9790005c3ecc42e64e32b51f20ec15119e013e2d
20
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 EdisonJwa
|
9790005c3e |
feat(beta): wire voice in/out end-to-end with push-to-talk (v0.2.0-beta.1)
Reaches the Internal Beta milestone of DEC-001's release sequence the
same day as Alpha. Adds voice capture and playback through the full
Flutter UI → FRB → Rust core → tsclientlib → server path.
Promotions from PoC:
poc/audio-capture-playback-spike → crates/chanora_audio/
New product code:
crates/chanora_audio/src/engine.rs — cpal capture and playback,
audiopus Opus VoIP encoder (48 kHz mono 20 ms frames), tsclientlib
AudioHandler for decode + jitter buffer + mix on playback,
push-to-talk gate, graceful playback-only fallback when capture
is unavailable.
crates/chanora_protocol/src/adapter.rs — extended with
voice_out_tx (clonable mpsc::Sender<OutPacket>) and
take_voice_in() (one-shot mpsc::Receiver<InboundVoice>); main
loop now interleaves outbound voice drain, event pumping, and
control-request handling.
crates/chanora_protocol/src/lib.rs — re-exports the few
tsproto_packets types (OutAudio, OutPacket, InAudioBuf,
AudioData, CodecType, Direction) that chanora_audio
legitimately needs. Documented as the single deliberate
cross-crate type re-export per SAD-067, justified by the
performance cost of a parallel type hierarchy on the 20 ms
voice frame.
core/chanora_core/src/lib.rs — ChanoraSession::start_audio,
set_ptt, audio_stats; disconnect now stops the engine first.
crates/chanora_bridge/src/api.rs — startAudio, setPtt,
audioStats commands and BridgeAudioStats DTO.
apps/chanora_flutter/lib/main.dart — "Start audio" button +
hold-to-talk PTT button with pressed/released visual state +
live stats line (TX/RX/PTT). Stats polled every 500 ms.
ARB:
Both en and zh-Hans gain startAudioAction, pttHoldToTalk,
pttTransmitting, audioStatsLine. Banner updated to
"Beta build — voice in/out wired; not production ready."
FRB config:
flutter_rust_bridge.yaml gains local: true so codegen resolves
the workspace member's library stem to "chanora_bridge" instead
of falling back to "UNKNOWN".
Empirical verification (2026-05-14, against cn.teamspeak.app):
cargo check + cargo test --workspace: all green.
flutter analyze: 0 issues.
flutter test: 4/4 passing including:
- test/alpha_e2e_test.dart (regression: Alpha still works)
- test/beta_e2e_test.dart (Beta: connect → startAudio →
PTT cycle → disconnect against cn.teamspeak.app).
Live smoke (cargo test alpha_smoke -- --ignored): 49 channels,
37 clients retrieved.
Capture stream open against the host PipeWire auto_null source
refused (snd_pcm_hw_params); engine correctly logged the warning
and continued in playback-only mode. TX=0 frames, RX=0 frames
reflects the headless null-source environment; on a real mic
host the encoder produces ~50 frames/second while PTT is held.
Honest Beta scope (NOT in this release):
- AEC / AGC / NS / HPF DSP (DEC-007..010): AudioEffects exists
as a struct but the filters are no-ops. Beta+ work.
- Production-quality resampler: current code is linear
interpolation. Beta+ work.
- Identity persistence via chanora_storage: still ephemeral.
- Push-to-Dart event stream: UI polls instead.
- chanora_diagnostics tracing-layer wiring: still scaffold.
- Mobile (Android) cdylib + UI: PoC-proven, not yet in product.
- Reconnect / network-loss recovery for the voice path.
Docs updates:
- docs/governance/product-decision-register.md bumped to v0.9.7
(Beta-milestone change-history entry; no row changes).
- docs/governance/poc-results-summary.md bumped to v0.6.0
(RISK-PoC-005 updated with Beta progress).
|
||
|
EdisonJwa
|
53b176b722 |
docs(governance): record Alpha milestone in PoC results summary (v0.5.0)
Updates RISK-PoC-005 to 'partially closed' and adds a v0.5.0 change
history entry documenting:
- the tsclientlib spike promotion into crates/chanora_protocol;
- the typed ChanoraSession in core/chanora_core wiring the
protocol API;
- the FRB 2.12.0 bridge wiring;
- the Flutter Alpha UI;
- the empirical verification path through alpha_e2e_test.dart
and alpha_smoke.rs;
- the v0.1.0-alpha.1 tag pointing at commit 3bb038c.
No other doc bumps are needed: the decision register stays at v0.9.6
(no new decisions), the audit reports stay at v0.9.3 (no new audit
evidence beyond what the PoCs already provided), the PoC plan stays
at v0.3.0 (all six PoC entries were already PASS).
|
||
|
EdisonJwa
|
4915ec0a1b |
feat(alpha): wire connect→snapshot→disconnect end-to-end (v0.1.0-alpha.1)
First Internal Alpha build per DEC-001. Closes the milestone of
'Flutter UI calls Rust via the typed bridge, Rust connects to a
TeamSpeak-compatible server through tsclientlib, returns a typed
snapshot, and disconnects cleanly.' Audio remains Beta scope.
Promotions from PoC:
poc/tsclientlib-connect-spike → crates/chanora_protocol/
New product code:
crates/chanora_protocol/src/{dto.rs,adapter.rs} — typed boundary
over tsclientlib. Tokio task owns the Connection; public
handle communicates via mpsc/oneshot. No tsclientlib types
cross out of the crate (SAD-067 / SysDes-011 / SysDes-029).
core/chanora_core/src/lib.rs — ChanoraSession composes the
protocol crate, enforces the DEC-006 single-connection
invariant.
crates/chanora_bridge/src/{api.rs,frb_generated.rs} — FRB 2.12.0
bridge per DEC-014. cdylib + staticlib + rlib. Typed
BridgeSnapshot / BridgeChannel / BridgeClient / BridgeError
DTOs. Process-wide OnceLock<Runtime> + OnceLock<ChanoraSession>.
flutter_rust_bridge.yaml at repo root.
apps/chanora_flutter/lib/main.dart — Alpha UI: server form,
connect button, channel tree, disconnect.
apps/chanora_flutter/lib/l10n/app_{en,zh}.arb expanded with the
Alpha key set; ARB metadata reaffirms ADR-008 for
server-provided content.
Generated Dart bindings under apps/chanora_flutter/lib/src/rust/.
Empirical verification (2026-05-14):
Workspace: cargo check + cargo test clean
(workspace tests: all green).
Bridge cdylib: target/release/libchanora_bridge.so produced
(~15 MB).
Flutter: flutter analyze clean; flutter test runs 3/3 green
including the alpha_e2e_test that drives the full
Dart → FRB → chanora_bridge → chanora_core → chanora_protocol
→ tsclientlib → UDP → cn.teamspeak.app
path. The captured logcat/stdout shows the tsproto resender
transitioning Connected → Disconnecting → Disconnected on
clean teardown.
Architecture changes:
- Removed the chanora_core ↔ chanora_bridge cyclic dependency.
chanora_core no longer knows the bridge exists; the bridge
maps from CoreError.
- chanora_bridge crate's #![forbid(unsafe_code)] lint relaxed
because FRB-generated glue legitimately uses unsafe at the
FFI boundary. Hand-written code remains unsafe-free.
Open follow-ups (NOT in this Alpha):
- Audio capture/playback wiring into chanora_audio
(Beta scope per DEC-001).
- Identity persistence via chanora_storage
(currently regenerated on every connect).
- Per-message diagnostics + redaction
(chanora_diagnostics still scaffold).
- Reconnect / network-loss recovery.
- Mobile (Android) build of the bridge cdylib + UI verification.
|
||
|
EdisonJwa
|
e0f34009d9 |
docs(governance): record product scaffolding paths (DEC-022)
Updates two documents to reflect the workspace + Flutter app
scaffold landed in the previous commit.
path-migration-map.md v0.9.2 -> v0.9.3:
Adds §3 Implementation Path Layout. Lists each subsystem's
canonical crate path alongside its SAD / SysDes / DEC authority.
Notes that the Flutter app is owned by Flutter tooling and is
not a Cargo workspace member.
CHANGELOG entry under [Unreleased]:
- Documents the seven new Cargo workspace members and the
invariants pinned at the workspace level.
- Documents the Flutter app scaffold, the DEC-004 minSdk = 28
override, and the DEC-015 English + Simplified Chinese ARB
catalogue (with the ADR-008 server-content reaffirmation).
- Records the empirical verification (cargo check + cargo test
+ flutter analyze + flutter test all clean).
|
||
|
EdisonJwa
|
974dda9601 |
feat(scaffold): create product workspace + Rust crates + Flutter app
Implements the canonical implementation directory layout adopted by
DEC-022 (register v0.9.5). Closes the scaffolding phase; no PoC code
has been promoted in yet (per proof-of-concept-plan.md §4 a PoC is
not product code unless explicitly promoted).
Rust workspace
==============
Top-level Cargo.toml declares seven workspace members:
core/chanora_core top-level Rust API + orchestration
crates/chanora_protocol tsclientlib isolation (SAD-067, SysDes-011/029)
crates/chanora_state state sync, reducers, deltas
crates/chanora_audio capture, DSP, Opus, jitter, mixer, playback
crates/chanora_storage non-secret DB + platform secure store
crates/chanora_diagnostics logs, redaction, export
crates/chanora_bridge typed Flutter/Rust DTOs
Workspace-wide pins:
license = "MIT OR Apache-2.0" (DEC-020)
rust-version = "1.95"
edition = "2021"
The Flutter app (apps/chanora_flutter) is NOT a Cargo workspace
member; it is owned by the Flutter / Gradle toolchain and is in the
workspace exclude array along with every poc/* spike.
Each crate ships:
* a Cargo.toml referring to workspace.dependencies pins;
* a lib.rs with #![forbid(unsafe_code)] + #![warn(missing_docs)],
a typed Error enum, and the public types relevant to the
subsystem's role per SAD §7.2;
* minimal unit tests so
running 1 test
test tests::defaults_match_decisions ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 1 test
test tests::it_compiles ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 1 test
test tests::session_can_be_constructed ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 1 test
test tests::marker_matches_poc ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 1 test
test tests::it_compiles ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 1 test
test tests::state_transitions_compile ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 1 test
test tests::it_compiles ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s is non-empty.
chanora_core's CoreError type-wraps every subsystem error via
#[from] so callers can match on origin without parsing strings.
chanora_audio's AudioEffects struct defaults all four effects to
true, matching DEC-007 (AEC), DEC-008 (AGC), DEC-009 (NS),
DEC-010 (HPF). A unit test pins this so a future regression that
flips a default fails immediately.
chanora_diagnostics exports REDACTION_MARKER = "[REDACTED]",
identical to the PoC's marker so audit grep patterns survive the
promotion.
chanora_bridge's BridgeError is Serialize + Deserialize so it can
flow across the FRB 2.x boundary (DEC-014).
Empirical verification: cargo check --workspace clean, cargo test
--workspace clean (7 unit tests + 7 doc-test runners, all passing)
against Rust 1.95.0 stable.
Flutter app
===========
apps/chanora_flutter created with .
DEC-004 applied: minSdk overridden to 28 in
android/app/build.gradle.kts with a comment that points back at the
decision register and forbids lowering it without re-opening DEC-004.
DEC-015 applied: shipped English + Simplified Chinese at MVP.
- pubspec.yaml gains flutter_localizations + intl + generate:true.
- l10n.yaml emits lib/l10n/generated/AppL10n (no synthetic package
— that was removed in Flutter 3.41+).
- lib/l10n/app_en.arb is the source of truth; lib/l10n/app_zh.arb
mirrors the key set in zh-Hans. ARB metadata explicitly
reaffirms ADR-008: server-provided content (channel names,
nicknames, welcome banners) is preserved verbatim and never
translated.
lib/main.dart and test/widget_test.dart were rewritten from the
template counter into a minimal localized scaffold that proves
both locales render correctly.
Empirical verification: reports no issues;
runs the two locale smoke tests and both pass.
|
||
|
EdisonJwa
|
0f418f1d7e |
feat(legal): resolve DEC-020 — dual-license under Apache-2.0 OR MIT
Closes the only previously-open decision in the register. Chanora is
now dual-licensed under either:
* Apache License, Version 2.0 (LICENSE-APACHE), OR
* MIT License (LICENSE-MIT)
at the recipient's option. This is the standard Rust-ecosystem
permissive model and is compatible with every direct dependency
in the PoC tree:
tsclientlib MIT OR Apache-2.0
flutter_rust_bridge MIT
cpal Apache-2.0
rusqlite MIT
keyring MIT OR Apache-2.0
hound Apache-2.0
ndk-context, jni, android_logger, regex, serde, tokio,
tracing, thiserror, zeroize, etc. MIT OR Apache-2.0
and with the Flutter framework's BSD-3-Clause.
Files added:
- LICENSE-APACHE Apache 2.0 license text.
- LICENSE-MIT MIT license text with the standard 2026 copyright
line.
Files updated:
- LICENSE Now the dual-license aggregator. Includes the standard
Apache-2.0 inbound-contribution clause ("Unless you
explicitly state otherwise, any contribution
intentionally submitted for inclusion in Chanora by you,
as defined in the Apache-2.0 license, shall be
dual-licensed as above, without any additional terms or
conditions.").
- NOTICE Rewritten with the dual-license declaration and an
inventory of direct dependencies with their upstream
licenses. Transitive deps remain to be enumerated by
build tooling (cargo about, Flutter LicenseRegistry).
- README.md §License section rewritten to point at LICENSE-APACHE
and LICENSE-MIT.
- docs/governance/product-decision-register.md v0.9.5 → v0.9.6:
DEC-020 status: Open → Accepted. §4 license row updated. §6
collapsed: every previously-Proposed or Open decision in the
register is now resolved. DEC-012 legal review remains as a
release-gating *work* item, but is not an open decision.
- docs/governance/poc-results-summary.md v0.3.0 → v0.4.0:
RISK-PoC-003 closed. DEC-020 row moved out of 'Still open'.
This is a license-model commitment, not a substitute for the
DEC-012 legal review. Per DEC-012 the actual legal review work
(transitive-dep OSS obligations, trademark registrability, final
sign-off on the non-affiliation wording) must still be completed
before any public/store release; that is sign-off work, not an
architectural decision.
Decision register state after this commit:
Accepted: 23 of 23 unique decisions
Open/Deferred: 0
Proposed: 0
|
||
|
EdisonJwa
|
a0d1c35461 |
docs(governance): owner confirmation on remaining 17 decisions (register v0.9.5)
Closes the 'Proposed / Owner Confirmation Required' state for every
decision in the register except DEC-020 (license, explicitly deferred
and now the only public-release-gating decision outstanding).
Accepted as recommended:
DEC-001 (Alpha → Beta → Public release sequence),
DEC-002 (all five platforms as MVP target, staged release allowed),
DEC-003 (iOS minimum: iOS 13),
DEC-005 (Android target SDK: Play-required API on upload date),
DEC-006 (single active server connection in MVP),
DEC-007/008/009/010 (AEC + AGC + NS + HPF defaults),
DEC-011 (platform-native audio first),
DEC-012 (legal/trademark/licensing review as a release gate),
DEC-013 (SQLite or equivalent for non-secret state),
DEC-016 (no automatic diagnostics upload),
DEC-017 (crash reporting disabled for MVP),
DEC-018 (product name: Chanora),
DEC-019 (drafted non-affiliation wording),
DEC-021 (Apple App Store SDK gate: Xcode 26+ / iOS 26 SDK+ on/after 2026-04-28).
Modified from the original recommendation by explicit owner ruling:
- DEC-004: Android minimum raised to API 28 (Android 9.0) from the
recommended API 24. Rationale: simpler audio path (AAudio stable
from API 28), narrower compatibility / privacy / scoped-storage
surface. Affects the Android spike's minSdk=24 in product code:
apps/chanora_flutter will need minSdk=28.
- DEC-015: MVP product language expanded to English + Chinese
(Simplified) from the recommended English-only. Rationale: the
demonstrated TS3-compatible-server audience (verified live against
cn.teamspeak.app) and broader TS3 audience include substantial
Chinese-speaking users. Adds zh-Hans translation, font, and
text-length-budget work to MVP. Server-provided content is still
preserved verbatim per ADR-008.
Still Open / Deferred:
- DEC-020 license model. The only remaining release-gating decision.
Documentation updates:
- product-decision-register.md → v0.9.5. §3 statuses updated, §4
renamed Recommended → Accepted with MODIFIED rows annotated,
§6 collapsed to DEC-020 only, §7 dated and statused for every
decision, change-history entry added.
- poc-results-summary.md → v0.3.0. §4 expanded with the
2026-05-14 owner-confirmation pass table. RISK-PoC-004 closed.
New RISK-PoC-006 (Android minSdk move 24 → 28) and RISK-PoC-007
(MVP language expansion to en + zh-Hans) added.
- CHANGELOG entry under [Unreleased].
|
||
|
EdisonJwa
|
4c64517e45 |
docs(governance): promote audio PoC to PASS; close mobile-Android half
Documentation update following the Android audio spike pass.
Decision register (v0.9.3 → v0.9.4):
- DEC-011.1 promoted from
'Accepted (desktop: cpal) / Deferred (mobile)'
to
'Accepted (desktop: cpal; Android: cpal-on-Oboe) / Deferred (iOS)'.
- Evidence pointer added: poc/audio-capture-playback-android-spike/
VERIFICATION.md.
PoC plan (v0.2.0 → v0.3.0):
- Audio row promoted from PARTIAL PASS to PASS.
- All six PoC plan entries are now PASS.
PoC results summary (v0.1.0 → v0.2.0):
- Audio row collapsed into one PASS spanning both spikes.
- RISK-PoC-001 narrowed from 'mobile audio' to 'iOS audio only'.
- Toolchain table expanded with Android NDK, cargo-ndk, AGP/
Gradle/Kotlin, jni/ndk-context/android_logger, and the test
device.
Cross-spike pointers updated:
- poc/audio-capture-playback-spike/VERIFICATION.md result and
follow-up sections updated to reference the Android spike.
- poc/README.md status table lists both audio spike directories.
CHANGELOG updated under [Unreleased].
|
||
|
EdisonJwa
|
ec21a880d2 |
feat(poc/audio): add Android mobile audio spike
Closes the mobile half of the PoC plan §2 audio exit criterion
left open by poc/audio-capture-playback-spike. The desktop and
mobile halves together fully retire the audio PoC.
Stack:
Kotlin (MainActivity) → JNI → Rust cdylib
→ cpal 0.16 → Oboe (AAudio / OpenSL ES) → Android audio HAL
Layout:
rust crate (src/lib.rs) — JNI_OnLoad, initContext,
playSine440, record1sToFile;
panic-catching at JNI boundary;
android_logger → logcat
android/ (Gradle 8.7, — minSdk 24, compileSdk 34, AGP 8.5.2.
AGP 8.5.2, Kotlin 1.9.24) cargoBuildRust task wraps cargo-ndk
-P 26 -t <abi> for all four ABIs;
wired into preBuild so AGP picks up
the produced .so files.
Verified on 2026-05-13 on a physical Motorola Moto G Stylus 5G
(2023), Android 14 SDK 34 arm64-v8a:
- Playback: 500 ms 440 Hz mono sine, 22,050 frames emitted at
44.1 kHz through cpal/Oboe/AAudio/device speaker.
- Capture: 1 s from default input, 42,624 frames written to
/data/data/app.chanora.poc.audio/files/chanora_poc_capture.wav.
File pulled via 'adb exec-out run-as ... cat' and confirmed
by file(1) as 'RIFF (little-endian) data, WAVE audio,
Microsoft PCM, 16 bit, mono 44100 Hz'. Header bytes
cross-checked against the reported frame count.
Notes:
- cpal links libaaudio (introduced API 26), so cargo-ndk targets
API 26 via -P 26 while the Android module's minSdk stays at 24
(DEC-004). API 24/25 devices would fall back to OpenSL ES at
runtime; not exercised here.
- JNI panic safety: every JNI entry point wraps its body in
std::panic::catch_unwind and a tracing panic hook routes
panic messages to logcat under tag 'ChanoraAudioPoC'. Without
this, cpal panicking inside an extern "system" function would
abort the process.
- The emulator AVD chanora-poc-api34 and its system image were
installed during Phase 0 but emulator verification was skipped
once the physical-device run succeeded. Real-device evidence
is stronger.
Surfaced finding: DEC-011.1 mobile half promoted from Deferred to
Accepted for Android in the same docs commit; iOS remains
explicitly Deferred (requires macOS + Xcode hardware).
Authority: PoC plan §2, DEC-011, DEC-011.1.
Not product code; not promoted into chanora_audio.
|
||
|
EdisonJwa
|
eca93a141e |
build(repo): gitignore Android/Gradle build artifacts
Adds .gradle/, local.properties, and **/jniLibs/**/*.so to the ignore list. The jniLibs .so files (600 KB - 950 KB per ABI) are produced by the cargoBuildRust Gradle task wrapping cargo-ndk; they are regenerated on every build and must not be tracked. Required by the next commit (the Android audio spike) which otherwise would try to track those four prebuilt libraries. |
||
|
EdisonJwa
|
271d23faf7 |
docs(governance): record PoC outcomes, owner decisions, and audit evidence
Closes Phases A and D of the post-PoC sequencing.
Decision register (v0.9.2 → v0.9.3):
- DEC-014 Accepted: flutter_rust_bridge 2.x pinned (closed by
poc/flutter_rust_bridge_hello).
- DEC-013.1 Accepted: rusqlite (bundled) (closed by
poc/sqlite-storage-spike).
- DEC-013.2 Accepted: Linux secure-storage backend policy —
Secret Service preferred, keyutils fallback (closed by
poc/secure-storage-spike; resolves SysRS-053 / SysRS-162
ambiguity).
- DEC-011.1 Accepted (desktop: cpal) / Deferred (mobile)
(closed by poc/audio-capture-playback-spike desktop half only).
- DEC-022 Accepted: canonical implementation directory layout per
the README sketch and SAD §7.2.
- DEC-020 explicitly Deferred by owner; remains a public-release
blocker.
Audit reports updated with empirical evidence:
- docs/security/secure-storage-audit-report.md v0.9.3:
SS-AUD-001/002/003/005/006 = PoC Pass with evidence pointers;
SS-TC-003 (Linux) Actual Result populated and Status = PoC Pass;
SS-AUD-004 cross-referenced to diagnostics-redaction PoC;
findings SS-FIND-001 (closed by DEC-013.2), SS-FIND-002 (keyutils
session caveat), SS-FIND-003 (non-Linux adapters still open).
- docs/security/diagnostic-redaction-audit-report.md v0.9.3:
REDACT-TC-001..010 = PoC Pass with evidence pointers; export
bundle policy §5 populated for every row; findings
REDACT-FIND-001 (regex coverage), REDACT-FIND-002
(tracing-layer integration), REDACT-FIND-003 (cross-spike
KnownSecretRegistry contract).
PoC plan (v0.1.0 → v0.2.0):
- Status column added to §2; outcomes recorded.
New doc:
- docs/governance/poc-results-summary.md v0.1.0 — single-page
reviewer-facing summary listing each spike's status, the
toolchain exercised, the owner decisions taken, the audit
coverage table, and open risks RISK-PoC-001..005 (mobile audio,
non-Linux secure-storage adapters, license, remaining
Proposed decisions, no product code yet).
This completes the post-PoC documentation work. Repo is at a clean
pause point: PoC code is committed, owner decisions are recorded,
audit reports carry empirical evidence, and the residual risks are
named in the summary doc.
|
||
|
EdisonJwa
|
181b3d329d |
docs(poc): add PoC index and record PoC outcomes in CHANGELOG
Adds poc/README.md as the top-level index across all six PoC spikes, recording status (5 PASS, 1 PARTIAL PASS), authority, and the non-promotion rule from proof-of-concept-plan.md §4. Updates CHANGELOG.md to enumerate the six spikes with their verification dates and to reference each spike's VERIFICATION.md. This completes Phase E of the post-bootstrap sequencing: E.1 git init + baseline import E.2 justfile E.3..E.8 six PoC spikes E.9 PoC index + CHANGELOG ← this commit |
||
|
EdisonJwa
|
d5b53996bc |
feat(poc/audio): add audio capture/playback spike (partial — desktop only)
Proof-of-concept addressing the audio exit criterion from
docs/architecture/proof-of-concept-plan.md §2:
"Capture/playback works on at least one desktop and one mobile
target."
PARTIAL PASS. The desktop half is verified on Linux; the mobile
half is NOT verified by this PoC and remains a documented open gap.
Implements via cpal (matching DEC-011 'platform-native first'):
- AudioCapture::record_to_wav opens the default input device,
handles f32/i16/u16 sample formats, down-mixes to mono, writes
16-bit PCM WAV via hound.
- AudioPlayback::play_wav opens the default output device, picks
a stream config matching the WAV, blocks until drained.
- synth_sine_wav produces a deterministic 440 Hz test signal for
headless verification of the playback path when no microphone
is available.
- Typed AudioError DTO with NoInputDevice, NoOutputDevice,
DefaultConfig, BuildStream, PlayStream, Wav, Io,
UnsupportedFormat arms.
Verified on 2026-05-13 (Linux + cpal + PipeWire). Capture stream
opened against the system default input; build failed against the
auto_null source (typed AudioError::BuildStream returned cleanly,
demonstrating the production error path); fallback to synth fired;
playback drove 24,000 frames to completion through
Rust → cpal → ALSA → pcm_pipewire → PipeWire → auto_null.
Both audio.rs tests pass.
Mobile gap (explicit, NOT closed):
- Android Oboe path not built or run.
- iOS AVAudioEngine path not built or run.
Surfaced finding for the decision register: DEC-011 does not pin an
audio crate. The PoC uses cpal; production code needs an owner
ruling, ideally after the mobile spike closes the gap.
Out of scope: DSP (HPF/NS/AEC/AGC), Opus encode/decode, jitter
buffer, mixer, latency measurement, bit-exact loopback, device
permission flows. These belong to chanora_audio.
Authority: PoC plan §2, DEC-011, SysDes audio subsystem.
Not product code; not promoted into chanora_audio.
|
||
|
EdisonJwa
|
06ec6f2965 |
feat(poc/diagnostics): add diagnostics-redaction spike
Proof-of-concept proving the diagnostics-redaction exit criterion from
docs/architecture/proof-of-concept-plan.md §2:
"Password and identity-secret samples are redacted."
Full coverage of the audit-report test matrix in
docs/security/diagnostic-redaction-audit-report.md §4
(REDACT-TC-001..010), plus two sanity tests.
The spike ships:
- RedactionPolicy: typed catalogue of regex rules
(identity-base64-blob, password-kv, ts3server-url-password,
authorization-bearer, linux/windows/macos user-path) with
optional capture-group narrowing.
- Structured-field redaction keyed on case-insensitive name
substrings (password, secret, token, ...).
- Bundle-level switches: chat and channel tree excluded by
default per audit-report §5.
- KnownSecretRegistry: literal-substring scrub for secrets the
host application has already loaded into memory (defense in
depth that regexes alone cannot guarantee — closes the gap
behind REDACT-TC-002).
- Length cap (MAX_PROTOCOL_STRING_LEN = 256) with truncation
marker for REDACT-TC-009.
- UTF-8 preserved in non-sensitive fields per REDACT-TC-010 /
ADR-008.
Test suite (12/12 PASS on 2026-05-13):
REDACT-TC-001 server password in connection data
REDACT-TC-002 identity secret in storage error (via KnownSecretRegistry)
REDACT-TC-003 server URL with password field
REDACT-TC-004 chat text excluded by default
REDACT-TC-005 channel name with Unicode excluded by default
REDACT-TC-006 nickname with Unicode preserved in safe field
REDACT-TC-007 local file path user segment minimized
REDACT-TC-008 mixed sensitive bundle (whole-bundle JSON scan)
REDACT-TC-009 long hostile protocol string truncated
REDACT-TC-010 multilingual safe text preserved
+ known-secret literal scrub
+ empty registered secret ignored
Out of scope: tracing-subscriber integration, diagnostic export
file format, memory/core dumps, performance, adversarial regex
evasion beyond trivial cases. These belong to chanora_diagnostics.
Authority: PoC plan §2, docs/security/diagnostic-redaction-audit-report.md,
SRS-093, SysRS-152/154/155.
Not product code; not promoted into chanora_diagnostics.
|
||
|
EdisonJwa
|
52e8d43f69 |
feat(poc/storage): add sqlite-storage spike
Proof-of-concept proving the SQLite-storage exit criterion from
docs/architecture/proof-of-concept-plan.md §2:
"Schema, migration, and repository pattern are demonstrated."
Also satisfies the SRS-089 acceptance criteria explicitly:
"Storage implementation uses an embedded local data store and
migration mechanism."
Implements:
- A forward-only Migrator over a fixed Migration list, tracking
the applied version via PRAGMA user_version. Each migration is
applied inside an IMMEDIATE transaction; rolled back on failure.
- Three canonical migrations (initial schema, add nickname,
add last_connected_at) demonstrating ALTER TABLE flows.
- A LocalDatabaseRepository implementing both BookmarkRepository
and SettingsRepository traits.
- Bookmark.identity_ref is a reference to a secret name, never
a secret value (cross-checked by the secure-storage spike's
SS-AUD-001/002 scans). This is the SAD-067 separation.
Test suite (11/11 PASS on 2026-05-13):
- migrator brings fresh DB to latest version
- migrator is idempotent (no-op when already current)
- migrator applies only pending versions (catch-up upgrade)
- migrator rejects out-of-order versions
- migrator rejects DB newer than known migrations (downgrade guard)
- failed migration rolls back atomically
- bookmark CRUD round-trip
- bookmark list ordered by recency
- bookmark UNIQUE(host, identity_ref) enforcement
- settings upsert + delete
- open creates file and persists across reopen
Surfaced finding for the decision register: DEC-013 does not pin a
SQLite crate. The PoC uses rusqlite with the bundled feature
(no system libsqlite3 dependency); production code needs an
owner ruling on rusqlite vs. sqlx vs. sea-orm.
Authority: PoC plan §2, SRS-089, SDD-077, SAD-067,
SysDes-033/036/049/091.
Not product code; not promoted into chanora_storage.
|
||
|
EdisonJwa
|
50c95b61ad |
feat(poc/storage): add secure-storage spike (Linux)
Proof-of-concept proving the secure-storage exit criterion from docs/architecture/proof-of-concept-plan.md §2: "Secret write/read/delete works through platform secure storage." Implements a typed SecretStorageRepository trait per ADR-006 (SecureStore + per-platform adapters) and a Linux adapter (the only adapter in PoC scope) that supports both equivalent Linux backends per SysRS-053/SysRS-162: Secret Service (libsecret) and kernel keyutils. The audit test suite covers: SS-AUD-001 identity secret absent from local DB (raw file scan) SS-AUD-002 server password absent from local DB SS-AUD-003 secrets absent from logs (Secret newtype redaction) SS-AUD-005 failure returns safe typed error (NotFound) SS-AUD-006 delete removes entry SS-TC-003 Linux round-trip set/get/delete Verified on 2026-05-13 against the local keyutils backend (cargo test runs need 'keyctl session -' to provide a valid session keyring under non-interactive shells, documented in the spike README). The CLI driver additionally observed a real locked gnome-keyring collection and exercised the typed-error → fallback path live. Surfaced finding for the decision register: DEC-013 does not pin a Linux secure-storage backend policy. Both Secret Service and keyutils are 'equivalent' per the requirements; production code needs an owner ruling. Out of scope: Windows DPAPI, macOS/iOS Keychain, Android Keystore, SS-AUD-004 (covered by diagnostics-redaction spike), SS-AUD-007/008 (process / migration items). Authority: PoC plan §2, ADR-006, SDD-078, SRS-091..095, SysRS-158..162. Not product code; not promoted into chanora_storage. |
||
|
EdisonJwa
|
2bbad5feb9 |
feat(poc/bridge): add flutter_rust_bridge hello spike
Proof-of-concept proving the Flutter/Rust bridge exit criterion from docs/architecture/proof-of-concept-plan.md §2: "Flutter can call Rust and receive event stream data." The spike exposes one synchronous fallible command (greet) returning a typed GreetResult / GreetError DTO, and one async event stream (counter_stream) emitting typed CounterTick events. The Flutter app demonstrates both flows on a Material 3 surface; the headless test suite in test/poc_verification_test.dart exercises the same API directly through dart:ffi. Verified on 2026-05-13 (Linux desktop, Flutter 3.41.9 / Dart 3.11.5, flutter_rust_bridge 2.12.0, Rust 1.95). All three tests pass: - greet() returns typed result for valid input - greet() surfaces typed error for empty input - counterStream() delivers the expected event sequence Authority: PoC plan §2, DEC-014 (typed Flutter/Rust bridge), SAD-068, SDD-079, SysDes-049. Naming note: the PoC plan lists this as flutter-rust-bridge-hello, but Dart pubspec.yaml package names require underscores; the directory uses underscores accordingly. Not product code; not promoted into chanora_bridge. Layout note: includes the full Flutter platform scaffold (android, ios, macos, windows, web, linux). Only the Linux desktop target has been built and verified. |
||
|
EdisonJwa
|
02c11ead7e |
feat(poc/protocol): add tsclientlib connect spike
Proof-of-concept proving the protocol-feasibility exit criterion from docs/architecture/proof-of-concept-plan.md §2: "Rust can connect to a compatible server/test double." The spike opens a tsclientlib connection, waits for the BookEvents state snapshot, subscribes to the server channel tree, prints server metadata and the channel tree with client names, and disconnects cleanly. Audio feature is disabled because audio is covered by a separate PoC. Verified on 2026-05-13 against cn.teamspeak.app (TeamSpeak 3 server 3.13.7); 36 channels and 5 online clients retrieved with full UTF-8 (CJK) preservation. See poc/tsclientlib-connect-spike/VERIFICATION.md for the captured run. Authority: PoC plan §2, SysRS-005, SysDes-011, SysDes-029. Not product code; not promoted into chanora_protocol. |
||
|
EdisonJwa
|
bdeab3b451 |
build(repo): add justfile to complete bootstrap v0.1.0
Adds the local task runner required by docs/governance/repository-bootstrap-plan.md v0.1.0 §3, with stubs for format, lint, test, verify-docs, and security-scan. Closes the last gap in repository-bootstrap-plan v0.1.0; CI workflow files remain deferred per the plan. |
||
|
EdisonJwa
|
f1bc9a6c85 |
chore(repo): initial baseline import (docs v0.9.2 + bootstrap)
Imports the v0.9.2 documentation baseline and the bootstrap files required by docs/governance/repository-bootstrap-plan.md v0.1.0 §3, minus the justfile (added in the next commit). This commit establishes the git history for the project. All previous work lived only as filesystem state with no version control. |