I'm using Django for my blog: http://www.mazelife.com/blog. My model for a blog post uses a slug as a primary key, and that field only allows letters, numbers and dashes to be used (in regex terms, [a-zA-Z0-9-]+)
When using the comments app, this causes and error when I try to call get_content_object_url on a comment object attached to a blog post.
Let's assume a hypothetical comment object, with the following properties:
comment.object_pk = "sample-blog-post-slug"
comment.content_type_id = 11
Step 1: comment.get_content_object_url() is called in my application
Step 2: in django.contrib.comments.models, this is in the get_content_object_url method, ln. 36 - 39:
return urlresolvers.reverse(
"comments-url-redirect",
args=(self.content_type_id, self.object_pk)
)
So, in this case, this code is executed:
urlresolvers.reverse("comments-url-redirect", args=(11, "sample-blog-post-slug"))
Step 3. urlresolvers.reverse attempts to match this candidate:
"comments/cr/11/sample-blog-post-slug/"
against this pattern, from django.contrib.comments.urls (ln. 17):
r'^cr/(\d+)/(\w+)/$'
The problem here is that \w sequence does not match "-" in the blog post slug, so urlresolvers.reverse raises a NoReverseMatch? exception.
I think that, for a primary key that will be used in a url, "-" is a vaild character and is probably even more prevalent than "_", which is matched by \w.
My proposed solution would be to modify django.contrib.comments.urls to match "-" characters in an object_pk. I've attached a patch to django.contrib.comments.urls that does this.
