Reputation: 2353
I'm currently using the following urls.py
:
api_patterns = [
url(r'^users/', include('users.urls', namespace='user')),
]
internal_patterns = [
# ...
]
urlpatterns = [
url(r'^api/', include(api_patterns)),
url(r'^internal/', include(internal_patterns)),
url(r'^admin/', include(admin.site.urls)),
url(r'^(?!(?:api|internal|admin)/)', MainView.as_view()),
]
The point of this config is to render MainView
if url doesn't have the api
, internal
or admin
prefix:
/api/users/...
— found/api/foo/
— not found/foo/
— foundHow can I make it simplier and more intent revealing?
Upvotes: 4
Views: 2020
Reputation: 173
For django >= 3 rather use re_path
:
from django.urls import re_path
urlpatterns = [
re_path(r'^.*',MainView.as_view())
]
Upvotes: 1
Reputation: 13058
I think your intent will be more clear if you do this in two urls:
url(r'^(api|internal|admin)/', SomeView.as_view()),
url(r'^.*', MainView.as_view())
MainView
will be executed only if a url does not begin with api, internal or admin.
SomeView
will be executed if a url begins with api/internal/admin but doesn't match the patterns above it. You can customize this view to either return a default 404 page, or perform other functions as you need.
Using your examples:
/api/users
will execute include(api_patterns)
/api/foo
will execute SomeView
/foo
will execute MainView
Edit
To address the first point in your comment: url patterns are regexes, while you can extract these into variables to remove duplication, it can make code hard to read. Here's one example for removing duplication (I'm sure you can come up with others):
d = OrderedDict([
(r'api', api_patterns),
(r'internal', internal_patterns),
(r'admin', admin.site.urls),
])
main_view_re = r'^!({})/'.format('|'.join(d.keys()))
urlpatterns = [url(r'^{}/'.format(k), include(v)) for k, v in d]
urlpatterns.append(url(main_view_re, MainView.as_view()))
Upvotes: 3
Reputation: 316
urlpatterns = [
url(r'^api/', include(api_patterns)),
url(r'^internal/', include(internal_patterns)),
url(r'^admin/', include(admin.site.urls)),
url(r'', MainView.as_view()),
]
Leaving no prefix would allow you to catch any URL that a user might try after the URL conf matches the api, internal and admin url's.
Upvotes: -2