Second half of "migrate a legacy wallet to a seed wallet": sweep all funds
into the new seed wallet, then adopt it as the primary wallet.
- Sweep (beginSweepToSeedWallet): z_mergetoaddress ["ANY_TADDR","ANY_ZADDR"]
-> the new wallet's z-address, at the default fee, limits 0,0. The opid is
tracked via the existing send-callback pipeline; the txid is persisted.
- Confirming gate (fund-safety): after the sweep, pollSweepStatus() polls the
tx confirmations + legacy balance. "Make this my wallet" only unlocks once
the sweep is MINED (>= 1 conf) AND the old wallet is empty — so wallet.dat is
never swapped while the funds could still bounce back or a remainder is left.
A too-big-for-one-tx remainder offers a "Sweep remaining" re-sweep.
- Adopt (beginAdoptSeedWallet, background): stop daemon -> move legacy
wallet.dat to a timestamped .bak (never deleted; restored on failure) ->
install the new wallet -> restart with -rescan. Exception-safe;
daemon_restarting_ gates tryConnect + is always cleared; skips the restart
while shutting down; beginShutdown joins the task first.
- Resume: a pending migration reopens at Sweep, or at Confirming if the sweep
txid is already recorded (re-derived from the chain) — never re-sweeps blind.
- Secret hygiene: the seed is wiped leaving ShowSeed, on any modal dismiss
(detected after BeginOverlayDialog), and in App::~App() (same fix applied to
the seed-backup dialog).
Built both variants; tests + hygiene pass. Two rounds of parallel adversarial
review (12 findings fixed, then a clean re-verify) gate this fund-moving code.
The migration modal uses English literals (as renderBackupDialog does); it gets
translated once the flow is finalized. Live sweep test against a funded wallet
is the next step.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The seed-wallet creator forced TLS on its RPC client, but the DRAGONX RPC is
plaintext HTTP on localhost (-tls=only applies to P2P, not the RPC; the main
GUI's conf has no rpctls key and connects over http too). Forcing https made
every connect fail with "SSL connect error", so the isolated create never
completed. Connect without TLS, matching the main daemon path.
Also add an env-gated (DRAGONX_SEED_CREATE_SMOKE=1) headless smoke in main()
that runs SeedWalletCreator::create() before GUI init and prints
ok/word-count/address (never the seed words) — used to validate the full C++
creator path without the GUI. Harmless in normal runs.
Validated end-to-end against the hd-transparent-keys (v1.0.3-dc45e7d9)
daemon: ok=1, words=24, a shielded z-address returned, temp datadir cleaned up.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
First, fund-safe step of "migrate a legacy wallet to a seed wallet": mint a
brand-new BIP39 mnemonic wallet in an ISOLATED throwaway datadir so the user
can later sweep funds into it. No funds move and the real wallet.dat / main
daemon are never touched — the isolated node runs concurrently on its own port.
- EmbeddedDaemon: one-shot isolated-datadir override (setNextStartOverride)
consumed on the next start(); a skip-default-port-check for an isolated
instance beside the main daemon; and a tcpPortInUse() probe. The datadir's
basename must be the assetchain name (DRAGONX) and the daemon reads
<datadir>/DRAGONX.conf itself, so no -conf is passed (both learned from
live testing against the hd-transparent-keys daemon).
- SeedWalletCreator (src/daemon): picks a free port, writes a throwaway conf,
starts an isolated dragonxd with -usemnemonic=1 -connect=0, waits for RPC,
calls z_exportmnemonic + z_getnewaddress (the sweep target), stops it, and
keeps/scrubs the temp datadir. Off-thread; the seed is wiped after use.
- App: beginCreateSeedWallet (background) -> pumpSeedMigration (main-thread
handoff) -> renderSeedMigrationDialog (Intro -> Working -> Show seed ->
Error), reusing the seed_display grid + clipboard auto-clear.
- Settings: "Migrate to seed…" button (full-node gated) + persisted pending
migration state (dest address + temp datadir) for Phase 2 to adopt.
Validated end-to-end against the hd-transparent-keys daemon: the isolated
node comes up in ~4s and returns a 24-word mnemonic + a shielded z-address.
Requires that daemon (the bundled Jun-28 build lacks z_exportmnemonic). The
migration modal uses English literals for now (as renderBackupDialog does);
the whole flow gets translated once Phase 2 finalizes it. Phase 2 (sweep +
adopt) is next.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
New full-node wallets now get a real 24-word recovery phrase, and users can
reveal/back it up from Settings.
- Pass -usemnemonic=1 to dragonxd. The daemon reads it only inside
GenerateNewSeed() when a wallet has no seed yet, so it is inert on existing
wallets (safe to pass unconditionally) and makes every freshly-created
wallet mnemonic-backed — its phrase is then exportable via z_exportmnemonic
and portable to SDXLite/ObsidianDragonLite. (Existing/legacy wallets are
unaffected and keep using the chat identity's z_exportkey fallback.)
- App::exportSeedPhrase() wraps z_exportmnemonic on the RPC worker with
sodium_memzero wiping on every path; it flags the daemon's "not derived
from a mnemonic" error as a legacy wallet rather than a failure.
- New "Back up seed phrase" modal (Settings -> Backup & Data, full-node
gated): reveals the phrase in a numbered word grid, copy (45s clipboard
auto-clear) + save-to-file (0600), wipes the secret on every close path
incl. dismiss-mid-fetch. Shared ui/windows/seed_display.h grid helper.
- One-time nudge (settings flag seed_backup_reminded) toasts mnemonic-wallet
users to back up their phrase; legacy wallets are not nagged.
- 15 new strings translated into all 8 languages (additive, 957->972 keys);
CJK subset font rebuilt for the new glyphs.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Each scripts/gen_<lang>.py held a hardcoded `translations` dict that it
wrote over res/lang/<lang>.json wholesale. Those dicts had drifted ~285
keys behind the committed JSONs (keys were added directly to the JSONs
over time), so running any generator silently truncated a language file —
a live regression trap. Nothing invokes them (not build.sh, CMake, or CI),
and the JSONs are now the maintained source of truth, so the generators are
pure duplication plus a footgun.
Retire all eight. The additive, non-destructive scripts/add_missing_
translations.py remains the i18n helper, and CLAUDE.md now documents the
"edit the JSONs directly and additively" workflow.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
tracePageName()'s switch over ui::NavPage was missing the two lite-only
tab values (LiteConsole, LiteNetwork), producing -Wswitch warnings in the
lite build. Add both, tracing them to the same labels as their full-node
siblings ("Console tab" / "Network tab") — exactly one of each pair shows
per variant. Trace-string only; no behaviour change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The HushChat UI labels were already translated, but the send-path
notifications (not-connected, no-z-address, waiting-for-reply, compose
failures, contact-request queued, lite busy) were still hardcoded English.
Route all eight call sites in app_network.cpp through TR() with seven new
keys (the three "no z-address" variants unified into chat_toast_no_zaddr),
add the English fallbacks in i18n.cpp, and translate the keys into de/es/fr/
ja/ko/pt/ru/zh — reusing each file's existing chat_* terminology (z-address,
contact request, message) for consistency. Additive only (950 -> 957 keys
per language). CJK subset font rebuilt to cover the new zh/ja/ko glyphs.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
checkAutoLock() called App::lockWallet(), which early-returns without an
rpc_ handle — so lite wallets, which have no daemon, never idle-locked and
stayed unlocked indefinitely. Route lite through lockLiteWallet() instead,
which locks the backend and tears down the chat session so no decrypted
store / unlocked DB key survives the idle lock. Full-node builds keep
lite_wallet_ null, so the branch is a no-op there and behaviour is unchanged.
The idle timer (last_interaction_, reset on ImGui input at the top of
App::update()) is already build-agnostic, so no other wiring is needed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add the 15 HushChat UI strings (nav label, locked/empty/select/waiting hints,
composer + new-conversation dialog, You / contact request / not-sent tags) to
res/lang/{de,es,fr,ja,ko,pt,ru,zh}.json. Previously they fell back to English.
Added directly to the JSON files (935 -> 950 keys, additive only): the
scripts/gen_<lang>.py generators are stale — they emit ~285 fewer keys than the
committed JSONs, so regenerating from them would silently drop translations.
(Pre-existing issue; noted for follow-up. The generators are left untouched.)
Rebuilt the usage-based CJK subset font (res/fonts/NotoSansCJK-Subset.ttf) from
the full JSONs so the new chat glyphs (聊天 / 对话 / 会話 / 답장, etc.) render —
verified every CJK code point in the full JSONs is covered (no glyph regression).
Verified: Linux + Windows build (embedded lang headers regenerated via xxd, font
re-embedded via INCBIN), hygiene clean; per-file diff is additive (15 keys, 0
deletions).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Outgoing echoes were recorded optimistically regardless of whether the broadcast
was actually submitted. Now broadcastChatMemos / broadcastChatMemosLite return
whether the send was submitted (false on immediate failures: not connected, no
spendable z-address, or a lite send already in progress), and the echo records a
ChatDelivery status (Sent / Failed). The Chat tab shows a "not sent" marker in
the error color on failed messages. The status persists in the DB (backward-
compatible: old rows deserialize as Sent).
(A later async failure — an opid that fails after submission — still surfaces via
the existing send-progress notification; only the submit gate is reflected here.)
Also removes the now-unused chat_readonly_note string (the composer replaced the
read-only footer). Test: delivery status round-trips through the DB. Verified:
Linux + Windows build with chat ON, ctest 100%, hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Flip DRAGONX_ENABLE_CHAT to default ON (a fresh configure now builds chat in),
and make full-node chat work for every wallet + any daemon.
Legacy compat: chat identity is derived from the wallet's mnemonic
(z_exportmnemonic) for a portable, SDXLite-compatible identity — but a legacy
random-seed wallet has no mnemonic, and released daemons don't have that RPC at
all. Provisioning now falls back to a stable z-address's spending key
(z_exportkey) in those cases: a functional, wallet-local identity that works on
any daemon. Since a chat identity is local (peers learn your public key from
your message headers, not your derivation), interop is unaffected; only
cross-client portability needs the mnemonic. The spending key is an in-memory
KDF input over a key the wallet already holds, wiped after use — no new exposure.
Stability: the chosen chat z-address (the reply-to in headers AND the legacy
identity source) is now persisted in settings (chat_reply_zaddr), so the
identity + reply address don't shift when new addresses are generated.
chatReplyZaddr() picks the smallest spendable z-addr once and reuses it.
CLAUDE.md updated to reflect the default flip. Verified: Linux + Windows build
with chat ON, fresh-configure default confirmed ON, ctest 100%, hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Security-review finding (low): on the lite variant, "Lock now" ran the backend
`lock` but never updated state_.locked — that only refreshes on the next ~2s
poll. Since chat teardown is driven by state_.isLocked() in maybeProvision
ChatIdentity, the decrypted in-memory store, the chat identity secret key, and
the seed-derived AEAD DB key all lingered in RAM for up to ~2s past an explicit
lock (the full-node path closes this immediately, since App::lockWallet sets
state_.locked synchronously).
New App::lockLiteWallet() mirrors the full-node behavior: on a successful lite
lock it sets state_.locked and tears down the chat session now (clearIdentity +
store().clear() + chat_db_.lock() + re-arm). The lite "Lock now" button routes
through it. (Note: the lite variant has no auto-lock at all — checkAutoLock ->
App::lockWallet early-returns for lite — which is a separate, general lite gap.)
Completes the end-to-end feature security review (1 confirmed finding of 3 raw,
now fixed). Verified: Linux + Windows build with chat ON, ctest 100%, hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add App::seedChatDemoData() + a "Seed demo chat" button in the DEBUG OPTIONS
card (shown only in chat-enabled builds, next to the screenshot sweep). It gives
the Chat tab a demo identity and injects a few sample conversations into the
in-memory store (NOT persisted — no DB attached, gone on restart) so the
screenshot sweep captures the tab's real UI (conversation list, thread bubbles,
composer, and the "waiting for reply" contact-request state) instead of the
empty "unlock" hint. No-op when the chat feature is off.
Verified: Linux + Windows build with chat ON, ctest 100%, hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The full-node receive harvest works off z_viewtransaction; the lite wallet's
transactions come from the backend instead, so lite chat receive was unwired.
App::ingestLiteChatMemos runs after each lite refresh: the backend lists one
entry per received (non-change) note — same txid, distinct position + decoded
UTF-8 memo — so a chat tx yields two entries (header + payload). Group the
Receive-kind notes by txid, feed them through the SAME extractHushChatTransaction
Metadata parser (now order-tolerant, so the daemon's output shuffle is a non-issue),
and thread the result into ChatService. The store dedups (txid+position), so
re-listing every refresh is harmless.
This closes the lite variant's chat loop end-to-end: identity (exportSeed), DB
unlock/load, receive (here), UI, and send (broadcastChatMemosLite). Gated by
DRAGONX_ENABLE_CHAT (default OFF). Verified: Linux + Windows build with chat ON,
ctest 100%, hygiene clean; caches restored to OFF.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The full-node daemon unconditionally shuffles a transaction's shielded outputs
(transaction_builder.cpp ShuffleOutputs, to hide the change position), so the
two 0-value HushChat memo outputs land at random note positions — the header is
NOT guaranteed to precede its payload. The receive parser paired a header only
with the NEXT payload in position order, so a shuffled send failed to pair (the
message was silently lost) ~50%+ of the time.
groupHushChatMemoOutputs now holds an orphan payload (one seen before its header)
and pairs it with the header when it arrives, so header↔payload pair regardless
of on-chain order. A chat tx carries exactly one header + one payload (the harvest
already skips change outputs), so this is unambiguous.
This is the complete fix for the ObsidianDragon ecosystem: full-node and lite both
receive through this one C++ parser, so full-node→full-node and full-node→lite now
work despite the shuffle, with no daemon fork and no lite-backend change. Context:
SilentDragonXLite (which hardcodes payload position==1 and forced the daemon/backend
question) is being replaced by ObsidianDragonLite. Lite send never shuffles.
Tests: shuffled receive (payload-before-header, and change-interspersed) pairs +
decrypts; all existing ordered-input tests unchanged. Verified: Linux + Windows
build, ctest 100%, hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Make composed messages actually reach the wire: broadcastChatMemos sends the
header + payload as two 0-value memo outputs to the peer's z-address (header
first, the lower memo position), on both variants.
- chat_outgoing: chatSendOutputs(memos, utf8Prefix) — the pure, testable memo
encoder. Full-node memos get a "utf8:" prefix (dragonxd rejects raw JSON and
then UTF-8-encodes on-chain, byte-identical to SilentDragonXLite's
Memo::from_str); the payload is NOT double-hex. Lite memos are raw UTF-8 (the
backend does Memo::from_str directly). Header is always output 0.
- App::broadcastChatMemos: full-node builds a two-recipient z_sendmany array
(amount 0, from the spendable reply z-address, default fee) via submitZSendMany
with markFeeGapRetry=true — deliberately, to suppress the fee-gap auto-retry,
which rebuilds a single-recipient tx and would drop the payload output. Lite
routes to broadcastChatMemosLite (two 0-value LiteSendRecipients, raw memos;
the backend accepts duplicate addresses + 0-value for exactly this pattern).
- Encoding + design established by a four-codebase mapping (wallet, daemon, SDXL,
lite backend) and an adversarial review of the wiring (0 confirmed findings).
- Tests: chatSendOutputs (utf8:-prefix + header-first for full-node, raw for
lite) + on-chain round-trip (strip "utf8:" -> the harvest parser re-pairs and
decrypts).
Remaining as a LIVE test (cannot be proven from source): that dragonxd returns
the memo under memoStr verbatim, that recipient-array order maps to note
position, that a 0-value memo-only tx relays, and full SDXL<->DragonX interop.
Gated by DRAGONX_ENABLE_CHAT (default OFF). Verified: Linux + Windows(mingw)
build with chat ON, ctest 100%, hygiene clean; caches restored to the OFF default.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add the outgoing side of HushChat: compose messages and start conversations.
The wire construction is the byte-exact inverse of the receive parser and is
proven by a self-consistent round-trip (build → parse → decrypt).
- src/chat/chat_outgoing.{h,cpp} (pure): buildOutgoingMessage encrypts via
encryptOutgoing and buildOutgoingContactRequest carries plaintext; both emit
the header memo JSON ({h,v,z,cid,t,e,p}, which nlohmann serializes
alphabetically to match SilentDragonXLite) + the payload memo, validating the
512-byte memo limit, the 64-hex peer key, and the "no leading '{'" rule for
request text. My public key goes in header "p"; the peer's key is the
encryption recipient.
- ChatService: composeMessage/composeContactRequest (encrypt with the held
identity) + recordOutgoing (echo an Outgoing ChatMessage into the store + DB —
we never harvest our own sends, so this is the only local record).
- App: sendChatMessage(cid,text) sources the peer z-addr + public key from the
conversation, composes, and records a random-id echo; startChatConversation
(zaddr,text) mints a random cid + composes a plaintext contact request;
chatReplyZaddr() picks a spendable z-addr. broadcastChatMemos() is the Phase-5
transport seam (network delivery + real-SDXL interop verification land there).
- Chat tab: a message composer (shown once the peer's key is known, else a
"waiting for reply" hint) + a "New conversation" modal that sends a contact
request to a z-address.
- Tests: outgoing round-trip through the receive parser + decrypt, contact-request
passthrough, validation guards, and ChatService compose + recordOutgoing echo.
Adversarially reviewed (crypto/interop, secret hygiene, logic, UI) — 0 findings.
Gated by DRAGONX_ENABLE_CHAT (default OFF). Verified: Linux + Windows(mingw)
build with chat ON, ctest 100%, hygiene clean; caches restored to the OFF default.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add the first user-visible HushChat surface: a Chat tab in the sidebar with a
two-pane, read-only conversation view — a conversation list on the left and the
selected thread on the right — reading from the App-owned ChatService store.
- New src/ui/windows/chat_tab.{h,cpp} (RenderChatTab(App*), mirroring the
Contacts tab). Conversations are sorted by most-recent activity; peer
z-addresses resolve to contact names via the address book when known; each
message shows sender + timestamp + wrapped body, with a contact-request tag
and a read-only footer (composing arrives in a later phase). Empty states
cover "wallet locked / identity not ready" and "no conversations yet".
- Sidebar wiring: NavPage::Chat + registry entry (static_assert stays balanced),
NavPageSurface + GetNavIconMD (ICON_MD_CHAT), dispatch in app.cpp, trace/sweep
page names, and App::chatService() accessor. The tab is gated on the new
WalletUiSurface::Chat, which returns DRAGONX_ENABLE_CHAT != 0 — so it only
appears in chat-enabled builds and is hidden (and unreachable) by default.
- i18n: English defaults for the chat nav label + hint strings.
Verified: Linux + Windows(mingw) build with chat ON, ctest 100%, hygiene clean;
caches restored to the OFF default. The screenshot sweep now includes the Chat
tab (sweepPageName "chat") for per-theme visual review.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Swap the in-memory-only chat store for durable sqlite persistence with real
per-transaction timestamps, encrypted at rest under a key derived from the
wallet's own seed (no passphrase, works on encrypted and unencrypted wallets).
- ChatDatabase (src/chat/chat_database.{h,cpp}): sqlite store mirroring
data::TransactionHistoryCache. unlockWithSecret(seed) derives a 32-byte AEAD
storage key and a wallet-partition tag via domain-separated keyed BLAKE2b
(generichash) contexts. Every record — bodies, peer z-addrs, threading,
timestamps — is crypto_aead_xchacha20poly1305_ietf-encrypted with a random
nonce and the wallet tag as associated data; even the dedup key is a keyed
hash of txid+position, so nothing about your conversations is in cleartext on
disk. Rows are partitioned per-wallet; a different seed sees nothing. Messages
are decrypted once at ingest then re-encrypted under the storage key, so
load() needs only the storage key, not the chat identity.
- ChatService: ingest() now stamps each message with its own transaction time
(txid->time map + fallback) and writes new (store-deduped) messages through to
the database; loadFromDatabase() rehydrates the in-memory read model on unlock.
- App: unlock the chat DB with the same seed in provisionChatIdentityFromSecret
and load prior history; lock the DB + clear the decrypted in-memory store on
relock and on lite-controller rebuild.
Adversarial review (4 confirmed findings, all fixed): don't provision if the
wallet locks mid-fetch (re-check isLocked at completion); wipe the serialized
plaintext temporary in append(); trim the seed into a separate fully-wiped
buffer (no residue past a shrunk size()); scrub the mnemonic copy in the RPC
json result.
Tests: ChatDatabase round-trip (persist/reload, field + order fidelity), dedup,
per-wallet isolation, lock inertness, and ChatService write-through + reload
without an identity. Gated by DRAGONX_ENABLE_CHAT (default OFF). Verified:
Linux + Windows(mingw) build with chat ON, ctest 100%, hygiene clean; caches
restored to the OFF default.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Complete the Phase-1 App integration for HushChat, now that the dragonx
`hd-transparent-keys` daemon exposes a BIP39 mnemonic (z_exportmnemonic)
that is byte-compatible with SilentDragonXLite. Both variants derive the
chat identity from the wallet's OWN seed phrase, so it is portable across
full-node and lite (same words -> same KDF input -> same identity) and
recoverable from the single wallet backup.
- App owns a dragonx::chat::ChatService; maybeProvisionChatIdentity() runs
each update() tick and, once the wallet seed is reachable and unlocked,
derives the identity via deriveChatIdentityFromSecret and sets it on the
service. Full-node fetches z_exportmnemonic off the UI thread via the RPC
worker; lite reads it synchronously through LiteWalletController::exportSeed.
Secrets are wiped on every path.
- Wire the previously-dead TransactionRefreshResult.hushChatMetadata sink:
both the full and recent transaction-refresh completions now ChatService
::ingest the harvested memos (before the result is moved) so incoming
messages are decrypted and threaded. The store dedups across both paths.
- Robust provisioning gates: skip + re-arm on relock (isLocked, mirrored for
both variants), don't fetch before the encryption state is known, and treat
a non-mnemonic wallet (RpcError) as identity-unavailable rather than
retrying every tick. rebuildLiteWallet re-arms so a server switch / re-open
re-derives from the newly opened wallet.
Gated by DRAGONX_ENABLE_CHAT (default OFF) via the constexpr
hushChatFeatureEnabledAtBuild() gate, so the wiring folds away in shipping
builds. Verified: Linux + Windows(mingw) build with chat ON, ctest 100%,
hygiene clean; caches restored to the OFF default.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 1 Steps 4-6 (the receive pipeline, minus the App/sync wiring which
lands next).
- chat_message: the in-memory ChatMessage model (direction, kind, txid, cid,
peer zaddr/pubkey, body, timestamp, payload_position). No libsodium.
- chat_store: threads messages by conversation_id and dedups by
(txid, payload_position) so re-scanning the chain never double-inserts.
- chat_service: owns the long-lived chat identity keypair (move-disabled,
wiped on destruction/clear) and the store. ingest() decrypts each Message
(drops undecryptable ones silently — no plaintext/memo logging), passes a
ContactRequest's plaintext through, and threads the result. No-op without an
identity.
Tests: a metadata batch decrypts into a threaded conversation; re-ingest
dedups; a contact request carries through; a wrong identity decrypts nothing;
no-identity ingest is a no-op.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 1 Step 1 (the linchpin). The receive path had nothing to decrypt:
HushChatTransactionMetadata carried only payload_position/size, and the
extractor discarded the paired payload memo + the header's "e"/"p". Widen the
metadata with sender_public_key_hex ("p"), secretstream_header_hex ("e"), and
payload_memo (ciphertext hex for a Message; request text for a ContactRequest),
populated from the already-parsed header/payload pair.
Add an end-to-end receive-path test: derive two identities, encrypt Alice->Bob,
wrap it as a HushChat header+payload memo pair, run it through
extractHushChatTransactionMetadata, and decrypt straight from the carried
metadata. Also asserts the harvest yields nothing when the feature is disabled.
Fields are only populated when featureEnabled (the extractor already returns
early otherwise), so an OFF build produces no decrypt material.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 1 crypto foundation (gated by DRAGONX_ENABLE_CHAT; the pure primitives
are always compiled + unit-tested, the feature gate lives at the future
service layer).
- chat_identity: derive the DragonX-native X25519 (crypto_kx) identity from a
stable per-wallet secret via a domain-separated keyed BLAKE2b KDF
(crypto_generichash, context "DragonX-HushChat-Identity-v1") into a clean
32-byte crypto_kx seed. Deterministic; skips SDXL's UTF-8-hex-seed quirk
(that's the Phase-4 import path). Per §5.6.
- chat_crypto: encryptOutgoing (server_tx) / decryptIncoming (client_rx) via
crypto_secretstream_xchacha20poly1305, byte-exact per Appendix A.3/A.4 so
DragonX interoperates with SilentDragonXLite. Single chunk, TAG_FINAL;
decrypt enforces both the Poly1305 tag and TAG_FINAL (stricter than SDXL).
Every session key / seed / stream state / plaintext scratch is sodium_memzero'd
on all paths; no secret is ever logged.
- Tests: encrypt->decrypt round-trip (incl. empty + UTF-8), identity
determinism, feature-gate + empty-secret handling, and malformed/tampered/
wrong-key inputs all fail safely.
Adversarially security-reviewed (3 lenses: secret hygiene, crypto correctness,
SDXL wire-interop) — confirmed byte-compatible with the SDXL reference in both
directions. Fixes from review: check the secretstream_push return before
reporting Ok; allow the empty-plaintext ciphertext (== ABYTES) so encrypt/decrypt
round-trip symmetrically; document the caller-owns-and-wipes secret contract.
Interop caveat: round-trip proves self-consistency; a real captured SDXL
message is still needed to prove wire-compat end to end.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Screenshot-sweep review follow-ups:
- sweepPageName() had its own page-name switch that lacked a Contacts case,
so the tab's screenshots landed under the fallback "page/" dir. Add the
case so they capture under "contacts/".
- The Z/T type badge (green/amber) washed out on light skins. Use darker,
more-saturated variants when IsLightTheme(); keep the brighter ones on dark.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 0d — the accessibility pass on the Contacts tab.
- Search box filters by label/address/notes (case-insensitive); a distinct
"no matching contacts" empty state.
- Sortable columns (ImGuiTableFlags_Sortable) — click a header to sort the
view by label/address/notes, ascending or descending.
- Keyboard nav (when the tab owns focus, no field/modal active): Up/Down move
the selection through the *visible* order, Enter edits, Delete deletes
(feeding the two-click confirm), Ctrl+C copies.
- Contrast: the address is no longer rendered as muted TextDisabled (it's the
row's key data) — normal legible text, with a coloured Z/T type badge.
Notes are legible too.
- The add/edit form focuses its first field on open (SetKeyboardFocusHere).
- The delete confirm is now visible: the Delete button relabels to
"Confirm delete?" while armed, instead of a transient toast.
Correctness: selection is tracked by STORAGE index, decoupled from the
filtered/sorted visible order, so edit/delete/copy always target the right
entry. (ImGui renders to a canvas with no OS accessibility tree, so this is a
keyboard/contrast/findability pass, not screen-reader support.)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 0c. The Send recipient field only offered Paste and a tx-history
suggestion list — no way to pick a saved contact. Add a compact contacts
icon button next to Paste that opens a picker popup over the App-owned
address book; selecting a contact fills the recipient (re-validated next
frame).
- New header-only src/ui/windows/contact_picker.h: ContactPickerPopup(id,
book, outBuf, outSz) — a reusable popup listing contacts ([Z]/[T] tag +
label + short address) that copies the chosen address into the caller's
buffer. Kept out of the material layer so the design-system headers stay
free of data/ deps; reserved for Chat "new conversation" later.
- send_tab: shrink the recipient input to make room for the icon button
(ICON_MD_CONTACTS), wire the popup to app->addressBook(). Coexists with the
existing Paste + suggestion affordances.
- i18n: send_contacts_button tooltip.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 0b. The address book was buried in Settings behind a modal dialog.
Make it a first-class "Contacts" tab (which will also become the chat
roster), rendered inline in the main content area.
- New NavPage::Contacts (after History; ICON_MD_CONTACTS) +
WalletUiSurface::Contacts. isUiSurfaceAvailable's `default: return true`
shows it in BOTH variants; uiSurfaceNeedsWalletData's default keeps it
usable before wallet data loads. All the touchpoints wired: NavPageSurface,
GetNavIconMD, the app.cpp dispatch case, the app_network.cpp tracePageName
case, and the `contacts` i18n label.
- New src/ui/windows/contacts_tab.{h,cpp}: RenderContactsTab lifts the
toolbar + table + add/edit modal out of address_book_dialog, rendered in a
BeginChild scroll region (peers_tab pattern) instead of an overlay; the
add/edit form stays a modal layered over the tab. Reuses the existing
address_book_* i18n keys and the dialogs.address-book schema.
- Delete address_book_dialog.{h,cpp}; remove its app.cpp render pump and the
dead App::show_address_book_. The Settings "Address Book…" button now
navigates to the tab (setCurrentPage) instead of opening the modal, so the
Tools & Actions grid layout is untouched.
- CMake: swap the dialog sources for contacts_tab.
Behavior-preserving move; search/sort/keyboard/contrast land in 0d. Visual
check pending the per-theme screenshot sweep.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase 0a of the Contacts/Chat plan. The address book was a file-scoped
singleton (`s_address_book` + `getAddressBook()`) trapped inside
address_book_dialog.cpp, reachable only from that TU. Promote it to an
App-owned member so the upcoming Contacts tab, the Send contact picker,
and a future Chat roster all read one source of truth.
- app.h: add `data::AddressBook address_book_;` + `App::addressBook()`
accessors, next to the other owned data models.
- app.cpp: load it once in App::init() (idempotent; missing file is fine;
purely local, no daemon dependency).
- address_book_dialog.cpp: delete the singleton + getAddressBook(); read
through the App* the dialog already carries (dropping the dead
`(void)app;`). show() is static so it can't reach the App — drop its
per-open reload (the book is authoritative for the app lifetime and
self-saves on mutation). Drop the now-unused <memory> include.
No behavior change; groundwork for the Contacts tab.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Centralise the remaining hardcoded English literals in two otherwise
fully-translated files (explorer_tab: 40 TR calls, shield_dialog: 23) so
they join the i18n system with an English source key + fallback:
- explorer_tab: the three search-error messages (invalid response, hash
not found, not-connected) now use explorer_* keys. Also guards a hash
lookup behind an active daemon connection and disables the block-detail
prev/next nav while a fetch is in flight (both were the reason these
error paths could fire).
- shield_dialog: the operation status/error strings (submitting, submitted,
failed, status, error-checking-status, shield/merge failed) now use
shield_* keys.
- i18n: add the new explorer_* and shield_* English source keys.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- mining_benchmark: when a thread benchmark finishes with no nonzero
hashrate sample (e.g. the pool never reported a rate), don't finish
silently — set a new `inconclusive` flag on ThreadBenchmarkUpdate and
let mining_tab warn. Keeps the state-machine core pure/no-I/O (it is
unit-tested), rather than calling the Notifications singleton from it.
- mining_controls: fold the two byte-identical idle-delay combo blocks
(non-scaling + thread-scaling branches) into one lambda parameterised
by combo id; removes ~30 lines of duplication.
- request_payment: drop the dead s_selected_addr_idx state and write the
address-selected comparisons as std::string == char* consistently.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Tighten input handling across several dialogs so bad/edge input can't
produce a doomed request or a confusing display:
- app_security: passphrase strength meter now factors character-class
diversity — an all-one-class string downgrades one tier so "aaaaaaaa"
no longer scores as high as a mixed one.
- block_info: clamp the height field to the chain tip and hide/deny
"Next" at the tip (was only yielding a raw RPC error).
- address_label: trim surrounding whitespace before saving; a
whitespace-only label clears it (mirrors clearing the icon).
- send_tab: re-clamp the amount when a Max send has its fee bumped in the
confirm popup (kept the total within budget) and NUL-terminate the
strncpy'd address/memo when a payment URI overwrites the form.
- settings_page: reject an empty/whitespace-only lite wallet path before
dispatching an unusable open/restore request.
- peers_tab: guard ExtractIP against an empty address.
- transaction_details: disable "View in explorer" when the configured
explorer URL is empty/whitespace so we never open a garbage link.
- app: seed-backup "Skip" now requires a second, deliberate click with a
fund-loss warning before it dismisses.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Address a cluster of low-severity robustness nits where an async
callback could be skipped (leaving a spinner spinning forever) or a
synchronous action could be re-fired mid-flight:
- app_network: add catch(...) fallbacks around the importPrivateKey and
submitZSendMany worker lambdas so a non-std throw can't escape the
worker and skip the main-thread callback (stuck "Importing…"/"Sending…").
- export_transactions: re-entrancy guard disabling Export while a write
is in flight.
- bootstrap_download: disable Cancel once clicked so the request is
visibly acknowledged and can't be re-fired before the worker stops.
- network_tab: give clear "already in list" feedback on a duplicate
server (keep the inputs, clear only the stale invalid-URL error) and
disable/relabel Refresh while a probe is in flight.
- key_export: offer Retry from the error branch (falls back to Reveal)
and guard against an empty/malformed address before dispatching.
- receive_tab: only enter the "generating…" state when a dispatch is
actually possible (guards a stuck spinner when disconnected).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Address-transfer dialog closed itself on submit (s_open=false the same frame),
so its in-dialog result screen was dead code. Keep the dialog open on submit:
s_sending drives the button to a disabled "Sending…" state, and the async
callback's result (success txid / error) now shows in the result screen with
its own Close button.
- Settings RPC connection editor was inert AND showed compile-time defaults.
Populate Host/Port/Username/Password from the auto-detected daemon config
(rpc::Connection::autoDetectConfig) and make the fields read-only — the RPC
credentials come from the daemon's DRAGONX.conf, so these now accurately
DISPLAY the live connection instead of pretending to be an editor (which did
nothing). The existing "auto-detected" note now matches the behaviour.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The last four (more involved) robustness items from the audit:
- deleteBlockchainData: post the deleted-item count to the main thread (via an
atomic, since Notifications isn't thread-safe) and show a completion toast
("Blockchain data deleted (N items). Daemon restarting…") — previously the
result was only logged.
- Pool mining: pool start has a connect delay with no feedback; announce
"Starting pool miner — connecting…" on a successful start and "Pool miner
connected and hashing." once the poll confirms it (contained pool_starting_
flag; the shared mining-toggle state machine is untouched).
- Force Quit (shutdown screen): gate it on the status text having STALLED (a
genuine hang) rather than a bare 10s timer — with a 20s hard-ceiling backstop —
and show a state-aware caution naming the stuck step (force-quitting mid daemon
flush risks the chainstate).
- Benchmark: require a confirming second click that first builds candidates to
estimate the duration ("Benchmark takes ~Ns and interrupts mining"), instead
of interrupting mining immediately.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Send: re-validate against LIVE state at the "Confirm & Send" click (connected,
not syncing, amount > 0, total <= available). The confirm dialog persists
across frames, so a balance drop / sync restart / fee bump after Review could
otherwise broadcast a now-invalid transaction; it now shows a clear message
instead of sending.
- Console: a malformed JSON argument ({...}/[...] that fails to parse) was
silently sent to the daemon as a plain string. Now the command fails with a
clear "Argument N is not valid JSON" line instead of being silently mangled.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Send: block Review from a view-only source (no spending key) with a tooltip,
instead of failing at submit; label the memo counter as bytes and warn in
colour once the field hits its cap (Sapling memos are byte-limited).
- Shield: show "No shielded (z) address yet — create one on the Receive tab
first" when the destination combo is empty, instead of an empty dropdown.
- Updater dialogs (xmrig/daemon): suppress the window X during an active
download/verify/extract so closing can't orphan/block on the worker — the
in-dialog Cancel is the way to abort.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
More robustness fixes from the audit (the import-key pattern, lower severity):
- Trim pasted input before use/validation: network add-server URL/label,
lite-wallet key import (+ block empty), validate-address, address-book entry.
- Confirmations for destructive/irreversible actions:
- Address-book Delete now needs a confirming second click (no undo).
- Console `stop` needs a confirming second `stop` (it shuts down the node).
- Wizard "Skip" encryption needs a confirming second click (keys stored
unencrypted).
- Send amount: normalize to 8dp (satoshi precision) on both the DRGX and USD
inputs so digits past 8dp aren't silently dropped between the preview/review
and what's actually sent.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The app-wide "import-key pattern" fixes (missing guards / untrimmed input / no
re-entrancy protection), from the robustness audit:
- Import-key dialog: unify the type indicator and the RPC dispatch on one
classifier (classifyPrivateKey is now prefix-aware — an "SK..." shielded key no
longer misroutes to the transparent RPC), trim manual input (not just Paste),
and gate both the indicator and the Import button on a shared
isRecognizedPrivateKey() so unrecognized/empty input can't be submitted.
- Shield: guard the submit button on connected + !syncing (like Send), with a
disabled tooltip, instead of firing into a raw daemon error.
- Mining: disable the pool Mine button when the payout address is empty
("enter a payout address first"), and trim the pool URL/worker on persist.
- Console: track in-flight RPC commands (atomic counter + busy() override) so the
input is disabled while a command runs instead of piling up on the worker.
- Maintenance (rescan/repair/delete-chain/reinstall-daemon): re-entrancy guard —
bail with "already in progress" instead of launching a duplicate destructive op.
- Bootstrap dialog: re-attach to an in-flight download on reopen instead of
resetting state and orphaning the running worker.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
From the app-wide robustness audit (calibrated to the import-key dialog). These
are the safety-critical fixes; several could cost users funds:
- Backup/export false success (FUND LOSS): exportAllKeys pre-seeded a header, so
a keyless result (the usual case when the wallet is encrypted+locked) still
looked non-empty and backupWallet wrote a private-key-less file and reported
"Backup saved". Now exportAllKeys returns an exported count; backupWallet and
the export-all dialog refuse to write / report success on 0 keys, disclose the
count, flag partial results as INCOMPLETE, and the backup dialog confirms
before overwriting an existing file (+ trims the path).
- Bootstrap: fail CLOSED — refuse to install an unverified multi-GB archive when
no checksum is published (was `return true`), mirroring the xmrig/daemon updaters.
- Restore-from-seed: require a valid BIP39 word count (12/15/18/21/24) with a live
"should be 24 words — you have N" hint instead of accepting any non-empty text.
- Encryption passphrase: detect leading/trailing whitespace and BLOCK with a
warning (not a silent trim, which would change the passphrase and lock the user
out).
- Receive payment QR: emit the canonical `drgx:` scheme with a URL-encoded memo
via a new shared util::buildPaymentUri (the request-payment dialog now routes
through it too, so they can't diverge); the old "dragonx:" + raw memo was
unparseable by the wallet's own scanner.
Also clamps the receive "Recent Received" list to the 4 most recent (companion to
the recent-lists commit).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Each pool row showed "<hashrate> N%" with nothing saying the % is the pool fee.
Make it "<hashrate> N% fee" and widen the left auto-balance card (25%/180dp ->
30%/210dp min) so the added text fits.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Lower the zebra-stripe contrast (scanline-line-alpha 4 -> 2; the derived dark
band softens with it) so the alternating rows are subtler.
- Give the input's terminal bar a 1px outline along its own edges (theme-aware),
matching the output panel's glass rim, and a near-white surface on light skins.
- Add a blinking terminal caret at the end of the input when it isn't focused.
- Report a daemon started before the wallet ("no wallet-managed EmbeddedDaemon"
but RPC connected) as "Running" instead of the misleading "no daemon".
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The overview "Recent Transactions" and the send tab "Recent Sends" lists showed
every matching transaction in a scroll region. Cap them at the 4 most recent
(state.transactions is sorted newest-first, so a simple head-limit). The receive
"Recent Received" list gets the same clamp in the Tier-1 robustness commit
(same file as the QR fix).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Lower-severity polish from the per-skin review:
- Market portfolio ratio bar: was full-saturation Success/Warning at a208 over a
white a10 track — the green read as "neon" on dark skins and the track was
invisible on light skins. Make it theme-aware: muted fills (a165) on dark, a
dark track on light so the bar is defined on white/pastel surfaces.
- Light-skin muted text: bump --on-surface-disabled 0.52 -> 0.62 on light/
color-pop-light/dune/marble/iridescent so secondary text (addresses, category
labels) clears the pastel/gradient backgrounds (analog of the dark-skin bump).
Remaining review nits are the "white-overlay-assumes-dark" track/hover pattern
(a broader separate sweep) and per-skin aesthetic trade-offs; deferred.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Address the high-severity, systemic legibility issues found in the per-skin
screenshot review (verified across all 9 skins):
- Status pills (history): "Confirmed" text was 55%-alpha (dim); make it full
opacity, and give status/shield pills a visible tinted fill (a48) + 1px border
(a90) so they stop vanishing on dark-red / near-white / gradient skins.
- Console light theming: light skins now use a near-white terminal surface, and
the log text/JsonBrace colors switch to the dark defaults in light themes (the
schema colors are dark-terminal-authored — honor them only in dark themes) so
text isn't pale-on-white.
- Dark-skin muted text: bump --on-surface-medium 0.85 / --on-surface-disabled
0.58 on dragonx/dark/color-pop-dark/obsidian (obsidian addresses were ~1.1:1).
- Disabled Send button: the disabled fill was hardcoded white (invisible on
light skins) — use a dark overlay in light themes.
- Light-skin inputs: stronger frame fill + border color and a 1px frame border
for light themes so fields (and buttons) have defined boundaries on white/
pastel surfaces.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Extend the inputs layer with the two remaining form-field idioms, pixel-identical:
- material::LabeledInputRow(label, labelPos, ...) — horizontal label-beside-input
(Text + SameLine + SetNextItemWidth + InputText). Adopt at the settings RPC
host/port/user/pass rows.
- material::LabeledInputMultiline(label, id, buf, size, flags) — label above a
sized multiline input. Adopt at the address-book notes and request-payment
memo (drops the redundant SetNextItemWidth that InputTextMultiline's explicit
size already governs).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
UI-standardization audit: the labeled single-line form field (a plain
ImGui::Text label above a fixed-width InputText/InputTextWithHint) was
open-coded across dialogs. Add material::LabeledInput(label, id, buf, size,
width=-1, hint=nullptr, flags=0) mirroring that idiom exactly — pixel-identical,
but a single chokepoint so input styling (label typography, frame, spacing) can
later evolve in one place. Returns the input's edited bool.
Adopt at the clean vertical (label-above) single-line sites: address-book
add/edit (label + address), shield from-address, export-all-keys + export-
transactions filename, validate-address input, request-payment label + URI.
Left as-is: horizontal label-beside-input rows (settings RPC host/port/user/
pass use Text + SameLine), multiline (InputTextMultiline), numeric (InputDouble),
combos, and hint-only inputs — none match the vertical single-line idiom.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
UI-standardization audit: the receive address-type filter (All/Z/T) was three
hand-rolled gapped ImGui::Button pills with per-segment PushStyleColor. Replace
with the shared material::SegmentedControl (rounded track + Primary inset-pill
on the active cell + centered caption labels), the same control the market
portfolio editor uses. Keeps the same right-aligned footprint (toggleTotalW)
and drives s_addr_type_filter from the returned clicked index. Verified across
all 9 skins via the screenshot sweep.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow-up to c83348a: convert the three mining solo-mode header toggles
(idle-mining, thread-scaling, gpu-aware) from the hand-rolled "draw active
pill + glyph, then hit-test" idiom to material::IconButton using the restBg
(active-pill) path. Each keeps its exact behavior pixel-for-pixel: a circular
Primary-tinted pill when active (alpha 60 idle / 40 scale+gpu), a state-colored
glyph (Primary when on, muted when off), hand cursor + tooltip on hover, and
the idle button's cursor save/restore (savedCur, restored at block end) is
preserved. Completes the clean icon-button adoption; the raw-rect IsRectHovered
sites remain intentionally separate (different hit mechanism).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
UI-standardization audit: ~66 call sites still used raw ImGui::Button /
ImGui::SmallButton instead of the app's canonical TactileButton, so those
buttons missed the standard glass fill + rim + tactile overlay (and the
light-theme flat treatment). Convert 59 plain action-button sites across the
wizard, security dialogs, settings, market/console/explorer tabs, and the
address dialogs to material::TactileButton / TactileSmallButton (drop-in:
same signature/return, args preserved verbatim). Danger buttons keep their
PushStyleColor wrappers — Tactile renders through them then overlays.
Deliberately left raw (not plain buttons): InvisibleButton hit-targets, the
frameless transparent-bg collapsible-header toggles (RPC/Debug), the icon-only
pagination chevrons, and the receive All/Z/T segmented filter — a glass rim
would clash with those intentionally borderless/segmented looks.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
UI-standardization audit: three draw patterns were copy-pasted across tabs and
dialogs. Promote each to a material:: helper and adopt at every verbatim site.
Output is pixel-identical (same colors/geometry threaded through as params).
- material::DrawProgressBar / ProgressBar — replaces 3 byte-identical rounded
fill bars in the daemon/xmrig/bootstrap download dialogs.
- material::CopyableTextOverlay — the click-to-copy affordance (hit button +
hand cursor + tooltip + hover underline + clipboard) shared by the peers
best-block hash and the 3 mining stats (difficulty/block/address). Returns
"copied this frame" so callers keep their own toast (no ui-layer dependency).
- material::DrawPill / PillSize — rounded bg + optional border + inset text
badge, shared by the explorer status pill, the market Z/T chip, and the
network official/custom tag.
Divergent one-offs left in place (transactions status pills use schema rounding
around pre-positioned multi-line text; the mining filter-pill is a button bg).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
UI-standardization audit found two dead, contradictory "how to draw a button"
surfaces with zero call sites:
- src/ui/material/components/buttons.h (353 lines: TextButton/Outlined/Contained/
Button(spec)/IconButton/FAB) — included by 3 mining files but never used.
- ApplyTactile (draw_helpers.h) — a retrofit-tactile-onto-plain-button wrapper,
never called.
Remove the file, the 3 dead includes, and ApplyTactile. TactileButton/StyledButton
remain the canonical button helpers. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
From the theme sweep review, three overarching light-theme fixes:
- Buttons: bump ImGuiCol_Button light overlays 13/20/28% -> 16/23/30% so the
small/secondary buttons read as clearly as the main ones.
- Dividers + outlines: raise --divider (0.12-0.14 -> 0.20) and --outline
(-> 0.24) across all light skins so section/list dividers are visible on white.
- Ratio/balance bars: deepen the muted --success/--warning greens/ambers in the
light skins that were pale (light, marble, dune) so the shielded/transparent
bars + success text are vivid (color-pop-light + iridescent were already vivid).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
From the theme screenshot sweep: input fields, dropdowns and amount boxes were
nearly invisible on light themes (Send form especially) — ImGuiCol_FrameBg was
black@4% vs the dark theme's white@7%, so boxes blended into the white surface.
Bump the light-theme FrameBg / hovered / active to 9/14/18% so inputs read as
defined boxes. Affects all light skins.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Organize sweep output into per-tab subfolders: <config>/screenshots/<tab>/<skin>.png
(was one timestamped folder with idx_skin_tab.png names).
- Fixed folder (no timestamp): subsequent sweeps overwrite the existing PNGs in
place instead of creating a new directory each run.
- Add an "Open location" TactileButton next to "Run screenshot sweep" that opens
the screenshots folder via Platform::openFolder(app->screenshotDir()).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Correct the placement: instead of a new "Debug options" button + dialog (and
moving the Verbose logging checkbox), rename the existing collapsible "DEBUG
LOGGING" section to "DEBUG OPTIONS" and put the "Run screenshot sweep" button at
the top of it, above the dragonxd daemon debug= categories. Restore the Verbose
logging checkbox to the wallet row (unchanged). Drop the now-unused
debug_options / debug_options_title i18n keys + the debug_options_open flag.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a debug tool that cycles every skin across every build-enabled tab and saves a
PNG of each to a timestamped folder under the config dir (screenshots/sweep_<ts>/)
— for gathering visual context on theme-specific UI work.
- App state machine (app_network.cpp): startScreenshotSweep() builds the
skin+enabled-page lists, saves the current skin/page, and steps through them;
updateScreenshotSweep() (top of App::render) pins the page + settles ~4 frames
after each skin/page change (skin switches reload TOML + reset the acrylic
capture, so they need to settle); wantsScreenshotThisFrame()/screenshotSweepPath()
/onScreenshotCaptured() coordinate with main.cpp. Restores the original skin/page
when done; nothing persisted.
- Framebuffer capture (main.cpp): read the finished frame and encode PNG via the
bundled miniz (tdefl_write_image_to_png_file_in_memory_ex). GL reads FBO 0 (RGBA,
bottom-up -> flip); DX11 copies the backbuffer to a staging texture + maps it
(BGRA, top-down -> channel-swap). Alpha forced opaque.
- Settings UI: move the Verbose logging checkbox off the wallet row into a new
"Debug options" button that opens a dialog housing the verbose toggle + a "Run
screenshot sweep" button. New i18n keys.
Both platforms build; no-crash smoke verified.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per the request for a flatter look like the Manage-portfolio modal's plain
buttons, render TactileButton FLAT on light themes: DrawButtonGlassOverlay now
early-returns in light themes with just a subtle defining edge (--rim-light),
skipping the white glass sheen (--glass-fill) and the tactile 3D depth (bright top
edge + the hardcoded bottom shadow that couldn't be removed via TOML).
ImGui::Button still draws the fill/hover/active (the darkened ImGuiCol_Button
overlays), so the button reads as a flat, defined element. Dark themes keep the
raised glass/tactile look. Re-adds the IsLightTheme() luminance helper (early in
the header so the overlay can use it). Applies to all TactileButtons app-wide in
light themes, not just Settings.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The settings-page buttons (TactileButton = ImGui::Button + a glass overlay) looked
faint in light themes: --glass-fill was a 55-62% WHITE wash that paled the button
body, and --rim-light was a 6-10% (sometimes light-colored) edge that was invisible
on light surfaces, so buttons had no defined border. Across all light skins:
- --glass-fill white wash 0.55-0.62 -> 0.20 (button body no longer washed out;
panels use the acrylic blur on capable systems, so only low-spec panel fills are
affected).
- --rim-light -> each skin's on-surface tint @ 0.22 (a defined dark button edge
instead of an invisible/light rim).
- --hover-overlay alpha -> 0.10 (visible hover).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the light-theme modal backing (19e3138, now reverted) with a root-cause
fix: darken the light themes' UI elements for contrast.
- Buttons: the light-theme ImGuiCol_Button overlays were barely-there grey
(5/8/12% black); bump to 13/20/28% so buttons read as buttons on light
surfaces (one edit in ApplyColorThemeToImGui covers all light skins).
- Text: raise --on-surface-medium (60-72% -> 86%) and --on-surface-disabled
(38-40% -> 52%) across the light skins (light, color-pop-light, dune, marble,
iridescent) — the dominant secondary/label text was too faint. Base
--on-surface was already near-black, left as-is.
The active theme is TOML-skin-driven, so the code GetMaterialLightTheme() is not
the live source; edits are in the skin TOMLs + the ImGui-style mapping.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Floating (BlurFloat) modals draw no card, so in a light theme the light content
sits on a light frosted backdrop with almost no separation (washed out — the
Manage-portfolio modal being the clearest case). Add a faint elevated sheet
(translucent Surface) + a defining edge (OnSurface) behind floating content in
LIGHT themes only, gated on a new IsLightTheme() (background luminance) helper.
Dark themes are unchanged (they already read fine). Kept translucent to preserve
the airy floating feel; respects "no dark backdrop in light theme".
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The PIN/passphrase input uses default ImGui framing, which is subtle on the glass
card — when focused there was no clear indication the field is active. Draw an
accent (Primary) ring around the input while it's focused/being edited so it's
obvious the field is ready for typing.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Regression from the full-window lock overlay: the full-window ##LockInputBlocker
InvisibleButton (drawn first) shadowed the lock card's widgets (PIN/passphrase
input, Unlock button, mode toggle) that overlap it, so clicks never reached them
(dead) and the app's global "hand cursor on clickable hover" fired for the whole
screen. Mark the blocker SetNextItemAllowOverlap() so the later widgets take
hit-test priority — the exact use case in ImGui's docs (invisible button covering
an area where subsequent items are added). The portfolio avoids this by putting
its content in a child window; the lock screen draws direct, so it needs the flag.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Address two GUI-review points:
1. Light themes no longer get a dark modal backdrop: DrawFullWindowBlurBackdrop
now tints toward the theme app background (WithAlpha(Background(), ...)) for
both the opaque base and the frost, so a light theme gets a light frosted
backdrop and a dark theme a dark one. Fixes all blur overlays, not just the
lock screen.
2. The lock screen now blurs the WHOLE window (sidebar included), not just the
content area: renderLockScreen opens a full-viewport borderless overlay window
(##LockOverlay, same pattern as the modal overlays / portfolio, opened from
within ##ContentArea) with an input blocker, instead of drawing in the content
child. Removed the theme-independent dark scrim (contradicted point 1).
Auth logic untouched. Needs GUI re-verify of all auth paths + that the whole
screen (incl. sidebar) obscures on lock and unlock returns cleanly.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
GUI review (light theme) showed the blurred backdrop is light there, washing out
the (now glass) card. Add a theme-independent dark scrim over the blur so the lock
screen reads as a clearly-dimmed "locked" state and the card stands out in both
light and dark themes — standard lock-screen dimming. Alpha tunable.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
GUI review showed the lock card was a flat light SurfaceVariant panel that stood
out against the dark glass cards of the other modals. Draw it with the same
GlassPanelSpec (rounding 16, fill 35, border 50) that BeginOverlayDialog uses, so
it renders as the same opaque-dark glass card on the blur backdrop. Visual only.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The lock screen no longer reads screens.lock-screen.backdrop-alpha (replaced by
the live-blur backdrop in c7f89e5). Remove the now-dead schema key.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The security DIALOGS (encrypt/decrypt/change-passphrase/remove-encryption, PIN
set/change/remove) already adopted the blur backdrop via the Phase 2 flip (they
use BeginOverlayDialog). This finishes Phase 4 by giving the bespoke lock screen
the same live-blur backdrop over the wallet content — restyle only, auth logic
(PIN/passphrase input, attempt counting, lockout timer) untouched.
The lock screen previously defaulted to backdrop-alpha=0 (wallet fully visible
behind the card); the blur now obscures it — a privacy improvement as well as a
consistency one. Capture-once on lock-open + resize (frame-gap detected via the
shared BlurCaptureStateFor), MarkBlurOverlayDrawn drives the theme-effect
suppression + acrylic re-capture on unlock. The opaque base in
DrawFullWindowBlurBackdrop guarantees the wallet is obscured even if blur is
unavailable (low-spec / acrylic-off / capture failure) — no content leak. UV
mapping is screen-space, so the content-area sub-rect blurs the correct region.
Needs GUI verification of every auth path (PIN/passphrase unlock, lockout).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The two Phase 3 "bespoke overlays" aren't content modals: the send-error is an
in-tab glass banner (already consistent — left as-is), and the shutdown screen is
a teardown screen where a blur backdrop would be semantically wrong (should read
as "closing") + risky on the teardown path. Per that finding, apply minor
consistency polish to the shutdown screen only:
- Share the overlay-scrim tone via a new material::OverlayScrimColor(opacity),
used by both the dialog scrim and the shutdown scrim (was a slightly different
hardcoded 0.06/0.06/0.08 → now the 0.04/0.04/0.06 dialog base).
- "Shutting Down" title → theme Warning() amber (was hardcoded gold).
- Force-Quit confirm destructive button → theme Error() (alpha-varied) instead of
hardcoded reds.
No behavior change; the shutdown scrim stays opaque (no blur).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per the user's selection, convert the large multi-section dialogs to the BlurFloat
style (floating content on the blur, plain heading, no glass card), each KEEPING
its authored width so width-coupled content is untouched:
- Export all keys, Console RPC reference, Explorer block detail
- Updater dialogs: daemon update, xmrig update, bootstrap download (multi-state)
Settings and Address book intentionally kept as cards (the modal Settings *window*
is dead code — show_settings_ is never set; the live Settings surface is the nav
page, not a modal; Address book has a nested edit that reads better contained).
All other confirm/input dialogs stay card-on-blur ("card for small"). Updater
dialogs are multi-state — GUI-check each state (confirm/progress/done/failed).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Convert two clean detail-view dialogs to the BlurFloat style (floating content on
the blur, plain h6 heading, no glass card) — the "float for large" half of the
plan, applied conservatively: keep each dialog's authored card width so its
width-coupled content (absolute SameLine positions, GetWindowWidth math) is
unchanged, and rely on the footer Close + outside-click for dismissal (the plain
heading drops the title-bar close X, matching the portfolio modal). Auto-height
preserved. A representative pair to GUI-review before rolling the treatment out to
the other large dialogs (or reverting if the card-on-blur reads better).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Grep-verified dead, zero live callers (the live paths all go through
BeginOverlayDialog and already got the blur backdrop):
- App::renderAboutDialog (app.cpp/app.h) — dispatch uses ui::RenderAboutDialog.
- src/ui/windows/import_key_dialog.{h,cpp} (ImportKeyDialog class) — live path is
App::renderImportKeyDialog.
- src/ui/windows/backup_wallet_dialog.{h,cpp} (BackupWalletDialog class) — live
path is App::renderBackupDialog.
Drop the two dead files from CMakeLists. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Flip the positional BeginOverlayDialog overload's default to blurBackdrop=true, so
every app dialog (~43 call sites) now renders its card on the live-blur backdrop
instead of the opaque scrim — matching the Manage-Portfolio look. Cards stay
auto-height (small dialogs stay small = "card for small") and render opaque via
the existing sole-consumer fallback (a translucent glass card would thrash the
single-slot blur cache against the radius-64 backdrop). Nested dialogs (address
book edit) already pass a distinct idSuffix, so their blur windows/capture keys
don't collide. A dialog that wants the old scrim can pass a spec with
blurBackdrop=false. Big visual change — needs a GUI sweep; trivially reverted.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The adversarial review found the ModalRenderGuard lagged the old
PortfolioEditorActive() signal by one extra frame on CLOSE (a possible 1-frame
"opaque panels" flash). Latch the overlay's open flag (already available via
spec.p_open) — "drew this frame AND still open after Esc/Close/outside handling" —
instead of just "drew last frame", so the guard drops and the acrylic re-capture
fire on the same frame the overlay closes, matching the old inline behavior on
both edges.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Dogfood the Phase 0 framework: RenderPortfolioEditor now opens via
material::BeginOverlayDialog(BlurFloat) + EndOverlayDialog instead of its own
inline overlay scaffold, and uses the promoted material:: SegmentedControl /
RightAlignX / BeginFadeScrollChild helpers (overlay_scroll.h). Delete the
portfolio's inline backdrop/capture-once state machine + its PfEditState capture
statics + the four local helper copies. App::render's ModalRenderGuard is now
driven by the modal-agnostic AnyBlurOverlayActiveLastFrame() (timing-equivalent),
and LatchBlurOverlayActive() at frame end handles the acrylic re-capture on close.
Framework tweak: floating cards pop the card WindowPadding after BeginChild (so
nested children don't inherit it) and center button labels, keeping the net
style-var count at 2 so EndOverlayDialog is unchanged and GlassCard stays
byte-identical. RenderPortfolioEditor: 357 -> 282 lines. Needs GUI verification
that the portfolio modal is visually/behaviourally unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Refactor BeginOverlayDialog to run through an OverlayDialogSpec + OverlayStyle
(GlassCard vs BlurFloat) with orthogonal flags (blurBackdrop / floatingContent /
plainHeading, fixed vs auto card height). The classic look is now GlassCard with
all flags off; the existing positional overload forwards to it → all ~43 current
dialogs are byte-identical (verified line-by-line). The BlurFloat path — live-blur
backdrop with per-overlay capture-once (frame-gap arming), floating content, plain
h6 heading — is implemented but unused until Phase 1 dogfoods it on the portfolio.
Also replace the unusable open-refcount with a per-frame drawn flag + last-frame
latch that invalidates the acrylic capture on the overlay's close transition.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
First step toward unifying all app modals on the Manage-Portfolio visual style.
Add, as shared material:: helpers (additive — no caller yet, no visible change):
- blur-overlay open refcount + 1-frame latch (Push/Pop/AnyBlurOverlayActive[LastFrame],
LatchBlurOverlayActive) so App::render's effect guard can be modal-agnostic;
- capture-frame query (IsCapturingBlurBackdrop) for scroll-fade suppression;
- RightAlignX and SegmentedControl (promoted verbatim from market_tab);
- BeginFadeScrollChild/EndFadeScrollChild in a dedicated overlay_scroll.h (kept
out of the widely-included draw_helpers.h because effects::ScrollFadeShader's GL
headers clash with SDL's in some TUs).
Also fix IsCurrentWindowOverlayDialog to prefix-match "##OverlayScrim" so suffixed
overlay windows are recognized by the modal-aware hover check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add market_iv_1h/1d/1w/1m and market_updated ("· Updated %s") keys and use them
in the chart interval strip (previously hardcoded "1H"/"1D"/"1W"/"1M") and the
pair-selector attribution line (previously a literal "· Updated %s"). Closes the
audit's i18n gaps in the market tab.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the balance summary (fiat/DRGX/BTC + shielded ratio bar) and the
draggable/resizable custom-group card grid (drag/resize gestures, dot grid,
click-to-edit) out of RenderMarketTab into mktDrawPortfolio(MktCtx); the local
x-coord `cx` renamed `cx0` (ctx param is `cx`). Drop the now-unused
gap/dl/glassSpec/buf locals. RenderMarketTab is now ~150 lines (from 962) — a
thin orchestrator: precompute -> build MktCtx -> hero/chart/pair/portfolio.
Verbatim bodies. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the ~250-line chart block (interval strip + 24h stats + refresh, plotted
curve with grid/area-fill/hi-lo+time labels/hover crosshair+tooltip, empty
state) out of RenderMarketTab into mktDrawPriceChart(MktCtx), recomputing cheap
locals and reading the precomputed series via the ctx. Drop the now-unused
hs/pad locals. Verbatim body. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Introduce a MktCtx struct holding RenderMarketTab's precomputed geometry/series,
built once after the precompute, and extract the combined hero card (glass panel
spanning header + chart, price/ticker/period-badge/trade button) into
mktDrawPriceHero(MktCtx). Cheap locals (dl/fonts/dp/market) recomputed in the
helper; the hero's local x-coord `cx` renamed `cx0` (the ctx param is `cx`).
Verbatim body. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the exchange/pair chip grid (wrapping flat chips + click-to-switch +
attribution line) out of RenderMarketTab into mktDrawPairSelector(app, registry,
availWidth), recomputing its own schema/fonts/market. Drop the now-unused ovFont
local. Verbatim body. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Lift the ~90-line [&]-capturing drawCard closure out of RenderMarketTab's grid
code into a file-static mktDrawCard(MktCardCtx, min, max, entry, hov). The
captured locals (dl/state/market/mktDp/pfInset/pfCardPad/sub1/capFont) are
bundled into a MktCardCtx bound once per frame. Verbatim body. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pull the ~30-line row-major cell placement (honor stored placement else
auto-place) out of RenderMarketTab into a pure, deterministic helper taking
(entries, cols, minW, minH). Also bump the grid resize-readout buffer 16->32 to
silence a worst-case -Wformat-truncation the extraction surfaced. No behavior
change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pull the three detail-pane section blocks out of RenderPortfolioEditor into
pfDrawAppearanceSection / pfDrawPriceSection / pfDrawAddressSection (each
recomputes its own dp/fonts/state; the section dispatch is a 3-line switch).
Verbatim move — same widgets, IDs and drawing. RenderPortfolioEditor drops from
~760 to ~357 lines. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Convert the four [&]-capturing lambdas (persist / buildWorking / workingMatches /
commitIfNeeded) into file-static pfPersist / pfBuildWorking / pfWorkingMatches /
pfCommitIfNeeded that operate on the s_pfEdit struct + take Settings*. Removes
the per-frame closure setup and lets the upcoming section helpers reuse them.
No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Collapse the loose market_tab.cpp statics into MarketViewState (s_mkt),
PfEditState (s_pfEdit) and PfGridDrag (s_grid). Pure mechanical rename
(compile-verified, no behavior change) that shrinks the global surface and gives
the upcoming function decompositions a single state handle per concern instead
of ~35 free names.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the no-I/O, no-ImGui series helpers (resampleHistory, bucketBySeconds,
sparklineSeries, chartSeries) out of market_tab.cpp into an inline header under
data/ (mirroring data/portfolio.h) and cover them with a testMarketSeries()
group in test_phase4.cpp (resample block averaging + partial blocks, window
bucketing + guards, sparkline fallback, chart time-window filtering + live
timestamp synthesis). market_tab now calls data::sparklineSeries/chartSeries.
No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Collapse the two identical equal-width segmented controls (the detail section
switcher Appearance/Price/Addresses and the address type filter
All/Shielded/Transparent) into one pfSegmentedControl(dl, origin, w, h, labels,
count, selected, font, idBase, dp) that returns the clicked index. Same track /
active-pill / centered-label drawing and per-cell hit-test as before; the
variable-width chart-interval strip is left as-is. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Dedup the identical min/max→rect projection shared by pfDrawSparkline and
pfDrawSparklineFilled into pfProjectSeries, and the repeated right-align cursor
math (footer Close, detail Revert/Save, summary Manage button) into
pfRightAlignX. No behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Audit cleanup of market_tab.cpp:
- Remove the write-only statics s_history_initialized and s_last_refresh_time
(assigned in several places, never read) and their now-dead assignments.
- Align the portfolio grid card's BTC value to %.8f to match the editor preview
(pfBuildDisplay) and the summary rows; it was the lone %.6f, so the same
balance rendered with different precision in the preview vs the card.
- Fix the stale s_pf_sel comment (no "-2 = unsaved draft" state exists).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Merge the two address-section rows into one: the All/Shielded/Transparent
segmented type filter sits on the left and Select all / Clear / Funded are
right-aligned on the same line, both vertically centered. Select-all is deferred
to a flag applied after the filtered set is built, so it still targets the
current filter (and stays idempotent via PortfolioEntryAdd's dedup).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Complete the explicit-save model: leaving the Manage-portfolio modal via Esc or
an outside-click now discards the current group's uncommitted edits (matching
the discard-on-switch behavior), while the Close button still commits. Only the
Save and Close buttons persist; every other exit path is a cancel.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Switching to another group in the Manage-portfolio list (or adding/deleting a
group) now discards the current group's uncommitted working edits instead of
auto-saving them — only the Save button (and closing the modal) persists. This
matches the explicit-save model signalled by the Revert/Save buttons and the
unsaved-changes dot: navigating between drafts is ephemeral until saved.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Move the group-scoped Revert/Save buttons into the detail container (sized to
the Add-entry height) so they read as lower-hierarchy than the modal-level
Close, which now stands alone in the footer. The detail body is wrapped in a
child that reserves the bottom row.
- Add an amber "unsaved changes" dot on the selected group row while its working
copy differs from the stored entry; also gate the detail Revert/Save on that
dirty state.
- Addresses: narrow the All/Shielded/Transparent segmented control (360px) and
right-align Select all / Clear / Funded on one line; give the Select all /
Clear pills more text padding.
- Appearance: shrink the icon search field and right-align it on the heading
line, with a "Search icons…" hint.
- Price: move the sparkline toggle onto the 24h line; the interval radios stay
below, enabled only when the sparkline is shown.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Rework the Manage-portfolio detail pane's Appearance and Addresses sections:
- Icon picker: search field inline with the "Icon" heading (filters by name,
hides the "None" slot while searching); grid now always fits 12 icons per
row, scaling cell size to width and centering the block.
- Both the icon grid and the address list scroll via a shared helper
(pfBeginScrollChild/pfEndScrollChild): smooth scrolling + the Settings-tab
edge-fade shader + a thicker, rounded scrollbar inset from the container edge
(bordered outer frame wrapping an inner scrolling child). Smaller container
radius. Fade gates off in low-spec and while the acrylic backdrop is being
captured — latched per-frame so the last capture frame no longer clashes with
the render-state reset.
- Addresses: All/Shielded/Transparent segmented control moved to its own row,
capped at 580px; Select all / Clear pills moved left of the Funded checkbox
on one left-aligned row (Funded vertically centered against the pills).
- Custom colour "+" swatch inset a few px so its selection outline no longer
clips at the modal's right edge.
- Remove the divider above the footer Revert/Save/Close buttons.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Appearance: icon-picker cells doubled (56px, iconXL glyphs); color swatches back to a
single row that hides overflow but always keeps the custom "+" cell visible at the right.
- Price: basis radios are now horizontal, matching the sparkline-interval radios.
- Addresses: All/Shielded/Transparent segmented control shrinks to sit on one line with
the Funded toggle to its right; Select all / Clear are rounded (pill) buttons with
padding, right-aligned on their own row.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Price: the basis and sparkline-interval dropdowns become radio groups (all options
visible), now that the segmented layout gives the section full height.
- Addresses: the All/Shielded/Transparent chips become a rounded segmented control
matching the top switcher. "Select shown" renamed to "Select all"; Select all + Clear
moved onto the Funded row, right-aligned. Address list gets inner WindowPadding.
- Appearance: 17 more preset colors (25 total), and the swatches now wrap into a grid
(with the custom-color cell) instead of a single overflowing row.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Swap the collapsible Appearance/Price/Addresses dropdown sections for a single
segmented pill control (one section shown at a time) — no dropdown headers. The
active section now owns the full detail height, so the icon grid and address list
fill the available space instead of fixed heights. State: s_pf_section (0/1/2)
replaces the three accordion open-flags.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Raising the backdrop radius alone would thrash the shared radius-keyed blur cache (the
glass panels blur at ~30). Instead, while a full-window blur overlay is active, glass
panels use their opaque fallback (IsFullWindowBlurOverlayActive) — they're covered by the
backdrop anyway — so the backdrop is the SOLE applyBlur caller and can use a strong custom
radius (64) with the cache staying coherent and frozen. The flag is toggled alongside the
theme-effect suppression by a RAII guard in App::render().
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Switch the live-blur behind the modal from a per-frame capture (which flickered from
per-frame re-blur non-determinism) to capture-once-on-open, re-captured only on resize.
- Insert the acrylic live-capture callback into the overlay draw list ONLY for a few
frames on open and whenever the viewport size changes; other frames reuse the frozen
blurred snapshot (blurCacheValid_), so there's no per-frame capture/re-blur cost. State:
s_pf_was_open (reset on close), s_pf_capture_frames, s_pf_capture_w/h.
Two fixes from an adversarial review of the timing/pipeline interaction:
- Do NOT override the backdrop blurRadius to 40. applyBlur caches one blurred buffer
keyed on radius, shared with every glass panel (market cards + the modal preview card,
radius ~30). A different radius thrashed that cache every frame and made the backdrop
sample the wrong blur, defeating the frozen-blur guarantee. Using the theme card's
radius keeps one blur serving all surfaces, frozen.
- Skip the blur on the overlay's first frame (allowBlur=false) — the live content isn't
captured until that frame's render, so blurring then would show stale/background pixels
for one frame; draw the plain opaque scrim instead.
Compiles on OpenGL (Linux) + DX11 (Windows). Needs on-device visual verification.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Fix the ImGui error on hovering a group row: the per-row delete was an
overlapping InvisibleButton + SetCursorScreenPos over the row Selectable. Replace
it with a manual IsMouseHoveringRect / IsMouseClicked hit-test (no overlapping
ImGui item, no cursor manipulation).
- Delete icon is larger (iconMed) and inset from the row edge (spacingMd).
- Footer no longer clipped: reserve 56px for the 40px buttons + separator.
- Manage-portfolio heading is now plain text (h6) with no title-bar background or
divider — content floats on the backdrop.
- Sidebar theme-effect borders no longer bleed through the overlay: suppress panel
theme-effects frame-wide (RAII guard in App::render via ui::PortfolioEditorActive)
while the modal is open, instead of only during the market render.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The acrylic system only blurred a frozen snapshot of the static app background. Add
an overlay-time re-capture so the Manage-portfolio modal blurs the LIVE tab content
behind it.
- AcrylicMaterial::captureLiveFramebuffer() (GL blit of FB 0 / DX11 CopyResource of the
backbuffer + no-backend stub) mirrors captureBackgroundDirect but bypasses the
dirtyFrames_ gate so it refreshes every frame the overlay is open, forcing a re-blur.
- ImGuiAcrylic::GetLiveCaptureCallback() returns a draw callback for it.
- market_tab: the modal inserts that callback (+ ImDrawCallback_ResetRenderState) at the
START of its overlay draw list, before the backdrop, so the capture holds the app UI
drawn below and the backdrop blurs it. On close, InvalidateCapture() re-captures the
background so other glass panels don't keep blurring the stale live capture.
- Panel theme-effects are suppressed during the market render while the modal is open
(their foreground-draw-list borders draw above all windows and would otherwise bleed
over the overlay). Market tab body is rendered again (no longer skipped).
- Backdrop draws the acrylic at full strength (fallbackColor.w=1) over an opaque base so
the blurred content reads clearly; lighter dim.
Design mapped via an Explore investigation of the capture/blur pipeline; compiles on
both OpenGL (Linux) and DX11 (Windows cross-build). Needs on-device visual verification.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- The per-row delete icon's SetCursorScreenPos left the ImGui cursor at the icon,
so the next group row started too high and rows overlapped. Capture the row's
end cursor after the Selectable and restore it after drawing the delete icon.
- Force centered button-label alignment for the modal (ButtonTextAlign 0.5,0.5)
so the footer/Add buttons read consistently.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Delete moved from the footer to a per-row trash icon on each group (shown on
hover/selected), with deferred deletion + selection fix-up.
- Removed the title-bar X; the footer is now right-aligned Revert / Save / Close
(larger buttons), Close commits + dismisses. Esc and outside-click still close.
- Add entry now immediately creates a persisted "Untitled" group and selects it
(no more invisible draft state); the -2 draft path is gone.
- Backdrop bleed-through fixed: when the modal is open the Market tab body is not
rendered, so its content and foreground theme-effect borders no longer show over
the overlay — the modal floats on the app's blurred/opaque backdrop.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Add button can no longer be cut off: it's now a normal widget below the group
list (in a group with the scroll child) instead of nested inside the list child.
- Stop the modal's 28x20 WindowPadding from cascading into the nested master/detail/
list children (pop it right after the card child begins) — that double-padding
was squeezing the master column and the Add button.
- Right pane navigability: collapsed accordion sections now show a compact summary
on the right (Appearance -> icon, Price -> basis, Addresses -> N selected), so a
group's state is readable without expanding. Section titles are consistent
title-case ("Appearance"/"Price"/"Addresses").
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Remove the centered glass card behind the Manage-portfolio content so it floats
directly on the blurred/opaque full-window backdrop. The little live-preview card
keeps its panel (it previews the actual group card).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Drop ImGuiWindowFlags_NoBringToFrontOnFocus on the overlay window — it stopped
the per-frame SetNextWindowFocus() from raising the modal, so it rendered behind
the market chart/cards. Now matches the proven BeginOverlayDialog behavior.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Combine the old list/edit modes into one master-detail view rendered as a
full-window overlay instead of a floating dialog.
- Backdrop: new material::DrawFullWindowBlurBackdrop draws an opaque base first
(so the sheet is opaque regardless of the UI-opacity slider — the acrylic path
multiplies its alpha by GetUIOpacity), then a strong full-window blur of the
backdrop, gated exactly like DrawGlassPanel so low-performance mode skips the
blur entirely (just the opaque scrim).
- Frame: bespoke full-window overlay (not BeginOverlayDialog) that mirrors the
scrim plumbing and preserves both overlay popup gotchas verbatim (no focus-steal
and no outside-click dismiss while a combo / color picker is open). Title-bar X,
Esc, and outside-click all dismiss.
- Left (master): selectable group list with accent bar + icon + name + value line,
and a pinned "Add group" button.
- Right (detail): live preview + label header, then a single-open accordion
(Appearance = color + outline opacity + icon; Price; Addresses) so only one
control group shows at a time — far less cluttered than the old two-column form.
- Footer: Delete / Revert / Save. Editing auto-commits named groups on navigation
and on close (preserving grid geometry); unnamed drafts are dropped.
- s_pf_editing tri-state replaced by s_pf_sel (+ accordion open-state); the Manage
button and grid-card click open with a valid selection.
Design + adversarial review done via multi-agent workflows (understand/design, then
a 3-lens ImGui-lifecycle / state-safety / backdrop review; zero confirmed defects).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Drop the extra gap the hero header reserved below the price row, so the interval
buttons sit closer to the top of the chart area.
- Place the divider just below the buttons with a comfortable gap (spacingMd)
between it and the plot, instead of centering it in the strip gap.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Default the price chart to the 1M range on open (was 1D).
- Move the hero divider from below the price row to below the interval buttons
(centered in the gap between the buttons and the plot).
- Remove the dot markers on the chart: no per-point dots, no current-price dot,
and no high/low dot anchors — just the clean line + area fill (the high/low
price labels and the hover crosshair remain).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Portfolio group cards:
- Always show the secondary DRGX amount, including on 2-cell cards.
- Larger group icon + label (subtitle1); value stays subtitle1 for consistency.
- Minimum group width is now 8 cells (default width 8).
Price chart:
- Area fill is now a flat translucent fill matching the group-card sparklines
(dropped the vertical gradient + its primitive-API helper).
- Interval buttons sit near the top of the chart area (removed the empty gap
above them) with a comfortable gap below before the plot.
- The change badge's suffix now matches the SELECTED range (1H->1h, 1D->24h,
1W->7d, 1M->30d) instead of the raw data span (which read 23h / 6d).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Chart interval buttons are now time RANGES, not bucket sizes: 1H = last hour,
1D = last 24h (5-min intraday), 1W = last 7 days, 1M = last 30 days (daily).
pfChartSeries filters the timestamped series to the window; X-axis "ago" labels
and the period-change span now derive from the real timestamps.
- Portfolio group cards redesigned: name (+ secondary DRGX) on the left; the value
with its % change stacked in the top-right corner; the sparkline fills the card
body. The sparkline now renders on 2-cell cards too (down to ~8px of body), with
tighter padding on short cards.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Builds on the timestamped historical series (pfChartSeries returns [time,price]):
- Gradient area fill: replace the flat translucent fill with a true vertical
gradient (solid at the line, fading to transparent) via the draw-list primitive
API (pfFillGradientArea).
- Period-aware change: the price badge now reports the change over the DISPLAYED
chart period with a span label derived from real timestamps (e.g. "+42% 1.0y"),
falling back to 24h for Live. Line/fill/marker colors follow that direction.
- Timestamped hover: the crosshair tooltip shows the real date/time at the hovered
point (time for Live/1H, date for day+ intervals) alongside the price.
- High/low markers: hollow dot + price label at the range's extremes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Portfolio group minimum height is now 2 cells (was 3). Added a 2-cell card tier
that also drops the sparkline (not just the footer) so a ~58px card keeps the
header + hero value legible.
- Price chart: stop drawing a dot on every point — for a dense historical series
(>40 points) just the clean line + fill, and always mark the latest point
(current price) with a dot + ring.
- Y-axis labels use the shared adaptive price formatter (e.g. $0.0143) instead of
a fixed 6 decimals, so they're cleaner and narrower.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Chart:
- The price chart now plots the historical CoinGecko series with a Live/1H/1D/1W/1M
interval selector (top-left of the chart). Live = in-session buffer; 1H uses the
intraday 5-min series, 1D/1W/1M the daily series (via pfSparklineSeries). X-axis
labels are interval-aware ("~5h"/"~3d"/"~2w"/"~4mo"). Market-tab open kicks the
self-throttled market_chart fetch so history populates promptly.
- 24H volume + market cap moved from their own row onto the chart's top strip
(right side, next to the refresh button), reclaiming the empty band above the
chart; the hero header now holds just the price row + separator.
- More bottom padding under the plot so the time labels and the DRGX/USDT pair
buttons aren't crowded against the card edges.
Portfolio groups:
- Accent-outline opacity is now configurable per group (0-100%, default 25%) via a
slider in the editor's Appearance column; applied to the card + preview outline.
- Resize grip enlarged (14->20px) and a live "W x H" size readout is drawn while
resizing, so shrinking a group to the 4x3 minimum is reliable and visible.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Portfolio groups now default to the 3-cell minimum height (was 4). The card
design adapts: below ~112px it drops the redundant DRGX footer and uses the
compact hero font so a 3-cell card stays legible.
- Group accent color is now a colored outline around the whole card (and the
editor's live preview) instead of a stripe on the left edge; the text no longer
reserves the old left-stripe offset.
- Price chart: trim the empty band between the 24H VOLUME stats and the chart
(smaller hero-header bottom pad), and keep the Y-axis price labels inside the
card on narrow windows (clamp to the card's left edge + bump min y-axis padding
40->54) so they no longer spill onto the tab background.
- Market tab no longer draws a scrollbar (NoScrollbar); the portfolio grid still
scrolls via the mouse wheel.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
While rearranging, the dot grid only reached two rows below the lowest card, so
the empty space beneath the cards showed no dots even though a card can be
dropped there. Fill the grid down to the bottom of the visible scroll viewport
(dl clip rect), keeping the two-row buffer as a minimum. The drop-target row is
already unclamped, so dropping into that lower area places the card and grows the
scrollable content.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Portfolio-group sparklines previously resampled the ~24-minute in-session price
buffer, so the hour/day/week/month intervals could never fill. Fetch real
historical USD series from CoinGecko /market_chart and back each interval with
appropriately-grained data.
- MarketInfo gains two timestamped series: price_chart_intraday (days=1, 5-min)
and price_chart_daily (days=365, daily), plus a fetch timestamp.
- App::refreshMarketChart() fetches both on the RPC worker via the TLS-verifying
util::httpGetString helper, self-throttled to ~30 min (historical data moves
slowly). Gated by getFetchPrices(); triggered on connect and each price tick.
- Pure NetworkRefreshService::parseCoinGeckoMarketChart() parses {"prices":
[[ms,price],...]} into (unix-seconds, price); malformed rows skipped. Unit-tested.
- market_tab pfSparklineSeries() maps interval -> series+bucket: minute = live
buffer; hour = intraday bucketed to 1h; day/week/month = daily bucketed to
1d/7d/30d. Falls back to the in-session buffer until the fetch populates.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the fill-to-height portfolio dashboard with a fine gapless grid of
~32px square cells whose column count scales to fill the tab width. The grid
extends past the window and the Market tab scrolls when cards are placed lower.
- Cells are a fixed square edge (>= ~32px), gapless; cards inset by 3dp so
neighbours don't touch. Column count = availWidth / 32dp (min pfMinW cols).
- Card minimum span is pfMinW x pfMinH (~128x96); layout/resize clamps to it.
New groups default to 6x4 cells.
- Content height reserves two extra rows below the lowest card so a card can be
dragged past the current bottom; the scroll child grows to match.
- Dot grid (shown while rearranging) softened and extended into the buffer rows.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Recommendation #1 + fill-to-fit:
- Group cards are now stat tiles instead of top-heavy rows: a header (icon +
name, with a 24h badge right), the value as a large hero centered in the
middle band, a footer DRGX amount, and a dedicated sparkline strip along the
bottom (soft fill under the line). Uses the taller square cells properly.
- The grid now fills the available height too: cell height is computed at render
time so the rows reach the bottom of the tab (no dead space), and the dot grid
extends to the bottom edge while rearranging. Cell height = availableHeight /
rows, so adding a row shrinks cells to keep filling. The hero font drops from
h4 to subtitle1 on short cells so it always fits.
Full-node + Lite build clean; ctest 1/1; hygiene clean. Interactive/visual —
needs a screenshot + drag/resize check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Grid cells are now square (cellH = cellW); column count is responsive with a
2-column minimum, and a group's minimum size is 2x1 cells. New groups default
to 2x1; drag/resize/auto-place all enforce the 2-wide minimum.
- Fix: editing a group no longer resets its grid placement/size. The editor's
Save now starts from the existing entry and only overwrites the form fields,
so grid_col/row/w/h (and anything else off-form) are preserved.
- Add a per-group sparkline interval (Minute/Hour/Day/Week/Month) next to the
Sparkline toggle. The price history (~1 sample/min) is resampled by averaging
each interval's worth of minute-samples into one point.
PortfolioEntry gains sparklineInterval and defaults gridW=2. Persisted in
settings.json. Full-node + Lite build clean; ctest 1/1; hygiene clean.
Note: the app only accumulates in-session minute samples, so Hour/Day/Month
show points only once enough data has been collected; true long-range history
(instant day/month charts) would need a CoinGecko market_chart fetch, which can
be added separately.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Turn the portfolio group area into a dashboard grid:
- Responsive column count (1-4 by width); each 1x1 cell is one card.
- Drag a card to move it to another cell; drag its bottom-right corner to
resize (span multiple cells). Both snap to the grid.
- A subtle dot grid appears while rearranging (dragging or resizing), with a
highlighted drop-target / resize footprint.
- Click (no drag past a 4px threshold) still opens the group editor.
- Grid placement (col/row) and span (w/h) persist per group in settings.json;
positions are resolved each frame — stored placement when it fits, else
auto-placed row-major into free cells. Dropping onto an occupied cell swaps.
PortfolioEntry gains grid_col/grid_row/grid_w/grid_h. The card render was
refactored into a drawCard() lambda reused for the in-grid, dragged, and
resizing states.
Full-node + Lite build clean; ctest 1/1; hygiene clean. This is an interactive
feature I can't verify without the GUI — needs drag/resize testing.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a per-group "Sparkline" toggle (live-market bases only, like 24h). When on,
the group card draws a faint price-trend line in its lower band — the group's
value tracks the DRGX price, so it plots the market price_history (normalized),
colored green/red by overall direction. Behind the text so amounts stay
readable. The editor's live preview reflects it too.
Persisted as show_sparkline in settings.json; wired through PortfolioEntry +
the editor working state + the pfDrawSparkline() helper shared by card+preview.
Full-node + Lite build clean; ctest 1/1; hygiene clean. Rendering change —
needs a screenshot check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
BeginOverlayDialog called ImGui::SetNextWindowFocus() every frame to keep the
dialog on top. That forced the scrim back to the front each frame and closed
any popup opened from inside the dialog — so the portfolio editor's price-basis
Combo and the custom color-picker popup flashed open then vanished, and clicking
a dropdown item that overhung the card could trip the outside-click-dismiss and
close the whole dialog.
Make the overlay popup-aware: compute overlayPopupOpen (IsPopupOpen with
AnyPopupId|AnyPopupLevel) and (1) skip SetNextWindowFocus while a popup is
showing so it stays focused/on-top, and (2) skip the outside-click-dismiss while
a popup is open so clicks inside it don't close the dialog. No change when no
popup is open, so existing dialogs are unaffected.
Full-node + Lite build clean; ctest 1/1; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Each portfolio group can now configure how it is valued and which price
fields its card shows:
- Price basis: Market · USD, Market · BTC, DRGX only, or Manual (a custom
price-per-DRGX + currency label). Manual mode values the group independently
of the live DRGX market price.
- Field toggles: show DRGX amount / show value / show 24h change (24h enabled
only for the live-market bases; value disabled for DRGX-only).
- The editor's left "Appearance" column gains a Price section (basis combo,
manual price + currency inputs, field checkboxes); the live preview and the
group cards both render through one pfBuildDisplay() helper so they stay in
sync. Defaults preserve the prior behavior (DRGX + USD).
Persisted in settings.json (price_basis / manual_price / manual_currency /
show_drgx / show_value / show_24h).
Note: "price source per exchange/pair" is limited to USD/BTC/Manual — the app
only fetches the aggregate DRGX price (USD + BTC); ExchangePair carries no
price, so valuing against an arbitrary pair's live price would need a new
market fetch (can be added separately).
Full-node + Lite build clean; ctest 1/1; hygiene clean. Rendering change —
needs a screenshot check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Reorder the Appearance column so Color sits above the icon grid, and let the
icon grid fill the remaining height of the left column (via
GetContentRegionAvail().y). Fixes the large empty space that sat below the
color row and gives the icon picker much more room.
- Add a custom-color circle after the preset swatches: it shows the current
color when it isn't a preset (else a "+"), and clicking it opens a
ColorPicker3 popup. The picked RGB packs straight into the group's color.
New i18n key portfolio_custom_color. Full-node + Lite build clean; ctest 1/1;
hygiene clean. Rendering change — needs a screenshot check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Full redesign of the group edit view (list mode unchanged):
- Two columns in a wider (780px) dialog:
* LEFT "Appearance": a live preview of the group card (icon + color + name +
summed DRGX/USD), then Label, the icon picker, and the color picker.
* RIGHT "Addresses": a search box, Shielded/Transparent/All type filters, a
"Funded" toggle, Select-shown / Clear, and the address list.
- Reworked address rows: a custom checkbox, a Z/T type chip, the address's own
label + icon (from the address book), a middle-truncated address, and the
balance right-aligned. Selected rows sort to the top (then by balance); the
whole row toggles selection. Replaces the old wall of raw address strings.
- Search filters by address or label; empty-state hint when nothing matches.
New i18n keys (portfolio_appearance/addresses_hdr/search/funded/select_all/
select_shown/group_name/no_addr_match). Full-node + Lite build clean; ctest 1/1;
hygiene clean. Rendering change — needs a screenshot check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Rename the portfolio heading "MY DRGX" -> "PORTFOLIO".
- Group cards are now clickable: clicking one opens the editor directly on that
group (edit mode), with a hover highlight + hand cursor.
- PortfolioEntry gains an `icon` (project_icons wallet-icon name) and a `color`
(packed IM_COL32 accent); both are persisted in settings.json.
- Editor edit-mode gains an icon picker (the same wallet icon set as the overview
tab's address-label dialog, via material::project_icons) with a "None" option,
and a color picker (preset accent palette + a default/no-color swatch).
- Group cards render the chosen icon to the left of the name and, when a color
is set, a left accent bar + the icon tinted in that color.
Full-node + Lite build clean; ctest 1/1; hygiene clean. Rendering change — needs
a screenshot check of the Market portfolio + the editor.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The hero-card header height was derived from an inflated schema value
(hero-card-height * vs), reserving far more vertical space than the price row +
separator + stats row actually use. Since the chart is drawn at the bottom of
that header, it started well below the "24H VOLUME"/"Market Cap" row, leaving an
empty band. Size heroHeaderH to the actual content so the chart begins right
after the stats.
Full-node build clean; ctest 1/1; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow-up to the Market tab rework, per feedback:
- Chart is more compact: a modest responsive height (~22% of the tab, floored
at 110px) instead of stretching to fill.
- Chart now has a time axis. The price series is one sample per ~60s market
refresh (RefreshScheduler::kPrice), so the x-axis is labelled in approximate
minutes-ago ("~45m … ~15m … now") instead of only "Now" — you can now read
the interval/span of each plot point.
- Portfolio: removed the top-level card background — the summary (fiat hero +
24h change + BTC, DRGX balance + Z/T, ratio bar) now floats directly under the
MY DRGX header, and the "GROUPS" heading is gone.
- Group cards redesigned: a max width (280px, shrinks on narrow tabs) so they
float rather than stretch; the group name is larger (subtitle font) top-left;
the DRGX + fiat amounts moved to the card's top-right corner; and more inner
padding around the content.
Removed the now-unused portfolio_groups i18n key. Full-node + Lite build clean;
ctest 1/1; hygiene clean. Rendering change — needs a screenshot check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Market tab is non-scrolling, but an oversized fixed-height chart ate ~65%
of the height and squeezed the portfolio into a cramped strip, with custom
groups rendered as tiny afterthought rows at the very bottom.
- Precompute section geometry (hero header + portfolio card height) up front and
let the price chart ABSORB the remaining vertical space, capped at its schema
height and floored at 130px*vs so it stays useful. The chart now shrinks to
make room instead of dominating. Replaces the barely-used SectionBudget.
- Portfolio summary refined: fiat value stays the hero with the 24h change beside
it and the BTC value right-aligned; DRGX balance + Z/T breakdown on the next
row; a full-width shielded/transparent ratio bar + % label.
- Custom groups now render as a wrapping grid of mini glass cards (label + DRGX +
fiat each), sized to the card width, instead of a cramped one-line list. A
"GROUPS" subheader separates them from the all-funds summary. The card grows to
fit the grid.
New i18n key portfolio_groups ("GROUPS"). Full-node + Lite build clean; ctest
1/1; hygiene clean. Rendering change — needs a screenshot check of the Market tab.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The "Manage portfolio" editor renders its address checklist inside the
overlay dialog's content child, which is ImGuiChildFlags_AutoResizeY (the
dialog sizes to its content). The checklist used a negative "fill remaining
space" height (-GetFrameHeightWithSpacing()*1.5f), which is meaningless in an
auto-resizing parent — there is no fixed height to subtract from — so ImGui
collapsed the child to ~0 and no addresses were visible, even with
state.addresses fully populated.
Size the list to its rows instead, capped at 40% of the viewport height so a
large wallet scrolls rather than running off-screen. Pre-existing bug from
when the editor was added; distinct from the earlier state.addresses
population fix.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two render-loop de-duplications, zero behavior change (verified byte-equivalent
on both the GL and DX11 backends via the mingw Windows cross-build):
- drawWindowBackdrop(app, bgDL, p0, p1, backdrop_active, lowSpec): the ~80-line
gradient/texture/acrylic-capture/WindowBg-alpha block was copy-pasted in the
main loop (with low-spec guards) and the resize-watcher lambda (without them).
Extracted once; the main loop calls it with lowSpec=isLowSpecMode() and the
resize watcher with lowSpec=false — each reproducing its CURRENT behavior
exactly (pure dedup, no change). The intentional differences around it
(resize's immediate present(0) vs the main loop's vsync, the main loop's
try/catch, the BeginFrame low-spec guard) are left untouched.
- handleDisplayScaleChange(window, newScale, savedSizeForScale, lastKnownW/H,
dpiTargetW/H, dpiResizeRetries, dpiResizePending): the SDL display-scale-change
handler was duplicated ~verbatim in the idle-wait and poll event paths
(differing only in debug-log text). Moved the body into one helper called from
both; the #ifdef DRAGONX_USE_DX11 g_borderlessDpi and #ifdef __APPLE__ scale
normalization are preserved.
Deliberately did NOT unify the full frame/present sequence — the resize path's
immediate present and lack of try/catch are intentional.
Full-node (GL) + Lite build clean; Windows/DX11 cross-build compiles main.cpp
clean; ctest 1/1; hygiene clean. NEEDS RUNTIME VERIFICATION: a live window
drag-resize (backdrop draws, no flicker) and a monitor DPI-scale change (window
resizes + fonts reload).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Settings::getDefaultPath and AddressBook::getDefaultPath each hand-rolled the
same per-platform #ifdef (Win SHGetFolderPath / macOS Application Support /
XDG ~/.config) with the DRAGONX_APP_NAME variant suffix. Route both through
util::Platform::getConfigDir() instead — one place owns the split — and add
the missing __APPLE__ branch to getConfigDir so it too returns the
macOS-native ~/Library/Application Support/<AppName>.
This also fixes a latent macOS inconsistency: transaction_history_cache
already resolves via getConfigDir(), which lacked the __APPLE__ branch, so on
macOS the tx cache landed in ~/.config while settings/addressbook went to
~/Library/Application Support. Now all three agree. No migration needed:
Linux/Windows paths are unchanged, and macOS packaging is not yet shipped.
Removes ~50 lines of duplicated #ifdef.
Full-node + Lite build clean; ctest 1/1; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Three UI de-duplications (rendering changes; the test suite skips the GUI, so
these need a manual screenshot check of the Balance/Peers/Transactions tabs).
1. Classic balance layout — the DEFAULT layout — hand-rolled its own ~550-line
address-list + recent-tx renderers that had DIVERGED from the shared
RenderSharedAddressList/RenderSharedRecentTx used by the other 9 layouts.
Deleted the inline copies and called the shared renderers (as the minimal
layout already does), so the default layout now gains set-label +
add-to-portfolio context items, custom address icons, drag-to-reorder,
keyboard nav, copy-flash, and fully TR()'d strings. Hero row and Classic
sizing (tabs.balance.classic address-table-height=340) preserved; addrH is
computed before the call exactly as before. This is a deliberate
feature-parity behavior change for the default layout.
2. peers_tab info cards — extracted drawStatCell() (label + value/em-dash with
the shared offset math) and one drawCardDivider() replacing two duplicate
divider lambdas; plain cells drive off a per-card loop. Bespoke cells
(Blocks "(X left)", copy-on-click Best Block, TLS check-icon) kept inline.
3. transactions_tab — one drawSummaryCard() replaces the three near-identical
Received/Sent/Mined blocks (same glass panel, icon, hover outline,
click-to-filter). Byte-equivalent.
Full-node + Lite build clean (net -651 lines, no new warnings); ctest 1/1;
hygiene clean. NEEDS SCREENSHOT VERIFICATION.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Extract App::restartDaemonAfterEncryption(taskName, announceRestartStatus)
from the two verbatim copies in encryptWalletWithPassphrase and
processDeferredEncryption. The two differed only in the async task name and
whether they set connection_status_ = "restarting_after_encryption" first
(the immediate encrypt path does; the deferred path does not) — both
parameterized. The 20x100ms settle loop, cancel/shutdown guards, and
stop/startEmbeddedDaemon sequence are preserved exactly.
Deliberately NOT done: the audit's "decrypt 5-deep callback pyramid" finding.
On inspection the nesting is load-bearing, not gratuitous — WalletSecurityWorkflow
is unsynchronized main-thread-only state, and each step's worker->MainCb bounce
is what keeps its mutations on the main thread (the UI reads snapshot()/importActive()
every frame); collapsing the pipeline onto one worker/async task would race the
UI, and the final restart+import stage requires an AsyncTaskManager::Token that
worker_->post can't supply. Left byte-for-byte intact.
Full-node + Lite build clean; ctest 1/1 (incl. WalletSecurity controller tests);
hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Six behavior-preserving consolidations from the audit:
- RPCClient::call() overloads share a private performCall() (payload dump +
curl_easy_perform + http code) and a static parseRpcResult() (error->RpcError
extraction). The timeout overload restores the prior timeout in both the
success and catch paths, exactly as before.
- The six UnifiedCallback methods delegate to one splitUnified() that builds
the (Callback, ErrorCallback) pair once (null-check preserved).
- One liteTrimCopy() in lite_connection_service replaces 5 file-local
trim-copy helpers across the lite slice.
- lite_wallet_controller's inline height JSON parse now routes through the
tested parseLiteHeightResponse().
- theme_effects.cpp: unpackRGB()/scaledAlpha() replace 14 RGB-unpack + 5
alpha-scale copy-paste blocks.
- settings.cpp load() uses loadScalar()/loadClamped() helpers for ~44 scalar
fields. Besides removing boilerplate this HARDENS loading: ~44 fields that
previously only checked contains() now also verify the JSON type, so a
malformed value in settings.json falls back to the default instead of
throwing/misreading. Custom cases (enum parses, legacy migrations, arrays)
stay inline; save() is unchanged.
Full-node + Lite build clean; ctest 1/1; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Three behavior-preserving consolidations in src/app_network.cpp:
1. Fold the three near-identical pending-send balance-delta blocks
(upsertPendingSendTransaction debit, removePendingSendTransactions
restore, applyPendingSendBalanceDeltas debit) into one
App::applyPendingSendDelta(from, signedAmount, includeAggregates). The
restore path's unclamped `+amt` is provably identical to the clamped
signed form (balance/amount are >=0 invariants), so a single clamped
helper reproduces all three exactly.
2. Collapse createNewZAddress/createNewTAddress into one
App::createNewAddress(bool shielded, cb) with thin public forwarders,
preserving the lite early-return, the dual push into the type list AND
the combined state_.addresses view, and the dirty-flag/refresh bookkeeping.
3. Extract App::submitZSendMany() shared by sendTransaction and
resendWithFeeGapWorkaround — the only differences (the TraceScope label
and the retry-only send_feegap_retried_opids_.insert) become parameters.
The in-flight counter, opid tracking, upsertPendingSendTransaction, and
pending_send_callbacks_ bookkeeping are byte-equivalent.
Full-node + Lite build clean; ctest 1/1; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
XmrigUpdater and DaemonUpdater carried byte-identical libcurl code —
httpGet, downloadToFile, and the write/progress callbacks — including the
security-critical TLS-verify (SSL_VERIFYPEER=1 / SSL_VERIFYHOST=2) and
max-body-size options, duplicated across two signature-verified
download-and-execute files that had to be hand-kept in sync.
Hoist that machinery into src/util/http_download.{h,cpp}:
- util::httpGetString(url, logTag)
- util::httpDownloadToFile(url, dest, logTag, maxBytes, onProgress)
threading progress + cancellation through a std::function (returning false
aborts the transfer) instead of the class `this` pointer. Each updater's
httpGet/downloadToFile become one-line forwarders that bind their own log
tag and archive-size cap (64 MiB miner / 256 MiB node, both preserved).
Every curl option is preserved byte-identically (verified by diffing the
setopt lines against the pre-change code — the only deltas are the
now-parameterized MAXFILESIZE cap and the XFERINFODATA payload). No behavior
change; the TLS/size-guard options now live in exactly one place.
Deliberately scoped DOWN per the audit's own guidance: the divergent
per-updater logic (release DTOs/parsers, checksum-table parsing, version
compare, and installResolved's per-member-SHA vs transitive-trust verify,
install strategy, and chmod scope) stays in each class — NOT merged behind a
base-class/virtual-hook layer, which would raise review cost on
signature-verified execute-code and risk a subtle security regression. The
shared-DTO/parser consolidation (~77 test refs of churn on stable code) is
left as a separate, lower-value follow-up.
Full-node + Lite build clean; ctest 1/1; source-hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Mechanical, behavior-preserving consolidations from the audit. Rendering
extractions were kept byte-equivalent; sites that couldn't be made identical
were left as-is (noted below).
Shared helpers:
- util::truncateMiddle(s,maxLen) / (s,front,back) in text_format.h — replaces
6 file-local middle-ellipsis truncators + several inline substr sites
(send/receive/transactions/balance_recent_tx/explorer + app.cpp + 3 dialogs).
Carries the maxLen<=3 guard, fixing the latent unsigned-underflow copies.
- material::LoadingDots() in draw_helpers.h — one animated-ellipsis source for
8 copy-pasted spinner sites (identical GetTime()*3 phase preserved).
- Reuse the existing FormatHashrate() in explorer_tab + peers_tab (dropped two
inline hashrate ladders).
- material::DrawButtonGlassOverlay() — the glass-fill/rim/tactile overlay block
shared by TactileButton / TactileSmallButton / schema TactileButton.
- material::CollapsibleHeader() — the invisible-button + label + chevron idiom
(3 of 5 settings_page sites; RPC/Debug headers skipped — non-identical).
- material::GlassCardScope (RAII) — the ChannelsSplit/Indent/DrawGlassPanel/
ChannelsMerge card scaffold (5 settings_page cards; About/send/receive skipped
— different padding / logo interleaving).
- material::DialogWarningHeader()/DialogConfirmFooter() — warning header +
50/50 Cancel/danger footer across the confirm dialogs; button height moved
from a hardcoded 40px into ui.toml (components.overlay-dialog.confirm-btn-height).
- File-local enterLowSpec()/exitLowSpec() collapse the 3 low-spec snapshot copies.
i18n: wrapped hardcoded English in the shared balance render paths and the
whole Lite lifecycle/security section in TR(), with English defaults added to
loadBuiltinEnglish() (res/lang/*.json left for the translation tooling — TR
falls back to English, so output is unchanged).
Full-node + Lite build clean; ctest 1/1; hygiene clean. These are rendering
changes — the Settings page (cards/headers/confirm dialogs), all tactile
buttons, and the Lite settings section warrant a screenshot check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Three P1 structural refactors from the audit; no behavior change.
1. Extract the ~180-line daemon-stdout rescan/witness parser out of the
~790-line App::update() into a pure, static, unit-testable
NetworkRefreshService::parseDaemonRescanOutput() returning a
DaemonRescanScan result struct (sits next to the existing static parse*
siblings). The scanning half moved verbatim; App::update() keeps the
state-application half, now reading scan.* fields. Adds
testParseDaemonRescanOutput() covering all six daemon signals + edge
cases against fixture log snippets — previously untestable without a
live daemon.
2. Extract the ~95-line global keyboard-shortcut block (Ctrl+, / theme
cycle / F5 / low-spec / effects / gradient / wizard) out of
App::update() into App::handleGlobalShortcuts().
3. Collapse the ImGuiAcrylic frontend fork: the GLAD and DX11 namespaces
were ~375 lines of semantically-identical code (the frontend makes zero
direct GL/DX calls — all backend work delegates to AcrylicMaterial, which
has a backend per API). Extended the single frontend's guard to
#if defined(DRAGONX_HAS_GLAD) || defined(DRAGONX_USE_DX11) and deleted the
DX11 duplicate, leaving the no-backend stub. Kills the Windows/DX11 drift
hazard (acrylic frontend changes now made once).
Verified: full-node + Lite build clean; ctest 1/1 (incl. the new parser
tests); source-hygiene clean; and the collapsed acrylic path was
compile-verified under DX11 via the mingw-w64 Windows cross-build
(ObsidianDragon.exe links).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two dead-code removals surfaced by the audit; no runtime behavior change.
1. Delete the lite-lifecycle "readiness" scaffold (~1,586 lines). The pure
dry-run evaluateLiteWalletServerLifecycleReadiness (whose own status text
says "wallet lifecycle execution is still disabled in this scaffold") had
zero callers; executeLiteWalletServerSelectionUi had zero callers; and
executeLiteWalletLifecycleUiRequest's only caller was a fallback in
settings_page reached only when app->liteWallet() is null. Deleted
lite_wallet_server_lifecycle_readiness.{h,cpp} and the readiness /
UI-execution machinery + enums + translation tables from both adapters,
keeping the genuinely-live helpers (liteConnectionSettingsFromAppSettings,
applyLiteConnectionSettingsToAppSettings, redactLiteServerSelectionValue,
and the LiteWalletLifecycleUiExecutionInput DTO the live create/open/
restore path still uses). The null-backend fallback now sets a plain
"Lite wallet backend unavailable" status. This is exactly the
readiness/scaffold pattern CLAUDE.md forbids regrowing.
2. Split the HushChat fixture/capture-manifest/seed-projection tooling
(~2,050 lines, incl. the libsodium seed-projection) out of
chat_protocol.cpp (1854 -> 284) and chat_protocol.h (586 -> 105) into a
new chat_fixture_tooling.{h,cpp} compiled ONLY into the HushChatFixtureCheck
dev tool — never into the app, lite, or test binaries (verified via nm).
The app keeps only the runtime slice (memo parsing + tx metadata
extraction) that network_refresh_service actually calls. Also moved the
decrypt-preflight/hex helpers, which likewise had no runtime caller.
Note: the HushChat split is on gated-off experimental code and should be
coordinated with the pending dormant-HushChat-content handling (commit
af06b8b) before the lite PR — done here as a pure mechanical split, no redesign.
Full-node + Lite + HushChatFixtureCheck build clean; ctest 1/1; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Four correctness fixes surfaced by the codebase audit:
1. Lite address-book path bug: AddressBook::getDefaultPath() hardcoded
"ObsidianDragon" while Settings::getDefaultPath() uses DRAGONX_APP_NAME,
so the Lite build wrote addressbook.json into the full-node app's config
dir instead of ObsidianDragonLite/. Now uses DRAGONX_APP_NAME on all
platforms (no macOS path change — the getConfigDir consolidation, which
would move the macOS dir, is deferred).
2. PIN-unlock lockout bypass: the PIN path duplicated unlockWallet's
success/lockout logic and its RPC-error branch bumped the attempt counter
but skipped the escalating-lockout math entirely — a PIN user hitting an
RPC error escaped the lockout curve. Extracted App::applyUnlockSuccess()
and App::applyUnlockFailure() and routed all unlock paths (passphrase,
PIN vault-fail, PIN RPC-fail) through them, so the lockout curve now
applies uniformly. (noVault stays a mode-switch, not a failed attempt.)
3. Witness/rescan reset drift: the ~11-line rescan+witness completion reset
was copy-pasted at four sites (app.cpp x2, app_network.cpp x2); adding a
witness field and missing a copy would leave stale progress. Folded into
App::resetWitnessRescanProgress(). Left the distinct phase-transition
reset in App::update() untouched (it sets witness_phase, not 0).
4. Truncation underflow: send_tab/receive_tab's size_t TruncateAddress
copies computed (maxLen - 3) without guarding maxLen <= 3, which wraps
and throws std::out_of_range on short inputs. Added the guard.
Full-node + Lite build clean; ctest 1/1; source-hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Market-tab portfolio (editor checklist + SumPortfolioBalance) reads the
combined WalletState::addresses view, but the full-node refresh path only ever
updated the authoritative z_addresses/t_addresses lists — rebuildAddressList()
was never called, so `addresses` stayed empty. Result: the manage-portfolio
modal showed no addresses to pick, and addresses added via the overview
right-click menu contributed no value to their entry (SumPortfolioBalance
found nothing in the empty combined list).
Rebuild the combined view in NetworkRefreshService::applyAddressRefreshResult
(the sole bulk updater of the address lists), and push newly created addresses
into the combined view immediately in the full-node create paths for
zero-latency parity with the overview (matching the lite branch). This also
repairs two other silently-broken full-node consumers of state.addresses: the
auto-shield target-address finder and the pool-mining transparent fallback.
Add a regression test (testWalletStateAddressListRebuild) covering the
empty-before / union-after rebuild and pending-send delta reflection.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Swap the "Install bundled" and "Test connection" daemon-binary buttons:
Install bundled now sits in the left common row (Check for updates | Refresh
| Install bundled); Test connection leads the right-aligned maintenance
group. Each button keeps its own disabled-predicates, action, and tooltip.
Give the daemon-binary Refresh button a unique ImGui ID
("Refresh##daemonRefresh") so it no longer collides with the theme-section
Refresh buttons, which triggered Dear ImGui's "2 visible items with
conflicting ID!" error. The label still displays "Refresh".
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the Backup & Data button row (Import/Export key, Export all, Backup, Export
CSV, and the full-node Download Bootstrap / Setup Wizard) into the WALLET card as
a "Backup & Data" sub-section under the privacy toggles + Tools & Actions, and
delete the standalone BACKUP & DATA card (its own glass panel + header). One
fewer top-level card; all actions preserved.
Full-node build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Rename the RPC collapsible header from "RPC CONNECTION" to "RPC Connection..." and
render it in the body2 font so it matches the "Tools & Actions..." toggle exactly.
Full-node build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the static "RPC CONNECTION" overline header with a full-width clickable
toggle (label + expand/collapse arrow, transparent button with hover) that
reveals the Host/Port/Username/Password fields + auto-detected caption only when
expanded — same idiom as the "Tools & Actions…" section. Collapsed by default
(rpc_expanded). Field rendering/bindings unchanged.
Full-node build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a button to the left of "Open data folder" that opens the ObsidianDragon
config folder (util::Platform::getObsidianDragonDir() — ~/.config/ObsidianDragon
etc.: settings, themes, logs), distinct from the daemon data/blockchain folder.
The two buttons render as a right-aligned group on the data-dir row, each an
explicit button-font width so the group's right edge lands exactly on the card
edge (no overflow). New i18n: settings_open_app_dir / tt_open_app_dir.
Full-node build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Remove the "Advanced" toggle. The four maintenance actions (Install bundled /
Rescan / Delete blockchain / Repair wallet) now render right-aligned on the same
row as Check-for-updates / Refresh / Test connection (wrapping to a right-aligned
next row only when the window is too narrow). Predicates preserved.
- Fix the Open-data-folder button overflowing the card's right edge (worse at high
display scaling): its width was estimated with the CURRENT font, not the button
font, so the estimate was wrong. Measure with the button font and give the button
an explicit width so its right edge lands exactly on the card edge regardless of
DPI/font-scale timing.
- Drop the now-unused node_advanced_expanded flag.
Full-node build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The right-aligned Open-data-folder button captured its Y from
GetCursorScreenPos() after the wallet-size text, by which point ImGui had already
advanced to the next line — so the button dropped below the data-dir row. Capture
the row's top Y before rendering the row and pin the button to it, so Data Dir /
Wallet Size / Open data folder sit on one line.
Full-node build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- SECURITY: the auto-lock + PIN controls now continue on the same row as the
encrypt/change/lock/remove controls (full-node), so the whole security band
fills the card width instead of leaving the right half blank. Lite (no encrypt
row) still starts auto-lock/PIN on a fresh row at the left edge.
- Daemon Binary "Advanced" (Install bundled / Rescan / Delete blockchain / Repair
wallet) now defaults to expanded (node_advanced_expanded = true) so the actions
are visible without a click; the toggle can still collapse them.
Full-node + lite build clean; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the column/grid/masonry machinery (which kept leaving empty gaps and
buried the daemon Advanced actions) with a single full-width card whose controls
are arranged horizontally to use the width:
- NODE: Data Dir + Wallet Size + Open-folder on one row.
- RPC: Host / Port / Username / Password four-across when wide (>=700), wrapping
to 2x2 when narrow; auto-detected + plaintext warning kept.
- SECURITY: encryption buttons + Auto-lock combo + PIN controls on horizontal
rows (factored into renderSecuritySection so lite gets auto-lock+PIN and
full-node adds the RPC-backed encrypt/lock/remove block — same gating as before).
- DAEMON BINARY: inline Installed/Bundled/status; Check/Refresh/Test row; and the
four maintenance actions (Install bundled / Rescan / Delete blockchain / Repair)
behind a now-prominent, clearly-clickable "Advanced ▾" toggle instead of the
easy-to-miss faint overline — fixing the "daemon options missing" report.
Lite keeps its existing wallet-lifecycle/backup/security block, full width.
Removed useGrid/stacked/column clips/dividers. Every control + state branch +
confirm flag + tooltip preserved (verified). Full-node + lite build clean; ctest
green; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Single-column left every row hugging the left with a blank right half (and a
split RPC row); two-column variants left a bottom-corner gap. Rework the card as
a 2x2 grid on wide full-node windows so both halves carry real content:
TOP-LEFT NODE/Data | TOP-RIGHT RPC
BOTTOM-LEFT SECURITY | BOTTOM-RIGHT DAEMON BINARY
with the two top cells sharing a row bottom, a thin vertical + horizontal
divider, and the Daemon "Advanced" destructive buttons dropping to a full-width
4-button strip below the grid when expanded.
Gated behind one `useGrid` flag (supportsFullNodeLifecycleActions() &&
availWidth >= node-grid-breakpoint, default 900). When false — narrow windows and
lite — it falls through to the existing single full-width column, unchanged
(one-flag-revertible). Wallet Size moved to its own row (removes a mid-row gap in
both layouts). RPC / Security / Daemon rebased to per-cell X/width via aliases;
Advanced buttons factored into one lambda so the disabled predicates live in a
single place. Design produced + adversarially critiqued via a multi-agent pass.
New: components.settings-page.node-grid-breakpoint (ui.toml), "rpc_connection"
i18n string. Full-node + lite build clean; ctest green; hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two-column layouts left an unbalanced empty gap for this content (Node/RPC vs
Security+Daemon can't be evened into two columns), so drop the split entirely:
Node -> RPC -> Security -> Daemon Binary now stack vertically at full card width.
This guarantees no horizontal gap and gives the RPC grid and the Daemon Binary
button rows the full width they want.
Implemented by forcing the existing vertical-flow path (`stacked`) always on, so
the section bodies are unchanged; removes the now-unused column-split math and the
node-sec-stack-width schema key.
Full-node + lite build clean; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The 60/40 split left the right (Security) column short while Node/RPC on the left
was tall, and Daemon Binary sat full-width below — leaving a large empty gap in
the card's bottom-right.
Rework it as a greedy two-column masonry: even 50/50 columns, and the Daemon
Binary section is placed into whichever column is currently shorter (clipped to
that column's width). In practice Node/RPC fills the left and Security + Daemon
Binary stack on the right, so the gap is filled and the two sides balance in
height. When the window is narrow the columns stack and Daemon Binary spans full
width below, as before. Daemon Binary now sizes to its column (dbColW) instead of
the full card width.
Full-node + lite build clean; source hygiene clean. Visual QA needed (column
balance + the daemon button rows in the narrower column).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The fixed 60/40 split left a large empty gap in the bottom-right: the SECURITY
column is short while NODE (data dir + RPC grid) is tall, and the DAEMON BINARY
row spans full width below both. Two changes:
- Widen the side-by-side split to 70/30 so NODE (the content-heavy column, esp.
the RPC grid) gets the room and the right-side gap shrinks.
- Move the "Advanced" destructive-actions expander out of the full-width DAEMON
BINARY row and into the right column below Security, filling the gap. Its four
actions now stack vertically to fit the narrow column.
Full-node + lite build clean; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The zebra banding is emitted per text row, so it stopped at the last line and
left the bottom fade-padding (up to 18% of the panel) unbanded — very visible
when scrolled/pinned to the bottom. After the text-row bands, continue the
alternating pattern (same parity, lineHeight pitch) down to the panel bottom so
the zebra fills the whole output area. Also covers the case where the content is
shorter than the panel.
Full-node + lite build clean; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The per-line scanline shade alternated on rowIndex = floor((cumulativeY +
yOffset) / lineHeight + 0.5), but cumulativeY includes the 2px inter-line gap
while the divisor is lineHeight alone. That drifts ~0.14 row per line and, with
the rounding, jumps the index by 2 every ~7 rows — so two adjacent rows get the
same shade and the alternation visibly breaks in patches.
Replace it with a true per-visual-row counter (one increment per drawn sub-row),
seeded with the first on-screen line's sub-row ordinal so the shading stays
stable while scrolling. Parity now alternates exactly every row.
Full-node + lite build clean; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a "Portfolio" submenu to the Overview address-list context menu: it lists
each Market-tab portfolio entry with a checkmark showing whether this address is
in it, and clicking toggles membership (add/remove), persisting immediately. If
no entries exist yet, a disabled hint points to creating one.
Full-node + lite build clean; ctest green; source hygiene clean. Completes the
Market configurable-portfolio item (last of the todo.md batch).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Build the configurable-portfolio UI on the persisted model:
- The portfolio card keeps the "all funds" summary and now renders each custom
entry below it as a compact row (label · summed DRGX · USD), live-computed via
SumPortfolioBalance over the wallet's per-address balances; the card grows to
fit the entries.
- A right-aligned "Manage…" button on the card header opens a modal editor
(material overlay dialog): list entries with Edit/Delete, "Add entry", and an
edit form = label field + quick "All shielded / All transparent / Clear"
selectors + a checklist of wallet addresses (with balances). Persists to
Settings on each mutation.
- i18n English defaults for the new portfolio_* strings.
- Test namespace fix (AddressInfo is dragonx::, not dragonx::data::).
Full-node + lite build clean; ctest green; source hygiene clean. Remaining:
the Overview-tab right-click "add/remove address to a label" integration.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Foundation for the Market tab's configurable portfolio (custom labels tied to
groups of addresses):
- Settings gains a PortfolioEntry {label, addresses[]} list, persisted in
settings.json as a JSON array-of-objects (same mechanism as lite_servers_),
with get/setPortfolioEntries().
- data/portfolio.h (header-only, ImGui-free): SumPortfolioBalance() sums a
group's DRGX balance against the live per-address list (unknown addresses
contribute 0), plus PortfolioEntryContains/Add/Remove group helpers.
- Unit-tested (testPortfolioHelpers): empty/known/unknown-address sums, and the
add/remove/contains membership semantics.
UI (editor dialog, Market card rework, Overview right-click) comes next.
Full-node + lite build clean; ctest green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The NODE + SECURITY split was a fixed 60/40 two-column layout regardless of
width, which cramps the RPC grid and wastes space on narrow windows. Make it
responsive: below a configurable content width (node-sec-stack-width, default
600) the two columns stack into one full-width column — SECURITY flows below
NODE — so each section gets the full width; side-by-side 60/40 is kept when
there's room. Column-geometry-only change (leftX/rightX/widths + the right
column's top Y); the content blocks are untouched. The rejoin already used
max(leftBottom, rightBottom), correct in both modes.
Full-node + lite build clean; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Combine the top two Market cards: the price-summary hero and the chart now share
a single glass panel (one panel spans the price/stats header + the chart drawn
below it; the chart no longer draws its own panel or an inter-card gap).
Simplify the stat strip from three columns to two — drop the niche BTC-price
column, keeping 24h Volume + Market Cap — and remove the "updated Ns ago"
staleness line (data freshness is still shown by the attribution "Updated" line
under the pair buttons).
Full-node + lite build clean; ctest green; source hygiene clean. Completes the
Market "round buttons + combine/simplify top cards" item.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the exchange combo dropdown + the horizontally-scrolling pair chip bar
with a single flat selector: every trading pair across all exchanges is shown as
a round button (pair name + dimmed exchange sublabel) that wraps to multiple
rows. Clicking one selects that exchange+pair, persists it, and refreshes market
data. The CoinGecko attribution + last-updated line moves under the buttons.
Removes the now-unused pair-scroll state (scroll/drag/arrow machinery) and the
pair-bar-height budget input.
Full-node + lite build clean; source hygiene clean. (Second half of the Market
"round buttons + combine cards" item — the hero/chart card merge is next.)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The About card draws its logo deferred in the left column, scaled to the card
height (cardMax.y - bottomPad). Because the buttons row is full-width at the
bottom of the card, the tall logo extended down over it. Clamp the logo to end
just above the buttons row (captured Y minus a small gap) instead of the card
bottom — aspect + reserved-width caps are unchanged.
Full-node + lite build clean; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Warm the Mining tab's data during App::init() (both off the UI thread) instead of
only when the user opens the tab or clicks a mine / auto-balance button:
- kick the idempotent one-shot `xmrig --version` detection, and
- kick a background pool-stats refresh over the known pools.
Gated on supportsPoolMining(). The pool-stats snapshot lands on a later frame and
the auto-balance driver's self-throttling still applies, so this is just an
earlier first fetch — hashrates + version are visible as soon as the tab opens.
Full-node + lite build clean; ctest green; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Work through the console items in todo.md:
- Lite filter toggles (#5): replace the single hasLogFilters() bool with a
ConsoleLogFilterCaps struct so each backend advertises which toggles apply.
Full node = daemon/errors/rpc-trace/app; lite = errors-only + app (its
diagnostics ring maps to the App/Error channels; the text filter is always
shown). Lite previously showed no toggles at all.
- Color-coded toggles (#6): each filter checkbox is tinted with its channel's
accent color (daemon=blue, errors=red, rpc=secondary, app=teal) via a new
channelAccentColor() that also drives the output's left accent bar — one source
of truth for channel color.
- Darker terminal look (#7): drop the light glass rectangle on the input; both
output and input now get a terminal-dark overlay (tabs.console.bg-darken-alpha,
default 110) and the input field blends into it (transparent frame bg).
- Monospace font (#8): bundle Ubuntu Mono (res/fonts/UbuntuMono-R.ttf, Ubuntu
Font License — same family as the existing Ubuntu fonts) via INCBIN, load it at
caption size as Type().mono(), and render console output + input in it so
pretty-printed JSON columns and terminal text align. Falls back to the
proportional caption font if unavailable.
Full-node + lite build clean; ctest green; source hygiene clean; sandboxed
startup smoke confirms the mono font loads + atlas builds without crashing.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add fold/unfold of JSON objects and arrays in console command results, built on
the refactored channel + model foundation.
- ComputeConsoleFoldSpans() (console_input_model): pure brace/bracket matcher
over a block of pretty-printed JSON lines; each opener line gets the offset to
its matching closer at the same indentation (empty/one-line blocks are not
foldable). Unit tested (nested objects, arrays, empty blocks, non-JSON, and a
string value that merely contains a brace).
- ConsoleModelLine gains foldSpan (relative, so it survives front-eviction by the
line cap) + a collapsed flag; ingest() carries the span and toggleCollapsed()
flips openers (main thread). addFormattedResult computes the block's spans and
ingests them atomically.
- computeVisibleLines() skips a collapsed block's interior (opener stays, its
foldSpan lines through the closer are hidden); folding is bypassed while a
filter is active so every match stays reachable.
- drawVisibleLines() draws a fold triangle in the free left gutter of opener
lines (result/JSON channels carry no accent bar there), toggled by a click in
the gutter cell, and appends a dim " ... }" / " ... ]" summary on collapsed
openers.
Full-node + lite build clean; ctest green (incl. fold-span + model-fold tests);
source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Finish the thin-view pass on the remaining large render methods (pure code
motion, no behavior change):
- renderToolbar (225 -> 99 lines): extracted drawToolbarStatus(), the
drawLogFilterToggles() checkbox group, drawFilterInput(), and drawZoomControls().
- renderInput: factored the command echo + built-in interception + submit into
submitConsoleCommand(exec, cmd) -> bool, leaving renderInput to own just the
glass panel + InputText + completion/history callback.
- renderCommandsPopup (194 -> 135 lines): removed the duplicated lowercase-and-
find filter logic (match-count loop and per-row filter now share
consoleCommandMatchesFilter()) and extracted the [optional]-param dimming loop
into drawConsoleCommandParams().
Full-node + lite build clean; ctest green; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
renderOutput() was a ~410-line method doing filter-building, mouse/keyboard
interaction, the per-line draw loop, the new-output pill, and the right-click
context menu all inline. Split it into a thin orchestrator that calls five
single-purpose private methods (pure code motion, no behavior change):
- computeVisibleLines() -> the filter -> visible_indices_ + match count
- handleOutputInteraction() -> wheel-up, selection drag lifecycle, Ctrl+C/A
- drawVisibleLines() -> accent bars, JSON indent guides, selection +
filter highlight, scanline capture, text
- drawOutputContextMenu() -> the right-click menu
- drawNewOutputIndicator() -> the jump-to-bottom pill
Ordering and every operation are preserved exactly (interaction still runs
after layout build and before the draw; selection bounds are recomputed post-
interaction; scanline_rows_ cleared before the draw pushes to it). Full-node +
lite build clean; ctest green; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pull the auto-scroll concern out of ConsoleTab into a focused, ImGui-free
ConsoleScrollController (console_scroll_controller.{h,cpp}). It owns the three
coupled fields that were loose on the god-class — auto_scroll_, the wheel-up
cooldown, and the new-lines-while-scrolled-up backlog count — and the state
machine tying them together: wheel-up detaches + starts a cooldown, the cooldown
gates the at-bottom re-enable check, backlog counting happens only while
detached, and re-pinning (checkbox / jump pill / reaching the bottom) clears it.
The view keeps only what needs ImGui: it measures scroll position, hands the
controller a plain `atBottom` bool + frame delta, and issues the actual
SetScrollHereY. ConsoleTab loses three more members and all the scattered
auto-scroll bookkeeping collapses to scroll_.* calls.
Adds testConsoleScrollController: pinned start ignores backlog, wheel-up detach +
cooldown gating, re-enable only when at bottom, checkbox toggle semantics, and
the jump-to-bottom pill. Full-node + lite build clean; ctest green; source
hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pull the text-selection concern out of ConsoleTab into a focused, ImGui-free
ConsoleSelectionController (console_selection_controller.{h,cpp}). It owns the
anchor/caret state and all the fiddly logic around it: the mouse-drag lifecycle
(beginDrag/updateDrag/endDrag), select-all, range ordering, text extraction over
the visible model, and the index shift applied when the buffer cap evicts lines
from the top.
ConsoleTab now delegates: the four scattered selection fields (is_selecting_,
has_selection_, sel_anchor_, sel_end_) and five helper methods (selectionStart/
End, isPosBeforeOrEqual, getSelectedText, clearSelection) collapse to a single
selection_ member plus a thin selectedText() wrapper. screenToTextPos (still
view-side — it needs the frame's layout) now returns the shared ConsoleTextPos,
so ConsoleTab::TextPos is retired.
The byte-offset / eviction math was previously only reachable through the live
ImGui renderer; it now has direct unit coverage (testConsoleSelectionController:
ordering, drag lifecycle, upward drag, select-all, partial/full/no-op eviction
shift, and filtered extraction). Full-node + lite build clean; ctest green;
source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Introduce ConsoleModel (console_model.{h,cpp}): an ImGui-free, thread-safe line
store. Producers on any thread call ingest(), which only briefly locks a small
pending queue; the main thread calls drain() once per frame to move pending
lines into the visible deque and enforce the line cap, reporting how many were
added and evicted from the front.
This fixes the real data race. Previously a single lines_mutex_ guarded both the
line container AND the UI-only interaction state (selection, scroll, auto-scroll
counters), and it was held across the whole render pass — yet render() touched
auto_scroll_ / new_lines_since_scroll_ outside that lock, racing the background
RPC-trace and app-logger producers that call addLine() on worker threads.
Now:
- addLine() only ingests (lock-free w.r.t. the model); safe from any thread.
- render() drains once per frame, then does the cap/selection/scroll bookkeeping
on the main thread (new drainModel()).
- The visible deque + all selection/scroll fields are main-thread-only, so every
lines_mutex_ lock is gone — including the one wrapping the entire renderOutput.
Adds testConsoleModel() covering ingest/drain ordering, the line-cap eviction
counts, view-only clear (queued lines survive), and an 8-thread concurrent
ingest/drain stress with no lost lines. Full-node + lite build clean; ctest
green; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Make ConsoleChannel the canonical semantic classification of every console
line. Producers (command executor, app log forwarders, result formatter) tag
each line with a channel; the UI derives both the text color and the left
accent-bar color from it at draw time, and the output filter keys off it.
This replaces the prior scheme of storing an ImU32 color per line and then
reverse-engineering the source from color-equality plus a "[daemon] "/"[xmrig] "
/"[app] "/"[rpc] " text prefix. Consequences:
- ConsoleLine now stores {text, ConsoleChannel} — no per-line color.
- addLine()/addRpcTraceLine() take a channel; the redundant text prefixes are
gone (the accent bar carries the origin).
- The executor's ConsoleAddLineFn hands a channel, not an ImU32, so the backend
no longer depends on ConsoleTab::COLOR_* — FullNode/Lite executors emit clean
text + channel.
- The theme-remap loop that rewrote stored colors on light/dark flip is deleted;
colors are resolved per-channel each frame (refreshColors on flip only).
- ConsoleOutputFilter drops its color fields; consoleLinePassesFilter() keys off
the channel. Test updated to the channel-based API.
- Added a Warning channel (amber severity peer of Error/Success) so the
notification + logger warning forwarders keep their color.
Full-node + lite build clean; ctest green; source hygiene clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Lift the console's word-wrap layout, screen->text hit-testing, and selection extraction
out of ConsoleTab into a pure, ImGui-free module (console_text_layout.{h,cpp}). Text
measurement is injected via a ConsoleTextMeasure interface — production uses an
ImFont-backed impl (identical CalcWordWrapPositionA/CalcTextSizeA calls as before), while
tests drive a fixed-width stub.
- BuildConsoleLayout: per-line wrap segments + heights + cumulative Y (was inline in
renderOutput).
- HitTestConsoleLayout: point -> (visibleRow, byte col) (was screenToTextPos).
- ExtractConsoleSelection: selected text across visible lines (was getSelectedText).
- ConsoleTab now holds a single `ConsoleLayout layout_` (replacing 4 mutable members;
drops the dead cached_wrap_width_) and calls the pure functions.
- New unit test testConsoleTextLayout covers wrapping, multi-row hit-testing, bottom
clamp, and filtered/unfiltered selection — the biggest previously-untested surface.
No behavior change (algorithm preserved verbatim). Full-node + lite build clean, ctest
green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Zero-behavior-change cleanup ahead of the console refactor:
- Delete dead API/members: addCommandResult (unused), handleSelection (declared, no
def), isAutoScrollEnabled (no callers), and scroll_to_bottom_ (written 8× but never
read) with all its writes.
- Drop the unused screenToTextPos `line_height` parameter (+ 4 call sites) and the unused
`segEnd` local — clears the two -Wunused warnings.
- Remove the toolbar filter checkboxes' dead `static bool s_prev_*` change-detectors,
whose only effect was setting the dead scroll_to_bottom_.
- Fix the console_new_lines format/arg mismatch (format takes one %d; the call passed a
spurious plural-suffix arg).
- Fix the channel/color incoherence: a "[daemon] error:" line kept red text but got a
blue daemon bar — the prefix no longer downgrades an error channel.
- Delete 11 orphaned lite_console_* i18n keys left over from the console merge (keep
lite_console_help_passthrough).
Full-node + lite build clean, ctest green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Draw faint vertical guides per 2-space nesting level on result (non-channel) lines, so
nested RPC JSON is easier to read. Draw-only in the per-line loop (like the channel bar),
so it doesn't touch the text layout or selection.
Note: collapsible JSON nodes were intentionally deferred — folding entangles with the
wrap cache / filter / selection line indices and warrants a runtime-verified pass.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Right-clicking the output now offers "Copy <value>" when a long alphanumeric token
(txid/blockhash/address, >=16 chars) is under the cursor — resolved via screenToTextPos.
Uses right-click so it never conflicts with the left-drag text selection.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
renderOutput already draws a clickable "N new lines" jump-to-bottom indicator
(console_tab.cpp ~996), so the pill added earlier duplicated it — and reused the
"console_new_lines" format string incorrectly. Remove the pill and the stray i18n entry;
the existing indicator stands.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Highlight the matched substring (case-insensitive) in every visible line with a
translucent yellow behind the text, drawn per wrap-segment like the selection highlight.
- Show a live match count next to the filter input ("N matches", red when zero).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Each line gets a `channel` inferred in addLine() from its color and any
"[daemon]"/"[xmrig]"/"[app]"/"[rpc]" prefix, which is then stripped. renderOutput draws
a thin colored bar in the existing left margin (padX) per line — node log (blue), mining
(amber), app (teal), rpc (secondary), command (accent), error (red). Drawn in the margin
so it never touches the text layout or selection; filtering is unaffected (it keys off
color, not the prefix).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Surface new_lines_since_scroll_ (already tracked) as a floating pill at the output
panel's bottom-right whenever the user is scrolled up with unseen output; clicking it
re-enables auto-scroll and snaps to the bottom.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The two consoles were separate implementations (the rich full-node ConsoleTab and a
minimal RenderLiteConsoleTab). Introduce ConsoleCommandExecutor to abstract the only
real differences — command execution + log source — so both variants share the one rich
terminal (the lite console now gains selection, zoom, copy, and JSON-colored output).
- console_command_executor.{h,cpp}: interface + FullNodeConsoleExecutor (dragonxd log
ingestion + RPC command execution, result queue) and LiteConsoleExecutor (lite
diagnostics ring + backend runConsoleCommand/takeConsoleResult, connection/sync status).
- ConsoleTab::render() now takes a ConsoleCommandExecutor&; log ingestion, command
submission, and command results flow through it. The RPC command-reference popup and
the daemon/errors/rpc-trace/app filter toggles are gated on the executor's capabilities;
the toolbar status dot + a status header come from the executor.
- Safety preserved: `clear`/`cls` is intercepted as a view-only clear in the shared UI so
it is NEVER forwarded (the lite backend's `clear` wipes tx history); command output is
not persisted to diagnostics.
- App owns the executor (created per variant) and routes both NavPages to console_tab_;
lite_console_tab.{h,cpp} deleted.
Needs runtime verification of BOTH consoles (full-node RPC + lite: clear, seed/export
secrecy, sync status, command results).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Fetch /coins/dragonx-2/tickers once per session (App::refreshExchanges, reusing the
price-fetch worker path) and build the exchange registry at runtime into
state.market.exchanges; parseCoinGeckoTickers groups tickers by venue. The Market tab
sources from this live list and falls back to the compiled-in registry (now including
OurBit + Nonkyc.io) when offline. Fixes the missing OurBit pair and auto-tracks future
listings.
- Fix stale selection defaults (TradeOgre / DRGX-BTC -> Nonkyc.io / DRGX/USDT) that never
matched the registry, and make the attribution generic ("Price data from CoinGecko").
- Reframe the single-asset portfolio card as a DRGX holdings dashboard: relabel to
"MY DRGX" and show the 24h change on the holdings value (colored by direction).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Draw the brand word "ObsidianDragon" in Ubuntu Medium (the closest weight the
project ships — no true 600 face) as a stand-in for semibold; append "Lite" in the
Regular font under DRAGONX_LITE_BUILD so the lite window reads "ObsidianDragonLite".
- Fix a misleading comment: pool mining (and thus auto-balance) is available in both
builds; only solo mining is full-node-only.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- About: draw the logo deferred, scaled to the card's full height (aspect-preserved,
capped to the reserved width) so there's no empty space below it.
- Node & Security: split the daemon toolbar — Refresh + Test connection stay visible;
the rare/destructive actions (Install bundled, Rescan, Delete blockchain, Repair
wallet) move behind a collapsible "Advanced" header (reusing the DEBUG LOGGING idiom).
- Lite gating: hide the DEBUG LOGGING card (dragonxd debug= categories, no daemon in
lite) and the RPC-backed encrypt/change/lock/remove block (lite has its own encryption
in the NODE column) behind full-node guards; label the lite left column "WALLET" not
"NODE". (Auto-lock + PIN remain in both.)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Address list: a mining-flagged address stays visible even at 0 balance when
"hide zero balances" is on (payout addresses shouldn't vanish).
- Drag-reorder: persist dense sort orders (0..N-1) for the whole visible list on
drop via App::reorderAddresses, instead of a pairwise swap that no-ops when both
rows are still at the default un-ordered state. First drag now always takes
effect, and explicit order keeps overriding the starred/type/balance sort.
- Tests: 0-balance mining row survives hide-zero; ordered non-favorite outranks a
favorite.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Split the pool-mode hashrate card into a 25%-width left card + 75% chart so the
Manual/Auto toggle and pool metrics no longer reflow the tab:
- left card: Manual|Auto toggle (with tooltips), a Local/Pool/Shares/Uptime
metric stack, and a fixed-height scrolling pool list — rows click-to-select in
Manual (reusing the existing save+restart path) and show a "mining here" dot in
Auto. Fixed height keeps the card constant as pools are added.
- mode-toggle row: keep the suggested-pools dropdown + auto-mode URL disable;
the toggle and pool list moved into the card.
- mining_controls: show the installed/running miner version ("Current: …")
before and during mining instead of "none".
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- app: periodic weighted-random auto-balance of the pool while pool-mining
(~30-min cadence; restarts the miner only when the pick actually changes),
plus snapshot + refresh accessors for the UI.
- app_network: resolve the xmrig algo per pool at start (pool.dragonx.cc =
rx/dragonx, pool.dragonx.is = rx/hush; custom hosts keep the setting).
- xmrig_manager: schema-aware pool-side hashrate readout (fixes a silent 0 for
Miningcore pools); expose the running miner's version from its HTTP API and a
cached `xmrig --version` detection so the UI can show it before mining.
- wallet_state: carry the running miner's version.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Data + service layer for supporting multiple mining pools:
- pool_registry: KnownPool table (pool.dragonx.is + pool.dragonx.cc) carrying
each pool's stratum host, xmrig algo, and stats-API schema; pure helpers
findKnownPoolByUrl / resolvePoolAlgo / parsePoolHashrate / chooseWeightedPool
(weighted-random by hashrate with incumbent stickiness).
- pool_stats_service: background, schema-aware hashrate fetcher (libcurl, TLS,
browser UA for Cloudflare) modelled on XmrigUpdater.
- settings: pool_select_mode (manual | auto_balance), string-serialized.
- i18n strings + unit tests (both stats schemas, algo resolution, weighted pick).
Note: pool.dragonx.cc's stratum host is us.dragonx.cc:3333 — the .cc domain is
only the Cloudflare-proxied web/API host and does not accept stratum on :3333.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Set project(ObsidianDragon VERSION 2.0.0) (was 1.3.0); the lite variant keeps
its own independent DRAGONX_LITE_VERSION (1.0.0). res/ObsidianDragon.manifest is
generated per-variant by configure_file from res/ObsidianDragon.manifest.in, so
it was wrongly tracked (it kept showing dirty, stamped with whichever variant
built last) - untrack it and gitignore it; the .in template remains the source.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Vendor the two local Rust crates that build the lite backend artifact into
third_party/silentdragonxlite/ (the qtlib C-ABI wrapper + the silentdragonxlitelib
core, with proto/res and all the lite-send fixes), and point
build-lite-backend-artifact.sh's default --backend-dir there, so the lite wallet
builds without the upstream SilentDragonXLite repo.
External build inputs are now only the Rust toolchain + git.dragonx.is + crates.io:
- the 6 librustzcash git deps point at the git.dragonx.is/DragonX/librustzcash
mirror (pinned rev acff1444), not git.hush.is;
- the Sapling params are gitignored (not committed, no Git LFS) - the build fetches
them from the git.dragonx.is/DragonX/zcash-params 'sapling-v1' release and verifies
their SHA-256 before rust-embed bakes them in (ensure_sapling_params).
For fully offline builds, cargo vendor into lib/vendor/ and add a vendored-sources
redirect (vendor/ is gitignored; the script symlinks it into the prepared dir).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
tools/lite_send_smoke drives the real SDXL backend's exact GUI send path
(newaddr/status/list/tree/send/rescan) for diagnosing shielded sends.
scripts/gen-lite-checkpoints.sh generates verified mainnet checkpoints from a
synced dragonxd (getblockhash + getblockmerkletree), self-checking against a
known checkpoint before emitting.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Network tab: glow only the active node, drop the left accent bar. Key-export
dialog: fix the lite-wallet "Not connected" failure by exporting the key
locally via the SDXL backend when there's no daemon; rework the layout to
wrapping click-to-copy fields with a side QR (empty placeholder when hidden),
85% modal width, HRP-preserving key chunking, and a centered, emphasized
warning. QR popup matched to the same sizing and click-to-copy address. Shared
field rendering extracted to widgets/copy_field.h so both dialogs stay in sync.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a full-node daemon updater (util/DaemonUpdater + daemon_download_dialog)
reachable from Settings -> NODE & SECURITY: downloads/verifies (SHA-256 +
enforced ed25519 signature) and atomically installs the latest dragonxd from
the project Gitea, with a "Restart daemon now" step. Add a shared "Browse all
releases..." picker (release_list_view) to both the miner and daemon updaters
so users can pin older/pre-release builds. Pure no-I/O cores
(daemon_updater_core / xmrig_updater_core) are unit-tested; sign-daemon-release.sh
signs release archives offline.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add material::Tooltip / BeginTooltip / EndTooltip wrappers (tooltip_style.h)
that scope a small window padding (8x4, dpi-scaled) and ~85% opacity to
tooltips only, then route the tooltip call sites through them. Menus and combo
dropdowns are untouched (they keep the global opaque PopupBg).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Use the new daemon's "Reading blocks for witness rebuild: <done> / <total>" as an
exact fraction: the reported total is the denominator directly, so the bar sweeps
0..1 smoothly instead of being held near the top by the peak-anchored remaining
heuristic (kept only as a fallback for older daemons that log bare "<n> remaining").
Also snap to 100% on the parallel rebuild's completion line ("rebuilt <n> note
witness cache(s) … using <t> thread(s)"), which otherwise logs no progress, so the
bar visibly finishes before the rescan-complete signal clears it.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The updated dragonxd (parallel witness rebuild) replaced the per-block
"Building Witnesses for block <h> <frac> complete, <n> remaining" log line with a
clean serial read counter "Reading blocks for witness rebuild: <done> / <total>".
Parse the new line and map it onto the existing phase-2 path (remaining = total -
done), so the witness progress bar shows done/total against this daemon. The old
"Building Witnesses" matcher is kept for backward compatibility with older daemons;
"Setting Initial Sapling Witness for tx … i of N" (phase 1) is unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The initial-witness pass ("Setting Initial Sapling Witness") and the cache walk
("Building Witnesses for block … remaining") interleave during a rescan — the daemon
does both per block. The phase selector picked phase 1 whenever a batch had only
initial-pass lines, so once the cache walk started an interleaved initial line would
flip the phase back to 1 and reset the bar to 0 every batch. Make the phase
upgrade-only (once the cache walk is seen it never drops back), so the reset happens
at most twice (→1, →2) and the cache-walk percentage advances monotonically.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Verified by running the app against a live node and watching a real rescan. Three
issues that only surfaced at runtime:
- Wrong RPC name: this daemon (hush/komodo) exposes the runtime rescan as
"rescan <height>", not bitcoin's "rescanblockchain". runtimeRescan() and
RPCClient::rescanBlockchain() used the bitcoin name and failed with "Method not
found" on every node. Corrected to "rescan".
- Witness/rescan progress never surfaced during a rescan: the daemon-output parser
that drives it was gated behind rpcConnected, but a heavy rescan holds cs_main so
getinfo times out and the RPC reads disconnected — silencing the parser exactly
when it's needed. The parser reads the daemon's stdout pipe (no RPC), so it now
runs whenever the daemon process is alive. It also now parses INLINE on the main
thread instead of via fast_worker_, so it can't be starved when the worker is
blocked on a getrescaninfo call (which waits on cs_main during a witness rebuild).
- Witness rebuild has TWO sub-phases with different scales — the initial-witness
pass ("Setting Initial Sapling Witness for tx <hash>, <i> of <N>") and the cache
walk ("Building Witnesses for block <h> <frac> complete, <n> remaining"). Tracking
them with one monotonic value pinned the bar at the initial pass's ~100% through
the whole cache walk. They're now tracked as distinct phases (witness_phase) with
their own monotonic progress and labels ("Setting witnesses" vs "Rebuilding
witnesses"), so neither resets/bounces and the long phase shows real movement.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The witness-rebuild bar reset repeatedly because the daemon's "Building Witnesses
for block <h> <frac> complete" line reports per-call progress: BuildWitnessCache is
re-invoked for each connected block and each call walks from its own start height to
the tip, so the fraction restarts every time. The earlier "Setting Initial Sapling
Witness for tx <i> of <m>" counter resets per call too, so neither is a usable
overall metric.
Derive a stable, monotonic percentage from the "<n> remaining" count instead: track
the largest "remaining" seen during the phase as the full span and show how far
remaining has fallen below it. The longest pass defines 0→100%; the short per-block
follow-up passes only nudge the bar near the end rather than resetting it. The
"Setting Initial" line now only marks the phase active. Per-phase tracking resets at
phase start and every rescan-completion site.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The daemon's post-rescan witness rebuild ("Building Witnesses for block ...")
is a distinct, often-long phase that previously showed only as an indeterminate
"Rescanning..." with no progress. Parse the daemon's witness-build log lines and
surface a dedicated progress indicator.
- Parse "Building Witnesses for block <h> <frac> complete, <n> remaining" (and the
earlier "Setting Initial Sapling Witness for tx ..., <i> of <m>") from daemon
output, extracting a 0..1 fraction and remaining-block count.
- New SyncInfo fields building_witnesses / witness_progress / witness_remaining,
cleared at every rescan-completion site (warmup-end, getrescaninfo poll, runtime
rescan callback, daemon-log "finished").
- Status bar shows "Rebuilding witnesses NN%" (priority over the generic rescan
text); the loading overlay (shown during -rescan warmup) gets a labelled witness
progress bar with the remaining-block count.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A wallet bootstrapped from a snapshot keeps its wallet.dat but never rescans, so
its spent-state is stale and the first send tries to spend already-spent notes and
is rejected. The startup -rescan flag can't fix it either: the snapshot lacks the
pre-snapshot block history -rescan needs, so it errors. The working fix is a runtime
rescanblockchain RPC from a height the snapshot actually has.
- Add App::runtimeRescan(startHeight): runs rescanblockchain via the worker, drives
the rescanning UI state, and owns completion via the RPC callback (getrescaninfo
is unavailable on this daemon). Suppresses the per-second mining/rescan pollers
and the Core/balance/tx refreshes while the daemon holds cs_main for the scan.
- Add App::detectLowestAvailableBlockHeight(): async binary search via getblock for
the lowest height whose block data is on disk → the snapshot base, and whether the
node still has full history.
- Auto-reconcile after bootstrap: both completion sites (wizard + Settings download
dialog) mark a pending rescan; once the daemon is back up and the tip is known,
detect the base and runtimeRescan() from it (or -rescan restart on a full node).
- Settings "Rescan Blockchain" now probes first: full-history nodes get the existing
-rescan restart; bootstrapped/pruned nodes get a prompt pre-filled with the
detected base height that runs the runtime rescan.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Daemon binary panel (info + the all-actions toolbar) lived inside the 60%-wide
NODE column, so the six-button toolbar clipped. Pull it out into a dedicated full-width
row rendered after the NODE + SECURITY columns reconcile, so it spans the whole card:
Installed | Bundled info side by side, status line, and the Install bundled | Refresh |
Test connection | Rescan | Delete blockchain | Repair wallet toolbar now have the full
container width and no longer clip. The NODE column keeps only the node/RPC info.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Move Test Connection / Rescan Blockchain / Delete Blockchain / Repair Wallet onto the
same line as and right after the Daemon binary buttons (Install bundled | Refresh), so
all node actions form a single toolbar row beneath the Daemon binary panel. Buttons are
auto-sized to pack onto one line; each disabled-state group (connection vs embedded
daemon vs bundle present) keeps its own guard. Removes the former two stacked button rows.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Delete Blockchain and Repair Wallet now sit on one line, paired with the same
uniform button width as Test Connection / Rescan Blockchain (a clean 2-column grid;
all maintenance buttons share one rowBtnW sized across the four labels).
- The Daemon binary panel lays Installed and Bundled side by side across the node
column (version + size·date | version + size) instead of stacked narrow lines, with
the status line spanning underneath and Install bundled / Refresh paired below.
- Shorten the install button label to "Install bundled" so it fits the shared width;
the tooltip still explains the full action. Date shown as YYYY-MM-DD.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Previously the wallet re-extracted the bundled dragonxd on startup whenever the
installed binary's size differed from the bundle ("stale" overwrite), which could
replace a node a user had deliberately placed in dragonx/.
Now dragonx binaries (dragonxd/cli/tx) are auto-placed ONLY when missing — never
auto-overwritten on a size mismatch (needsParamsExtraction + extractEmbeddedResources).
Params/asmap keep their size-based refresh; a daemon dropped next to the wallet exe
still takes priority and is never touched.
Replacing the daemon is now an explicit action: Settings → "Daemon binary" reports the
installed binary's version (scanned from the file), size and modified date, compares it
to the version bundled in this build, and offers an "Install bundled daemon" button.
That stops the node, overwrites dragonxd/cli/tx with the bundled copies (waiting for the
process to release the file lock), and restarts — wallet/keys/chain data untouched.
Adds resources::{getInstalledDaemonInfo,getBundledDaemonInfo,reextractBundledDaemon}
(+ a version-string scanner) and App::reinstallBundledDaemon().
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
dragonxd's z_sendmany picks notes to cover the recipient total (nTotalOut) but not
the miner fee, then rejects the build unless the selected notes cover amount+fee
(rpcwallet.cpp:5312 vs asyncrpcoperation_sendmany.cpp:278). So a shielded send whose
largest notes sum exactly to the amount fails with "Insufficient shielded funds,
have H, need H+fee" despite ample balance — e.g. sending exactly 2.0 from an address
whose biggest note is 2.0.
Since the failure is async (reported via the opid poll), detect it there: when a
shielded send fails with that message and the selected total H >= the requested
amount (selection covered the amount but stopped one note short of the fee — vs a
genuine shortfall where H < amount), re-issue the send once with a tiny self-output
(= fee) back to the from-address. That lifts the daemon's selection target past the
boundary so it grabs another note and can cover the fee; the recipient still receives
the exact amount. Retries are tracked so a second failure surfaces normally (no loop).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
renderLoadingOverlay() rendered the daemon's recent output twice during startup:
a bare 4-line centered tail (section 2c, init/warmup only) and the styled
terminal-style box (section 4, always shown when the embedded daemon exists).
The bare tail was a strict subset of the box, so the same dragonxd output showed
stacked twice. Remove the redundant bare tail; keep the terminal box (which also
matches the shutdown screen's panel).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
When a note's stored record is corrupt or its tx isn't in the canonical chain,
z_sendmany fails to build a valid sapling spend proof even after a full -rescan,
because a plain rescan replays witnesses but keeps the existing tx/note records.
The zcashd repair for this is -zapwallettxes=2, which deletes all wallet tx/note
records and rebuilds them from the chain (keys/addresses preserved).
Adds a RepairWallet lifecycle operation that mirrors the existing -rescan plumbing
(one-shot zapOnNextStart flag on the embedded daemon; -zapwallettxes=2 implies and
supersedes -rescan), an App::repairWallet() that reuses the rescan status UI (so the
status bar + warmup-end completion detection apply), and a confirmed "Repair Wallet"
button + dialog in Settings → node maintenance (embedded daemon only).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two related fixes for the post-bootstrap "send fails / rescan stuck at 99%" trap:
1) Rescan completion now keys off warmup-end. A -rescan runs entirely inside daemon
warmup (every RPC returns -28 until it finishes), so warmup completing IS the rescan
completing. The old detectors relied on getrescaninfo (which some daemons answer with
"Method not found") or a "Done rescanning"/bench log line the daemon may never print,
leaving the status bar stuck at 99% — so users killed the rescan before it finished.
When warmup ends and a rescan was confirmed active, clear the rescan state, flip to
100%, refresh history/balance, and toast completion.
2) z_sendmany failures that mean stale shielded note data (shielded-requirements-not-met,
missing sapling anchor, invalid sapling spend proof, bad-txns-sapling-*) now append a
plain-language hint telling the user to run a full rescan, instead of surfacing only the
raw daemon string.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The lite status bar showed "blocks: 0" once fully synced. The backend's
`syncstatus` only includes synced_blocks/total_blocks WHILE actively scanning;
at rest it returns just {"syncing":"false"}, so the parsed syncedBlocks was 0
and became state.sync.blocks. On the synced refresh path, additionally query the
backend `height` command (wallet last-scanned height, a fast local read) and use
it as the synced height/tip, so the block count is correct at rest.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
On first run the lite welcome screen's "Restore from seed" button only showed a
hint toast and bounced the user to Settings, dismissing the welcome with no
wallet open — it never prompted for a seed. Add a real restore step to the
welcome wizard: a seed-phrase field + optional birthday height, which calls
beginRestoreWalletAsync() (same server failover as create/open), shows
"Restoring…" progress, then completes (wallet syncs) or surfaces the error to
retry. The seed buffer is wiped on success/Back and in finish().
(The Settings -> Lite -> Restore path already prompted for a seed; this fixes
the first-run welcome path.)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The lite SilentDragonXLite backend stores its wallet in its own directory
(dirs::data_dir()/silentdragonxlite — %APPDATA%\silentdragonxlite on Windows,
~/.silentdragonxlite on Linux, ~/Library/Application Support/silentdragonxlite on
macOS), NOT the full-node getDragonXDataDir() (…/Hush/DRAGONX). The newly added
lite "Open data folder" button opened the wrong (full-node) directory.
Add Platform::getLiteWalletDataDir() mirroring the backend's get_zcash_data_path
for the "main" chain, and point the lite button at it. The full-node button is
unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Lite send/shield, unlock, and key-import failures were shown only as transient
toasts — impossible to copy. Route them through liteLog() so they also appear in
the lite Console (which has a Copy button), alongside the lifecycle/open/sync
errors the controller already logs:
- send/shield broadcast failures (App broadcast-result delivery)
- wallet unlock failure
- key import failure (controller; logs the error text only, never the key)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add an explicit button in Settings that opens the wallet/blockchain data
directory (getDragonXDataDir()) in the OS file manager via the existing
Platform::openFolder(). Placed in the full-node connection section (next to the
data-dir path, which was only a subtle clickable link) and in the lite section
(always available). i18n strings added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a maintenance option for the lite wallet to re-download and re-scan every
block from the lite server — useful when balances or history look wrong.
- LiteWalletController::startRescan() runs the backend `rescan` command (which
clears the wallet's synced block cache and re-syncs from its birthday) on a
detached thread, reusing the existing sync progress/refresh machinery: it
resets syncDone_ so refreshModel() shows progress again and refreshes data on
completion. No-op if no wallet is open or a scan is already running.
- scanInProgress() exposes the initial-sync-or-rescan state.
- Settings (lite, open wallet) gains a "Redownload blocks" button behind a
confirmation modal, disabled while a scan is running. i18n strings added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The header and coin logos load disk-first (for dev builds / theme drop-ins) and
fall back to the copies embedded in the exe. The portable single-file build has
no res/img/ folder beside it, so the disk read always failed and logged
"LoadTextureFromFile: failed to read ..." before the (successful) embedded
fallback. Guard each disk load with std::filesystem::exists() so the missing
file is skipped silently and we go straight to the embedded logo — no error
line, logos unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The console mixed RPC traces, daemon output, and the wallet's own "[app] ..."
log lines with no way to hide the latter. Add an "App" checkbox alongside the
existing Daemon/Errors/RPC toggles. Since [app] lines share COLOR_INFO with
other info text, the filter matches them by their "[app] " prefix rather than by
color. Default on; unit test + i18n added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
On a fresh open the status-bar peer count stayed 0 until the Network tab was
opened. refreshData() — the one-shot refresh run on connect / warmup-complete /
unlock — only refreshed peers when the active tab was Peers, so on any other tab
nothing populated the count until a tab visit forced it. Refresh peers
unconditionally there so the count appears right after connecting; the periodic
20s Peers timer (all tabs) keeps it current after that.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a fuzzy mode to the explorer search: a non-numeric, non-full-hash query now
filters the list to cached blocks whose hash (or height text) contains the query
substring, live as you type. Backed by a new ExplorerBlockCache::searchBlocks()
(SQLite LIKE with escaped wildcards), memoized per query so it doesn't hit the DB
every frame. Exact queries still navigate precisely: a block height re-anchors
the list, and a full 64-char hash is resolved via RPC. Row clicks still open the
detail modal. Empty results show "No matching cached blocks".
Note: fuzzy matching covers cached (browsed/prefetched) blocks only — the daemon
has no partial-hash index — while exact height/hash lookups reach any block.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Typing in the explorer search ran an exact lookup that popped the block-detail
modal. It now updates the recent-blocks LIST as you type: a block height
re-anchors the list to that height (offline-friendly, no RPC), and a complete
64-char hash is resolved to its height and the list jumps there (a txid still
shows the inline tx view) — all without a modal. Clearing the box returns to the
recent (tip) blocks. Row clicks still open the detail modal for an explicit
full view. Removed the now-unused fetchBlockDetailByHash.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The explorer search only ran on Enter or the Search button. Now it also fires
automatically ~350ms after the user stops typing, once the query is resolvable
— a block height (all digits) or a complete 64-char hash/txid. Partial hex is
ignored so it won't flash "invalid query" mid-type, per-keystroke RPC spam is
avoided via the debounce, and the same query isn't re-run. Enter/button still
work for an immediate search.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The Rescan blockchain and Restart daemon buttons fired immediately on click —
both are disruptive (long offline rescan / connection drop) and easy to hit by
accident. Route them through confirmation modals, matching the existing
delete-blockchain / clear-ztx confirmations: the button now sets a confirm flag
and an overlay dialog performs the action only on explicit confirm. New i18n
strings added with English defaults.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The peer count in the status bar is state_.peers.size(), refreshed only by
getpeerinfo — and the peers refresh interval was 0 (disabled) on every tab
except Peers. So the count never changed until you opened the Peers/Network
tab. Give peers a slow 20s cadence on all tabs (30s on Console); the Peers tab
keeps its fast 5s for the live list. During sync this is still overridden by
kSyncProfile (peers 0) so it can't contend with block download. Test updated to
the new intervals.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
When a newer wallet build embeds a newer daemon, extractEmbeddedResources()
detects the size change and tries to overwrite dragonxd.exe in the daemon dir —
but the write is a plain truncating ofstream, which fails silently if the file
is locked. A running (or just-killed, handle-not-yet-released) daemon locks the
.exe on Windows (and Linux returns ETXTBSY), so the stale binary was kept and
the wallet kept launching the old daemon version.
If the direct write fails, move the stale binary aside to "<name>.old" (renaming
a running/locked executable is permitted on both Windows and Linux — the running
process keeps the moved copy) and write the fresh one at the original path. A
best-effort pass removes leftover .old files once the old process has exited.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
build.sh bundles whatever daemon binary already sits in prebuilt-binaries/, and
setup.sh only rebuilds a platform's daemon when its flag (--win/--mac) is passed
— so a daemon left over from an older source revision silently shipped in the
wallet (the Network tab showed dragonxd v1.0.1 while the source was v1.0.2).
Add a stale-daemon guard: compare the vX.Y.Z baked into each prebuilt daemon
against CLIENT_VERSION_* in the checked-out dragonx source. On the present/skip
and --check paths it now prints either "matches dragonx source" or a STALE
warning naming both versions and the rebuild command, plus a summary reminder at
the end of the daemon section. Version is read with grep -a (no binutils/strings
dependency); no-ops cleanly when the source or a binary is absent.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The History badge counts transactions with confirmations==0, iterating the raw
transaction list. Autoshield transactions have two legs sharing one txid, and
the send leg parsed from z_viewtransaction carries confirmations=0 even when the
transaction is long confirmed (the receive leg holds the real count). So the
badge counted those stale legs and stuck at a non-zero number (e.g. 7) with no
pending transactions.
Treat a txid with ANY confirmed leg as confirmed, and count UNIQUE unconfirmed
txids rather than legs — so confirmed multi-leg transactions don't inflate the
badge and genuinely pending ones still count once each.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a sort selector next to the type filter with four modes: Newest first
(default), Oldest first, Largest amount, Smallest amount. The mode folds into
the merged-list memoization cache key (so the list re-sorts only when the mode
changes) and the comparator branches on it, keeping txid as a deterministic
tiebreak. Changing the sort resets to page 1.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Marking/unmarking a mining address triggered a long history reload: it called
invalidateShieldedHistoryScanProgress() + forced a transaction refresh, which
re-scans every z-address over many RPC cycles. But "mined" vs "receive" is a
pure function of the LOCAL mining-address set — the daemon knows nothing about
it — so a chain re-scan is pointless.
Relabel the affected rows in the in-memory history directly and persist just
those to the encrypted SQLite history cache. The History tab updates instantly
(its display cache rebuilds on the type change), with no daemon round-trip and
no reload. Only re-save when something actually changed.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Unmarking (or marking) a mining address didn't change the history. The refresh
re-scanned the affected transaction as "receive", but appendMissingPreviousTransactions
carries over not-yet-rescanned prior transactions and dedupes by txid+TYPE — so
the stale "mined" copy was carried over right alongside the fresh "receive", and
the change never appeared.
Re-label state_.transactions in setMiningAddress() the moment the flag changes
(mined vs receive is just whether the receiving address is mining-flagged). The
History tab updates instantly, and the next refresh's carry-over now matches the
fresh scan instead of duplicating the old label. The reclassified list is also
persisted via the existing cache save.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
History streams in over many refresh cycles (the incremental shielded scan
walks every z-address), so the first batch appears long before the list is
complete — with no indication more is still coming. The existing loading banner
deliberately goes quiet once any rows are on screen.
Track whether the first full shielded scan has finished
(initial_history_scan_complete_) and, until it has, surface a progress percentage
(fraction of z-addresses scanned) in transactionRefreshProgressText() — which the
History tab already renders as its pulsing loading indicator. Goes quiet once the
first scan completes; routine per-block re-scans don't re-trigger it. Reset on a
full history invalidation (rescan / session reset) so it shows again on reload.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The Sent summary card showed 0 while selecting the Sent filter listed
transactions. The card counted only plain "send" rows and deliberately excluded
both legs of an autoshield pair as an "internal move", but the list shows the
merged "shield" row under the Sent filter. With only shielding transactions and
no plain sends, the card read 0 against a non-empty Sent list.
Count each shield pair toward the Sent card (with the shielded receive-leg
amount, which is what the merged row displays), so the card and the filter agree.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
History looked unsorted because every merged "shield" row carried
confirmations=0, and the list sorts 0-conf (pending) transactions to the very
top. So long-confirmed shielding transactions floated above newer ones — and
when the type filter was switched off "All" they vanished (shield rows only
match the "Sent" filter), which read as "transactions disappear when sorting".
Root cause: the autoshield merge set the row's confirmations to
min(send, recv). Both legs are the SAME transaction (one real confirmation
count), but the send leg (parsed from z_viewtransaction) routinely arrives with
confirmations=0, so min() picked 0. Use max() to take the populated value.
Also give the sort a txid tiebreak so same-block transactions keep a stable
order instead of reshuffling every time a new block bumps confirmations.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A rescan ran to completion but the status bar stayed at "Rescanning 99%"
forever. The daemon-output parser only treated "Done rescanning"/"Rescan
complete" as finished, but this daemon prints neither — it logs the rescan
benchmark timing line exactly when the scan ends:
2026-... rescan 16760577ms
then resumes normal block processing. So the parser saw the last
"Still rescanning ... Progress=0.99" and never the finish, leaving it stuck.
- Recognise the " rescan <N>ms" bench line as completion (it ends in "ms",
which the "Still rescanning"/"Rescanning..." progress lines never do).
- When the parser reads "Still rescanning" straight from the daemon log, mark
rescan_confirmed_active_ — hard proof the scan is running that doesn't depend
on catching a getrescaninfo warmup error, so the RPC completion path can also
fire after the daemon leaves warmup. Clear it on finish.
The parser reads the daemon's debug.log via the controller (not RPC), so this
completes the rescan UI even if the RPC connection hasn't re-established yet.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
While the daemon processes -rescan it sits in RPC warmup and rejects every call
with -28 ("Rescanning..."). The balance/tx/address refreshes already skip warmup
(state_.warming_up), but the 1-second mining poll didn't — so getmininginfo fired
the whole rescan and flooded the log with "getMiningInfo error: Rescanning..."
(~680 entries in one capture).
Gate refreshMiningInfo() on !state_.warming_up like the other refreshes. The
getrescaninfo progress poll still runs (it's how the warmup/rescan is tracked).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clicking Settings → Rescan restarted the daemon with -rescan correctly, but the
progress poll fired "Blockchain rescan complete" the instant it was clicked,
then showed nothing for the entire (multi-hour) rescan — so it looked broken.
Cause: the very first getrescaninfo poll runs before the daemon has restarted
and hits the still-running pre-restart daemon, which answers rescanning=false.
The completion branch took that as "done", cleared the rescanning flag, and the
real rescan then ran invisibly. (Confirmed from a Windows debug-log capture: an
instant OK{"rescanning":false}, then ~6400 warmup errors over ~5h, all swallowed.)
Fixes:
- Gate completion on a new rescan_confirmed_active_ flag that's only set once we
actually observe the rescan running, so a pre-restart rescanning=false can't be
misread as completion.
- While the daemon is in -rescan RPC warmup it rejects every call with the live
phase as the message ("Loading block index..." -> "Rescanning..."). Treat that
as proof-of-progress: surface it as rescan_status and mark confirmed-active,
instead of silently swallowing it. The status bar keeps its animated
"Rescanning..." for the whole run, then reports complete when warmup ends.
- Read rescan_progress whether the daemon returns it as a string or a number
(the get<std::string>() would have thrown on a numeric field).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The opid poll called z_getoperationstatus(["opid"]) to check a specific
operation, but this daemon rejects the filtered form with "JSON value is not a
number/array as expected" (a UniValue error returned as an RPC error). The
poll's catch swallowed it, so every completed send stayed stuck on "Waiting
for operation" forever — confirmed via a Windows debug-log capture showing the
throw on every 2s cycle. The no-arg form works (verified in the console).
Call z_getoperationstatus with no arguments (returns ALL operations) and filter
to the opids we're tracking in parseOperationStatusPoll(). The parser now skips
any operation whose id isn't in the requested set, so unrelated/old operations
can't fire a spurious error toast or pollute send state. The stale-opid logic
is unchanged (the no-arg form still reports in-progress ops, so a genuinely
pending opid is never misread as stale).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A completed send could spin forever on "Waiting for operation (N)". Root
cause: onDisconnected() stopped fast_worker_ but kept the unique_ptr, so
onConnected()'s `if (!fast_worker_)` guard never restarted it — after the
first reconnect (daemon warmup, restart, any RPC blip) the fast lane stayed
dead for the whole session.
The opid poll was the only fast_worker_ user that posted to it directly with
no fallback, so it alone broke: its post() landed on a stopped thread, the
result MainCb never ran, opid_poll_in_progress_ stuck true, and the poll never
fired again — leaving the operation (already "success" on the daemon, with a
txid) untracked.
Two fixes:
- onDisconnected() now reset()s fast_worker_ after stop(), so onConnected
recreates and starts a fresh one (restores the fast lane for all its users,
not just the poll).
- the opid poll now falls back to worker_ when the fast lane isn't running,
matching every other fast_worker_ call site — defense in depth.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A prior change passed the user-selected fee to z_sendmany as a fixed-decimal
string (mirroring the recipient amount). But the daemon reads the fee param
with UniValue::get_real(), which rejects a string with "JSON value is not a
number as expected" — breaking every z_sendmany send (surfaced via the
address-to-address transfer feature).
Pass the raw double instead. get_real() parses it directly and accepts any
number notation (including the "5e-05" form of a small fee), so this is
correct for all fee values. The recipient "amount" stays a fixed-decimal
string on purpose — that field is parsed with ParseFixedPoint, which a
scientific-notation double would break.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Three layout fixes in the export-key modal, all symptoms of widths/heights
authored as raw pixels while text scales with the user's font setting:
- "Copy to Clipboard" no longer clips — the Show/Hide · Copy · QR buttons are
auto-width (size 0) so they always fit their label;
- those buttons now share one font, so Show/Hide matches Copy (was a smaller
toggle-button font);
- the read-only address and key fields are sized to the wrapped text instead
of a fixed 60/80px, removing the empty space below their value.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The overlay dialog's content child is AutoResizeY, but the glass card behind
it was drawn to a fixed viewport ratio — leaving a tall band of empty glass
below short dialogs (e.g. the key-export modal had a gap under its Close
button). Measure the rendered card height each frame and reuse it next frame
to draw the glass to the content; fall back to (and stay capped at) the ratio
so tall dialogs are unchanged and can't run off-screen.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
From a screenshot at a non-default font scale: the red WARNING box clipped its text
("...balance, but" cut off) because it used a fixed 80px child height while the text inside
scales with the font. Make the warning box auto-size to its content (ImGuiChildFlags_AutoResizeY)
so it never clips at any scale, and scale the address / key read-only field heights by
Layout::dpiScale() for the same reason. Complements the card-width scaling fix.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Every BeginOverlayDialog is passed a raw pixel card width (550, 620, …), but the fonts and
spacing inside scale with Layout::dpiScale() — which includes the user's font-size setting. At
any non-default scale the content outgrew the fixed card, so text overflowed the card edge and
elements misaligned. Scale the card width by dpiScale() (no-op at the default 1.0 scale) and clamp
it to the viewport so a large scale can't push it off-screen. Fixes all overlay dialogs at once.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Auto-clear: the Copy button now routes through App::copySecretToClipboard, so a copied
private/viewing key is wiped from the clipboard after ~45s (same protection as the seed) with
a "auto-clears" notice — instead of the raw SetClipboardText that left it indefinitely.
- QR: once the key is revealed, a Show/Hide QR toggle renders the key's QR inline (via the same
GenerateQRTexture/RenderQRCode widget the Receive tab uses) for scanning into another wallet.
The QR texture is cached, regenerated on key change, and freed on hide/close/dismiss; hiding the
key also hides its QR.
- Actions row tightened to Show/Hide · Copy · QR, and the key + QR texture are now cleared on any
dismissal (Close button, scrim click, Esc), not just the Close button.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The address list supported two drag gestures that collided: dragging a row onto another
transferred funds, dragging into a gap reordered. Since rows are contiguous, a reorder-drag was
almost always over another row, so it triggered a fund transfer instead of reordering.
Disambiguate by WHERE on the target row the drag is released (user's suggestion): the top/bottom
~30% edge bands = reorder (an insertion line is shown), the centre = transfer (the row highlights).
A zero-balance row or an off-row drop always reorders. Tooltip and i18n hint updated to match.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
First slice of decomposing balance_tab.cpp (3449 lines). The five rendering helpers used by every
balance layout — UpdateBalanceLerp, RenderCompactHero, RenderSharedAddressList (599 lines, the
drag-reorderable address list), RenderSharedRecentTx, RenderSyncBar — are moved verbatim into
balance_components.cpp. balance_tab.cpp is now 2680 lines.
Clean extraction: the helpers' interactive statics (drag/copy/hide/show) are function-local and
move WITH them; the only file-scope state they share is the balance-lerp animation values
(s_dispTotal/Shielded/Transparent/Unconfirmed) and s_generating_z_address, now non-static and
declared `extern` in balance_components.h (defined once in balance_tab.cpp, so both TUs share the
same objects). RenderCompactHero's default arg moved to the header declaration. The layouts (still
in balance_tab.cpp) call the helpers via the new header.
Verified: full-node + Windows + lite build (links cleanly -> extern state resolves), tests,
hygiene. This touches every layout's address list / recent-tx / hero / sync bar, so needs a
hands-on pass across the balance layouts before the next slice.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Final slice of decomposing mining_tab.cpp. The ~529-line "Mode toggle" section (SOLO | POOL
segmented control + pool URL/worker inputs) is moved verbatim into RenderMiningModeToggle().
mining_tab.cpp is now 311 lines (was 2628) — just the tab dispatch, thread-sync glue, benchmark
advance, section-budget setup, and four card calls.
State the toggle mutates is passed BY REFERENCE so behaviour is identical: the pool-mode flag,
the settings-dirty flag, and the pool URL / worker char[256] buffers (the text inputs write into
them) — passed as char(&)[256] references and named with their original identifiers so the body
stays byte-identical.
Verified: full-node + Windows + lite build, tests, hygiene. Audit #10 complete: the 2628-line
monolith is now five focused files (earnings, stats, controls, mode-toggle + the 311-line shell).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
BeginOverlayDialog dismisses on a click outside the card via IsMouseClicked (mouse-down). When
the dialog is opened by a button that fires on the same frame (e.g. the mining tab's
"Update miner…" button), that opening click is still registered as an outside-click, so the
dialog opens and instantly closes — it just "flashes". Skip the outside-click dismissal on the
frame the scrim window first appears (ImGui::IsWindowAppearing()); normal outside-click closing
is unaffected on every subsequent frame. Fixes all overlay dialogs, not just the xmrig updater.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The wallet is meant to ship xmrig embedded (HAS_EMBEDDED_XMRIG -> getXmrigPath() extracts it
at runtime), but build.sh only embedded a flat prebuilt-binaries/xmrig-hac/xmrig.exe — while
the published DRG-XMRig archive ships the binary inside a versioned subdir
(drg-xmrig-6.25.3-win-x64/xmrig.exe). So xmrig.exe was never present, HAS_EMBEDDED_XMRIG stayed
undefined, and the Windows wallet ran with no miner: "xmrig binary not found", pool mining and
the thread benchmark both fail.
build.sh now extracts xmrig.exe (flattened) from the matching win-x64 zip when a raw binary
isn't already staged, so the existing embed step fires. (Checks the extracted file rather than
unzip's exit code, which is non-zero when a glob matches nothing.) Verified: --win-release now
logs "Extracted xmrig.exe", stages it (6.7M), and defines HAS_EMBEDDED_XMRIG=1.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Third and largest slice of decomposing mining_tab.cpp. The ~843-line "Controls" card (CPU-core
grid + drag-to-select, mining start/stop button, benchmark + miner-update controls) is moved
verbatim into RenderMiningControls(). mining_tab.cpp is now 839 lines (was 2628 originally).
The most coupled section, so mutated state is passed BY REFERENCE — the benchmark
(ThreadBenchmark&), selected thread count (int&), and drag state (bool&/int&) — with local
reference aliases so the body stays byte-identical and interactions (drag, benchmark, start/stop)
behave exactly as before. Read-only context is passed by value/const; the compiler verified
const-correctness. Local statics inside the block moved with it.
Verified: full-node + Windows + lite build, tests, hygiene, no startup crash.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Second slice of decomposing mining_tab.cpp. The ~313-line "Hashrate + Stats" card (stat values +
hashrate chart / live-log view) is moved verbatim into RenderMiningStats(); mining_tab.cpp is now
1680 lines (was 1992 after slice 1, 2628 originally). Body byte-identical apart from a s_pool_mode
alias; the chart/log toggle statics (s_show_pool_log/s_show_solo_log) moved with the card, and the
log buffer was already a function-local static. No App dependency in this section.
Verified: full-node + Windows + lite build, tests, hygiene, clean smoke start. Pending hands-on
visual check before the next slice.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
First incremental slice of decomposing the 2628-line mining_tab.cpp monolith (one giant
RenderMiningTabContent function). The ~636-line "Earnings" card section is moved verbatim into
RenderMiningEarnings(); mining_tab.cpp is now 1992 lines and calls it with the immediate-mode
layout context as parameters (draw list, fonts, scale/spacing, glass spec, pool-mode flag).
Behavior-preserving by construction: the body is byte-identical (the only additions are a
`const bool s_pool_mode = poolMode` alias and a local scratch `buf` so the moved code keeps its
original identifiers). The earnings-filter static moved with the card it belongs to. The
compiler surfaced every enclosing dependency, which became explicit parameters.
Verified: full-node + Windows + lite build, tests, hygiene, clean smoke start. Pending hands-on
visual check of the Earnings card before extracting the next section.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The sync throttle (kSyncProfile) covers the core/transactions/addresses/peers timers, but
getmininginfo runs off the separate 1s Fast timer and so still polled ~every 5s during
sync — another cs_main contender slowing block connection. Skip it while syncing unless the
user is on the Mining tab or actively mining (where live stats are wanted). Completes the
"no RPC contention during sync beyond the 10s progress poll" goal.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The full-node wallet polled the daemon at the per-tab cadence regardless of sync state.
On the Peers/Network tab that meant getpeerinfo every 5s + core every 5s + a full
transaction scan on every new block — and blocks arrive fast during sync. Each of those
calls takes the daemon's cs_main lock, the same lock block connection needs, so the node
synced noticeably slower than on the lightweight Console tab (core 10s, no peer polling).
Make the refresh cadence sync-aware:
- RefreshScheduler::kSyncProfile {core 10s, transactions/addresses/peers disabled} is applied
to ALL tabs while state_.sync.syncing, and reverts to the per-tab profile when sync ends.
applyRefreshPolicy() picks the profile; update() re-applies it on the syncing<->synced
transition. This suppresses getpeerinfo and the per-block tx scan during sync (that data is
incomplete mid-sync anyway) — every tab now syncs as fast as Console.
- collectCoreRefreshResult(rpc, includeBalance): skip z_gettotalbalance (wallet lock + cs_main)
while syncing; only getblockchaininfo runs, which is also what drives sync-progress detection.
applyCoreRefreshResult already leaves the balance untouched when balanceOk is false.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- rpc_client::callRaw: a daemon error object is no longer assumed to carry a string
"message" — a malformed error now yields a clean "RPC error: <dump>" instead of throwing
a json type-exception from .get<std::string>().
- sendTransaction (full-node): add a single-flight guard so a rapid double-click can't issue
two z_sendmany before the first returns its opid. The lite path already guarded this; the
send form guards it in the UI, but the controller entry point now does too.
(#9 from the audit was mostly false positives on verification — all popen sites already
null-check and the xmrig download FILE* path has no throwing calls. The payment-URI
checksum idea was dropped: the send flow already checksum-validates the recipient before
broadcasting, and tightening the parser would reject the placeholder addresses the existing
test relies on; added a comment noting this is format-only by design.)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- rpc_client: wipe the plaintext "user:password" temporary with sodium_memzero after
base64-encoding it into the auth header (std::string doesn't zero its buffer on
destruction).
- connection: the auto-generated DRAGONX.conf holds rpcuser/rpcpassword in plaintext but
was written with the default umask (often world-readable 0644). Restrict it to owner
read/write after creation so another local user can't read the credentials.
- app: copying a seed phrase / private key to the clipboard now arms an auto-clear —
App::copySecretToClipboard() copies the secret and, after 45s, wipes the clipboard IF it
still holds that secret (compared via a stored hash, never the plaintext). Wired into the
lite first-run wizard's seed Copy and the Settings export-secret Copy, with a
"clipboard auto-clears in 45s" notice. pumpSecretClipboardClear() runs each frame.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Per-frame hot paths in the immediate-mode UI were allocating needlessly:
- Address filtering in the balance tab rebuilt a std::string filter per address AND
containsIgnoreCase() lower-cased two fresh copies per call — ~6×N allocations/frame
on large wallets. New util::containsIgnoreCase(string_view, string_view) is
allocation-free, and the filter is now built once outside the loop.
- Four duplicated "time ago" implementations (balance_tab_helpers, balance_recent_tx,
send_tab, transactions_tab) are consolidated into util::formatTimeAgo (localized long
form) + util::formatTimeAgoShort (compact "5s ago"), preserving each call site's exact
display style. Both use snprintf, no per-row string concatenation.
- The send-tab address-suggestion scan (a walk over the whole tx list) is memoized on the
typed text + tx count, so it no longer recomputes every frame while the user pauses.
New src/util/text_format.{h,cpp}; the two existing containsIgnoreCase/timeAgo definitions
now delegate to it. Added to both the app and test targets (test target also gains i18n.cpp,
which text_format's localized path needs).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two issues shared one root cause: the shielded-receive scan marked each z-address "scanned
at the EXACT current tip," but a new block (~36s on DRGX) advances the tip and invalidates
every prior per-address scan. A wallet with more z-addresses than one refresh cycle can
scan therefore never reached "all scanned at tip" — so shieldedScanComplete stayed false
and transactions_dirty_ stayed true forever, which (a) kept the history-refresh banner lit
and the full rescan churning every cycle, and (b) blocked maybeFinishTransactionSendProgress
(it waited on transactions_dirty_), leaving the send-progress indicator stuck on.
Fix 1 — completion tolerance. Add TransactionRefreshSnapshot::shieldedScanTipTolerance: an
address counts as fresh if its last scan is within N blocks of the tip (0 = old strict
behavior, so existing tests are unchanged). The app scales N with the z-address count
(2 + count/96, capped at 50), so a multi-block pass can COMPLETE before its earliest scan
goes stale. This also throttles full rescans to ~N blocks instead of every block —
transactions_dirty_ clears, the banner stops, and CPU/RPC churn drops. Already-fresh
addresses are skipped, so the per-block cost falls back to just the (cheap) transparent
listtransactions.
Fix 2 — send-progress gate. maybeFinishTransactionSendProgress() no longer waits on the
transaction history scan (transactions_dirty_ / Transactions job): the sent tx is already
shown via the optimistic pending insert, and the spend is reflected once the balance
refresh lands, so it now finishes on the address/balance signal alone.
Test: a tolerant snapshot skips recently-scanned addresses (shieldedAddressesScanned == 0,
shieldedScanComplete) while a strict one re-scans them.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The banner was driven by transactions_dirty_, which refreshTransactionData() sets to
!shieldedScanComplete. The shielded-receive scan marks each z-address "scanned at tip,"
but every new block (~36s on DRGX) advances the tip and invalidates all prior per-address
scans, so for a wallet with more z-addresses than the per-cycle budget (8 on History) the
scan can never catch the tip — shieldedScanComplete stays false, transactions_dirty_ stays
true, and the banner stayed lit indefinitely.
Decouple the user-facing banner from that perpetual background scan:
- A just-sent transaction being enriched still surfaces (the user is waiting on it).
- Once any history is displayed, stay quiet for routine background refreshes — new receives
still appear live as they're scanned.
- The loading banner now only shows during the genuine INITIAL load (nothing displayed yet)
and send enrichment.
This is a UI-visibility fix; the underlying per-block full shielded rescan (and the related
send-progress flag that maybeFinishTransactionSendProgress gates on transactions_dirty_) are
separate follow-ups.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The History tab rebuilt its entire display list on every render frame: indexing all
transactions by txid, merging autoshield send+receive pairs into "shield" rows, and
std::sort-ing the result — O(N log N) plus several heap allocations at ~60fps, only to
show one 50-row page. The data is already sorted newest-first by the refresh service,
so the per-frame sort was redundant on top.
Memoize the merged+sorted list, rebuilding only when the underlying transactions
actually change. The cache key is a cheap, allocation-free FNV-1a fingerprint over the
display-relevant fields (count, last update time, and each tx's confirmations /
timestamp / type+address first char) — a new block bumps every confirmation so the key
changes and we rebuild; otherwise (the common read/scroll case) the cache is reused.
Filtering, search, and pagination still run per-frame over the cached list (cheap linear
scans that depend on interactive state).
Also document that App::shouldRefreshTransactions() is block-height/dirty driven (not
interval-gated) — the Transactions timer only paces the check; the recent-poll handles
between-block mempool/unconfirmed deltas.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The 1.3.0 lite-capability work gated isUsingEmbeddedDaemon() on the compile flag
DRAGONX_ENABLE_EMBEDDED_DAEMON (in 1.2.0 it was hardcoded true, so the daemon
always launched). The lite branch in CMakeLists set that flag OFF with
`CACHE BOOL ... FORCE`, which POISONS the build dir's cache: a later full-node
reconfigure of the same dir keeps the forced-OFF value (the full-node branch
never re-asserts it), so embeddedDaemonAvailable=false and the wallet extracts
dragonxd but never starts it — exactly the reported "unpacks dragonxd.exe but
does not start the daemon, manual start works."
Note the two gates are independent: the binary is EMBEDDED/extracted via build.sh
(HAS_EMBEDDED_DAEMON), while LAUNCHING is gated by DRAGONX_ENABLE_EMBEDDED_DAEMON
— so they diverged (extract yes, launch no).
The forced cache write was also pointless: makeWalletCapabilities() already
forces the embedded-daemon capability off for any lite build via
`fullNodeBuild && embeddedDaemonCompiled`, so lite never launches a daemon
regardless of the flag.
Fix:
- CMakeLists: remove the FORCE cache poisoning (the root cause).
- build.sh: set DRAGONX_ENABLE_EMBEDDED_DAEMON explicitly per variant (ON for
full-node, OFF for lite), mirroring the existing DRAGONX_BUILD_LITE handling,
so an already-poisoned build dir is HEALED on the next build rather than
silently keeping the stale OFF.
Verified: a poisoned Windows cache (=0) flips to =1 on reconfigure; full-node
builds define =1, lite =0; tests + hygiene green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The connection probe (getinfo) used a 30s request timeout, so when something on
the local RPC port accepts TCP but never answers — a daemon still loading the
block index, or a wedged/foreign occupant — every attempt blocked the full 30s
before the wallet could retry or update its status. That is the "stuck, timing
out every 30s" behaviour users hit.
A healthy local daemon answers getinfo in milliseconds, and a warming one
returns -28 just as fast, so a long hang on localhost only ever means trouble.
Probe localhost with an 8s timeout (remote/TLS keeps the 30s budget). The
per-call override restores the persistent 30s afterwards, so normal RPC calls
that legitimately take longer are unaffected — only the probe fails faster, so
the wallet retries promptly and reflects "initializing" / recovery within
seconds of the daemon becoming ready.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The daemon can spawn successfully (CreateProcess OK) and then exit immediately —
a missing runtime DLL, wrong architecture, corrupt binary, datadir lock, etc.
EmbeddedDaemon's crash monitor already builds a detailed reason for this
(translated Windows exit code, e.g. "STATUS_DLL_NOT_FOUND — required DLL not
found", plus the launch command and a debug.log tail) and stores it in
lastError(), but it runs on a background thread and was never shown. The result
was the exact symptom users reported: the wallet unpacks dragonxd.exe, looks
"stuck connecting", and the node silently dies-and-respawns until it gives up —
with no visible reason (manually starting dragonxd works, so the wallet then
connects to it).
tryConnect now watches the daemon's crash count (on the main thread, where it
already logs daemon state) and surfaces each NEW crash's lastError() once, as a
sticky error notification, with a concise "Couldn't start dragonxd" status. The
counter resets on a successful connect (alongside the daemon's own crash-count
reset), so a later crash re-notifies.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two failure modes left the wallet stuck on a silent "connecting / Starting
dragonxd…" spinner with no path forward:
1. Stale external-daemon latch. EmbeddedDaemon::start() sets
external_daemon_detected_ whenever the RPC port was busy at a prior attempt
and never re-checks it, so tryConnect's no-config branch trusted that latch
and waited forever for a config the phantom would never write — even after a
stale/half-dead process freed the port. Now the port is re-evaluated LIVE
(EmbeddedDaemon::isRpcPortInUse()) each attempt: if it's genuinely busy we
keep waiting (and, after a bounded ~20s with no config, warn that whatever
owns the port isn't a usable DragonX node and how to fix it); if it's free we
fall through and start our own daemon.
2. Silent start failure. When startEmbeddedDaemon() failed (binary not found,
Sapling params missing, spawn failure) the status stayed on "Starting
dragonxd…" with the real reason only in a VERBOSE log. Now the reason
(daemon_controller_->lastError()) is surfaced once as a sticky error
notification, with a short "Couldn't start dragonxd" status.
Both counters reset on a successful connect so the messages re-arm for the next
disconnect. Lite is unaffected (tryConnect returns early for lite builds).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The lite backend's litelib_execute() is the same command interface as
silentdragonxlite-cli (balance, info, height, list, notes, addresses, sync,
syncstatus, new, send, shield, encrypt, …), so the lite Console can be a real
interactive console — like the full-node RPC console — instead of a read-only
diagnostics log.
Controller: add an async arbitrary-command runner mirroring the broadcast
pattern — runConsoleCommand() splits "<command> [args]" (the first token is the
command, the remainder is passed as the single arg string litelib_execute
expects, since it does NOT whitespace-split), runs the bridge call on a detached
thread that captures the shared bridge (never `this`), and delivers the result
to a main-thread slot drained by takeConsoleResult(). Results are NEVER routed
through LiteDiagnostics (seed/export can return secrets).
Console tab: a command input (Enter to run, Up/Down history via the shared
console_input_model helpers) over a unified scroll buffer that interleaves the
automatic diagnostics events with user command I/O, colour-coded, with the live
status header preserved. The input is disabled while a command runs.
Two backend footguns are intercepted at the UI layer before forwarding:
`clear` (the backend command WIPES wallet tx history — re-bound to clearing the
view, what the user expects) and `quit`/`exit` (would only save; the embedded
backend must stay running with the app).
Test: runConsoleCommand drives the fake backend (info -> raw response; "new zs"
-> exercises the command/arg split; blank line rejected).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The full-node Console tab already streams the daemon's output, but during
startup the user is held on the loading overlay (wallet-data tabs are blocked),
so they can't watch progress without navigating away. Surface the last few
console lines the node printed (UpdateTip height=…, "Verifying blocks…", etc.)
directly under the status/description on the overlay while initializing or
warming up, so progress is visible where the user is already looking.
Full-node only (guarded on daemon_controller_); each line is trimmed and
ellipsis-truncated to one row. Reuses DaemonController::recentLines().
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
When the full-node connect probe (getinfo) times out, the daemon is reachable
at the TCP level but busy initializing (loading the block index, verifying,
activating best chain, …) and won't answer RPC. The wallet only recognized the
JSON-RPC -28 warmup reply, so a raw socket timeout fell through to a bare,
alarming "Connection failed" retry with no indication of what the user was
waiting on.
Add a daemon-initializing UI state that drives the existing loading overlay:
- WalletState::daemon_initializing — daemon up/launching but not serving yet
(distinct from warming_up, which needs a -28 reply).
- App::applyDaemonInitStatus() infers the current phase from the daemon's own
console output (scanning recent lines for Loading/Verifying/Activating/
Rescanning/Rewinding/Pruning) and the latest block height, producing a
friendly title + description, e.g. "Processing blocks… (Block 123456)".
- The connect loop calls it from the daemon-starting and external-detected
branches: a timeout -> "reachable but initializing", a connect refusal ->
"launching, waiting to come online". Cleared on a real connect.
- The loading overlay now shows the description for daemon_initializing too,
and the status-bar amber indicator covers it (so Peers/Console tabs without
the overlay still explain the wait).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The Settings page drove the controller's synchronous createWallet/openWallet/
restoreWallet, which blocks the UI thread on the (often flaky) lightwalletd and
gives up after the first server. Add a generic async lifecycle path that mirrors
the async-open failover but carries the full request (passphrase, restore seed/
birthday/account/overwrite):
- beginCreateWalletAsync / beginOpenWalletAsync / beginRestoreWalletAsync run
on a detached thread that builds its OWN local LiteWalletLifecycleService
from captured value copies + the shared bridge (never `this`, so it can
safely outlive the controller). Each request type's serverUrl override field
feeds the failover: try the preferred server, then every other usable
default; stop on the first ready wallet or a structural block; keep the
preferred server's error on total failure. The request's secrets are wiped
once the attempt finishes.
- pumpLifecycleResult() finalizes on the main thread (flip walletOpen, persist,
start sync) and caches the result for the UI; wired into App::update next to
pumpAsyncOpen(). beginAsyncLifecycle() now also yields to an in-flight
lifecycle request so the auto-open loop can't race it on the same bridge.
- settings_page kicks off the async op, disables the button while in flight,
and polls the cached result each frame for the status/summary.
Tests: testLiteWalletControllerAsyncLifecycleFailover covers async create (with
passphrase) and restore failing over preferred->fallback, plus all-servers-down.
Co-Authored-By: Claude Opus 4.8 <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>
- Restore: live "N / 24 words" count, a one-line birthday explanation, and a guard
that rejects a restore unless all 24 words are entered (the secret scrubber still
wipes the input on the early return).
- Backup: "Show seed" now also shows the birthday (needed to restore quickly) with a
"back this up too" note, a stronger "only way to restore" warning, and a "Save to
file" button that writes the seed + birthday to an owner-only (0600) file in the
config dir via the atomic-write helper.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Creating a wallet was one-click and silent — it never showed the seed, relying on
the user to later find Settings -> Show seed, which is an easy-to-miss fund-loss
risk. Replace the first-run prompt with a 3-step guided flow mirroring the upstream
SilentDragonXLite wizard:
1. Welcome (Create / Restore / Later) — unchanged entry.
2. Reveal: after createWallet, read the seed back via exportSeed and show all 24
words (numbered grid) + the birthday, with a strong "only way to restore"
warning, plus Copy. ("Skip" leaves the wallet created, seed backable later.)
3. Verify: tap the words in order (shuffled chips) to confirm the backup before
finishing; out-of-order taps are rejected with a hint.
The seed is held only for the wizard and securely wiped (sodium_memzero) on finish.
Builds clean for full-node, lite, and Windows cross-compile.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The failover overwrote outcome.error on every attempt, so a total open failure
reported whichever (often broken) fallback was tried last — e.g. lite5's
"CertNotValidForName" — instead of what the user's preferred server actually did.
Keep the first (preferred) server's error as the summary so "Open failed: …" names
the actionable reason; the per-server attempts are still in the Console log, and
the warmup flag is still set if any server was warming up.
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>
A lite build compiled without the SDXL backend (DRAGONX_ENABLE_LITE_BACKEND off,
i.e. built with --lite instead of --lite-backend) leaves the controller null, so
the wallet never opens and the UI shows a silent "disconnected" state. The Console
status now states the cause and the fix directly ("Lite backend not linked in this
build (rebuild with --lite-backend)") instead of a vague "unavailable".
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The Console could look empty if the wallet produced few events. Make it useful
in every state and remove a cross-platform footgun:
- Add a live status header read straight from the controller (connected /
connecting / disconnected, sync %, and the last open error) — independent of the
diagnostics event log, so the Console always shows the current connection +
wallet-open state even when the log is sparse.
- Move LiteDiagnostics::instance() into a single .cpp so there is exactly one
instance across the binary, rather than relying on the linker folding an
inline-function static across translation units (a known fragility, especially
on mingw/Windows — the most likely cause of a stuck-empty event log there).
Verified the writer and reader share one instance on Linux; builds clean for
full-node, lite, and Windows cross-compile; tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The lite variant had no visibility into why a wallet failed to open — just a
"disconnected" spinner. Add a lite-only Console tab (full-node keeps its RPC
console) that shows a live diagnostic log.
- LiteDiagnostics: a small thread-safe, bounded ring buffer (header-only). The
controller writes to it from its background threads: each failover server
attempt and result, wallet open/create/restore outcomes, sync start, and
blocked-open reasons. The App logs controller (re)builds with the preferred
server.
- lite_console_tab: a terminal-styled, read-only view of the log (newest at the
bottom, error/success lines coloured) with Clear / Copy / Auto-scroll. Reachable
even when the wallet is locked (it's diagnostics, no secrets). Registered as
NavPage::LiteConsole, gated lite-only via WalletUiSurface::LiteConsole.
A unit test drives an open-with-failover and asserts the log records the
connection attempt and the successful open. Built clean for full-node, lite, and
Windows cross-compile.
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>
The wallet auto-open is a one-shot (lite_autoopen_done_), but rebuildLiteWallet()
creates a fresh, closed controller — so switching the lite server from the Network
tab (rebuildLiteWallet force=true), or any later rebuild, left the wallet
permanently closed ("disconnected" spinner) because auto-open never fired again.
Re-arm the one-shot (and clear the surfaced open-error) in rebuildLiteWallet so
the next update() tick reopens the existing wallet against the new server. This is
the recovery path when the configured lightwalletd server is unreachable: the
Network tab surfaces the failure reason, the user picks a reachable server, and
the wallet reopens. Also makes the Network tab's apply-immediately server switch
actually take effect.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Both variants hardcoded "ObsidianDragon" as the per-user config folder
(settings.json, themes, the lite_rollout cache), so the lite app and the
full-node app shared one settings.json. That cross-variant pollution can leave
the lite server selection in a bad state — and since openWallet() contacts the
selected lightwalletd server, a wrong/empty server URL there makes an existing
wallet fail to open (a silent "disconnected" spinner).
Use DRAGONX_APP_NAME (already "ObsidianDragon" / "ObsidianDragonLite" per variant)
for the config-dir name in Settings::getDefaultPath, Platform::getConfigDir and
getObsidianDragonDir (and the theme-setup exe-name probe). Full-node is unchanged;
lite now reads/writes %APPDATA%\ObsidianDragonLite (and ~/.config/ObsidianDragonLite),
so it starts from a clean, isolated config and uses default servers.
Note: the lite wallet file itself lives in the litelib backend's own data dir
(unaffected); this isolates the GUI config only.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The startup auto-open of an existing lite wallet discarded openWallet()'s result,
so when initialize_existing failed (e.g. the lightwalletd server is unreachable)
the UI just showed a "disconnected" spinner with no reason — and DEBUG_LOGF is
compiled out of release builds, so there was no way to see why. Capture the
failure: store the reason, show it in the Network tab status line (in place of
"no wallet open"), and raise a notification. Cleared once a wallet opens.
This doesn't change open behaviour — it makes a stuck open diagnosable.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
stop()-ing a worker that is mid curl_easy_perform joined on the UI thread, so a
slow/hung transfer froze the UI until the request timeout. Add RPCClient::
requestAbort() (a thread-safe atomic read by a curl progress callback that aborts
the transfer), and call it before stopping the workers on disconnect
(onDisconnected) and shutdown (beginShutdown + the synchronous fallback). The
flag is cleared on each connect() so a fresh connection never starts aborted.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The send screen labelled any prefix+length match as a "Valid" address, so a
mistyped address that still matched the pattern passed the gate. Add pure,
offline checksum validation — Base58Check (transparent R-addresses) and Bech32
(Sapling zs-addresses) — and require it in the validity check. Both verifiers are
version-byte/HRP agnostic (the HRP is taken from the string, the Base58 checksum
is chain-independent), so a correct implementation never rejects a genuine
address while catching transcription errors. Works for both build variants
(no daemon round-trip), unit-tested against standard BIP173 / Base58Check vectors.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
removeVault() overwrote vault.dat with zeros then unlinked it, but never flushed
to stable storage, so the zeros could stay in the OS cache and never reach disk.
flush + fsync before unlink on POSIX (still best-effort on CoW/SSD, but now does
what it claims).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
resetJobs() cleared the in-progress flags but left generations_ untouched, so a
refresh WorkFn still executing on the worker when a disconnect cleared state_
could pass completeDispatch's generation check and apply last-connection data
onto the new session. Bump every job's generation in resetJobs() so any
pre-reset ticket is treated as stale.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- importKey routed transparent vs. shielded purely by the first character, which
can mis-route (e.g. testnet/regtest WIFs). On failure, try the other import
command before reporting an error (each validates the encoding, so a wrong
command rejects rather than mis-imports). The key copy is wiped after both tries.
- Clamp the shield dialog's fee input to [0, 1] DRGX, mirroring the UTXO-limit
clamp, so a negative or fat-fingered huge fee can't be submitted.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Balance card hashrate now uses the shared FormatHashrate() (TH/GH/MH/KH/H)
instead of a bespoke two-tier KH/s formatter.
- Recent-tx rows show the full untruncated address on hover — two z-addresses can
truncate to the same first/last window — and the truncate helpers guard maxLen<=3.
- Remove the unused viewTop/viewBot "viewport culling" locals in the tx list
(pagination already bounds per-frame work).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The transaction-status overlay decided error vs. success styling by searching the
status string for "Error"/"Failed" — so under a non-English locale a failed send
rendered as a green success. Drive it from the existing s_status_success flag
instead. Also show the USD-mode DRGX preview at 8 dp so it matches the confirm
panel and the amount actually sent (was 4 dp).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- testAtomicFileWrite: Platform::writeFileAtomically creates dirs, overwrites,
leaves no .tmp, and honors owner-only perms.
- failureByOpid assertion in the operation-status poll parser test.
- testTransactionHistoryCachePrunesOldWallets: a save under a new identity prunes
the prior identity's snapshot.
- testLiteSendShowsRecipientFromOutgoing / testLitePartialRefreshKeepsPriorAddressBalances.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Many strings are used directly as printf/ImGui format strings, and translations
are loaded from user/installer-modifiable JSON with no validation. A translated
value that drops or changes a conversion specifier would be passed to printf with
mismatched varargs (undefined behavior) on a wallet screen.
overlayTranslations() now compares each translated value's argument signature
against the English source and keeps English on mismatch. Also adds the
send_status_unconfirmed string used by the deferred-send-result path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
GetAvailableBalance() read state.addresses[s_selected_from_idx], but the index
desyncs from s_from_address (the value actually debited) after an address-list
refresh, and is left at -1 when the source is chosen from another tab's "Send
from this address" — which made the sufficiency check see a 0 balance and wrongly
block a valid send. Look the balance up by matching the source address string.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- A Send record carries its recipient in outgoing_metadata, not the top-level
address/memo, so sent txs showed a blank destination + memo. Surface the first
recipient (single-recipient case) into the transaction list.
- A tolerated partial refresh where the notes/utxo command failed (addresses
present, spendable outputs absent) zeroed every per-address balance, which looks
like fund loss. Preserve the last-known per-address balances in that case.
- Retry the post-send/shield save once on transient failure instead of ignoring
the result (the backend does not auto-save after send/shield).
- An unparseable broadcast response now uses cautious wording ("status could not
be confirmed — check Transactions before retrying") rather than implying a hard
failure, avoiding a blind double-spend retry.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Write vault.dat atomically and 0600 (it holds the PIN-encrypted passphrase, so
a world-readable copy enables an offline brute-force of the short PIN), and
chmod the tx-history SQLite + its WAL/SHM sidecars to 0600 on open.
- The tx-history snapshot and key-salt rows are keyed on a hash of the full
address set, which changes whenever a new address is generated — orphaning the
prior hash's full-history blob and salt forever. pruneOtherWallets() now drops
rows for every non-live wallet hash on each save, bounding the database.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Market chart now plots the real accumulated price_history instead of a
rand()-generated curve; the hover tooltip no longer claims a specific "Xh ago"
price and the x-axis only labels the truthful "Now" point. Falls back to the
existing empty state until there are >=2 real samples.
- Transactions summary cards exclude autoshield legs (same txid send + receive-to-z)
so a shield isn't double-counted into both Sent and Received, matching the list.
- Send/Receive sync banners use verification_progress like every other surface,
instead of the blocks/headers ratio that over-reports during early sync.
- Fix printf format/type mismatches: %.0f<-int (market % shielded), %d<-size_t
(peer counts), %ld<-int64_t (peer byte counters, wrong on Windows).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The connection state machine never tore down on a lost connection: refresh-loop
RPC errors were swallowed, rpc_->isConnected() stayed true after a daemon
crash/restart/socket drop, and the UI showed stale balances with no reconnect.
Several operations also ran synchronous curl straight from ImGui handlers.
- Add handleLostConnection(): after N consecutive cycles where BOTH core RPCs
fail (warmup excluded, so no reconnect loop), disconnect so update()'s
reconnect branch re-enters tryConnect().
- Move banPeer/unbanPeer/clearBans and key export/import onto the worker thread
(import requests a rescan that could freeze the UI for the curl timeout).
- Run the block-info dialog's two chained RPCs on the worker thread (+ guard the
getblockhash result type).
- Detect daemon warmup via the JSON-RPC -28 code (new RpcError carrying the code;
message text preserved so 401/warmup string-matching is unaffected), and widen
CONNECTTIMEOUT to 10s for remote/TLS hosts (2s localhost).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
z_sendmany returns an opid immediately; the tx is built/signed/broadcast
asynchronously afterward. The send path showed "Transaction sent successfully!"
and cleared the form on opid receipt, so a later async failure contradicted it.
Shield/merge stored the opid only in a dialog-local static (never polled), and
auto-shield ran a blocking z_shieldcoinbase on the UI thread and discarded its
opid — async failures of all three were silently lost.
- Add App::trackOperation(opid) so shield/merge/auto-shield register with the
shared opid poller (failures surface, balances refresh on completion).
- Defer the full-node send's success/failure to the poller via per-opid callbacks
(parseOperationStatusPoll now exposes failureByOpid); the "Sending..." spinner
covers the finalizing window, and the form is kept until terminal status.
- Dispatch auto-shield through the worker thread and use the configured fee.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
settings.json and addressbook.json were written in place with a bare ofstream —
a crash or power loss mid-write truncated the file, and on the next launch the
parse failure silently reset every preference (hidden/favorite addresses, labels,
pool workers, language, theme, lite-server list) because the next save overwrote
the corrupt file with defaults.
Add Platform::writeFileAtomically() (temp file -> fsync -> atomic rename; dir
fsync on POSIX, MoveFileEx on Windows; optional owner-only 0600) and route both
saves through it. On a parse failure, quarantine the unreadable settings file to
settings.json.corrupt-<ts> instead of clobbering it.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The full-node send built the recipients array and called z_sendmany with only
(fromaddress, amounts) — dropping the minconf and fee positional args. The whole
fee-tier UI (Low/Normal/High, send-max math, the confirmation fee) was collected
and shown but never sent, so the daemon silently applied its own default fee and
the Low/High tiers were cosmetic.
Pass {from, recipients, 1, fee}, with the fee formatted fixed-decimal so the
daemon's ParseFixedPoint accepts it (a small double like 0.00005 would otherwise
serialize to "5e-05" and be rejected).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a status panel at the top of the Network tab driven by the live WalletState:
- Connection: a colored dot + Connected / Syncing / Not connected, with the in-use server host
(or "Random server") and its latency on the right.
- Sync: "<pct>% · <walletHeight> / <chainHeight>" while syncing (with a thin progress bar),
"Synced · block N" when complete, or "No wallet open" when disconnected.
Reads app->state().sync (populated by the lite refresh: progress / wallet+chain height / complete)
and state().connected (= walletOpen). Advances with a Dummy so the bounds grow correctly.
Both variants build; suite passes; hygiene clean; lite GUI smoke OK.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Each server card advanced to the next via a bare SetCursorScreenPos, which ImGui won't use to
extend the scroll region's content height ("Code uses SetCursorPos() to extend window boundaries
... submit an item e.g. Dummy() afterwards"). Beyond the warning, this meant cards past the fold
wouldn't scroll. Advance with an ImGui::Dummy(cardW, gap) below each card so the content height
grows correctly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A lite-wallet-only "Network" tab (full-node keeps the Peers tab; exactly one shows per variant)
to manage lightwalletd servers, replacing the basic selector that was in Settings.
- Card list of servers with per-server latency + status dot, DNS host + resolved IP, and an
Official/Custom pill. Official DragonX servers get a glowing outline.
- Pick a server (Sticky) by clicking its card, or toggle "use a random server" (Random mode);
selection applies immediately (App::rebuildLiteWallet(force=true) tears down + rebuilds the
controller against the new server and resyncs — its dtor detaches the uninterruptible sync
thread, so this doesn't block).
- Add custom servers; hide/unhide servers (persisted set, revealed by a "Show hidden" toggle).
- Latency/IP come from a new background probe (util/LiteServerProbe): libcurl CONNECT_ONLY does
the TCP+TLS handshake (works for gRPC lightwalletd, no HTTP response needed), recording
APPCONNECT_TIME as latency and CURLINFO_PRIMARY_IP. Auto-runs on tab open + a Refresh button.
Wiring: WalletUiSurface::LiteNetwork (gated !fullNodePagesAvailable) + NavPage::LiteNetwork in
the sidebar + app.cpp dispatch; settings gains a hidden-servers set; isOfficialLiteServer() added
to lite_connection_service. The Settings page lite-server selector + its plumbing are removed
(single source of truth = the tab).
Reuses the existing server model (LiteServerPreference, Sticky/Random, selectLiteServer) and UI
primitives (DrawGlassPanel, ThemeEffects glow, peers-tab ping-dot idiom). Unit-tested
(liteServerHost, isOfficialLiteServer) + an env-gated live probe (verified vs lite.dragonx.is:
online, latency, IP). Both variants + lite-backend build; suite passes; hygiene clean; GUI
smoke-launched without crash.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The linux/windows release packaging did `rm -rf "$out"` on the whole output dir, so building
ObsidianDragonLite into release/<os>/ wiped the ObsidianDragon artifacts already there (both
variants share release/linux and release/windows). Remove only the CURRENT variant's prior
artifacts (by APP_BASENAME, which can't cross-match — "ObsidianDragon-*" excludes
"ObsidianDragonLite-*"), so full-node and lite releases coexist.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Relocate the miner-update control from a standalone full-width button into the mining-control
header row, immediately left of the benchmark button:
- The button now shows the latest available version ("Update <tag>"), with the current installed
version as text to its left ("Current: <tag>" / "none").
- A one-shot background version check (util::XmrigUpdater::startCheck) runs the first time the pool
section is shown, so the latest tag can be displayed; until it arrives the button reads
"Update miner…". Clicking opens the existing dialog; disabled (greyed, with tooltip) while the
miner is running.
- New i18n keys: xmrig_update_short, xmrig_current, xmrig_none.
Both variants build; suite passes; GUI smoke-launched without crash.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The full-node app and ObsidianDragonLite are now versioned separately:
- project() VERSION -> 1.3.0 (suffix cleared); DRAGONX_LITE_VERSION -> 1.0.0.
- A DRAGONX_APP_VERSION* set (resolved per variant in the lite/full block) feeds the generated
header (version.h.in), the Windows VERSIONINFO/.rc + manifest, and the build summary — so each
variant reports its own version. The .rc/manifest name fields also follow DRAGONX_APP_NAME so a
lite .exe's properties read "ObsidianDragonLite".
- build.sh resolves the release-filename version per variant by parsing CMakeLists (single source
of truth) instead of a hardcoded string.
Also fixes a latent variant-bleed: build.sh now passes DRAGONX_BUILD_LITE and
DRAGONX_ENABLE_LITE_BACKEND explicitly (ON *and* OFF), so switching variants in a shared build dir
can't reuse a stale cached value (a prior --lite build was making a subsequent full-node build
produce the lite name/version).
Both variants build + report the right version (full 1.3.0, lite 1.0.0); suite passes.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a "Miner updater (xmrig)" section to CLAUDE.md: the update flow + verification
(TLS + archive SHA-256 + enforced ed25519 signature against a pinned key), and the
release-process consequence — every drg-xmrig release must be signed
(scripts/sign-xmrig-release.sh) with the .sig uploaded per archive, or the in-app
updater refuses it; the signing secret key stays offline (gitignored), only the base64
public key is pinned in source.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replace the English string literals in the miner-update dialog + the "Update miner…" mining-tab
button/tooltip with TR() keys, and register their English text in i18n.cpp's loadBuiltinEnglish()
(the in-code English fallback that non-English locales overlay). Reuses the existing cancel/close/
retry keys. Labeled values use a "%s %s" literal format with a TR'd label (no -Wformat-security
risk). Non-English locales fall back to English for the new xmrig_* keys until translations are
added to res/lang/*.json.
Both variants build; suite passes; hygiene clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Now that the release publishes a valid .sig per archive (verified against the pinned key for
linux/win/macOS), enable enforcement and fix two bugs that the newer multi-platform release
(v6.25.3, which added a macOS build) exposed:
- kXmrigRequireSignature = true: refuse any install whose release doesn't publish a valid
ed25519 signature over the archive. Verified live end-to-end against the signed v6.25.3
(archive SHA-256 + signature -> install).
- Drop the redundant inner-binary SHA-256 check. It keyed on the inner filename, but both the
linux and macOS archives contain a binary literally named "xmrig", so the two "xmrig (…)"
checksum lines collided in the map and the linux install compared against the macOS hash ->
spurious "could not verify" failure. The whole archive is already verified (SHA-256 +
signature), so every extracted member is authentic by transitivity — the per-member check
added nothing but ambiguity.
- Fix the macOS platform token: the asset is named "...-macos-x86_64.zip", not "...-macos-x64",
so selectXmrigAsset never matched it. currentXmrigPlatformToken() now returns "macos-x86_64"
on Intel macs (arm64 has no build -> Unavailable). Added a matcher test for the macOS naming.
Both variants build; suite stable (0 failures / multiple runs); live require-mode install verified.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Pin the ed25519 public key in xmrig_updater.h, activating signature verification in soft mode
(kXmrigRequireSignature=false): a release's ".sig" asset is verified when present, but an
unsigned release still installs on TLS + SHA-256. Verified live against the current release
(v6.25.2, which ships no .sig yet) — still installs.
- gitignore *.ed25519.key / *.ed25519.pub.b64 so a signing secret key can never be committed.
- Add a unit test that the pinned key decodes to a valid 32-byte ed25519 key (a malformed paste
fails the build, not silently disabling verification).
Bug fix (found via a flaky test): verifyXmrigSignature trimmed trailing whitespace BEFORE the
raw-64-byte check, so a raw signature whose last byte equals '\n'/'\r'/space/tab (~1.6% of
signatures) was corrupted and rejected. Now base64 is tried first (safe to trim) and the raw
path uses the exact untrimmed bytes. Added a deterministic regression test that forces a
whitespace-terminated raw signature. Suite is stable (0 failures in 10 runs; was ~3/8).
Also de-brittled the live integration test: it no longer pins a release-specific binary hash
(reaching Done already means the worker verified the binary against the release's own checksum).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Rewrite scripts/sign-xmrig-release.sh to use OpenSSL (>= 1.1.1) instead of PyNaCl, so signing
needs no Python deps. OpenSSL's ed25519 is PureEdDSA (RFC 8032) — interop-verified against the
wallet's libsodium crypto_sign_verify_detached (script-produced .sig -> VERIFY-OK; tamper ->
VERIFY-FAIL). keygen/pubkey/sign subcommands; emits base64 raw-64-byte signatures as <file>.sig.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Closes the supply-chain gap the review flagged: today the archive and its SHA-256 share one
trust root (the release body), so a compromised/edited release can ship an arbitrary binary
that still "verifies". This adds authenticity via a detached ed25519 signature checked against
a public key PINNED IN THE BINARY (not fetched), using libsodium's crypto_sign_verify_detached.
Opt-in / soft rollout:
- kXmrigSignaturePublicKeyBase64 in xmrig_updater.h is EMPTY by default -> signatures are not
checked and behavior is unchanged (TLS + SHA-256 only). Paste the base64 public key to enable.
- Once a key is pinned, an install verifies a "<archive>.sig" asset (base64/raw 64-byte ed25519
signature over the archive bytes) when present; kXmrigRequireSignature=true additionally
refuses installs that publish no signature.
- The check runs after the SHA-256 check, over the same already-read archive bytes; refuses on
a missing key-but-required, unreachable .sig, or invalid signature.
- verifyXmrigSignature + selectXmrigSignatureAsset are pure (libsodium only) and unit-tested:
valid base64 + raw-64-byte signatures verify; tampered data, wrong key, and malformed/empty
inputs all fail closed. Cross-tool interop verified (Python stdlib base64 == sodium base64).
- scripts/sign-xmrig-release.sh: keygen / sign / pubkey helper (PyNaCl = same libsodium ed25519)
to produce the .sig assets and the public key to pin.
No behavior change until a key is pinned. Both variants build; suite passes; live worker
re-verified (signatures off by default).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Addresses confirmed findings from the multi-lens review of the updater:
- Cancelable + live progress (was: download uncancelable, progress stuck at 0%, closing
the dialog mid-download blocked the UI thread on the worker join). Wire a libcurl
CURLOPT_XFERINFOFUNCTION that publishes byte counts and returns abort when cancel() is
requested; add a Cancel button. The dialog's destructor now aborts the transfer promptly,
so closing mid-download no longer freezes the UI.
- Graceful "unavailable" instead of a red error on platforms with no published build
(macOS / ARM): new terminal State::Unavailable rendered neutrally, not as a failure.
- Install-time running guard (TOCTOU): App::isPoolMinerRunning() re-checked in the dialog
before each install, so a dialog opened before mining started can't replace a live binary.
- Size caps: CURLOPT_MAXFILESIZE on the download and a per-archive-member ceiling before
decomphressing into memory, to bound an attacker-controlled archive.
- Distinguish a local read failure of the downloaded archive from a checksum mismatch
(was reported misleadingly as "possible tampering").
- Reword the dialog's verification note to "checked against the release's published SHA-256
checksum" (integrity, not authenticity — see the signing note below).
Not fixed here (needs your input): WinRing0x64.sys has no per-file hash published, but it is
covered by the verified archive checksum (it is inside the verified zip); and the release is
not cryptographically signed — checksums and binary share one trust root. Adding a pinned-key
ed25519/minisign signature is the real supply-chain hardening and needs an offline signing key
+ a release-process change.
Both variants build; suite passes; live worker re-verified end-to-end on linux-x64.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Wires util::XmrigUpdater into the GUI:
- ui/windows/xmrig_download_dialog.h: a modal (mirrors BootstrapDownloadDialog) that drives
the updater — Checking -> Up-to-date/Update-available -> Downloading/Verifying/Extracting ->
Done/Failed, with a progress bar and a "verified against its published checksum" note. On
success it persists the installed release tag to settings. Rendered each frame from App::render.
- mining_tab: an "Update miner…" button in the pool section, disabled (with a tooltip) while
xmrig is running so a live binary is never replaced.
- settings: persist the installed DRG-XMRig tag (xmrig_version) for update detection.
Both variants build; suite passes; GUI smoke-launched without crashing.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds util/XmrigUpdater: a background-thread service (mirrors util/Bootstrap) that pulls
the latest DRG-XMRig release from the project's Gitea, verifies it, and installs the miner
binary into the daemon directory. Service layer only; the mining-tab UI hook comes next.
Flow: GET /api/v1/repos/DragonX/drg-xmrig/releases/latest -> pick the asset matching this
platform (…-linux-x64.zip / …-win-x64.zip; no macOS build -> graceful "unavailable") ->
download (libcurl, TLS verified) -> verify the archive SHA-256 -> extract with miniz,
flattening the versioned subdir the archive nests the binary in -> verify the extracted
binary's SHA-256 in memory before writing it -> atomic install (+chmod +x on POSIX). On
Windows also extracts WinRing0x64.sys; config.json/README.md are skipped.
Security (download-and-execute): TLS is verified, and BOTH the archive and the inner binary
are checked against the SHA-256 checksums published in the release body (parsed as
"<hex> <name>" lines) — install is refused on a missing or mismatched checksum.
Split into a pure core (xmrig_updater_core.cpp: release parse, asset/platform match, checksum
parse, SHA-256) and the curl/miniz worker (xmrig_updater.cpp). The core is unit-tested against
a real captured release fixture (tests/fixtures/xmrig/release_latest.json); an env-gated
(DRAGONX_TEST_NETWORK=1) integration test exercises the worker live and was verified end-to-end
on linux-x64 (inner binary SHA-256 matches the published value). Both variants build; suite passes.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The mac-release path was mostly ObsidianDragon-hardcoded, so `--lite --mac-release`
would produce a broken bundle. Make it variant-aware, mirroring the linux/win lite
handling that already keys off APP_BASENAME + should_bundle_full_node_assets:
- SDL3 rpath fix, the launcher script + its .bin pair, and CFBundleExecutable now
follow ${APP_BASENAME} (ObsidianDragonLite), so the bundle's executable resolves.
- Lite variant gets its own CFBundleName/CFBundleDisplayName ("DragonX Wallet Lite"),
CFBundleIdentifier (is.hush.dragonx.lite), DMG filename (DragonX_Wallet_Lite-…)
and volume name, so it can coexist with the full-node app.
- Full-node assets (daemon, Sapling params, asmap) were already gated out for lite;
the lite backend artifact is auto-selected for the macos platform by the existing
--lite-backend logic, and CMAKE_LITE_ARGS already reaches the mac configure.
Authored + validated on Linux (bash -n; launcher heredoc, plist, and DMG naming
render correctly for the lite variant) but NOT yet built/run — that needs macOS or
osxcross, neither available here. CLAUDE.md updated to reflect the wired-but-unverified
status; remaining M5b is verifying it on a Mac plus CI backend-artifact build + signing.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds a fail-open, local-only gate that decides whether the lite wallet may run,
so a post-release issue can disable it and rollout can be staged — without any
phone-home (privacy posture: no runtime network fetch; the per-install rollout
bucket is a hashed, never-transmitted local id).
- wallet/lite_rollout_policy.{h,cpp}: a pure decision core. Order — emergency env
kill-switch (absolute) -> local override -> manifest gates (global enable /
version floor-ceiling / blocklist / staged-rollout permille) -> fail-open allow.
Plus a JSON manifest loader (missing/invalid -> fail-open) and FNV-1a bucketing.
- Threads the decision through LiteWalletController -> LiteWalletLifecycleService:
new availability() reason RolloutDisabled blocks create/open/restore and surfaces
the gate's user-facing message via the lifecycle status.
- App::rebuildLiteWallet() resolves it from: DRAGONX_LITE_KILL_SWITCH (env), the
lite_rollout setting (auto/force_on/force_off), and a locally-cached manifest at
<config-dir>/lite_rollout.json. install id generated once via libsodium.
- Settings: persist lite_rollout override + the install id.
A signed remote fetcher can populate the manifest cache later without touching the
policy. Unit-tested (version compare, bucketing, override/env precedence, manifest
gates, staged rollout, loader fail-open, controller integration) and runtime-verified
on Linux (env kill-switch, manifest disable, control sync). Both variants build;
full suite passes; hygiene clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
~9,988 lines of header-only UI code that no compiled translation unit reached,
verified by transitive include-reachability from every .cpp plus a symbol sweep
(all 28 component classes — Snackbar, Ripple, NavDrawerSpec, TabBarSpec,
TransitionManager, … — had zero references in live code):
- src/ui/material/ component library: the material.h umbrella, components/*
(app_bar, cards, chips, dialogs, inputs, lists, nav_drawer, progress, slider,
snackbar, tabs, text_fields), and the animation system (elevation, motion,
ripple, transitions, app_layout) — 19 headers. Kept the live helpers the app
actually uses directly: color_theme, colors, type/typography, draw_helpers,
layout, project_icons, and components/buttons (included by mining_tab).
- src/ui/screens/ layer: main_layout, home_screen, send_screen, etc. — the
original screen stack and the only consumer of the dead component library.
The live UI runs through ui/windows/ (34 .cpp) + ui/pages/.
- src/embedded/resources.h: a superseded dragonx::embedded::Resources duplicate;
the app uses src/resources/embedded_resources.h.
None were in CMakeLists or included by live code, so the build is unaffected.
Both variants build; full test suite passes; source-hygiene check clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
lite_backend_artifact_{contract,resolver}.{cpp,h} (~1,960 lines) were
app-linked but never invoked: all 14 public entry points
(evaluateLiteBackendArtifactContract/Resolver, evaluateLiteBackendActivation-
Readiness, the resolve*/...Name helpers) had zero callers in the app, the
lite_smoke tool, build scripts, or surviving tests. The real backend load
path (LiteClientBridge::linkedSdxl) uses direct litelib_* externs, and the
DRAGONX_ENABLE_LITE_BACKEND symbol check is done in CMake against the symbols
inventory (FATAL_ERROR on a missing symbol) — not via these C++ files. The
files were saturated with churn markers (disabled / dry-dispatch / scaffold).
- Delete the four artifact files and their 8 CMakeLists references.
- Drop the orphaned test cruft in test_phase4.cpp: the contract include,
5 type aliases, and 3 never-called helpers (heapConstructPlanResult,
makeReadyLiteBackendArtifactProvenance, liteBackendArtifactContractHasIssue)
left over from the already-removed bridge-runtime tests.
- Correct the CLAUDE.md lite-wallet description (it credited these files with
backend validation that CMake actually performs) and drop the stale
lite_bridge_runtime mention.
Both variants build; full test suite passes; source-hygiene check clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
lite_bridge_runtime.{cpp,h} was ~25k lines of dry-dispatch / dynamic-loader
scaffolding that the shipping wallet never used: 0 of its 122 public types
reached the app binary. The only live code on the bridge path was the
owned-string memory-safety helper — LiteClientBridge::linkedSdxl() already
loads the backend via direct litelib_* externs in lite_client_bridge.cpp.
- Extract LiteBridgeOwnedString + liteBridgeRuntimeTakeOwnedString into
src/wallet/lite_owned_string.{h,cpp} (the copy-before-free / free-once /
wipe / "Error:"-classify boundary), with the runtime-friend coupling removed.
- Point lite_client_bridge.cpp at the new header.
- Delete lite_bridge_runtime.{cpp,h} and the 16 runtime-only tests +
their fixtures/aliases in test_phase4.cpp; keep the 5 owned-string tests
(retargeted) and restore testGeneratedResourceBehavior, which had been
caught in the runtime-test line range.
- Swap the CMake source/header references.
Both variants build; full test suite passes; source-hygiene check clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The lite-wallet v2 plan was the last tracked lite doc. Fold its still-live
content — current status, remaining M5b work (macOS/CI/signing/rollout), and the
push plan — into a concise "Lite wallet status" section in CLAUDE.md (the
canonical project doc), then move the full milestone plan to docs/_archive/
(untracked) alongside the other lite design docs.
Result: docs/ has no tracked markdown; tracked .md is now just repo essentials
(README, CONTRIBUTING, CODE_OF_CONDUCT, SECURITY, CLAUDE.md). No dangling links.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Both are header-only, in no CMake target, #included nowhere, and their only
symbols (ScrollFadeRT, DrawScrollFadeMask) are referenced nowhere:
- src/ui/effects/scroll_fade_fbo.h — superseded by the shader-based
scroll_fade_shader.h (the implementation actually used by settings_page).
- src/ui/material/gpu_mask.h — a GPU blend-mask helper never integrated.
App + test build clean after removal; tests pass; hygiene clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Move 8 dated-snapshot / dormant-feature docs to docs/_archive/ (git-ignored,
kept locally), leaving only repo essentials + the active lite plan tracked:
- docs/codebase-audit-2026-04-27.md, docs/codebase-overview.md — "current as of
2026-04-27" snapshots, superseded by CLAUDE.md and the v2 plan.
- docs/ui-static-state.md — Phase-9-era UI static-state review snapshot.
- docs/chat-port-feasibility-2026-05-06.md, docs/chat-protocol-spec-2026-05-06.md
— superseded/old-"Batch"-framing docs for the dormant, gated-OFF chat module.
- tests/fixtures/hushchat/{README,CAPTURE_MANIFEST,IMPORT_CHECKLIST}.md -> docs/
_archive/hushchat/ — human docs (not tool input) for the dormant chat fixtures;
the .json fixtures the HushChatFixtureCheck tool globs remain tracked.
These docs only cross-referenced each other (no code/CMake/script refs); no
dangling tracked links remain. Tracked .md (non-libs): 14 -> 6.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Consolidate the lite-wallet documentation down to the single active plan
(lite-wallet-implementation-plan-v2-2026-06-04.md). The 8 prior design/planning
docs — the superseded v1 plan, its runtime-promotion-matrix, the two phase2
runtime-bridge plans, and the four backend artifact/signing design docs — are
moved to docs/_archive/ (added to .gitignore), preserving them locally as
reference while decluttering the tracked tree.
The v2 plan's References section is rewritten to be self-contained: it points to
docs/_archive/ for the historical design docs and to the actual shipping
mechanisms (scripts/build-lite-backend-artifact.sh, lite_backend_artifact_*,
lite_bridge_runtime.cpp) so there are no dangling tracked links. No code,
CMake, or scripts referenced these docs.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Summarize the 2026-06-05 session in the v2 plan: M1–M5a + encryption complete,
GUI wired with lite wording, ~3.2k lines cleanup, Linux+Windows packaging
verified, both variants build clean, runtime-verified on Linux. Notes the
remaining M5b infra (macOS/CI/signing/rollout) and the push plan.
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>
When the user confirms a send on a locked encrypted lite wallet, show an unlock
modal (passphrase -> unlockWallet) instead of letting the backend reject it with
"Wallet is locked". After unlocking, the user re-confirms the send (the form is
preserved). Balances remain viewable while locked; only spending needs unlock.
- send_tab: the Confirm-and-send button routes to App::requestLiteUnlock() when
getWalletState().isLocked(), else sends as before.
- App::renderLiteUnlockPrompt(): centered modal, passphrase (Enter submits),
Unlock/Cancel; the passphrase buffer is sodium-zeroed after every path.
Full-node unaffected (gated on liteWallet()/isLocked()). Builds clean, launches
clean, tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a "Security" subsection to Settings → Backup & keys (open wallet only) that
wires the encryption controller methods to the UI:
- Unencrypted wallet: passphrase field + "Encrypt wallet".
- Encrypted + locked: "Unlock" (passphrase) ; Encrypted + unlocked: "Lock now".
- Encrypted: passphrase + "Remove encryption" (decrypt).
- Status line reflects the result; state shown from WalletState.isEncrypted()/
isLocked() (kept current by the controller's encryptionstatus refresh poll).
Secret hygiene: the passphrase inputs (lite_enc_pass / lite_dec_pass) are
sodium-zeroed immediately after each action and when the wallet closes while the
section was open.
Runtime-checked: app auto-opens a wallet and the new encryptionstatus worker poll
runs clean (no errors); tests pass; hygiene clean.
Follow-ups (not yet): a send-time unlock prompt and a startup lock-screen overlay
for an encrypted+locked wallet (today: unlock via Settings; balances remain
viewable while locked).
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>
Replace the bare "land on main UI with a No-wallet overlay" first-run with a
lite welcome modal, shown when no wallet file exists yet (lite_wallet_ present,
not open, walletExists() false):
- "Create new wallet" — one-click createWallet({}); on success, notifies the user
to back up their recovery phrase and navigates to Settings (Backup & keys),
where the seed can be revealed/copied via the existing backup UI.
- "Restore from seed" — navigates to Settings (Lite wallet request → Restore).
- "Later" — dismiss for the session.
Routes to the already-built + verified create/restore/backup flows rather than
re-implementing seed display in the modal (no new secret-handling surface).
Dismissed once an action is chosen; never shown again once a wallet exists.
Full-node is unaffected (renderLiteFirstRunPrompt() returns early when
lite_wallet_ is null). English i18n built-ins added.
Verified: fresh-HOME lite launch shows the prompt, clean run + shutdown, no
crash/RPC noise; tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
isFirstRun() keys off the full-node `blocks/` data dir, which never exists in
lite — so the daemon/blockchain setup wizard (download node, extract blockchain,
daemon status) fired in lite, where none of it applies and it has zero
lite-awareness. Gate the wizard on !isLiteBuild(); lite goes straight to the main
UI, where the "No wallet open — create or open one in Settings" prompt guides new
users to the lite create/open flow. Full-node behavior is unchanged
(isFirstRun() && !isLiteBuild() == isFirstRun() there).
Completes the lite daemon-wording sweep: the other full-node surfaces are already
lite-gated — daemon settings via supportsFullNodeLifecycleActions(), RPC settings
in the isLiteBuild() else-branch, and Console/Peers/Explorer hidden via
isUiSurfaceAvailable.
Verified: true first-run in lite (fresh HOME) no longer starts the wizard; clean
launch + shutdown, no daemon noise. tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
In lite builds there is no daemon, and isConnected() now tracks the lite wallet,
so the full-node "not connected / waiting for daemon" wording was misleading when
no wallet is open. Add two strings (lite_no_wallet, lite_no_wallet_short; English
built-ins, so other languages fall back until translated) and use them in lite:
- receive/send address preview + receive empty-state overlay + send "can't send"
tooltip + transactions empty state -> "No wallet open [— create or open one in
Settings]" instead of daemon wording.
- Status bar: the red indicator shows "No wallet open" (not "Disconnected") in
lite; the P2P peer count is skipped (lite has no peers); and the redundant
full-node connection-detail line is suppressed (connection_status_ set to
"Connected"/"" from the lite wallet state).
Full-node wording unchanged (all gated on isLiteBuild()). Build + run clean
(no RPC noise), tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Auto-open: on the first update() tick (kept off init() so a slow
initialize_existing network call can't freeze startup before the window), if a
wallet file exists, open it. initialize_existing needs no passphrase — it loads
the file; a previously-synced + saved wallet resumes from its height (fast)
instead of rescanning from the checkpoint. Adds LiteWalletController::walletExists()
(bridge.walletExists on the connection's chain) + a chainName_ member.
RPC-refresh gating: the earlier connected=walletOpen() fix (so the wallet UI is
enabled in lite) had a side effect — the full-node periodic + per-page RPC
refreshes (mining/balance/peers/txs, and setCurrentPage's immediate refresh)
gate on state_.connected, so they began firing in lite and failing
("X error: Not connected"). Re-gate those on ACTUAL RPC connectivity
(rpc_ && rpc_->isConnected()) instead of the lite proxy. Full-node is unchanged
(state_.connected ⟺ rpc connected there); lite no longer issues any RPC.
Runtime-verified in WSLg with a pre-seeded wallet: app auto-opens (Starting
Mempool + sync begins), and "Not connected" / getMiningInfo / RPC-connect noise
all drop to 0 — a fully clean lite run. tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Runtime monitoring of ObsidianDragonLite (WSLg) showed the full-node RPC connect
state machine running in the lite build — `tryConnect()` fired every ~5s and
failed ("Couldn't connect to server / no daemon"). It's called unconditionally
from the main loop with no lite guard.
Worse than noise: `state_.connected` (App::isConnected()) was therefore ALWAYS
false in lite, and it gates the wallet UI — receive_tab disables the new-address
button + shows "not connected", send_tab disables send, transactions_tab shows
not-connected. So the M3/M4 GUI wiring was effectively unreachable: a lite user
could never generate an address or send, even with an open, synced wallet.
Fix:
- tryConnect() no-ops in lite builds (isLiteBuild()), so no RPC attempts.
- App::update() derives state_.connected from lite_wallet_->walletOpen() each
frame — a non-blocking proxy for "lite backend operational" (a wallet opens
only after a successful backend init against the lite server). This enables the
wallet UI once a wallet is open.
Full-node is unaffected (both branches are runtime-gated: isLiteBuild() is false
and lite_wallet_ is null there).
Verified by re-running the app: RPC connection attempts dropped from 7/30s to 0;
clean launch (GL 4.2) + clean shutdown; tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`build.sh --lite-backend --win-release` now cross-compiles a working
ObsidianDragonLite.exe with the real SDXL backend:
- Artifact platform follows the cross target: when only --win-release is
requested, auto-select build/lite-backend/windows/ (previously always the host
artifact, which would link a Linux .a into a Windows .exe).
- Link the Win32 system libs a Rust x86_64-pc-windows-gnu staticlib pulls in
(rustls/schannel, ring, dirs, std) via DRAGONX_LITE_BACKEND_EXTRA_LIBS. The set
is rustc's `--print native-static-libs` for the backend (winapi_* shims mapped
to real mingw import libs); all 21 exist in mingw-w64.
Verified end to end on Linux:
- scripts/build-lite-backend-artifact.sh --platform windows cross-builds the
backend to x86_64-pc-windows-gnu (~105 MB .a); rustls/ring cross-compile clean
(no openssl blocker); all required litelib_* symbols present.
- build.sh --lite-backend --win-release -> release/windows/ObsidianDragonLite-
<ver>.exe (PE32+ GUI x86-64, INCBIN-embedded, ~170 MB) + zip, with the same
full-node-asset exclusion as Linux.
Not yet done: running the .exe on real Windows (cross-compiled only). Plan
updated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`./build.sh --lite-backend --linux-release` produces a working
ObsidianDragonLite zip + AppImage (SDXL backend linked statically). Verified the
lite bundle excludes all full-node assets (dragonxd, dragonx-cli, sapling
params, asmap.dat) and includes res/ + xmrig (pool mining works in lite). CMake
falls back to FetchContent SDL3 when system SDL3 is absent, so the release build
has no system-SDL3 prerequisite. release/ is gitignored.
Remaining M5b (Windows/macOS packaging, CI artifact build + signing,
kill-switch/rollout) is infra/CI, not locally verifiable.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add `lite_smoke --keys`: create a fresh wallet and exercise the M4/M5
spend/backup commands (new-address, export, seed, save) against the real linked
SDXL backend, verifying each response's JSON shape with nlohmann. SECRET-SAFE:
seed and private-key VALUES are never printed — only field presence/shape and
counts (no send/shield, which would broadcast).
Verified live (isolated HOME, throwaway wallet shredded after):
new z shape_ok=1 new t shape_ok=1
seed has_seed=1 has_birthday=1 (REDACTED)
export is_array=1 count=4 has_private_key=1 (REDACTED)
save result_success=1
Confirms the controller's newAddress / exportSeed / exportPrivateKeys / save
parsing matches real backend output.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a "Backup & keys" section to the lite Settings page, shown only for an open
wallet, wiring the M4 controller backup/import surface into the GUI:
- "Show seed" / "Show private keys" -> exportSeed() / exportPrivateKeys();
the revealed secret is displayed read-only (TextWrapped, no extra copies) with
Copy and "Hide & wipe" controls.
- "Import key" (password input) -> importKey() (auto-detects WIF vs shielded);
do_import_sk just records the key + saves (no synchronous rescan), so this is
safe on the UI thread — history appears after the next sync.
Secret hygiene: the revealed-backup buffer is sodium-wiped via
secureWipeLiteSecret on hide, on a new export (overwrite), and if the wallet
closes while revealed; each export also wipes the controller's result copy; the
import input buffer is zeroed immediately after submission.
Lite app + full-node variant build/link clean; controller methods already
covered by testLiteWalletControllerM4; hygiene clean. GUI behavior itself isn't
auto-verifiable here.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Route the existing receive/balance/send UI to the lite controller in lite builds,
with no per-tab UI changes — the existing buttons just work:
- App::createNewZAddress / createNewTAddress: lite branch calls
lite_wallet_->newAddress() (synchronous local key derivation), injects the new
address into WalletState so the UI selects it next frame, and invokes the
receive-tab callback. Placed before the full-node !connected guard.
- App::sendTransaction: lite branch builds a LiteSendRequest (DRGX -> zatoshis,
memo; `from`/`fee` ignored since the backend selects inputs and adds the fee),
fires the controller's async broadcast, and stashes the send_tab callback.
- App::update: drains takeBroadcastResult() and delivers txid/error to the stored
callback, so the send_tab's existing "sending.../sent" flow works unchanged.
All branches guard on lite_wallet_ (null in full-node). Verified: lite app +
test suite + full-node variant all build/link clean; hygiene clean.
Backup/import UI (export seed/keys, import) is deferred — it needs new
secret-display UI rather than an existing button.
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>
Each of these classes wrapped an existing free function with a one-line
delegating method and was never instantiated anywhere (verified: no references
outside their own translation unit, not even within their own .cpp beyond the
definition) — the redundant "wrapper layer" pattern CLAUDE.md warns against:
- LiteWalletLifecycleUiExecutionAdapter -> executeLiteWalletLifecycleUiRequest
- LiteWalletServerSelectionUiExecutionAdapter -> executeLiteWalletServerSelectionUi
- LiteWalletServerLifecycleReadinessPlanner -> evaluateLiteWalletServerLifecycleReadiness
- LiteBackendActivationReadinessAdapter -> evaluateLiteBackendActivationReadiness
The live free functions (the actual entry points used by the UI/runtime) are
unchanged. Both targets build, test suite passes, source-hygiene clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The lite-wallet tree carried a second, unused refresh+readiness architecture
that never reached the shipping binary — exactly the churn CLAUDE.md warns
against. The live refresh path is controller -> gateway.refresh ->
mapLiteWalletRefreshResult -> applyLiteRefreshModelToWalletState; this parallel
stack was dead weight.
Verified unused (their public types/functions are referenced only within the
cluster), then deleted (8 files / 16 incl. headers):
- lite_wallet_refresh_service (LiteWalletRefreshService + gateway adapters)
- lite_wallet_app_refresh_coordinator
- lite_wallet_app_refresh_orchestrator
- lite_wallet_refresh_readiness_policy
- lite_wallet_state_apply_plan
- lite_wallet_state_apply_executor
- lite_wallet_sync_app_refresh_integration
- lite_wallet_sync_execution_readiness
Severed three thin couplings into the cluster from live files:
- state_mapper: dropped the dead mapLiteWalletRefreshServiceResult and switched
its include from refresh_service.h to gateway.h (where the live
LiteWalletRefreshResult/Bundle DTOs actually live).
- server_lifecycle_readiness: dropped the unused syncLifecycleInput member +
converter and the sync_app_refresh_integration include.
- artifact_resolver: relocated the three LIVE artifact-input structs
(LiteWalletSdxlArtifact{Symbols,}Input, LiteWalletLinkedBackendReadinessInput)
out of sync_execution_readiness.h — their only real consumers — into
artifact_resolver.h, then dropped the include.
Also removed the dead DRAGONX_LONG_LITE_BATCH CMake machinery (its source var
was empty; on Windows it generated a broken lite_batch90_receipt_plan.cpp that
#included an empty path) and the stale .cpp/.h entries in CMakeLists.
Lite source files: 44 -> 30. Lite + full-node configure, both targets build,
test suite passes, source-hygiene clean.
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>
The LiteWalletController was constructed once at App::init() with the lite
connection settings known at startup; changing the lite server in Settings
persisted to disk but never reached the live controller, so the new server had
no effect until the next launch.
Factor the construction into App::rebuildLiteWallet() and call it after a
successful server-selection save. The rebuild deliberately preserves a live
session: if a wallet is already open (and possibly mid-sync), it no-ops and the
new selection applies on the next controller build, rather than discarding the
open wallet and its uninterruptible in-flight sync.
Closes the last remaining HIGH from the session audit.
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>
applyLiteRefreshModelToWalletState now derives each address's balance by summing its unspent
notes/utxos (excluding spent and unconfirmed-spent outputs) instead of the aggregate-only
zeros, so the Receive/Balance UI shows per-address amounts. The notes parser shape is
confirmed against do_list_notes in the backend source.
testLitePerAddressBalances covers the summing + spent-exclusion. Completes M2.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lite_smoke: add --restore-recent (restore a throwaway wallet at birthday≈tip) and factor
the data-shape checks (non-blocking commands first). Finding: the backend downloads from a
fixed checkpoint regardless of birthday, so first sync is ~30 min and balance/list block
until synced — a full live data run is impractical.
Verified all refresh parsers against the real backend without a full sync:
- live run: info/addresses/syncstatus parse_ok=1 (addresses z=1/t=6 on a restored wallet).
- via the authoritative Rust source (commands.rs / lightclient.rs):
- balance do_balance fields match parseLiteBalanceResponse.
- list do_list_transactions: sends use outgoing_metadata (no top-level address), receives
use address+amount; parseTransactionRecord already branches correctly.
- syncstatus was the only mismatch (fixed in the prior commit).
No parser changes needed beyond syncstatus. M2 refresh path verified end-to-end at the
shape level.
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>
- LiteWalletController owns a background std::thread worker that, once a wallet is ready,
refreshes every ~4s and publishes a copyable LiteWalletAppRefreshModel under a mutex.
Worker auto-starts on lifecycle-ready and is stopped+joined in the destructor. status_
is written only on the main thread; walletOpen_/syncStarted_ are atomic.
- App::update() calls takeRefreshedModel() and applies it into state_ on the main thread
(WalletState is non-copyable, so the model crosses the thread boundary, not the state),
so the existing Balance/Receive/Transactions tabs populate from lite data.
- refreshWalletState() refactored onto refreshModel() (pure, worker-safe).
- testLiteWalletControllerWorkerProducesModel verifies the worker publishes a populated
model (stable across repeated runs). Builds clean in all configs.
Real-backend smoke (lite_smoke --refresh now runs real output through the parsers) found
two integration bugs, documented in the plan for follow-up:
- syncstatus parser requires synced_blocks/total_blocks but the real idle response is
{"syncing":"false"} (string), so it fails to parse when not actively syncing.
- the first data query (balance/list) blocks on a full chain sync, which would hang the
worker's shutdown join — needs a cancel/timeout path.
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>
Preserve the previously-uncommitted lite wallet implementation and related dev WIP
under version control:
- src/wallet/ lite services: client bridge, bridge runtime, connection, lifecycle,
sync, gateway, result parsers, state mapper, artifact contract/resolver, refresh
services, UI adapters, wallet_backend/capabilities. (Includes two small M1 fixes:
lifecycle walletReady now parses the response; default chain name -> "main".)
- src/chat/ chat protocol; tests/fixtures/ (lite + hushchat); tools/hushchat_fixture_check.cpp;
scripts/build-lite-backend-artifact.sh.
- Pre-existing modified app_network/security/wizard, network_refresh_service, sidebar,
mining_tab, bootstrap dialog, and version headers captured as-is.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- docs/lite-wallet-implementation-plan-v2-2026-06-04.md: vertical-slice plan that
supersedes the v1 plan (now banner-marked); carries over the inherited artifact/
signing/phase-2 design docs for reference.
- scripts/check-source-hygiene.sh: pre-commit/CI guard rejecting >80-char filenames
and chained churn-token names, to stop the deleted "_plan"/"_batch" scaffolding
from regrowing.
- CLAUDE.md: repository guidance for future sessions.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add --lite build flow and ObsidianDragonLite target naming, hide full-node pages/features in lite mode, enforce pool-only mining in lite, and include chat port feasibility audit documentation.
Add an encrypted SQLite transaction history cache with cached tip metadata and
per-address shielded scan progress so startup and full refreshes avoid
re-scanning every z-address while still invalidating on wallet/address/rescan
changes.
Improve wallet history loading by paging transparent transactions, preserving
cached shielded and sent rows, keeping recent/unconfirmed activity visible, and
classifying mining-address receives. Show z_sendmany opid sends immediately in
History and Overview, pin pending rows through refreshes, and apply optimistic
address/balance debits until opids resolve.
Add timestamped RPC console tracing by source/method without logging params or
results, reduce redundant refresh/RPC calls, and cache Explorer recent block
summaries in SQLite.
Expand focused tests for transaction cache encryption, scan-progress
persistence/invalidation, history preservation, operation-status parsing,
pending send visibility, and Explorer/RPC refresh behavior.
Replay cached outgoing viewtransaction entries during transaction refresh so shielded sends created from the wallet remain in the History tab after send tracking is cleared.
Keep incomplete tracked sends retryable, preserve cached send timestamp/confirmation metadata, and emit a send placeholder from gettransaction metadata when viewtransaction enrichment is not yet available.
Add regression coverage for cached sends, retryable empty entries, placeholder sends, and send txid cleanup behavior.
- Add expanded address icon picker with search, bottom-aligned actions, and improved modal sizing
- Embed a pickaxe icon font subset and wire it into typography/address icon rendering
- Track view-only shielded addresses and prevent sends from non-spendable z-addresses
- Improve address transfer dialog sizing, max amount handling, and text clipping
- Tune main header layout values in ui.toml
- Update README, codebase overview, and third-party license documentation
s_dropTargetIdx was reset to -1 unconditionally each frame, including
the release frame. Since drop target detection runs in PASS 2 (after
the drop handler), the target was always -1 when checked. Only reset
while mouse button is held so the previous frame's value is preserved.
Also bump version to 1.2.0-rc1 and add release notes.
- Add text scaling for section labels (TOOLS, ADVANCED) in sidebar
- Separate explorer_section key from explorer nav label to fix ALL CAPS
- Shorten long sidebar translations: es/pt settings, pt overview, ru tools/advanced
- Fix explorer translations from ALL CAPS to proper case in all languages
- New BootstrapDownloadDialog accessible from Settings page
- Stops daemon before download, prevents auto-restart during bootstrap
- Confirm/Downloading/Done/Failed states with progress display
- Mirror support (bootstrap2.dragonx.is)
- Add bootstrap_downloading_ flag to prevent tryConnect() auto-reconnect
- Right-align Download Bootstrap + Setup Wizard buttons in settings
- Add 100 missing i18n keys to all 8 language files (de/es/fr/ja/ko/pt/ru/zh)
- Includes bootstrap, explorer, mining benchmark, transfer, delete blockchain,
force quit, address label, and settings section translations
- Update add_missing_translations.py with new translation batch
- Rewrite RenderSharedAddressList with two-pass layout architecture
- Add drag-to-transfer: drag address onto another to open transfer dialog
- Add AddressLabelDialog with custom label text and 20-icon picker
- Add AddressTransferDialog with amount input, fee, and balance preview
- Add AddressMeta persistence (label, icon, sortOrder) in settings.json
- Gold favorite border inset 2dp from container edge
- Show hide button on all addresses, not just zero-balance
- Smaller star/hide buttons to clear favorite border
- Semi-transparent dragged row with context-aware tooltip
- Copy-to-clipboard deferred to mouse-up (no copy on drag)
- Themed colors via resolveColor() with CSS variable fallbacks
- Keyboard nav (Up/Down/J/K, Enter to copy, F2 to edit label)
- Add i18n keys for all new UI strings
Replace fragile Dummy()-based cursor flow with a deterministic two-pass
layout system:
- Pass 1: compute exact Y positions for all elements (pure math)
- Pass 2: render at computed positions using SetCursorScreenPos + draw list
Eliminates the dual-coordinate mismatch that caused persistent centering
and overflow bugs. Height is computed once, not estimated then measured.
Also tune sidebar spacing via ui.toml:
- button-spacing: 4 → 6
- section-gap: 4 → 8
- Add section-label-pad-bottom (4px) below category labels
- bottom-padding: 0 → 4
- Fix blk/s calculation that was inflated ~10x due to resetting the
time baseline every frame instead of only when blocks advanced
- Add decay when no new blocks arrive for 10s so rate doesn't stay stale
- Add 7 missing translation keys (timeout_off/1min/5min/15min/30min/1hour,
slider_off) to all 8 language files so settings dropdowns translate
- Show language names in native script (中文, Русский, 日本語, 한국어)
Instead of blocking the entire UI with "Activating best chain..." until
the daemon finishes warmup, treat warmup responses as a successful
connection. The wallet now:
- Sets connected=true + warming_up=true when daemon returns RPC -28
- Shows warmup status with block progress in the loading overlay
- Polls getinfo every few seconds to detect warmup completion
- Allows Console, Peers, Settings tabs during warmup
- Shows orange status indicator with warmup message in status bar
- Skips balance/tx/address refresh until warmup completes
- Triggers full data refresh once daemon is ready
Also: fix curl handle/header leak on reconnect, fill in empty
externalDetected error branch, bump version to v1.2.0 in build scripts.
- Fix peer timer calling refreshEncryptionState() instead of
refreshPeerInfo(), so the Network tab now auto-updates every 5s
- Reorder RPC error handling so warmup messages (Loading block index,
Verifying blocks, etc.) display in the status bar instead of being
masked by the generic "Waiting for dragonxd" message
- Replace hardcoded IP addnodes with node.dragonx.is, node1–4.dragonx.is
in both daemon launch params and auto-generated DRAGONX.conf
- Add max_connections setting (persisted, default 0 = daemon default);
passed as -maxconnections= flag to dragonxd on startup
- Show blocks/sec in status bar during sync with exponential smoothing
(e.g. "Syncing 45.2% (12340 left, 85 blk/s)")
instead of initial burst performance. Previously the benchmark used a
fixed 20s warmup + 10s peak measurement, which reported inflated
results on thermally constrained hardware (e.g. 179 H/s vs actual
sustained 117 H/s on a MacBook Pro).
- Adaptive warmup with stability detection: mine for at least 90s,
then compare rolling 10s hashrate windows. Require 3 consecutive
windows within 5% before declaring thermal equilibrium (cap 300s)
- Average-based measurement: record mean hashrate over 30s instead
of peak, reflecting real sustained throughput
- Start candidates at half the system cores — lower thread counts
are rarely optimal and waste time warming up
- Add CoolingDown phase: 5s idle pause between tests so each starts
from a similar thermal baseline
- Adaptive time estimates: use observed warmup durations from
completed tests to predict remaining time
- UI shows Stabilizing when waiting for thermal equilibrium past
the minimum warmup, Cooling during idle pauses"
Security (P0):
- Fix sidebar remaining interactive behind lock screen
- Extend auto-lock idle detection to include active widget interactions
- Distinguish missing PIN vault from wrong PIN; auto-switch to passphrase
Blocking UX (P1):
- Add 15s timeout for encryption state check to prevent indefinite loading
- Show restart reason in loading overlay after wallet encryption
- Add Force Quit button on shutdown screen after 10s
- Warn user if embedded daemon fails to start during wizard completion
Polish (P2):
- Use configured explorer URL in Receive tab instead of hardcoded URL
- Increase request memo buffer from 256 to 512 bytes to match Send tab
- Extend notification duration to 5s for critical operations (tx sent,
wallet encrypted, key import, backup, export)
- Add Reduce Motion accessibility setting (disables page fade + balance lerp)
- Show estimated remaining time during mining thread benchmark
- Add staleness indicator to market price data (warning after 5 min)
New i18n keys: incorrect_pin, incorrect_passphrase, pin_not_set,
restarting_after_encryption, force_quit, reduce_motion, tt_reduce_motion,
ago, wizard_daemon_start_failed
Split monolithic refreshData() into independent sub-functions
(refreshCoreData, refreshAddressData, refreshTransactionData,
refreshEncryptionState) each with its own timer and atomic guard.
Per-category timers replace the single 5s refresh_timer_:
- core_timer_: balance + blockchain info (5s default)
- transaction_timer_: tx list + enrichment (10s default)
- address_timer_: z/t address lists (15s default)
- peer_timer_: encryption state (10s default)
Tab-switching via setCurrentPage() adjusts active intervals so
the current tab's data refreshes faster (e.g. 3s core on Overview,
5s transactions on History) while background categories slow down.
Use fast_worker_ for core data on Overview tab to avoid blocking
behind the main refresh batch.
Bump version to 1.1.2.
ObsidianDragon-agent/ is now a standalone git repo (future submodule)
so AI configuration files are not pushed to the main repository.
- Remove copilot-instructions.md and ARCHITECTURE.md from main tracking
- Remove symlinks from .github/ and docs/
- Add ObsidianDragon-agent/ and .github/ to .gitignore
- Create .github/copilot-instructions.md with project coding standards,
architecture overview, threading model, and key rules for AI sessions
- Add module description comments to app.cpp, rpc_client.cpp, rpc_worker.cpp,
embedded_daemon.cpp, xmrig_manager.cpp, console_tab.cpp, settings.cpp
- Add ASCII connection state diagram to app_network.cpp
- Remove /.github/ from .gitignore so instructions file is tracked
The console tab was passed fast_rpc_ even before its async connection
completed, causing 'Not connected to daemon' errors despite the main
RPC being connected and sync data flowing. Fall back to the main
rpc_/worker_ until fast_rpc_ reports isConnected().
- Set CMAKE_OSX_DEPLOYMENT_TARGET and CMAKE_OSX_ARCHITECTURES before
project() so they propagate to all FetchContent dependencies (SDL3, etc.)
- build.sh: native mac release builds universal binary, detects and
rebuilds single-arch libsodium, verifies with lipo, exports
MACOSX_DEPLOYMENT_TARGET; dev build uses correct build/mac directory
- fetch-libsodium.sh: build arm64 and x86_64 separately then merge with
lipo on native macOS; fix sha256sum unavailable on macOS (use shasum)
- Add pool mining thread benchmark: cycles through thread counts with
20s warmup + 10s measurement to find optimal setting for CPU
- Add GPU-aware idle detection: GPU utilization >= 10% (video, games)
treats system as active; toggle in mining tab header (default: on)
Supports AMD sysfs, NVIDIA nvidia-smi, Intel freq ratio; -1 on macOS
- Fix idle thread scaling: use getRequestedThreads() for immediate
thread count instead of xmrig API threads_active which lags on restart
- Apply active thread count on initial mining start when user is active
- Skip idle mining adjustments while benchmark is running
- Disable thread grid drag-to-select during benchmark
- Add idle_gpu_aware setting with JSON persistence (default: true)
- Add 7 i18n English strings for benchmark and GPU-aware tooltips
Extract txids from completed z_sendmany operations and store in
send_txids_ so pure shielded sends are discoverable. The network
thread includes them in the enrichment set, calls z_viewtransaction,
caches results in viewtx_cache_, and removes them from send_txids_.
- Shutdown: 3-phase stop (wait for RPC stop → SIGTERM → SIGKILL) prevents
LevelDB flush interruption on macOS/APFS that caused full re-sync on restart
- dbcache: auto-detect RAM and set -dbcache to 12.5% (clamped 450-4096 MB)
on macOS (sysctl), Linux (sysconf), and Windows (GlobalMemoryStatusEx)
- Import key: pass user-entered start height to z_importkey and trigger
rescanblockchain from that height for t-key imports
- Bump version to 1.1.1
Build & setup:
- Fix setup.sh and build.sh for macOS (bundle daemon, xmrig, sapling params, asmap.dat into .app)
- Fix CMakeLists.txt libsodium linking for macOS
- Fix incbin.h to use __DATA,__const section on macOS
- Remove vendored libsodium-1.0.18 source tree (use fetch script instead)
- Remove prebuilt-binaries/xmrig (replaced by xmrig-hac)
- Add .DS_Store to .gitignore
Rendering & UI:
- Use GLSL #version 150 and OpenGL 3.2 Core Profile on macOS
- Force dpiScale=1.0 on macOS to fix Retina double-scaling
- Set default window/UI opacity to 100% on Mac/Linux
- Add scroll fade shader guard for macOS GL compatibility
- Add ImGui error recovery around render loop and mining tab
Daemon & bootstrap:
- Fix getDragonXDataDir() to return ~/Library/Application Support/Hush/DRAGONX/ on macOS
- Fix isPortInUse() with connect() fallback (no /proc/net/tcp on macOS)
- Increase daemon watchdog timeout from 3s to 15s
- Add daemon status indicator (colored dot + label) in wizard bootstrap phases
Mining tab:
- Fix EmbeddedDaemon::getMemoryUsageMB() crash on macOS (was using Linux /proc)
- Fix XmrigManager::getMemoryUsageMB() to use ps on macOS instead of /proc
- Restructure RenderMiningTab with wrapper pattern for exception safety
- Fix default pool URL to include port (pool.dragonx.is:3433)
- Mining tab: sync s_selected_threads with actual thread count when idle
thread scaling adjusts threads (solo via genproclimit, pool via
threads_active), skipping sync during user drag
- Auto-lock: bypass lock screen overlay when xmrig pool mining is active
so the mining UI remains accessible
- Import key dialog: add clipboard hover preview with transparent overlay
on the input field, inline key type validation next to title (matching
send tab paste button pattern), configurable via ui.toml
DRAGONX_DEBUG was defined unconditionally, causing D3D11CreateDevice() to
request the debug layer via D3D11_CREATE_DEVICE_DEBUG. This layer is only
available on machines with the Windows SDK or Graphics Tools installed,
so the call fails with DXGI_ERROR_SDK_COMPONENT_MISSING on regular user
machines — causing the app to silently exit.
- Linux --linux-release now produces both ObsidianDragon-Linux-x64.zip
(raw binaries) and ObsidianDragon.AppImage (single-file)
- Windows --win-release keeps standalone exe alongside zip with binaries
- Bundle sapling-spend.params and sapling-output.params in Linux release
- Replace all hardcoded English strings with TR() translation keys across
every tab, dialog, and component (~20 UI files)
- Expand all 8 language files (de, es, fr, ja, ko, pt, ru, zh) with
complete translations (~37k lines added)
- Improve i18n loader with exe-relative path fallback and English base
fallback for missing keys
- Add pool-side hashrate polling via pool stats API in xmrig_manager
- Introduce Layout::beginFrame() per-frame caching and refresh balance
layout config only on schema generation change
- Offload daemon output parsing to worker thread
- Add CJK subset fallback font for Chinese/Japanese/Korean glyphs
Mine-when-idle:
- Auto-start/stop mining based on system idle time detection
- Platform::getSystemIdleSeconds() via XScreenSaver (Linux) / GetLastInputInfo (Win)
- Settings: mine_when_idle toggle + configurable delay (30s–10m)
- Settings page UI with checkbox and delay combo
Console tab:
- Shell-like argument parsing with quote and JSON bracket support
- Pass JSON objects/arrays directly as RPC params
- Fix selection indices when lines are evicted from buffer
Connection & status bar:
- Reduce RPC connect timeout to 1s for localhost fast-fail
- Fast retry timer on daemon startup and external daemon detection
- Show pool mining hashrate in status bar; sidebar badge reflects pool state
UI polish:
- Add logo to About card in settings; expose logo dimensions on App
- Header title offset-y support; adjust content-area margins
- Fix banned peers row cursor position (rawRowPosB.x)
Branding:
- Update copyright to "DragonX Developers" in RC and About section
- Replace logo/icon assets with updated versions
Misc:
- setup.sh: checkout dragonx branch before pulling
- Remove stale prebuilt-binaries/xmrig/.gitkeep
Windows identity:
- Add VERSIONINFO resource (.rc) with ObsidianDragon file description
- Embed application manifest for DPI awareness and shell identity
- Patch libwinpthread/libpthread to remove competing VERSIONINFO
- Set AppUserModelID and HWND property store to override Task Manager cache
- Link patched pthread libs to eliminate "POSIX WinThreads" description
Address creation (+New button):
- Move z_getnewaddress/getnewaddress off UI thread to async worker
- Inject new address into state immediately for instant UI selection
- Trigger background refresh for balance updates
Mining tab:
- Add pool mining dropdown with saved URLs/workers and bookmarks
- Add solo mining log panel from daemon output with chart/log toggle
- Fix toggle button cursor (render after InputTextMultiline)
- Auto-restart miner on pool config change
- Migrate default pool URL to include stratum port
Transactions:
- Sort pending (0-conf) transactions to top of history
- Fall back to timereceived when timestamp is missing
Shutdown:
- Replace blocking sleep_for calls with 100ms polling loops
- Check shutting_down_ flag throughout daemon restart/bootstrap flows
- Reduce daemon stop timeout from 30s to 10s
Other:
- Fix market chart fill artifact (single concave polygon vs per-segment quads)
- Add bootstrap checksum verification state display
- Rename daemon client identifier to ObsidianDragon
Diagnostics & logging:
- add verbose logging system (VERBOSE_LOGF) with toggle in Settings
- forward app-level log messages to Console tab for in-UI visibility
- add detailed connection attempt logging (attempt #, daemon state,
config paths, auth failures, port owner identification)
- detect HTTP 401 auth failures and show actionable error messages
- identify port owner process (PID + name) on both Linux and Windows
- demote noisy acrylic/shader traces from DEBUG_LOGF to VERBOSE_LOGF
- persist verbose_logging preference in settings.json
- link iphlpapi on Windows for GetExtendedTcpTable
Security & encryption:
- update local encryption state immediately after encryptwallet RPC
so Settings reflects the change before daemon restarts
- show notifications for encrypt success/failure and PIN skip
- use dedicated RPC client for z_importwallet during decrypt flow
to avoid blocking main rpc_ curl_mutex (which starved peer/tx refresh)
- force full state refresh (addresses, transactions, peers) after
successful wallet import
Network tab:
- redesign peers refresh button as glass-panel with icon + label,
matching the mining button style
- add spinning arc animation while peer data is loading
(peer_refresh_in_progress_ atomic flag set/cleared in refreshPeerInfo)
- prevent double-click spam during refresh
- add refresh-button size to ui.toml
Other:
- use fast_rpc_ for rescan polling to avoid blocking on main rpc_
- enable DRAGONX_DEBUG in all build configs (was debug-only)
- setup.sh: pull latest xmrig-hac when repo already exists
RPC client:
- Add call() overload with per-call timeout parameter
- z_exportwallet uses 300s, z_importwallet uses 1200s timeout
Decrypt wallet (app_security.cpp, app.cpp):
- Show per-step and overall elapsed timers during decrypt flow
- Reduce dialog to 5 steps; close before key import begins
- Run z_importwallet on detached background thread
- Add pulsing "Importing keys..." status bar indicator
- Report success/failure via notifications instead of dialog
RPC caching (app_network.cpp, app.h):
- Cache z_viewtransaction results in viewtx_cache_ across refresh cycles
- Skip RPC calls for already-cached txids (biggest perf win)
- Build confirmed_tx_cache_ for deeply-confirmed transactions
- Clear all caches on disconnect
- Remove unused refreshTransactions() dead code
Peers (app_network.cpp, peers_tab.cpp):
- Route refreshPeerInfo() through fast_worker_ to avoid head-of-line blocking
- Replace footer "Refresh Peers" button with ICON_MD_REFRESH in toggle header
- Refresh button triggers both peer list and full blockchain data refresh
Mining (mining_tab.cpp):
- Allow pool mining toggle when blockchain is not synced
- Pool mining only needs xmrig, not local daemon sync
- Fix z_importwallet to use full path instead of filename only
- Add rescanBlockchain() method that restarts daemon with -rescan flag
- Track rescan progress via daemon output parsing and getrescaninfo RPC
- Display rescan progress in status bar with animated indicator when starting
- Improve dark theme card contrast: lighter surface-variant, tinted borders, stronger rim-light
2026-02-28 15:06:35 -06:00
402 changed files with 84768 additions and 29822 deletions
**Every UI layout constant must be defined in `res/themes/ui.toml` and read at runtime via the schema API.** Never hardcode pixel sizes, ratios, rounding values, thicknesses, or spacing constants directly in C++ source files. This is critical for maintainability, theming support, and hot-reload.
### Schema API reference
The singleton is accessed via `schema::UI()` (header: `#include "../schema/ui_schema.h"`).
- Spacing that already goes through `Layout::spacing*()` or `spacing::dp()`
### Legacy system: `UILayout`
`UILayout::instance()` is the older layout system still used for fonts, typography, panels, and global spacing. New layout values should use `schema::UI().drawElement()` instead. Do not add new keys to `UILayout`.
Or with the C++ toml++ parser (which is what the app uses):
```bash
cd build && make -j$(nproc)
```
### Build
```bash
# Linux
cd build && make -j$(nproc)
# Windows cross-compile
./build.sh --win-release
```
## Plans
When asked to "create a plan", always create a new markdown document in the `docs/` directory with the plan contents.
## Icons: Use Material Design icon font, never Unicode symbols
**Never use raw Unicode symbols or emoji characters** (e.g. `↓`, `↗`, `⛏`, `🔍`, `📬`, `⚠️`, `ℹ`) for icons in C++ code. Always use the **Material Design Icons font** via the `ICON_MD_*` defines from `#include "../../embedded/IconsMaterialDesign.h"`.
### Icon font API
| Method | Size | Fallback |
|---|---|---|
| `Type().iconSmall()` | 14px | Body2 |
| `Type().iconMed()` | 18px | Body1 |
| `Type().iconLarge()` | 24px | H5 |
| `Type().iconXL()` | 40px | H3 |
### Correct usage
```cpp
#include "../../embedded/IconsMaterialDesign.h"
// WRONG — raw Unicode symbol
itemSpec.leadingIcon = "↙";
// CORRECT — Material Design icon codepoint
itemSpec.leadingIcon = ICON_MD_CALL_RECEIVED;
// WRONG — emoji for search
searchSpec.leadingIcon = "🔍";
// CORRECT — Material Design icon
searchSpec.leadingIcon = ICON_MD_SEARCH;
// For rendering with icon font directly:
ImGui::PushFont(Type().iconSmall());
ImGui::TextUnformatted(ICON_MD_ARROW_DOWNWARD);
ImGui::PopFont();
```
### Why
Raw Unicode symbols and emoji render inconsistently across platforms and may not be present in the loaded text fonts. The Material Design icon font is always loaded and provides consistent, scalable glyphs on both Linux and Windows.
### Audit for Unicode symbols
Before completing any task that touches UI code, search for and replace any raw Unicode symbols that may have been introduced. Common symbols to look for:
| Unicode | Replacement |
|---------|-------------|
| `▶` `▷` | `ICON_MD_PLAY_ARROW` |
| `■` `□` `◼` `◻` | `ICON_MD_STOP` or `ICON_MD_SQUARE` |
| `●` `○` `◉` `◎` | `ICON_MD_FIBER_MANUAL_RECORD` or `ICON_MD_CIRCLE` |
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
## What this is
ObsidianDragon is a portable, full-node GUI wallet for DragonX (DRGX), written in C++17 using SDL3 + Dear ImGui (immediate-mode). It drives a `dragonxd` full node over JSON-RPC and can embed/extract the daemon itself. A separate **Lite** variant (`ObsidianDragonLite`) drops the full node and instead talks to an external lite-wallet backend library.
## Build & run
`build.sh` is the single entry point for all builds. `setup.sh` (repo root) installs/validates dependencies.
```bash
./build.sh # Dev build (native, no packaging) -> build/linux/bin/ObsidianDragon
./build.sh --lite # Dev build of the Lite variant -> build/linux/bin/ObsidianDragonLite
./build.sh --clean # Wipe the build dir first
./build.sh --linux-release # Release zip + AppImage -> release/linux/
./build.sh --win-release # Windows cross-compile (mingw-w64) -> release/windows/
./setup.sh --check # Report missing build deps without installing
```
Dev builds use `build/linux/` (or `build/mac/`). To re-build incrementally without re-running CMake config: `cmake --build build/linux -j$(nproc)`.
The wallet connects to the daemon using credentials in `~/.hush/DRAGONX/DRAGONX.conf` (`rpcuser`/`rpcpassword`/`rpcport`). It searches for `dragonxd`/`dragonx-cli` binaries in the **executable's own directory first**, so dropping custom node builds next to the wallet binary overrides the bundled ones.
## Tests
Tests live in `tests/test_phase4.cpp` — a single large translation unit using a custom assertion harness (`EXPECT_TRUE`/`EXPECT_EQ`/`EXPECT_NEAR` macros, one `main()`, exit code = failure count). `include(CTest)` enables `BUILD_TESTING=ON` by default, so the `ObsidianDragonTests` executable is built alongside the app.
```bash
cd build/linux && ctest --output-on-failure # run the suite
./build/linux/bin/ObsidianDragonTests # run the binary directly (same thing)
```
There is no per-test filtering — it is one binary that runs every assertion. The suite exercises the services layer, lite-wallet bridge, and pure helpers (parsers, formatters, model classes) without launching the GUI. Fixtures are under `tests/fixtures/` (path injected as `DRAGONX_TEST_FIXTURE_DIR`).
## Architecture
**Entry & main loop.**`src/main.cpp` owns SDL3 window creation, ImGui/OpenGL(or DX11 on Windows) setup, and the frame loop. The `App` class is the central controller; because it is large it is split across four files that all implement the same class:
-`src/app.cpp` — core lifecycle, the per-frame `render()`, tab dispatch
**RPC.** All daemon calls go through `src/rpc/` (`rpc_client`, `connection`, `rpc_worker`). **Never block the main/UI thread with synchronous network I/O — dispatch through `RPCWorker`** (async). `rpc/types.h` holds the shared DTOs.
**Services** (`src/services/`) hold the non-UI state machines that the `App` owns: `NetworkRefreshService` + `RefreshScheduler` (polling/refresh of balance, peers, txs on intervals) and the `WalletSecurity*` controller/workflow stack (encryption & unlock flows).
**UI** (`src/ui/`): `windows/` are the tabs and dialogs (one pair per screen, e.g. `send_tab`, `mining_tab`, `console_tab`), `pages/` are multi-section screens (Settings), `material/` is the design-system layer (the live helpers `color_theme`, `colors`, `type`/`typography`, `draw_helpers`, `layout`, `project_icons`, `components/buttons`), `schema/` loads the TOML UI schema/skins, `effects/` is GL post-processing (blur/acrylic).
**Lite wallet** (`src/wallet/`): the bridge to an external `litelib_*` C-ABI backend. `lite_client_bridge` loads the backend (via direct `litelib_*` externs in `linkedSdxl()`) and owns each Rust string through `lite_owned_string` (copy-before-free / free-once). On top sit `lite_connection_service`, `lite_sync_service`, `lite_result_parsers`, `lite_wallet_gateway`, `lite_wallet_state_mapper`, and `lite_wallet_lifecycle_service`, all driven by `lite_wallet_controller`. The real frontend entry points are `lite_wallet_lifecycle_ui_adapter` and `lite_wallet_server_selection_adapter` (used by `src/ui/pages/settings_page.cpp`); everything else is reachable through them. (The prebuilt-backend symbol check for `DRAGONX_ENABLE_LITE_BACKEND` is done in CMake against the symbols inventory — see below — not in C++.)
> ⚠️ **Do not regrow the `_plan`/`_batch` churn.** This directory previously held ~160 dead `lite_wallet_*_plan` / `*_batch*_receipt_custody_acceptance_confirmation_archive_handoff_*` files (filenames up to 250 chars) — auto-generated scaffolding that never reached the shipping binary. They were deleted. When extending lite-wallet behavior, **edit the named service/bridge/runtime files in place**; never add another "promotion/receipt/custody/handoff/stewardship" wrapper layer. `scripts/check-source-hygiene.sh` (wired as a `.git/hooks/pre-commit` hook) blocks >80-char filenames and chained churn-token names — run it in CI too.
**Chat** (`src/chat/*`): the HushChat protocol port (Contacts/Chat tabs, seed-derived identity, secretstream crypto, seed-encrypted sqlite store, two-variant send/receive transport). Runtime behavior is gated by `DRAGONX_ENABLE_CHAT`, now **default ON** (the sources always compile; the flag folds the feature away at runtime via `hushChatFeatureEnabledAtBuild()`). Full-node chat derives its identity from the wallet's mnemonic (`z_exportmnemonic`, portable/SDXLite-compatible) or, for legacy/non-mnemonic wallets, a stable z-address spending key (`z_exportkey`).
## Build variants & feature gating
Variants are selected with CMake options (set by `build.sh` flags), surfaced to C++ as compile definitions:
-`DRAGONX_BUILD_LITE` (`--lite`) → `DRAGONX_LITE_BUILD` define; renames the app to `ObsidianDragonLite` and excludes embedded-daemon / full-node assets (Sapling params, asmap, dragonxd).
-`DRAGONX_ENABLE_LITE_BACKEND` → links a real external lite backend. Requires `--lite`, link mode `imported`, ABI `sdxl-c-v1`, and a symbols inventory file (built by `scripts/build-lite-backend-artifact.sh`); CMake hard-fails if any required `litelib_*` symbol is missing. The backend **source is vendored in-tree** at `third_party/silentdragonxlite/` — the `qtlib` C-ABI wrapper (`lib/`, produces `libsilentdragonxlite.a`) and the `silentdragonxlitelib` core (`silentdragonxlite-cli/lib/`, with `proto/` + `res/`). `build-lite-backend-artifact.sh` defaults `--backend-dir` there, so the lite wallet builds **without** the upstream SilentDragonXLite repo. External build inputs are limited to the **Rust toolchain (rustc/cargo 1.63)** plus two project-controlled sources on `git.dragonx.is`: the librustzcash crates come from the mirror `git.dragonx.is/DragonX/librustzcash` (the 6 `git =` deps in the core `Cargo.toml`, pinned to rev `acff1444…`), and the **Sapling params are not committed** (gitignored) — the build fetches them from the `git.dragonx.is/DragonX/zcash-params` release `sapling-v1` and verifies their SHA-256 before rust-embed bakes them in (`ensure_sapling_params`; override the URL with `SAPLING_PARAMS_BASE_URL`). Other crate deps come from crates.io. For a fully offline build, `cargo vendor` into `third_party/silentdragonxlite/lib/vendor/` and add a `vendored-sources` redirect to `lib/.cargo/config.toml` (the build script symlinks `vendor/` into its prepared dir if present); `vendor/` is gitignored.
-`DRAGONX_ENABLE_CHAT` → `DRAGONX_ENABLE_CHAT` define gating the chat module.
Guard full-node-only code paths with `#if DRAGONX_LITE_BUILD` / chat code with `DRAGONX_ENABLE_CHAT`.
## Lite wallet status
The Lite variant is **functionally complete and runtime-verified on Linux + Windows** (work lives on branch `cleanup/lite-plan-churn`, **local-only — not pushed yet**):
- **Implemented:** lifecycle (create/open/restore + auto-open on startup), sync, refresh, send / shield / import / export / seed, persistence (the backend does *not* auto-save after sync/send/shield — the controller triggers `save` at those points), and passphrase **encryption** (encrypt/unlock/lock/decrypt + Settings UI + send-time & startup unlock; the backend locks immediately on `encrypt`). All controller-tested against the fake backend (`tests/fake_lite_backend.h`) and smoke-verified against the real SDXL backend via `tools/lite_smoke` (incl. a full sync). GUI is wired end-to-end with lite-appropriate wording; the full-node RPC connect loop / wizard / daemon strings are gated out of lite (lite "online" is derived from `lite_wallet_->walletOpen()`, not RPC).
- **Packaging:** `./build.sh --lite-backend --linux-release` (zip + AppImage, **verified**) and `--win-release` (cross-compiled `.exe`, **verified**; first build the Windows backend artifact with `scripts/build-lite-backend-artifact.sh --platform windows`). macOS `--lite-backend --mac-release` is **wired but not yet verified on this Linux box** (needs macOS/osxcross): the `.app`/launcher/rpath/`CFBundleExecutable` follow `ObsidianDragonLite`, full-node assets are skipped, and the lite variant gets its own `CFBundleName` ("DragonX Wallet Lite"), bundle id (`is.hush.dragonx.lite`), and DMG name so it can coexist with the full-node app. All variants correctly exclude full-node assets.
- **Rollout / kill-switch (implemented):** `wallet/lite_rollout_policy.{h,cpp}` is a pure, fail-open gate (local-only, no network) feeding `LiteWalletLifecycleService::availability()` (new `RolloutDisabled` reason). Inputs: the emergency env var `DRAGONX_LITE_KILL_SWITCH` (absolute — not even `force_on` bypasses it); a `lite_rollout` setting (`auto`/`force_on`/`force_off`); and an optional **locally-cached** manifest at `<config-dir>/lite_rollout.json` (`global_enabled`, `min_version`/`max_version`, `blocked_versions`, `rollout_permille`, `message`) keyed for staged rollout on a hashed, never-transmitted per-install id. A signed remote fetcher can populate that cache later without touching the policy. Resolved in `App::rebuildLiteWallet()`; the disable message surfaces via the lifecycle status. Unit-tested + runtime-verified (env / manifest / control).
- **Remaining (M5b):** verify the wired macOS `--lite` packaging on a Mac/osxcross, CI backend-artifact build + signing.
- **To publish:** rename branch → `feat/lite-wallet`, base the PR on `dev` (the full-node UX is already there), and handle the dormant gated-OFF HushChat content bundled in commit `af06b8b`.
The detailed milestone plan and design history (the v2 plan, backend artifact/ABI/signing design docs, the v1 plan, chat specs, etc.) are kept **untracked** under `docs/_archive/`.
## Miner updater (xmrig)
The mining tab's pool section has an **"Update miner…"** button that downloads/verifies/installs the latest DRG-XMRig from the project Gitea (`util/XmrigUpdater` + `ui/windows/xmrig_download_dialog.h`). Flow: query `git.dragonx.is/api/v1/repos/DragonX/drg-xmrig/releases/latest` → pick the asset for this platform (`linux-x64` / `win-x64` / `macos-x86_64`; no match → "Unavailable") → libcurl download (TLS verified) → verify the archive **SHA-256** (from the release body) **and** a detached **ed25519 signature** → miniz-extract the binary (flattening the versioned subdir) into `resources::getDaemonDirectory()`. The whole archive is verified, so extracted members are trusted by transitivity (no per-member hash check). The pure, no-I/O core is split into `xmrig_updater_core.cpp` for unit tests; an env-gated (`DRAGONX_TEST_NETWORK=1`) test exercises the worker live. A **"Browse all releases…"** button (the `/releases` list, newest first, pre-releases included) lets users pin an older or pre-release build — same verify/install path via `startInstallRelease()`; the picker UI is shared with the daemon updater (`ui/windows/release_list_view.h`).
**Signature verification is enforced** (`kXmrigRequireSignature = true` in `src/util/xmrig_updater.h`), checked against the public key pinned in `kXmrigSignaturePublicKeyBase64`. **Consequence for releases:** every `drg-xmrig` release MUST ship a detached signature per archive or the in-app updater refuses it. To cut a release: build the archives, then `scripts/sign-xmrig-release.sh sign <secret.key> <archive.zip>...` (OpenSSL-based, no extra deps) and upload each `<archive>.sig` as a release asset alongside its `.zip`. The signing **secret key must stay offline** (it is gitignored: `*.ed25519.key`); only its base64 public key is pinned in the source. To rotate the key, regenerate (`scripts/sign-xmrig-release.sh keygen`) and update `kXmrigSignaturePublicKeyBase64`. An emergency env override is not provided — disabling verification means setting `kXmrigSignaturePublicKeyBase64` empty (and rebuilding).
## Daemon updater (dragonxd)
Settings → **NODE & SECURITY → DAEMON BINARY** has a **"Check for updates…"** button that downloads/verifies/installs the latest **dragonxd full node** from the project Gitea — the full-node sibling of the xmrig updater (`util/DaemonUpdater` + `ui/windows/daemon_download_dialog.h`, pure no-I/O core in `daemon_updater_core.cpp`; gated full-node-only via `supportsFullNodeLifecycleActions()`). Flow: query `git.dragonx.is/api/v1/repos/DragonX/dragonx/releases/latest` → pick the archive for this platform (`linux-amd64` / `macos` / `win64`; no match → "Unavailable") → libcurl download (TLS verified) → verify the archive **SHA-256** (parsed from the release body's markdown **checksum table**, not xmrig's `<hash> <name>` lines) **and** a detached **ed25519 signature** → miniz-extract the three executables (`dragonxd`/`dragonx-cli`/`dragonx-tx`, flattening the versioned subdir) into `resources::getDaemonDirectory()`. The archive also bundles Sapling params/asmap, which the updater deliberately leaves to the wallet's own resource extraction. Install is **atomic and safe while the node runs** (POSIX `rename()` replaces the in-use binary; Windows moves the locked `.exe` aside to `.old`); the new binary takes effect on the **next daemon start**, so the Done screen offers **"Restart daemon now"** (`App::restartDaemon()`). A **"Browse all releases…"** button (shared `release_list_view.h` picker) lets users pin a specific/older/pre-release node build via `startInstallRelease()` — with a downgrade caution, since an older binary may not match current chain data.
**Signature verification is enforced** (`kDaemonRequireSignature = true` in `src/util/daemon_updater.h`), checked against `kDaemonSignaturePublicKeyBase64`. **Consequence for releases:** every `dragonx` release MUST ship a detached `<archive>.sig` per platform archive or the in-app updater refuses it (as of v1.0.2 the releases publish SHA-256 but **no** signatures yet — sign + upload them to enable in-app updates). To cut a release: `scripts/sign-daemon-release.sh sign <secret.key> dragonx-<ver>-{linux-amd64,macos,win64}.zip` (OpenSSL-based) and upload each `.sig` next to its `.zip`. The signing **secret key stays offline** (gitignored `*.ed25519.key`; this repo's is `dragonx-daemon.ed25519.key`); only the base64 public key is pinned. To rotate: `scripts/sign-daemon-release.sh keygen` and update `kDaemonSignaturePublicKeyBase64`. The generic SHA-256 / ed25519 primitives are shared with the miner updater (`util::sha256Hex` / `util::verifyXmrigSignature`).
## Versioning
The version has a **single source of truth**: `project(... VERSION 1.2.0 ...)` plus `DRAGONX_VERSION_SUFFIX` in `CMakeLists.txt`. CMake generates `build/.../generated/dragonx_generated_version.h` from `src/config/version.h.in`. Do not hand-edit generated version output or hardcode version strings — bump the `project()` version in `CMakeLists.txt`.
## Conventions
- **C++17.** Match the surrounding code's style per file.
- **Icons:** use the Material Design icon font defines (`ICON_MD_*`); never raw Unicode glyphs.
- **UI layout values** belong in `res/themes/ui.toml`, read via `schema::UI()` — do not hardcode pixel sizes/offsets in code.
- **i18n:** user-facing strings are translated via `src/util/i18n`; the English source of truth is the `strings_[...]` map in `src/util/i18n.cpp`, and the per-language translations live as the **source of truth** in `res/lang/` (`de`, `es`, `fr`, `ja`, `ko`, `pt`, `ru`, `zh`). **Edit those JSONs directly and additively** — write with `json.dump(..., indent=4, sort_keys=True, ensure_ascii=False)`; never bulk-regenerate/overwrite a whole file (that path silently dropped ~285 keys/language before). `scripts/add_missing_translations.py` back-fills only the keys missing from a JSON (non-destructive), and `scripts/build_cjk_subset.py` rebuilds the CJK subset font (`res/fonts/NotoSansCJK-Subset.ttf`) after new CJK glyphs are added.
- **Commits:** the history uses Conventional Commits (`feat(scope): …`, `fix(scope): …`). PRs target `master`.
message(WARNING"xxd not found; runtime language JSON files will be copied, but embedded build/generated/embedded/lang_*.h files will not be regenerated")
The wallet will automatically connect to the daemon using credentials from \`~/.hush/DRAGONX/DRAGONX.conf\`.
The wallet will automatically connect to the daemon using credentials from `~/.hush/DRAGONX/DRAGONX.conf`.
### Using Custom Node Binaries
The wallet checks its **own directory first** when looking for DragonX node binaries. This means you can test new or different branch builds of `hush-arrakis-chain`/`hushd` without waiting for a new wallet release:
@@ -128,12 +134,13 @@ The wallet checks its **own directory first** when looking for DragonX node bina
[[ -f "$XMRIG_LINUX"]]&&{ cp "$XMRIG_LINUX""$dist_dir/"; chmod +x "$dist_dir/xmrig"; info "Bundled xmrig";}|| warn "xmrig not found — mining unavailable in zip"
cp -r bin/res "$dist_dir/" 2>/dev/null ||true
# ── Zip ──────────────────────────────────────────────────────────────────
ifcommand -v zip &>/dev/null;then
(cd"$out"&& zip -r "$DIST.zip""$DIST")
info "Zip: $out/$DIST.zip ($(du -h "$out/$DIST.zip"| cut -f1))"
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.