Group the notepad into work packages, move two reliability items to M5

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01H7LwYgJfpkbLCTeiAf8U4A
2026-08-16 16:20:48 +02:00
parent b1e1b4d501
commit 19970b0ad8
2 changed files with 87 additions and 96 deletions
@@ -6,117 +6,95 @@ Where TYPE could be BUG, FEAT, PERF, CHORE and SCOPE could be UX, UI, FLOW, NODE
 Don't write temporary reasons for deferring a task in the task description (only strategical reasons should be noted).
 Deferring because out of scope is fine, but don't mention deferring than.
 Open work is grouped into packages, each one session's worth of work with its
 own file territory so two can run at once. Packages are ordered by priority;
 so are the items inside them. `Deferred` holds what stays open on purpose,
 each with the condition that should reopen it.
 ## Open
- FEAT/UX: unify the layout between dashboard and flows: 1) In the flows: adapt the "overview" from the dashboard; clicking flows in the sidebar leads to a searchable overview of flows like in the dashboard. Then the "flowbar" get's merged into the bottom toolbar (status and edit button; add flow button goes in the overview) with the name of the current flow still being displayed at the top for reference. 2) adopt the floating bottom toolbar from the flow viewport and the click widget to edit behavior (make sure inputs like sliders and buttons stay interactable). Also adapt the new flow-title (sync UI) and the new edit button in the toolbar to trigger a panel for renaming the flow (and more later). The same floating right panel as in the flows should also be the way to go to edit a widget. Then in the toolbar should be a "View" button which opens the dashboard in a new browser tab (renders without the sidebar or any edit option) with a link to be used in a wallpanel later
+### A — Dashboard rework
- FEAT/UI: allow resizing of dashboard widgets by dragging edges and add the same dotted background we see in the flows viewport (again, sync UI (see above))
+
- FEAT/UI: add a graph widget to the dashboard
+Closes the M4 "visualization flows" remainder. Depends on B: it adopts the
 flow canvas' toolbar, side panel and title chrome, so those move first.
 - FEAT/UX: unify the layout between dashboard and flows: 1) In the flows: adapt the "overview" from the dashboard; clicking flows in the sidebar leads to a searchable overview of flows like in the dashboard. Then the "flowbar" get's merged into the bottom toolbar (status and edit button; add flow button goes in the overview) with the name of the current flow still being displayed at the top for reference. 2) adopt the floating bottom toolbar from the flow viewport and the click widget to edit behavior (make sure inputs like sliders and buttons stay interactable). Also adapt the new flow-title (sync UI) and the new edit button in the toolbar to trigger a panel for renaming the flow (and more later). The same floating right panel as in the flows should also be the way to go to edit a widget.
 - FEAT/UI/MOBILE: a "View" button in the toolbar opens the dashboard in a new tab on a full-bleed route — no sidebar, no edit affordances, the link a wall panel is pointed at. Dashboards render in the padded admin shell today, so a panel gets the sidebar, footer and a `max-w-7xl` column.
 - FEAT/UI: allow resizing of dashboard widgets by dragging edges and add the same dotted background we see in the flows viewport (again, sync UI (see above)). Widgets are sized with a wider/narrower control today; the document already stores layout per breakpoint, so only the editor changes.
 - FEAT/UX: allow setting a grid size for the dashboard to adjust to different wallpanel sizes (goes into the new dashboard panel setting; see above)
 - FEAT/UI: add a graph widget to the dashboard. The chart widget type is stored, validated and backed by message history already, but nothing draws it and the editor's add row filters it out. Needs a charting library and `--chart-*` tokens: one designed sequential scale, reintroduced when this lands. The node sparkline draws one series in `--primary` and needs none.
 - BUG/UI: strictly type a dashboard widget; i.e. wiring a boolean switch to a float message (or wiring no publish/consume at all) should flag the node as error -> reuse error indicator from the flows
- FEAt/UX: allow setting a grid size for the dashboard to adjust to different wallpanel sizes (goes into the new dashboard panel setting; see above)
+
 ### B — Flow canvas and editor UX
 - FEAT/UX: allow copy-pasting of nodes (also between flows)
 - FEAT/UX: introduce general keyboard shortcut system. As a first test: when hitting ctrl-s we should catch this keyboard shortcut to a) publish a flow b) save (and apply) code changes when in the code editor
 - FEAT/UX: in delay and schedule nodes: help creating cron schedules by providing a legend; Also: if interval is entered, this should automatically create a cron according to the interval set
 - FEAT/UX: python function should automatically update according to what the node consumes or produces
 - BUG/UI: there are two places to rename a flow; we should only go for the one in the top of the panel name (so that the UX is identical to the ones from the nodes). The actual rename should happen with dedicated "confirm" button (same for the nodes)
 - FEAT/UI: publishing and discarding are only reachable while no side panel is open, since the floating chrome hides for the panel. Editing a node's code and publishing it is therefore close-panel-then-publish.
 - BUG/UI: when the edges pulse on message, they change their color (go from slightly darker back to normal) at the end again. Double check that after the animation they end up with the same color as before to not have this double-pulse effect
 - BUG/UI: clicking outside the panel does not discard the flow edit panel
 - FEAT/UX: for the same type of node show suggestions in the setting fields based on other nodes; I.e. if I create an mqtt node and I already configured one in another flow, allow selecting the values of the exiting mqtt node as a drop-down option (identical to consumables) when clicking inside the corresponding field in the new node (same for all other nodes)
 - BUG/UI: the graph showed in the node edit panel should also be shown for a specific edge inside the pop-up panel when clicking the edge
 - PERF/FLOW: every save rebuilds the whole pipeline. Fine at the current flow count; rebuild
  only the touched flow when it starts to show.
 - FEAT/UI: the chart widget is stored and validated but not drawn — it is filtered out of
  the editor's add row. Needs a charting library and the `--chart-*` tokens below.
 - FEAT/UI: dashboard widgets are sized with a wider/narrower control rather than dragged.
  A grid library would give drag-and-resize and per-breakpoint layouts; the document
  already stores layout per breakpoint, so only the editor changes.
 - FEAT/UI/MOBILE: dashboards render in the padded admin shell, so a wall panel gets the
  sidebar, footer and a `max-w-7xl` column. A full-bleed shell would suit a panel better.
 - FEAT/UI: reintroduce `--chart-*` tokens as one designed sequential scale when the first
  chart lands. The node sparkline draws one series in `--primary` and needs none.
 - FEAT/FLOW: single-stepping a paused flow. The work queue it needed now exists: pausing
  parks claimed items per flow and resuming drains them, so a step button is a matter of
  popping one parked item instead of all of them.
 - FEAT/UI: interrupting a node that is already running. Pause holds nodes that have not been
  submitted yet; one already executing runs to completion.
 - FEAT/UI: flow-level edits (title, inputs) go through `FlowPanel.onChange` and stay outside
  the undo stack.
 - BUG/UI: the enlarged panel (for code editing) should still maintain its floating style
- FEAT/UI: no screen for alerting. Channels and rules are API-only (`/alerts/config`),
+- FEAT/UI: a "Bug" icon on the node error bubble opening the logs panel at that node's traceback. The panel has the traceback; wiring the bubble to open and filter it needs the panel's open state lifted into `FlowEditor`.
-  so setting up a phone notification means calling the endpoint by hand.
+- FEAT/UI: flow-level edits (title, inputs) go through `FlowPanel.onChange` and stay outside the undo stack.
- FEAT/UI: there is no screen for managing the secrets store itself. A node parameter marked
+- BUG/UI: `SidePanel`'s mobile branch does not set `data-testid` on the `SheetContent`, so `[data-testid=node-panel]` does not exist on a phone. Mobile specs cannot address the panel.
  `x-secret` offers the stored secrets, but they can only be created through the API. We should introduce ad dedicated page (in the sidebar) to handle secrets in a centralized way
 - BUG/UI: `SidePanel`'s mobile branch does not set `data-testid` on the `SheetContent`, so
  `[data-testid=node-panel]` does not exist on a phone. Mobile specs cannot address the panel.
 - CHORE/API: node source saves carry no version precondition, so two clients editing the same
  node's code are last-writer-wins. The flow document is what the optimistic lock protects;
  code files would need their own, and an exact-match one produces false conflicts against a
  single client's own interleaved flow and source saves.
 - FEAT/UI: publishing and discarding are only reachable while no side panel is open, since the
  floating chrome hides for the panel. Editing a node's code and publishing it is therefore
  close-panel-then-publish.
 - FEAT/UI: no screen for the OAuth clients an agent registers. They can only be listed or
  removed in the database, so withdrawing one agent's access means deleting its rows or
  rotating `OAUTH_PRIVATE_KEY_FILE`, which cuts off all of them.
 - CHORE/INFRA: `requires-python` is capped below 3.14 because the MCP SDK wants a newer
  starlette there than the pinned `sentry-sdk<2` allows. Lift the cap when sentry-sdk moves
  to 2.x.
 - FEAT/FLOW: input discretization drops the trailing edge — if a producer goes quiet inside
  the window, the consumer never sees the last value. Delivering it at the end of the window
  needs a timer per (node, port).
 - CHORE/FLOW: shared node sources bypass the draft/publish split. Editing one writes the
  library copy and reloads immediately, since the code is not any single flow's to hold back.
  Deliberate, but it means a shared node is the one thing publish does not gate.
 - CHORE/UI: `make test-backend` cannot reach Postgres while the integrated stack is up —
  `compose.local.yml` does `db: ports: !reset []`. Run it against the container's address, or
  move the suite inside the compose network.
 - CHORE/UI: the Playwright specs run against the development stack and leave their users and
  flows behind, which is why the flowbar filled with `test_flow_*`. Give them their own data
  or clean up after themselves, as `pytest` now does.
 - CHORE/UI: the standalone Vite origin (`make dev-frontend` on :5173) is still not in
  `BACKEND_CORS_ORIGINS`. `scripts/setup.sh` regenerates the list on every `make init` and
  `compose.local.yml` re-declares it as a container variable that wins, so the origin has to
  be added in both places to work.
 - CHORE/INFRA: `test-backend.yml` runs pytest on the runner host, so it keeps the fixed
  container name and port 5432 that `compose.ci.yml` resets everywhere else. Two concurrent
  runs of that one job collide.
 - FEAT/UI: an endpoint's edge routes straight across the graph, so it can pass behind a
  node that sits between the lane and the node it wires to. Readable, but a routed edge
  would be tidier.
 - CHORE/FLOW: a node function returning something other than a dict raises `AttributeError`
  in `Node._to_messages` rather than a named error. Outputs are keyed by port, so a non-dict
  cannot be one — say so where the return value is mapped.
 - PERF/UI: the app's entry chunk is 680 kB (210 kB gzipped) and exceeds the warning
  threshold. React Flow and Monaco are already lazy; a manualChunks split measured no better,
  so this needs route-level work on the shell rather than chunking config.
- CHORE/INFRA: `bun run --filter frontend build` fails on this workspace with
+### C — Node settings UX
  `crypto.hash is not a function` — Vite 7 wants Node 20.12+ and the host has 18. The
  Docker image builds fine, so it only bites local bundling; `bunx tsc` still type-checks.
 - PERF/UI: (deferred for now) the Monaco chunk is 2.6 MB. It only loads when a node panel opens, but the
  editor could be trimmed further or swapped for CodeMirror if that becomes a problem.
 - CHORE/FLOW: four modules of `app/flow/nodes/` are still excluded from strict mypy
  (`[[tool.mypy.overrides]]` in `pyproject.toml`): `mqtt`, `http`, `influx` and `delay`. They
  carry prototype typing, `croniter` ships no stubs and `influxdb_client` does not re-export
  its names. `base` and `mlp` are checked; shrink the rest as each integration is revisited.
- FEAT/NODE: the connectors only read. Enable the write paths with someone watching:
+Node panel only, so it can run alongside A once B has landed.
  WF-RAC `setAirconStat` (needs an operatorId registered with the unit first, which is
  itself a write) and Art-Net `transmit`.
 - CHORE/NODE: `wfrac` reports `mode` as "unknown" while the unit is off, because the
  mode bits hold a value outside the known set. Faithful to the reference decoder, but
  "off" would read better.
- FEAT/NODE: the second WF-RAC unit (the one Node-RED addresses with operatorId "0")
+- FEAT/UX: for the same type of node show suggestions in the setting fields based on other nodes; I.e. if I create an mqtt node and I already configured one in another flow, allow selecting the values of the exiting mqtt node as a drop-down option (identical to consumables) when clicking inside the corresponding field in the new node (same for all other nodes)
-  closes the connection on an anonymous read. It likely wants an account registered;
+- FEAT/UX: in delay and schedule nodes: help creating cron schedules by providing a legend; Also: if interval is entered, this should automatically create a cron according to the interval set
-  the first unit answers without one.
+- FEAT/UX: python function should automatically update according to what the node consumes or produces
 ### D — Admin screens
 Three backends without a face. API-first holds — the OAuth one needs its
 endpoints written before the screen.
 - FEAT/UI: there is no screen for managing the secrets store itself. A node parameter marked `x-secret` offers the stored secrets, but they can only be created through the API. We should introduce a dedicated page (in the sidebar) to handle secrets in a centralized way. The engine's own "secret not found" error already points the user at such a screen.
 - FEAT/UI: no screen for alerting. Channels and rules are API-only (`/alerts/config`), so setting up a phone notification means calling the endpoint by hand.
 - FEAT/UI: no screen for the OAuth clients an agent registers. They can only be listed or removed in the database, so withdrawing one agent's access means deleting its rows or rotating `OAUTH_PRIVATE_KEY_FILE`, which cuts off all of them.
 ### E — Engine
 - FEAT/FLOW: single-stepping a paused flow. The work queue it needed now exists: pausing parks claimed items per flow and resuming drains them, so a step button is a matter of popping one parked item instead of all of them.
 - BUG/FLOW: `Pipeline.resume()` re-executes a flow's nodes but does not appear to unpark what pausing parked, so parked items may only drain on the next trigger. Confirm against `queue.unpark` callers before building the step button on top.
 - FEAT/FLOW: input discretization drops the trailing edge — if a producer goes quiet inside the window, the consumer never sees the last value. Delivering it at the end of the window needs a timer per (node, port).
 - CHORE/FLOW: a node function returning something other than a dict raises `AttributeError` in `Node._to_messages` rather than a named error. Outputs are keyed by port, so a non-dict cannot be one — say so where the return value is mapped, which is the one place every caller routes through.
 ### F — Infrastructure and dev hygiene
 - CHORE/UI: the Playwright specs run against the development stack and leave their users and flows behind, which is why the flowbar filled with `test_flow_*`. Give them their own data or clean up after themselves, as `pytest` now does.
 - CHORE/UI: `make test-backend` cannot reach Postgres while the integrated stack is up — `compose.local.yml` does `db: ports: !reset []`. Run it against the container's address, or move the suite inside the compose network.
 - CHORE/UI: the standalone Vite origin (`make dev-frontend` on :5173) is still not in `BACKEND_CORS_ORIGINS`. `scripts/setup.sh` regenerates the list on every `make init` and `compose.local.yml` re-declares it as a container variable that wins, so the origin has to be added in both places to work.
 - CHORE/INFRA: `test-backend.yml` runs pytest on the runner host, so it keeps the fixed container name and port 5432 that `compose.ci.yml` resets everywhere else. Two concurrent runs of that one job collide.
 - FEAT/INFRA: MQTT broker and InfluxDB compose services for local development. The node types exist; a local broker would make them testable without external hardware.
 - CHORE/FLOW: four modules of `app/flow/nodes/` are still excluded from strict mypy (`[[tool.mypy.overrides]]` in `pyproject.toml`): `mqtt`, `http`, `influx` and `delay`. They carry prototype typing, `croniter` ships no stubs and `influxdb_client` does not re-export its names. `base` and `mlp` are checked; shrink the rest as each integration is revisited.
 ### G — Connector write paths
 Needs someone watching the real hardware, so it is not a background task.
 - FEAT/NODE: the connectors only read. Enable the write paths with someone watching: WF-RAC `setAirconStat` (needs an operatorId registered with the unit first, which is itself a write) and Art-Net `transmit`.
 - FEAT/NODE: the second WF-RAC unit (the one Node-RED addresses with operatorId "0") closes the connection on an anonymous read. It likely wants an account registered; the first unit answers without one.
 - CHORE/NODE: `wfrac` reports `mode` as "unknown" while the unit is off, because the mode bits hold a value outside the known set. Faithful to the reference decoder, but "off" would read better.
 ## Deferred
 Open on purpose. Each names what should bring it back.
 - PERF/FLOW: every save rebuilds the whole pipeline. Fine at the current flow count; rebuild only the touched flow when it starts to show.
 - PERF/UI: the app's entry chunk is 680 kB (210 kB gzipped) and exceeds the warning threshold. React Flow and Monaco are already lazy; a manualChunks split measured no better, so this needs route-level work on the shell rather than chunking config.
 - PERF/UI: the Monaco chunk is 2.6 MB. It only loads when a node panel opens, but the editor could be trimmed further or swapped for CodeMirror if that becomes a problem.
 - CHORE/API: node source saves carry no version precondition, so two clients editing the same node's code are last-writer-wins. The flow document is what the optimistic lock protects; code files would need their own, and an exact-match one produces false conflicts against a single client's own interleaved flow and source saves. Revisit with the M5 multi-user work.
 - CHORE/FLOW: shared node sources bypass the draft/publish split. Editing one writes the library copy and reloads immediately, since the code is not any single flow's to hold back. Deliberate, but it means a shared node is the one thing publish does not gate.
 - CHORE/INFRA: `requires-python` is capped below 3.14 because the MCP SDK wants a newer starlette there than the pinned `sentry-sdk<2` allows. Lift the cap when sentry-sdk moves to 2.x.
 - CHORE/INFRA: `bun run --filter frontend build` fails on this workspace with `crypto.hash is not a function` — Vite 7 wants Node 20.12+ and the host has 18. The Docker image builds fine, so it only bites local bundling; `bunx tsc` still type-checks.
 - FEAT/UI: an endpoint's edge routes straight across the graph, so it can pass behind a node that sits between the lane and the node it wires to. Readable, but a routed edge would be tidier.
 ## Blocked
 - FEAT/UI: a "Bug" icon on the node error bubble opening the logs panel at that node's
  traceback. The panel now has the traceback; wiring the bubble to open and filter it needs
  the panel's open state lifted into `FlowEditor`.
 - FEAT/INFRA: MQTT broker and InfluxDB compose services for local development. The node types
  exist; a local broker would make them testable without external hardware.
 - CHORE/INFRA: `bun install` inside the frontend Docker build intermittently fails with
  "Fail extracting tarball" for several packages at once, and succeeds on a plain rebuild.
  It looks like concurrent extraction under memory pressure. Pin down or retry in the
@@ -23,6 +23,9 @@ existing flow engine reachable and persistent precedes new feature breadth.
 - [x] Design token contract: root `DESIGN-GUIDELINES.md`, per-repo `DESIGN.md`,
      byte-identical token blocks verified by `make design-check`
 - [ ] CI on Codeberg (Forgejo Actions): pre-commit, backend tests, Playwright, compose smoke
 - [ ] Soak and chaos harness: sustained load with the state backend, the broker and
      the engine itself taken away underneath it. The reliability layers were verified
      by hand against a live instance; keeping them verified needs a harness
 ## Phase 1 — Backend: management
@@ -118,6 +121,9 @@ Rust, optimised for throughput. Executes nodes and distributes them across worke
 - [ ] Parallel invocation of stateless nodes over independent input sets, to
      keep I/O delay minimal (stateful I/O nodes keep serializing via the
      `synchronous` mechanism)
 - [ ] Run user Python nodes out of process. One occupies a worker thread until it
      returns today, so a runaway node cannot be bounded by a timeout nor cancelled
      from the canvas — both fall out of the isolation
 - [ ] Extract node execution from the Python prototype into a Rust engine
 - [ ] Worker distribution and load balancing across capable devices
 - [ ] Input/output validation at the node boundary
@@ -152,6 +158,9 @@ React + Vite, primarily desktop but usable on mobile. See `docs/architecture/str
 - [ ] Device assignment per node, selectable from compatible devices
 - [ ] Test-node affordance on the canvas
 - [ ] User management screens
 - [ ] Screens for what the API already owns alone: the secrets store, alert
      channels and rules, and the OAuth clients an agent registers — the last
      needs its management endpoints written first
 - [x] Mobile-friendly canvas: touch connect, full-screen node panel
 - [ ] Installable as a PWA (`vite-plugin-pwa`)
@@ -163,7 +172,11 @@ Shares components with the admin view. See `docs/architecture/structure.canvas`
 - [x] User-defined dashboard layout with edit and view modes: dashboards are their own
      documents, widgets bind to message names, and the input ones publish back. View
      mode is plain CSS grid, so a panel that only displays loads no editing code
- [ ] Responsive layout targeting wall panels, mobile and desktop
+- [ ] Chart widget drawing a message's history, with the one sequential
      `--chart-*` scale it needs. The type and its history are stored and served
      already; nothing draws it
 - [ ] Layout by dragging rather than a width control, a grid size per dashboard,
      and a full-bleed route so a wall panel loads no admin chrome
 - [ ] Per-device view
 ## Phase 5 — Website and docs