' on a
# thread-local variable.
if getattr(self._local, 'populating', False):
return
self._local.populating = True
lookups = MultiValueDict()
namespaces = {}
apps = {}
language_code = get_language()
for pattern in reversed(self.url_patterns):
if isinstance(pattern, RegexURLPattern):
self._callback_strs.add(pattern.lookup_str)
p_pattern = pattern.regex.pattern
if p_pattern.startswith('^'):
p_pattern = p_pattern[1:]
if isinstance(pattern, RegexURLResolver):
if pattern.namespace:
namespaces[pattern.namespace] = (p_pattern, pattern)
if pattern.app_name:
apps.setdefault(pattern.app_name, []).append(pattern.namespace)
else:
parent_pat = pattern.regex.pattern
for name in pattern.reverse_dict:
for matches, pat, defaults in pattern.reverse_dict.getlist(name):
new_matches = normalize(parent_pat + pat)
lookups.appendlist(
name,
(
new_matches,
p_pattern + pat,
dict(defaults, **pattern.default_kwargs),
)
)
for namespace, (prefix, sub_pattern) in pattern.namespace_dict.items():
namespaces[namespace] = (p_pattern + prefix, sub_pattern)
for app_name, namespace_list in pattern.app_dict.items():
apps.setdefault(app_name, []).extend(namespace_list)
if not getattr(pattern._local, 'populating', False):
pattern._populate()
self._callback_strs.update(pattern._callback_strs)
else:
bits = normalize(p_pattern)
lookups.appendlist(pattern.callback, (bits, p_pattern, pattern.default_args))
if pattern.name is not None:
lookups.appendlist(pattern.name, (bits, p_pattern, pattern.default_args))
self._reverse_dict[language_code] = lookups
self._namespace_dict[language_code] = namespaces
self._app_dict[language_code] = apps
self._populated = True
self._local.populating = False
@property
def reverse_dict(self):
language_code = get_language()
if language_code not in self._reverse_dict:
self._populate()
return self._reverse_dict[language_code]
@property
def namespace_dict(self):
language_code = get_language()
if language_code not in self._namespace_dict:
self._populate()
return self._namespace_dict[language_code]
@property
def app_dict(self):
language_code = get_language()
if language_code not in self._app_dict:
self._populate()
return self._app_dict[language_code]
def _is_callback(self, name):
if not self._populated:
self._populate()
return name in self._callback_strs
def resolve(self, path):
path = str(path) # path may be a reverse_lazy object
tried = []
match = self.regex.search(path)
if match:
new_path = path[match.end():]
for pattern in self.url_patterns:
try:
sub_match = pattern.resolve(new_path)
except Resolver404 as e:
sub_tried = e.args[0].get('tried')
if sub_tried is not None:
tried.extend([pattern] + t for t in sub_tried)
else:
tried.append([pattern])
else:
if sub_match:
# Merge captured arguments in match with submatch
sub_match_dict = dict(match.groupdict(), **self.default_kwargs)
sub_match_dict.update(sub_match.kwargs)
# If there are *any* named groups, ignore all non-named groups.
# Otherwise, pass all non-named arguments as positional arguments.
sub_match_args = sub_match.args
if not sub_match_dict:
sub_match_args = match.groups() + sub_match.args
return ResolverMatch(
sub_match.func,
sub_match_args,
sub_match_dict,
sub_match.url_name,
[self.app_name] + sub_match.app_names,
[self.namespace] + sub_match.namespaces,
)
tried.append([pattern])
raise Resolver404({'tried': tried, 'path': new_path})
raise Resolver404({'path': path})
@cached_property
def urlconf_module(self):
if isinstance(self.urlconf_name, str):
return import_module(self.urlconf_name)
else:
return self.urlconf_name
@cached_property
def url_patterns(self):
# urlconf_module might be a valid set of patterns, so we default to it
patterns = getattr(self.urlconf_module, "urlpatterns", self.urlconf_module)
try:
iter(patterns)
except TypeError:
msg = (
"The included URLconf '{name}' does not appear to have any "
"patterns in it. If you see valid patterns in the file then "
"the issue is probably caused by a circular import."
)
raise ImproperlyConfigured(msg.format(name=self.urlconf_name))
return patterns
def resolve_error_handler(self, view_type):
callback = getattr(self.urlconf_module, 'handler%s' % view_type, None)
if not callback:
# No handler specified in file; use lazy import, since
# django.conf.urls imports this file.
from django.conf import urls
callback = getattr(urls, 'handler%s' % view_type)
return get_callable(callback), {}
def reverse(self, lookup_view, *args, **kwargs):
return self._reverse_with_prefix(lookup_view, '', *args, **kwargs)
def _reverse_with_prefix(self, lookup_view, _prefix, *args, **kwargs):
if args and kwargs:
raise ValueError("Don't mix *args and **kwargs in call to reverse()!")
text_args = [str(v) for v in args]
text_kwargs = {k: str(v) for (k, v) in kwargs.items()}
if not self._populated:
self._populate()
possibilities = self.reverse_dict.getlist(lookup_view)
for possibility, pattern, defaults in possibilities:
for result, params in possibility:
if args:
if len(args) != len(params):
continue
candidate_subs = dict(zip(params, text_args))
else:
if set(kwargs) | set(defaults) != set(params) | set(defaults):
continue
matches = True
for k, v in defaults.items():
if kwargs.get(k, v) != v:
matches = False
break
if not matches:
continue
candidate_subs = text_kwargs
# WSGI provides decoded URLs, without %xx escapes, and the URL
# resolver operates on such URLs. First substitute arguments
# without quoting to build a decoded URL and look for a match.
# Then, if we have a match, redo the substitution with quoted
# arguments in order to return a properly encoded URL.
candidate_pat = _prefix.replace('%', '%%') + result
if re.search('^%s%s' % (re.escape(_prefix), pattern), candidate_pat % candidate_subs):
# safe characters from `pchar` definition of RFC 3986
url = quote(candidate_pat % candidate_subs, safe=RFC3986_SUBDELIMS + '/~:@')
# Don't allow construction of scheme relative urls.
if url.startswith('//'):
url = '/%%2F%s' % url[2:]
return url
# lookup_view can be URL name or callable, but callables are not
# friendly in error messages.
m = getattr(lookup_view, '__module__', None)
n = getattr(lookup_view, '__name__', None)
if m is not None and n is not None:
lookup_view_s = "%s.%s" % (m, n)
else:
lookup_view_s = lookup_view
patterns = [pattern for (possibility, pattern, defaults) in possibilities]
if patterns:
if args:
arg_msg = "arguments '%s'" % (args,)
elif kwargs:
arg_msg = "keyword arguments '%s'" % (kwargs,)
else:
arg_msg = "no arguments"
msg = (
"Reverse for '%s' with %s not found. %d pattern(s) tried: %s" %
(lookup_view_s, arg_msg, len(patterns), patterns)
)
else:
msg = (
"Reverse for '%(view)s' not found. '%(view)s' is not "
"a valid view function or pattern name." % {'view': lookup_view_s}
)
raise NoReverseMatch(msg)
class LocaleRegexURLResolver(RegexURLResolver):
"""
A URL resolver that always matches the active language code as URL prefix.
Rather than taking a regex argument, we just override the ``regex``
function to always return the active language-code as regex.
"""
def __init__(
self, urlconf_name, default_kwargs=None, app_name=None, namespace=None,
prefix_default_language=True,
):
super().__init__(None, urlconf_name, default_kwargs, app_name, namespace)
self.prefix_default_language = prefix_default_language
@property
def regex(self):
language_code = get_language() or settings.LANGUAGE_CODE
if language_code not in self._regex_dict:
if language_code == settings.LANGUAGE_CODE and not self.prefix_default_language:
regex_string = ''
else:
regex_string = '^%s/' % language_code
self._regex_dict[language_code] = re.compile(regex_string)
return self._regex_dict[language_code]