steid

1	# Current
2
3	> Keep this file short. One active step, one ordered backlog. Completed work moves to
4	> [progress.md](progress.md). If this file starts reading like a changelog, it has
5	> drifted — that's exactly what went wrong last time.
6
7	## Active: Milestone 2 — Profile page
8
9	Goal: `/{handle}` is the real profile page — public, working signed out, and the
10	frame that repos, writing, and projects hang off later. It replaces the Milestone 0
11	placeholder.
12
13	Out of scope: repos, posts, and projects don't exist yet, so there is nothing to
14	list. Avatars, following, and anything social. Organisation profiles beyond what falls
15	out for free.
16
17	### Phase 1 — the page
18
19	Shippable on its own: a public profile that renders signed out.
20
21	- [x] Settle URL shape — handles at the root, routes grouped under prefixes
22	([0004](decisions/0004-root-handles-grouped-routes.md))
23	- [x] Reserved-handle denylist in `OrgName::new`; auth routes moved under `/auth/`
24	- [ ] Decide `module_router!` vs explicit `#[page]` paths
25	- [ ] `PublicProfile` read model + `view_profile` use case — must not carry email
26	- [ ] `/{handle}` page: label, handle, bio, and the section frame; 404 on unknown,
27	case-insensitive
28	- [ ] Assert `/auth/login` and `/api/me` still route once `/{handle}` exists at the root
29	- [ ] `/` redirects to the owner's profile once claimed, retiring the placeholder
30	- [ ] `/api/users/{handle}` — same read model, public JSON
31
32	### Phase 2 — make it yours
33
34	A profile you can't change is a stub. This is what makes it a portfolio page.
35
36	- [ ] Migration: `orgs.bio`
37	- [ ] Flash messages — the first edit form needs success and failure feedback, and
38	every form after it inherits whatever we build here
39	- [ ] `/{handle}/settings` — edit display name and bio, owner only, enforced in the use
40	case (settings belong to the org, and this scales to organisations)
41	- [ ] Owner-only affordances on the profile (edit link)
42
43	### Done when
44
45	Signed out, `/{handle}` renders the owner's display name and handle and nothing
46	private. An unknown handle 404s. The owner can set a display name and bio and see them
47	on the page. `/api/users/{handle}` returns the same public view.
48
49	### Watch for
50
51	- Do not leak email. `describe_identity` carries email, and `/api/me` returns it —
52	correctly, because that endpoint describes the caller to themselves. The public
53	profile needs its own read model; reusing `Identity` would publish the owner's
54	email address to anonymous visitors. This is the single most likely mistake in this
55	milestone.
56	- Route precedence between static routes and `/{handle}`. The reserved list stops a
57	user owning `auth`, but it does not stop the router matching `/api/me` against
58	`/{handle}/{x}` and shadowing the real route. Static-over-parameterised is near
59	universal, so this is an assertion when the route lands, not a blocking spike.
60	- Case-insensitive handles. Storage is `collate nocase` and `OrgName::new`
61	lowercases, so `/JamesGill` must resolve rather than 404.
62	- Authorization on settings belongs in the use case, taking an `Actor` — not in the
63	page. Otherwise `/api` gets a different answer from the web form.
64	- A profile is public. This is the first page rendering for anonymous visitors by
65	design, so anything private must be gated explicitly rather than by assuming a
66	session exists.
67	- Reserve handles early. Adding to the denylist later is a breaking change for
68	whoever holds that handle ([0004](decisions/0004-root-handles-grouped-routes.md)).
69
70	### Settled
71
72	- Settings live at `/{handle}/settings`. They belong to the organisation, which
73	scales to real organisations in Milestone 7 without moving.
74	- Render the full frame from the start, including sections with nothing in them.
75	A profile page that renders almost nothing is a poor start for a portfolio-first
76	product; the shape of the page is part of what is being built, not scaffolding for
77	it.
78
79	### Carried over — small, unblocked
80
81	- No rate limiting on `/auth/login` or `/auth/setup`.
82	- `sweep_expired` is never called, so expired session rows accumulate. Expiry is
83	enforced on read, so this is tidiness, not a hole.
84	- CSRF. `SameSite=Lax` covers the common case; whether forms also want tokens is
85	still undecided. Phase 2 adds a form, so this is the natural time to settle it.
86	- Styling. Everything is unstyled HTML. Topcoat ships Tailwind without Node, and
87	`steid-backup/AGENTS/UI.md` has a full OKLCH system to mine. Cheaper now than at ten
88	pages — see [ui.md](ui.md).
89
90	## Backlog
91
92	Ordered. Pull from the top.
93
94	1. Milestone 3 — Writing. Posts, markdown rendering, `/{handle}/posts/{slug}`.
95	*Open question: is writing actually the first portfolio feature, or is it
96	projects/showcases?*
97	2. Milestone 4 — Repo model. `Repository` entity, `Visibility`, `create_repo`,
98	bare repo on disk at `{data_dir}/{org}/{repo}.git`. Watch the DB-plus-filesystem
99	atomicity problem — see [architecture.md](architecture.md#db-plus-filesystem-writes).
100	3. Milestone 5 — Git over HTTP. `git http-backend` subprocess, PATs over HTTP
101	Basic. See [0001](decisions/0001-git-over-http-not-ssh.md).
102
103	## Open questions
104
105	- Topcoat is early (v0.5.0, first released 2026-07-22, breaking changes expected
106	by its own authors). Expect churn that isn't feature work.
107	- Body size limits will reject large pushes at Milestone 5 — `topcoat-router` has a
108	`body_limit` layer that needs raising on the git routes. Recorded here because it
109	will surface as a confusing failure rather than a clear one.
110	- Topcoat ships Tailwind without Node, which reopens the design system attempt #1
111	dropped purely to avoid an npm build step — see [ui.md](ui.md).
112
113	## Routing findings (Milestone 0)
114
115	- Topcoat 0.5 requires rustc ≥ 1.95. On an older toolchain `cargo add topcoat`
116	silently resolves to an empty `topcoat v0.0.0` placeholder instead of failing. Local
117	stable is now 1.97.1.
118	- `Router::builder().discover()` collects `#[page]`-annotated items at link time,
119	so pages can live in any module. Layering is our choice, not the framework's.
120	- `module_router!` derives each URL from the module tree rather than a path string.
121	Still deferred. Application routes now group cleanly (`auth/login`, `api/me`), but
122	handles sit at the root ([0004](decisions/0004-root-handles-grouped-routes.md)), so a
123	parameterised root segment still has to coexist with static ones. Worth checking how
124	`module_router!` handles that before committing to it.
125	- Path and query params are read from `Cx` via `path_param!` / `#[query_params]`, not
126	injected as handler arguments. Parses are memoized per request.
127	- Layouts wrap by path prefix and nest outermost-first, and a layout can catch a page's
128	`NotFoundError` to render a branded 404.
129	- `HOST` / `PORT` configure the bind address, so `STEID_LISTEN_ADDR` is gone.
130	- `Body` is a boxed `http_body::Body` used for both requests and responses, with
131	`into_data_stream()` to read and `Body::new()` to wrap a stream — pack data can
132	stream both directions without buffering. This is what makes Milestone 5 viable.