Clarify when the user can abort the payment request algorithm

[marcoscaceres]

closes #796

The following tasks have been completed:

• Confirmed there are no ReSpec errors/warnings.
• Modified Web platform tests (link)
• Modified MDN Docs (link)

Implementation commitment:

• Safari (link to issue)
• Chrome (link to issue)
• Firefox
• Edge (public signal)

Optional, Impact on Payment Handler spec?
None

---

Preview | Diff

[marcoscaceres]

So, we had in there that there could be a timeout or some other means of aborting. However, I think we were going down the wrong path... I've routed the "timeout" and the user hitting "esc" down the "user aborts the payment request" path instead. I also removed the "[[updating]]" check to allow for this.

[marcoscaceres]

@domenic, if you have time, can you please sanity check this?

[domenic]

The duplication between "user aborts the payment request" and "abort the update" made this hard to review, but in the end I think the switch to that algorithm is fine. We should consider deduplicating those two algorithms though, if we can.

[marcoscaceres]

Filed bug about de-dupe abort the update and user aborts. #814

[marcoscaceres]

As this makes provisions for an optional UI feature (a "cancel" button or user triggered action, like a hitting "esc"), I'm ok with merging this as is. It doesn't really add any new conformance requirements. It just clarifies behavior that matches what browsers already do.