Every node type persisted its size in `nodes.width` / `nodes.height` except
`groupRect`, which stashed it inside the `custom_colors` JSON next to its
colours. The columns already existed and were simply unused for zones, so
this was an inconsistency rather than a missing-column workaround, and it
put geometry in a blob that otherwise holds style.
The serializer now writes the columns for a zone too, and strips the legacy
`width`/`height` keys out of the blob so the two cannot drift apart and
leave an older canvas reading a stale size.
No data is lost on upgrade:
- `_backfill_zone_size` copies the blob geometry into the columns at
startup. It only fills a column that is still NULL, so it cannot overwrite
a size set since; it parses the JSON in Python rather than with
`json_extract`, so it does not depend on the SQLite build carrying JSON1;
and an unreadable row is skipped without costing the others their size.
Re-running it is a no-op.
- the reader still falls back to the blob, covering a payload the backfill
has not reached — an older server, or an import.
Standalone mode is unaffected: it stores React Flow nodes verbatim, so the
size was always on `node.width` / `node.height` there.
The four serializer tests that pinned the size to the blob now assert the
columns, since that is the behaviour being changed.
ha-relevant: yes
Since 3.3.0 the inventory row owns a device's services and properties, and
every node drawing that device rendered the row wholesale. One row shared by
several canvases meant one rendering: a service the scanner fingerprinted
appeared on every canvas at once (users reported Uptime Kuma and Synology DSM
on hosts running neither — both are port-only signatures), and a property added
on one schematic showed up on all the others.
Order and visibility are presentation, so they move to the node. `display_view`
records, per node, which of the row's services and properties it draws and in
what order, keyed by `port|protocol|name` and by lowercased property key so the
view survives an edit to the fact itself. The facts stay on the row: hiding is
per node, deleting is still device-wide.
An item the view does not list is reported hidden rather than dropped, so what
a scan finds next is one toggle away instead of pushed onto every canvas.
The wire shape is unchanged. A client already sends its lists in display order
with their `visible` flags, so the view is read back out of them rather than
asking for a second field, and `visible` is only stamped when something is
hidden.
Upgrades keep what each canvas showed:
* from 3.2.0, the backfill seeds each node's view from its own legacy columns
before they are dropped;
* from 3.3.0-3.3.2 those columns are gone and the row holds the union of every
canvas, so the layout is recovered from the backup `_backup_db` took before
the 3.3.0 migration — the newest one whose `nodes` table still has the
columns, read read-only and matched by node id;
* with no usable backup the row is the seed, so every canvas keeps showing
exactly what it shows today and only later additions are held back.
Refs #347
ha-relevant: maybe
Upgrading 3.2.0 -> 3.3.0 could leave a database where approving a device —
or creating any node — failed with `NOT NULL constraint failed: nodes.status`.
The 3.3.0 migration moves the device facts off `nodes` onto `device_inventory`
and then drops the columns. The drop is skipped when a node is still unlinked,
because those columns are the only remaining copy of its facts. But a 3.2.0
`nodes` declares `status`, `services`, `properties` and `show_hardware` NOT NULL
with no server-side default, and the 3.3.0 model no longer writes them — so a
skipped drop bricks every later INSERT.
It skipped because one node killed the whole backfill: the loop shared a single
session, so an IntegrityError raised by autoflush lost every link made before it.
- Give each node its own savepoint. A node whose merge violates a constraint is
skipped and logged on its own; the rest still link. `skipped` joins the stats
reported at boot.
- Relax NOT NULL on the retained legacy columns when the drop is skipped, so the
database stays writable while later boots retry the backfill. Values are kept;
only the constraint goes. Guarded on the PRAGMA notnull flag, so it runs once.
- Share one `_rebuild_nodes` between the drop and the relax: it restores
`PRAGMA foreign_keys = ON` in a finally (the connection returns to the pool)
and clears a `nodes_new` left by a failed attempt.
- Match IEEE addresses case-insensitively, and never write one another inventory
row already owns — `device_inventory.ieee_address` is UNIQUE, and a duplicate
is one way the backfill was raising in the first place.
Tests build the real v3.2.0 schema from the tag: the existing legacy-migration
fixture declares `status` nullable, which is why this reached a release.
Fixes#351
ha-relevant: no
Completes the split: `nodes` now holds only how a device is drawn on one
canvas, and every device fact reaches the API from the inventory row.
- The device columns are removed from `nodes` (SQLite table rebuild, the
same shape as the existing device_inventory and canvas_state rebuilds).
The drop is skipped, and logged, while any non-furniture node is still
unlinked: those columns are the last copy of that node's facts. The
backfill therefore reads them with raw SQL — by the time it runs, the
model no longer declares them.
- The status checker iterates devices, not nodes: one check per device
however many canvases draw it, writing `status_live` / `last_seen` /
`response_time_ms` on the row. `/ws/status` messages carry `device_id`
and the node ids they light up. Hidden devices are not probed.
- Readers repointed: scanner (last_scan lands on the row), proxmox (its
node tier collapses into the inventory tier, keeping only the cluster
handles), zigbee/zwave (one property refresh serves every canvas), rack
inventory, liveview, stats, node dedupe.
- `POST /scan/pending` merges into the row that already describes the host
instead of minting a second one — one device is one row, whichever way it
was documented.
- Standalone keeps parity: the canvas blob gains `devices`, split on save
and hydrated on load. A blob written before the split still reads.
The rack inventory had a related bug: a mount that names a node explicitly
printed the mount's device rather than the pinned node's. It now reads the
node's own row.
Tests that built a node with device columns are ported to the link; where a
behaviour genuinely moved (properties refresh once on the row, last_scan is
the device's) the assertion moved with it rather than being dropped.
ha-relevant: yes
A device drawn on three canvases was three independent copies of the same
facts. Point every node at the Device Inventory row it draws, and let that
row own what the device *is* — the node keeps only how it is drawn.
- nodes.device_id -> device_inventory.id, ON DELETE SET NULL. NULL for
canvas furniture (group / groupRect / text), which describes nothing
physical. Deleting a node never deletes the row.
- services/inventory_sync holds the shared rules: matching by ieee > ip >
mac (per token, so 10.0.0.4 never matches 10.0.0.40), a property union on
key, a service union on (port, protocol, name), and the backfill that
links every pre-existing node.
- The backfill is non-destructive by construction: it writes device_id and
fills the row, and deletes nothing. Nodes are visited oldest-edit-first,
so where two canvases disagree on a scalar the most recently edited wins,
while properties and services stay unioned — nothing any canvas recorded
is lost. A second boot finds nothing to do.
- The wire shape does not change: GET /canvas hydrates the device fields
from the row, and a save routes them back to it. Editing a node's IP on
one canvas now shows on every other canvas holding that device.
- approve / bulk-approve set device_id instead of owning a copy, and a new
canvas node joins (or mints) its row. Rows minted this way are tagged
with a new `canvas` discovery source and get their own inventory filter.
- DetailPanel offers "Open in inventory"; standalone, which has no
inventory, is not offered it.
test_racks' "reports what the canvas node knows" seeded a second inventory
row for a host that already had one — a state a node create can no longer
produce. Its seeding order is swapped so the node links to the row; every
assertion is unchanged.
ha-relevant: yes
The Device Inventory was a read-only discovery log: rows arrived from a
scan or an import and went stale. Everything a user curates — properties,
services, notes, hardware, check method — could only be edited on a canvas
node, and never came back.
Give the inventory row the fields a node carries and a way to write them:
- device_inventory gains label, type, notes, cpu_count/cpu_model/ram_gb/
disk_gb, show_hardware, check_method/check_target, last_seen, last_scan,
response_time_ms and updated_at. Live reachability goes in a new
status_live: `status` already means the pending/approved/hidden lifecycle
and the two must not be conflated.
- PATCH /api/v1/scan/pending/{id} applies only what the client sends, so
editing one field never clears the rest. Lifecycle and discovery
bookkeeping stay owned by the approve/hide routes and the importers.
- InventoryDeviceCreate carries the same fields, so a hand-made entry needs
no create-then-PATCH round trip.
- InventoryDeviceModal becomes show *and* edit, reusing the canvas editors
(PropertyList, ServiceModal) rather than growing a second implementation.
- InventoryEntry moves to types/index.ts, its home; the modal re-exports it
so existing call sites are untouched. NODE_TYPE_GROUPS moves to
utils/nodeTypeGroups so both type pickers share one vocabulary.
ha-relevant: maybe
The rack's link picker read `/nodes`, so it only ever offered devices
someone had already approved onto a logical canvas — two rows on a homelab
holding 74 inventory entries. A device on no canvas is still the record of
a real box, and is exactly what a rack is built out of.
`DevicePickerModal` replaces `NodePickerModal` and lists the Device
Inventory itself. Picking an entry calls the new `relinkDevice`, which
repoints the mount's `deviceId`, adopts that entry's node, status and —
unless the user renamed the plate — its label. One entry, one mount: a row
another plate stands for is not offered, and the store refuses it anyway.
The placeholder a rack-created plate left behind is dropped through the new
`DELETE /api/v1/scan/pending/{id}`, which refuses a device a rack still
mounts (409): foreign keys are off at runtime, so the mount would be left
naming a row that no longer exists.
`LinkedDevicePanel` becomes "Linked device" and now prints what discovery
found even when nothing on a canvas answers for the device; only the
canvas-side rows go missing, under a "Not on a logical canvas." note.
Also renames `pending_devices` to `device_inventory` (and
`pending_device_links` to `device_inventory_links`), with the Python and
TypeScript names that followed it. "Pending devices" was the scanner's word
for a queue of finds awaiting approval; the rows outlive approval, are
edited by hand and are what a rack mounts. Routes, payload keys and MCP
tool names are a published contract and are unchanged — `/scan/pending/*`
and the `pending_devices` key in `/stats` stay as they are.
The rename migration runs before `create_all`, or an empty new table would
be created beside the populated old one and every scanned device would read
as gone; it repairs that state too, for anyone whose app already started
mid-upgrade. Foreign keys are switched on for the rename so SQLite rewrites
the `REFERENCES` clause in `rack_devices`.
ha-relevant: maybe
Clicking a cable — in patch mode or out of it — selects it and opens a
right rail. It carries the physical facts (type, colour, label) and
`properties`: the same NodeProperty records the logical canvas already
uses for nodes, each with its own "show on canvas" eye. What is ticked
visible is drawn on a small plate at the midpoint of the run, so an
export carries its lengths and VLANs.
The property editor moves out of DetailPanel into
components/common/PropertyList so both canvases share one implementation
rather than growing a second.
Persistence: rack_cables gains label_visible and properties, added by an
idempotent migration in _try_migrate; the response schema coerces the
NULLs legacy rows read back as.
Also drops the 40 % plate fade that came with cables-on. It let the rail
strips and the U grid show through the mounted gear, which read as a
rendering bug — cables are drawn above the plates anyway.
ha-relevant: maybe
Reconcile the same physical device discovered by both the nmap IP scan and
the Proxmox importer into a single inventory row, keyed on MAC. Previously
each path only deduped by IP, and the importer captured no MAC (and no IP for
stopped guests), so most guests double-listed.
Backend:
- mac_utils.normalize_mac: canonical MAC (lowercase, ':'-separated), the
cross-source join key. Normalized on write and on compare.
- proxmox_service: capture the guest NIC MAC agent-free from the net0 config
(qemu virtio=<MAC>, lxc hwaddr=<MAC>); works for stopped guests. Resolver
now returns (ip, mac).
- proxmox persist: match existing Node/PendingDevice by ieee OR ip OR MAC;
fill mac, keep the vm/lxc type, union sources.
- scanner persist: match PendingDevice by ip OR MAC; fill the IP a Proxmox
import lacked, keep a pve row's type, union the scan source. Stamp query
matches raw + normalized MAC (legacy-safe).
- Multi-source tags: new PendingDevice.discovery_sources JSON column so a
merged device shows under both the IP and Proxmox filters. Idempotent
migration backfills from discovery_source (legacy NULL-scalar rows with an
IP become ["arp"]). _sources_after_merge preserves a scanned row's IP origin
through the merge without tagging a pure Proxmox guest.
- Import now broadcasts a scan update on completion so an open inventory
reloads without a manual refresh.
Frontend:
- pendingSources: sourceBuckets/orderedSources map discovery_sources to filter
buckets; a device with ["arp","proxmox"] matches both filters and renders
both badges. PendingDevicesModal filter + badges use them.
Tests: MAC normalization, config MAC capture, cross-source merge both
directions, legacy-row IP-tag preservation, no-false-IP-tag guard, refresh
broadcast, and the frontend bucket mapping.
ha-relevant: maybe
The pending_devices.properties column is added by an idempotent migration, so
existing rows have properties = NULL. PendingDeviceResponse typed it as a list,
so GET /scan/pending 500'd on any pre-existing device.
- Coerce NULL/non-list properties to [] in PendingDeviceResponse.
- Backfill existing NULL rows to '[]' in init_db migrations.
- Regression test: /scan/pending returns 200 with a legacy NULL-properties row.
ha-relevant: maybe
Add a Proxmox VE importer that reads the /api2/json REST API with a read-only
API token and drops hosts (proxmox), VMs (vm) and LXC containers (lxc) onto the
canvas as typed nodes with run state and hardware specs (vCPU/RAM/disk).
- Backend: proxmox_service (httpx) + proxmox routes (test-connection, import,
import-pending, config). Two-tier dedupe — merge onto an existing scanned node
by IP, else synthetic pve-{host}-{vmid} identity. Update-in-place, never
deletes. Host->guest rendered as a 'virtual' edge via the pending-link flow.
- Security: token is env-only (PROXMOX_TOKEN_*), never written to disk by the
app, never returned by any endpoint; errors are credential-sanitized.
- Auto-sync: optional scheduled re-import into pending (APScheduler job).
- PendingDevice.properties carries specs through approve (+ migration).
- Frontend: ProxmoxImportModal, sidebar entry, pending inventory source filter,
Settings auto-sync section, proxmoxApi client.
- Docs: docs/proxmox-import.md, README + FEATURES sections, .env.example keys.
- Tests: backend service/router/scheduler, frontend modal/client/pending.
ha-relevant: maybe
Edge render (solid/dashed/dotted) and stroke width were hardcoded per
edge type. Expose both as user settings.
- Custom Style modal (Edges): line-style buttons, 1-4x width slider,
live preview; left-list swatch renders the actual line.
- Per-edge EdgeModal: same controls; line style follows the type preset
live until overridden.
- Renderer applies line_style/width_mult over BASE_STYLES (width scales
markers + animation overlays); unset keeps the type default look.
- Persist line_style/width_mult through serializer, canvas save, and the
edges API (new nullable columns, idempotent migration).
ha-relevant: yes
Add optional filled-triangle arrowheads at either end of an edge,
independently toggleable per edge (EdgeModal) and as per-edge-type
defaults (CustomStyleModal). Arrowheads are custom inline <marker> defs
filled with the live stroke colour so they recolour reactively with
custom_color / vlan / selected state. Persisted frontend (serializer)
and backend (edge columns + schemas + runtime migration).
Also fix two dedupe layers that silently dropped legitimate parallel
links between the same two devices:
- store: React Flow addEdge() connectionExists dropped a second edge
with matching source+target when handles were null/equal. Build the
edge with a unique id and append directly.
- render: rewireEdgesForCollapse deduped ALL edges by src->tgt key even
when nothing was collapsed, filtering real parallel edges out of the
visible set. Restrict the anti-mesh dedupe to rewired collapse stubs.
Tests: marker render, per-edge/per-type UI, store apply, serializer
round-trip, backend edge/canvas persistence, parallel-edge regressions.
ha-relevant: yes
The backend only stored bottom_handles, so top/left/right_handles were dropped
on canvas save and reset to defaults on reload — a left/right snappoint would
vanish after reload.
- models.py: add top_handles (default 1), left_handles (0), right_handles (0)
- database.py: ALTER TABLE migrations for the three columns
- schemas: add fields to NodeBase, NodeUpdate, and the canvas-save node schema
- tests: save+reload round-trip and default-fallback coverage
ha-relevant: no
The Show Port Numbers toggle reset on every reload because the backend
never stored it: the column was missing from the Node model and the field
was stripped by NodeSave/NodeBase/NodeUpdate schemas, so canvas/save
silently dropped it.
Add the show_port_numbers column (idempotent migration), and the field to
the node schemas so it round-trips through save/load.
Add regression tests covering persistence and the default-false case.
Fixes#184
ha-relevant: yes
Make designs (canvases) fully user-manageable: create with a chosen name
and icon, rename, change icon, and delete. Replaces the hardcoded
"New Electrical Design" button with a generic "New Canvas" flow.
- Add Design.icon column + migration that backfills legacy rows
(electrical -> zap, others -> dashboard)
- DesignModal: name input + curated lucide icon picker (create + edit)
- Sidebar switcher gains per-canvas edit/delete; delete guards the last
canvas and confirms
- designStore: addDesign/updateDesign/removeDesign with active reassignment
- Fix data loss on design switch: abort load when the save fails and keep
unsaved edits; skip the save-old step when the previous canvas was deleted
- designsApi create/update carry icon; design_type kept for back-compat
Tests: backend design CRUD (icon + cascade + last-canvas guard), designStore
actions, designIcons resolver, DesignModal create/edit/validation.
ha-relevant: yes
Mirrors IP scan flow: POST /zigbee/import-pending now creates a
ScanRun(kind=zigbee, status=running) and returns immediately.
Networkmap fetch + pending upsert run in the background, status
transitions to done/error when finished.
Frontend: import modal closes on submit, scan history shows the
run with a ZIG/IP kind chip and toasts on completion. Pending
modal auto-refreshes when run finishes.
scan_runs.kind column added (default 'ip', idempotent migration).
Existing zigbee tests refactored to exercise _persist_pending_import
directly (background tasks don't see the test session); route test
verifies the run is created with kind=zigbee.
Pre-push code review flagged the inline `with suppress(OperationalError)`
ALTERs as silently eating real errors. The project still uses inline
init_db migrations (CLAUDE.md aspires to Alembic but the runtime path
is inline), so as a minimum: add a `_try_migrate` helper that
distinguishes already-applied (debug log) from genuine failures
(warning log). Apply it to the Zigbee-related migrations only — the
legacy ALTERs above remain unchanged in this commit's scope.
Coordinator auto-approves to a canvas Node; routers/end devices land in
pending_devices keyed by IEEE. Discovered parent->child edges are stored
in pending_device_links so that approving a pending device later
auto-creates the Edge once both endpoints exist as canvas Nodes.
- new POST /api/v1/zigbee/import-pending (default mode in modal)
- new pending_device_links table; ieee_address on nodes + pending_devices
- pending_devices.ip migrated to nullable (table rebuild on existing DBs)
- approve / bulk-approve return auto-created edges; sidebar pushes them
into the canvas store with bottom -> top-t handles
- ZigbeeImportModal: radio toggle pending vs canvas; reset on close
- PendingDeviceModal: zigbee badge, IEEE/LQI/vendor/model rows, services
hidden for zigbee
- Sidebar pending row: ZIG source badge, LQI badge, friendly_name fallback
- SearchBar: null-safe IP, also searches friendly_name and ieee_address
- Tooltip trigger uses asChild to avoid nested-button hydration error
- Add VERSION file at repo root as single source of truth for app version
- frontend/vite.config.ts reads VERSION file instead of package.json
- backend config.py exposes APP_VERSION read from VERSION (dev) or /app/VERSION (Docker)
- database.py backs up DB to homelab.db.back-{version} before running migrations
(skipped if DB doesn't exist or backup already exists — fully idempotent)
- Dockerfile.backend and Dockerfile.frontend copy VERSION into the image
- Add test_db_backup.py with 4 tests covering create/skip/idempotent/version cases
Add waypoints JSON column to edges table, include it in all edge
schemas (EdgeBase, EdgeUpdate, canvas CanvasEdge) and add the
idempotent ALTER TABLE migration so existing databases are upgraded
on next startup.
- canvasStore.onConnect: include animated in edge data object (was silently dropped)
- Backend schemas: normalize animated bool/int to string ('none'/'snake'/'flow') via field_validator
- ORM model: change animated column from Boolean to String
- DB migration: convert existing 0/1 boolean rows to 'none'/'snake' strings
Add NodeResizer to BaseNode so users can drag corners to resize any node.
Persist width/height through the full stack: DB model, schemas, canvas
save/load route, and migration for existing databases.
Add tests covering save, update, clear, and load of node dimensions.
- Add animated toggle per edge in EdgeModal (cyan switch, "Flow Animation")
- SVG-native <animate> element for reliable cross-browser dot animation
- Persist animated field: backend model, schemas (EdgeBase/EdgeUpdate/EdgeSave), DB migration
- Include animated in App.tsx edgesToSave serialization so it survives save/reload
- Add animated: bool to EdgeData TypeScript type
On fresh installs the data/ dir may not exist, causing SQLite to create a
stub file without write permissions. mkdir(parents=True, exist_ok=True) runs
at import time before any DB operation.
Also add data/.gitignore to prevent homelab.db from ever being committed.
- Add 65+ icons across 7 categories (Infrastructure, Media, Monitoring,
Storage, Security, Automation, Dev & Containers, Communications) covering
popular self-hosted apps: Home Assistant, Jellyfin, Plex, Grafana, Portainer,
Pi-hole, Vaultwarden, Gitea, Nextcloud, Node-RED, Frigate, etc.
- New nodeIcons.ts utility with ICON_REGISTRY, ICON_MAP and resolveNodeIcon()
- Inline icon picker in NodeModal: collapsible panel with search + grid grouped
by category; click to select, click again or Reset to revert to type default
- BaseNode uses resolveNodeIcon() so custom icon renders live on canvas
- Add custom_icon field to NodeData type, NodeBase/NodeUpdate schemas, Node ORM
model, and database.py idempotent ALTER TABLE migration
- Add custom_color field to EdgeData type, Edge DB model and all schemas
- HomelableEdge applies custom_color as stroke override (before selected highlight)
- EDGE_DEFAULT_COLORS extracted to utils/edgeColors.ts (react-refresh compliant)
- EdgeModal: color picker row showing effective color (custom or type default)
with hex value, Reset button when custom color is active
- Auto-migration adds custom_color column to edges table
- Add resolveNodeColors() utility merging type defaults with per-node overrides
- Default colors per node type (cyan=isp/router/lxc/ap, green=switch/nas,
purple=server/vm, orange=proxmox, amber=iot, gray=generic)
- Remove glowColor prop from BaseNode — colors now come from node data
- ProxmoxGroupNode uses resolveNodeColors for group border/header/icon
- NodeModal: Appearance section with 3 color swatches (border, background, icon)
— click swatch to open native color picker; Reset to defaults button
- custom_colors persisted as JSON in DB (backend model + schemas + migration)
- 7 new unit tests for resolveNodeColors covering all node types + partial overrides
- Double-click any link to open Edit Link modal (type, label, VLAN ID, delete)
- Add updateEdge / deleteEdge actions to canvasStore
- Add container_mode field to proxmox nodes (backend model, schemas, migration)
- NodeModal shows container toggle for proxmox type (defaults ON)
- ProxmoxGroupNode renders as regular BaseNode when container_mode is OFF
- setProxmoxContainerMode store action handles structural changes atomically
(children parentId/extent, node dimensions)
- CanvasContainer filters edges between container proxmox and its children
- Canvas load respects container_mode when assigning parentId/extent
Pouzor