Repository navigation
Replies: 2 comments
|
Thanks for this. Most of the mechanics are already in main, so here's what exists and what I'll add. Already in main (
What's missing, and I'll implement:
So the readiness boundary stays where the resync is defined today, in |
0 replies
|
Landed in 0fa3fc7, and checked with the installed PWA on an iPhone after the phone slept.
It ships in the next release. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Could MailFlow expose an explicit data-readiness state after a PWA resumes or the WebSocket reconnects?
A connection can be healthy while the visible list and counts still reflect an older generation. The proposed UX is to keep a compact catching-up indication until the session is confirmed and the active list, categories and counts have refreshed; show a bounded retry action if catch-up fails. A successful health request alone should not mark the view current. In-flight refresh requests need replay when a newer event arrives.
This is a feature/design proposal based on a fork prototype, not a claim that a specific failure has been reproduced on clean upstream main. It is distinct from provider authentication reports and the conversation account-scope fix #607. Related category semantics: #597.
The prototype has generation/session guards and synthetic transport tests for backend restart and missed refreshes. Reference: https://github.com/shady2k/mailflow/tree/integration/upstream-20261014 . Desktop/mobile WebKit tests do not establish installed iOS PWA behavior; that still needs device validation. Is this behavior suitable for upstream, and where should the readiness boundary live?
All reactions