django.urls utility functions

reverse()

如果在你的代码中,需要使用一些类似url模板标签,Django提供了下列功能:

reverse(viewname, urlconf=None, args=None, kwargs=None, current_app=None)

viewname 可以是一个url模式名称或一个可调用的视图对象。例如,给定以下 url:

  1. from news import views
  2. path('archive/', views.archive, name='news-archive')

you can use any of the following to reverse the URL:

  1. # using the named URL
  2. reverse('news-archive')
  3. # passing a callable object
  4. # (This is discouraged because you can't reverse namespaced views this way.)
  5. from news import views
  6. reverse(views.archive)

If the URL accepts arguments, you may pass them in args. For example:

  1. from django.urls import reverse
  2. def myview(request):
  3. return HttpResponseRedirect(reverse('arch-summary', args=[1945]))

You can also pass kwargs instead of args. For example:

  1. >>> reverse('admin:app_list', kwargs={'app_label': 'auth'})
  2. '/admin/auth/'

args and kwargs cannot be passed to reverse() at the same time.

If no match can be made, reverse() raises a NoReverseMatch exception.

The reverse() function can reverse a large variety of regular expression patterns for URLs, but not every possible one. The main restriction at the moment is that the pattern cannot contain alternative choices using the vertical bar ("|") character. You can quite happily use such patterns for matching against incoming URLs and sending them off to views, but you cannot reverse such patterns.

The current_app argument allows you to provide a hint to the resolver indicating the application to which the currently executing view belongs. This current_app argument is used as a hint to resolve application namespaces into URLs on specific application instances, according to the namespaced URL resolution strategy.

The urlconf argument is the URLconf module containing the URL patterns to use for reversing. By default, the root URLconf for the current thread is used.

注解

The string returned by reverse() is already urlquoted. For example:

  1. >>> reverse('cities', args=['Orléans'])
  2. '.../Orl%C3%A9ans/'

Applying further encoding (such as urllib.parse.quote()) to the output of reverse() may produce undesirable results.

reverse_lazy()

A lazily evaluated version of reverse().

reverse_lazy(viewname, urlconf=None, args=None, kwargs=None, current_app=None)

It is useful for when you need to use a URL reversal before your project’s URLConf is loaded. Some common cases where this function is necessary are:

  • providing a reversed URL as the url attribute of a generic class-based view.
  • providing a reversed URL to a decorator (such as the login_url argument for the django.contrib.auth.decorators.permission_required() decorator).
  • providing a reversed URL as a default value for a parameter in a function’s signature.

resolve()

The resolve() function can be used for resolving URL paths to the corresponding view functions. It has the following signature:

resolve(path, urlconf=None)

path is the URL path you want to resolve. As with reverse(), you don’t need to worry about the urlconf parameter. The function returns a ResolverMatch object that allows you to access various metadata about the resolved URL.

If the URL does not resolve, the function raises a Resolver404 exception (a subclass of Http404) .

class ResolverMatch

  • func

    The view function that would be used to serve the URL

  • args

    The arguments that would be passed to the view function, as parsed from the URL.

  • kwargs

    The keyword arguments that would be passed to the view function, as parsed from the URL.

  • url_name

    The name of the URL pattern that matches the URL.

  • route

    The route of the matching URL pattern.

    For example, if path('users/<id>/', ...) is the matching pattern, route will contain 'users/<id>/'.

  • app_name

    The application namespace for the URL pattern that matches the URL.

  • app_names

    The list of individual namespace components in the full application namespace for the URL pattern that matches the URL. For example, if the app_name is 'foo:bar', then app_names will be ['foo', 'bar'].

  • namespace

    The instance namespace for the URL pattern that matches the URL.

  • namespaces

    The list of individual namespace components in the full instance namespace for the URL pattern that matches the URL. i.e., if the namespace is foo:bar, then namespaces will be ['foo', 'bar'].

  • view_name

    The name of the view that matches the URL, including the namespace if there is one.

A ResolverMatch object can then be interrogated to provide information about the URL pattern that matches a URL:

  1. # Resolve a URL
  2. match = resolve('/some/path/')
  3. # Print the URL pattern that matches the URL
  4. print(match.url_name)

A ResolverMatch object can also be assigned to a triple:

  1. func, args, kwargs = resolve('/some/path/')

One possible use of resolve() would be to test whether a view would raise a Http404 error before redirecting to it:

  1. from urllib.parse import urlparse
  2. from django.urls import resolve
  3. from django.http import Http404, HttpResponseRedirect
  4. def myview(request):
  5. next = request.META.get('HTTP_REFERER', None) or '/'
  6. response = HttpResponseRedirect(next)
  7. # modify the request and response as required, e.g. change locale
  8. # and set corresponding locale cookie
  9. view, args, kwargs = resolve(urlparse(next)[2])
  10. kwargs['request'] = request
  11. try:
  12. view(*args, **kwargs)
  13. except Http404:
  14. return HttpResponseRedirect('/')
  15. return response

get_script_prefix()

get_script_prefix()

Normally, you should always use reverse() to define URLs within your application. However, if your application constructs part of the URL hierarchy itself, you may occasionally need to generate URLs. In that case, you need to be able to find the base URL of the Django project within its Web server (normally, reverse() takes care of this for you). In that case, you can call get_script_prefix(), which will return the script prefix portion of the URL for your Django project. If your Django project is at the root of its web server, this is always "/".