The background 2 s refresh worker re-queried and re-copied the whole
wallet every cycle (a fresh TLS refresh plus a multi-megabyte JSON
build, copies and DOM parse), even when nothing had changed, so CPU and
the network never went idle.
The worker now skips the full refresh and the re-publish on a cycle that
is provably unchanged, while never letting the UI show stale confirmed
balance or transactions:
- The lite bridge keeps a monotonic mutation epoch, bumped after any
command not on a read-only allow-list, so a local change made on any
thread (send, shield, import, new address, encrypt/unlock/lock,
console) is detected at one choke point.
- Each cycle the worker runs one cheap sync probe (one connection, the
mandatory block ingest) and a local height read, and refreshes only
when the scanned height advanced, the mutation epoch moved, the
encryption state changed, a sync is in progress, or a ~10 s heartbeat
elapsed; otherwise it skips. Direct/synchronous callers are not gated.
- A failed probe backs off instead of running a second sync, so a dead
server is not retried back-to-back.
Confirmed funds can never be stale beyond one cycle (every confirmed
change advances the height or bumps the epoch); only unconfirmed mempool
rows can lag up to the heartbeat, and the spend path re-queries the
backend directly rather than reading the gated model. Average idle cost
drops from three data connections per cycle to about one.
Tests: a bridge mutation-epoch test and a worker change-gate test
(skip-when-unchanged; refresh on height advance, local mutation and
encryption change) against the fake backend.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A failed backend sync was silently swallowed: the reply was never
inspected, so a sync that errored out left the wallet showing a
"syncing" state that never completed and never reported a problem.
- The sync and rescan replies are now parsed. On failure the partial
progress is saved and the sync is retried with exponential back-off
(5s doubling up to 60s, six attempts in all), then an explicit failed
state is published: done, sync not complete, and the backend's error
text carried through.
- A failed rescan is retried as a plain sync, because a rescan has
already cleared the wallet and re-running it would throw away the
partial progress just saved.
- A terminal error, such as a reorg deeper than the wallet can unwind,
stops the retries immediately instead of waiting out the attempts.
- Recovery is observed, not guessed: once a later refresh's server-bound
reads all succeed, the failed state clears itself.
Behaviour change: a sync failure now surfaces to the user as a failed
state with a message, instead of a frozen "syncing", and the client
retries on its own with back-off.
Tests: a parser case for the success, failure, and terminal replies, and
fake-backend controller cases for retrying after a failure, giving up
after the attempt cap, a terminal reorg stopping at once, a failed
rescan retrying as sync, cancel stopping the back-off, and the failed
state self-healing on a good refresh.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
importKeyCore guessed the key type from its first character, then
retried a failed import with the other command. That retry is what sent
shielded keys to the backend's transparent-key decoder, which panicked
on them and poisoned the wallet lock. A WIF typo that was still
valid base58 also reached the backend unchecked.
- The key is classified first, and exactly one backend command runs, or
none:
- secret-extended-key-... and zxview... go to "import";
- a Base58Check WIF with version 0xBC or 0x80 and the compressed flag
goes to "timport";
- everything else is refused before any backend call.
There is never a retry with the other command. A testnet (0xEF) WIF
is not accepted, because the Lite app is mainnet-only.
- New fixed, untranslated reasons: "uncompressed transparent keys are
not supported by the Lite wallet", a full-node incoming viewing key
(zivks) is explained as such, and anything else is "unrecognized key
format". The key is never echoed.
- The backend's refusal while the wallet is encrypted now maps to "wallet
is encrypted - remove encryption to import a private key". It is
checked before "locked", because unlocking would not help.
- base58Decode reserves its buffers up front, and it and
decodeBase58Check wipe their copies of the decoded bytes with
sodium_memzero on every return path, since the input can be a private
key.
- exportSeed rejects the all-zero-entropy phrase (23 x "abandon" + "art").
A backend built before the seed-export fix returns it for a wallet
whose lock failed halfway (seed wiped, wallet still unlocked); it is
never a real backup and must not seed a chat identity.
- An open that the backend recovered from the .bak ("OK: recovered from
backup; damaged file kept as <name>") is written to the lite log.
Tests: the retry control now expects one "import" and no "timport", and
"typo-key" makes no backend call; the controller's transparent-import
test uses a real compressed WIF. A new classifier test covers both WIF
versions, the shielded and viewing keys, a single call on failure, the
uncompressed, zivks, testnet, bad-flag and bad-checksum refusals, the
encrypted reason and its lite-log line, the zero-entropy seed and the
recovery log line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New address, Import key, Encrypt and Remove encryption called the
backend on the UI thread. Each ends in the backend's save, which waits on
the lock a scan holds for its whole duration and every refresh read holds
briefly, so during a scan the window froze until the scan ended. A create
or restore with a passphrase encrypted and saved just after its first
scan had started: for a restore, a freeze for the whole initial scan. A
server switch destroyed the old controller on the UI thread and joined
its refresh worker, which can sit for minutes in a read against a dead
server (each balance, notes or list read retries for ~62 s).
- The four ops run one at a time on a background lane and their results
are applied on the main thread; Settings shows "Working…" meanwhile.
Key and passphrase copies are wiped after the call, or when a queued
op is dropped.
- A create or restore encrypts and saves (an open unlocks) before the
first scan starts, so the save takes milliseconds and the scan runs on
the encrypted file.
- Import, Encrypt, Remove encryption and Receive's New address are
disabled while a scan runs ("Wait for blockchain to sync"); Redownload
blocks is disabled while an op is queued.
- A switch retires the old controller without blocking (queued ops
dropped, pending ones reported failed, its client shut down) and
destroys it on a background thread. It no longer saves on the way out:
the new client re-scans the blocks since the last save, as after a
restart.
- A switch waits for an op in flight, at most 15 s, so the op can't
finish and save on the old client while the new one owns the wallet
file. The Network tab already shows the new selection meanwhile.
- The refresh worker stops between backend commands, skips the remaining
server reads once one fails, and then retries after 4, 8, then 15 s
instead of every 2 s.
- Quit waits up to 2.5 s for an op in flight, e.g. Encrypt's save.
Encrypt and Remove encryption now report a failed save: the backend
returns it as a JSON error that the old check never saw, so the failure
was reported as success. The trimmed copy of a key typed into Import is
now wiped too, and stopping the refresh worker can no longer miss its
wakeup.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Mirror the async-open path for wallet creation. beginOpenExisting() and
beginCreateWallet() now both delegate to beginAsyncLifecycle(bool create),
which runs the backend init on a detached thread and walks the failover
server list (preferred server first, then all usable defaults), reporting
the preferred server's error on total failure. The first-run wizard's
Create button drives this through a non-blocking "creating" poll state so
the UI no longer freezes while the backend contacts a (possibly flaky)
lightwalletd. The created seed response is securely wiped immediately and
read back via exportSeed for the reveal/verify steps.
Safe because litelib_initialize_new contacts the server before writing any
wallet file and LightClient::new errors if a wallet already exists, so a
failed candidate leaves no partial state.
Tests: fake backend's initialize_new now honors the dead/warmup server
substrings; testLiteWalletControllerOpenFailover gains a create-failover
case (preferred dead, fallback good -> walletOpen).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
When the preferred lightwalletd server is reachable but warming up (JSON-RPC -28
/ "Activating best chain"), the failover treated it like a dead server and fell
through to the others, so the wallet didn't open until the next 20s retry — even
though the healthy server was ready within seconds.
Detect the warmup error during failover, flag it on the open outcome
(lastOpenWasWarmup()), and have the App retry on a short ~4s interval in that case
instead of 20s, so the wallet opens promptly once warmup clears. A unit test
covers a warming-preferred + dead-fallback open setting the flag.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Opening an existing lite wallet ran synchronously on the UI thread and used a
single server, so a dead/unreachable lightwalletd server froze startup for the
connect timeout and then stranded the wallet ("disconnected" spinner) — and the
DragonX lite servers are flaky (often several down at once).
Add LiteWalletController::beginOpenExisting() / pumpAsyncOpen(): the open runs on
a background thread (mirroring the sync/broadcast shared-lifetime pattern — it
captures only shared_ptrs + value copies, never `this`), trying the preferred
server first and then every other usable default until one succeeds. The main
thread finalizes the result (flips walletOpen, starts sync) or records the reason.
The rollout gate is still checked up-front on the main thread.
App: auto-open now calls beginOpenExisting() and pumps it each tick, retrying on
a 20s interval so a transient outage self-heals once a server returns; a failed
open surfaces its reason (notification + Network tab) instead of a silent spinner.
Tested: a fake bridge that fails specific servers exercises both
preferred-dead -> fallback-opens and all-dead -> fails-with-reason. Built clean
for full-node, lite, and Windows cross-compile.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Startup lock screen (soft): once the first refresh reveals the auto-opened wallet
is encrypted+locked, show the unlock modal on launch (reusing renderLiteUnlockPrompt,
one-shot per session). Soft by design — balances stay viewable via viewing keys
while locked, so the user may dismiss and browse read-only; only spending needs
the passphrase.
Real-backend verification: add `lite_smoke --encrypt` (create -> encryptionstatus
-> encrypt -> lock -> unlock, checking flags; passphrase never printed). Running it
against the real SDXL backend showed encrypt LOCKS immediately
(after encrypt: encrypted=1, locked=1) — the backend removes spending keys right
after encrypting. The controller already relays encryptionstatus faithfully (UI is
state-driven, so unaffected), but the fake modeled encrypt->unlocked; corrected the
fake (encrypt -> encrypted+locked) and the test sequence (encrypt -> unlock -> lock
-> decrypt) to match real behavior.
Builds clean, tests pass, hygiene clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Wire the backend passphrase-encryption commands into LiteWalletController:
- encryptWallet / decryptWallet (take passphrase by value, securely wipe it,
save after), unlockWallet / lockWallet (bring spending keys into/out of
memory), and encryptionStatus() -> {encrypted, locked}. All return
failure-safe results; errors arrive as {"error":..} or "Error:" (handled).
- Fold encryptionstatus into refreshModel() (polled every cycle, available even
mid-sync since it reads local wallet state) and apply it in
applyLiteRefreshModelToWalletState, so WalletState.isEncrypted()/isLocked()
track the backend — which gates the existing locked/auto-lock UI.
Backend contracts verified against the SDXL source: encrypt/unlock/decrypt take
the passphrase as the single arg; lock takes none; encryptionstatus returns
{"encrypted","locked"}; ops return {"result":"success"} / {"error":..}.
Tests: testLiteWalletControllerEncryption drives encrypt -> lock -> unlock ->
decrypt via encryptionStatus(), checks empty-passphrase + closed-wallet rejection,
and that the status folds into WalletState. Fake models the state machine.
GUI wiring (encrypt in Settings, unlock prompt / lock action) is the follow-up;
the backend create flow remains unencrypted by default until encrypt is run.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Verified against the SDXL Rust source that the backend auto-saves only on
new-address / import / rescan; it does NOT save after sync, send, or shield, and
litelib_shutdown merely sets a flag. So without intervention a first sync
(~30 min) and any sent transaction are lost on restart.
The controller now triggers the backend `save` at exactly the right points:
- after the detached `sync` completes — and BEFORE syncDone_ is set, so a
syncComplete() observer always sees a fully persisted wallet;
- after a successful send / shield (the doSend/doShield cores; skipped on
failure so a failed broadcast doesn't write);
- a guarded best-effort flush in the destructor, only when syncDone_ and no
broadcast is in flight, so shutdown never blocks on the wallet lock held by an
uninterruptible scan or in-progress proving;
- plus a public saveWallet() for explicit/periodic saves.
Wallet-file crash recovery (.dat / .dat.bak rotation) is already handled inside
the backend.
Tests: testLiteWalletControllerM5Persistence proves saves fire after
sync/send/shield and explicit saveWallet(), and do NOT fire on a failed send or
with no wallet open (fake gains a save counter). Plan doc updated; M5b
(packaging/CI/signing/rollout) remains.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add the spend & backup surface to LiteWalletController, with the real SDXL
backend contracts verified against the Rust source:
- send / shield: ASYNC (detached broadcast thread + takeBroadcastResult() slot,
mirroring the sync thread's shared-lifetime pattern, since sapling proving can
take seconds), plus synchronous *Blocking cores for tests. send uses the
JSON-array form ([{address,amount,memo}]) because litelib_execute passes the
whole args string as ONE argument (no whitespace split) — the space-separated
CLI form would never parse. send/shield report failure via {"error":..} in the
body (NOT an "Error:" prefix), so the result is derived from the parsed JSON.
- importKey: auto-detects transparent WIF (U/5/K/L -> timport) vs shielded key
(-> import); takes the key by value and securely wipes it before returning.
- exportPrivateKeys / exportSeed: synchronous local reads returning SECRET
material (flagged: no logging; caller wipes after the user saves the backup).
- broadcast thread is detached in the dtor (captures shared bridge + flag + slot,
never `this`), so it is safe to outlive the controller.
Tests: testLiteWalletControllerM4 drives send (success / no-recipients /
{"error":..} / async-slot delivery / pre-open rejection), shield, export, seed,
and import (shielded + WIF + pre-open). Fake backend returns the real command
shapes + a g_liteFakeSendFails error toggle.
GUI wiring (send_tab button, backup/import UI) is deferred like the M3 UI hop
(GUI-unverifiable here). Plan doc updated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Re-audited this session's lite-wallet changes (originally written at medium
effort) and fixed the genuine issues found:
- walletReady (open path): litelib_initialize_existing returns the bare string
"OK", which is NOT valid JSON, so the previous `json::accept(value)` check
marked a *successful* open as not-ready. Key off a non-empty success response
instead (the bridge already maps "Error:"/null to failure). Drops the now
unused nlohmann include.
- sync progress: while the detached sync thread is still running, syncDone_ is
authoritative — don't surface the backend's transient idle syncstatus
({"syncing":"false"} -> parser progress=1.0/complete=true) as a misleading
100%/done. Force complete=false and zero the bogus 1.0 in the progress model.
- per-address balance: also exclude `pending` outputs (notes/utxos from an
unconfirmed received tx) so per-address figures match confirmed/available.
- secret wiping: the settings page left the page-local request copies
(input.request.*Request.{passphrase,seedPhrase}) unwiped, and the
validation-only fallback path wiped nothing. Replace the single-path memzero
with an RAII scrubber that wipes both the UI char buffers and the request
string copies on every return path.
- concurrency: document that concurrent bridge->execute() is intentionally
unguarded — litelib serializes wallet access internally via
Arc<RwLock<LightWallet>>, so a C++ mutex is unnecessary and would defeat the
sync/syncstatus concurrency the design relies on. syncLaunched_ -> atomic.
Tests: fake backend now returns the real init shapes (seed object for
create/restore, bare "OK" for open) and a new open-path case guards the
walletReady regression. Removed an unreliable alloc==freed leak assert from the
thread-bearing controller test (kept in the thread-free bridge test). Also fixed
a stray CMake indent and removed ~220MB of untracked build/debug scratch.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- LiteWalletController::newAddress(shielded) runs the backend "new" command ("zs"/"R" ->
do_new_address), parses the ["addr"] response, and returns the new address; the next
refresh lists it. Fast (local derivation), safe on the UI thread.
- fake_lite_backend returns ["zs1fakenew"]/["R1fakenew"] for "new" by args.
- testLiteWalletControllerNewAddress covers shielded/transparent + no-wallet error.
Also confirmed (no code needed): the sync-progress indicator already works for lite —
balance_tab reads state.sync.* which M2b-3 populates. Per-address balances landed in M2.
Remaining M3 is pure UI wiring (receive_tab button -> newAddress, loading/empty states),
which isn't verifiable without a GUI session.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
LiteWalletGateway::refresh() aborted the entire refresh on the first command whose bridge
call or parse failed — which turned a single real-backend shape mismatch (e.g. syncstatus)
into a total, empty-everything refresh. Since the balance/addresses/list real shapes are
still unverified and we've already hit shape drift twice, make refresh resilient:
- Run every planned command; assembleLiteWalletRefreshBundle already skips failed results.
- result.ok = any usable data came back (bundle.complete still reflects all-succeeded).
- One command's failure now degrades gracefully — the other sections still populate.
testLiteWalletGatewayRefreshSkipsFailedCommand (fake balance returns invalid JSON) asserts
the refresh still succeeds with addresses/transactions/info populated and balance skipped.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The backend `sync` command is a blocking, uninterruptible full chain scan (do_sync(true);
does not honor the shutdown flag), and balance/list block until synced. Previously
startSync() ran on the main thread (would freeze wallet creation) and the worker could
block, making the destructor join() hang at shutdown.
Redesign:
- bridge is now std::shared_ptr<LiteClientBridge>, shared with a detached sync thread so
detaching is safe and litelib_shutdown isn't called while a running sync still holds the
bridge; the controller's own ref prevents premature shutdown during normal operation.
- startSync() launches the blocking `sync` on a detached thread (non-blocking; never joined).
- refreshModel() gates on syncDone_: while syncing it publishes syncstatus progress only;
once synced it does the full balance/addresses/list refresh (now fast).
- destructor joins only the fast poll worker and detaches the sync thread -> no hang.
- syncComplete() accessor added.
Tests (deterministic, via a blocking-sync fake; counters made atomic for the detached
thread): testLiteWalletControllerShutdownDoesNotHangDuringSync (destructor returns <1.5s
with sync blocked); refresh/worker tests wait for syncComplete()/a balance-bearing model.
Stable across repeated runs; lite+backend and full-node apps build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The real backend returns syncstatus as idle {"syncing":"false"} (string) or in-progress
{"syncing":"true","synced_blocks":N,"total_blocks":M} (commands.rs:83-87), but
parseLiteSyncStatusResponse hard-required the block fields and failed whenever the wallet
wasn't actively syncing — so sync/progress never updated in the real app.
- Read "syncing" as a string; require synced_blocks/total_blocks only when syncing=true;
idle => complete, synced/total 0.
- fake_lite_backend syncstatus now uses the real "syncing":"true" shape.
- testLiteSyncStatusParserRealShapes covers idle, in-progress, and missing-counts-while-syncing.
- Verified against the live backend via lite_smoke --refresh (syncstatus parse_ok=1).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Shared-bridge refactor (litelib is a global singleton; every LiteClientBridge calls
litelib_shutdown() on destruction, so services must not each own one):
- LiteWalletLifecycleService, LiteWalletGateway, LiteSyncService now take a non-owning
LiteClientBridge*; LiteWalletController owns the single bridge and passes &bridge_.
Sync + controller refresh:
- LiteSyncService::startSync executes the real "sync" command (was a stub).
- LiteWalletController: startSync() (auto-fires when a wallet becomes ready) and
refreshWalletState(WalletState&) — polls syncstatus, runs gateway.refresh(), maps the
bundle, applies balances/addresses/transactions/sync into WalletState.
Tests:
- fake_lite_backend.h returns command-shaped JSON (per tests/fixtures/lite/result_parsers.json).
- testLiteWalletControllerRefreshPopulatesState drives the full path against the fake.
- Surfaced + worked around a real integration issue: parseLiteInfoResponse requires
latest_block_height and the gateway aborts the whole refresh on the first command's
parse failure (fragile vs partial backend responses; hardening tracked for M2b-3).
Verified: ctest green; lite+backend, full-node, lite-no-backend apps + lite_smoke build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
DanSandClaude Opus 4.8