search

w3c / payment-method-id

Payment Method Identifiers specification
https://w3c.github.io/payment-method-id/
Other
23 stars 20 forks source link

Identifier equivalence does not handle failure #8

Closed annevk closed 7 years ago

annevk commented 8 years ago

URL parsing can fail, especially without base URL.

zkoch commented 7 years ago

We've decided to use absolute URLs, so I don't think this is relevant any longer. Closing.

annevk commented 7 years ago

How would it not be relevant? How do you test whether it's an absolute URL?

webpayments-specs commented 7 years ago

Well... since it is a requirement on an author, not an implementor, we don't really need to test that strictly speaking. It's not that I don't appreciate the "how many angels on the head of a pin" arguments about URL parsing, I just don't think they matter as much when we are talking about something that is not end-user facing. While it is true that these URLs can be dereferenced, I expect that in reality they will be compared to items in cache as string literals.

So at the end of the day, while there is a requirement that payment method URLs be absolute, what really matters to a merchant is that the strings match up. And that's what I expect to test.

On Tue, Sep 13, 2016 at 2:24 AM, Anne van Kesteren <notifications@github.com

wrote:

How would it not be relevant? How do you test whether it's an absolute URL?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/w3c/webpayments-method-identifiers/issues/8#issuecomment-246597434, or mute the thread https://github.com/notifications/unsubscribe-auth/AQ8khoU-BOQ6bjGEPNaoekjT7u00wRhlks5qpk-hgaJpZM4JiM5r .

Shane McCarron Projects Manager, Spec-Ops

halindrome commented 7 years ago

My bad - I replied to the email instead of coming in here.... that comment above is from me.

annevk commented 7 years ago

But you need to define that. Otherwise some implementations end up treating https://example, https://example/, https://example:443/, and https://EXAMPLE/ as identical, whereas others don't.

halindrome commented 7 years ago

Isn't the reference to the URL spec and its definition of URL Parser sufficient? What additional text would you propose?

annevk commented 7 years ago

Well if you keep the current text, you need to handle failure since https://url.spec.whatwg.org/#url-equivalence expects two URLs and not failure as argument. Again, I don't understand why this issue was closed.

zkoch commented 7 years ago

We can re-open and close when the PR comes out.