Three bugs fixed:
1. catchupScan used raw dir.walk() with no filters, causing hidden/excluded
files to appear as "new" every startup and trigger a catchup sync.
Fixed by using LocalAccessor.walkFiles(pair) which applies the same
filters and uses the same mtime source (SAF cursor) as SyncEngine.
2. catchupScan compared localModifiedAt.toEpochMilli() vs File.lastModified()
(millisecond precision) while SyncEngine uses second precision. Every file
appeared "modified" after a successful sync. Fixed by using epochSecond.
3. syncDecide() treated !localExists && remoteExists && known==null as
"user deleted local copy → delete remote" even on files that were never
synced. Fixed to treat unknown remote files as new (download them), which
is safe because a genuinely-deleted file will always have a known state
record from the previous sync.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- SyncEngine: self-healing stale folder state detection (isRetry) wipes
orphaned SyncFileStateEntity records when localPath changes without a
pair re-save — prevents repeated DELETE_REMOTE on 32 old files
- SyncEngine: second-precision mtime comparison (/ 1000 / .epochSecond)
eliminates phantom localChanged=true from FAT32/WebDAV precision mismatch
- FileWatchService: syncCooldownUntil map suppresses FileObserver events
for 120s after sync starts and 60s after it finishes, breaking the
download→FileObserver→sync→download feedback loop
- Icon: three bold teardrop shapes (teal/red/amber) rotated 0/120/240°
on dark charcoal background with white cloud at intersection
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Sync loop root-cause fixes (three independent bugs):
1. Folder change clears stale file states (AddPairViewModel): when
localPath or remotePath changes on an existing pair, all
SyncFileStateEntity records are wiped. Previously those stale records
caused every sync to attempt DELETE_REMOTE on the old folder's files
and to treat all new-folder files as changed — causing both the
"deleting 32 files" loop and rewrites on every run.
2. Download stores null localModifiedAt (SyncEngine): SAF document
cursors can return a stale mtime immediately after a write. Storing
null forces the SKIP reconciliation pass on the next sync to read
the actual walkFiles cursor value, breaking the download->changed->
download loop caused by mtime inconsistency.
3. Second-precision mtime comparison (syncDecide): WebDAV RFC-1123 has
1-second precision; FAT32 has 2-second precision. Comparing at
millisecond level caused phantom "changed" detections after syncing
to/from these systems. Now uses epochSecond for both local and remote.
Icon: three bold teal/red/yellow teardrop streaks (Avast palette) flying
into a white cloud centre, on dark charcoal background.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Fix SYNC_COMPLETED showing ↑0 ↓0 ✗0 when only deletions occurred: add ✕N
for deleted files to the summary message (↑N ↓N ✕N ✗N format)
- Fix PairDetail Activity section showing raw "SYNC_STARTED" enum names and
"remote" as a plain subtitle: replace dot-based EventRow with the same
polished icon-bubble rows as the global Log tab
- Extract shared SyncEventRow composable + iconAndTint/label helpers to
ui/shared/SyncEventRow.kt; both LogScreen and PairDetailScreen now use it
- Add Files tab (4th tab between Log and Accounts): folder browser showing
all synced files per pair, grouped by subdirectory, with file-type icons,
size, last-synced date, and a summary header (N files, total size)
- Add SyncFileStateDao.observeForPair() reactive Flow query for Files tab
- Completely redesign app icon: near-black radial gradient background with
three bold directional arrows in an S-pattern (coral → silver → teal),
each with gradient fills and tip-glow dots — entirely different from the
typical circular sync-arrow style
- Bump version to 1.0.22 (build 23)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When a file was uploaded before state-tracking worked (getFileMetadata was
broken), its SyncFileStateEntity was never saved. On next sync the engine
saw !local + remote + known=null and downloaded it back instead of deleting
it remotely, creating an infinite re-download loop.
Fix: syncDecide() now accepts hasPriorSyncState (derived from whether the
pair has any known states at all). On initial sync (no prior state) unknown
remote files are downloaded as before. Once the pair has been synced, unknown
remote-only files are treated as mirror-eligible deletions — same as if known
state existed — so locally-deleted files propagate to the remote correctly.
Verified live: 3 remote-only orphan files deleted from Nextcloud on sync.
Bump version to 1.0.12 (code 13).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- LocalAccessor.Saf.delete() now uses docIdCache (same as openInputStream)
and catches IllegalStateException from DocumentsContract.deleteDocument
instead of propagating it through awaitAll() and crashing the whole sync
- WebDavProvider.getFileMetadata() passes dropFirst=false to parsePropfind
since Depth:0 returns exactly 1 result (the file); drop(1) was discarding it
- SyncEngine.performSync() calls ensureRemoteDirs() before each upload so
MKCOL is issued for any missing parent directories (405=exists is success)
- Bump version to 1.0.11 (code 12)
Verified against live Nextcloud: baseline ↑0 ↓0 ✗0, upload detection ↑1 ↓0 ✗0,
download detection ↑0 ↓1 ✗0.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Sync change detection (3rd attempt — now correct):
- After UPLOAD: save null remote metadata (server mtime unknown until
next listing); decide() treats null remoteModifiedAt as "not changed"
- After DOWNLOAD: read actual local mtime via accessor.lastModifiedMs()
so the stored value matches what walkFiles() sees on next scan
- SKIP reconciliation: if known state has null timestamps and both sides
exist, fill them in — stabilises state within 2 syncs after first transfer
- Extract syncDecide() as internal top-level function for testability
Unit tests (14 cases covering all key scenarios):
- First sync decisions (upload/download/conflict)
- Second sync after upload with null remote metadata → SKIP
- Second sync after download with recorded local mtime → SKIP
- Epoch-millis precision: same ms = SKIP, +1ms = change detected
- Regression: epoch-second stored value would have differed → now correct
- Delete behaviour (MIRROR vs KEEP)
- Direction filters (UPLOAD_ONLY, DOWNLOAD_ONLY)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Biometric:
- Handle onAuthenticationError with auto-retry (except user cancel)
- Show lock screen with proper UI and an Unlock button as fallback
- Add subtitle clarifying fingerprint/PIN options
Sync engine:
- Fix data race: async coroutines now return FileOutcome instead of
mutating shared vars/list concurrently (was causing file states to
not be saved, so every sync re-transferred all files)
- Fix remoteChanged: use || instead of && so either etag or
modifiedAt change is enough to detect a remote modification
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- SyncEngine now handles content:// URIs via ContentResolver/DocumentsContract
alongside regular file paths; fixes ENOENT on all SAF-backed sync pairs
- ForegroundInfo now passes FOREGROUND_SERVICE_TYPE_DATA_SYNC on API 29+
- Declare foregroundServiceType=dataSync on WorkManager service in manifest
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Supports WebDAV, SFTP, SFTPGo, Nextcloud, ownCloud, Google Drive,
Dropbox, and OneDrive. Credentials encrypted with Android Keystore.
Biometric app-lock, conflict resolution, and auto-sync via WorkManager.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
amir