Proposal (Was: When are and aren't two URLs the same?)

[Keith Howe]

On 4/21/06, Johannes Ernst <jernst+lists.danga.com at netmesh.us> wrote:
> There is no more #8 -- you mean the trailing slash? Just say that any
> trailing slashes make no difference.

It was more that I'm not convinced trailing slashes should be ignored.
They have very real implications about how a server does its lookup
(unlike the other points raised, such as default ports and such), so
I'm not sure it is valid to try and make this assumption at all.

While it is valid to assume that an empty path component is identical
to a root path component (as covered in 3.2.3 of RFC2616), that cannot
be extended further. e.g.:

1. http://joe.example.com - path component is ""
2. http://joe.example.com/ - path component is "/"
3. http://joe.example.com/id - path component is "/id"
4. http://joe.example.com/id/ - path component is "/id/"

By RFC2616, 1 and 2 are identical, but 3 and 4 are not, and I think
this behavior should be true for Yadis as well.

In fact, I think it would be best to just replace the rules 4, 5, and
6&7 as I stated them with a rule that comparisons should also conform
to section 3.2.3 of RFC2616.

> Was this intentional that this is off-list?

No, just an oversight on my part.

- K.Howe