Protecting every file of a live space also protected the ones a commit had just
superseded, and gave rejected uploads immortality. The distinction that matters
is narrower: a space with NO plan_url has had its image detached and may want it
back; a space that has one can only be holding its own rejects. Attachments:
staging folders keep the hour, marker folders get the month.
Also: the layout event test asserted the order of separately fired bus events,
which nothing promises — it came back [2,1,3] in CI.
Deployed v1.46.3 to my own instance, restarted, and the startup sweep deleted
both floor plans: config/houseplan/plans/ went from f1.svg + f2.png to empty.
The backup is a SecureTar, so they are gone.
The rule was wrong, not the code. v1.46.0 introduced collection that treats
'nothing references this right now' as abandoned and gives it an hour. But
detaching a plan — switching a space to 'draw' — is a normal, reversible action,
and the editor's own comment says the file stays on disk. Those two plans had
been detached for weeks; every pass since v1.46.0 was entitled to remove them,
and the one that finally ran did.
New rule, one for every path:
* superseded by a commit (was in the old revision, is not in the new) — goes
immediately; that is the one thing a commit knows for certain;
* belongs to a space or marker that still exists — never collected, at any
age, because unreferenced is not abandoned;
* a per-dialog staging folder (up_*) — one hour, unchanged: by construction it
only ever holds an upload from a dialog that was never saved;
* anything else — thirty days.
The flag I added an hour ago is gone with it: two rules for the same
question is how this happened. Tests updated to the new grace, plus two that pin
the distinction directly.
I am sorry about the files.
The startup sweep resolved its runtime data with get_data(hass), which lists
only LOADED entries — during async_setup_entry the entry is still
SETUP_IN_PROGRESS, so it always got None and degraded to removing streaming
temporaries. The real collection was then 24 hours away, and an instance that
restarts more often than that never ran it at all. It closes over the
object created a few lines above instead; the callback is unregistered with the
entry, so that matches the lifecycle.
The test that was meant to prove the previous fix passed for the wrong reason:
it seeded the strays BEFORE config/set, which collects too, so nothing was left
for the restart to find. Now seeded after the save, plus two more — one firing
the interval callback on its own, and one running a reload and a save
concurrently to assert the accepted config never references a file the sweep
removed (they share the write lock; this pins that they must).
Docs: CHANGELOG.md + CHANGELOG.ru.md + TESTING.md + STATUS.md.
HP-1461-01: collection was tied to config/set, which is the right scope for
what a commit supersedes but leaves a file nobody references with no future
write to notice it — cancel a dialog after the upload finished, drop the
connection just after, or call the upload API directly. The daily sweep added
in v1.46.1 only removed streaming temporaries, so the documented 'a cancelled
attachment is collected an hour later' did not hold on an instance nobody
edits. The scheduled pass now loads the stored config under the same write_lock
a commit uses and runs collect_attachments/collect_plans with it as BOTH sides:
nothing counts as superseded, referenced files are preserved, aged unreferenced
ones go. Doing it under the lock keeps it from deciding on a snapshot a commit
is about to replace.
HP-1461-02: _reloadLayoutOnly captured the dirty set AFTER flushing the pending
write, and the flush empties it first — so during a real drag (where a write is
already scheduled) the snapshot was empty and the server's older position was
merged over the user's move. The snapshot is taken before the flush, by value,
and a _sentPos map now holds positions that are sent but unacknowledged, which
closes the same window for a write that was already in flight.
Tests: the upload test now cancels the request task for real (the previous one
claimed to and only walked error paths); smoke_layout_sync schedules a genuine
debounced write and delays it — verified failing on a v1.46.1 build with
exactly the reported symptom; a new backend test reloads the entry and asserts
the scheduled sweep takes an aged cancelled attachment and an orphan plan while
keeping everything the config still references.
Docs: CHANGELOG.md + CHANGELOG.ru.md + ARCHITECTURE.md + TESTING.md + STATUS.md.
HP-1460-01: v1.46.0 stopped overwriting attachments, but picking a free name
and taking it were two steps. Two uploads racing between them agreed on the
same name, both answered 200, and one set of bytes replaced the other;
files/migrate had the same check-then-copy gap. reserve_filename now claims the
name with O_CREAT|O_EXCL as it picks it, and both paths use it. It also splits
the extension off the RAW name and budgets the stem against MAX_FILENAME
including the collision tag — a maximal name lost its '.pdf' and then grew past
the limit, so the view sanitised the request back to a different name and the
attachment 404'd for good.
HP-1460-02: cleanup lived in an 'except Exception', which CancelledError walks
past, only one tmp_path was tracked, promotion had no finally, and the
collector only walks marker folders — an aborted transfer stranded a .upload-*
that nothing would ever remove. An outer finally owns every temporary, a second
'file' part is refused, promotion failure cleans up, and sweep_upload_temps
runs at setup, daily, and inside the commit-scoped collector. Chunks are
batched to 1 MB per disk task instead of one per 64 KB.
HP-1460-03: the layout event reached the static card and not the full one, so
two full cards diverged until a reload. The full card subscribes now and
re-reads ONLY the layout, keyed on its revision. Two hazards handled: it
records revisions it produced itself, and the reaction is deferred ~200 ms
because the event can beat the reply to our own write over the same socket;
positions dragged but not yet sent are flushed and merged on top, so a fix for
a stale UI cannot become a lost drag.
Tests: smoke_layout_sync (fails on a v1.46.0 build), four pure tests for atomic
reservation incl. 20-thread concurrency and the length boundary, a backend test
walking every failing exit path of an upload, and — as the report asked — an
HA-harness test that a repair issue disappears with its space.
Docs: CHANGELOG.md + CHANGELOG.ru.md + ARCHITECTURE.md + TESTING.md + STATUS.md.
unique_filename produced 'manual (2).pdf'; HouseplanContentView sanitises the
name in the REQUEST too, turning ' (2)' into '_2_', so the file was written and
then 404'd. The same pattern was already in files/migrate, so a rebind that hit
a name collision has been producing dead links. Both use the shared helper now,
with '-2', which round-trips sanitize_filename — asserted.
HP-1454-01 (high, release blocker): an uploaded SVG plan opened directly is a
top-level document of Home Assistant's own origin, so a <script> inside it
reaches the session's localStorage and API. Uploading needs write access, which
by default every authenticated user has. SVG responses now carry a sandbox CSP;
only SVG, because a CSP on a PDF can break the browser's viewer and a raster
image has nothing to disable. Verified in Chromium both ways: the script runs
without the header and does not with it.
HP-1454-02: attachment uploads wrote straight to <marker>/<filename>, outside
the config transaction — a cancelled dialog or a rejected save left the stored
url serving new bytes, and every new icon shared one 'new' folder, so two of
them attaching manual.pdf pointed at one file. Uploads take a free name, a new
icon gets a per-dialog staging folder promoted on an accepted save, and
config/set collects superseded and aged-orphan attachments like it does plans.
HP-1454-03: the debounce spaced out the starts of a write, not the writes. A
save slower than 500 ms let the next edit go out with the same expected_rev;
the server accepted the first, rejected the second, and the conflict handler
reloaded over the local copy. Writes are chained now — one in flight, each with
the revision the previous returned.
HP-1454-04: _openPairsCache keyed on room ids and links only, so an aspect
change or a dragged vertex left open boundaries and their glow cuts at old
coordinates. It keys on the rendered model object now — the same invalidation
the model cache already has, not a second strategy. The fingerprint also gained
an O(1) geometry roll-up per room.
HP-1454-05: outer collections were capped, inner ones were not. Limits for
poly points, open_to, controls, pdfs, text and url lengths, plus a total
serialized size cap; legacy is dropped server-side.
HP-1454-06: upload streams to a temp file and downloads use FileResponse, so a
50 MB manual no longer costs ~100 MB of RSS per transfer.
HP-1454-07: spaceModels() dropped room.settings, so the static card ignored the
per-room fill override. HP-1454-08: layout had no revision on point-wise writes
and no event, leaving static cards stale forever; it now keeps a revision,
returns it and fires houseplan_layout_updated. HP-1454-09: repair cleanup only
walked existing spaces, so a deleted space kept its warning. HP-1454-10:
serialize-javascript pinned past two advisories.
Tests: smoke_svg_sandbox (proves both directions), smoke_config_writer and
smoke_render_parity (both verified failing against a v1.45.4 build), six pure
tests for attachment collection and inner limits, four HA-harness tests for the
CSP, non-overwriting uploads, the size cap and layout revisions.
Docs: CHANGELOG.md + CHANGELOG.ru.md + ARCHITECTURE.md + TESTING.md + STATUS.md.
R5-1: the backend signs each path independently and answers successfully with
whatever it managed, skipping (and logging) the rest. The card read any
successful call as 'the batch is done', cleared the backoff for every path in
it, then wrote only the urls that came back — so a path the backend kept
skipping was asked for again on every render, the exact amplification the
backoff was added to stop. A path now counts as signed only when the answer
carries a url for it; the others back off individually, keys that were not
requested are ignored, and onUpdate fires only when a new signature landed.
R5-2: docs/STATUS.md still described main as holding releases up to v1.40.1 and
quoted test counts several releases old, while the version line beside them was
kept current — a handoff reader got a wrong branch model and less coverage than
exists. Branch roles are now accurate, and the counts are gone rather than
corrected: scripts/inventory.mjs (npm run inventory) prints them from the tree,
so there is nothing left to drift.
Tests: three unit cases for empty/partial/foreign-key answers, verified to fail
against a v1.45.3 checkout; a backend test pinning the partial-success contract
by making async_sign_path raise for one path of two.
Docs: CHANGELOG.md + CHANGELOG.ru.md + TESTING.md + STATUS.md.
Version distance is not calendar distance. v1.26.0 shipped 2026-07-21 and the
report came in on 2026-07-27; the project itself is three weeks old. The point
stands and is unchanged — nothing in the suite could have caught it, because the
option list and the schema were written in two languages and never compared —
but the 'year and a half' was wrong.
The device editor has offered display='value' since v1.26.0; MARKER_SCHEMA
accepted only badge/ripple/icon_ripple. Picking it produced
not a valid value for dictionary value @ data['config']['markers'][n]['display']
and since one rejected marker fails the whole config write, the user could not
save the plan at all until the setting was undone. Reported by @RemyRoux with
the exact error text, 2026-07-27 — a year and a half after the feature shipped.
The schema now accepts it, and the class of bug is closed rather than the
instance: DISPLAY_MODES, TAP_ACTIONS, SPACE_FILL_MODES and ROOM_FILL_MODES are
exported from src/logic.ts, the editors render their options from them, and a
backend test parses those lists out of the TypeScript source and asserts the
schema accepts every one (and rejects a bogus value). Reverting the one-word
schema fix fails that test, which is the check that was missing.
Plus an HA-harness test saving a config that contains a value-display marker —
the exact call the user's card was making.
Docs: CHANGELOG.md + CHANGELOG.ru.md + TESTING.md + STATUS.md.
R4-1: collecting superseded plan files runs after the configuration is already
durable, but an error listing the directory propagated out of config/set. The
client saw a failure for a revision the server had committed, and its retry
came back as a conflict. collect_plans now reports 0 instead of raising, and
config/set logs and proceeds — the event fires, the revision is returned.
R4-2: the pending set was cleared when a batch went out, not when it came back,
so every render during an in-flight content/sign queued another request: six
calls where one was needed, and unbounded on a socket that is slow rather than
busy. Queued and in-flight are separate states now; a failure backs off (2 s
doubling to 60 s) instead of retrying on the next frame; an in-flight entry
expires after 15 s so a promise that never settles cannot wedge retries; a late
answer after dispose() no longer renders.
Tests: test/signing.test.mjs — eight cases with hand-settled promises, verified
against a v1.45.1 checkout where four of them fail (2 sign calls instead of 1,
no backoff, a late answer rendering after teardown). Backend: a broken
collector still yields a successful save whose revision the next CAS accepts.
Pure collector: a disappearing directory returns 0.
Docs: CHANGELOG.md + CHANGELOG.ru.md + ARCHITECTURE.md + TESTING.md + STATUS.md.
R3-1 (high): v1.45.0 made the upload safe but left deletion to the client —
after a successful save the card asked the backend to remove everything but the
file it had just committed. Two open editors cannot be ordered: a delayed
request from one deleted the plan the other had just saved, leaving the
accepted configuration pointing at nothing, the exact damage copy-on-write was
introduced to prevent.
houseplan/plan/cleanup is removed. config/set collects inside its own write
lock from the two configurations that bracket the commit (plans.collect_plans):
a file the old revision referenced and the new one does not is superseded and
goes; any other unreferenced upload waits out PLAN_ORPHAN_TTL_S, because a
fresh one may belong to a transaction that has not committed yet. The collector
lives in a pure module so it can be reasoned about and unit-tested without the
HA harness.
R3-2: houseplan-space-card signed its plan url and threw the result away —
getCardSize() mutated a throwaway model while render() rebuilt its own from the
config, so the <image> requested the protected path and got 401 on every
render. Both cards now share ContentSigner (src/signing.ts), which also gives
the static card batching, expiry handling and periodic re-signing. is
released in finally: one failed request no longer wedges a url for the life of
the page.
Tests: five backend interleaving cases from the report, six unit tests for the
pure collector, smoke_space_card_bg (verified to fail against a v1.45.0 build:
the raw url reaches the DOM and no retry happens). 57 smokes, 124 unit, 22
backend-pure.
Docs: CHANGELOG.md + CHANGELOG.ru.md + ARCHITECTURE.md + TESTING.md + STATUS.md.
R2-1 (high): plan replacement committed filesystem state before the config CAS.
The upload wrote the final name and unlinked the other extension, so a rejected
config write left the live plan already replaced — or the stored config
pointing at a deleted file. Uploads now go to <space>.<token>.<ext> and delete
nothing; houseplan/plan/cleanup runs only after the config write is accepted.
The '.' separator is load-bearing: a space id cannot contain one, so cleaning
'f1' can never reach the files of 'f1-attic'.
R2-2: the backend signs at most MAX_SIGN_PATHS (200) per request and ignores
the rest silently, while the card sent its whole cache in one call and trusted
any cached entry forever — past 200 attachments the later ones stopped being
refreshed and expired for good. Requests are chunked to the shared constant,
entries carry their issue time (aging urls keep rendering while a replacement
is fetched, expired ones are dropped), and the cache is pruned to urls the live
config still references.
R2-3: areaClimate() rescanned the whole registry per room and per measurement.
areaClimateMap() classifies once and returns Map<area,{temp,hum}>, memoized on
hass identity so fresh states are always observed. Smoke measurement: 133
registry scans per update with 44 rooms before, 2 after, flat in room count.
Also: smoke_ux_fixes wrote its screenshot to a hard-coded /tmp path and could
not run on Windows.
Tests: smoke_plan_upload_reject, smoke_sign_cap, smoke_climate_once (all fail
on v1.44.8), three backend tests for versioned plan names and cleanup scoping,
unit tests for chunk/referencedContentUrls and areaClimateMap.
Docs: CHANGELOG.md + CHANGELOG.ru.md + ARCHITECTURE.md + TESTING.md + STATUS.md.
Found on the owner's install: the image lands in /config/houseplan/plans, the
space keeps plan_url=null, the plan never shows and re-saving does not help.
_saveSpaceDialog held a reference to the space object across the await that
uploads the file. _reloadConfigOnly() — which runs on every
houseplan_config_updated event — REPLACES _serverCfg, so that reference became
an orphan: plan_url, aspect, title and every display setting were written into a
detached object while the save shipped the untouched config. In 'create' mode
the whole new space was lost the same way.
- upload first, then touch the config; no reference is held across an await.
- _saveConfigNow() sets _cfgWriting like the debounced writer, so a revision
arriving mid-save defers its reload instead of replacing the config (audit L2
extended to this path).
- demo/smoke_plan_upload_race.mjs: on v1.44.7 the sent config still carries the
OLD plan_url and the created space is missing; passes here. The demo's
config/get now returns a fresh object, as a real server does — returning the
same reference is what hid this class of bug from the smoke layer.
- DEVELOPMENT.md: the deploy target is custom_components/houseplan/frontend/,
and deploy verification must go over HTTP. A copy placed next to __init__.py
is served by nobody — that cost two deployments today.
- docs: CHANGELOG.md + CHANGELOG.ru.md + TESTING.md + STATUS.md.
The card signs content urls because a browser cannot authenticate an <image
href>. But _display() was called inside _buildModel(), and the space model is
memoized on the config fingerprint — so the UNSIGNED url froze in the cache and
the signature, which did arrive, never reached the element. The plan never
loaded, and the browser kept hitting the unsigned path: 401, which Home
Assistant reports as a failed login attempt from the viewer's own IP (that is
how the owner spotted it). PDF links were unaffected: they already resolved at
render time.
- _buildModel() keeps the raw plan_url; the render pass calls _display().
- _display() returns '' for an unsigned content url instead of the plain path,
and the <image> is not emitted at all until the signature lands — no 401, no
spurious login-attempt warning.
- _resign() replaces 'drop everything and re-request': the previous urls are
kept until the new ones arrive, so a wall tablet never blanks.
- demo/smoke_plan_signed.mjs: reproduces on v1.44.6 (href stays ?v=..., never
?authSig=), passes here. TESTING.md row added.
- docs: CHANGELOG.md + CHANGELOG.ru.md + STATUS.md.
After v1.44.5 read the area registry instead of visible icons, every hidden
temperature entity in the area became a candidate, including ones measuring
something other than room air. Verified against a live 60-area install: a NAS
processor temperature, kettle water, a 90 C sauna heater and a virtual
better_thermostat all leaked into room averages.
- areaClimate(): skip entity_category (diagnostic/config), skip EXCLUDED_DOMAINS
platforms, skip entity ids naming a non-air medium (water/coolant/flow_temp/
return_temp/target/setpoint/chip/cpu/processor/board/device_temp/batter/
freezer/fridge/oven/kettle/boiler).
- rules.ts: kettle/thermopot -> mdi:kettle, sauna/harvia -> mdi:hot-tub, so they
no longer fall through to the generic thermometer rule.
- test: all four real false positives asserted out, one real sensor left.
- docs: CHANGELOG.md + CHANGELOG.ru.md + STATUS.md snapshot.
- areaClimate() walks the HA registry for the area instead of the list
of VISIBLE icons: a thermometer hidden by curation or by the user was
silently dropped from the room card, tooltip and temperature fill
(field report). Curation still filters fridges/TRVs; the auto icon is
used on purpose so a custom marker icon cannot change what a device
measures; an explicit per-room source still wins
- room tooltip no longer says 'open the area' — room clicks were removed
in v1.40.1 (the link icon does it)
- +1 unit test (120); both changelogs updated
B2: the HTTP upload view failed OPEN when the config entry was
unavailable while the WS path failed closed — both now share one
may_write() policy helper (new auth.py) that denies non-admins when the
policy cannot be read.
B5: _finite now guards room rects, polygon vertices, view_box and
opening coordinates, not just layout positions; the declared
MAX_OPENINGS cap is finally enforced.
L4 (sub-item): every drag pipeline captures the pointer through the
tolerant helper (an inactive pointerId used to kill device/label/resize
drags); decor shapes gained a bounds clamp so they cannot be dragged far
outside the plan and persisted there.
+2 backend tests (16); both changelogs updated in this commit
- 10 most recent releases translated; older entries stay English-only
- policy updated in STATUS.md and CONTRIBUTING: user-visible changes go
into BOTH changelogs in the same commit (the user base is largely
Russian-speaking — see the Telegram chat)
- cross-links between the two files and from both READMEs
Matysh