34 Commits

Author	SHA1	Message	Date
Maurice	5d0fac3026	feat(mcp): place pictures, POI search, GPX export and a biased place search ... image_url was in no places tool although the service persists it and the picker in the UI sets a URL, which is not a file transport question. search_place passed two of the service's four arguments, so an unbiased query resolved to the wrong continent. get_place_details gained expand for reviews and the editorial summary. It defaults to off because that field mask bills as a Google Enterprise SKU, and it now honours the admin kill switch the REST route checks: an instance owner who turns Place Details off to stop the billing was still billable through this one surface.	2026-08-28 14:20:02 +02:00
Maurice	1ad2e92142	fix: make the app keyboard-reachable, and close four DoS-able patterns ... The bulk of what SonarQube still had open on the reliability side, read site by site rather than swept. Two clusters are worth calling out because they are defects, not lint. **Four regexes could be made to hang the process.** The worst is `/.+@.+\..+/` in the forgot-password path: three unbounded `.+` that all overlap, unanchored, on an endpoint that needs no login and whose schema is a bare z.string() under a 100kb body cap. A thousand '@' characters had not finished after 100 seconds — one request pins the event loop, and Node has one. EMAIL_REGEX in auth.helpers is the same shape on register and profile update (310ms for 8kb). The two tag strippers over an uploaded KML description and over Commons metadata took 0.52s on a 16k run of '<'. Each rewrite was proven equivalent by brute force against the original — millions of strings for the coordinate patterns, 671,876 for the tag stripper — because a subtly different regex on a parsing path is worse than the finding. `<[^>]+>` could not be fixed in the pattern: `[^>]` matches `<` too, so the only unambiguous form changes what happens to malformed markup. It is now a scan in common/stripHtmlTags.ts, quirks included — `<>` is left alone, `<a<b>` is one tag — with the equivalence pinned by a test that compares against the old regex over every string up to length six. **And roughly 500 places could only be reached with a mouse.** Clickable divs became real buttons where the markup allowed (52 of them), took role + tabIndex + Enter/Space where they had to stay divs, and modal backdrops took role="presentation" — the honest role, since a backdrop is a mouse shortcut for a Cancel control that is already in the tab order, and making it a button plants a nameless full-screen tab stop around its own contents. Alongside those: `.sort()` calls that now say the order they always had, `{n && <X/>}` guards that rendered a literal "0" for a zero-byte file and for a place at latitude 0, an optimistic budget write that stored a boolean where the contract and both transports use 0/1, and the Number statics. sonar-project.properties documents the six that were deliberately left: two wiki patterns with no overlapping quantifier pair whose "fix" would stop parsing `[see [1]](url)`, four tar header fields bounded at 12-155 bytes by the format, two djb2 hashes where the 32-bit wrap is the algorithm, two `| 0` that are also coercing undefined to 0 on a path where the API sends no user_id, and the MCP session maps Sonar reads as empty because they are filled from another module.	2026-08-25 19:26:38 +02:00
Maurice	2c2d008dc2	fix: clear every open Sonar security issue, and half the reliability ones (#2037) ... * fix(security): mint idempotency keys properly, and resolve the extractor by path Three findings out of a sweep of every open Sonar security issue. The other 49 turned out to be false positives and are documented in sonar-project.properties rather than changed — each entry scoped to a file, never a whole rule, so a future Math.random() on an auth path still fails. The idempotency key is the one that can lose data. Both producers fell back to Math.random when crypto.randomUUID was missing, and randomUUID needs a secure context — TREK's own quickstart is http://localhost:3000 and LAN installs run on http://192.168.x.x, so on the deployments the README documents the fallback is what actually runs. The axios one was `Math.random().toString(36).slice(2)`, which is not length-stable: 0.5 yields a single character and 0 yields the empty string, which the interceptor reads as "no key" and skips deduplication for entirely. Two writes sharing a key collapse into one — the second gets the first one's cached response and the edit is gone with no error anywhere. Both now come from utils/randomId, which has the getRandomValues rung that works in insecure contexts too. The extractor probe stored the bare binary name, so every extraction re-resolved it through whatever PATH held at that moment. The official image sets KITINERARY_EXTRACTOR_PATH and never reaches that branch, but a bare-metal install without it ran an unqualified name from a PATH it does not control. It now walks the search path itself and keeps the absolute path of the file it verified; execFileSync also drops the /bin/sh hop the old probe needed. PATH comes through app-config because process.env is walled off in src/. * fix: four tests and sorts that were not doing what they claimed Sonar reported 43 high-impact reliability issues. Reading all of them, four are real; the rest are cosmetic or would be made worse by the suggested change. budget.test.ts hand-built `paid: true`, but both emitters send `paid: body.paid ? 1 : 0` and the handler stores the value verbatim — the test asserted a payload the server never sends. The line below it already used a number. admin.controller.test.ts had a `.resolves` assertion with no await. The promise was dropped, so that line could not fail the test. The two to-do category lists sorted with a bare .sort(), which is byte order: every accented name files behind the whole ASCII range, so in a German trip "Übernachtung" lands after "Zelt". TREK ships 23 locales, so this is real data. One side effect worth knowing: katColor picks a category's dot colour by its index, so re-sorting can move which category shows which colour — it already moves whenever a category is added, and the colour carries no meaning. Deliberately not changed: the two `| 0` truncations (Math.trunc is not equivalent — `| 0` wraps at 32 bits and these carry millisecond timestamps), the two .map() callbacks whose second parameter is declared and used, and the sha1 in thumbnail.service (its own comment says a swap would orphan every existing thumbnail). * refactor: give every button a type, and prefer the Number statics Two mechanical sweeps, no behaviour change intended in either. A <button> with no type attribute is type="submit" per the HTML spec. Outside a form that does nothing today, but it turns into a surprise the moment the markup is wrapped in one, which is what Sonar's S9011 is warning about. 870 buttons get an explicit type="button". Two things were left alone on purpose: buttons that take a {...spread}, because the spread may carry a type and the merge order decides which wins, and anything inside a <form> — there the answer is a judgement call, and checking all eleven form files found every real submit button already carrying type="submit". The Number sweep covers only the three globals that are exactly equivalent to their Number twin: parseInt, parseFloat and NaN. isNaN is NOT in it — the global coerces its argument, so isNaN('abc') is true where Number.isNaN('abc') is false, and renaming it would change what the code does. It reaches server and plugin-sdk as well as the client; both copies of egress-policy.ts got the identical edit, so the mirror the SDK's parity test guards stays intact.	2026-08-25 15:22:38 +02:00
Maurice	38f3cb3101	fix(server): the second pass over this branch's own fixes ... An adversarial review of what had already landed here turned up three that were wrong, and they are the reason for this commit. The CSP change removed 'unsafe-eval', which is not leftover: heic-to's libheif build initialises embind through new Function(), and without it every iPhone .heic upload fails in the browser. 'wasm-unsafe-eval' alone was tried once already and was not enough (93b51a0b). Put back, with the reason written down this time and the test pinning it rather than its absence. The idempotency in-flight guard never fired. finalize() runs when the handler's observable completes, but Nest writes the response - and therefore stores the row - several microtasks later, so the waiting request always woke up before the row existed and ran the handler a second time. It releases from the point the row is written now; finalize is the backstop for handlers that answer another way. The test drove the two in the opposite order from Nest, which is why it passed either way. And the trip-scoping check treated an id that resolves to nothing like one belonging to someone else. accommodation_id has no foreign key, so shortening a trip's dates leaves a booking pointing at a gap - which update() has always healed by clearing the link, and which would have become a permanent 400. The rest: a reply to a soft-deleted chat message is refused and the create path blanks the quote like the list path does; the ws-token buckets are keyed by account rather than by IP, where every user behind one NAT shared a ceiling; the weather and vacay caps read through the stream, since a chunked response has no content-length to check; the Google host gate takes country domains again; abandoned response bodies are cancelled instead of left to GC; the boot failure prints its errno; and a share link that is not yours to read now says so instead of reporting that the journey has none. TICKETJSON gets a real repair: the in-place GLOB fix only helps installs that have not run it yet, so an appended migration puts back the note text the case-insensitive LIKE cut off, for rows where what it stored is not valid JSON. A genuine receipt always parses, so it cannot be touched. Wiki: the audit log records the IP Express resolves after TRUST_PROXY, not the leftmost X-Forwarded-For entry a caller can write. That is the right way round, but it means the hop count has to be accurate - said plainly in both places now.	2026-08-23 11:59:12 +02:00
Maurice	97ebe23d3b	fix(server): bound every outbound fetch ... The rule is a timeout, a size cap and boundary validation on anything that leaves the process, and several call sites had only the timeout. A link preview, a maps URL resolve, a wiki fallback and a holiday lookup all read the whole body into memory, so a caller who picks the URL picks how much memory the server spends - a few concurrent multi-hundred-megabyte responses is the whole heap. Two of them also fetched more than their name suggests. resolve-url fell through to a full page GET for any host once it found no coordinates, which makes the server a general-purpose outbound proxy wearing its own IP. The weather URLs took lat/lng straight from the caller into the query string. readCapped/exceedsDeclaredLength collect the pattern transit.service.ts and the plugin installer had each grown separately, so there is one of it now. Callers that only scan the head of a document (og: tags, a coordinate regex) keep the truncated bytes and find fewer fields; callers that must have the whole thing reject it.	2026-08-23 11:59:12 +02:00
jubnl	76c8c349b1	Fold nest mcp (#2017) ... * refactor(server): fold @trek/nest-mcp into server/src/nest-mcp * docs(server): point nest-mcp references at server/src/nest-mcp * remove tracked md file	2026-08-22 12:55:24 +02:00
jubnl	305b113d5f	Feat/storage driver (#1949) ... * feat(storage): storage key validation, driver contract suite, LocalDriver * feat(storage): MirrorDriver composite backend * feat(server): add TREK_UPLOADS_DIR env var for the uploads root * feat(storage): settings-backed registry, StorageService facade, module wiring * test(storage): pin src/nest/storage coverage thresholds * feat(storage): multer upload options factory * refactor(files): spool trip-file uploads through storage * refactor(collab): spool note attachments through storage; normalize files/ prefix * refactor(journey): spool journey media uploads through storage * refactor(trips): spool trip cover upload through storage * refactor(places): spool place-image and collection uploads through storage * refactor(auth): spool avatar upload through storage * fix(server): consolidate allowed_file_types onto AllowedFileTypesService * fix(storage): default defParamCharset to utf8 in the upload factory * test(db): re-point the trailing-migration guard at the trip_files prefix strip * fix(storage): age-gate the boot spool sweep * test(files): key FILE-P02 residue checks on upload size * test(storage): raise src/nest/storage coverage pin * refactor(platform): serve /uploads routes through storage * refactor(files): serve trip-file downloads through storage * refactor(memories): serve photo bytes through storage * refactor(maps): stream place-photo bytes through storage * refactor(share): stream shared place-photo bytes through storage * refactor(journey): serve public journey assets through storage * fix(share): answer place-photo cache miss with 204 like the maps proxy * test(server): close serving-swap gaps surfaced by the full tier * test(storage): raise src/nest/storage branches pin * refactor(place-photos): route place-photo cache through storage * refactor(places): reclaim place images and collection covers through storage * refactor(trips): trip covers through storage * refactor(auth): avatar reclaim through storage * refactor(collab): note-file deletion through storage * refactor(files): trash deletion through storage * refactor(memories): trek photo cache internals through storage * refactor(memories): thumbnail service category-addressed * refactor(memories): immich upload reads through storage; drop uploads-root * refactor(journey): photo deletion through storage * fix(memories): reclaim orphaned trek photo cache objects in the ttl sweep * feat(journey): journey-thumbs orphan sweep and delete-site derived-artifact reclaim * test(server): close slice-4 gaps surfaced by the full tier * test(coverage): raise slice-4 domain pins * refactor(backup): list, existence and delete through storage * refactor(backup): download through storage sendToResponse * refactor(backup): create backup through per-category storage enumeration * refactor(backup): restore and upload-restore through storage * refactor(backup): auto-backup retention through storage * refactor(server): boot dir-creation to driver init * test(server): close slice-5 gaps surfaced by the coverage tier * test(coverage): raise slice-5 backup pin * test(server): resolve async backupFileExists mocks for the tests typecheck * test(server): pin ranged-read stat and zero-byte roundtrip in the storage driver contract * feat(server): TREK_S3_* env wiring — schema, s3 namespace, all-or-nothing boot rule * feat(server): S3 driver core — aws-lite client seam, stat/delete, driver-owned deadlines * refactor(server): type the aws-lite S3 client via s3-types reference * feat(server): S3 driver getStream and list — ranged full-object stat, iterator pagination * feat(server): S3 driver put — bounded-peek routing, Upload inactivity deadline * fix(server): S3 driver Upload monitoring — Transform interposition, stream cleanup * feat(server): env-declared s3-main storage backend; settings-declared s3 rejected * test(server): env-gated S3 driver contract suite against MinIO * fix(server): S3 driver zero-byte put — work around aws4 duplicate Content-Length signing defect * fix(server): scope the zero-byte put workaround to buffered bodies; bind test MinIO to localhost * ci(server): MinIO-backed S3 driver contract job * ci(server): align s3-contract shim comment and MinIO bind with references * docs: S3 storage backend configuration and off-box backup mirroring * docs: restore README formatting, keep S3 additions only * ci(server): make live-MinIO contract runs manual-only (AIStor licensing) MinIO's community line is discontinued entirely; the maintained successor, AIStor, requires a license key that can't safely live in a shared CI runner. The s3-contract CI job is removed; live-provider contract runs against MinIO/AIStor are now a manual, license-gated procedure documented in docker-compose.minio-test.yml's header. * chore(server): final-review polish — stale CI refs, dockerignore key, boundary pins, zero-byte tempfile contract case * expose admin ui * feat(shared): storage backend type registry, wire schemas, category relocation * feat(server): accept settings-declared s3 backends — shared-schema options, decrypt-at-build, registry preview() * feat(server): remove TREK_S3_*/TREK_UPLOADS_DIR — tombstone boot rule, built-in override as the relocation path * feat(server): deriveSecurity.encryptionKeySet and storage secret plumbing (gate, encrypt, unmask, mask, redact) * chore(server): drop storage-secrets import that only Task 8's probe helper will use * feat(server): seed-once storage-config.json boot import — fail closed on any invalid seed * feat(server): registry snapshot() — effective storage config with provenance * feat(server): api/admin/storage GET/PUT — masked secrets, preview-gated saves, wired reload, audits * feat(server): api/admin/storage/test — per-target ephemeral probes, mirror targets probed individually * test(server): e2e coverage for api/admin/storage — guards, managed refusal, envelope, probes * test(server): give e2e harness DBs the app_settings table the storage registry reads at boot * test(coverage): raise pins after the storage admin-config slice * fix(server): final-review polish — encryption-gate wording, all-field sentinel guard, README/e2e notes * test(server): storage slice-1 review follow-ups — mirror s3-replica probe decrypt, all-encrypted seed without key, test-route 401 * feat(i18n): storage admin domain + tab label in every locale * feat(client): adminApi storage endpoints — getStorage, updateStorage, testStorageBackend * feat(client): storage admin draft-document helpers + relative-time util * feat(client): schema-driven storage BackendForm — fields by kind, mask echo, encryption banner * feat(client): admin storage panel — backends list, category map, health strip, one-PUT save * feat(client): storage tab on the admin page * feat(client): storage admin panel and section on the phone shell * fix(client): block renaming a storage backend onto an existing name * feat(server): allow mirrors on any storage category — lift the backups-only rule * feat(i18n): replicas-on-primary mirror strings in every locale * feat(client): mirror fold/synthesize model helpers for replicas-on-primary * feat(client): BackendForm mirror-targets composer; raw mirror type hidden * feat(client): desktop storage panel folds mirrors into their primaries * feat(client): phone storage panel folds mirrors into their primaries * fix(client): strip removed backends from remaining mirror replicas — no phantom refs * feat(storage): retire photos as a configurable category — served-only via the ServedCategory seam * feat(i18n): descriptive storage category names and descriptions in every locale * feat(client): descriptive category names, descriptions and id badges in the storage panel * docs(storage): fix stale nine-category comments after the eight-category shrink * docs: storage configuration via the admin UI and seed file — README + compose * docs(wiki): Admin Storage page and storage cross-links * docs(server): storage domain blueprint entry; TREK_PLACE_PHOTO_DIR tombstone clarifier * refactor(server): drop the storage env-var tombstones — the variables never shipped * docs: storage config is simply new in v4 — drop the removed-variables story * feat(shared): backfill-status and usage wire schemas on the storage admin state * feat(server): MirrorDriver.backfill — prefix-scoped, size-compared, cancellable replica catch-up * feat(server): one-at-a-time storage backfill jobs + the storage events bridge * feat(server): storage usage scan — nightly cron, on-demand service, persisted with computedAt * feat(server): replica_failure admin notifications — events bridge, hourly debounce, channel registration * feat(i18n): replica-failure notification text in every locale * feat(server): storage backfill and stats-refresh admin endpoints * feat(client): adminApi backfill and stats-refresh endpoints * feat(i18n): storage sync and usage strings in every locale * feat(client): desktop storage panel — sync now, usage display, backfill polling * feat(client): phone storage panel — sync now, usage display, backfill polling * fix(client): per-category usage line on the phone storage panel * docs: sync now, usage stats and replica-failure notifications * test(server): admin scope now includes replica_failure, not just version_available Commit 6db135da added replica_failure to ADMIN_SCOPED_EVENTS but left NPREF-005b asserting the admin scope contained only version_available. Update the expectation to match. * test(server): cover storage backfill/stats/jobs branches to hold the coverage pin * fix(client): guard storage admin polling against stale responses racing a save A poll GET issued before a save's PUT resolves could land after the save applied its fresher response, overwriting it with the pre-save world and re-deriving the draft from that stale snapshot. useStorageAdmin now tracks a state-application sequence number and drops any refreshState() response whose sequence has moved on, or that resolves while a save is in flight. Also guard the fire-and-forget refreshState() calls in startBackfill/ cancelBackfill against unhandled rejections, matching the poll's style. * fix(server): report 404 before 409 when a backfill start names an unknown mirror startBackfill checked the global busy flag before validating that :name is a mirror routed by some category, so POST .../backends/garbage/backfill answered 409 instead of 404 whenever a sync was already running. Resolve and validate the target first (BackfillTargetError -> 404), then check busy (BackfillBusyError -> 409). * feat(client): two-column category grid on the storage panel * feat(server): plugin files.getContent reads through the storage layer * feat(server): plugin files.create writes through the storage layer * refactor(server): drop FilesService.resolveFilePath, last direct-fs seam in the plugin surface * fix(server): map StorageInvalidKeyError to the plugin file accessibility envelope * feat(storage): allow credentialed backends without an explicit ENCRYPTION_KEY Secrets are still always encrypted at rest; without an explicit key the implicit data-directory key is used. Removes the plaintext-secret refusal from the admin PUT and the seed import, the encryptionReady GET field, the save-blocking banner in both admin shells, and the now-consumerless security.encryptionKeySet env derivation. Docs now recommend (not require) an explicit key, since the implicit key rides inside backup archives. * test(client): pin the stale-poll release to the pre-save GET in FE-ADMIN-STOR-024 The stub deferred EVERY poll GET and rebound the release handle to the newest one, so a tick landing after the save was released with the pre-save payload — a response no real backend can produce, which the seq guard rightly applied. Only the first (genuinely stale) GET is held now; later GETs answer with the world the mock server currently holds. Also drops the encryptionReady fixture field removed from the schema in the previous commit. * fix(storage): backfill puts carry content-type, stat failures report op stat, progress clamps * feat(storage): category migration job — copy, flip, delta sweep, reclaimable tally * feat(storage): migration start/cancel endpoints and client API * test(client): storage admin fixtures carry the new migrations field * fix(storage): migration POST answers 200 and validates its body via the shared schema * fix(storage): cancel running storage jobs when a config save removes their backend * fix(notifications): in-app replica-failure row carries the suppressed count * feat(client): migration prompt, queue, and progress on the storage panel * fix(client): migration queue holds a synchronous single-flight lock * feat(client): migration prompt and progress on the mobile storage panel * fix(server): register category-migration routes in the managed-route allow list StorageAdminController.migrationStart/migrationCancel were added under the controller-level @ManagedForbidden() decorator but never added to MANAGED_ROUTE_ALLOW_LIST, so validateManagedRoutes() threw "Managed surface changed" on every buildApp() call, failing 46 test files (any test booting the real app). * test(server): cover category-migration copy/sweep branches to hold the storage coverage pin Adds MIG-005 (a target object already matching size/key is skipped in the copy phase itself, not just the delta sweep) and MIG-006 (a copy failure during the delta sweep is counted without undoing the flip that already happened). src/nest/storage/**'s branch coverage was 83.91%, just under the 84% pin. * docs(storage): category migration — move existing objects on reassign * fix(storage): close the pre-flip cancel window and cancel re-routed migrations A cancel landing between the last copy iteration and the category flip was silently ignored, and a config save that re-routed the migrating category to a third, still-defined backend never stopped the in-flight migration. Also guards assignCategory against writing an unknown backend name, and fixes a missing `suppressed` notification param being treated as truthy suppression. * fix(client): migrations keep mirrors, queue waits on backfills, sweep failures surface Reassigning a category onto a mirrored primary now posts the migration to the mirror's wire name instead of the bare primary, so replication survives the move. The queued-migration effect also waits out a running backfill (the server's one-job rule spans both), and a done migration with sweep failures now surfaces a doneFailures line in both shells, translated into every locale. * ci(server): run the S3 contract suite against MinIO Resurrects the s3-contract job removed in 6fbff835, adapted for AIStor: writes the MINIO_LICENSE_KEY secret to a license file, boots the licensed container, and runs the contract suite hermetically (no --coverage) so the storage coverage pin keeps holding from unit tests alone. Gated to same-repo PRs since fork PRs have no access to the license secret. * fix(server): backup archives uploads from a remote primary createBackup() enumerated uploads through withLocalFile, which downloads a remote object to a temp file, hands it to the callback, then deletes it before archiver ever reads it — archiver reads entries lazily during finalize(), so a remote (S3) primary silently backed up zero upload bytes. Add StorageService.getLocalPathOrNull(category, name): a narrow locality probe that returns a real path only when one exists on disk right now (the fs-existence check is the fail-safe half of the contract — a driver-reported path that vanished between list() and here falls back too, not just a path-less remote driver). The enumeration loop uses it directly when available (zero-copy, unchanged for a local primary) and otherwise streams the object into a per-backup staging dir under the backups spool, cleaned up in the existing finally alongside zipSpool/dbSnap. Also wire archive.on('warning', reject): archiver treats a missing/unreadable entry as a warning, not an error, and skips it by default — which would silently drop the entry from an otherwise "successful" backup. * fix(server): client aborts stop 500ing and leaking streams storage.service.ts sendToResponse: the local res.sendFile branch now treats ECONNABORTED/ECONNRESET/ERR_STREAM_PREMATURE_CLOSE/syscall=write as success, mirroring res.sendFile's own default callback (the contract storageStaticHandler already implements caller-side). The remote branch replaces the hand-rolled pipe+Promise with pipeline(stream, res), which destroys the source stream on any failure — fixing the leaked S3 keep-alive socket on a client abort — while swallowing the same client-abort error set; a real source error still rejects so callers keep their miss contracts. This centralizes the fix for every caller that streams through sendToResponse: files-download, journey-public, photo-resolver, trek-photo-cache, backup downloads, and /uploads/photos. trek-exception.filter.ts now no-ops (destroying the socket) when headers are already sent, so no filter path ever writes onto a response in flight. maps.controller.ts and share.controller.ts's place-photo bytes proxies switch their direct getStream+pipe to pipeline() with the same abort-tolerant catch, keeping the existing 204-empty fallback for pre-header failures. * fix(storage): mirror read fallback consults replicas on primary miss MirrorDriver.getStream/stat now fall back to replicas on a primary miss (StorageNotFoundError / null stat), not just an infra error — binding decision: a read succeeds if ANY member holds the object. Only a deterministic StorageInvalidKeyError skips fallback. stat() gets a dedicated fallback loop instead of reusing fromReplicas: a replica returning null means "keep looking," which the generic catch-and-continue helper would have short-circuited on. StorageService.sendToResponse/withLocalFile stop treating composite driver.stat() as proof the object is at localPath; both now fs-check the path itself and fall through to the stream branch when it's missing, which is what actually reaches a MirrorDriver's replicas. * feat(storage): remote serving parity — content-type, validators, range Local backends get content type, ETag/Last-Modified, conditional GET, Range and HEAD for free from res.sendFile; the proxying branch every remote backend (and a local primary whose file vanished) falls through to had none of it — every request re-streamed the whole object with no cache validator and no way to seek. sendStreamed() reproduces the contract from ONE pre-flight driver.stat(): a miss throws before any header is set (the caller's 404/204/next() contract), and that same stat then answers 304s, 416s, HEADs and suffix ranges without ever opening a body. The decisions live in a new pure module, http-serving.ts: - validatorsFor: the driver's ETag, else express/send's weak W/"<size hex>-<mtime hex>". mtimeMs 0 (S3 can report it) means no Last-Modified and no derived tag — serving uncached beats a 1970 header that reads as fresh forever. - isNotModified: If-None-Match (weak compare, *, list) wins over If-Modified-Since, per RFC 9110 §13.2.2. - decideRange: single range only, inclusive clamp, suffix computed from the pre-stat size, unsatisfiable → 416, If-Range strong-exact or full entity. A multi-range or malformed spec is answered with the full object rather than multipart/byteranges. If-Match/If-Unmodified-Since stay out of scope. Range math uses the pre-stat size as the single source of truth, the ranged getStream stat is never re-consulted, and the pipeline abort semantics are unchanged. S3Driver now surfaces the bucket's ETag on stat()/getStream() (on a ranged GET it is still the full object's tag). aws-lite strips the tag's surrounding quotes — verified live against AIStor RELEASE.2026-08-07 — and an unquoted entity-tag is not a valid HTTP ETag, so the driver restores exactly the quotes the SDK removed and nothing else. * fix(server): key rotation covers storage.backends The S3 storage backend's secretAccessKey is encrypted at rest inside the app_settings row keyed storage.backends, but scripts/migrate-encryption.ts didn't know about it — after a key rotation that secret stayed encrypted under the old key and silently failed to decrypt on the next read. The script now re-encrypts secretAccessKey for every s3 backend inside the storage.backends JSON array, using a pinned copy of the secret-field list (it deliberately can't import @trek/shared). The parity test pins that copy against storageSecretFields('s3') so a new secret field can't be missed silently. * fix(server): restore reloads the storage registry restoreFromZip closed and reopened the DB but never reloaded the storage registry, so uploads rehydration ran against the STALE pre-restore driver/category map instead of the config the restored archive's own travel.db carries (audit #4, major). StorageService gains reloadConfig(), a narrow passthrough to the (unexported) registry's reload() — same shape as health(). backup.impl's restoreFromZip now calls it right after reinitialize() and right before the uploads wipe/rehydrateUploads/plugin staging, so rehydrated bytes land where the restored config says. When reopen fails (reinitFailed), reload and rehydration are skipped entirely — already reported as "restart required", and there is no live DB handle for the registry to read anyway. * fix(server): category migration rewrites keys across prefix changes Category migration copied objects to the destination backend using the SOURCE key verbatim. A mode-A ('' prefix) source migrating onto a prefixed backend (or vice versa) landed objects at keys the registry could never resolve post-flip (audit #8). Extracts StorageRegistryService.keyPrefixFor(category, backendName) from build()'s inline conditional so startMigration can compute the destination prefix before the flip. runMigration now maps every destination key (destPrefix + key.slice(sourcePrefix.length)) in the copy skip-check, the put, and the delta sweep; the reclaimable tally stays source-keyed. Equal-prefix migrations are unaffected (identity mapping). * fix(plugins): await the now-async place/collection services in RPCs PlacesService.update/remove and CollectionsService.deletePlace became async when they started deleting storage objects, but their plugin RPC handlers still called them synchronously: the null/false refusal check compared against an always-truthy Promise, and places.update broadcast a Promise instead of the resolved place. Both handlers now await the service call; mapCollectionError is async-aware (accepts a sync-or- async thunk) so the sync callers (create/update/savePlace/copyToTrip) keep working through the same path. Verified the RPC host already awaits handler results, so no change was needed there. * fix(storage): backfill Sync now gains a deletion sweep The Sync now backfill copied primary objects to replicas but never removed replica objects the primary no longer has, leaving mirror read-fallback able to serve stale data indefinitely after a failed replica delete. backfill() now sweeps each replica post-copy: any replica key absent from the primary key-set is re-checked against primary.stat (a race gate against the primary regaining the key) before being deleted. Adds a deleted counter end-to-end (driver, jobs service, shared schema, both admin shells, all 23 locales) and a new ReplicaFailure 'list' op for a replica whose listing itself fails during the sweep. * feat(storage): optimistic concurrency for admin config saves A stale admin PUT of storage settings could silently undo a category migration's just-flipped assignment. Adds a version counter (app_settings row storage.config_version) bumped inside the same transaction by both writers — applyConfig and the migration flip (assignCategory). applyConfig compares the submitted version first and throws StorageConflictError, mapped to 409 ahead of the blanket 400. The client carries the draft's loaded version through every local edit and maps a 409 to a distinct "settings changed underneath you" message, refreshing state while preserving the dirty draft for review. * fix(storage): surface a failed config load + redact secretAccessKey Registry load() now records the exact parse/validate failure as lastLoadError() instead of silently falling back to last-good config or built-in defaults; StorageAdminService exposes it as configError on the admin state (storageAdminStateSchema extended additively), and AdminStoragePanel renders it as a warning banner naming the error. Save stays enabled — it's the recovery path — but an admin can no longer save over a broken stored config without knowing. Also adds 'secretaccesskey' to globalMiddleware's SENSITIVE_KEYS request-log redaction list, so storage admin PUT bodies never leak the S3 secret into debug logs. * fix(client): admin storage panel state races Migration prompt now recomputes computeMigrationCandidates(draft, state) on render and again at confirm time, so a stale open-time snapshot never drives stripCategories or the migration queue; adds a Cancel button to the dialog. A failed startMigration POST explicitly clears the remaining queue with a toast naming the dropped categories and refreshes state, instead of leaving it invisibly stranded; a "Queued: X, Y" line renders while the queue is non-empty. Testing a mirrored row now expands the draft primary + replicas client-side and probes each individually, instead of sending the mirror stub to the server (which resolved targets from its own live snapshot, silently testing saved options over unsaved draft edits). A completed/cancelled/errored backfill now renders its terminal status alongside the Sync now button instead of hiding it, so it can be re-run without a page reload. * fix(storage): refuse shared replica backends so the sync sweep cannot cross categories A mirror's "Sync now" sweep makes each replica MATCH the primary: it lists the replica under every prefix the mirror's categories occupy and deletes the keys the primary doesn't hold. The backups category's prefix is '' — the replica's entire root — so making a category-serving backend (say uploads-local) a replica of a backups mirror meant the first sync deleted every user upload. Nothing forbade the config: validateConfig only checked that mirror targets existed and weren't nested, and the admin picker offered every backend. The sweep is correct, so the fix is at the config level: - validateConfig refuses a backend that is both a mirror replica and a category target (directly, or as the primary of a mirror a category routes to), and refuses a backend replicating two mirrors whose swept prefixes overlap — '' overlaps everything. Disjoint prefixes stay legal. - assignCategory (the migration flip, the other writer of the settings rows) refuses routing a category onto a backend that is currently a replica. - The admin mirror-target picker (both shells) only offers backends that serve no category; already-selected targets stay listed so they can be unchecked. A category assigned to a replica still surfaces the server's 400 verbatim. Existing shared-replica fixtures were rewritten onto dedicated replicas. * fix(client): a 409 conflict save is recoverable without a page reload After a 409 the admin storage panel was a dead end. refreshState leaves a dirty draft's version alone on purpose (audit #7: the background poll must never float a mid-edit draft past a migration flip it never carried), setDraft re-attaches that same version to every later edit, and save() resubmits it — so every retry 409s again and only a full page reload got the operator out, while the banner told them to "review and save again". The version block stays. Instead the conflict banner now offers the one action that can clear it: discard the local edits and reload the saved settings — a fresh GET applied through applyState, re-deriving the draft (and its version) and clearing dirty/saveError. Bumping the version in refreshState, or only in the 409 branch, was rejected: either resurrects audit #7, since a full-document save would then revert a flip the draft never carried. storage.saveConflict now points at that action instead of implying a plain re-save works (all 23 locales). The escape lives in the shared hook, so the phone panel renders it too. * fix(client): deflake the queued-migration-waits-out-backfill panel test FE-ADMIN-STOR-037 seeds a running backfill, so the panel polls GET from first render — but the stub served the post-save world (files: off-box) before the PUT had landed, which no real server would do. A poll resolving while the confirm dialog was open made the confirm-time candidate recompute see from === to, so nothing was queued and the migration never POSTed. The GET stub now stays faithful (pre-save world until the PUT lands, usage included so candidates keep their object counts), and the test deterministically forces the once-flaky interleaving: it waits for polls to land while the dialog is open and asserts the candidate survives them. * fix(storage): surface the underlying cause in probe target errors The admin storage Test probe reported only the driver wrapper's outer message ("put failed for '…' on 'probe-s3'"), dropping the cause that says WHY — connection refused vs missing bucket. Flatten the error's cause chain into the probe result, appending an S3-style code when it isn't already in the message. * fix(storage): surface the underlying cause in replica-failure reports Same defect as the probe fix, one layer over: MirrorDriver's reportReplicaFailure captured only the wrapper's message, so replica-failure notifications said "put failed" without the why. Move the probe's cause-chain flattener into storage.types.ts as describeError and use it in both call sites. * fix(storage): strip ContentType from multipart Upload calls (MalformedXML) Every multipart S3 put carrying a contentType — every backup zip and photo over the 10MB threshold — failed with 400 MalformedXML. Upstream @aws-lite/s3's Upload leaks the ContentType param into its final CompleteMultipartUpload headers ({ ...xml, ...paramHeaders }), clobbering application/xml, which flips @aws-lite/client's payload serializer to JSON — so the server receives a JSON parts list where XML belongs. Confirmed live against MinIO AIStor; any S3 refuses the same. uploadMonitored now drops ContentType before s3.Upload. Multipart objects are stored without a content-type, which TREK never reads back (serving derives Content-Type from the key/caller, never object metadata); single-request PutObject paths keep sending it. Also closes the coverage gap that let this ship: the live contract suite never crossed MULTIPART_THRESHOLD, so s3.Upload had never run against a real server — it now round-trips a >10MB put carrying a contentType. --------- Co-authored-by: Maurice <mauriceboe@icloud.com>	2026-08-22 11:33:55 +02:00
Maurice	28765e220b	fix(places): keep the Google feature id, and give list imports the free address lookup ... Two halves of #1954, both about a place arriving through a list coming out thinner than the same place pasted as a single link. The feature id was never dropped on purpose. googleFtidFromMapsUrl only looked at the `?ftid=` query parameter, while a share link that resolveGoogleMapsUrl has followed to its final hop carries the id in the `data=!…!1s0x…:0x…` path blob — the same run the coordinates are already parsed out of two lines further down. So every pasted maps.app.goo.gl link lost it, and the Maps button opened a name search rather than the entry the link pointed at. The parser now reads the blob too, but only on a /place/ URL: a /maps/dir/ route carries one !1s per waypoint and the first is the origin, which would attach the wrong place. List imports never had a reverse lookup at all. The only pass behind them is the Google enrichment, which returns immediately without a Maps key, so an instance without one imported a list and left every address empty forever. They now get the same free Nominatim lookup the single-link path has always had: detached after the import answers, on the throttled background lane so it never queues in front of someone's search, through COALESCE so it can only fill an empty column, and each row pushed over the websocket so the sidebar fills in without a reload. Capped at 250 places — past that it stops being a backfill and starts being bulk geocoding. Not included: an osm_id for imported places. Single links do not produce one either, and one derived from a zoom-18 reverse lookup would name the building or the street rather than the POI.	2026-08-18 17:40:56 +02:00
Maurice	1a0834b866	fix(maps): resolve the Places key per instance, not per whoever is asking ... Searching worked for the admin and 403'd for every other member. getMapsKey read the calling user's own row first and then fell through to "any admin" via LIMIT 1 with no ORDER BY, so the admin who pasted the key hit their own row while everyone else got a non-deterministic stranger's. That fallback is what CLAUDE.md forbids. The Places and Unsplash keys bill the install, not the account that pasted them in, so they become instance configuration: an encrypted app_settings row, resolved as env -> instance -> the caller's own row, the same shape #1772 chose for the LLM endpoint. The per-user column stays as the last step because /me/api-keys has always accepted a personal key. The trailing migration carries an existing install's key into the instance row so nothing degrades to OpenStreetMap on upgrade, but only where that key already was the whole install's: as soon as a second row holds one, it is left alone, because the instance row outranks everybody's own column and an upgrade must not move a member onto someone else's key and bill. Two things from the same report: The key test button probed with field mask places.displayName while the search asks for places.rating and more, which is a separate SKU and can be restricted on its own, so the button could pass while every search failed. It now probes the resolved key with the search's own mask. Saving API keys wrote no audit line. It now writes one settings.api_keys_update row naming which fields changed, never their values, comparing cleartext because the random IV makes every ciphertext look different. Closes #1939	2026-08-17 14:48:23 +02:00
Maurice	3a0604b3f5	feat(config): put the operator-owned surfaces behind TREK_MANAGED (#1946) ... * feat(config): put the operator-owned surfaces behind TREK_MANAGED The switch landed last release with nothing behind it. This is what it turns off, plus the four fixes that came out of walking every one of these surfaces. Nothing here changes anything for an install that does not set the flag. That is not a hope, it is what managed-inert.test.ts asserts: the mode stays off for any unparseable value, no other variable can turn it on, every settings key stays writable, and splitManagedKeys hands back the same object rather than a copy so no write path can quietly diverge on self-hosted installs only. WHAT THE MODE WITHHOLDS Fourteen routes carry @ManagedForbidden: key validation, static MCP tokens, JWT rotation, the OIDC admin pair, restore and auto-backup, plugin sideload and reload, the Ollama model list and pull, and the SMTP test send. Each names the reason it is not the instance admin's, and the reason is mandatory. Eighteen settings keys are filtered instead of refused. SMTP, the OIDC toggles and the WebAuthn pair share one handler with keys the admin does own, and the llm_* family is reachable through per-user settings, the instance defaults and the addon config alike, so a route-level refusal would be both too coarse and incomplete. A locked key is skipped and reported as managed_keys; the rest of the save goes through. Refusing the request would take the registration toggles down with the SMTP block. GET /api/auth/me/settings stops returning the three API keys in the clear. It is the only endpoint that decrypts a stored credential back to the caller, which is fine when the admin pasted it in and wrong when the operator supplied it. Two multipart routes check in the handler rather than the guard: a guard throws before the parser drains the body and the client gets an ECONNRESET instead of the 403 (PROFILE-015). They still carry the marker so the boot gate sees them. WHAT KEEPS IT FROM DRIFTING validate-managed-routes.ts refuses to boot on a marking that is not on the reviewed list, a list entry whose route lost its marking, or a marking with no reason. The third case has no equivalent in the guard gate next door and is the one that matters most: a blank reason destroys the only record of why a control was taken away, and nothing recovers that later. It found two wrong entries in my own list while I was writing this. managed-keys.test.ts asserts that every key in ADMIN_SETTINGS_KEYS and DEFAULTABLE_USER_SETTING_KEYS belongs to exactly one of two sets. A key added to either list belongs to neither until somebody decides, and that is the failure this prevents: the lists grow, the mode does not. PLACES_API_BASE AND PLACES_API_KEY The nine Places endpoints spell out googleapis.com, so an install that wants those calls to leave through something of its own has no way to say so. One variable, substituted where every call already funnels through. A second one supplies the key, and when it is set the users table is not consulted at all — a per-user key would route around whatever the endpoint counts, and those columns are readable by any admin of the instance. FOUR FIXES THAT ARE NOT ABOUT THE MODE An admin setting somebody's password now bumps password_version and clears their MCP tokens. An admin sets a password for one reason, and until now verifyJwtAndLoadUser kept accepting every cookie the intruder already had: the one action taken to lock them out was the one action that did not. The four OIDC provider calls go through the SSRF guard. Three are worse than they look — the token, userinfo and JWKS URLs come out of the discovery response, so they are chosen by the other end, not by an admin. safeFetchLlm is renamed safeFetchAdminConfigured and keeps its old name as an alias, because it is the variant that allows loopback and LAN: a self-hoster with Authentik on localhost:9000 must keep working. The token exchange follows no redirect at all, since one would hand client_secret to a second host. DELETE /api/admin/users/:id/mfa clears another account's TOTP. The passkey half has existed since passkeys landed; the TOTP half never did, which left "somebody on the trip lost their phone" with no answer short of an operator in the database. Never for the caller themselves. TREK_PLUGIN_PERMISSIONS=off refuses to boot when the install is centrally administered. A warning is right for a machine its owner chose to trust; where the person who set the flag and the people whose data is in the database are not the same person, fail closed. TEXTS Four strings lose the phrase "self-hosted" because they describe somebody else's software — an Ollama endpoint, an AirTrail instance, a service a plugin talks to — and read better without it in both modes. The one string that describes TREK itself keeps it, and the mode picks a second key instead, so self-hosters lose nothing. A grep gate in shared/ keeps the canonical locale honest, because the parity check compares the file set and the top-level keys and never looks at a value: re-wording a string is invisible to CI today, and so is putting it back. * feat(maps): supply the map credential with the instance when centrally administered MAPBOX_ACCESS_TOKEN joins PLACES_API_KEY as an operator-supplied value. SettingsService injects it when the settings are read, and picks GL as the provider if the reader is still on Leaflet, since a token is always present. The key moves from the customer set to the locked set. It is the one locked key whose value is public by design — a pk.* token has to reach the browser to be useful — so this is not about secrecy: a per-user save landing on top of the instance token would only break the map for that user. Both settings fields for it go away with it. * fix(auth): count the operator key when reporting whether Google features work has_maps_key only looked at the users table, but getMapsKey prefers the PLACES_API_KEY env over anything stored there. On an install that supplies the key that way and has an empty column, every lookup worked while the client was told there was no key and hid the search, the autocomplete and the place photos. Found by walking the running instance rather than the code: the map came up, the search box did not. * feat(admin): hide the settings a centrally administered install cannot act on A control that answers 403 is worse than no control: the AI section rendered the refusal as "Request failed with status code 403", which tells an admin they broke something rather than that the decision is not theirs. So the surfaces go rather than being disabled in place: - AI parsing and AirTrail leave the addon list, and the AI addon route now refuses outright. The endpoint, the model and what a document costs come with the instance, so there is no instance decision left — not even whether it runs. - Immich and Synology Photos leave it too. Both connect to a server the reader runs at home, and there is no route from here to one. - The SMTP block, the admin webhook and ntfy targets, and the passkey RP ID and origins follow the same rule: they configure something the operator owns, and a wrong RP ID on a shared parent domain reaches past this instance entirely. - The About tab goes. It is about the project — what TREK is, where to file a bug, where to support it — and is aimed at the person running the software. The version and the link to the source move to the sidebar footer, on both the desktop page and the mobile twin, because AGPL 13 wants that offer wherever people reach the software over a network. - The update check stops asking GitHub and reports nothing available. Separately, the version footers lose the "self-hosted" suffix everywhere, not just here. It read as a label on the software rather than on the install, and the version alone says what it needs to. * feat(config): refuse to boot centrally administered without an at-rest key Measured on a running instance rather than assumed: with ENCRYPTION_KEY unset, the at-rest key falls back to a file in the data volume, and backupService bundles that file into every archive it builds. The archive is downloadable by any instance admin, so the encryption ships with its own key in the same zip. The skip in backupService already handles this — it only omits the file when the key came from the env — but nothing made sure the env actually carried one. Provisioning sets it; forgetting it was silent. validateEnvAtBoot now carries a second kind of check next to the schema: cross- field rules that only apply with the flag set, for combinations where every value is individually valid and the result is wrong. Empty without the flag, so an install that never asked to be centrally administered cannot meet it. * feat(client): add an attachment point for install-specific screens Two empty exports on the client, an empty module on the server, and the four places that read them. Nothing here adds a route, a nav entry or an endpoint — the arrays are empty and the module contributes nothing, which managed-ext.test pins so it stays that way. Read unconditionally, never behind a flag. On the server a conditional import would tie the module graph to an environment variable, and the graph is built once per app while the flag is read live; on the client an empty array costs nothing to spread, so every build walks the same path instead of one build walking a path nobody tests. Anything attached here follows the same rules as the rest of src/nest and src/components: a real module with a controller and a provider, bodies through the Zod pipe, routes under the global guards, tests. * test(shared): move the wording gate out of src so shared typechecks again CI installs shared with its own dependencies only (npm ci --workspace shared), so there is no @types/node and the `node:fs` import inside src/ failed to compile. The tests passed the whole time — vitest supplies the module — which is why this only showed up in the typecheck step. The walk moves to scripts/, next to i18n-parity.mjs, and the spec imports it the same way that one does. It also runs on its own now (`node scripts/i18n-selfhost-wording.mjs`), which is what you want when CI names a line you have to go find. * test(server): repair the call sites typecheck:tests caught tests/ sits outside the build tsconfig, so none of this showed up in a green test run — which is exactly the gap `npm run typecheck:tests` exists to close. Three kinds, all mine: - Seven `import type` paths for RuntimeEnvService carried one `../` too many. Type-only imports are erased, so vitest never resolved them and nothing failed. - VersionCheckJob and PluginsController grew a constructor parameter and eleven hand-wired `new` calls kept passing the old count. The argument arrived as undefined and no assertion happened to read it. - ConditionContext grew `managed`, and the shared context object in the conditions suite did not. The condition also had no test of its own, so it has one now: both directions, since the registry uses `is: false` to keep a notice away and a notice that only belongs on a centrally administered install would use `is: true`.	2026-08-15 16:42:27 +02:00
Maurice	c9193e0621	fix(maps): answer a keyless place lookup with an empty result, not a 400 ... An instance without a Google key already runs on the OpenStreetMap stack: search falls back to Nominatim, autocomplete does too, photos fall back to Wikimedia. The two details lookups were the exception. Both threw a 400 with "Google Maps API key not configured" the moment they were handed a Google place id. A Google id has no OpenStreetMap equivalent, so without a key there really is nothing to resolve. That is an empty result, not a caller mistake. The callers already treat a null place as a miss and fall back to the text-search path, so they were doing the right thing with the wrong signal, at the cost of an error round-trip and a logged exception each time. It bites whenever a place that was saved while a key existed is opened after the key is gone, and on any instance that never had one but imported Google-sourced places from a shared trip. The declared return type said the place was non-nullable while the OSM branch could already return null and the wire contract has always had it nullable. It now says what it does.	2026-08-14 23:42:51 +02:00
Maurice	f02a4b5654	feat(maps): bill a place search as one session instead of per keystroke ... Google charges autocomplete per request unless the requests carry a session token. With one, the keystrokes of a single search and the details lookup that ends it are charged together, which is also how a user experiences it: they looked up one place. TREK sent no token at all, so every debounced keystroke was its own billable request. On a busy trip that is the largest line on the Places bill, and it is paid by whoever owns the key, which for a self-hosted instance is the person running it. A session opens with the first suggestion request and closes when the user picks a result. It has to close on the abandoned paths too: a token carried into the next search would merge two searches into one billed session and scope the suggestions wrongly. So the search box clearing it, the text-search fallback and the error paths all end it. The token is validated on the way in. The autocomplete body goes through the Zod pipe, the details query does not, so the controller shape-checks it there. Anything that does not match Google's URL-safe 36-character form is dropped rather than forwarded, which falls back to per-request billing instead of breaking a lookup. A cached details hit never reaches Google and so never closes its session. That is billing-neutral: an unclosed session is charged as a plain autocomplete session, which is still cheaper than the per-request behaviour it replaces.	2026-08-14 23:42:51 +02:00
Maurice	a39c31cb48	Hide permanently closed places from search (#1887) ... A place that has shut down for good is never the answer to "where should we go", so it should not be in the results at all. Google knows this and says so in businessStatus, we just never asked for the field. It goes into the field mask and CLOSED_PERMANENTLY drops out. CLOSED_TEMPORARILY stays - a restaurant on holiday next month is still worth planning around - and results without the field are parks and viewpoints rather than businesses, so they stay too. OSM states it a few different ways. Re-tagged objects (disused:amenity, abandoned:shop) never matched the selectors to begin with, but plenty keep their original tag and gain a marker instead, and those did come back. Same for opening_hours=closed/off.	2026-08-11 17:04:09 +02:00
Maurice	3a7c8e1d26	Make a bind failure fatal again, and undo four contract tightenings ... An adversarial audit over the three merged migration PRs. Sixteen findings survived triple refutation; these are the ones with a defect behind them. The serious one is mine, from the gateway move. ws registers `error: this.emit.bind(this, 'error')` on the http server it attaches to, and the adapter now runs inside buildApp, before listen(). So from that moment the http server HAS an error listener and Node stops throwing on one. The adapter's handler was `() => {}`, which meant EADDRINUSE was forwarded there and dropped: the process survived, the listen callback never ran, so the banner never printed and the scheduler never started. A container that looks healthy and serves nothing. Reproduced, then fixed at both ends: index.ts owns the fatal decision on the http server, and the adapter logs instead of swallowing. Four DTOs were tightening contracts rather than describing them, which is the thing the pipe-before-handler ordering makes easy to do by accident: - OauthClientCreateDto replaced 'Name is required' with a Zod field report, and pre-empted the handler's 403 on an instance with the MCP addon off. - BookingImportConfirmDto demanded per-item fields the service never reads, so a caller assembling items itself got a 400 about `source`. - TagCreateDto/TagUpdateDto typed name and color as strings; the legacy route bound whatever arrived, so a numeric name used to create a tag. - PluginUserSettingsUpdateDto typed config as a record, where the handler answers 200 for anything it cannot use. Two of my own tests were false-green, which the audit demonstrated by mutation rather than by reading. WSGW-008 asserted on frames that never carry the user object at all, so it passed with the password_version strip deleted; it reads the retained identity now. GEO-009 made the same assertion before and after the write; it distinguishes a cached negative from a miss now, through a hasCached seam that says what getCached cannot. Both were re-run against the mutation they exist to catch, and both fail on it. booking-import geocodes every venue and endpoint in one request loop, up to thirty sequential calls. Folding maps onto the shared Nominatim cursor put that loop on the interactive lane, so an import starved every concurrent keystroke for half a minute. It takes the background lane now, which is what the two-lane split was built for. Also: the README's in-container bridge list had outlived one entry and was missing another, and the journey e2e still mocked a module this branch deleted.	2026-08-11 09:58:37 +02:00
Maurice	328b5fcb2b	One Nominatim client for atlas and maps ... Atlas throttled to the 1.1s OSM asks for and cached its answers in a 50k map; maps did neither, so five interactive routes fired straight at the service while the two atlas backfill loops politely waited their turn. From Nominatim's side that is one instance ignoring the rate limit. Both go through geo/nominatim.client.ts now. Three things the naive merge would have got wrong: - The shared client sends UA, the instance-identifying User-Agent, not the generic string atlas used. A shared generic UA is throttled harder. - One cursor across five interactive routes and two background loops would have put a keystroke behind however many places the backfill enqueued, so there are two lanes over the one cursor and background yields. - The cache key carries the query shape. Atlas asks at zoom 3 and 8, maps at 18; a coordinate-only key answers a street lookup with a country. The throttle cursor and the cache stay module state, the way the permissions and FX caches do, because the bridges build their collaborators outside the container and a second copy of the cursor is the bug being removed here. GeocodingService owns only the cleanup timer's lifecycle, which atlas used to start as an import-time side effect. Also folds the three haversine implementations into common/geo.ts, keeping the clamped form: asin of a value a hair over 1 is NaN, and a NaN distance fails every comparison it feeds instead of throwing. tests/setup.ts zeroes the interval. Nineteen maps cases drive these paths back to back with real timers and were sleeping nineteen seconds for it.	2026-08-10 17:08:54 +02:00
Maurice	ad6640cdb2	Fill the gaps Google leaves from the free sources	2026-08-10 12:24:15 +02:00
Maurice	0cd67675c5	Describe the chain when nothing describes the branch	2026-08-10 12:24:15 +02:00
Maurice	e321bfcf1b	Match a place name across its inflected forms, and forget an empty answer quickly	2026-08-10 12:24:15 +02:00
Maurice	880e258438	Show a week of opening hours as a week	2026-08-10 12:24:15 +02:00
Maurice	0c8788ce44	Give a Google place its free half back	2026-08-10 12:24:15 +02:00
Maurice	8f9af76204	Ask the sources that know the place before the ones that know the area	2026-08-10 12:24:15 +02:00
Maurice	f5a483da64	Get the pictures right and stop reusing stale answers ... Three things you could see on dev1. Pictures were "roughly right" rather than right, because a coordinate search only knows what was photographed within 300 m. Wikidata's P18 is one picture somebody chose to represent that exact object - "Zurletzteninstanz.jpg" for the restaurant, not a statue down the road - so it now leads, then the Commons category, and geosearch is the last resort it should always have been. Small venues carry a wikidata tag far more often than a Commons category, so this is also what reaches the places the other sources miss. Descriptions stayed on Wikipedia after the switch to Wikivoyage because the week-long result cache predates it. It carries a format version now; a release that changes the shape or the sources invalidates its own entries instead of serving last week's for another six days. Ratings and opening hours were being fetched from Google and then dropped on the floor - the fact list only looked at OpenStreetMap tags. Both are shown now, whichever provider supplied them. Also: the column follows a place opened from the map, not only one picked from search. A POI tap arrives as prefillCoords and never touched the selection, so the column kept showing the previously added place. Pictures are a tight three-column grid and only the picture in play is credited in full, on one line. Crediting all of them at two lines each drowned the column, and the licence attaches to the one that gets used - whose credit also stays visible in the inspector after saving. The request carries the place record the dialog already fetched, so the server skips a details lookup it was making in parallel with the client's own. That lookup was measured at 12.8 s for a large OSM relation.	2026-08-10 12:24:15 +02:00
Maurice	d98d597581	Make the detail column useful for places no encyclopaedia describes ... A restaurant got nothing: no wiki article, and a coordinate search around it returns whatever was photographed nearby rather than the place itself. Three changes, all on free sources: OpenStreetMap facts. A tagged restaurant carries its cuisine, its opening hours, a link to its menu, outdoor seating, step-free access and diet tags. Those arrive with the details lookup the dialog already makes, so showing them as chips costs nothing and is the single biggest gain for exactly the places that were empty before. Only positive facts are listed - "no outdoor seating" is noise. Commons categories. Where a place has a wikimedia_commons tag, its category is pictures OF it; geosearch falls back in when the category is empty. Wikivoyage before Wikipedia. Same API, same licence, but it describes a place for someone about to visit rather than opening with area in square kilometres. Two sentences instead of three, because this sits beside a form. The column also follows the section-heading idiom Vacay and the dashboard share (uppercase overline, wide tracking), its header matches the Saved places panel beside it, and a lead picture with a row underneath replaces the two-column grid that left a hole on every odd count. Credits wrap over two lines instead of truncating - a cut-off name credits nobody. Saved places stops after ten with a Show more button; search and filters still run over everything, not just the visible page. Form spacing is one rhythm (12px between groups, 8px inside, 4px label to field) and Notes matches Description in height. The OSM notice fits on one line and no longer claims there are no photos, which stopped being true when the column started fetching them from Wikimedia. The clipboard hint is gone; paste still works.	2026-08-10 12:24:15 +02:00
Maurice	1f1ed630fc	Fetch the enrichment pictures concurrently ... The column showed "details could not be loaded" for every place. The endpoint was answering correctly, just after ten seconds, and the shared axios instance gives up at eight. The downloads sat in a loop: five Commons hits meant five round trips and five decodes in sequence, even though the slot limiter already allows five at once. Both provider listings now run together and the bytes are fetched concurrently, which brings a warm request back to about a second. Candidate keys move with it. They now number the provider's list rather than the surviving strip, so a picture that fails to download leaves a gap instead of renumbering the ones behind it - a key has to point at the same file on the next request or the cached result goes stale against the photo cache. The client asks for enrichment with a longer timeout, and the Wikimedia calls carry one of their own: a cold TLS handshake from a fresh container was measured at eight seconds on its own, and no pictures is a better answer than a request that never comes back.	2026-08-10 12:24:15 +02:00
Maurice	23e76cc3ec	Let the maps providers hand back candidates, not just one picture ... Commons geosearch already returned five files and we threw four away, so a whole strip costs nothing extra. It now comes back with the licence metadata too: without LicenseShortName and the file description page there is no way to credit a CC BY-SA image, and a picker full of them needs that. Wikipedia descriptions resolve through the OSM wikipedia tag rather than guessing the article from the place name, which lands on the wrong article for anything ambiguous. Google photo references and photo bytes are separate calls now so a caller can list without downloading, and the editorial summary has its own lean field mask - asking for it through the expanded details would drag reviews along and move the call into the pricier tier.	2026-08-10 12:24:15 +02:00
Maurice	c7fed4b1f8	refactor(server): make the place-photo cache a provider (#1833) ... The last file in be-places-helpers-fold, and the one that had to be a provider rather than a move: it carries three in-memory structures that only do their job if the process has exactly one of them. recentFailures stops a screenful of markers hammering an API that is down inFlight stops a stampede when several requests miss the same placeId knownOnDisk saves an existsSync per cache hit A second instance gives the stampede guard its own inFlight map, so two simultaneous requests for the same uncached place both call the provider. That is why the scheduler takes this one from the container instead of constructing it: SchedulerDeps gains a placePhotos entry alongside backups, filled from app.get() in index.ts — the shape PR #1827 was built for. The module-scope mkdirSync becomes onModuleInit, and GOOGLE_PHOTO_DIR is read per call through RuntimeEnvService rather than frozen at import. That is not cosmetic: TREK_PLACE_PHOTO_DIR is the seam the tests use to keep off the real uploads tree, and its own test used to have to import the module dynamically to set the variable in time. The __dirname fallback moved a level with the file and is corrected accordingly. Consumers: maps, places and share inject it; places.helpers.reclaimPhotoCache takes it as a parameter, since it is a free function. Six test seams follow. Four path mocks become injected stubs with the same mock functions behind them; the collections case keeps the REAL implementation, because what it asserts is that removeIfUnreferenced counts collection_places as a reference (#1081) — a stub there would assert nothing. The share e2e overrides the provider in its container. services/ is down to 14 top-level files, from 44 when this started.	2026-08-07 20:27:19 +02:00
Maurice	6dda05f672	refactor(server): move the stateless helpers into nest/common (#1823) ... Eight files in services/ hold no state and touch no database: avatarUrl, conflictResult, demo, passwordPolicy, timezoneService, cookie, and the two crypto helpers. They are imported from 65 places, 62 of them already inside src/nest, so having them live in the legacy layer was pure inertia. They stay plain functions rather than becoming providers. Three importers sit outside the container and would have no way to reach an injectable one: db/migrations.ts runs before the app is built, and middleware/auth.ts and middleware/mfaPolicy.ts run beside it. Wrapping these in DI would buy ceremony and nothing else. mfaCrypto and apiKeyCrypto move together into common/crypto — they read the same ENCRYPTION_KEY, and the directory says so. services/ drops from 44 files to 36, 9,870 lines to 8,493. One trap worth recording: src/app-config and src/nest/app-config both exist. cookie.ts imported '../app-config' from services/, which resolved to the former; after the move the same string silently pointed at the latter, which has no readEnv. It is now explicit about which one it wants. Nothing else changes — no signatures, no behaviour, no tests rewritten beyond their import paths.	2026-08-07 15:14:14 +02:00
jubnl	4f0f80ad69	chore: move place backend to proper DI ... Fold the 1029-line services/placeService.ts into the DI-native PlacesService: the CRUD + ratings SQL, the GPX/KML/KMZ importers and the Google/Naver list importers become methods over the injected DatabaseService, byte-identical statements, COALESCE semantics and the If-Match conflict protocol (#1135). The pure half — frozen XML parsers, the KMZ unpacker, the dedup predicates, the Google hex-id parsers, reclaimPhotoCache — moves to places.helpers.ts (maps.helpers precedent). The 10-tool mcp/tools/places.ts registrar plus the trek://trips/{tripId}/places resource move onto the decorator registry as places.mcp.ts. search_place travels with them because its gate is places:read, not the read-only geo group, and now injects MapsService. TripsService, DaysMcp, BookingImportService and the plugin RPC host (PluginHostDepsFactory's 21st constructor dep) all inject PlacesService, so the domain needs no bridge at all. The two assignments.bridge imports that places.mcp.ts keeps are a deliberate cycle break: AssignmentsModule imports DaysModule and DaysModule now imports PlacesModule for days.mcp.ts's place creation, so injecting AssignmentsService there would close DaysModule -> PlacesModule -> AssignmentsModule -> DaysModule. Same seam reservations.mcp.ts uses, same trade trips.bridge.ts documents. Also folds services/placeEnrichment.ts in — its DB/websocket/Maps half becomes PlacesService methods over the injected DatabaseService/RealtimeService/ MapsService, its pure match selector joins places.helpers.ts — which retires nest/maps/maps.bridge.ts with its last consumer. Adopts places.dto.ts for the seven grandfathered body contracts and clears every PlacesController entry from body-contract-allow-list.ts. Two consequences worth knowing: the bespoke 400s 'Place name is required', 'ids must be an array of numbers' and 'URL is required' are now the ZodValidationPipe envelope, and because a pipe runs before the handler a malformed body 400s ahead of the trip-access 404 it used to follow — the same trade the todo and trips migrations took. placeBulkUpdateRequestSchema.ids drops .min(1) so the endpoint's empty-list short-circuit stays reachable. Fixes four defects the relocation had faithfully carried, each with a failing-first regression test: - create dropped lat/lng of exactly 0, so a place on the equator or the prime meridian lost its coordinates (`x || null` cannot tell absent from zero). - duration_minutes: 0 was unsettable — replaced by the 60-minute default on create, and read as "absent" by COALESCE on update. - the journey delete hook fired on unscoped ids: DELETE /:id, POST /bulk-delete and the delete_place MCP tool could detach another trip's journey entries for an id they then refused. All three scope first via the new PlacesService.scopedIds, mirroring the guard PluginHostDepsFactory already had. bulk_delete_places also moves its hook ahead of the DELETE — journey_entries.source_place_id is ON DELETE SET NULL, so running it afterwards left the entries as orphans. - the place search treated % and _ as LIKE wildcards; a bare % returned the whole trip. The three clauses now carry ESCAPE with the term escaped. Plus hardening on the Google list fetch: the response is capped at 8 MB (declared content-length and post-read length, the transit.service precedent) and its JSON.parse is guarded, so a malformed provider payload produces the existing 'Invalid list data received from Google Maps' 400 instead of throwing. Quirks deliberately preserved: the non-COALESCE route_color (#776), currency/transport_mode staying unclearable, importGpx returning null for a file that parses but yields no usable geometry, update skipping the If-Match check when the stored updated_at is null, and every string-valued `x || null`. Tests move with their case IDs: placeService.test.ts and the wrapper suite merge into unit/nest/places.service.test.ts, kmzUnpack into places.helpers.test.ts, placeEnrichment into places.enrichment.test.ts. The places e2e converts to the DI-native pattern (real SQL against temp-db DDL), tools-places.test.ts is kept and extended so it exercises the decorator path through the test registry, and the plugin-host and booking-import path mocks become constructor stubs.	2026-07-30 17:07:12 +02:00
jubnl	09f7cc4811	fix(maps): repair the quirks preserved by the DI fold	2026-07-29 20:27:30 +02:00
jubnl	4706112516	feat(server): adopt zod DTOs for the maps body contracts	2026-07-29 20:21:09 +02:00
jubnl	2e40c19249	chore: move maps backend to proper DI	2026-07-29 20:20:57 +02:00
Maurice	50600b74cd	fix(maps): stop the photo endpoint answering 404 for photo-less places (#1727) ... Opening one trip fires a 404 per place without a photo, 30 to 60 of them from one IP in seconds. CrowdSec and fail2ban read that as scanning; the reporter had three users banned within five minutes. A place having no photo is not an error and no longer answers as one. OSM places were also sent to Google as node:/way:/relation: ids, which can only return 400 INVALID_ARGUMENT — a billed round-trip and about 1.3s per place. getPlacePhoto already guarded coords: this way; the OSM prefixes now get the same treatment, as does getPlaceDetailsExpanded, which had no guard. The negative cache separates a place having no photo from a failed provider call instead of expiring both after five minutes and replaying the storm.	2026-07-29 15:14:21 +02:00
Konstantinos Thermos	e100c52676	fix(maps): localize map POI pin titles to the user's language	2026-07-24 11:36:30 +02:00
Maurice	ad893eb1cc	Release 3.1.0 (#1185) ... * Phase 0 — NestJS + Zod foundation harness (F1–F8) (#1050) Co-hosted NestJS app behind the existing Express server via a strangler-fig dispatcher, sharing the same better-sqlite3 connection and JWT httpOnly cookie. Additive and dormant: default routing stays on Express, Nest only serves its own /api/_nest diagnostics until a module opts in. F1 @trek/shared Zod contract package; F2 Nest bootstrap co-hosted (fall-through, single Dockerfile/port); F3 shared better-sqlite3 provider; F4 JWT cookie auth guard (+ @CurrentUser, admin guard); F5 Zod validation pipe + error-envelope parity; F6 Nest test + coverage gates; F7 per-prefix strangler toggle (env, default Express); F8 CI build/typecheck/test/coverage. Remaining F4/F6/F8 checklist items (trip-access + permission levels + MFA policy, e2e harness/seed + 80% gate, Nest↔Express parity test, Playwright PR-comment workflow) are tracked on the first consuming module cards (L1/A1/C1). * feat(weather): migrate /api/weather to the NestJS pilot module (L1) (#1053) First strangler migration (L1): /api/weather is served by a NestJS module. - @trek/shared/weather Zod contract; Nest controller byte-identical to the legacy Express route (paths, query params, status codes, { error } bodies, lang default, ApiError/500 passthrough). Service reuses getWeather/getDetailedWeather (+ shared cache; MCP tools unchanged). - Strangler routes /api/weather to Nest by default; the legacy Express route + its migration-time parity test were decommissioned in this PR. - Frontend (FE2): weatherApi typed against the @trek/shared WeatherResult contract. - Harness: reusable Nest-vs-Express parity harness, e2e harness (temp SQLite + seed/cookie helpers, real JwtAuthGuard), src/nest coverage gate raised to >=80%, src/nest test guide. - Verified end-to-end on a prod mirror (dev1): 401/400/200 via Nest with real Open-Meteo data, Express route gone. * fix(packing): multiply item weight by quantity in bag/total weight calcs (#898) Quantity now counts toward bag and total weights. Generalised to an itemWeight() helper used by every weight sum (bag totals + max, unassigned, grand total; sidebar + bag modal) with unit tests. * feat(i18n): add Korean (ko) translation (#977) Korean translation by @ppuassi, topped up to full en.ts key parity. Language registration follows separately. * feat(i18n): add Japanese (ja) translation (#829) Japanese translation by @soma3978, at full en.ts key parity, registered in supportedLanguages + TranslationContext. * Add Turkish (tr) translation + language registry (#1029) Turkish translation by @SkyLostTR, at full en.ts key parity, registered in supportedLanguages + TranslationContext. * i18n: register Korean + add Ukrainian translation (#1055) Korean translation by @ppuassi (#977) — now registered. Ukrainian by @JeffyOLOLO (#902) — lifted onto a clean branch. Both at full en.ts key parity (2258 keys). * chore: fix monorepo build pipeline and migrate shared to built package (#1056) * chore: fix monorepo build pipeline and migrate shared to built package - Root package.json: add workspace scripts (dev, build, test, test:cov, test:e2e) that delegate to actual scripts in shared/server/client workspaces - shared: add tsup build step (CJS + ESM dual output, .d.ts); consumers now import from the built dist instead of raw TS source via path aliases - server: replace tsc-alias with tsconfig-paths (tsc-alias mangled node_modules paths); fix MCP SDK path aliases to point to root node_modules (../node_modules) - server/scripts/dev.mjs: delay node --watch until tsc -w signals first-pass done, eliminating the spurious restart on every dev startup - client/vite.config.js + vitest.config.ts: remove @trek/shared path alias (no longer needed now that shared is a proper package) - Consolidate package-lock.json at the workspace root; drop per-workspace lock files * chore: fix test script to reflect root package.json * chore: add missing lint and prettier script in root package.json * fix(ci): build shared before tests; fix vitest MCP SDK alias paths vitest.config.ts aliases pointed at ./node_modules/ (server-local) but packages are hoisted to the root node_modules/ in the npm workspace — changed to ../node_modules/. CI jobs now install and build shared before running server/client tests so that @trek/shared's dist/ exists when vitest resolves the package. * fix(docker): update Dockerfile and CI for monorepo workspace structure Dockerfile: - Add shared-builder stage that produces @trek/shared dist before client and server stages need it - Each build stage carries root package.json + package-lock.json so npm can resolve @trek/shared as a workspace dependency - Production stage installs via workspace context (npm ci --workspace=server --omit=dev) so node_modules/@trek/shared symlinks to shared/dist correctly - Copy server/tsconfig.json into the image so tsconfig-paths/register can find the MCP SDK path aliases at runtime - CMD cds into /app/server before starting node so tsconfig-paths baseUrl resolves and ../node_modules points to /app/node_modules - Remove mkdir for /app/server (now a real dir); keep symlinks for uploads/data docker.yml version-bump: - Replace manual per-workspace cd+npm-version calls with single: npm version --workspaces --include-workspace-root --no-git-tag-version (mirrors the version:* scripts in root package.json) - git add now references root package-lock.json; adds shared/package.json .dockerignore: add shared/dist package.json: fix version:prerelease preid (alpha → pre) * fix(tests): use in-memory SQLite per worker in test mode vitest pool:forks spawns parallel worker processes that all called initDb() on the same data/travel.db, causing SQLite "database is locked" and "duplicate column name" races. When NODE_ENV=test each fork now gets an isolated :memory: DB so migrations run independently with no file contention. * chore(ci): add ACT guards to skip DockerHub steps in local act runs act sets ACT=true automatically. Guards added: - docker login: if: ${{ !env.ACT }} - build outputs: type=docker (local load) when ACT, push-by-digest when CI - digest export/upload: if: ${{ !env.ACT }} - merge job: if: ${{ !env.ACT }} - release-helm job (docker.yml): if: ${{ !env.ACT }} - version-bump git push (docker.yml): wrapped in [ -z "$ACT" ] shell guard Run locally with: ./bin/act -j build -W .github/workflows/docker.yml \ -P ubuntu-latest=catthehacker/ubuntu:act-latest * fix(ci): move ACT guards to step level; add guards to security.yml env context is invalid in job-level if conditions — moved all ACT guards down to individual steps. Also guards docker login + scout in security.yml so act can run the build-only part of that workflow. * fix(ci): skip git fetch and tag logic in act (no remote access in local containers) * Revert "fix(ci): skip git fetch and tag logic in act (no remote access in local containers)" This reverts commit 67cf290cda. * Revert "fix(ci): move ACT guards to step level; add guards to security.yml" This reverts commit f92b95e054. * Revert "chore(ci): add ACT guards to skip DockerHub steps in local act runs" This reverts commit 797183de08. * fix(docker): add musl optional deps so alpine builds find native rollup/sharp binaries npm prunes libc-constrained optional deps to the host libc (glibc) when generating the lockfile, leaving no musl entry for Alpine containers. Declaring the x64/arm64 musl variants as explicit root optionalDependencies forces them into the lockfile so npm ci on Alpine can install them. Covers shared-builder (tsup/rollup) and client-builder (vite/rollup + sharp icon generation) for both linux/amd64 and linux/arm64 CI targets. * fix(docker): copy client dist into server/public so the server resolves static files correctly The server runs from /app/server and serves static files relative to that directory, so the client build output must land at /app/server/public, not /app/public. * feat(planner): real road routes (OSRM) with travel-time connectors (#1060) * feat(planner): real road routes (OSRM) with travel-time connectors Replace the straight-line "as the crow flies" route with real OSRM road geometry (FOSSGIS routed-car/-foot) and an Apple-Maps style render (blue casing under a lighter core) on both the Leaflet and Mapbox GL maps. Routes are off by default and toggled per session, with a driving/walking mode switch in the day footer. Each day shows per-segment travel time/distance connectors between places, computed from the OSRM legs and split at transport bookings. Also redesigns the day header for visual consistency: vertical number+weather capsule, name with a divider before the date, subtle hotel/rental pills that stay on one line, and a hover-revealed 2x2 action square (edit / add transport / add note / collapse). Drops the Google Maps button. * test(planner): update route hook tests for calculateRouteWithLegs * remove route_calculation setting, always use OSRM routing (#1064) The per-user route_calculation toggle was a second, hidden on/off layer on top of the day footer's show-route button, and made it easy to end up with straight-line routes for no obvious reason. Drop the setting entirely: routing is always on, the footer toggle stays the single switch. Old stored values are simply ignored (settings are key-value, no migration needed). * chore: move i18n to shared package (#1066) * chore: move i18n to shared package * chore: move server translations to shared package and apply linter and prettier on entire shared package * feat(dashboard): upcoming reservations endpoint + travel-stats country/distance Adds GET /api/reservations/upcoming for the dashboard widget, switches travel-stats to the same country source as Atlas (manual + place-derived, ISO codes), and a distance service for flown km. * i18n(dashboard): dashboard keys across locales * feat(dashboard): boarding-pass hero, atlas row, live widgets + modal portal fix Reworked dashboard layout: boarding-pass hero with hover + days-left countdown, atlas stats row with real flags, searchable currency widget, editable timezone widget, new-trip FAB. Modals now portal to document.body to avoid inheriting dashboard-scoped button/font styles. * i18n(dashboard): sync all locales to one key set + German copy-dialog strings Brings every locale's dashboard namespace to the same 149-key set (missing keys backfilled from English) and translates the previously English-only copy-trip dialog into German. * refactor(dashboard): replace hardcoded strings with i18n keys Hero, atlas row, trip cards, filters, currency and timezone widgets now resolve all visible copy through t() instead of hardcoded English/German. * feat(i18n): add Greek translation (#1061) * i18n: complete Turkish (tr) translation (#1075) Fill in the remaining ~2100 UI strings in shared/src/i18n/tr so Turkish matches the English catalog. Brand names, URLs, and technical placeholders are left untranslated by design. * chore: prettier + lint * chore: enforce prettier & lint on shared package * feat: Updated border of map markers to reflect category color. (#1062) * feat(dashboard): mobile layout, glass UI, context bottom nav + OIDC PKCE (#1079) * feat(dashboard): mobile layout, glass tiles, plain-text countdown, place photos - Rework the mobile dashboard: cover hero, separate boarding-pass card, trimmed atlas (trips + days only), stacked widgets - New floating bottom tab bar with a centred context-aware + button (new trip / place / journey / entry depending on the page) - Move profile + notifications into a small top strip on the dashboard - Desktop: glassmorphic tiles (light + dark), neutral dark palette, plain-text countdown module, real place photos in the boarding pass * i18n(dashboard): translate new dashboard keys across all locales Fill the dashboard-rework keys (hero, atlas, fx, tz, upcoming, copy dialog, aria labels, countdown) that were left as English placeholders, plus the new startsIn/aria keys, for all 19 languages. * feat(oidc): send PKCE (S256) in the OIDC login flow The OIDC client now generates a code_verifier per login, sends the S256 code_challenge on the authorize request and the code_verifier on the token exchange. Works whether the provider has PKCE optional or required (fixes login against providers that require PKCE, e.g. Pocket ID). * Migrate TREK 3 to NestJS + React 19 (shared Zod contracts) (#1087) * Migrate TREK 3 to NestJS + React 19 with a shared Zod contract layer Brownfield strangler migration of the backend onto NestJS modules (auth, trips, days, places, assignments, packing, todo, budget, reservations, collab, files, photos, journey, share, settings, backup, oidc, oauth, admin, atlas, vacay, weather, airports, maps, categories, tags, notifications, system-notices) served through a per-prefix dispatcher, keeping the existing SQLite/better-sqlite3 DB and JWT httpOnly cookie auth, with behavioural parity for every route. Client: React 19 upgrade, "page = wiring container + data hook" pattern across all pages, per-domain Zustand stores bound to @trek/shared contracts, and decomposition of the large components (DayPlanSidebar, PackingListPanel, CollabNotes, FileManager, MemoriesPanel, PlacesSidebar, CollabChat, SystemNoticeModal, BudgetPanel, PlaceFormModal, ...) into focused render units backed by in-file hooks. Apply the shared global request pipeline (helmet/CSP, CORS, HSTS, forced HTTPS, the global MFA policy and request logging) to the NestJS instance as well, so a migrated route is protected identically to the legacy fallback rather than bypassing it. * Finish the NestJS migration — drop the legacy Express app NestJS now serves the whole surface: every /api domain plus the platform routes (uploads, /mcp, the OAuth/MCP SDK + /.well-known metadata and the production SPA fallback). Removed server/src/app.ts, all of server/src/routes/* and the strangler dispatcher; index.ts and the integration suite share a single buildApp() bootstrap so prod and tests can't drift. - Platform/transport routes extracted to nest/platform/platform.routes.ts and mounted before app.init() — Nest's router answers an unmatched request with a 404, so a route registered after init is never reached. The SPA fallback is a NotFoundException filter and the catch-all uses a RegExp (Express 5's path-to-regexp rejects a bare '*'). - New modules: memories (/api/integrations/memories — the Journey gallery's Immich/Synology proxy), addons (GET /api/addons) and the cross-trip GET /api/reservations/upcoming. - TrekExceptionFilter reproduces the old multer / err.statusCode handling so upload rejections keep their 400/413 { error } body and non-ASCII filenames survive (defParamCharset). - addTripToJourney and the MCP get_journey_share_link tool gained the trip-access check they were missing. - Re-pointed the 34 integration tests + the websocket test onto the Nest app; removed the now-meaningless Express-vs-Nest parity tests and a few orphaned client components. * Restore the reset-password rate limit and fix copyTrip reservation links Two correctness/security gaps the NestJS migration introduced: - POST /api/auth/reset-password lost its per-IP rate limiter. Restore it (5 attempts / 15 min on a dedicated bucket, same as the old resetLimiter) so reset tokens can't be brute-forced unthrottled. Covered by AUTH-019. - copyTripById did not copy reservations.end_day_id (a day reference — now remapped through dayMap like day_id) or needs_review, so a duplicated trip lost multi-day transport end-day links and reset the review flag. * Clean up dead code, dedupe helpers, fix the reset-password contract - Remove server exports orphaned by the Express removal: the immich album-link helpers, seven route-only service exports, getFileByIdFull; de-export internal-only helpers (utcSuffix). - De-duplicate verifyTripAccess (9 identical copies -> services/tripAccess.ts) and avatarUrl (3 -> services/avatarUrl.ts); name the bcrypt cost (BCRYPT_COST) and the email regex (EMAIL_REGEX). Public API unchanged. - resetPasswordRequestSchema declared `password`, but the client sends and the service reads `new_password` — rename it so the contract matches and the client types resolve. - Make ATLAS-013 deterministic: stub the admin-1 GeoJSON download instead of fetching ~4600 features from GitHub during the test (it hung the suite). * Make the client typecheck runnable (vitest/vite ambient types) The client had no `typecheck` script and tsc couldn't even start (the baseUrl deprecation errored out, same as server/shared already silence). Add `ignoreDeprecations: "6.0"` to match the other workspaces, a `typecheck` npm script, and a src/vite-env.d.ts referencing vite/client + vitest/globals so tsc knows the test globals (describe/it/expect/vi). This turns ~3600 phantom "Cannot find name" errors into a real, measurable count (~590 actual type errors remain, to be worked down). Type-only; no runtime change. * Derive client domain types from the shared schema contracts Add entity/response Zod schemas to @trek/shared (place, trip, assignment, day, budget, packing, reservation), each matched against the producing server service, and re-export them from client types.ts instead of the hand-written duplicates that had drifted (name/title, amount/total_price, owner_id/user_id, cover_url/cover_image, ...). Updates the call sites and test fixtures the corrected types surfaced; type-only, no runtime behaviour change. * chore(db): log swallowed errors in addon-disable migration + guard against destructive migrations The migration that disables the legacy "memories" addon swallowed any error in an empty catch, as did ~30 other catch blocks in the migration runner (column adds, the journey rebuild, index probes). Replace each silent catch with the existing console.warn('[migrations] ...') log so failures are visible. Control flow is unchanged: every step stays non-fatal, nothing new is thrown. Add a static guardrail test that scans the migration source and fails when a new destructive statement (DROP TABLE / DROP COLUMN / TRUNCATE / DELETE FROM / ALTER ... DROP) appears outside a reviewed allowlist, and when an empty/silent catch block is reintroduced. The existing destructive statements are all legitimate table rebuilds or bounded cleanups and are recorded in the allowlist with a reason. * Re-check SSRF on every redirect hop when resolving short links Replace the one-shot checkSsrf + fetch(redirect:'follow') in the maps and place short-link resolvers with safeFetchFollow, which follows redirects manually and re-runs checkSsrf against the DNS-pinned IP of each hop (max 5). A redirect to an internal/loopback address is now blocked even when the initial URL is public, while legitimate cross-host redirects (goo.gl -> maps.google.com) still resolve. * Reject WebSocket tokens minted before a password change Stamp the user's password_version onto the ephemeral ws token and verify it on connect, closing the socket (4001) when it no longer matches, so a token issued before a password reset can't be replayed. Tokens minted without a version are treated as version 0, matching the JWT pv-claim semantics. * fix(i18n): guard locale key parity and finish the OAuth consent page strings Every non-en locale now exposes the exact same flat key set as en. Keys that had drifted out of sync are backfilled with the English source value (tagged en-fallback) so t() resolves a real string instead of relying on the silent runtime fallback; no existing translation was touched and no key was removed. Add a parity test that imports each aggregated locale bundle and asserts its key set matches en, with a diagnostic listing of any missing/extra keys. This complements the file-level check in shared/scripts by guarding the merged export the app actually serves. Finish internationalising OAuthAuthorizePage: the ~15 remaining hardcoded English chrome strings now go through oauth.authorize.* keys (English source in en, en-fallback placeholders elsewhere). Markup and behaviour are unchanged. * Add semantic theme color tokens to Tailwind Map the CSS theme variables from src/index.css (:root light / .dark dark) to named Tailwind utilities — bg-surface, text-content, border-edge, bg-accent and their variants. This gives components a Tailwind-native target for the theme colors so we can replace inline `style={{ ... 'var(--...)' }}` with utility classes without changing the rendered values. * Surface silent store failures to the user and validate API responses in dev Reservation toggle, todo/packing toggle and budget reorder were swallowing API errors after rolling back, so the user saw the change silently snap back with no explanation. Route those failures through the existing toast channel (new store/notify.ts bridges to window.__addToast, the same channel SystemNoticeBanner uses); the reservation toggle re-throws so ReservationsPanel's own translated toast finally fires. Also wire the existing parseInDev/checkInDev response validation into the maps and notification-test endpoints to catch contract drift in dev. * Migrate static theme inline styles to Tailwind utilities and extract page sub-components Replace the static, color-only inline `style={{ ... 'var(--bg-primary)' ... }}` props with the new semantic Tailwind utilities (bg-surface, text-content, border-edge, ...) wherever the result is byte-identical; dynamic/conditional theme styles and hardcoded status colors are left inline. Extract the Atlas country-search autocomplete, the Admin update banner, and two Journey dialogs into their own presentational components to shrink the oversized page files, keeping behaviour and markup identical. * Remove the unrouted photos page and its dead photo components PhotosPage was never wired into the router and its usePhotos hook read a tripStore photos slice that was never implemented; the Photos gallery, lightbox and upload components were only reachable through it. Per-trip photos now live in the Journey gallery (Immich/Synology). Removed the dead page, hook and components — the live Journey PhotoLightbox is a separate component and stays. * Resolve the remaining client type errors and the trip.title navbar bug Drive the client typecheck to zero without any/ts-ignore: convert the tripId route param to a number once at the page boundary so it matches the numeric props and store actions it feeds, fix trip.name -> trip.title (the wire field is title, so the old read rendered blank in the files/offline views), and tighten the scattered handler-arity, DOM-cast and untyped-payload sites. No runtime behaviour change. * Convert the remaining dynamic and hardcoded inline styles to Tailwind utilities Second styling pass over the components and pages: move conditional theme colors into className ternaries (bg-accent / bg-surface-hover etc.), turn reused CSSProperties constants into className constants, and express static hardcoded hex/rgba colors as Tailwind arbitrary values so the exact rendered colour is preserved. Truly dynamic styling (computed geometry, gradients, multi-part shadows, data-driven colours, the undefined --sidebar/--nav layout vars) stays inline as it cannot be expressed as a static class. Updated three component tests that asserted the old inline active-state styles to assert the equivalent utility class instead. Verified: client typecheck 0, full client suite green, and a live light/dark render check in the dev server confirms the semantic theme tokens resolve correctly (the earlier 'transparent popups' were a stale dev server that pre-dated the tailwind.config token addition, not a code issue). * Add eslint flat-config for client and server and gate typecheck, lint and pages in CI client and server had lint scripts but no eslint config (only shared was linted in CI). Add flat configs mirroring shared's stack (js + typescript-eslint recommended + eslint-config-prettier) plus the client's react-hooks/react-refresh plugins. Pre-existing patterns in this never-linted code (explicit any, require() in the CommonJS server, empty catches, exhaustive-deps) are set to 'warn' rather than 'error' so the gate passes at 0 errors without a repo-wide reformat — these can be ratcheted to errors over time. Wire blocking typecheck + lint + lint:pages steps into the client and server CI jobs (now that both typechecks are clean) and promote the server typecheck from informational to blocking. * Decompose the remaining God Components into hooks, helpers and sub-components FE6: split the oversized page and panel components into thin layout shells plus colocated use<Component> hooks, .constants.ts, .helpers.ts (with tests) and presentational sub-components, following the established 'logic in a hook, render in slices' pattern. Behaviour, markup, classes and effect order are unchanged. Largest reductions: PackingListPanel 1598->42, FileManager 1055->36, AdminPage 1525->167, BudgetPanel 1266->146, JourneyDetailPage 2822->547, PlacesSidebar 945->66, CollabChat 861->106, CollabNotes 1417->532. DayPlanSidebar's drag-and-drop render body was left intact (ref-identity sensitive) and only its toolbar/modals/constants were extracted. * Fix duplicate React keys in the file-assign place list When a place is assigned to the same day more than once it appeared twice in a day's list, so the place-button key={p.id} collided and React warned about duplicate keys. Key by place id + render index so siblings stay unique. Pre-existing in the old FileManager; behaviour unchanged. * Format the shared package and drop an unused import to satisfy the lint gate The i18n and schema changes added code that wasn't prettier-formatted, and place.schema.ts imported categorySchema without using it. Run prettier over shared and remove the import so 'npm run lint' + 'format:check' pass. * Install all workspaces in the server CI job so SWC's native binary is present The server vitest config transforms via unplugin-swc, which needs @swc/core's platform-specific native binary. A workspace-scoped 'npm ci --workspace server' skips that optional dependency, so vitest failed to load the config on the Linux runner. Use a full 'npm ci'. * Re-resolve dependencies with npm install in the server CI job for SWC Full 'npm ci' still skipped @swc/core's Linux native binary because the committed lockfile was generated on Windows and lacks the Linux optional-dep install metadata. 'npm install' re-resolves and fetches the platform-matching binary, which the server's unplugin-swc transform needs to load vitest.config.ts. * Install @swc/core's Linux binary explicitly in the server CI job Neither npm ci nor npm install fetched @swc/core-linux-x64-gnu on the Linux runner because the lockfile was generated on Windows and lacks the Linux optional-dep metadata. Add a step that installs the matching @swc/core-linux-x64-gnu version (no-save, no-lockfile) so unplugin-swc can load the server's vitest config. * Use legacy-peer-deps when installing the SWC Linux binary in CI The explicit @swc/core-linux-x64-gnu install re-resolved the tree and hit the pre-existing lucide-react/react-19 peer conflict that the lockfile was generated around. Add --legacy-peer-deps so the step matches the project's resolution and installs the binary. * Keep the lockfile when installing the SWC binary so other deps stay pinned Dropping --no-package-lock made npm re-resolve the whole tree and upgrade eslint, whose newer recommended config flagged no-useless-assignment as an error in the server lint step. Keep the lockfile so only @swc/core-linux-x64-gnu is added and every other dependency (incl. eslint) stays at its locked version. * Fix a batch of reported bugs: Atlas regions, planner overlays, imports, Safari modals (#1094) * Start the Journey date picker week on Monday (#1078) The Journey entry date picker started the week on Sunday (firstDow = getDay(), headers Su-first) while every other picker (CustomDateTimePicker, VacayCalendar) starts on Monday. Align it: Monday-first leading offset ((getDay()+6)%7) and Mo-first weekday headers. * Fix Taiwan resolving to CN-TW in the Atlas country search (#1049) natural-earth gives Taiwan ISO_A2='CN-TW' (a subdivision-style value) with ADM0_A3='TWN'. The dynamic A2_TO_A3 augmentation added 'CN-TW'->'TWN', which then overwrote the legitimate TWN->TW entry in the reverse map, so Taiwan's country option resolved to 'CN-TW' — unresolvable by Intl.DisplayNames (no name, broken flag, not searchable). Only augment A2_TO_A3 with real 2-letter codes. * Drop empty leftover dateless days when a trip gets a shorter dated range (#1083) generateDays kept all unused dateless placeholder days after switching to an explicit (shorter) date range, so day_count (COUNT(*) FROM days) stayed inflated. Delete the empty leftovers (no assignments/notes/accommodations) like the dateless path already does, while preserving any that still hold content. Adds TRIP-SVC-017. * Render GPX and route overlays once the Mapbox style has loaded (#1036) The GPX and route geojson effects ran before the map 'load' event had attached their sources, so on the first paint they hit the early return and never re-ran. Add mapReady to their dependencies so they fire again the moment the sources exist. * Convert HEIC trip and journey covers to JPEG before upload (#1085) HEIC/HEIF covers coming straight off an iPhone could not be rendered in the preview or stored as a usable image. Route both cover pickers through normalizeImageFile, the same conversion the journal entry editor already uses, so the file becomes a JPEG before it leaves the browser. * Name GPX routes and tracks after their source file so multiple imports stick (#1054) Unnamed routes and tracks all fell back to the same generic 'GPX Route' / 'GPX Track' label, so the name-based import dedup dropped every one after the first - importing several files (or one file with several tracks) only kept a single place. Derive the default name from the source filename with an index suffix when a file holds more than one geometry, thread the filename down through the controller, and let the import modal take more than one file at a time. Adds PLACE-SVC-037/038. * Namespace the modal backdrop class so content blockers stop hiding it (#1027) Generic class names like .modal-backdrop sit on the cosmetic filter lists that content blockers (1Blocker, EasyList Annoyances) ship, and get hidden with display:none. The shared Modal - used by New Trip and Add Place - carried that class, so Safari users running such a blocker saw the modal silently fail to open with no error and no network request. Rename it to .trek-modal-backdrop. * Highlight GB regions by resolving England/Scotland/Wales/NI to finer admin-1 codes (#1067) A zoom-8 reverse geocode of a UK place only resolves to the constituent country (GB-ENG/SCT/WLS/NIR), but Natural Earth's admin-1 polygons for GB are counties and boroughs (GB-LND, GB-MAN, GB-CON, ...). Those four codes match no polygon, so places in England never highlighted in the Atlas while CH/IT/NL/etc. worked. When a GB lookup lands on a constituent country, re-resolve it at a finer zoom where Nominatim exposes the county/borough code the polygons actually carry. Other countries keep the exact zoom-8 behaviour. Adds ATLAS-UNIT-021. * Surface the real place-search error instead of a generic toast (#1092) When a place search or detail lookup fails, the backend already forwards the upstream reason - including descriptive Google Places API messages such as 'Places API (New) has not been used in project ... or it is disabled'. The planner discarded it and always showed 'Place search failed', so a key that is mis-enabled, unbilled, or pointed at the legacy API instead of Places API (New) looked like an unexplained silent failure. Show the server-provided message when present, and stop the Atlas bucket-list search from swallowing its error without a trace. * Await the async cover normalization in the TripFormModal paste test (#1085) handleCoverSelect now normalizes the pasted file before previewing it, so URL.createObjectURL is called a microtask later. The assertion moves into waitFor; a non-HEIC file still passes through unchanged. * fix(pwa): removed orientation from the manifest (#1058) * fix(journey): raise PhotoLightbox z-index above MobileEntryView (#1101) * feat(transport): add bus, taxi, bicycle, ferry and other transport types (#1105) Closes #718. Adds five new transport reservation types alongside the existing flight/train/car/cruise: bus, taxi, bicycle, ferry and a generic 'transport_other' catch-all. The new types are treated as first-class transports everywhere — the transport modal, day plan, route calculation, map overlays, file grouping and the PDF export — and are translated across all 20 locales. A dedicated 'transport_other' value is used for the catch-all so existing 'other' bookings are not reclassified as transport. * feat(reservations): native booking-confirmation import via KDE KItinerary (#1102) * feat(reservations): native booking-confirmation import via KDE KItinerary Adds a two-step preview → confirm flow for importing booking emails, PDFs, PKPass and HTML confirmations. The server invokes the KDE kitinerary-extractor binary, maps JSON-LD schema.org output to TREK reservation shapes, and persists via the existing createReservation pipeline (accommodations, budget, places, WebSocket broadcasts). - NestJS BookingImportModule: preview + confirm endpoints under /api/trips/:tripId/reservations/import/booking{,/confirm} - KitineraryExtractorService: spawns the binary, filters stderr noise, handles QDateTime (@value) timezone-aware datetimes - kitinerary-mapper: FlightReservation, TrainReservation, BusReservation, BoatReservation, LodgingReservation, FoodEstablishmentReservation, RentalCarReservation, EventReservation → typed preview items - BookingImportService: auto-creates place rows; geocodes venues without coordinates via Nominatim (name+address → address → name fallback); resolves day IDs for accommodation linking - BookingImportModal: drag-and-drop multi-file upload, preview cards with type icons, per-item exclude toggle, confirm step - Shared Zod contracts: BookingImportPreviewItem, PreviewResponse, ConfirmRequest, ConfirmResponse — consumed by controller, service, API client and modal - Dockerfile: node:24-trixie-slim runtime; amd64 downloads KDE static binary + locales; arm64 installs libkitinerary-bin + symlinks to fixed path; ENV KITINERARY_EXTRACTOR_PATH set for both arches - /api/health/features exposes { bookingImport: boolean } so the UI hides the Import button when the binary is absent - i18n keys (English), wiki docs, API.md, README one-liner * i18n: add booking import translations for all 19 non-English locales Adds 17 reservations.import.* keys and undo.importBooking to ar, br, cs, de, es, fr, gr, hu, id, it, ja, ko, nl, pl, ru, tr, uk, zh, zh-TW. * chore: enforce i18n parity * docs(wiki): add KItinerary local setup instructions to dev environment guide * feat(costs): rework Budget into Costs — Splitwise-style, multi-currency, mobile (#1106) * fix(journey): authorize reads of the journey share link GET /api/journeys/:id/share-link now requires journey access (canAccessJourney), matching the create/delete share-link routes and the get_journey_share_link MCP tool. Returns no link when the caller lacks access to the journey. * feat(costs): rework Budget into Costs — Splitwise-style, multi-currency, mobile Renames the Budget addon to "Costs" (UI only) and reworks it into a Tricount/ Splitwise-style cost tracker: multiple payers per expense, equal split across chosen members, settle-up with persisted history + undo, 12 fixed categories, per-expense currency with live FX conversion to a user-set display currency (Settings -> Display), and locale-correct money formatting. Adds a desktop and a dedicated mobile layout. A migration backfills existing budget items (single payer, split members, currency). Closes #551 (per-expense currency). Also switches the app font to self-hosted Poppins (Geist for secondary subtext), replacing the Google Fonts CDN dependency. * fix(costs): neutral dashboard dark palette + liquid glass, full page width, entry-count badge - Dark mode used a warm oklch palette that read brownish; switch to the neutral zinc tokens used by the dashboard (#121215 bg, #f4f4f5 ink) and add a subtle backdrop-blur glass on cards. - Costs now uses the full available page width on desktop instead of a 1280px cap. - Render the expense count next to the Expenses title as a badge. - Adapt budget/journey unit tests to the new payer-based settlement model and the Costs rename (category default 'other', Costs tab/CostsPanel). * fix(costs): drop the entry-count badge, always show row edit/delete actions Removes the count badge next to the Expenses title and makes the per-row edit/delete actions permanently visible (no longer hover-only) on desktop too. * feat(costs): currency-native money formatting, custom select/date, rename addon to Costs - Format every amount in its own currency convention (symbol position, grouping and decimal separators) regardless of app language, via a currency->locale map (EUR -> '12,00 €', USD -> '$12.00', JPY -> '¥12', ...). Previously Intl used the app locale, so EUR showed the symbol in front under an English UI. - Use TREK's CustomSelect (searchable, with symbols) and CustomDatePicker in the add/edit expense modal instead of the native <select>/<input type=date>. - Rename the 'Budget Planner' add-on to 'Costs' in the admin list (display only; id/tables/permissions/MCP stay 'budget') via seed + a migration for existing DBs. * feat(auth): configurable session duration via SESSION_DURATION Adds a SESSION_DURATION env var (ms-style strings: 1h, 7d, 30d, ...) controlling how long a session stays valid before re-login. It drives both the trek_session JWT exp claim and the cookie maxAge from one source, so they never drift. Invalid values warn at startup and fall back to the default (24h — unchanged). The MFA challenge token and MCP OAuth tokens keep their own TTL. Implements the request from discussion #946. Documented in the env-var wiki page, .env.example and docker-compose.yml. * feat: Passkey (WebAuthn) login (#1111) * feat(auth): passkey (WebAuthn) login — server endpoints, schema + admin toggle Add @simplewebauthn/server registration and primary (discoverable) login ceremonies under /api/auth/passkey, a webauthn_credentials + single-use webauthn_challenges schema (migration), the instance-wide passkey_login toggle (default off) enforced before auth by a guard, and require_mfa satisfaction via a verified passkey. RP ID/origin come only from server config (webauthn_rp_id/origins -> APP_URL), never request headers. * feat(auth): passkey enrolment, login button + admin settings UI PasskeysSection in account settings (add/rename/remove with a current-password step-up), a 'Sign in with a passkey' button on the login page, the admin enable + RP-ID/origins controls, and a per-user admin reset action. * i18n(auth): passkey strings across all locales Add login/settings/admin passkey keys to en and all 19 translated locales. * chore: update kitinerary version * Backend/frontend hardening & consistency cleanups (#1113) * refactor(auth): session token validation and password-change consistency * refactor(journey): entry field allow-list and public share-link consistency * refactor(mcp): align tool authorization with the REST permission checks * chore: input validation and sanitisation touch-ups (uploads, pdf, maps, backup, csp) * feat: optimize routes around accommodation, confirm note deletions (#1123) Optimize day routes around the accommodation When a day has an accommodation set, the route optimizer now treats it as the day's home base: it optimizes a loop that leaves the hotel and returns to it, so the stop nearest the hotel comes first. On a transfer day - checking out of one hotel and into another - the route runs from the first hotel to the second instead. The optimizer also gained a 2-opt pass on top of the nearest-neighbor ordering, which removes the crossings the greedy pass used to leave behind. A new display setting ("optimize route from accommodation", on by default) lets you turn the anchoring off. Confirm before deleting notes Deleting a plan note or a collab note now asks for confirmation first. On phones and tablets the edit and delete icons sit close together and were easy to mis-tap, which deleted notes with no way back. * fix: miscellaneous bug fixes (#1139) * fix(share): serve place thumbnails in shared trip links (#1100) Google-sourced place photos are stored as image_url pointing at the JWT-guarded /api/maps/place-photo/:placeId/bytes endpoint, so they 401 for an unauthenticated shared-trip viewer and render as broken images. Rewrite place image_url values in the shared payload to a public, token-scoped proxy (/api/shared/:token/place-photo/:placeId/bytes) and add an unguarded SharedController route that validates the token and that the place belongs to its trip before streaming the cached bytes. Mirrors the existing JourneyPublicController precedent. No client changes needed. * fix(atlas): replace Natural Earth with geoBoundaries for up-to-date regions (#1119) Atlas sourced country and sub-national boundaries from Natural Earth's GitHub `master` at runtime. That data is stale (e.g. it still shows Norway's pre-2020 counties such as Oppland/Hordaland) and depicts some contested territory in unwanted ways (nvkelso/natural-earth-vector#391), so Natural Earth is dropped entirely. - Country borders (admin0) now come from the geoBoundaries CGAZ composite; sub-national regions (admin1) from per-country gbOpen, which carries ISO 3166-2 codes. A new script (server/scripts/build-atlas-geo.mjs) normalizes and quantizes them into committed gzipped bundles under server/assets/atlas, read server-side at runtime (no network at boot, no GitHub CSP allowlist entry). - New GET /addons/atlas/countries/geo serves the country layer; the client fetches it from the API instead of GitHub. - A migration reconciles manually-marked visited_regions against the new bundle (valid code -> keep; region name still matches -> re-code; curated merge crosswalk for renamed reforms; else leave intact), with UNIQUE-safe dedup. bucket_list and visited_countries hold only invariant alpha-2 country codes, so they are untouched. - Attribution added (NOTICE.md + README) per geoBoundaries CC BY 4.0. Closes #1119 * fix(packing): make templates admin-only to create, usable by members Creating a packing-list template was gated only by trip access, so any trip member could create one from the Lists feature, while applying a template silently failed for non-admins because the apply dropdown was populated from the AdminGuard-protected /api/admin/packing-templates endpoint. - save-as-template now returns 403 for non-admins; the Save-as-Template button is hidden unless the user is an admin (both the TripPlanner toolbar and the inline packing header). - add member-accessible GET /api/trips/:tripId/packing/templates so the apply dropdown lists templates for any trip member; client fetches from it instead of the admin endpoint. Closes #1120 Closes #1121 * fix(packing): show bag tracking to non-admin members The global Bag Tracking toggle was only readable via the admin-gated GET /api/admin/bag-tracking, so non-admin trip members got 403 and the weight fields, bag circles, and BAGS sidebar never rendered (#1124). Surface the flag through the already-authenticated GET /api/addons (loaded into the client addon store on app start for every user); the packing hook reads it from the store instead of the admin endpoint. The admin write path stays admin-gated and unchanged. * Fix a batch of reported bugs (#1145) * fix(maps): fall back to OSM/Wikipedia for place photos and normalize non-standard language codes (#1137) * fix(auth): refuse password reset for OIDC/SSO-linked accounts (#1129) * fix(docker): ship server/assets (airports + atlas geo) in the runtime image (#1133, #1119) * fix(unraid): point the template at a PNG icon Unraid can render (#1073) * fix(offline): serve cached file blobs when offline or on network failure (#1046, #1069) * fix(map): centre the selected pin in the visible map area above the bottom panel (#1125) * fix(pdf): render persisted place-photo proxy URLs as images (#1130) * fix(planner): show the selected place category in the edit form (#1134) * fix(dashboard): collapse list-view trip cards to a compact row on mobile (#1132) * Support multi-leg (layover) flights (#1146) * feat(transport): support multi-leg (layover) flights in the booking form A flight booking can now hold an ordered chain of airports (e.g. FRA -> BER -> HND) instead of a single departure/arrival pair. The route is entered as a list of waypoints with a '+ add stop' button; each stop carries its own arrival and departure time plus the airline/flight number of the segment leaving it, while the whole booking keeps one price. Stored without a schema change: the existing reservation_endpoints rows carry the ordered waypoints (from/stop/to by sequence) and a metadata.legs array holds the per-leg detail. Top-level metadata (departure_airport/arrival_airport/airline/ flight_number) mirrors the first and last leg, so a single-leg flight persists exactly as before and legacy readers keep working. * feat(planner): show each flight leg as its own day-plan entry, ordered by time A multi-leg flight now expands into one entry per leg (BER -> FRA, then FRA -> HND), each on its own day with its own times, instead of a single span. Each leg is an addressable slot (reservation id + leg index) so places and notes can be dropped into the layover gap between legs; the per-leg position is persisted in metadata.legs[i].day_positions and survives a reload. Day-plan items are now ordered chronologically: anything with a time (a place's time, a flight leg, a timed note) sorts by that time, and untimed items inherit the time of the item before them so they stay where they were placed. * feat(planner): show the full multi-stop route in the bookings panel The route row now lists every waypoint (FRA -> BER -> HND) by sequence instead of just the first and last airport. * feat(map): draw multi-leg flights as connected legs with a marker per airport Both the Leaflet and Mapbox overlays now render a flight over all its waypoints: one great-circle arc per leg and a marker at every airport, with the label showing the full route and the summed distance. A single-leg flight is unchanged. Also drops the floating stats badge that was drawn on transport arcs. * fix(map): centre a clicked place above the bottom inspector panel Selecting a place panned/flew it to the dead centre of the screen, where it sat behind the detail card. Both overlays now bias the target into the visible area above the bottom panel (Leaflet offsets the pan by the inspector inset; Mapbox passes the padding to flyTo). * feat: show the full multi-stop flight route in PDF and calendar export The PDF day list and the ICS export now render the whole route (FRA → BER → HND) for a multi-leg flight instead of just the first and last airport, falling back to the flat metadata for single-leg flights. The ICS keeps a single event per booking. * feat(import): group connecting flight legs into one multi-leg booking When a booking confirmation contains several flight legs sharing a PNR that connect at the same airport with a short layover (under 24h), they are now imported as a single multi-leg booking (from/stop/to endpoints + metadata.legs) instead of one booking per leg. A round trip (same PNR, multi-day gap) stays two separate bookings, and a single flight is unchanged. * i18n: translate the new flight-route strings into all languages * i18n: translate the Costs page into every language The Budget → Costs rework left the new costs.* strings untranslated in every non-English locale (they fell back to English). Translate them across all supported languages. * Revert "fix(map): centre a clicked place above the bottom inspector panel" This reverts commit 0936103f04. * Explore places on the map, planner route fixes, and instance-wide Mapbox (#1147) * feat(maps): add an OSM POI search endpoint (category within a viewport) New /api/maps/pois queries OpenStreetMap via Overpass for places of a category (restaurants, cafes, hotels, sights, …) inside a bounding box. OSM-only by design — it never calls Google, even when a Google key is configured. * feat(map): explore nearby places on the trip map (OSM category pill) A floating, icon-only pill over the planner map lets you toggle a POI category and see those OpenStreetMap places in the current view; clicking a marker opens the add-place form pre-filled (name, address, website, phone). Single-select with a 'search this area' action after the map moves. Renders on both the Leaflet and Mapbox maps, and can be turned off in settings (discussion #841). * fix(planner): anchor timed places when optimising and route transports by location - The day optimiser no longer reshuffles places that have a set time — they stay anchored to their time, like locked places. - The route now uses a transport's departure/arrival location as a waypoint when it has one (e.g. a flight's airport), instead of breaking the route at every booking; transports without a location are ignored for routing but still show their leg's distance/duration under the booking. * feat(admin): instance-wide Mapbox defaults in default user settings Admins can set a shared Mapbox token (plus style, 3D and quality) as instance defaults, so the whole instance can use Mapbox without each user pasting their own key. Users without their own value inherit it via the existing admin-defaults merge; the shared token is stored encrypted (discussion #920). * Reorder whole days and insert a day (#589) (#1148) * feat(days): reorder whole days and insert a day at a position Adds reorderDays + insertDay to the day service and a PUT /days/reorder route (plus an optional position on create). Day rows stay stable so a day's assignments, notes, bookings and accommodations ride along by id; on a dated trip the calendar dates stay pinned to their slots while the content moves across them, and each booking's date is re-stamped onto its day's new date (time-of-day preserved) so day_id stays consistent. Renumbering uses the two-phase write to avoid the UNIQUE(trip_id, day_number) collision, and a move that would invert an accommodation's check-in/out span is rejected. * feat(planner): reorder days from a toolbar popup, and add days A new toolbar button opens a popup listing the days; drag a row by its grip or use the up/down arrows to reorder, and add a day from there. Reorders apply optimistically with rollback and sync over WebSocket; the day headers are left untouched, so the existing place drop-targets are unaffected. * i18n: add day-reorder strings across all languages * Map/planner/dashboard polish and small community features (#1155) * feat(planner): reorder days in a modal instead of a dropdown The day-reorder control opened a small anchored dropdown; move it into the shared Modal (portal, dimmed backdrop, Esc/backdrop close) so it matches the Add activity dialog. Drag handles, up/down arrows and the day badges are unchanged. * feat(map): explore reliability, Mapbox popups + compass, region-biased search POI explore: clamp oversized viewports, query the Overpass mirrors in parallel (first valid response wins) with a per-request timeout and a short-lived cache, and surface a retry when every mirror fails - so it returns results at any zoom instead of timing out. Mapbox renderer: add the place/POI hover popups (name, category, address, photo) the Leaflet map already had, plus a compass pill next to the explore pill that resets the view to north. /api/maps/search: accept an optional locationBias to fix foreign-region bias and expose Google's place types in the result. * feat(dashboard): list-view and mobile polish Use the Archived status label for the filter and show Open dates for trips without dates; drop the unused settings button next to the view toggle. Desktop list view renders the date as a stat-style block separated from the counts. Mobile list rows are stacked (slim cover banner + centred date), trip actions stay visible (touch has no hover), and the hero card's hover lift is disabled on touch; small spacing fix under the sidebar. * feat: small community-requested options Raise the plan-note subtitle limit to 250 characters and add more note icons. Expose is_archived and cover_image on the update_trip MCP tool. Add place coordinates to the PDF export. Allow creating a category from an existing to-do, and add a show/hide toggle on the admin password fields. * test(shared): bump day-note subtitle limit assertion to 250 * test: align specs with the new search param order and archive label Keep lang as the 3rd positional arg of the maps search controller so the existing unit test stays valid, and forward locationBias as the 4th. Add the now-used Popup to the MapViewGL mapbox mock, switch the dashboard archive-filter query to the Archived label, and expect the 4-arg search call. * fix(packing): add more bag colors so sub-bags stop repeating (#1156) The auto-assigned bag palette only had 8 colors, so the 9th bag reused the first one. Double it to 16 (keeping the existing 8 and their order) and keep the server and client lists in sync - both cycle BAG_COLORS[count % length]. * fix(packing): respect per-item quantity in bulk import (#1157) * AirTrail integration: import flights & two-way sync (#214) (#1158) * feat(admin): register AirTrail as an integration addon Off by default; toggle lives in Admin -> Addons with a Plane icon. The per-user connection (URL + API key) follows in integration settings. * feat(integrations): add per-user AirTrail connection Settings -> Integrations gains an AirTrail section: instance URL + Bearer API key (encrypted at rest via apiKeyCrypto), a self-signed-TLS opt-in and a test-connection check. Served by a small Nest controller under /api/integrations/airtrail, gated on the airtrail addon and SSRF-guarded. The key is per-user, so it only ever returns that user's own flights. * feat(transport): import flights from AirTrail Adds an AirTrail Import button next to Manual Transport that lists the user's AirTrail flights and highlights the ones inside the trip dates. Selected flights become reservations linked to their AirTrail origin (external_* columns), deduped against flights already in the trip, then broadcast to every member. The mapping resolves airports, airport-local times and flight metadata; the linkage is what the two-way sync rides on. * feat(transport): badge AirTrail-linked flights as synced Linked reservations show an 'AirTrail synced' badge, or 'no longer synced' once the flight is gone from AirTrail. * feat(transport): keep TREK and AirTrail flights in sync both ways A scheduled poll reconciles each connected owner's flights: field edits (detected by snapshot hash, since AirTrail has no updated_at) flow into the linked reservation and broadcast live; a flight deleted in AirTrail keeps the TREK row but stops syncing. Editing a linked flight in TREK pushes back to AirTrail under the importer's credentials, preserving the existing seat manifest; if the owner disconnected the link detaches so the poll can't revert the local edit. Deleting in TREK never touches AirTrail. * i18n(airtrail): add AirTrail strings across all locales * test(airtrail): cover flight mapping, timezones and snapshot hashing * fix(airtrail): reduce airline/aircraft objects to codes The flight list/get response returns airline and aircraft as joined objects ({icao, iata, name, ...}), not bare codes. Mapping them straight through produced '[object Object]' titles and stored objects in metadata, which crashed reservation rendering. Extract the ICAO/IATA code instead, and title flights by their flight number. * fix(airtrail): clear error on non-JSON responses, tolerate /api in URL A misconfigured instance URL made AirTrail serve its SPA/login HTML, and the raw JSON.parse failure surfaced as 'Unexpected token <'. Surface an actionable message instead, and strip a pasted trailing /api so the base URL still resolves. * feat(transport): sync AirTrail edits on trip open, not just on the poll Add a per-user on-demand sync (POST /integrations/airtrail/sync) triggered when a connected user opens a trip, so AirTrail-side edits appear right away instead of waiting up to a full poll cycle. Lower the background poll from 15 to 5 minutes as a safety net. * fix(transport): refresh imported AirTrail flights without a reload loadTrip doesn't fetch reservations, so a freshly imported flight only appeared after a full page reload — use loadReservations instead. Also show flight dates in the user's locale format (e.g. 13.06.2026) rather than the raw ISO string. * style(settings): align AirTrail connection with the photo-provider layout Match the Immich section: stacked URL/key fields, a ToggleSwitch for self-signed TLS, and a Save / Test-connection row with a status badge. * feat(transport): add a seat field when editing flights The transport editor only offered a seat field for trains; flights had none even though imports store metadata.seat. Show and persist a seat for flights too. * style(transport): match the AirTrail button height to Manual Transport * feat(transport): put the flight seat next to flight number and sync it to AirTrail Move the seat from a standalone row to the per-leg flight details (beside the flight number), stored per leg in metadata.legs[].seat with the first leg mirrored to metadata.seat. On push, set the seat number on the user's own AirTrail seat (the one with a userId), leaving co-passengers untouched; import/poll read that same seat back. * refactor(planner): move the AirTrail trip-open sync into useTripPlanner Page containers must not own state/effects (lint:pages). Same logic, relocated from the page into its data hook. * test(db): pin the region-reconciliation test to its schema version The test re-ran 'the last migration' assuming the reconciliation is last; it no longer is once later migrations are appended. Pin to version 135 and re-run from there (the appended migrations are idempotent). * Various fixes: 2FA autofocus, viewer-timezone times, duplicate place guard (#1159) * fix(auth): autofocus the 2FA code input when the MFA step appears (#767) * fix(notifications): show notification and admin times in the viewer timezone (#1149) SQLite CURRENT_TIMESTAMP is UTC but the string has no Z, so the client parsed it as local time. Normalize in-app notification created_at to ISO-UTC, and stop forcing the admin user table to render in the server timezone. * fix(places): warn before adding a duplicate place (#1152) Manually adding a place did not check the existing pool, so the same POI could land in Unplanned twice. Flag a likely duplicate by Google Place ID, name or near-identical coordinates and require a confirming second click to add anyway. * fix(planner): make route tools reachable in mobile day plan sheet (#1142) * wiki: update dev env * wiki: small precision in dev env * fix(planner): make route tools reachable in mobile day plan sheet On mobile, selecting a day closes the plan sheet immediately, so the route tools footer (Route toggle / Optimize / routing profile) - gated on the selected day - was never reachable. Desktop was unaffected. - Add showRouteToolsWhenExpanded prop to DayPlanSidebar: when set, route tools render on any expanded day with 2+ assigned places - Make handleOptimize accept an explicit dayId (defaulting to selectedDayId, preserving desktop behavior) - Keep the distance/duration pill gated on the selected day, since routeInfo belongs to the selected day's calculated route - Enable the prop on the mobile plan sheet in TripPlannerPage * fix(planner): correct route-tools prop doc and dev-environment wiki - Reword the showRouteToolsWhenExpanded JSDoc to list the controls the footer actually renders (Route toggle / Optimize / travel profile); there is no "Open in Google Maps" action in that block. - Wiki: drop the non-existent server test:parity script, document the real shared i18n:parity checks, and fix the i18n note (the translation layer already lives in @trek/shared, it is not "upcoming"). --------- Co-authored-by: jubnl <jgunther021@gmail.com> Co-authored-by: Maurice <mauriceboe@icloud.com> * feat(places): enrich list-imported places via the Places API (#886) (#1161) * feat(places): enrich list-imported places via the Places API (#886) Google/Naver list imports only carry a name and coordinates, so the places open as bare pins — the Maps tab jumps to coordinates, with no photo, address or open/closed. Add an opt-in "Enrich places via Google" toggle to the list-import dialog, shown only when a Google Maps key is configured. When enabled, after the (fast, unchanged) import the server runs a background pass that re-resolves each place by name — biased to and validated against the imported coordinates so a common-name search cannot overwrite the wrong place — and fills the empty address/website/phone/photo columns plus the resolved google_place_id, pushing each row over the live sync. Opening hours and the proper Maps link then work on demand from the stored id. Enrichment only fills empty fields, runs detached so a long list never blocks the import, and no-ops when no key is configured. * fix(places): use the ToggleSwitch component for the enrich toggle Match the rest of the app — the import-enrichment opt-in used a raw checkbox; swap it for the shared ToggleSwitch (text left, switch right) like the settings toggles. * fix(maps): bound place-photo cache growth (Wikimedia + Google) (#1174) The place-photo cache (uploads/photos/google) grew unbounded: a Wikimedia geosearch path cached full-res originals despite requesting a 400px thumb, the writer applied no size guard, nothing reclaimed orphaned files, and backups archived the whole re-derivable cache verbatim. - Prefer the scaled `thumburl` over the full-res `info.url` in the Commons geosearch fallback. - Downscale any cached image to <=800px JPEG via the existing jimp dep, with a safe fallback to the original bytes on decode failure. - Add sweepOrphans() (orphaned meta rows + stray files) wired into the scheduler (startup + nightly), and removeIfUnreferenced() called on place delete for prompt reclamation. - Exclude the re-derivable photo/trek caches from backups; restores self-heal as the cache dirs are recreated at startup. * fix(sync): remap temp ids, prevent id collisions, surface failed mutations (#1175) Closes three offline BLOCKERs from the PWA audit: - B1: offline edits/deletes of an offline-created entity were lost. The negative temp id was baked into the PUT/DELETE url and never rewritten after the CREATE returned a real id, so dependents 404'd and were dropped. Dependents now carry a {id} placeholder + tempEntityId; flush builds a tempId->realId map and durably rewrites still-queued dependents on CREATE success (survives flush boundaries / reloads). - B2: tempId = -(Date.now()) collided within a millisecond, overwriting an optimistic row. Replaced with a monotonic nextTempId() minter. - B3: any 4xx marked the mutation failed with no rollback and no signal, and the badge ignored failed rows. Terminal failures now roll back the phantom optimistic CREATE; 401/408/425/429 are treated as retryable; failedCount() is surfaced in OfflineBanner (red pill) and OfflineTab. * fix(maps): make offline tiles cover real trips (cap coherence + zoom-clamp) (#1177) Closes BLOCKER B5 — the offline map was blank for most real trips: - The Workbox 'map-tiles' cache held only 1000 entries while the prefetcher budgeted ~3413, so prefetched tiles were evicted on arrival. Both caps are now a coherent 12288 (~180 MB), kept in sync with cross-referencing comments. - prefetchTilesForTrip skipped a trip entirely when its all-zooms estimate exceeded the cap, so region/road-trip bboxes got no tiles. Removed the all-or-nothing guard; prefetchTiles already fills zooms low→high and stops at the budget, so large trips now cache the zooms that fit instead of nothing. * fix(security): stop cross-user offline data leak on shared devices (#1176) Closes BLOCKER B4 — three reinforcing paths could serve one account's cached data to the next user on a shared device: - The Workbox 'api-data' cache keyed trip/user-scoped GETs by URL only (cookie-blind). Changed to NetworkOnly; offline reads come from the per-user IndexedDB cache via the repo layer instead. - IndexedDB had no per-user scoping. The Dexie connection is now scoped per user (trek-offline-u<id>) behind a Proxy so the ~19 importers keep a stable binding; login opens the user DB, logout deletes it and returns to the anonymous DB. - logout() was fire-and-forget and racy: background flush/syncAll could re-seed the DB after the wipe. It is now async and ordered — close an auth gate, unregister sync triggers, disconnect, clear caches, delete the user DB — and flush()/syncAll() bail when the gate is closed. * fix(db): scope, evict, and cap the offline blob cache (H3) (#1178) Blob cache previously leaked forever: clearTripData omitted it, entries had no trip discriminator, and there was no size/count bound, so file blobs survived trip eviction and could starve the map-tile cache for quota. - BlobCacheEntry gains tripId + bytes; Dexie v3 adds a tripId index with a backfill upgrade (legacy rows -> tripId -1, bytes from blob.size) - clearTripData purges the trip's blobs in-transaction - enforceBlobBudget() evicts oldest-by-cachedAt past 200 entries / 100 MB - tripSyncManager threads tripId/bytes into puts and enforces the budget * fix(repo): fall back to Dexie when a network read fails (H2) (#1179) Repos gated reads on raw navigator.onLine and the online branch had no try/catch, so a captive portal or connected-but-no-internet (navigator.onLine lying "true") threw a network error instead of serving the good cached copy — blanking the trip even though Dexie held it. - new onlineThenCache(onlineFn, cacheFn) helper: reads the cache when offline, and on a network-level failure (Axios error with no HTTP response). A genuine HTTP error (4xx/5xx — the server responded) is rethrown so callers still set error state / navigate, not masked by a stale cache. - gates only on navigator.onLine, NOT the connectivity probe: the probe is a coarse global flag and one failed health check would otherwise divert every read to the (possibly empty) cache even when the request would succeed. - every repo list/get read path routed through it (reads only — writes still go through the mutation queue so failures surface) - tests: captive-portal fallback, HTTP-error rethrow, non-Axios rethrow * fix(store): reset and uniformly hydrate trip-scoped slices in loadTrip (H4, H5) (#1180) loadTrip only replaced the first slice group, so budget/reservations/files from a previous trip stayed visible after switching trips (data exposure on a shared screen). Those three also loaded via separate tab-gated effects, so they never hydrated offline for an unopened tab. - resetTrip() clears every trip-scoped slice (keeps global tags/categories) and runs at the top of loadTrip, so a switch can't leak the prior trip's data - loadTrip now hydrates budget/reservations/files through their repos alongside the rest (non-fatal catches), making offline hydration uniform - useTripPlanner drops the redundant loadFiles + reservations/budget effects; tab-gated lazy reloads stay as on-demand refresh - tests: cross-trip no-leak, uniform hydration, resetTrip * fix(sync): re-hydrate active trip store on reconnect/online (H1) (#1181) setRefetchCallback was dead code, so on reconnect the queue flushed and Dexie re-seeded but the open trip's Zustand store was never refreshed — a collaborator's edits made while we were offline didn't appear until navigating away and back. - new tripStore.hydrateActiveTrip(): silent refresh of the active trip's collaborative state (days/places/packing/todo/budget/reservations/files), no resetTrip and no isLoading toggle so there's no splash on reconnect - syncTriggers wires setRefetchCallback to it (WS layer awaits the flush hook first) and re-hydrates open trips after the online-event syncAll; cleared on unregister - websocket exposes getActiveTrips() for the online-event path - tests: refetch wiring + ordering, silent hydrate without reset/splash * fix(server): lengthen idempotency key TTL to survive multi-day offline (H6) (#1182) The nightly cleanup deleted idempotency keys older than 24h. The TREK client replays queued mutations with their X-Idempotency-Key on reconnect, so a device offline longer than a day had its keys GC'd before it returned — the replayed POST was then treated as new and created a duplicate. - raise the TTL to 30 days (DEFAULT_IDEMPOTENCY_TTL_SECONDS), overridable via IDEMPOTENCY_TTL_SECONDS - extract purgeExpiredIdempotencyKeys(now, ttl, db) (mirrors cleanupOldBackups) with an injectable db, and have the cron job call it - tests: 30-day default eviction, 25-day key retained (was dropped at 24h), env override H7 (exactly-once across the lost-response window) is deferred: a correct fix must store the response in the same DB transaction as the entity write. Doing it in the generic interceptor (reserve-before-handler) cannot store the real response body for the crash case, which would break the client's temp->real id remapping on replay (mutationQueue.flush relies on the entity in the body). It needs a per-service change and is tracked separately. * fix(realtime): correct assignment:created echo dedup (H11) (#1183) When X-Idempotency/X-Socket-Id let an own-echo through, the assignment:created dedup had two bugs: it keyed on place id, so (1) a legitimate second assignment of a place already on the day was silently dropped, and (2) the temp-version reconciliation matched place?.id === placeId, letting undefined === undefined collapse place-less rows onto each other. - dedup now keys on assignment id (exact-id duplicate -> no-op) - temp (negative-id) optimistic rows are reconciled only when a real placeId matches, replacing just that row; a sibling temp of another place is untouched - everything else appends, including a genuine 2nd assignment of the same place - tests: 2nd-of-same-place kept, correct temp picked among siblings, place-less rows don't collapse Note: the broader own-echo suppression relies on X-Socket-Id being sent; this fixes the client-side fallback when an echo slips through. * fix(pwa): persist offline storage + Mapbox offline policy (H8, H9) (#1184) H8: prefetched tiles and file blobs could be evicted under storage pressure (worsened by opaque tile responses inflating the quota ~7MB each), blanking the offline map right when a traveler needs it. Request persistent storage at app init so the browser exempts our caches from eviction. We deliberately keep tile requests no-cors (a cors switch would break self-hosted/custom tile providers without CORS headers), so persistence is the safe mitigation rather than de-opaquing responses. H9: Mapbox GL users had no offline map at all — no runtimeCaching matched the Mapbox hosts. Add a StaleWhileRevalidate rule for api.mapbox.com / *.tiles.mapbox.com so visited areas are available offline (best-effort; full pre-download still requires the Leaflet renderer, now documented). - new sync/persistentStorage.ts requestPersistentStorage(), called from main.tsx - vite.config: mapbox-tiles SW cache rule - MapViewAuto / tilePrefetcher comments document the offline-maps policy - tests for the persist helper (granted / already-persisted / absent / rejects) * ci(security): only fail Docker Scout on fixable CVEs Add only-fixed so the scan no longer fails on vulnerabilities with no upstream fix available (e.g. base-image OS packages), and only flags actionable, fixable findings. * build(docker): rebuild gosu with a current Go toolchain Debian's apt gosu ships an old Go stdlib that the image CVE scan flags (1 critical + several high, all in golang/stdlib). Build gosu from source with a current Go toolchain and copy the static binary in instead; the runtime behaviour is unchanged — gosu still drops root to node at startup. * build(deps): bump tsx's esbuild to 0.28.1 (GHSA-gv7w-rqvm-qjhr) The production image's last image-scan finding was esbuild 0.28.0, pulled in transitively by tsx. Pin tsx's esbuild to 0.28.1 (within tsx's ~0.28.0 range) to clear GHSA-gv7w-rqvm-qjhr. Lockfile-only; no runtime change. * feat(auth): add "Remember me" checkbox to extend session lifetime (#1189) Adds a "Remember me" checkbox to the login form (single responsive page, covers mobile + desktop). Unchecked (default) issues the existing SESSION_DURATION JWT with a browser-session cookie (no maxAge); checked issues a longer-lived JWT plus a persistent cookie sized by the new SESSION_DURATION_REMEMBER env var (default 30d). The choice is threaded through the MFA verify leg so it survives the step-up. Register/demo logins keep their current persistent behaviour. * chore(ssrf): include lookup error code in error message * fix(backup): restore from Docker, fail-fast on shadowed /app, bundle encryption key (#1193) (#1197) * fix(backup): restore uploads through symlinked dir and bundle encryption key (#1193) Restoring a backup inside Docker threw ERR_FS_CP_DIR_TO_NON_DIR because /app/server/uploads is a symlink to the mounted /app/uploads volume and cpSync (dereference:false) refuses to overwrite the symlink node with a directory. The DB was swapped before this failing copy, so users saw restored data but missing upload files (trip covers). Resolve the symlink with realpathSync before copying so the merge targets the real directory; no-op on a plain dir, so non-Docker behavior is unchanged. Also bundle the at-rest encryption key (data/.encryption_key) into the backup so a restore onto a different install can decrypt stored secrets (API keys, MFA, SMTP/OIDC). Skipped when ENCRYPTION_KEY is provided via env (the file is not the source of truth then). On restore the key is swapped back if the archive carries one; a restart is required for the in-memory key to take effect. * fix(docker): fail fast when a volume shadows /app (#1193) Mounting an old volume at /app hides the image's node_modules and dist, so startup crashed with a cryptic "Cannot find module 'tsconfig-paths/register'". Add a CMD preflight that detects the missing app files and exits with actionable guidance. Document in the README that only /app/data and /app/uploads should be mounted, never /app. * fix: ssrf test * fix(places): fall back to search when autocomplete details lookup fails (#1192) (#1198) Clicking an auto-suggest dropdown item did a second /maps/details lookup that could fail (details kill-switch off, an overloaded OSM Overpass mirror behind a proxy, or any upstream error), dead-ending on "Place search failed" while the search button stayed reliable. handleSelectSuggestion now treats a missing or coordinate-less details result (or a thrown error) as a miss and falls back to the text-search path the search button uses, applying the first result. The error toast only fires if the fallback also returns nothing. Adds tests for the previously untested suggestion-click path. * fix(planner): scroll long place description/notes on mobile (#1195) (#1199) The place details card (PlaceInspector) clipped long description/notes with no way to scroll. The content area is a flex column whose children (description/notes) had the default flex-shrink: 1, so once the card hit its maxHeight cap they compressed to fit and their overflow:hidden clipped the text instead of overflowing into a scroll region. - Make the content area a bounded scroll region (flex: 1 1 auto, minHeight: 0, overflowY: auto, momentum + overscroll containment). - Pin description/notes with flexShrink: 0 so they keep natural height and the card overflows into the scroll instead of clipping. - Pin header/footer with flexShrink: 0 so they stay fixed while scrolling. - Add wordBreak/overflowWrap to the description div to fix horizontal clip. * Day plan: hotel travel times at start/end + login toggle polish (#1206) * fix(login): use the shared toggle for the stay-signed-in option * feat(planner): show hotel travel times at the start and end of a day * fix(login): give the stay-signed-in toggle an accessible name and fix its test * fix(trips): keep the day-count field empty when cleared and validate it (#1204) (#1207) * docs(readme): refresh dashboard, costs and trip screenshots (#1208) * docs(readme): refresh dashboard, costs and trip screenshots * docs(readme): correct outdated info (React 19, NestJS, 20 languages, Costs rename, passkeys, AirTrail, notifications) * chore: update all dependencies (#1209) * chore: update all dependencies * chore: remove lint errors * fix(client): restore typecheck after dependency bump vitest 4 types vi.fn() as Mock<Procedure | Constructable>, which no longer assigns to the strictly-typed onUpdate prop; type the mock explicitly. TS6 + the new transitive @types/node 25 stopped auto- including node builtin module types, so import('node:buffer') failed; add @types/node as a direct client devDependency and a scoped node type reference in the one test that needs it. * test: fix constructor mocks for vitest 4 Reflect.construct semantics vitest 4 resolves new-invoked mocks via Reflect.construct, which rejects arrow-function implementations (including mockReturnValue sugar) as non-constructable. Convert mapbox-gl and better-sqlite3 mocks that the code instantiates with new to regular function implementations. * fix(planner): only route to multi-day transport endpoints on their pickup/drop-off days (#1210) (#1212) * chore: move to Frankfurter API for exchange rate (#1214) * Restore nest coverage to >=80% after the #1209 dep bump (istanbul provider + branch tests) (#1213) * fix(server): set oxc:false in vitest so the SWC transform survives the Vite 8 bump * fix(server): switch coverage to the istanbul provider (v8 under-reports branches on Vite 8 + Vitest 4) * test(nest): cover controller/service branches to clear the 80% coverage gate * fix(planner): correct transfer-day hotel legs and connect them to transports (#1215) When you change hotels on a day, the morning bookend leg showed the hotel you check into instead of the one you slept in whenever the morning stay didn't end exactly on that day — both bookends collapsed onto the arriving hotel. The morning hotel is now picked by "checked in earlier and still in range" rather than "checks out today", which also fixes the route optimizer's start anchor for the same case. The bookend legs now connect to the first/last located waypoint of the day — a place or a transport endpoint (a car return, a taxi or train arrival) — so the hotel-to-transport drives are included too. * feat(transports): add kitinerary import-from-file button to Transports tab * docs(config): document SESSION_DURATION_REMEMBER across deployment artifacts Add SESSION_DURATION_REMEMBER to docker-compose, .env.example, README env table, Helm chart (values + configmap passthrough), the Unraid template, and the Unraid install guide. Where the base SESSION_DURATION was also absent (README, charts, Unraid) add the pair so the Remember-me variable has context. --------- Co-authored-by: gzor <risenbrowser@web.de> Co-authored-by: ppuassi <34529179+ppuassi@users.noreply.github.com> Co-authored-by: sss3978 <106522699+soma3978@users.noreply.github.com> Co-authored-by: SkyLostTR <onurluerin@gmail.com> Co-authored-by: Julien G. <66769052+jubnl@users.noreply.github.com> Co-authored-by: Dimitris Kafetzis <39215021+Dkafetzis@users.noreply.github.com> Co-authored-by: Ahmet Yılmaz <70577707+sharkpaw@users.noreply.github.com> Co-authored-by: jubnl <jgunther021@gmail.com> Co-authored-by: jufy111 <40817638+jufy111@users.noreply.github.com> Co-authored-by: Larinel <bodink7@gmail.com> Co-authored-by: rossanorbr <48014819+rossanorbr@users.noreply.github.com>	2026-06-16 22:22:45 +02:00