NoReverseMatch with keyword argument uidb64 with Django 2.0

In Django 2.0 and 2.1 you should call decode() after base64 encoding the uid, to convert it to a string:

message = render_to_string('acc_active_email.html', {    'user': user,    'domain': current_site.domain,    'uid': urlsafe_base64_encode(force_bytes(user.pk)).decode(),    'token': account_activation_token.make_token(user),})

See the note in the Django 2.0 release notes for more info.

In Django 2.2+, urlsafe_base64_encode returns a string, so there is no need to decode.

message = render_to_string('acc_active_email.html', {    'user': user,    'domain': current_site.domain,    'uid': urlsafe_base64_encode(force_bytes(user.pk)),    'token': account_activation_token.make_token(user),})

It should be possible to write code that is compatible with Django <= 1.11, 2.0-2.1, and 2.2+, by using force_text. Note the following is untested.

from django.utils.encoding import force_textmessage = render_to_string('acc_active_email.html', {    'user': user,    'domain': current_site.domain,    'uid': force_text(urlsafe_base64_encode(force_bytes(user.pk))),    'token': account_activation_token.make_token(user),})

You can drop the force_text and use the second code snippet once you drop support for Django < 2.2.

For newer versions of Django, you can use the slug syntax. For example:

path(    'activate/<slug:uidb64>/<slug:token>/',    views.activate_account,     name='activate')

Maybe its stupid since the URL is not dynamic but it worked...

path('activate/(?P<uidb64>[0-9A-Za-z_\-]+)/(?P<token>[0-9A-Za-z]{1,13}-[0-9A-Za-z]{1,20})/$', views.activate, name='activate'),