The AI-parsing base URL and model inputs had no autoComplete, so a
browser password manager could drop the saved login email into the
base URL field. In the admin addon config onBlur then fired a model
lookup against e.g. "admin@trek.local", which the server rejected
with 400. Mark the base URL and model inputs as type=url /
autoComplete=off in both the admin addon config and the per-user
connection section.
The Dockerfile copies server/reset-admin.js (the admin recovery
script), but .dockerignore also listed it, so it was stripped from
the build context and the image build failed with a not-found error.
Drop the ignore entry so the COPY resolves again.
Adds unit coverage for saveUnsplashCover (host check, content-type
and size limits, download failure), the searchUnsplashPhotos error
and success paths, and the PUT handler internalising a hot-link.
Updates the existing PUT tests for the now-async handler.
Previously a selected Unsplash photo was stored as a remote
images.unsplash.com hot-link, so covers broke offline and on link
rot. The trip PUT handler now fetches the picked image through the
SSRF guard and saves it under uploads/covers, rewriting cover_image
to the local path (502 if the download fails). Also debounces the
cover search so a slow earlier request can no longer overwrite newer
results, drops a dead userId parameter, and reverts an unrelated
vite proxy change.
On mobile the bottom-nav "+" always created a new place (except on the Costs tab,
where it added an expense). It now matches the active trip tab: Bookings adds a
reservation, Transports adds a transport, Costs adds an expense, and everything
else (Plan, plus tabs that have no create modal — Lists / Files / Collab) keeps
adding a place.
Follows the existing ?create=<intent> pattern: BottomNav.useCreateAction emits the
per-tab intent, and useTripPlanner consumes create=reservation|transport to open
the booking / transport modals (both already mounted at page level). Place and
expense were already wired; this just extends the mapping.
Tests: 4 new BottomNav cases (plan/bookings/transports/costs → correct intent +
navigate target); client tsc clean, full client suite green (2855).
Implements mauriceboe/TREK#1349
On mobile the location (GPS) FAB sat at bottom: calc(var(--bottom-nav-h) + 12px),
which only clears the bottom nav. When a day is selected, DayDetailPanel slides
up over the map from bottom: navh+20 and spans nearly full width at z-index
10000, covering the button's band — so the button was hidden behind it.
DayDetailPanel now publishes its live measured height to a root CSS var
--day-panel-h (ResizeObserver, reset to 0 on unmount), and both map renderers
lift the button above the panel when it's open, reusing the hasDayDetail prop
they already receive:
hasDayDetail
? calc(var(--bottom-nav-h) + 20px + var(--day-panel-h) + 12px)
: calc(var(--bottom-nav-h) + 12px)
Applied to both the Leaflet (MapView) and GL (MapViewGL) renderers. When the
panel closes, hasDayDetail is false and the offset falls back to the bottom-nav
value. Desktop is unaffected — the button is mobile-only.
Tests: new DayDetailPanel case asserting --day-panel-h is published and reset on
unmount; client tsc clean, full client suite green (2851).
Opening the installed PWA offline showed Chrome's "no internet" page instead of
the cached app. On boot the axios response interceptor reacts to a failed
request with no response by probing /api/health; the probe collapsed "genuinely
offline" and "edge-proxy auth wall" into a single reachable=false, so the
interceptor unregistered the service worker and reloaded — straight into a dead
network. navigator.onLine is true on mobile while offline, so the existing guard
didn't help. This also defeated the offline data layer (withOfflineFallback,
authStore's offline branch), which runs later in the chain.
Fix: connectivity.probe() now returns a discriminated state
('online' | 'offline' | 'proxy-wall'). A fetch that throws, or navigator.onLine
false, is 'offline'; a cross-origin redirect (CF Access, via redirect:'manual'
→ opaqueredirect) or an HTML auth wall (Pangolin) is 'proxy-wall'. The
interceptor only tears down the SW on 'proxy-wall'; on plain offline it lets the
request reject so the cached shell + IndexedDB serve the app. CF Access /
Pangolin reauth still works — the proxy always presents a reachable redirect or
HTML wall, which the probe now detects positively.
Regression dates to v3.0.16 (#964), surfaced by the 3.1.0 rewrite.
Tests: 6 new connectivity cases (offline/online/proxy-wall discrimination);
client tsc clean, full client suite green (2850).
The mapper read only `departureScheduled`/`arrivalScheduled`, but those columns
are optional in AirTrail and stay null for manually-entered flights — where
`departure`/`arrival` are the only times set. So the import dropped the departure
clock (date-only) and the whole arrival (no date, no time), exactly as reported.
AirTrail's own rule is "use departure if available, otherwise fall back to
departureScheduled". Mirror that: prefer the scheduled instant, fall back to the
primary departure/arrival, in mapFlightToReservation, normalizeFlight, and the
sync hash. Hashing the resolved instant means flights already imported without a
scheduled time re-sync once and pick up their clock automatically; flights that
do have scheduled times are unaffected (no spurious re-sync).
Tests: 3 new mapper cases (fallback mapping, picker preview, hash tracking);
two existing cases that asserted the scheduled-only behaviour updated to the
"neither time set" case. Full server suite green (4085).
This branch added a clamp-to-nearest-day fallback to resolveDayIdFromTime so an
imported booking whose exact date has no day row still lands on a day. After
rebasing onto dev, that collided with #1288's resyncReservationDays, which
relies on the original "null when no exact day" semantics to leave a booking
whose date now falls outside the range untouched — instead it snapped to an edge
day (TRIP-SVC-019 failed: expected day_id kept, got the clamped one).
Make clampToNearest an opt-in parameter (default true, preserving the import
behaviour for create/update) and have resyncReservationDays pass false, so
out-of-range bookings keep their day_id. Full server suite green (4082).
The Swedish (sv) locale landed on dev (#1325) after this branch added the
AI-parsing settings/reservation keys to the other locales, so sv was missing
them — strict i18n key parity failed after rebasing onto dev. Adds the 3
reservations.import.* and 17 settings.aiParsing/aiAlwaysRetry keys in sv.
BackgroundTasksWidget (mounted globally in App) subscribes via addListener/removeListener from api/websocket, but the global test mock didn't export them, so every test that renders <App/> threw on mount. Add the two stubs. (Surfaced now that the page-pattern check passes and the client test step actually runs.)
The new LLM extraction router shipped with little branch coverage, dropping src/nest below the 80% gate. Add unit tests for routeExtraction (flights/single/union/error paths, deterministic booking-wide fill), the native Ollama format client, the provider factory, the local-router service path with its type-aware text cap, the flat->schema.org mapper's remaining reservation types, and the background import-jobs runner. Also remove the now-unused validate.ts (only its FlatLike type was still referenced; moved to flat-schemas).
TripPlannerPage held a useEffect (the background-import → review bridge), which trips the page-pattern check (pages must stay wiring containers). Move the effect and its store/IndexedDB wiring into useTripPlanner where the rest of the import-review state already lives.
hasIntegrations gated the tab on memories/mcp/airtrail only, so a user with just the llm_parsing addon enabled saw no Integrations tab and could not reach the AI parsing config. Include llmEnabled in the gate.
Adds an "AI parsing" section under Settings -> Integrations where a user can choose the LLM provider, model, base URL, API key and multimodal option used for booking extraction. This per-user config applies when an admin has not configured an instance-wide model. Reuses the existing encrypted user settings: the API key is stored encrypted, never prefilled, and a blank field keeps the stored one. Adds settings.aiParsing.* across all 20 locales.
The source document was only kept in memory on the background task, so a page reload during the (now always-LLM ~25s) parse lost it and the booking saved without its file. Store the uploaded files in IndexedDB keyed by job id; the review loads them from there when the in-memory copy is gone, and a 1h TTL prunes abandoned imports.
During the per-item import review the booking isn't saved yet, so the Costs section showed an empty 'Create expense' even though a linked cost will be created on save. Show the parsed price (amount + category) as the pending linked expense so the user can verify it up front. Reuses existing i18n keys.
A long single-booking PDF (e.g. an 11-page rental voucher) spent ~200s on CPU prompt-eval at the 16k cap, though its data sits in the first ~2k. Cap non-flight docs at 6k (flights keep 16k for all legs). Also make the rental operator a required field so the car gets a real title.
The saving client gets no budget:created echo (X-Socket-Id) and the create response omits the linked budget item, so the booking's Costs section and the Costs tab stayed stale until a manual reload. Reload the budget items right after a create that carried a budget entry.
After dropping the vendor templates, the model skipped the (often unlabeled) Expedia-style hotel address — making address a required schema field forces it to emit the street-address line, restoring the booking's location/place. Also hint the rental company so a car booking gets a real title instead of the generic fallback.
Now that a capable instruct model (Qwen3-8B, thinking off) reads name/address/dates/legs reliably across formats, the per-vendor template short-circuit distorted more than it fixed: brittle on layout variations and overriding the better model output. Remove the template layer; the model extracts the structure and Schicht 2 backfills the confirmation/total and takes the currency from the document's own symbol (correcting model misreads like ¥→$). Per-type prompts now also ask for address and price/currency.
A/B against the prior default (qwen2.5:7b) on CPU showed Qwen3-8B is both faster and more accurate on tricky/multilingual booking docs (correct Airbnb year+price, correct DisneySea admission date), once thinking is disabled — which the router now does. Feature it as the recommended pull, keep qwen2.5:7b as the fallback.
Hybrid/reasoning models (Qwen3 and similar) default to emitting reasoning tokens, which collide with Ollama's format-grammar constraint — on CPU this produced null/unparseable output and blew the latency budget (qwen3:8b: null or 300s timeouts vs ~20s with thinking off). Send think:false on the /api/chat call; Ollama ignores it for non-thinking models (verified on qwen2.5:7b), so it's safe and unlocks the stronger Qwen3 family.
Keep the uploaded files on the background task and hand them to the review flow, so each reviewed booking pre-fills its Files with the document it was parsed from (uploaded with the booking on save). The two modals also adopt the shared resolveDayId helper.
Code-audit clean-ups: share one normCurrency between the router and the templates, lift the duplicated nearest-day resolver into formatters.resolveDayId, drop two needless as-unknown-as casts at the fillBookingWideFields call sites, restore routeExtraction's doc comment, and give the broker template readable names. Plus recognise ¥/JPY and fall back to a standalone symbol amount, so a Klook-style voucher whose price sits far from any label still yields a cost.
Imported bookings auto-create their linked budget items server-side, but the saving client suppresses its own budget:created echo, so the Costs list stayed stale until a manual reload. Reload the budget items when the review session ends.
A reviewed transport (e.g. a rental car) arrived with only its parsed pick-up/return dates and no day_id, so the modal kept just the time and saved a bare "HH:MM" with no date. Resolve start/end day from the parsed dates (exact match, else nearest trip day) so the booking lands on the right days.
reservation_endpoints.lat/lng are NOT NULL, so saving a reviewed transport whose pick-up/return couldn't be geocoded threw a 500 and lost the whole booking (dates, linked cost). Skip those rows; the dates still persist on reservation_time/reservation_end_time.
Persist the background-import tasks (id/trip/status only) and re-fetch each job's status on mount, so a parse still running when the page reloads keeps its widget instead of vanishing; expired jobs (404) are dropped and a restored 'done' task re-fetches its items.
Apply the deterministic confirmation-code and total fill to vendor-template results too (not just model output), and require the captured reference to contain a digit so a bare 'Confirmation'/'Reference' label no longer grabs the next prose word.
Pull the hotel/rental fields these vendors print in a stable text layout (name, address, stay/pickup dates, price, reference) deterministically, so the import stops depending on the local model for them. Handles German long/abbreviated months and English dates incl. 12-hour and comma forms.
Reviewing an imported booking saves it through the normal reservation
form, which dropped the parsed price (so no linked cost was created) and
only created the accommodation when both nights matched a trip day.
Carry the parsed price into a linked cost on save, and create the
accommodation from whichever day the check-in/out dates resolve to.
Parsing a booking can take a while on a CPU host, so don't hold the
upload modal open for it. The async import endpoint returns a job id
right away; the parse runs server-side (one at a time per user) and
pushes progress over the user's WebSocket, and a small widget in the
bottom corner tracks it while the user keeps navigating and editing.
A finished job opens the per-item review from the widget.
The single-shot prompt was unreliable on multi-leg flights and longer
documents, and slow on a CPU host. For the local provider, run a small
router instead:
- deterministic vendor templates first, with no model call at all
- exactly one grammar-enforced call per document via Ollama's native
`format` (flights as a flat array of legs, everything else as one flat
reservation, the type picked from keywords or a union schema)
- booking-wide fields (booking reference, total price, the overnight
arrival day) filled deterministically from the text afterwards, and
dates coerced to ISO so a natural-language date can't slip through
Recommend qwen2.5 in the AI-parsing settings instead of NuExtract.
Instead of writing parsed items straight to the trip, the import opens the
normal edit modal pre-filled for each one, so you can check and fix it before
saving — useful when a model guesses a wrong date or address. Hotels gained an
editable address field; on save an existing place is matched by name, otherwise
the reviewed address is geocoded and a new place is created.
The provider picker is the shared CustomSelect now and the form is split into
clear sections rather than a flat stack of inputs. NuExtract 2.0 2B is the
recommended default — fastest on a CPU-only host and MIT licensed; the 4B
carries a non-commercial licence, so it's no longer flagged as recommended.
NuExtract isn't an instruct model — fed a plain chat prompt it just echoes the
schema back. Detect a NuExtract model by id and talk to it the way the model
cards document: the JSON template inlined in a single user message, no system
prompt, no json_schema, temperature 0. Its flat result is mapped back to the
same KiReservation shape the rest of the pipeline already uses, so nothing
downstream changes; every other model keeps the generic prompt.
Money is taken as a verbatim string and parsed locally (German "1.580,22 €"
otherwise comes back as 1.49772), a rental car's pickup/return ride the from/to
fields so a stray form label doesn't become the location, and a lodging with no
name falls back to its address instead of being dropped.
A freshly imported hotel links to an accommodation that lives outside the
trip store, so loadTrip alone left the reservation edit modal with blank
place/date fields. Reload the accommodations list once the import finishes.
When a confirmation carries a total price, record it as a real expense
linked to the reservation (in the matching Costs category) instead of
leaving the amount in metadata only. Gated on the Costs addon.
Request and map root-level seat/class/platform and a total price/currency into reservation metadata (shown on the card; price reuses the existing label). Read both the root and reservationFor and tolerate common field-name aliases (priceAmount, priceCurrencyISO4217Code, fareClass, ...) since models name these inconsistently. Also capture event/attraction venue telephone + url onto the auto-created place, matching lodging/restaurant.
On a GPU-less host the model's prompt-eval time scales with input length and dominates total latency. Booking details sit at the top of a confirmation, so capping the extracted text at 4000 chars (was 8000) roughly halves extraction time (~50s warm for a capable local 7B model) with no loss of fields on real hotel/rental confirmations. Tunable if a long multi-segment itinerary needs more.
- rental car: request+map dropoffLocation, emit pickup->return from/to endpoints, set a location string (G1/G2/G3). - geocode endpoints (stations/stops/terminals/rental desks) on confirm via Nominatim; mapper now emits coordless named endpoints and confirm persists only the geocoded ones (G6). - assign every dated booking to the nearest trip day so it still shows when slightly out of range, and keep hotel accommodation from vanishing when a check date misses (G5/G10). - fix bus mislabelled as train + add bus_number metadata (G7/G8), flag malformed boats (G9), accept root start/end time for events (G11). - raise the local-LLM timeout to 300s for CPU-only Ollama.
client: the import call inherited the global 8s axios timeout and aborted long LLM extractions even though the server finished it; remove the timeout. server: raise the OpenAI-compatible LLM timeout 60s->180s (a cold Ollama model can take ~45s to first token). server: cap extracted text to 8000 chars before the LLM - multi-page T&C tails (30k+ chars) overflowed the context window, truncating the relevant head and making CPU inference crawl; booking details sit at the top.
Booking import only fell back to the LLM when each user flipped an 'always retry with AI' toggle, so by default files kitinerary returned nothing for just failed. Run the fallback automatically whenever the AI Parsing addon is on (fallback-on-empty); drop the now-redundant per-user toggle and its setting.
Maurice