Repo-tracked context so a fresh Claude Code session (this machine or a
remote/cloud one) gets accurate project context without relying on this
session's local, machine-specific memory: repo conventions (no
co-author trailers, flag copyleft deps explicitly, scope security
gates broadly -- each backed by a real past incident), testing/deploy
workflow, and pointers into the existing docs.
docs/architecture.md's sequence diagram and boot-flow text still
described the pre-widget-system single-photo-queue model (e.g. "force-
advance to next queued photo") even though that was fully replaced by
the widget system across this branch's recent history -- fixed, and
added docs/widgets.md distilling the widget system's actual design
(data model, grid placement, compositor, button-action dispatch) as
current-state documentation, including the still-open Phase 6 cleanup
(legacy Frame columns not yet dropped) as a known gap.
Two widgets of the same type (e.g. two Photos widgets) both showed up
as plain "Photos" with no way to tell which was which. The buttons API
now includes each widget's grid placement plus the frame's grid dims;
the UI derives a rough position ("top-left", "right", etc.) from that
and only appends a number+position suffix when a type actually
repeats on the frame, leaving the common single-widget-per-type case
unchanged.
Whiteboard rendering isn't fully reliable yet -- tag it (alpha)
everywhere it's user-facing (widget label, add-widget button, dialog
title, Settings' WebDAV section) via one shared WIDGET_LABELS map
(moved to common.js so both the Layout canvas and the new Configuration
tab section can use it).
Button assignments: a new "Button assignments" card on the
Configuration tab lets you assign an ordered list of (widget, action)
bindings to each physical NEXT/BACK button -- add/remove/reorder, all
autosaved. Backed by new GET/PUT /api/frames/{id}/buttons endpoints;
PUT replaces a button's whole list in one atomic, fully-validated call
rather than separate add/remove/reorder endpoints. Device-side
consumption already existed (routers/device.py's _run_button_actions);
this is the UI for what was previously only reachable via the default
migration mapping.
Small header thumbnail showing exactly what the frame is currently
displaying -- same widget compositor /frame/image uses, handed back as
a plain PNG (image_pipeline.render_panel/render_placeholder gain an
as_png option) instead of packed native-panel bytes. New session-authed
GET /api/frames/{id}/preview exposes it; click-to-refresh plus a slow
60s poll on the frame header so it doesn't hammer Immich/calendar
sources just for a header thumbnail.
Once set, device_token_ack permanently locks out id-only requests
(auth.require_device) -- fine for a device that still has its token,
but a reprovisioned device has wiped its own token locally and had no
way back in short of a manual DB edit. The device's captive portal
always redirects the phone to /claim?device_id=... after
(re)provisioning, so reopen the handshake window there instead,
scoped to a logged-in owner/linked user of that frame.
showStatus() always wrote to the page-level #result div, which sits
behind a <dialog>'s backdrop -- a save inside a widget's gear-icon
dialog produced a message the user couldn't see without closing the
dialog first. It now prefers a .dialog-result element inside whichever
<dialog> is currently open, falling back to #result everywhere else.
The dialog is a flex column with its body scrolling independently so
.dialog-result stays pinned as a visible footer regardless of scroll
position -- otherwise a save message on a long form (e.g. the calendar
dialog) could land off-screen below the fold with no visible feedback
at all.
The dialog calls /albums via window.FRAME_API, but that's repointed to
this widget's own API base while a dialog is open (see frame_layout.js)
-- /albums is frame-level (lists the owner's whole Immich library, not
scoped to one photo widget), so it needs window.FRAME_BASE_API instead.
The button silently 404'd since the resulting URL never existed.
Replaces the Photos/Calendar/Whiteboard tabs with a single Layout page
(now the frame's landing route) where each widget gets a gear icon
opening a dialog scoped to that specific widget's own settings. This
was the missing piece for genuinely independent same-type widgets --
"the Calendar tab" never made sense once a frame could hold more than
one calendar widget with different settings.
Data layer: FrameCalendar re-keyed from frame_id to widget_id, so each
calendar widget has its own independent included-calendars set. The
rekey runs as an unconditional post-startup step (like the existing
widget backfill), not a numbered migration -- it depends on calendar
widgets already existing, which themselves come from that same
backfill step, not from schema migration. Registering it as a numbered
migration would have run it first during a real upgrade, silently
dropping every row; caught by a new test that exercises the raw-SQL
upgrade path instead of the fresh-install create_all() shortcut every
other migration test takes.
API layer: every endpoint that used to assume "the frame's widget of
this type" (photo queue/thumbnail/preview, calendar select/color/
tasks/weather, whiteboard source/browse/preview) moved into
api_widgets.py under /api/frames/{id}/widgets/{widget_id}/..., with a
new require_widget_view/control dependency pair mirroring the existing
frame-level ones. Device status (battery/last-seen/firmware) got its
own frame-level /status endpoint, split out of the old photo-specific
/queue it used to piggyback on -- fixes the status bar going silently
blank on any frame without a photo widget.
UI layer: each widget type's existing settings markup/JS was ported
into a dialog partial + an explicit init/close function pair (the
content is now fetched and injected on demand, not loaded at page load
time). window.FRAME_API is repointed to the open dialog's widget-scoped
API base for its duration and restored on close; a separate
window.FRAME_BASE_API stays stable for the always-present header/
status-bar scripts.
Caught during manual browser testing: the consolidated config-save
endpoint initially expected a JSON body while the copied-over dialog JS
posts form-urlencoded data (the old convention) -- fixed to match, with
new HTTP-level test coverage that would have caught it immediately.
Adds a deploy job to the existing build-and-push workflow: SSHes into
the deploy host as a dedicated espressoframeuser account and runs
docker compose pull && up -d. Runs only after build-and-push succeeds,
using a key/host pulled from repo secrets (DEPLOY_SSH_KEY, DEPLOY_HOST,
optional DEPLOY_PORT).
The placement canvas sized itself with CSS aspect-ratio plus
percentage widths/heights on the widget boxes. aspect-ratio isn't
supported on every mobile browser this app gets viewed from, and
without it a percentage height on an absolutely-positioned child
collapses to 0 against an indeterminate-height ancestor -- the canvas
rendered with no visible size at all, just unstyled labels and
buttons floating in normal document flow. Confirmed working on
desktop, broken on the reporter's phone.
Canvas and widget-box geometry is now computed and applied in pixels
from JS (measuring the wrap's own width, deriving height from the
grid's row/col ratio), recalculated on window resize. Also gives
.widget-box's color-mix() background a plain-color fallback for the
same class of older-browser gap.
Adds the actual "Android home screen" placement experience: a new
Layout tab with a pointer-driven canvas for dragging/resizing widgets
and adding new ones from a type picker. Backed by a new
routers/api_widgets.py (create/move/delete), which re-validates
bounds, minimum footprint, and no-overlap server-side regardless of
what the client already checked. A widget added without an explicit
position lands in the first open space that fits it
(grid.find_open_rect), so users don't have to hunt for empty space
themselves.
Also fixes a real latent bug this surfaced: changing a frame's
orientation swaps the widget grid's long/short axis, which left
existing widget placements out of bounds on the new grid with no
render-time safeguard. Orientation changes now reset the layout to a
single full-panel widget (keeping the first widget's type, dropping
the rest), with a client-side confirm before it happens.
calendar_render.py's _build_* functions now take a real target box
and pick font sizes/margins from three discrete size tiers (nearest
pixel-area fit) instead of always laying out at full panel size and
resizing after the fact -- a calendar widget placed smaller than the
full panel gets an actually-legible layout instead of shrunk text.
Month view falls back to agenda below the smallest tier, where 7
columns can no longer stay readable.
The old photo-inlay split (inlay_region/_content_region/_paste_inlay)
is deleted along with it -- arbitrary widget placement already
subsumes what a fixed half-panel split did, and every call site has
passed photo_inlay=None since the Phase 2 cutover.
Also adds HTTP-level test coverage for GET .../preview/calendar,
which had none before this -- it's what caught a stale photo_inlay
kwarg left over from the _build signature change that would have
TypeError'd on every request.
device.py's mode-keyed dispatch is replaced by a real compositor:
load a frame's widgets, compute pixel rects via app/grid.py, render
each through its widget module, and composite with render_panel.
Physical NEXT/BACK buttons now execute each frame's assigned
FrameButtonAction rows instead of one hardcoded per-mode action.
api_frames.py, manage.py, and common.py's build_manage_content are
repointed to read/write the frame's widget config rows instead of
the old Frame columns, and every settings page (Photos/Calendar/
Whiteboard tabs) now pre-fills its form from the same widget config
the write endpoints actually save to -- previously the read and
write sides would have silently diverged. The old mode selector and
photo-inlay checkbox are removed along with their now-inert wiring;
arbitrary widget placement subsumes what the fixed inlay split did.
Ships together with Phase 1 (per-type render/action modules) since
splitting the read/write cutover across deploys would have left
settings changes with no visible effect.
New app/widgets/ package (photos.py, calendar.py, whiteboard.py, plus
the WIDGET_TYPES registry) -- the widget-system analogue of
routers/device.py's old RENDERERS/ADVANCE_RENDERERS/BACK_RENDERERS,
generalized from "one mode owns the whole panel" to "each widget renders
into its own region and responds to named button actions." Each module
exposes render(db, frame, widget, target_w, target_h) -> Image.Image
(never raises -- a widget's own fetch hiccup falls back to a small
placeholder rather than taking the whole panel's render down) and an
ACTIONS registry for NEXT/BACK button assignment.
Supporting changes needed to give the widget modules something to call,
all mechanical/behavior-preserving for every existing caller:
- image_pipeline.py: render_panel(regions, ...) generalizes render_frame's
tail (paste, enhance once, overlay once, quantize once, pack once) from
one photo to N regions -- not a restructuring, since calendar mode's
photo-inlay feature already pastes a second composed image onto the
canvas before that single shared pipeline runs.
- photo_queue.py: advance_forced/back_forced/remove_from_rotation/
get_current take an explicit `frame` param now that `cfg` won't always
be the Frame itself once photo-queue state moves to PhotoWidgetConfig
-- caught a real latent bug while doing this: get_current was reading
refresh_interval_s off `cfg`, but that's a frame-level wake-cadence
setting, not something that becomes per-widget, so it now reads that
off `frame` explicitly instead.
- routers/common.py: list_assets/fetch_source_and_faces take album_id/
display_mode directly instead of a whole Frame (both only ever read
that one attribute off it); new get_or_refresh_*_for_widget siblings
of the existing calendar/weather/tasks/whiteboard cache helpers, read/
writing the new per-widget config tables -- the Frame-scoped originals
are untouched and still what routers/device.py's actual dispatch calls
until the Phase 2 cutover.
26 new tests (95 total): render_panel size/placement/orientation
coverage, and per-widget-type render/action tests (unconfigured ->
placeholder, a fetch failure -> placeholder not a crash, actions mutate
the right state). Full suite passes; diff-reviewed to confirm
device.py's actual RENDERERS dispatch and the old Frame-scoped
get_or_refresh_* bodies are unchanged, so this is safe to deploy on its
own despite being step 1 of a two-step cutover (see the project plan on
why the *next* step, not this one, has to ship atomically).
First step of replacing Frame.mode (one renderer owns the whole panel)
with an Android-home-screen-style widget system -- a frame will hold N
independently placed/sized widgets (photos/calendar/whiteboard), each
with its own config/state, plus fully user-assignable NEXT/BACK button
actions. Full plan at .claude/plans/prancy-snacking-iverson.md.
This phase is additive only and changes no existing behavior -- nothing
reads these new tables yet:
- models.py: Widget (placement) + PhotoWidgetConfig/CalendarWidgetConfig/
WhiteboardWidgetConfig (per-type 1:1 extension tables, matching this
codebase's existing convention of dedicated tables for naturally-scoped
state rather than one wide table) + FrameButtonAction (ordered
(widget, action) bindings per physical button).
- grid.py: pure snap-to-grid placement math, defined relative to the
panel's long/short axis so it stays valid across
logical_render_size(orientation)'s genuine width/height swap for
portrait, not just a rotation applied at the end.
- db.py: widget_locked(), the widget-scoped equivalent of frame_locked()
-- deliberately still locks at frame granularity (not a new per-widget
lock) to avoid a new class of multi-lock deadlock bugs.
- migration.py: _migration_16 creates the new tables; a separate
_ensure_widgets_backfilled() (ORM-based, not raw SQL -- much less
error-prone for this much per-mode branching) gives every existing
frame a widget reproducing its exact current mode/settings, so
upgrading changes nothing about what a frame displays or what its
buttons do. calendar_photo_inlay frames specifically get two widgets
(calendar + photo, split like the old inlay did) rather than silently
losing the photo half.
10 new tests covering fresh-install backfill, re-run idempotency, the
photo-inlay two-widget case, whiteboard's check_now button mapping, and
migration_16's actual CREATE TABLE path against a simulated pre-existing
database (not just the fresh-install create_all() shortcut). Full suite
(69 tests) passes.
Root cause of "whiteboard stuck on old content, no errors anywhere":
the Node sidecar had crashed at some point and, since it was just a
bare backgrounded process with nothing supervising it, stayed dead
permanently. Every refresh since then hit connection-refused, which
get_or_refresh_whiteboard treats as a soft failure and falls back to
the last successfully cached image -- so it looked exactly like a
caching bug from the outside, silently, forever, with no error visible
anywhere except a crash trace that had already scrolled out of the log
buffer.
Wrap it in a restart loop instead of a bare `&` so a future crash (a
still-unknown third jsdom/Excalidraw edge case, most likely) is a
few-second hiccup instead of a silent permanent outage.
Whiteboard and tasks-source save/clear used to just tell the user to
reload the page to see the change. Both endpoints always assign a
successful "set" to the calling user, so the new state is fully known
client-side already -- rewrite the "Currently using/showing ..." block
in place instead, no server round trip or reload needed.
Also adds .gitea/workflows/firmware-build-check.yml: builds both board
variants (devkit, xiao) on every push touching firmware/**, unlike
firmware-release-build.yml which only builds on a version.txt bump.
Verified both builds succeed locally against the actual ESP-IDF
toolchain before wiring this in.
64 tests covering: auth/setup and the CSRF gate, the "owner adds their
own data, anyone linked can mute it" permission pattern shared across
calendar-select/tasks-source/whiteboard-source, migration correctness
(fresh install, idempotent re-run, expected columns), battery estimate
outlier rejection, calendar_feed's fetch/merge/partial-failure handling,
webdav_client's fetch/list-directory, the whiteboard force-refresh
throttle bypass and browse endpoint, and render-size invariants across
calendar views/orientations.
No DB/HTTP fixtures need Docker, Node, or a real Immich/CalDAV/WebDAV
server -- a fresh temp SQLite file plus a couple of small local HTTP
servers as test doubles cover it all. Table data is wiped and reseeded
between tests rather than relying on SQLAlchemy's transaction-rollback
isolation pattern, which needs a pysqlite event-listener workaround
app/db.py's engine doesn't have and has no reason to gain just for tests.
Wired into .gitea/workflows/server-docker-build.yml as its own job that
build-and-push now depends on, so a failing suite blocks the image push
rather than just running alongside it for show.
File picker: an optional WebDAV browse root in Settings
(User.webdav_base_url) plus a plain-PROPFIND directory listing
(webdav_client.list_directory) power a "Browse..." panel on a frame's
Whiteboard tab, so a file can be clicked into rather than typing its
exact WebDAV URL. Manual URL entry still works unchanged either way.
Force-refresh: get_or_refresh_whiteboard takes a force flag that skips
the fetch throttle entirely; the Whiteboard tab's refresh button now
passes it, so clicking it always re-fetches and re-renders instead of
possibly just re-showing the same cached image from within the last
~20 minutes.
exportToSvg's font-embedding path (base64 @font-face rules) goes through
the browser FontFace API, which jsdom doesn't implement -- that's what
crashed fontFacesStylesGenerator after the previous global-shimming fix
got past startup. skipInliningFonts avoids that path entirely; resvg
already falls back to system fonts for rasterizing regardless, so
embedded fonts were never going to affect the final PNG.
@excalidraw/utils is a browser bundle -- it references devicePixelRatio,
location, matchMedia etc. as bare globals the way inline <script> code
would, not as window.foo. Only copying window/document/navigator onto
Node's global left everything else undefined, so the very first bare
reference threw a ReferenceError as soon as the module loaded. Copy
jsdom's entire window onto global instead, enable pretendToBeVisual so
jsdom actually populates devicePixelRatio/requestAnimationFrame, and
stub matchMedia since jsdom doesn't implement it at all.
The 413 from git.thumeit.com was on a single blob PUT, not the whole
image -- split the Node.js apt install and the render-service npm
install into several smaller RUN layers (purging curl/gnupg in the
same layer they're installed in, cleaning npm's cache after each
package) so no single pushed blob is as large as before.
Docker defaults to HTTPS for any non-docker.io registry, and
10.0.0.246:3000 (a bare LAN IP:port) isn't serving valid HTTPS -- the
push kept going out over HTTPS and failing. Adds a buildkitd config
telling BuildKit specifically to use HTTP for that host.
This only fixes the actual build+push step (BuildKit). The "Log in"
step runs a plain `docker login` through the classic Docker CLI/daemon
instead, which doesn't read this config -- that one still needs
10.0.0.246:3000 added to "insecure-registries" in the actual Docker
daemon's /etc/docker/daemon.json on whatever host runs the Gitea
Actions runner, followed by a daemon restart. That's runner-host
infrastructure outside this repo.
Cloudflare (fronting git.thumeit.com) started rejecting image pushes
with 413 Payload Too Large once whiteboard mode's Node runtime + native
resvg bindings made the image significantly bigger than before. Switched
the build/push target to a LAN-local registry (10.0.0.246:3000) that has
nothing in front of it to hit that limit, and updated
docker-compose.yml.example to match.
New third mode alongside photos/calendar: fetches a .whiteboard file
over plain WebDAV (Basic auth -- generic, not Nextcloud-specific) and
renders it via a small Node.js sidecar using Excalidraw's own real
export code (@excalidraw/utils + @resvg/resvg-js, no headless browser),
since a .whiteboard file turns out to be Excalidraw scene JSON, not an
image. The sidecar runs as a second process inside this same container
(Dockerfile installs Node, start.sh backgrounds it before exec'ing
uvicorn) rather than a separate docker-compose service -- lightweight,
stateless, reachable only at 127.0.0.1 from the Python process, nothing
worth independently scaling.
The rendered PNG is treated exactly like a photo from there on --
composed/quantized through the existing image_pipeline (letterboxed,
never cropped) rather than a second parallel rendering pipeline.
WebDAV credentials support the common "it's actually the same Nextcloud
account as my CalDAV" case (an explicit opt-in checkbox, not silently
inferred) while still working with any WebDAV server generically.
Frame-level source (URL + owning account) follows the same owner-
controls-their-own-data permission split as calendar sources and the
week view's task list: only the account owner can point a frame at it,
anyone linked can clear it.
Honest limitation: this environment has no Node.js/npm, so
render-service/ is written carefully against each library's documented
API (verified via the npm registry, including transitive dependency
licenses after the CalDAV/AGPL surprise earlier this session) but has
never actually been executed. First real docker build is the first
true test -- see render-service/README.md.
Both are photos-specific settings (rotation order, how a photo's aspect
ratio gets reconciled with the panel), not device-wide configuration --
they belonged on the Photos tab next to the album/queue settings, not
Configuration. No backend change: both already save through the shared
partial-update /config endpoint regardless of which tab's form sends
them.
calendar_week_start's fixed-weekday anchor ("start on the most recent
Monday") stops making sense once the view isn't a literal calendar
week, so a non-7-day week view now starts calendar_week_start_offset
days from today instead (0 = starts today, negative/positive = past/
future) -- calendar_week_start still governs at the default 7 days,
unchanged.
Also hides the Calendar tab's week-only fields (days to show, layout,
start offset) unless View is actually set to Week, and further hides
the new start-offset field specifically when Days to show is 7 (where
it has no effect). "Week starts on" stays visible for Month too, since
it actually still applies there.
- Day count (2-10, was fixed at 7) -- 5 days trims the weekend clutter
without losing the grid format.
- Layout choice: days side by side (original behavior) or stacked
vertically as full agenda-style sections (reuses _draw_agenda_day,
same approach _build_today_tomorrow already used for a fixed 2 days).
- Optional task list (CalDAV VTODO collections only -- a plain ICS
subscription doesn't meaningfully have one) that takes the space of
one day slot instead of adding an extra one. Same owner-controls-
their-own-data permission split as calendar sources: only the
calendar's owner can point a frame's task list at it, but anyone
linked to the frame can clear it.
Browse-offset paging now moves by N days (was hardcoded to weeks),
identical to the old behavior when days=7. Changing the day count
resets the browse offset, same reasoning as changing views already did.
A single noisy ADC/regulator glitch (see firmware/main/battery.c)
survives the existing recharge filter: whichever way it reads, one of
the two steps around it (into a dip, or out of a spike) still looks
like an ordinary drop and got averaged straight into the remaining-
time estimate, letting one bad reading swing it dramatically.
Added a MAD-based modified z-score outlier check on top of the
existing recency-weighted average. Had to special-case the standard
MAD degenerating to exactly 0, which happens whenever more than half
the steps share the same value -- the norm for battery data (most
wakes cost the same small integer percent), and exactly the shape a
single spliced-in glitch among a steady discharge rate has, so the
naive case would have let the outlier this is for sail straight
through. Falls back to mean absolute deviation there instead.
Verified against constructed glitch scenarios in both directions
(spurious dip and spurious spike): estimate now comes out identical
to the same series with the glitch removed entirely.
Switched from the monochrome emoji font to color: NotoColorEmoji's
embedded CBDT bitmap glyphs, rasterized once at their native 109px
size and scaled to the target row height (unlike normal vector text,
color bitmap glyphs aren't stored at arbitrary sizes). Confirmed by
rendering an actual agenda row through the real quantizer that the
dithered-to-6-color result still reads clearly, not just muddy noise.
Falls back to the monochrome font if a deployment's Pillow/FreeType
wasn't built with embedded color bitmap support, so this degrades
instead of crashing or showing nothing.
The earlier fix stripped emoji instead of rendering them, which wasn't
what was asked for. Event titles now draw with two fonts: the usual
default font for text, and a vendored monochrome emoji font (Noto
Emoji, OFL-1.1) for actual emoji runs, so they show up as real glyphs
instead of a tofu box or nothing at all. Monochrome rather than color,
since reliably rendering COLR/CBDT color glyphs depends on how Pillow's
FreeType was built -- not something to depend on across deployments.
Also: events sharing the exact same title and time across different
calendars (e.g. a shared family event synced onto more than one
person's calendar) now collapse into one row instead of showing twice,
with a color bar split between every contributing calendar so it's
still clear whose event it is.
ImageFont.load_default() has no emoji glyphs, and PIL/FreeType don't
skip an unsupported codepoint -- they substitute a visible ".notdef"
tofu box, which read as a rendering glitch rather than "not supported."
Stripped instead (covers the standard emoji Unicode blocks, skin-tone
modifiers, and the ZWJ used to combine them into one glyph), with
surrounding whitespace collapsed.
Each linked person can pin one of the panel's four non-black/white
colors (Yellow/Red/Blue/Green) to their own calendar instead of relying
on calendar_render.py's old auto-cycle-by-owner-name order -- owner-only,
like adding a calendar in the first place. Colors resolve against
whichever palette a frame actually renders with (including a custom
Advanced configuration override), so a pinned "Blue" stays this frame's
actual blue. Event color bars/dots are also bigger and rounded now
across agenda/week/month views, easier to tell apart at a glance.
Weather: multiple cities per frame, geocoded via Open-Meteo (no API
key), shown above the event list on agenda/today & tomorrow/week views
-- never month, no room for it there. Hand-drawn sun/cloud/rain/snow/
thunderstorm icons (no new font/icon asset, same primitives-only
approach the rest of calendar_render.py already uses). City geocoding
handles "City, State" qualifiers Open-Meteo's own search doesn't
(disambiguates same-named cities, e.g. the three "Portland"s).
CalDAV fix: events were never showing despite calendars discovering
fine, because fetch_calendar_events relied on the calendar-query
REPORT's server-side time-range filter, which real servers implement
inconsistently (confirmed against a real server, not just guessed --
reproduced locally with Radicale). Switched to fetching every event
unfiltered and doing all date-window filtering/expansion client-side,
same approach already used for plain ICS feeds.
Calendar rule/grid lines were light gray, which dithers away to
near-invisible on the 6-color e-ink palette -- now black.
CalDAV accounts (Nextcloud, Fastmail, iCloud, ...) can now be linked
alongside the existing single ICS subscription, since one account can
expose several calendars. A frame's Calendar tab now lists calendars
per person rather than one opt-in per person: your own row shows every
calendar you have available with a full add/remove toggle, while other
linked users' rows show only calendars they've included, toggleable
off (mute) but not on -- only a calendar's owner can add it to a
shared frame. FrameCalendar replaces the old single-boolean
UserFrame.calendar_included; existing opt-ins are migrated forward.
The old estimate used a single linear percent/second rate from the
current discharge cycle's battery_history, which resets to empty on
every recharge -- so "not enough data yet" kept showing up despite the
frame having plenty of history overall, and the rate it did compute was
tied to whatever refresh interval produced it (changing the interval
didn't move the estimate until enough new history accumulated under
the new setting).
Now pulls the last 100 rows from the permanent battery_log table
instead, and averages the *per-wake* percent drop (not per-second) --
recharge jumps are skipped rather than counted as negative drain,
flat/zero-drop wakes still count so the rate isn't overstated, and
more recent steps are weighted more heavily. The per-wake rate then
converts to wall-clock time using the frame's current
refresh_interval_s and quiet-hours settings, so halving the refresh
interval roughly halves the estimate immediately, and quiet hours
correctly stretches it out (fewer wakes/day at the same per-wake cost).
Frame name (pencil-icon inline edit) and the Photos/Calendar mode
selector now live in the page header, shared across all four per-frame
pages instead of being buried in the Configuration form -- so renaming
a frame or flipping its mode no longer requires navigating to a
specific tab first.
Calendar settings move out of a conditionally-hidden card on the
Configuration tab into their own dedicated tab (new /frames/{id}/
calendar route), visually greyed out when the frame is in Photos mode
but still fully usable so calendar settings can be configured ahead of
switching modes. Also wires the week-start setting into the UI for the
first time (the column/backend support landed earlier but had no
control anywhere).
Responds to post-launch feedback on calendar mode: configurable
week-start day for week/month views, crisper non-antialiased text
(threshold-masked instead of drawn straight, so Floyd-Steinberg
dithering doesn't speckle glyph edges), a color-coded/proportionally
filled battery icon on the manage overlay, word-wrapped placeholder
text so "Calendar isn't set up yet" no longer clips in portrait, photo
inlay support extended from agenda-only to every view, and a fix so
manage-overlay face labels reposition correctly when a photo inlay is
active (they previously assumed the photo filled the whole canvas).
Also adds a fourth calendar view, "Today & Tomorrow" -- a two-day
agenda that reuses the same per-day row-layout helper the single-day
agenda view already has.
calendar_feed.py/calendar_render.py: fetch/merge per-user ICS feeds,
render agenda/week/month views. manage_overlay.py: composites the
manage-button overlay server-side (QR, battery, location/date,
share-QR, face labels), reused by every render mode. device.py/common.py
wire both together: mode dispatch for /frame/image+advance+back, and
the &manage=1 flag. Plus UI (frame_config.html Calendar card, settings
calendar URL field) and the icalendar/recurring-ical-events deps.
Deletes manage_qr_overlay.c/.h; frame_client.c's manage-button flow is
now one fetch with &manage=1 instead of on-device QR/text generation
plus separate photo-info/face-labels requests.
New columns, all ADD COLUMN with inert defaults -- no existing frame's
behavior changes until mode is explicitly switched to "calendar":
- users.calendar_ics_url: one personal iCal/CalDAV subscription per
user, same shape as the existing per-user immich_url/immich_api_key.
- user_frames.calendar_included: explicit per-(user,frame) opt-in,
default off. Being linked to a frame does not by itself contribute
your calendar to it -- each person's calendar is their own data to
share, not something a frame's controller decides on their behalf.
- frames.calendar_view/calendar_photo_inlay/calendar_browse_offset:
per-frame display settings and NEXT/BACK navigation state.
- frames.calendar_checked_at/calendar_cached_events/calendar_fetch_summary:
the throttled merge-fetch cache, same shape as the existing
firmware_update_checked_at/firmware_gitea_latest_version pattern.
A report was flagged as "the battery got recharged" (resetting
battery_history and stats_recharge_cycles, and re-arming the low-battery
alert) whenever it came in >= RECHARGE_JUMP_PCT above the single
immediately-previous report. That's exactly what a real recharge looks
like, but it's also exactly what a normal reading looks like right
after one noisy low report: e.g. 60, 59, 58, then a stray 53, then back
to a perfectly normal 58 -- 58 >= 53+5 falsely read as a recharge.
Now compared against the max of the last RECHARGE_LOOKBACK (3) reports
instead of just the one before it, so a lone stray reading doesn't get
to set the bar a normal reading then trips. A real recharge still needs
to clear all of them, so genuine recharges are still caught immediately
(verified: 18% -> 90% still triggers, history still resets).
Paired with the firmware-side battery.c change (trimmed-mean ADC
sampling) that reduces how often a stray reading like the 53 above
happens in the first place.
Sometimes a single reading came in noticeably off from the real trend
(a regulator/RF transient during sampling), and the next normal reading
would then look like a big jump relative to that bad one -- server-side,
enough to misfire the recharge-cycle heuristic (see the paired server
commit). Went from 8 raw-averaged samples to 16, sorted, with the 3
extreme samples on each end dropped before averaging the remaining 10 --
a handful of outliers can no longer skew the result the way a plain
average let them.
tfaour