Move sign-in, forgot-password and reset-password onto TanStack Form v2 #419

New issue

Closed

opened 2026-10-15 12:34:35 +00:00 by root · 7 comments

root commented

2026-10-15 12:34:35 +00:00

(Migrated from gitlab.core-project.space)

Parent

#410

What to build

Sign-in, forgot-password and reset-password run on the shared TanStack Form v2 module. A Player sees the same fields, checks, refusals and screens as today: one refusal for wrong credentials, credentials no Account could have refused without a request, the "sent" confirmation, the invalid-link screen.

Seam: S1 — each page with the auth harness and MSW; S2 — accounts.spec in apps/e2e, unchanged.

Acceptance criteria

The three pages use the shared module; no per-field useState remains on them.
Validators call the existing credential checks and return our refusal unions.
Server refusals and the rate limit are returned from onSubmit, never thrown.
Focus moves to the first invalid field after a refused submit.
The existing sign-in, forgot-password and reset-password tests pass unchanged in what they assert.

## Parent #410 ## What to build Sign-in, forgot-password and reset-password run on the shared TanStack Form v2 module. A Player sees the same fields, checks, refusals and screens as today: one refusal for wrong credentials, credentials no Account could have refused without a request, the "sent" confirmation, the invalid-link screen. Seam: S1 — each page with the auth harness and MSW; S2 — `accounts.spec` in `apps/e2e`, unchanged. ## Acceptance criteria - [ ] The three pages use the shared module; no per-field `useState` remains on them. - [ ] Validators call the existing credential checks and return our refusal unions. - [ ] Server refusals and the rate limit are returned from `onSubmit`, never thrown. - [ ] Focus moves to the first invalid field after a refused submit. - [ ] The existing sign-in, forgot-password and reset-password tests pass unchanged in what they assert.

root commented

2026-10-15 12:34:35 +00:00

(Migrated from gitlab.core-project.space)

marked this issue as related to #410