mirror of
https://github.com/QwenLM/qwen-code.git
synced 2026-07-10 01:29:17 +00:00
68 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Dragon
|
067cfbba62
|
docs: consolidate design docs and plans under docs/ (#6417)
Design docs and implementation plans were scattered across .qwen/design, .qwen/plans, and docs/superpowers. The .qwen/ locations are git-ignored, so docs written there never got tracked, while docs/design already held the richer, version-controlled set. Consolidate everything under docs/design and docs/plans, relocate two stray root docs into docs/design, and repoint the references left dangling by the move (moved-doc cross-links and a few source comments). Also update AGENTS.md and the feat-dev skill so the documented workflow writes new design docs and plans to the tracked docs/ locations. Co-authored-by: DragonnZhang <dragonzhang1024@gmail.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
zhangxy-zju
|
1783ae86f3
|
docs(web-shell): document chart renderer integration (#6353)
* docs(web-shell): document chart renderer integration * docs(web-shell): describe daemon-backed chart artifacts * docs(web-shell): clarify chart ref validation layers |
||
qqqys
|
b23f888d73
|
[codex] add proactive channel loop tools (#6287)
* feat(channel): add proactive loop tools * fix(channels): stabilize proactive loop routing * fix(channels): gate loop tools in shared sessions * fix(channels): tighten channel loop tool routing * fix(channels): close loop tool review blockers * fix(dingtalk): preserve markdown tables * fix(dingtalk): use app token for reactions * fix(channels): scope loop tools to active caller * fix(channels): preserve group session metadata * fix(channels): normalize loop targets * test(cli): cover settings cron disable path * fix(channels): address dingtalk review suggestions * fix(dingtalk): restore table normalization * fix(channels): mark loop tool failures * fix(channels): tighten loop mcp protocol handling * test(channels): cover loop tool guard paths * fix(channels): await loop mcp registration * test(channels): preserve base proactive target default * refactor(channels): clarify loop target promotion * fix(channels): harden loop recurring input * fix(channels): ack loop mcp notifications * fix(channels): preserve legacy loop targets * test(channels): cover channel loop wiring paths * fix(channels): retry skipped loop mcp registration * fix(channels): keep promoted loop targets visible * fix(channels): harden loop mcp input logging |
||
Heyang Wang
|
3bf0fa0af0
|
Feat: LSP Server support hot reload (#5953)
* feat(core): Add LSP server config hot-reload support - Implement reconcileServerConfigs to diff desired vs current LSP configs and apply minimal add/remove/restart operations with a serialized reconcile queue - Add configHash utility to detect config changes via stable hashing - Add lspConfigWatcher in CLI to watch .lsp.json and trigger reconciliation on file changes - Extend LspServerManager with per-server config hash tracking and detailed debug logging - Add design docs for LSP runtime reinitialization and hot-reload overview - Include comprehensive unit tests for all new modules * refactor(cli): Extract registerLspHotReload from main function Move the LSP config file watcher setup and reconciliation logic into a dedicated module-private function registerLspHotReload, reducing the size and nesting depth of the main startup flow. Added a JSDoc summarizing responsibilities, early-return conditions, and the AppEvent.LspStatusChanged side effect. * fix(lsp): release server resources during reload * fix(lsp): address hot reload review feedback * fix(lsp): harden hot reload reconciliation * docs(lsp): update hot reload design notes * fix(lsp): harden hot reload retry semantics * fix(lsp): harden hot reload lifecycle * fix(lsp): harden hot reload lifecycle * fix(lsp): isolate hot reload recovery paths * fix(lsp): align command probes and replay tracking * fix(lsp): prevent crash restarts during shutdown * fix(lsp): preserve reload state across failures * fix(lsp): cancel reloads during shutdown * fix(lsp): handle socket startup races * fix(lsp): harden command probe env and socket startup * fix(lsp): report skipped reload and restart states * fix(lsp): harden hot reload lifecycle cleanup * chore: add one comment for `Object.create(null)` --------- Co-authored-by: heyang.why <heyang.why@alibaba-inc.com> Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> |
||
|
Heyang Wang
|
5d0733f79c
|
feat(core): stabilize tool schema declaration order (#6339)
Make tool declaration ordering deterministic so prompt-cache prefixes do not depend on asynchronous registration history. - Sort function declarations by canonical tool name after existing visibility filtering - Preserve deferred, revealed, and alwaysLoad filtering semantics - Add tests covering deferred tools, revealed tools, and MCP registration order - Document the prompt-cache motivation and next-step cache break detection plan Co-authored-by: heyang.why <heyang.why@alibaba-inc.com> |
||
ChiGao
|
b13032d3ae
|
docs(design): daemon side-channel coordination (A1/A2/A4/A5) (#4511)
Co-authored-by: jinye <djy1989418@126.com> Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> Co-authored-by: 易良 <1204183885@qq.com> |
||
|
zhangxy-zju
|
e9a7917d5e
|
feat(web-shell): support compact echarts full data blocks (#6232)
* feat(web-shell): support custom code block rendering * fix(web-shell): harden custom code block rendering * docs: add skill capability gating design * fix(web-shell): make chart skill host supplied * docs(web-shell): write chart skill in English * docs(web-shell): document full-data chart payload * docs(web-shell): use dataset-backed chart payload * feat(web-shell): add echarts full-data renderer * chore(web-shell): keep chart skill host supplied * fix(web-shell): show loading for streaming chart blocks * style(web-shell): polish echarts full-data renderer * fix(web-shell): harden custom code block language parsing * fix(web-shell): harden echarts full-data renderer * fix(web-shell): polish echarts renderer followups * fix(web-shell): reuse enhanced table for chart data * fix(web-shell): recover chart renderer after errors * fix(web-shell): harden chart option handling * fix(web-shell): harden chart data rendering * fix(web-shell): tighten chart renderer guardrails * fix(web-shell): update chart fallback title * fix(web-shell): polish chart renderer review fixes * feat(web-shell): support compact echarts full data blocks * docs(web-shell): add chart skill template * fix(web-shell): address chart review suggestions * fix(web-shell): preserve punctuation language aliases * fix(web-shell): address chart follow-up review * fix(web-shell): harden chart ref resolution * fix(web-shell): cover chart sanitizer follow-ups * fix(web-shell): address chart review follow-ups * fix(web-shell): address chart review leftovers * fix(web-shell): close chart review gaps * fix(web-shell): handle latest chart review |
||
callmeYe
|
fe3dd93e8f
|
Add sessionless workspace memory forget and dream (#6227)
* feat(serve): add sessionless memory forget and dream * fix(serve): thread abort through memory forget * fix(serve): address workspace memory review feedback * fix(serve): address memory review follow-up * fix(memory): harden forget review paths * fix(serve): classify memory availability failures * fix(serve): document memory task capacity tiers * fix(memory): address review edge cases * chore: remove mobile-mcp formatting noise |
||
|
ChiGao
|
9658dccfbb
|
feat(daemon): add session artifact APIs (#5895)
* docs: add session artifacts daemon API design * docs: tighten session artifacts design scope * docs: frame artifacts API as complete v1 capability * docs: address artifacts review follow-ups * docs: clarify artifacts reset boundary * docs: clarify batch hook artifact flow * docs: address latest artifact design audit * docs: tighten artifact event and store semantics * docs: simplify artifact v1 merge policy * docs: resolve artifact v1 review blockers * docs: tighten artifact trust and retention semantics * docs: close artifact v1 boundary gaps * feat(daemon): add session artifact APIs * fix(daemon): harden session artifact semantics * fix(sdk): update daemon browser bundle budget * fix(daemon): tighten artifact ingestion boundaries * fix(daemon): cache artifact workspace realpath * fix(daemon): sanitize artifact add dispatch input * docs(daemon): align artifact change wire shape * fix(daemon): harden artifact status validation * test(daemon): cover artifact acp dispatch * test(daemon): update artifact capability baseline * fix(daemon): clear workspace locator on published artifacts * fix(core): forward post-tool batch artifacts * fix(daemon): harden artifact status refresh * fix(daemon): guard artifact event ingestion * test(daemon): cover non-strict artifact drops * fix(core): align artifact display validation * fix(daemon): serialize artifact store operations * chore(daemon): clarify artifact publisher tool name * fix(daemon): coordinate artifact route mutations * fix(daemon): harden artifact refresh comparison * fix(daemon): harden artifact ingress edge cases * fix(daemon): guard artifact rpc mutations during archive * fix(daemon): gate session metadata mutation auth * fix(daemon): harden artifact route boundaries * fix(channels): compact drained group history * fix(daemon): address artifact review findings * fix(daemon): address artifact review follow-ups * fix(daemon): preserve hook artifact success output * fix(daemon): handle artifact review edge cases * fix(daemon): address artifact review hardening * test(daemon): cover artifact review edge cases * fix(daemon): validate hook artifact aggregation * fix(daemon): improve artifact ingestion diagnostics * fix(daemon): address artifact review feedback * fix(daemon): address artifact review feedback * fix(daemon): harden session artifact ingress * fix(daemon): harden artifact edge cases * fix(daemon): tighten artifact path validation * fix(daemon): address artifact review races * fix(daemon): surface artifact path inspection errors * fix(daemon): forward batch hook artifacts in ACP * fix(daemon): clean artifact bridge metadata * test(daemon): cover artifact store edge cases * fix(daemon): resolve artifact file url symlinks * fix(daemon): harden artifact ingestion paths * fix(daemon): harden artifact review paths * test(daemon): cover artifact tool name sync * fix(daemon): harden artifact republish validation * chore(daemon): remove unrelated artifact PR churn * fix(daemon): address artifact review gaps * test(daemon): cover artifact url rejection * chore(daemon): drop unrelated formatting churn * chore(daemon): update settings schema * fix(daemon): harden artifact validation * fix(daemon): tighten artifact event validation * docs(core): clarify artifact env flag comment * test(cli): align soft failure artifact expectation * fix(daemon): address artifact review edge cases * fix(daemon): enable artifact metadata recording * fix(daemon): harden artifact store review paths --------- Co-authored-by: 秦奇 <gary.gq@alibaba-inc.com> Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> |
||
|
ChiGao
|
c302e3ec29
|
feat(daemon,sdk): resumable /acp session stream (Last-Event-ID) + opt-in SDK transports export (#5852)
* fix(daemon): resume /acp session stream via Last-Event-ID (recover mid-turn content)
The `/acp` Streamable-HTTP session event stream was live-only: it emitted no
SSE `id:` sequence and ignored a `Last-Event-ID` reconnect header. When a
control-plane proxy idle-closed the long-lived SSE mid-turn, every content
frame the daemon produced during the gap (`session/update` carrying
agent_thought_chunk / agent_message_chunk) was lost — the turn still settled,
so the UI showed "done" with an empty/truncated body, and only a re-send
recovered it (tracked as §1.8 in the integration notes).
The replay engine already exists and is battle-tested on the REST surface:
EventBus assigns a monotonic per-session `id`, keeps a bounded ring, and
`subscribeEvents({ lastEventId })` replays `id > lastEventId` before live
events flow. This wires the `/acp` transport to it — no eventBus/bridge change.
- transport-stream / sse-stream / ws-stream: `send(message, id?)`. SSE emits an
`id:` line when `id` is present (mirrors REST `formatSseFrame`); WS ignores it
(stateful, no replay).
- connection-registry: `sendSession(…, id?)` threads the cursor; the pre-attach
session buffer stores `{ frame, id? }` so a buffered frame keeps its `id:`.
- dispatch: `translateEvent` passes `event.id` for bus events; `pumpSessionEvents`
forwards `lastEventId` to `subscribeEvents`.
- index: the `GET /acp` session branch reads `Last-Event-ID` (strict
decimal-only parse, same rule as REST) and passes it to the pump.
Bus-originated frames (session/update, request_permission, daemon notifies)
carry an `id:`; JSON-RPC responses and synthetic terminal frames do not, so
they don't burn a slot in the resume sequence. Backward compatible: clients
that send no `Last-Event-ID` get live-only behaviour as before, and `id:`
lines are inert for clients that ignore them.
Design: docs/design/daemon-acp-http/sse-resumable-stream.md
* fix(daemon): make /acp resume actually engage — session-stream grace/reclaim + replay guards
Addresses three review Criticals on the §1.8 plumbing: on its own the
`id:`/`Last-Event-ID` wiring never fired in the real close-then-reconnect flow,
and once it does fire two replay-correctness gaps become reachable.
1. Session-stream grace/reclaim (the core fix). A transport-level session-stream
close used to run the FULL `closeSessionStream` teardown — removing
ownership, aborting the in-flight prompt, detaching the bridge client. In the
real EventSource/proxy order (old socket closes first, then reconnect) that
meant the reconnect carrying `Last-Event-ID` was rejected 403 before the
cursor was read, and the prompt was already aborted — so replay had nothing
to resume. Now a transport close DETACHES (`detachSessionStream`): it stops
only the stream + subscription and keeps the binding, ownership, prompt, and
bridge-client alive for a grace window (`SESSION_GRACE_MS`, mirrors
`CONN_GRACE_MS`). A reconnect within the window reclaims (clears the timer);
otherwise the grace timer runs the full teardown, bounding runaway cost. Full
teardown stays immediate for explicit `session/close` and connection destroy.
The GET handler branches on `stream.isClosed` (transport close → grace;
pump-ended-while-open → full close).
2. No double-delivery (buffer ↔ ring overlap). `attachSessionStream` records the
max bus id flushed from the pre-attach buffer; the GET handler advances the
replay cursor to `max(Last-Event-ID, lastFlushedEventId)` so the ring replay
doesn't re-emit an already-flushed frame.
3. Idempotent `permission_request` under replay. `translateEvent` reuses the
existing `conn.pending` entry for a `bridgeRequestId` (re-sends the same
outbound id) instead of minting a second id+entry — no orphan pending, no
duplicate prompt on a ring-replayed permission.
Also: extract `parseLastEventId` to a shared `serve/sse-last-event-id.ts` used
by both REST and `/acp` (no drift; logs the rejected value); log `lastEventId`
in the pump error.
Tests: real close-then-reconnect order (200 not 403 + prompt not aborted);
overflow Last-Event-ID; replayed permission reuses pending id; registry
grace/reclaim + buffer-flush-preserves-id. Full acp-http suite green (216).
* feat(sdk): expose ACP transports via opt-in ./daemon/transports subpath
The resumable ACP-over-HTTP transport (AcpHttpTransport, native
supportsReplay + Last-Event-ID) and the negotiateTransport factory were
reachable only from source paths inside the monorepo — the published
`@qwen-code/sdk/daemon` barrel intentionally omits them to keep its
budget-checked browser bundle lean, so external consumers (agent-web)
had no import path short of forking.
Add a separate opt-in subpath `@qwen-code/sdk/daemon/transports` that
ships AcpHttpTransport / AcpWsTransport / AutoReconnectTransport /
RestSseTransport / negotiateTransport as their own browser+node bundle.
The default `./daemon` barrel and its byte budget are unchanged, so
REST-only consumers stay tree-shaken and pay nothing for the transports.
Also add a `fetchFn` option to NegotiateTransportOptions so callers can
inject auth/proxy/test fetch instead of the hardcoded global.
- build.js: emit dist/daemon/transports.{js,cjs}; reuse the node-builtin
guard for the new browser bundle (no size budget — it legitimately
ships the transports) while keeping the default barrel's budget check.
- daemon/index.ts: update the rationale comment to point at the subpath.
- daemon-transports-surface.test.ts: lock the runtime + type surface and
the package.json exports entry.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): resume cursor must not skip in-flight-lost frames
Round-3 review (qwen-code-ci-bot) flagged a silent-frame-loss Critical in
the §1.8 resume path I added: `resumeCursor = max(Last-Event-ID,
lastFlushedEventId)` advances the ring-replay cursor past the buffer, but a
frame sent to the now-dead socket yet never received by the client has a bus
id BELOW the buffer's ids and ABOVE the client's cursor — so the max() skips
it and the ring replay never re-emits it. Exactly the proxy idle-close
mid-turn frame §1.8 is meant to recover.
Fix without trading loss for duplicates: a buffered bus event is ALSO in the
EventBus ring (it was published there to get its id), so the ring replay
started at the client's cursor is the single delivery path for every bus
event after the cursor. `attachSessionStream` now takes the resume cursor and,
when resuming, does NOT flush id-bearing buffered frames — the ring owns them,
delivering each exactly once including the in-flight-lost frame. Id-less
frames (JSON-RPC replies via `replySession`, not ring events) are still
flushed — their only delivery path. The GET handler sets
`resumeCursor = lastEventId` verbatim; `lastFlushedEventId` is removed.
Also from the same review:
- sse-last-event-id `safeLogValue`: strip ALL C0 control chars + DEL (not just
CR/LF) so a crafted `Last-Event-ID` can't smuggle ANSI ESC / null bytes onto
an operator's terminal via stderr.
- ws-stream: regression test asserting `send(msg, id)` keeps the WS wire frame
bare JSON (no SSE `id:` framing leak).
- connection-registry: resume-path test (id-bearing frames skipped, id-less
reply still flushed); design doc updated.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(daemon): inline resumeCursor alias to lastEventId
Review nit (yiliang114): after the prior commit dropped the `max()` logic,
`resumeCursor` is a pure alias for `lastEventId`. Use `lastEventId` directly in
the `pumpSessionEvents` call and the error log; drop the alias.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* docs(daemon): refresh stale resume comments + add gap-delivery test
Round-4 review (qwen-code-ci-bot, against the now-corrected resume model):
- connection-registry: the attachSessionStream CONTRACT comment still cited a
`promptAbort?.abort()` call in the index.ts onClose handler that an earlier
commit removed. Rewrite it to describe the current model — each stream's pump
has its own abort controller and teardown is identity-guarded in
`onPumpSettled`, so installing the new stream first makes the old stream
settle into detach-with-grace rather than tearing down the in-flight prompt.
- dispatch: the stream_error frame comment ("no bus id, so no SSE id: line")
contradicted the code passing `event.id`. Make it truthful: pass the cursor
through if present; a synthetic terminal frame has no id so none is written.
- connection-registry.test: add the explicit detach → produce gap events →
reattach → flush-exactly-once test (the PR's core value prop at the registry
layer), incl. a second reattach asserting the buffer drained.
The two Criticals in the same review referenced `resumeCursor` /
`lastFlushedEventId` / `Math.max`, all removed in prior commits — obsolete
against current code (answered + resolved on the threads).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): preserve stream order on resume + harden grace/permission paths
Round-5 review (wenshao + qwen-code-ci-bot):
- [Critical, wenshao] Out-of-order completion on resume. attachSessionStream
flushed id-less buffered JSON-RPC replies (e.g. a session/prompt result that
landed during the detach gap) immediately — ahead of the ring replay that
redelivers the content chunks preceding them, so a client could see "prompt
complete" before the body (the truncated-body failure §1.8 fixes). Now on
resume those id-less frames are DEFERRED in the buffer; the event pump
releases them via flushBufferedSessionFrames once the replay boundary
(replay_complete / state_resync_required) passes, preserving original order.
Fresh connects (no cursor, no replay) still flush the whole buffer in order.
- [Critical, ci-bot] Permission auto-denied during the reconnect grace window:
a permission_request arriving while binding.stream is detached cancel-denies,
so a client reconnecting within grace can't vote. The structural fix (defer
the vote across grace) belongs with the §1.7 permission-coordination
follow-up; here, log an operator breadcrumb when it fires during grace, and
document the synchronous-translateEvent INVARIANT the direct binding.stream
.send relies on.
- [ci-bot] Stale-stream detach is now tested (reclaim installs s2; a late s1
close is a no-op — no teardown, no grace re-arm). Grace-expiry teardown now
logs a breadcrumb so a vanished session is distinguishable from explicit
close. TS4111: bracket-access the index-signature exports entry in the SDK
surface test. transports browser bundle now has a size budget
(MAX_TRANSPORTS_BROWSER_BUNDLE_BYTES = 48KB; current ~29KB).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): route session-scoped /acp responses so prompts don't hang
[Critical, wenshao] The published AcpHttpTransport could hang real
session/prompt + config requests. Its subscribeEvents() opened REST
GET /session/:id/events and only sendRequest's connection-scoped stream
resolved responses — but the daemon's replySession() routes session-scoped
JSON-RPC replies onto the session-scoped /acp stream, which the transport
never read. So a session/prompt reply was never observed → the pending
request never settled.
Switch subscribeEvents to the session-scoped /acp stream (GET /acp +
Acp-Session-Id) — the resumable §1.8 stream the daemon puts session replies
on — and dispatch each raw JSON-RPC frame by shape:
- response (id, no method) → resolve the shared pending map (the fix)
- notification (method, no id) → DaemonEvent via denormalizeAcpNotification,
stamped with the real bus id from the SSE
`id:` line (the synthetic denormalizer id is
not resume-compatible; supportsReplay=true
now tracks the authoritative cursor)
- session/request_permission → surfaced as a permission_request event so
consumers still see prompts (responding to
the vote is the §1.7 follow-up)
The connection-scoped stream still carries replies to connection-level
requests (initialize, session/new). Adds 4 subscribeEvents unit tests
(stream selection + headers, notification→event+busId, response consumed-not-
yielded, permission surfaced). Full SDK suite green (1062).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk,daemon): harden /acp SSE parser + grace/flush observability
Round-6 review (qwen-code-ci-bot), all additive / no behavioural side effects:
- [Critical] Unbounded SSE buffer in the new AcpHttpTransport session-stream
parser → OOM (tab crash for browser consumers). Add a 16 MiB cap mirroring
parseSseStream's MAX_BUF_CHARS, and reuse parseSseStream's CRLF-aware
`consumeFrames` splitter (now exported) instead of an inline `\n\n` scan —
closing the CRLF, multi-line `data:` join, and trailing-CR gaps in one go.
- Deferred-flush ordering race: the pump's post-loop safety flushDeferred()
now runs only on a non-aborted exit. An abort means the stream was
detached/reclaimed; flushing there could drain the deferred reply onto a
reclaiming stream ahead of its own replay (reintroducing the out-of-order
delivery the deferral prevents). On error the frames stay buffered for the
next attach — never lost.
- Grace reclaim now logs (detach + grace-expiry already did) so the reconnect
trail is complete for operators.
- sse-last-event-id doc: corrected the "shared by REST and ACP" claim — after
the #5809 serve-route split REST keeps its own copy; unifying them would
touch REST, so it's deferred (this PR keeps REST untouched).
Thread on a full deferred-flush integration test: the ordering invariant is
already locked at the unit layer (flushBufferedSessionFrames defer test +
gap-delivery test); a full-HTTP timing test against the FakeBridge would be
flake-prone, so it's intentionally not added.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon,sdk): harden /acp resumable stream from review round 7
Address review-pr findings on the §1.8 resumable stream, all additive /
backward-compatible (REST untouched, no behavioural side effects):
- connection-registry: guard flushBufferedSessionFrames against a closed
stream so deferred replies stay buffered for the next reconnect instead
of being dropped onto a dead socket. Keep the synchronous in-order
enqueue (SseStream serializes via one writeChain) — an await-per-frame
drain would let a live event interleave between deferred frames and
reorder the very replies this deferral preserves (W1).
- connection-registry: log at the moment of detach so an operator can
measure the real disconnect→reconnect gap against the grace window.
- index: route sessionId through logSafe() in the event-pump error log,
matching every other log line this PR adds (terminal-escape hardening).
- AcpHttpTransport: remove the abort listener in the finally block so a
long-lived signal reused across reconnects doesn't accumulate listeners.
- AcpHttpTransport: parse the SSE `id:` cursor with the server's strict
/^\d+$/ + MAX_SAFE_INTEGER rule instead of lenient Number() (rejects
proxy-mangled hex/exponential/empty cursors).
- AcpHttpTransport: document that an unparseable non-empty data frame is a
corrupt frame (not a heartbeat); tracing it is a follow-up once the SDK
grows a logger (the package lint config forbids console).
- tests: add sse-last-event-id.test.ts (parseLastEventId accept/reject +
safeLogValue control-char stripping/truncation) and a
flushBufferedSessionFrames closed-stream-retains-buffer case.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon,sdk): round-8 review hardening for /acp resumable stream
All additive / backward-compatible (REST untouched, no behavioural side
effects):
- AcpHttpTransport: attach a no-op catch to abortPromise so an
already-aborted signal at entry (loop never enters, Promise.race never
consumes the rejection) can't surface as an unhandled rejection.
- AcpHttpTransport: document that opts.maxQueued does not apply to the
/acp transport (the session stream is backed by the daemon's
server-controlled EventBus ring; there is no client-tunable queue to
forward it to) — intentionally ignored, not silently mis-applied.
- index: run err.message through logSafe() in the event-pump error log
(CR/LF/ANSI in a bridge error string would otherwise reach stderr raw),
and add operator breadcrumbs for the previously-silent onPumpSettled
branches (pump-ended-while-open full close; superseded-stream no-op),
completing the detach/reclaim/grace trail.
- tests: assert subscribeEvents writes Last-Event-ID on the outbound GET
when resuming and omits it on a first connect (the resume cursor must
reach the wire), plus an already-aborted-signal case that would fail on
an unhandled rejection without the catch above.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): flush deferred /acp replies on replay_complete only
The EventBus emits `state_resync_required` BEFORE the replay frames (the
`epoch_reset` and `ring_evicted` paths both fall through to the replay
loop and still emit `replay_complete` at the end). The pump was releasing
the deferred id-less replies on EITHER boundary, so on a resync-triggering
resume the buffered `session/prompt` result was flushed ahead of the
replayed content chunks — the exact truncated-body reordering §1.8 fixes
(client sees "done" before the body).
Flush on `replay_complete` only. The live-only case (no cursor ⇒ no replay
⇒ no `replay_complete`) is still covered by the pump's post-loop safety
flush. Add an over-the-wire integration test (resume with a reply buffered
during the detach gap, bridge replays resync → content → replay_complete)
asserting the reply lands AFTER the replayed content; verified it fails
against the previous dual-boundary flush. Design doc updated.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): close two §1.8 grace/replay-ordering holes
Both additive / in-scope / no REST change:
- Replay-window reply ordering (connection-registry, dispatch): the
resumptive-attach deferral only covered id-less replies ALREADY buffered
from the detach gap. A prompt that finished AFTER the new stream attached
but BEFORE replay drained went straight out live via `sendSession`,
overtaking replay frames not yet sent. Add a per-binding `replayPending`
flag (armed on resumptive attach, cleared on `replay_complete` in
`flushBufferedSessionFrames`) and route `replySession`'s out-of-band
replies through a new `sendSessionReply` that defers while it's set.
In-band pump frames keep using `sendSession`, so the `replay_complete`
frame itself can't be deferred (which would deadlock the release).
- Connection reaper vs session grace (index, connection-registry): the
conn-stream-close reaper treated only LIVE session streams as activity,
so a session detached into its own `SESSION_GRACE_MS` window (stream
undefined, graceTimer armed) didn't count — the connection could be
reaped at `CONN_GRACE_MS`, 404-ing the imminent session resume and
aborting the in-flight prompt early. Add `hasRecoverableSession()` and
treat a grace-armed session as activity in the reaper guard.
Unit tests for both at the registry layer.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): thread query params into ACP transport route extractors
The exported ACP HTTP/WS transports reduced request URLs to `pathname`
before the route table built JSON-RPC params, so every query parameter
from the REST-style DaemonClient helpers was dropped — e.g.
`readWorkspaceFile('a.ts', { maxBytes: 123 })` (`/file?path=a.ts&maxBytes=123`)
produced `_qwen/file/read` with `params: {}`. Same for `/file/bytes`,
`/stat`, `/list`, `/glob`, and `context-usage?detail=true`.
Pass `parsedUrl.searchParams` into `extractParams` and coerce each query
value to the type the daemon's ACP handlers require — the daemon validates
`maxBytes`/`line`/`limit`/`offset` as real numbers and `detail` as the
boolean `true`, neither of which a raw query string satisfies. Helpers
`strParam`/`numParam`/`boolParam` keep the per-route extractors terse.
`query` is optional so the existing path-only extractors are unaffected.
(`/workspace/voice/transcribe` has no ACP route at all — separate gap,
binary audio doesn't belong on the JSON-RPC transport.)
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): route session-stream replies via a background pump (no-subscriber prompt)
The daemon answers POST /session/:id/prompt (and session/cancel,
set_config_option, set_mode, set_model) with 202 and routes the JSON-RPC
result onto the SESSION stream via replySession — not the connection
stream the transport pumps. So a DaemonClient that calls prompt() but
never iterates subscribeEvents had nothing reading that reply, and
sendRequest()'s pending promise never resolved → prompt() hung forever.
For these session-reply methods, sendRequest now opens a reference-counted
background session-reply pump (GET /acp + Acp-Session-Id) that routes
JSON-RPC responses to `pending`, released when the request settles. It's
suppressed when a subscribeEvents consumer is already iterating that
session (tracked via activeSessionSubscriptions) — the daemon's session
stream is single-reader, so a competing GET would detach the consumer's;
in that case the consumer already routes the reply (the W2 fix). The pump
skips notifications and permission requests (method-bearing frames) so a
permission request id can't be mis-routed onto a pending response slot.
All five methods require an owned session, so the pump's GET is always
authorized. Disposed pumps are aborted in dispose().
Verified the new test times out without the pump and passes with it.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve): sequence /acp deferred replies by bus watermark + harden grace/reap
Address review on the §1.8 resumable-stream fixes:
- replayPending is now set from the current attach mode every time
(resume arms, fresh connect clears) so an aborted resume that skipped
its boundary flush can't strand the flag and buffer every later reply
forever (MsyIq, MylZ4).
- Deferred out-of-band replies carry a watermark (anchorId = bus head at
produce time) and release only once the pump delivers through that id,
via per-event releaseDeferredSessionReplies + endReplayDeferral at
replay_complete. A result produced during a slow replay no longer jumps
ahead of tail content still flowing as live events behind the boundary
(MsyIt). Unanchored fallback replies still release at the boundary.
- Connection reap re-evaluates after a session reclaim grace expires
(connGraceExpired + onSessionGraceExpired), so a conn blocked from
reaping by a then-recoverable session no longer lingers to the 30-min
idle sweep (MsyIs).
- Wrap the grace-timer teardown in try/catch so a throwing detach
callback can't crash the daemon from a bare setTimeout (MylZ8).
- sse-last-event-id reuses the shared logSafe sanitizer (covers C1 +
Unicode bidi) instead of a narrower divergent regex (M1isz); refresh
stale replayPending/flush JSDoc (MselO).
Unit tests cover the replayPending reset, watermark ordering, grace
expiry hook, and grace-timer try/catch.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): scope pending sweeps per stream + reset bus cursor on invalid id
Address review on the ACP HTTP transport:
- Tag each pending request with its routing scope (connection vs a
sessionId). A connection-stream failure now sweeps only conn-scoped
pendings, so it can't reject a session/prompt the session stream is
about to resolve; the session reply pump mirrors this for its own
scope (MselM).
- An invalid id: line later in an SSE frame resets the bus cursor to
undefined rather than carrying a stale earlier value into the event
(MselW).
- Strengthen the W2 response-routing test to register a pending request
and assert the frame RESOLVES it, not merely that it isn't yielded
(MylZ-). Add tests for the per-stream sweep partition and the id reset.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): accept `kind`-tagged _qwen/notify envelopes (don't drop resume signals)
The daemon's session-stream translateEvent stamps `_qwen/notify` events
under `kind` (state_resync_required, replay_complete, stream_error,
model_switched, …), but denormalizeAcpNotification read only `type` and
returned undefined for them — so subscribeEvents silently dropped every
such event. During a ring-overflow resume the SDK would never see
state_resync_required and would apply replayed events to stale state.
Read `params['type'] ?? params['kind']` (preferring `type`, so other
producers are unaffected) and add an SDK test feeding a `kind`-tagged
notify through subscribeEvents (M2bvl).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): reject session-scoped pendings when the subscription stream closes
A `session/prompt` reply routed through an active subscribeEvents consumer
(no reply pump is started while a subscription is live) would hang if that
session SSE stream closed before the reply arrived: the connection-stream
catch only sweeps conn-scoped pendings, and subscribeEventsInner's finally
cleaned up the reader but never the pendings.
Sweep session-scoped pendings in that finally too, gated so it only fires
when this is the session's last delivery route (no other active
subscription — the ref-count still includes self here — and no reply pump),
mirroring the reply-pump and connection-stream sweeps. Add a regression
test (M2iHz).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): harden ACP SSE readers + reply-pump handoff + empty query param
Address the latest review wave on the transport:
- pumpConnStream now bounds its unread SSE buffer with the same
MAX_SSE_BUF_CHARS guard the two session readers already have — the OOM
vector (a server that never emits a `\n\n` boundary) was open on 1 of 3
readers — and attaches the no-op abortPromise.catch() crash guard (M3BYQ).
- pumpSessionReplies mirrors subscribeEventsInner's abort handling: named
listener ref removed in finally (no leak on a clean drain of a reused
signal) + abortPromise.catch() so a pre-aborted signal can't surface an
unhandledrejection; and it throws the HTTP status on a non-OK response so
the failure is diagnosable rather than a silent void return (M3BYT, M3BYY).
- subscribeEvents aborts any existing background reply pump for the session
before opening the consumer stream. The single-reader session stream
detaches the pump anyway; aborting it skips its teardown sweep so it can't
spuriously reject the very `session/prompt` the consumer now delivers (M3BYa).
- numParam treats an empty value (`?maxBytes=`) as absent, not Number('')===0
(M3BYd).
Tests: empty-numeric-param omission, and the reply-pump abort-on-subscribe
handoff (pump aborted, its pending not rejected).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve): log a breadcrumb when the replySession anchor is unavailable
The getSessionLastEventId fallback (deferring a reply unanchored when the
ACP binding briefly outlives the bridge session) was silent. Emit a scoped
stderr breadcrumb so an operator can tell that benign teardown race apart
from an unexpected bridge regression that starts exercising the fallback
(M3BYf).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): validate content-type in pumpSessionReplies before parsing as SSE
pumpSessionReplies fed any 2xx body straight to the SSE frame parser. A
non-SSE response (an HTML error page / a JSON proxy error injected by a
CDN) would be consumed as garbage or hang the pump waiting for `data:`
lines that never arrive — strictly weaker validation than its sibling
subscribeEventsInner, which already guards content-type.
Mirror that guard: between the res.ok check and getReader(), reject a body
that isn't text/event-stream (cancelling it first). Add a test that a
no-subscriber session/prompt whose reply pump GET returns text/html
rejects instead of hanging (M3pAM).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): close reply-pump handoff strand race + scope-guard reply resolution
Address the latest review wave:
- subscribeEvents now removes the aborted reply pump's map entry
SYNCHRONOUSLY, not just aborting it. Otherwise, if the subscription
exited before the pump's async `.finally` deleted the entry, BOTH
stranded-pending guards missed (the consumer sweep saw the entry still
present and deferred; the pump's sweep skipped on abort) — a live
session/prompt stayed in `pending` forever. Synchronous removal makes the
consumer sweep deterministically responsible (M3w6Y).
- Reply resolution (both the session-reply pump and the subscribeEvents
consumer path) now skips a reply whose pending is scoped to a DIFFERENT
session — defense-in-depth against a future daemon misroute silently
cross-delivering across the SDK boundary (M3w6d).
- denormalizeAcpNotification prefers a NON-EMPTY `type`; an empty-string
`type` no longer wins over a valid `kind` and drops the event (M3w6i).
Tests: reply-pump handoff happy-path (delivers/resolves) + strand case
(rejects, not stranded); empty-`type`→`kind` fallback; the SSE buffer cap
firing; the unanchored-reply hold/release branches; and connGraceExpired
reset on reconnect (M3w6e, M3w6f, M3w6g).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): sweep session pendings in the subscribe wrapper + carry pump error
Two follow-ups on the reply-pump handoff:
- The session-scoped pending sweep moves from subscribeEventsInner's
read-loop finally to the subscribeEvents WRAPPER finally. The read-loop
finally only runs once the pump reaches the loop; a fast failure (fetch
reject / non-OK / wrong content-type, all before the loop) skipped it and
stranded the pending. The wrapper finally always runs, so it covers the
fast-fail path too (M4DWq).
- ensureSessionReplyPump captures the pump's error (HTTP 401/404, wrong
content-type) and rejects swept pendings WITH it instead of a generic
message, so a caller can tell auth failure from a network drop (M4DWx).
Test: a 401 on the session GET (inner throws before its read loop) still
rejects the in-flight session-scoped pending via the wrapper sweep.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): carry the subscription error into the wrapper sweep + guard tests
Follow-ups on the reply-pump handoff:
- The subscribeEvents wrapper sweep now rejects with the actual cause of the
subscription's exit (captured from a try/catch around the inner generator)
instead of a hard-coded generic message. On the fast-fail path (401 /
wrong content-type thrown before the inner read loop) this wrapper finally
is the only sweep that fires, so the caller now sees the real failure —
parity with the reply-pump's pumpError reason (M4W9a).
Tests:
- the fast-fail sweep reason carries the 401 (not a generic message);
- the M3pAM non-SSE rejection asserts the content-type cause reaches the
caller (proves pumpError propagation) (M4W9g);
- cross-session scope guard, both the consumer and the reply-pump
resolution paths: a reply on session A's stream must not resolve a pending
scoped to session B (M4W9e).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve): guard onSessionGraceExpired in the grace timer against an uncaught throw
The session grace-expiry setTimeout protected closeSessionStream with a
try/catch but called the owner-supplied onSessionGraceExpired callback
outside it. From a bare setTimeout, an uncaught throw there would crash the
whole daemon — the same hazard the teardown guard exists for. Wrap it in
its own try/catch (separate from teardown, so the conn-reap re-check still
runs even if teardown threw). Add a regression test (M4i9z).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): make conn-stream pump CRLF-aware; align replay opt-in doc
The connection-scoped SSE reply pump split frames with an LF-only
`buf.indexOf('\n\n')`. A server or proxy emitting `\r\n\r\n` frame
separators produces no `\n\n` substring, so the scan never found a
boundary: the unread buffer grew to the OOM cap and the pump threw,
leaving every connection-scoped JSON-RPC reply unresolved. Reuse the
shared CRLF-aware `consumeFrames` splitter (and strip a trailing CR
per data line) so the conn pump frames exactly like the session
readers. Add a regression test that delivers a conn-scoped reply over
`\r\n\r\n` and asserts it resolves.
Also update the design doc: the in-repo SDK `AcpHttpTransport` opts in
to replay in this PR (`supportsReplay = true` + resends Last-Event-ID),
so the backward-compat note no longer reads as "keeps false until it
opts in". Only the external agent-web transport flip stays deferred
(already listed under Out of scope).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon,sdk): log replay-deferral arm; document session-reply routing invariant
Add a stderr breadcrumb when a resume arms `replayPending`: while armed,
`sendSessionReply` defers every out-of-band reply until the pump delivers
`replay_complete`. If that sentinel never arrives (a dropped frame or a
pump error), the replies stay buffered indefinitely with no other trace —
the log gives operators a starting point. Silent on a fresh connect (no
deferral). Covered by a new test.
Also strengthen the `SESSION_STREAM_REPLY_METHODS` doc comment: name the
authoritative daemon call sites (dispatch.ts), spell out the hang failure
mode if the set drifts, and record a build-time grep / shared-constant
enforcement as a follow-up (a cross-package invariant the SDK can't type-check).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): use bracket notation for _meta index-signature access (TS4111)
`extractParams` returns `Record<string, unknown>`, so dot access to
`params._meta` violates `noPropertyAccessFromIndexSignature` (set in the
root tsconfig). The esbuild bundle path doesn't typecheck, so CI's build
stayed green, but strict `tsc --noEmit` reports 6 × TS4111 at these sites
(added with the query-param routing change). Switch all six to
`params['_meta']`. Purely syntactic — runtime behavior is unchanged.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): don't silently hang conn-scoped requests on a failed conn stream
`pumpConnStream` swallowed two failure paths: a non-2xx / no-body `GET
/acp` did a bare `return`, and read-loop errors were caught and dropped.
Either way the pump promise RESOLVED, so `openConnStream`'s `.catch`
never ran — connection-scoped JSON-RPC pendings stayed in the map
forever, and `connStreamAbort` was never cleared, so `ensureConnStream`
saw it non-null and never reopened the stream (every later 202 request
hung with no pump to deliver its reply).
- A non-2xx / missing-body response now throws (HTTP status in the
message) so the catch sweep rejects the conn-scoped pendings.
- The read-loop catch rethrows real errors and only swallows an
intentional abort (dispose / reconnect, which owns its own cleanup).
- `openConnStream` clears `connStreamAbort` in a `.finally` (guarded on
controller identity) so the stream reopens on the next request after
ANY settle — clean close, error, or abort.
Regression test: a 500 `GET /acp` rejects the conn-scoped pending (leaves
session-scoped ones for their own stream) and the next request reopens.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon,sdk): conn-stream error propagation, listener cleanup, buffer eviction, param parsing
Address review findings on the resumable /acp stream and exported SDK
transports — all additive/backward-compatible, REST untouched:
- openConnStream: reject connection-scoped pendings with the pump's REAL
error (HTTP 401/503, network drop) instead of a generic message, mirroring
ensureSessionReplyPump.
- pumpConnStream: keep the abort listener in a named ref and remove it in
finally so a long-lived signal reused across reconnects doesn't accumulate
listeners (mirrors the session readers).
- sendRequest: remove the abort listener on the happy path (the `{ once: true }`
listener self-removes only when the signal fires), preventing per-call
listener buildup on a shared caller signal.
- pushCapped: under a content flood, evict a REPLAYABLE id-bearing frame (the
ring redelivers it) before an irreplaceable id-less deferred reply — dropping
the latter would hang the session/prompt caller — and log the dropped id.
- acpRouteTable.boolParam: treat a present-but-empty value (`?detail=`) as
absent, matching numParam, so `{ detail: false }` isn't forwarded for an
unset param.
- connection-registry resume flush: hoist the `splice(0)` snapshot into a named
local to make the re-entrant copy-semantics invariant visible.
Tests: boolParam empty-value omission; pre-attach buffer keeps the id-less
reply under a 400-frame content flood. Document two exported-transport
limitations (permission voting; session RPC awaited inside the subscribeEvents
loop) as §1.7-adjacent follow-ups in the design doc.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): never evict an irreplaceable id-less reply from the pre-attach buffer
The previous pushCapped change preferred evicting id-bearing (ring-replayable)
frames, but left a degenerate hole: when the buffer fills with ONLY id-less
deferred replies (no id-bearing entry exists), findIndex returned -1, dropIndex
fell back to 0, and the oldest deferred JSON-RPC reply was evicted — silently
hanging its session/prompt caller, the exact failure the guard exists to
prevent (wenshao).
Fix: when there is no replayable id-bearing frame to evict, do NOT drop —
append and let the id-less replies exceed the soft cap. The cap is a memory
bound against a CONTENT flood (id-bearing frames); id-less replies are bounded
by the number of in-flight session RPCs the client actually issued
(client-controlled, tiny), so they can't run away in practice. Log once when
over the soft cap. The connection buffer (no id accessor) keeps its FIFO
eviction unchanged.
Test: 300 all-id-less replies buffered past the 256 cap are all delivered on
reconnect, none evicted.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): hard ceiling + transition-only logging for the id-less reply buffer
Follow-ups on the prior id-less-eviction fix (wenshao):
- Defense-in-depth HARD cap. The soft-cap path never drops id-less replies,
relying on "id-less replies are RPC-bounded" — true today but enforced only
by convention. Add HARD_BUFFERED_FRAMES_CAP (4× soft = 1024): past it, drop
the oldest id-less reply and log loudly, so a future non-RPC-bounded producer
or a buggy client can't grow the daemon heap without limit.
- Log at the soft-cap transition only (buf.length === MAX_BUFFERED_FRAMES), not
on every over-cap push — the comment said "once" but it logged linearly with
over-cap depth (~44 lines for 300 entries).
Tests: assert the soft-cap warning fires exactly once for a 300-entry overflow;
new test that 1100 id-less replies are bounded at the 1024 hard cap (oldest
dropped, newest kept, loud breach log emitted).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon,sdk): release all deferred replies when replay evicted frames; guard conn pump against session-scoped pending
When ring replay overflows and emits state_resync_required, the watermark
anchor guarantee is void (the anchored frame may have been evicted), so
hold-until-watermark could freeze deferred session replies indefinitely.
Track eviction through the pump loop and flush ALL buffered session frames
at replay_complete in that case instead of waiting on the watermark.
Also harden the SDK conn-stream pump: never resolve a session-scoped pending
entry from the connection stream (scope guard), and document the fresh-attach
(non-resumptive) caveat for ensureSessionReplyPump.
Tests: add FakeBridge.getSessionLastEventId so integration replySession no
longer throws (anchorId now reachable); cover the eviction cascade-release
path, the conn-stream session-scope guard, and shared reply-pump ref-counting.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(daemon): flush deferred replies on mid-replay iterator error; cover anchored watermark e2e
On an iterator error mid-replay the catch path re-throws, which drives
onPumpSettled; while the session stream is still open that takes the
closeSessionStream branch (full teardown, not a detach-with-grace), so any
still-deferred session replies in the binding buffer were dropped rather than
preserved. Flush them in the catch before signalling stream_error — same
safety flush as the happy-path completion (the iterator has terminated, so no
content frame can still race ahead of them). Correct the now-inaccurate
happy-path comment that claimed error-path frames stay buffered.
Add an end-to-end transport test for the anchored watermark path: with a real
getSessionLastEventId, a deferred reply is held through pre-watermark content
and released ON its anchor mid-replay, before replay_complete — distinguishing
the watermark release from the unanchored release-at-boundary path.
Document two deferrals in the design doc: response-replay idempotency for an
already-resolved permission (a conformant client dedupes on _meta.requestId;
full re-send belongs with the permission-coordination follow-up) and an
automated guard for the SESSION_STREAM_REPLY_METHODS drift.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* feat(daemon,sdk): log replay effectiveness; de-shadow pump sweep; document resume/permission edges
Add an operator breadcrumb at replay completion (resumed-from cursor, delivery
high-water mark, bus replayed count, eviction flag) so 'did resume recover the
gap?' is answerable from server logs.
Rename the reply-pump sweep loop variable so it no longer shadows the outer
pump-map entry (unrelated types).
Clarify why the resume path drops id-bearing buffered frames (the event pump is
aborted on detach, so only id-less out-of-band replies accumulate during the
gap; ring replay owns id-bearing recovery and eviction is signalled via
state_resync_required). Document two opt-in-transport edges as
permission-coordination follow-ups: the no-subscriber reply pump's GET stream
causing an agent permission_request to be routed to the pump and dropped, and
why an automated SESSION_STREAM_REPLY_METHODS drift guard needs dataflow (the
prompt reply is decoupled from its case block).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
---------
Co-authored-by: 秦奇 <gary.gq@alibaba-inc.com>
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Co-authored-by: qwen-code-dev-bot <qwen-code-dev-bot@users.noreply.github.com>
|
||
|
qqqys
|
68348e236a
|
feat(channels): qwen tag — RFC + Phase 0 (multiplayer channel-resident agent) (#5888)
* docs(channels): add "qwen tag" RFC — channel-resident multiplayer agent Design for a persistent, multiplayer agent that lives in a chat channel (DingTalk-first), built on the existing channel adapters (qwen channel start, packages/channels/*) and the qwen serve daemon rather than a new service. Covers the phased plan: Phase 0 multiplayer identity, Phase 1 proactive engine (scheduler + cold-group push, daemon migration), Phase 2 channel memory + governance, plus tradeoffs and resolved design decisions. Part of #5887 Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * feat(channels): qwen tag Phase 0 — multiplayer identity & group-safe commands In a thread-scoped group every member shares one session, which surfaces a few gaps in the channel layer. Phase 0 closes them as a backward-compatible increment on the existing AcpBridge channel path: - Inject a [sender] marker into the prompt for group turns so the agent can tell speakers apart; skipped for 1:1 chats and for already-prefixed re-entries. - Add Envelope.alreadyPrefixed so collect-mode coalescing does not double-prefix the already-tagged buffered text. - Require "/clear confirm" in groups — a bare /clear no longer wipes the shared session for the whole channel; DMs still clear directly. - Add a read-only /who reporting channel / workspace / session scope without creating a session. - Drop DingTalk group messages with no conversationId, so the shared session is never keyed on the expiring sessionWebhook. - Fix the stale ChannelConfig.dispatchMode JSDoc (runtime default is 'steer'). Adds unit tests for sender attribution, the collect double-prefix guard, the group /clear confirmation, and the read-only /who. Part of #5887 Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): scope group clear confirmation * fix(channels): address review on qwen tag Phase 0 - Restrict /clear in a shared (thread) group to config.allowedUsers (when set) and require an explicit "confirm"; only the shared case is gated — DMs and per-user groups clear directly. Reconcile the RFC (OD-4) with this approach (a hyphenated /clear-channel isn't parseable; per-member owner-gate waits on the identity model, OD-3/OD-11). - Sanitize the injected [sender] marker (strip brackets/CR/LF, cap length) so a crafted nick can't break out of or spoof the attribution tag. - Surface to stderr when a collect-mode coalesced re-entry drops buffered turns instead of swallowing the error. - Narrow Envelope.alreadyPrefixed to the literal `true` (internal-only flag). - Expose DingtalkChannel.isUnroutableGroupMessage and test the group-without-conversationId drop; replace flaky setTimeout sequencing in the collect test with vi.waitFor; add tests for /who (active + private scope) and /clear authorization in a shared group. Part of #5887 Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): preserve group command routing * fix(channels): harden group routing (map leak, injection, slash attribution) Address the surviving Phase-0 review criticals: - /clear now purges every per-session map (sessionQueues, activePrompts, collectBuffers), not just instructedSessions, so a long-running gateway doesn't leak dead session entries. - QQ group adapter sanitizes the self-prefixed sender name. It sets alreadyPrefixed, which bypasses ChannelBase's [..]/newline/length guard, so a crafted QQ nickname could otherwise inject brackets/newlines. - Unrecognized slash commands in a shared group keep their [sender] attribution; recognized commands (a local handler or a forwarded agent command) still reach the agent verbatim. - Sanitize quoted referencedText (strip control chars, cap at 500) so it can't inject newlines/instructions or balloon the prompt. - /who reports only the workspace basename, not the absolute cwd. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): close Phase-0 review criticals (quote injection, /clear cancel, drain log) - referencedText: strip the wrapper's own delimiters (" [ ]) in addition to control chars, so a quoted message can't break out of [Replying to: "..."] and inject its own top-level instructions. - /clear: cancel any in-flight bridge.prompt() for the cleared session(s) and drop their buffered follow-ups before purging the maps, so a running turn can't deliver a stale response into — or resurrect — the cleared session. - collect-drain failure log now includes the sessionId and last sender so concurrent sessions are diagnosable. - Extract the duplicated sender-name sanitizer into a shared sanitizeSenderName helper used by both ChannelBase and QQChannel; reuse isSharedGroupSession for the /who scope note. - Tests: quote-breakout payload, /clear cancels in-flight, /who in a DM, single-scope hasSession/removeSession, and a sanitizeSenderName unit test. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): make group slash-command pass-through race-free and invalidate queued turns on /clear CRITICAL 1 (sender attribution race): availableCommands is populated asynchronously by the ACP available_commands_update notification, so a registration check could see an empty list on a fresh session and wrongly [sender]-prefix the first real /command — corrupting it into plain text. Stop inline-prefixing ANY slash-shaped message: pass it through verbatim so it always parses as a command regardless of load/registration state. Also widen the command token to include - and : so /compress-fast and /git:commit parse as commands. Tradeoff: an UNREGISTERED group slash command no longer carries [sender] attribution (inverse of the earlier R2 ask) — not breaking real commands is the safer default; flagged for maintainers. CRITICAL 2 (/clear vs queued followups): a teammate turn that entered handleInbound() before /clear confirm had already captured the prev.then() chain and would still run bridge.prompt() against the just-cleared session. Add a per-session generation counter bumped up-front by /clear; a queued turn snapshots it at enqueue and bails if it was bumped before the turn dequeued. Tests: fresh-session/empty-availableCommands pass-through, hyphenated command, unrecognized command (flagged behavior), non-slash still prefixed; /clear genuinely awaits a PENDING in-flight turn before confirming; /clear confirm invalidates an already-queued followup (no resurrection); collect-drain failure logs the lost count with sessionId + sender to stderr. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): only skip group [sender] tag for real slash commands The group attribution check keyed off the lenient parseCommand(), which accepts slash-prefixed paths (e.g. /tmp/foo) as commands, so that prose reached the model without the [sender] tag. Decide command-vs-prose with a dedicated isSlashCommand() that mirrors the CLI's classifier (cli ui/utils/commandUtils.ts isSlashCommand): reject //, /*, a bare /, and any path separator in the first token. Slash-prefixed paths and comments now keep their speaker attribution, while real commands (/compress, /git:commit) still pass through verbatim. Purely lexical, so it stays race-free against the async command list. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): bound /clear's wait on a wedged in-flight turn doClear awaited active.done after cancelSession, but active.done only resolves in the prompt() finally. If the ACP child is wedged (stuck tool call, not reading stdin, or crashed without closing), cancelSession may throw (swallowed) or succeed yet the prompt never returns, so active.done never resolves and /clear — and the whole channel — hangs forever. Race the wait against a CLEAR_CANCEL_TIMEOUT_MS (3s) timeout and purge anyway when it wins: the per-session maps are still purged and the generation already bumped, so a turn that settles later is invalidated. Cancellation stays best-effort. Tests: a wedged session whose active.done never resolves still completes /clear (purges every map, replies "Session cleared") within the timeout via fake timers — no real wait(ms); plus /clear Confirm / CONFIRM (mixed-case) is accepted, guarding the handler's .toLowerCase(). Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): test QQ group slash branch, log dropped queued turns, reclaim cleared generations Address review feedback on the Phase-0 qwen-tag /clear + group-slash work: - QQChannel: cover the previously-untested `isSlash` branch of handleGroup — a group `/clear` is forwarded verbatim (no `[sender]` tag) and must NOT set `alreadyPrefixed`. Catches a regression that always sets it. - ChannelBase: when a followup turn is dequeued after `/clear` bumped the session generation, log the drop (with sessionId) instead of returning silently, so an unanswered queued message is diagnosable. - ChannelBase: reclaim the bumped `sessionGenerations` entry once the cleared session's queue drains (or immediately when nothing was queued), so a long-running gateway no longer leaks one entry per `/clear`. Reclamation is deferred and guarded so it can't delete an entry a still-queued turn needs, and a wedged in-flight turn can't block `/clear`. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): harden qwen-tag group session gating, cancel bounds, and prompt sanitization Address the latest PR #5888 review batch (4 Criticals + suggestions): - Treat sessionScope 'single' as a SHARED group session: the shared-scope predicate now covers both 'thread' and 'single' (via SHARED_SCOPES), so a 'single'-scoped group can no longer bypass /clear confirm + allowlist gating and wipe the channel-wide __single__ session with a bare /clear. - Bound the steer-mode wind-down wait: race active.done against CLEAR_CANCEL_TIMEOUT_MS (and fire-and-forget the cancel) so a wedged ACP child can't pin the session queue forever, matching the /clear hardening. - Make /clear's cancelSession request fire-and-forget so a wedged cancel request can't hang /clear before the bounded active.done wait even starts. - Strip Unicode line/paragraph separators (U+2028/U+2029) and bidi overrides (U+202A-U+202E, U+2066-U+2069) from sender names and quoted text via shared sanitize helpers (sanitizeSenderName, new sanitizeQuotedText); route referencedText and attachment filenames through the shared quoted sanitizer. - Add sender/conversation context to the generation-bail drop log and the DingTalk unroutable-group-message drop log. - Tests: fix TS4111 bracket-access in ChannelBase tests (test-inclusive typecheck clean); add load-bearing coverage for single-scope sharing, wedged steer, wedged /clear cancel, Unicode/bidi sanitization, and the generation-reclamation guard fire paths. - Correct the isSlashCommand JSDoc to document the intentional bare-'/' divergence from the CLI classifier; note SessionRouter's by-sender scan does not match single-scoped keys (latent today). Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): stop block-streaming on /clear, not just cancel it doClear flipped active.cancelled = true but never called active.stopStreaming(), unlike the /cancel handler. In block-streaming channels, text already buffered in the BlockStreamer could still be emitted by the idle timer during/after /clear, leaking a stale response into the just-cleared session. Mirror /cancel: also tear down the streamer on the cancelled prompt in the clear path. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): gate /clear for single-scope DMs and unstick wedged steer queue Two follow-up fixes in ChannelBase, both on my recent fixes. CRITICAL 1: `single` scope maps EVERY sender — group OR DM — to the one `__single__` session, but the shared-session guard also required `isGroup`, so anyone who could DM the bot could bare-/clear the channel-wide session without the confirm + allowedUsers gate. The shared-session predicate now treats `single` as shared regardless of isGroup (`thread` stays group-only); /help and /who follow the same predicate. CRITICAL 2: the bounded steer wait stopped `await active.done` from hanging, but the replacement turn was still chained behind the wedged turn's never-resolving sessionQueues tail, so the follow-up (and every later message) hung forever. On the steer timeout we now re-seed the chain (prev = resolved) so the next turn starts a fresh chain; the wedged turn stays cancelled, so a late settle still can't deliver a stale response. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): sanitize attachment filePath and tighten isSlashCommand The attachment filePath was embedded raw in the prompt while the filename beside it was sanitized. Adapters build the path by basename()-ing the user-supplied filename, so its last segment carries the same attacker-controlled chars (brackets, newlines, U+2028, bidi overrides), letting a crafted filename inject prompt lines via the path. Route the rendered path through the same sanitizeQuotedText neutralization; att.filePath itself is left intact, and benign names render unchanged so the agent's read-file tool still resolves them. Also tighten isSlashCommand to require parseCommand()'s token charset ([a-zA-Z0-9_:-]+ plus an optional @botname). A non-command-shaped input like /cafe or a zero-width-laden token previously skipped the group [sender] tag yet was not a runnable command, reaching the shared session as unattributed prose. It now keeps its attribution. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): guard activePrompt cleanup against replacement-turn clobber The per-turn `finally` in ChannelBase deleted `activePrompts[sessionId]` unconditionally. With the steer-mode bounded wait, this corrupts steer protection: when turn A wedges (its `bridge.prompt()` never resolves), the `CLEAR_CANCEL_TIMEOUT_MS` race times out (`steerWedged`), and turn B starts a fresh chain and re-seeds `activePrompts` with its own entry. When A's prompt finally settles and reaches its `finally`, the unconditional delete removes B's entry — so a later turn C sees `activePrompts.get(sessionId) === undefined` and silently loses steer protection (it forwards verbatim instead of cancelling + re-prompting). Fix: capture this turn's own `ActivePrompt` and compare-and-delete — only clear the entry if it is still ours (`activePrompts.get(sessionId) === promptState`). Sibling-map audit (same later-settling-wedged-turn clobber): - collectBuffers: also gated on the same `stillCurrent` flag. A replaced wedged turn must not drain the buffer the live replacement turn now owns (reachable via mixed-mode single-scope, where a steer turn and collect follow-ups share one session). Behavior-preserving: in all serialized flows the turn is still current at its finally, so the drain runs exactly as before. - sessionQueues / sessionGenerations / instructedSessions: not mutated in the per-turn finally (only in /clear), so no per-turn clobber. Queued turns are already guarded against /clear by the sessionGenerations counter. - promptState.resolve(), bridge.off, streamer.stop, onPromptEnd: per-turn-owned cleanup, kept unconditional (a replaced wedged turn must still release any steer/clear waiter racing its done promise). Adds a deterministic test (manual deferred, no timers/sleeps) reproducing A wedged -> B replaces -> A settles late, asserting B's entry survives and a following turn C still engages steer protection. Reverting the guard fails it. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): guard onPromptEnd on superseded turns and strip control chars from sender names When a steer-replacement turn B re-seeds a session after the bounded wind-down wait times out, a wedged predecessor turn A can settle late and run its finally. onPromptEnd hides a session/chat-scoped working indicator (e.g. the Telegram/WeChat typing indicator keyed by chatId), so A's previously-unconditional onPromptEnd would stop B's indicator while B is still working. Guard onPromptEnd with the same stillCurrent identity check that already protects the per-session map cleanup, so a superseded turn can no longer clear the successor's indicator. A's own teardown (textChunk listener, streamer, promptState.resolve) stays unconditional. Bring sanitizeSenderName to parity with its sibling sanitizeQuotedText by stripping C0/DEL control chars: a crafted display name with \x07/\x1b otherwise reaches the [name] prompt tag. Mirror the same strip in the qqbot send.test.ts mock so a control-char regression in the real helper is caught. Add a diagnostic stderr line when a steer abandons a wedged turn, matching the existing queue-drop log. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): preserve path chars in attachment filePath; trim sender names Addresses ci-bot review on the channel prompt-sanitization paths. filePath over-sanitization (the main regression): a prior hardening pass routed the rendered attachment path through sanitizeQuotedText, which strips the `[`, `]`, and `"` characters. Those are valid, common filesystem path characters (e.g. a Next.js `app/[slug]/page.tsx` route, a quoted segment, or a space in a folder name), so stripping them advertised a `saved to:` path that does not exist on disk and broke the agent's read-file tool. A path rendered alone on its own line cannot use brackets/quotes/spaces to break out of that line, so the only real injection risk is characters that break or reorder the line. Add a path-safe neutralizer, sanitizePromptPath, that strips ONLY C0/DEL controls (incl. CR/LF), the Unicode line/paragraph separators, and bidi overrides (the PROMPT_UNSAFE_INVISIBLES set), preserving everything else byte-intact and without capping length. The human-readable fileName label keeps sanitizeQuotedText (bracket-stripping is fine for a quoted label). steer generation-capture ordering: in steer mode the session generation was snapshotted AFTER the bounded wind-down wait. A concurrent /clear (e.g. another sender's clear-confirm on a shared session) that bumps the generation DURING that wait was therefore invisible: the snapshot read the post-clear value, the dequeue equality guard still held, and the turn ran against the just-cleared session. Capture preSteerGeneration BEFORE the wait and use it for the guard so a /clear during the wait is detected and the turn bails. sender-name fallback: sanitizeSenderName now trims after the length cap and returns an 'unknown' default, so a name made entirely of strippable chars (brackets/newlines) no longer renders an anonymous bracket tag. Both call sites embed the result as a bracket tag with no fallback of their own, so the default lives in the shared helper; the qqbot test mock mirrors the new contract. Adds load-bearing, mutation-checked tests for each change (deterministic, fake-timer driven for the steer/clear race; no real-time waits). Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): stop streamer on steer-cancel and align slash-command trim Two ci-bot review findings on the steer/slash-command paths in ChannelBase. 1. steer-cancel must stop the BlockStreamer, not just cancel. The steer handler flipped active.cancelled but never called stopStreaming(), unlike the sibling doClear path. cancelled alone only suppresses NEW chunks: text already buffered in a wedged turn's streamer is still flushed by the idle timer (~1500ms), which fires before the 3000ms steer wind-down bound, so it leaks into the chat after the replacement turn has begun. Mirror doClear and stop the streamer immediately after cancelling. 2. parseCommand now trims its input so it agrees with isSlashCommand (which already trims). Before, " /help" (leading space, common from IME/copy-paste) made isSlashCommand return true — suppressing the group [sender] tag — while parseCommand returned null, so the command reached the agent unattributed. Trimming closes that attribution gap; it changes nothing else (the no-whitespace path is unaffected, and /tmp/foo still has no handler). Also document the forward-looking late-cancel concern at the steer cancel site: cancelSession is keyed only by sessionId and the replacement prompt reuses it. The shipping AcpBridge sends cancel+prompt over one in-order stdin stream with cancel enqueued first, so the child always processes the cancel before the new prompt — the replacement turn is safe today. A future network/daemon bridge with cancel latency could reorder them; the proper fix is turn-scoped cancellation (a Bridge-contract change), deferred to avoid an API break here. Tests: a steered wedged turn with buffered streamer text invokes stopStreaming so the idle-timer flush can't deliver stale text after the replacement turn begins; " /help" in a group is handled as a command (no [sender] tag, no forward) while /help and /git:commit still parse. Both are mutation-checked. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): block shell commands in shared sessions; dedupe cancel path Phase 0 ships shared/group sessions (sessionScope 'thread' and 'single') but has no per-sender trust model — the [sender] marker is explicitly not a trust boundary. The pre-existing bang (`!`) handler in ChannelBase runs arbitrary host shell commands, so in a shared session ANY participant could run `!rm -rf /`. Gate it with the same isSharedSession predicate used for the destructive-/clear confirm gate: refuse with a one-line notice in shared sessions, keep direct execution in a 1:1 session (the lone user is the operator). isSharedSession is promoted from a closure to a private method so both gates share one source of truth. Extract the duplicated cancel + bounded-wait sequence from doClear and the steer branch into a private cancelAndAwaitActive() helper (the duplication previously caused steer to miss stopStreaming()). steer keeps its pre-wait generation snapshot and uses the boolean result; doClear ignores it. Also: trim the over-verbose steer/clear comment blocks to the load-bearing why; cap sanitizePromptPath at 1024 chars (defense-in-depth) like its siblings; and make the qqbot send.test mock use the real sanitizeSenderName via vi.importActual instead of an inline re-implementation that can drift. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): run onPromptEnd after /clear; only skip it for superseded turns The per-turn `finally` guarded `onPromptEnd` behind `stillCurrent` alone. When `/clear` cancels an in-flight turn with NO replacement it deletes that turn's `activePrompts` entry, so if the cancelled prompt settles late `stillCurrent` is false and `onPromptEnd` was skipped — leaking the platform cleanup several adapters do there (Telegram clears its typing interval, DingTalk recalls the working reaction, Weixin clears typing, Feishu finalizes card state). The `stillCurrent`-only guard was only meant to stop a STEER-superseded turn from clobbering its replacement's indicator, but it wrongly also caught the no-replacement `/clear` case. Distinguish the two: add `ActivePrompt.superseded`, set on the OLD turn only when a steer replacement actually takes over the session slot. The `finally` now runs `onPromptEnd` UNLESS the turn was superseded; the `activePrompts.delete` and the collect-drain stay `stillCurrent`-gated (a superseded turn must not drain the replacement's buffer). Regression + collect-drain tests added with mutation checks. Also addresses review nits: extract a shared command-token regex constant so parseCommand and isSlashCommand can't drift; report `single` scope as "shared channel-wide" in /who (it is shared across all DMs and groups, not just one group); append a truncation ellipsis in sanitizeQuotedText so a cut quote/filename is detectable; and include the message text in the generation-bail drop log so an ignored message is diagnosable. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): require the slash-command token to immediately follow the slash isSlashCommand used trimmed.slice(1).trimStart() before taking the first token, so a space after the slash (`/ foo`) still classified as a command. parseCommand's regex requires the token immediately after `/`, so it returned null for the same input. In a shared group session that divergence suppressed the [sender] attribution (isSlashCommand true) while running no command (parseCommand null), letting `/ foo` reach the agent unattributed — the exact failure the file's comment warns about. Remove the .trimStart() so the token must immediately follow the slash; `/ foo` now splits to an empty first token and is treated as prose, agreeing with parseCommand. Normal commands (/help, /git:commit, /compress-fast, /cmd@bot) are unaffected. Adds an invariant test that isSlashCommand and parseCommand agree on `/ foo` (both false) and /help (both true), plus a behavioral test that `/ foo` keeps its [sender] tag. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): don't clobber a post-/clear turn's indicator; tidy refusal + diagnostics Addresses wenshao's review on PR #5888: - Refusal text: `!` shell commands are "disabled in shared sessions" (drop "group" — isSharedSession is also true for single-scope DMs, which are not groups). - /clear diagnostic: capture cancelAndAwaitActive's result and log on timeout, mirroring the steer "abandoned a wedged turn" message, so a wedged /clear is observable instead of silently "worked". - /clear indicator clobber: a /clear that evicts a wedged turn never set `superseded`, so the turn's late-settling finally still ran onPromptEnd and killed the indicator a turn started AFTER the /clear now owns. Add a `clearEvicted` flag: /clear runs the wedged turn's own onPromptEnd at eviction time (no replacement exists yet) and marks it clearEvicted; the late finally then skips onPromptEnd. onPromptEnd now fires exactly once for the evicted turn, at clear-time. The finally guard becomes `stillCurrent || (!superseded && !clearEvicted)`. ActivePrompt carries the originating chatId/messageId so the eviction can target the right indicator. - Dropped-queued-turn log: sanitize the attacker-controlled message text (render newline visibly, strip C0/DEL incl. CR/ESC) before it reaches an operator's terminal, matching every other embed path. Tests: complete the single-scope DM `!` refusal test (not-called + length + not-forwarded); update the group/DM/1:1 refusal assertions to the new text; re-point the "/clear settles late" test to clear-time cleanup; add a replacement-after-/clear test asserting a late settle does not end a later turn's indicator; strengthen the dropped-turn-log test with control-char input. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): fix root tsc errors in channel tests; gate /who; sanitize dingtalk nick log Root `tsc --noEmit` failed on 9 type errors in two channel test files that per-package builds and `--workspace` typechecks miss (both exclude test files). Fix the test types only — no production type changes: - qqbot/dingtalk: `QQChannel`/`DingtalkChannel` come from `await import()` as values, so annotate instances with `InstanceType<typeof X>`. - dingtalk test config was missing the required `token` field of `ChannelConfig`. - qqbot mock: bracket-access the index-signature `router` and type the assignment so `{}` is assignable to `Record<string, unknown>`. /who: gate it to authorized senders in shared sessions, mirroring /clear — /who leaks the workspace basename, so non-members shouldn't see it. dingtalk: sanitize `senderNick` with the shared `sanitizeSenderName` before writing it to stderr, so a crafted nick with CR/LF/control chars can't fragment or inject log lines. !-shell gate: refuse `!` shell commands in ALL groups (gate on `envelope.isGroup`, not just shared sessions) — a user-scope group is not a shared session yet is still multi-operator, so members could reach the host shell. The refusal now runs BEFORE router.resolve, so a refused command never creates a session. Single-scope DMs stay refused. tests: the qqbot/dingtalk suites mock `@qwen-code/channel-base` and pull the real `sanitizeSenderName` via the package export, which resolves to base/dist and broke clean package-local runs. Alias the specifier to base's source in each vitest config (mirrors cli) so the suites run without a prior `tsc --build` of base. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): bump generation and release indicator on steer-abandon, mirroring /clear When a steer abandons a wedged turn it re-seeds a fresh queue chain but, unlike doClear, did neither of the two protections doClear runs on a wedged-turn eviction. Both bugs surface when mixed dispatch modes collapse onto one session (single/user scope). FIX 1 (generation bump): a followup queued behind the wedged turn stays on the now-orphaned chain. Without bumping sessionGenerations, when the wedged turn late-settles the followup PASSES the dequeue guard and runs the unguarded activePrompts.set, clobbering the live steer replacement -> two concurrent bridge.prompts on one session (duplicated responses + double tool execution). Bump the generation up-front (skipping if a /clear raced the wind-down wait) and advance the replacement's captured snapshot so it proceeds while the orphaned followup bails. FIX 2 (messageId-scoped indicator leak): the supersede path never released the abandoned turn's OWN indicator. A CHAT-scoped indicator is re-seeded by the replacement, but a MESSAGEID-scoped one (per-message reaction/card keyed on the inbound messageId) is keyed on the abandoned turn's messageId and leaks until disconnect. Run the abandoned turn's onPromptEnd at steer-time and mark it superseded so its late finally still skips onPromptEnd (released once, no double-fire, replacement untouched). Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): steer waits for the running turn instead of racing a concurrent replacement In steer dispatch, when a new message arrived while a prompt was running, ChannelBase cancelled the running turn (best-effort), did a BOUNDED wait, and on TIMEOUT (the old turn wedged, not finished) PROCEEDED to start a replacement bridge.prompt() on the SAME sessionId while the old prompt was still active. As wenshao flagged, this is bridge-unsafe: - DaemonChannelBridge.prompt() REJECTS while the prior prompt is still marked active, so the replacement is silently dropped. - Both bridges collect/emit chunks keyed by sessionId ONLY, so the abandoned turn's late chunks mix into the replacement's stream (duplicated/stale output). Fix (wenshao option (a)): steer now best-effort cancels the running turn and CHAINS the new turn onto the existing session queue tail, so it runs only AFTER the old turn's finally has actually run (onChunk detached, activePrompts cleared, indicator released). The cancel stays — it makes the old turn wind down sooner — but steer never proceeds while a turn is still active, eliminating the unsafe concurrency entirely. Steer now differs from followup only by that best-effort cancel. Removed the steer-only concurrency machinery that only existed to host a concurrent replacement: the steer-path bounded wait (cancelAndAwaitActive call), the steerWedged flag and fresh-chain re-seed, the preSteerGeneration capture, the steer-time sessionGenerations bump, the steer-time onPromptEnd, and the superseded flag (plus its finally guard term). /clear's OWN protections — its up-front generation bump, eviction-time onPromptEnd, and the clearEvicted finally guard — are SEPARATE and preserved; /clear genuinely evicts and still needs them. cancelAndAwaitActive is retained, now used only by /clear. BEHAVIORAL CHANGE: steer no longer force-interrupts a genuinely wedged turn; it cancels it and waits for it to finish before the new turn runs. Turn-scoped cancellation/routing (so a new turn can run without waiting for a wedged predecessor) is wenshao option (b) — it needs an API change across every adapter and is the deferred enhancement, out of scope here. Folded in two further #5888 review items: - [Critical] QQ slash-command audit-log injection (qqbot/src/QQChannel.ts): the audit process.stderr.write interpolated the RAW, attacker-controlled senderName (event.author.username) and cleanText BEFORE sanitization, so a crafted nick or message with CR/LF/ANSI escapes could forge or corrupt operator audit logs. Hoisted `safeName = sanitizeSenderName(senderName)` above the audit log and now log a neutralized command string (cap 80, render \n visibly, strip C0/DEL), mirroring ChannelBase's dropped-turn log and the DingTalk hardening already in this PR. The prompt-path sanitizeSenderName usage is unchanged. - [Suggestion] single-scope DMs were not attributed (base/src/ChannelBase.ts): the [sender] prefix was gated on envelope.isGroup alone, but sessionScope:'single' collapses every sender's DM into one __single__ session (already treated as shared by the !-gate, /clear confirm and /who), so different people merged into one unattributed conversation (the RFC-R4 gap Phase 0 closes). The gate is now (envelope.isGroup || sessionScope === 'single') — deliberately NOT isSharedSession, which is false for user-scope groups that must keep attribution. Tests: removed the obsolete steer-concurrency cases and added cases for the new steer behavior (new turn starts only after the old completes; the abandoned turn's late chunks cannot reach the new turn). Added an audit-log sanitization test (QQ) and single-scope-DM / user-scope-group / 1:1-DM attribution tests (base). /clear's eviction protections are unchanged and still pass. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): neutralize NEL/C1 in prompt text; dedupe shared-session auth gate Address wenshao's review on the Phase-0 qwen-tag channel base. - sanitize: PROMPT_UNSAFE_INVISIBLES now also neutralizes the C1 control block (U+0080-U+009F), which includes NEL (U+0085), a Unicode line break (UAX#14 BK) that renders as a new line. Without it a display name like "Alice<NEL>system: ..." or a crafted reply quote could inject a prompt line. Applies to both sanitizeSenderName and sanitizeQuotedText (and the shared path sanitizer) since they share the set. Tests cover NEL + a second C1 (CSI U+009B) for sender names and quoted text; mutation-checked. - ChannelBase: the two best-effort cancelSession() calls (the /clear wind-down wait and the steer pre-cancel) no longer swallow the IPC failure with an empty catch. They now log channel name + sessionId + reason to stderr, matching the existing /cancel log style, so a wedged turn is diagnosable. Still fire-and-forget (not awaited). - ChannelBase: extract the verbatim shared-session authorization gate shared by /clear and /who into a private isAuthorizedForSharedSession() predicate (isSharedSession + allowedUsers check). Behavior is unchanged; each caller keeps its own rejection wording. - tests: add a GROUP-path case that /help@mybot is treated as a command (no [sender] prefix, not forwarded to the agent); the existing @botname test only covered a DM. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): strip C1/NEL from audit-log text too, matching the prompt path The prompt path now neutralizes the C1 control block (U+0080-U+009F, incl. NEL U+0085, a Unicode line break) via PROMPT_UNSAFE_INVISIBLES, but the inline stderr audit-log sanitizers still stripped only C0/DEL. A crafted message carrying NEL/C1 therefore survived into the dropped-/queued-turn log (ChannelBase) and the QQ slash-command audit log, where many terminals render NEL as a newline, forging an extra [channel] log line. Extend both inline strips to also cover the C1 block (C0 + DEL + C1), matching the prompt path's union, and refresh the now-stale comments. Tests feed a NEL (U+0085) and a C1 char (U+009B) through both log paths and assert neither survives; reverting the strip to C0/DEL fails the new NEL assertions. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): gate /status and /cancel in shared sessions; code-point-safe truncation Three follow-ups to the shared-session authorization work: - /status: add the same isAuthorizedForSharedSession gate /who uses, so a non-member of a shared session with a non-empty allowedUsers list can no longer read its session/access state. Non-shared (DM/per-user) is unchanged. - /cancel: gate the destructive abort behind isAuthorizedForSharedSession like /clear (auth gate only, no confirm), so a group member can't cancel another user's in-flight turn on a shared session. 1:1 and authorized users unchanged. - sanitize: truncate sanitizeSenderName/sanitizeQuotedText/sanitizePromptPath on Unicode code-point boundaries (Array.from) instead of UTF-16 code units, so a cap landing mid-surrogate-pair (e.g. an emoji) can't leave a lone surrogate that renders as the replacement character downstream. The ellipsis logic in sanitizeQuotedText still keeps the result within maxLen code points. Adds tests for each (incl. mutation-checked gates and an emoji-at-cap case). Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): audit-log shared /clear and blocked ! shell; test steer stopStreaming Address ci-bot review on PR #5888: - [Observability] Emit a stderr audit line on a successful clear of a SHARED session (channel, sessionId, sanitized sender + stable senderId). A 1:1 DM clear is single-participant and is not logged. - [Observability] Emit a stderr audit line when a group/shared member's `!` host-shell command is refused, so operators can detect blocked attempts. Sender name is sanitized; the command payload is not echoed. - [Test gap] Add a steer test asserting the best-effort cancel calls active.stopStreaming (spying on the running turn's prompt), plus tests for both new audit lines and the DM no-log case. The /cancel and /status shared-session auth gates flagged by ci-bot were already present on this branch and are left unchanged. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): gate steer-cancel by authorization; close audit-log + attribution injection gaps Address review on PR #5888 (qwen-tag-phase0): - [Critical] The /cancel auth gate was bypassable via the default `steer` dispatch mode: any normal message cancels the running turn, so an unauthorized member of a shared session - blocked from /cancel - could abort another user's active turn just by sending a message. The steer branch now checks isAuthorizedForSharedSession FIRST and, when unauthorized, breaks to normal queuing (the message chains onto the session queue tail and runs AFTER the active turn instead of aborting it). - [Suggestion] Audit-log sanitizers missed U+2028/U+2029 and the bidi overrides (log-line spoofing / trojan-source). Exported PROMPT_UNSAFE_INVISIBLES and added a shared sanitizeLogText(text, maxLen) helper that applies BOTH that set AND the C0/DEL strip (and caps length), used at both audit-log sites (ChannelBase dropped-turn log + QQ slash-command audit log) so the defense can't drift apart. - [Suggestion] The [sender] attribution prefix was suppressed for any command-SHAPED text, so unrecognized "/x\n[SYSTEM]: ..." reached the agent unattributed. Added a synchronous isRecognizedCommand() (locally registered commands + the bridge.availableCommands snapshot) and now suppress the prefix only when the text is BOTH a command shape AND a recognized command; unrecognized command-like text keeps its [sender] tag. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): recognize command aliases per-session so attribution doesn't break them isRecognizedCommand decides whether to suppress the group [sender] attribution prefix, but matched only against availableCommands[].name. The ACP parser also accepts command ALIASES (parseSlashCommand altNames), so a valid alias like /summarize (alias of /compress) or /login (alias of /auth) was classified as unrecognized, rewritten to "[Alice] /summarize", and then run as PLAIN CHAT (the leading / is gone) instead of executing. It also read the bridge's GLOBAL availableCommands snapshot, which in DaemonChannelBridge can belong to another session. - Carry aliases through the available-command contract: add an optional altNames?: string[] to AvailableCommand, emit them on the wire in _meta.altNames (ACP's extension point; a top-level altNames would be an excess-property error against the SDK type), and lift them in both bridges via readAvailableCommandAltNames. Omitted when absent, so alias-free entries stay byte-identical. - Match name AND altNames, against THIS session's command list: isRecognizedCommand now takes sessionId and reads getAvailableCommands(sessionId) when the bridge exposes it (DaemonChannelBridge), falling back to the global getter (AcpBridge, single-agent, inherently session-correct). Stays synchronous: a real command sent before the snapshot loads keeps its tag (safe default). - Security intent intact: genuinely-unrecognized command-shaped text (e.g. /x\n[SYSTEM]: ...) still keeps its [sender] tag. Also fold in two #5888 review items in ChannelBase.ts: - Steer auth gate: audit the silent steer->queue downgrade to stderr for an unauthorized member (operator-visible only; no per-message reply), matching the /cancel,/clear,/who,/status gates' observability. - /clear eviction: a clear-time onPromptEnd that throws would abort the purge, leaving the evicted turn in activePrompts so its late finally (stillCurrent || !clearEvicted) re-runs onPromptEnd and clobbers a newer turn. Set clearEvicted first and catch+audit the throw so the purge always runs (turn becomes non-current) and the late finally skips. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): match agent commands case-sensitively; add steer wedge watchdog isRecognizedCommand lowercased the token before comparing it to agent command names/aliases, but the CLI's parseSlashCommand matches agent commands CASE-SENSITIVELY (`cmd.name === part`, `cmd.altNames?.includes(part)`). So a wrong-case token like `/SUMMARIZE` was "recognized" here — suppressing the `[sender]` attribution tag — yet ran NO command in ACP, which then forwarded the raw text unattributed, reopening the injection where a crafted `/SUMMARIZE\n[SYSTEM]: …` second line reaches a shared group as an apparent system directive. parseCommand now also returns the typed-case `raw` token; isRecognizedCommand matches AGENT commands on `raw` (case-sensitive, mirroring the CLI) while LOCAL commands keep their existing case-INSENSITIVE match (registerCommand lowercases the stored name; handleInbound dispatches by the lowercased token). The steer chain-and-wait path lost the old "abandoned wedged turn" log, so a hung predecessor bridge.prompt() could silently deadlock the session with no observability. Add a diagnostic-only watchdog: the steer branch arms a timer (unref'd) that, if the predecessor is still the active prompt after CLEAR_CANCEL_TIMEOUT_MS, emits a stderr line pointing at /clear for recovery; the chained `.then()` disarms it as its first statement once the predecessor's tail resolves. No concurrency change — chain-and-wait is untouched. Also simplify the late-finally onPromptEnd guard from `stillCurrent || !clearEvicted` to `!clearEvicted`: clearEvicted is set ONLY by /clear's eviction, which then unconditionally deletes activePrompts (its try/catch around the clear-time onPromptEnd guarantees the purge even if it throws) and never re-inserts the same promptState, so `clearEvicted` implies `!stillCurrent` and the dropped term was unreachable. Tests: wrong-case `/SUMMARIZE` keeps `[sender]` (with a mutation note); wrong- case local `/HELP` still dispatches locally; fake-timer watchdog tests for the wedged (logs) and settled (timer cleared, no log) paths. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): align agent command recognition with parseSlashCommand; guard finally onPromptEnd Address maintainer review on the qwen-tag Phase 0 channel adapter. - [Critical] Guard the normal-completion onPromptEnd in the per-turn finally. onPromptEnd runs platform-adapter cleanup (typing interval, working-reaction recall, card finalize) - network/IO that can throw. An uncaught throw skipped activePrompts.delete (session leak), promptState.resolve (active.done never settled, so a later /clear falsely logged "abandoned a wedged turn"), and the collect-buffer drain - and the rejection, swallowed by the queue tail's .catch(() => {}), silently dropped every later turn. Wrap it in try/catch with a stderr log, matching the eviction-path treatment. - [Critical] End the channel/agent command-recognition divergence by matching the AGENT branch of isRecognizedCommand EXACTLY as the CLI's parseSlashCommand does: the FIRST whitespace token after the leading '/', case-SENSITIVELY, WITHOUT stripping an @suffix. parseCommand's '@'-stripped, lowercased token diverged from the agent (PARSE_COMMAND_RE drops '(?:@\S+)?'), so /compress@x or /Compress were "recognized" here (tag suppressed) yet ran no command there and reached the model unattributed. Matching the exact token leaves wrong-case / @suffix / injection- shaped tokens UNRECOGNIZED -> they keep their [sender] tag (attributed), exactly as the agent treats them. No prompt rewrite: a /compress@otherbot aimed at another bot must not run here. LOCAL commands keep case-insensitive dispatch. - [Suggestion] Include the originating chatId/messageId (sanitized) in the "/clear abandoned a wedged turn" log so oncall can correlate the stuck turn. - [Suggestion] Clarify the alreadyPrefixed JSDoc: it is also set by the QQ adapter on real self-prefixed inbounds, which sanitize the embedded name at the source - so the flag does not bypass sanitization (verified; no behavior change needed). - [Suggestion] Replace the blind `as unknown as {...}` bridge cast on the recognition path with a typed AgentCommandsProvider interface (optional members), so a future rename/return-type change is type-checked instead of breaking at runtime. - [Suggestion] Validate altNames' shape: isAvailableCommand now rejects a non-array altNames, and the recognition site guards the alias check with Array.isArray, so a malformed wire payload can't throw at the `.includes` call. Tests: throwing-onPromptEnd cleanup + collect-drain + log; exact-token recognition (verbatim /compress and /summarize alias; tag kept for /SUMMARIZE, /COMPRESS, /compress@x, and /compress@x + [SYSTEM] line); wedged-turn log carries chat/message; malformed altNames does not throw; isAvailableCommand drops a malformed-altNames entry. Restored stderr spies in the two steer-watchdog tests' finally. Co-Authored-By: Qwen-Coder <noreply@qwen.ai> * fix(channels): address qwen tag review followups --------- Co-authored-by: Qwen-Coder <noreply@qwen.ai> Co-authored-by: qwen-code-dev-bot <qwen-code-dev-bot@users.noreply.github.com> Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> |
||
jinye
|
7f71a8512d
|
docs(daemon): refresh daemon docs for recent PRs (wave 2) (#5954)
* docs(daemon): update developer docs for recent daemon PRs - Add Last-Event-ID client reconnect guide (10-event-bus, 13-sdk-daemon-client) - Add cross-connection vote routing section (04-permission-mediation) - Add new capability tags: daemon_status, workspace_permissions, workspace_trust, workspace_github_setup, workspace_voice, workspace_voice_transcription, voice_transcribe (11-capabilities-versioning) - Add new event types: trust_change_requested, github_setup_completed, extensions_changed, mid_turn_message_injected (09-event-schema) - Fix _meta.serverTimestamp source description (09-event-schema, 10-event-bus) - Fix async function* syntax in SDK example (13-sdk-daemon-client) - Sync event/capability counts across all docs (43->47 events, 67->75 tags) * docs(daemon): add workspace remember design doc (PR #5884) Design document for the sessionless workspace remember API proposed in PR #5884. Covers API endpoints, task lifecycle, implementation details, events, error handling, and SDK integration. Status: Proposed (not yet merged). |
||
易良
|
f33dd61f8a
|
fix(core): stop repeated truncated write_file/edit retries from looping (#5934)
* fix(core): stop repeated truncated edit retries * fix(core): default output tokens to the model limit instead of the 8K cap The 8K CAPPED_DEFAULT_MAX_TOKENS made normal large responses (esp. file writes) truncate, forcing a truncate->escalate round-trip and, worst case, a retry loop. Default to the model's declared output limit instead; the existing escalation + multi-turn recovery stay as the truncation backstop. The 8K cap was a slot-reservation optimization. Claude Code keeps the same cap but gates it behind a feature flag that defaults OFF for third-party providers; qwen-code's providers are all third-party / OpenAI-compatible / self-hosted, so matching that default-off behavior is the safe choice. The capacity tradeoff stays opt-in via QWEN_CODE_MAX_OUTPUT_TOKENS. Refs #5756 * fix(core): use a truncation-specific stop directive for repeated truncated writes * docs: update max tokens configuration wording |
||
|
Heyang Wang
|
1344f34147
|
feat(mcp): reconcile MCP servers live on settings change (#5561)
* feat(mcp): reconcile MCP servers live on settings change Hot-reload MCP servers when settings.json changes (issue #3696 sub-task 3): editing mcpServers / mcp.allowed / mcp.excluded now connects, disconnects, or restarts only the affected servers in place, without restarting the session or losing conversation context. - Part A: Config runtime setters + reinitializeMcpServers incremental reconcile; align the shared-pool path with the #4615 pending-approval gate - Part B: SettingsWatcher subscriber (hotReload.ts), gated on a mcpServers + gating-list diff; flip the three MCP schema keys to hot-reloadable - Part D: re-fire the approval modal for a gated server left pending by an edit - Part E: /mcp shows why a gated server was skipped (pending / rejected) - Record connection fingerprints on the bulk and lazy-connect paths so an edit to a server first connected via those paths is not silently dropped - Design doc (en/zh) incl. the admission-stance boundary clarification Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> # Conflicts: # packages/cli/src/config/settingsSchema.test.ts # Conflicts: # packages/cli/src/ui/components/mcp/steps/ServerDetailStep.tsx # Conflicts: # packages/cli/src/gemini.tsx * feat(mcp): reconcile MCP servers live on settings change Hot-reload MCP servers when settings.json changes (issue #3696 sub-task 3): editing mcpServers / mcp.allowed / mcp.excluded now connects, disconnects, or restarts only the affected servers in place, without restarting the session or losing conversation context. - Part A: Config runtime setters + reinitializeMcpServers incremental reconcile; align the shared-pool path with the #4615 pending-approval gate - Part B: SettingsWatcher subscriber (hotReload.ts), gated on a mcpServers + gating-list diff; flip the three MCP schema keys to hot-reloadable - Part D: re-fire the approval modal for a gated server left pending by an edit - Part E: /mcp shows why a gated server was skipped (pending / rejected) - Record connection fingerprints on the bulk and lazy-connect paths so an edit to a server first connected via those paths is not silently dropped - Design doc (en/zh) incl. the admission-stance boundary clarification Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(mcp): harden hot-reload teardown and reconcile (review follow-ups) Address reviewer findings on the MCP hot-reload changes: - Extract purgeServerRegistries() and use it at every teardown path, fixing the discovery-timeout handler which leaked prompts/resources (only tools were purged) for a server that stalled tools/list past the timeout. - Surface reconcile failures via AppEvent.LogError so a failed settings edit is visible to the user, not just under --debug. - Make a single-session config edit to a discovery filter (trust / includeTools / excludeTools) reconnect the server so discover() re-applies it: connectionIdOf stays transport-only; add singleSessionConnectedKeyOf and rename connectionFingerprints -> connectedConfigKeys. - Make a coalesced reinitializeMcpServers await the in-flight pass + its drain (store mcpReconcilePromise) so the caller no longer emits approval events / logs "complete" before its change is applied; coalesced callers share the failure. - Assert removeResourcesByServer in the fingerprint-change tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * feat(mcp): bound hot-reload MCP admission and explain why servers are unavailable (#3696) (#5561) - K: treat the startup --allowed-mcp-server-names flag as an immutable upper bound — a runtime settings edit may narrow MCP admission within it but never widen beyond it; with no flag, settings fully drive admission. - H: preserve an explicit `mcp.allowed: []` as deny-all (don't collapse to undefined / allow-all), matching boot semantics, and make mcpGatingEqual distinguish absent (allow-all) from [] (deny-all) so the change reconciles. - B: classify why an MCP server is unavailable (removed / not_allowed / excluded / pending_approval) and route the tool-not-found message to the right recovery action; track removals against the gating-independent merged map (dropping the prev-effective snapshot param). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * docs(mcp): document hot-reload admission bound, deny-all, and unavailable reasons (Part F) Reflect the K/H/B changes in the sub-task 3 design doc: add Part F (CLI --allowed-mcp-server-names as an immutable upper bound, mcp.allowed: [] as deny-all, and getMcpServerUnavailableReason routing the tool-not-found message), and fix the now-superseded "settings can widen beyond the startup allowlist" admission-stance note and verification item 11. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * test(serve): pre-approve gated MCP servers in daemon baseline harness The pool/daemon discovery path now honors #4615 pending-approval gating, so the workspace-scoped MCP servers the amplification suite declares in .qwen/settings.json are skipped as pending and never spawn (the suite timed out waiting for grandchildren). Add approveWorkspaceMcpServers() to the harness (keyed by the realpath workspace to match the daemon's canonicalized --workspace) and pre-approve the fixtures before boot, mirroring simple-mcp-server.test.ts. --------- Co-authored-by: heyang.why <heyang.why@alibaba-inc.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
ChiGao
|
a7b9c5d915
|
feat(tui): partition tool display by type — collapse read/search, show mutation tools individually (#5661)
* feat(tui): remove tool group borders and collapse completed tool results Remove round borders from ToolGroupMessage, CompactToolGroupDisplay, and InlineParallelAgentsDisplay. Completed tools now default to a single collapsed header line with dimColor styling. Executing/error/confirming tools continue to show their full result block. Part of #4588 (Track 3: Simplify tool-call rendering). Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): gate collapse on compact mode and fix innerWidth calculation - Only collapse completed tool results in compact mode, preserving full visibility in non-compact mode - Subtract 2 from innerWidth to account for ToolMessage paddingX={1} - Update snapshots to reflect removed borders Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): address review feedback on collapse and visual alignment - Gate isDim on compact mode so non-compact tools stay fully styled - Add paddingX={1} to CompactToolGroupDisplay for left-edge alignment - Delete Border Color Logic test block (borders removed) - Add compact-mode test coverage for Error/Executing/Pending/forceShowResult - Clean up stale border references in comments Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * feat(tui): unify tool output with semantic summaries Replace the dual compact/normal mode tool output with a single unified mode. Completed tools always show a semantic overview line ("Read 3 files, edited 2 files") instead of dumping full results. - Add buildToolSummary() for category-based semantic summaries - Remove compactMode gate from shouldCollapse and isDim in ToolMessage - Make all-completed tool groups use CompactToolGroupDisplay - Remove unused useCompactMode hook calls from ToolMessage Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(tui): add buildToolSummary unit tests and fix stale comment - Add 10 dedicated unit tests for buildToolSummary covering edge cases - Fix stale comment referencing old compactMode gate logic Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): address audit findings for unified tool output - Add Canceled status to allComplete check in ToolGroupMessage - Move memory-only group rendering before showCompact to prevent them being swallowed by CompactToolGroupDisplay - Fix LLM summary duplication: absorbedCallIds now tracks completed groups in non-compact mode; HistoryItemDisplay no longer bypasses summaryAbsorbed when !compactMode - Update StandaloneSessionPicker test for new compact rendering - Fix design doc category order example and add missing rendering rules Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): address inline review findings - Add SHELL_COMMAND_NAME and @ file-reference pseudo-tools to TOOL_NAME_TO_CATEGORY mapping for correct category classification - Fix height calculation test to use Executing status so expanded path is actually exercised - Update stale comment about empty toolCalls behavior Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): remove unused compactMode import in HistoryItemDisplay Fixes CI build failure caused by TS6133 (noUnusedLocals) — the compactMode destructure became dead code after the summary gating was moved to summaryAbsorbed. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * ci: trigger re-run with updated merge ref Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * refactor(tui): partition tools by type instead of completion status Align tool display with read/search collapse pattern: only information-gathering tools (read, search, list) are collapsed into a summary line; mutation tools (edit, write, command, agent) always render individually with their results. - Remove compactLabel prop and cross-group merge logic from MainContent - Add isCollapsibleTool predicate and partition in ToolGroupMessage - Remove shouldCollapse gate from ToolMessage (results always shown) - Update CATEGORY_ORDER to search → read → list → command → ... - Update tests and snapshots to match partition-based rendering Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): collapse text/ANSI output for completed tools Only string and ANSI results are hidden for completed (Success/Canceled) tools — diff, plan, todo, and task results always render since they carry non-repeatable information the user needs to review inline. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): unify summary format and show results in error-expanded groups - buildToolSummary always uses count format ("Read 1 file") instead of description for single tools, ensuring consistent display across all collapsible tool groups - Pass forceExpandAll to forceShowResult so sibling Success tools in error-expanded groups keep their results visible for diagnostics - Add memory-only group test coverage (Recalled N memories / Wrote N) Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(tui): improve coverage for partition logic and result collapse - Add isCollapsibleTool unit tests (read/search/list → true, others → false) - Add partition tests: pure collapsible → summary, mixed → summary + individual - Add forceExpandAll tests: error group bypasses partition, all siblings get forceShowResult - Add diff bypass test: completed Success tool with diff result is not collapsed - Update MockTool to surface forceShowResult flag for assertion - Update snapshots reflecting forceExpandAll → forceShowResult propagation Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): align CompactToolGroupDisplay style with ToolInfo Replace dimColor={!isActive} with bold to match ToolInfo's styling in ToolMessage. Completed summary lines now render at normal brightness with bold text, consistent with sibling non-collapsible tools in the same group. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): address review findings on result collapse and memory safety - Only collapse results for collapsible tools (read/search/list), not MCP/WebFetch/other tools whose output is the answer - Only collapse on Success, not Canceled (partial output may be useful) - Exclude errored memory ops from allMemOpsComplete early return so error details remain visible - Add memory badge to all-collapsible early return path - Simplify forceShowResult to forceExpandAll (per-tool conditions were already subsumed by group-level flags) - Account for collapsible summary row in height budget - Remove vestigial mergedHistory alias - Remove tool_group from compactToggleHasVisualEffect (compact mode no longer affects tool rendering) - Fix outdated JSDoc in buildToolSummary Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * docs(tui): sync design doc with actual implementation Update unified-tool-output.md to reflect: - Type-based partition (isCollapsibleTool) instead of completion-based - Count format for all summaries ("Read 1 file" not "Read package.json") - shouldCollapseResult gated on collapsible tools and Success only - Bold styling for all tool names (isDim removed) - tool_use_summary renders unconditionally (absorption removed) - forceShowResult simplified to forceExpandAll - compactToggleHasVisualEffect no longer triggers on tool_group Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(tui): update AppContainer test for compactToggleHasVisualEffect change tool_group no longer triggers a visual effect on Ctrl+O, so the test now expects refreshStatic to be skipped for tool-group-only histories. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): restore per-tool forceShowResult and harden edge cases - Revert forceShowResult to per-tool computation so shell output capping is preserved for successful siblings in force-expanded groups - Exclude Canceled tools from collapsible partition so partial output stays visible via individual ToolMessage rendering - Extract shared MemoryBadge (deduplicate IIFE in all-collapsible and mixed paths) - Add memory badge height to staticHeight budget in mixed path - Add legacy ToolDisplayNamesMigration entries (SearchFiles, FindFiles, ReadFolder, Task, TodoWrite) to TOOL_NAME_TO_CATEGORY - Fix stale comment referencing removed showCompact Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(tui): add coverage for canceled partition, memory badge, and legacy names - Canceled collapsible tool renders individually (not absorbed into summary line), preserving partial output visibility - Mixed group with memory counts renders memory badge alongside collapsible summary and individual tools - Legacy display names (SearchFiles, FindFiles, ReadFolder, Task, TodoWrite) map to correct categories in isCollapsibleTool and buildToolSummary Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): document dual effect of isCollapsibleTool and add ANSI collapse test Update isCollapsibleTool JSDoc to document both decision points (ToolGroupMessage partition and ToolMessage.shouldCollapseResult) so future maintainers understand that adding a category suppresses both grouping and result output. Add test for ANSI renderer branch of shouldCollapseResult to ensure ANSI output from collapsible tools is also collapsed. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(tui): address review round 6 — type fix, stale refs, and test gaps - Fix TS2740: add required AnsiToken properties to ANSI collapse test - Rename mergedLengthRef → visibleHistoryLengthRef to match removed mergedHistory concept - Fix design doc Rule 5: forceExpandAll forces results only for triggering tools, not all siblings - Update compactToggleHasVisualEffect JSDoc: tool_group no longer checked, only gemini_thought items affected by compact mode - Fix diff bypass test: use collapsible tool name (ReadFile) so shouldCollapseResult is actually exercised - Add all-collapsible memory badge test covering the early-return path Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(tui): add isUserInitiated and memory-only error test coverage - Add test: user-initiated group renders all collapsible tools individually (bypasses partition into summary line) - Add test: memory-only group with errored tool falls through to expanded path instead of showing compact badge Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> --------- Co-authored-by: 秦奇 <gary.gq@alibaba-inc.com> Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> |
||
|
jinye
|
a8863c203c
|
refactor(cli): Finish serve kebab-case filenames (#5604) | ||
|
jinye
|
b4705b2534
|
refactor(cli): rename serve files to kebab-case (#5592)
Rename the PR1 serve and daemon adapter files from issue #5576 to kebab-case and update current imports, tests, comments, and developer docs to match. Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> |
||
|
Heyang Wang
|
a8eb824fb7
|
feat(config): add settings file change detection via chokidar watcher (#3696) (#4933)
Co-authored-by: heyang.why <heyang.why@alibaba-inc.com> |
||
Zqc
|
abfdcc112b
|
fix(core): prevent OOM in auto-memory extraction during /quit (#5147) (#5181)
The auto-memory extraction background task caused FATAL ERROR: Reached heap limit after /quit because buildTranscriptMessages() ran .replace() over every history message synchronously, but the resulting text was never consumed downstream. Changes: - extract.ts: delete buildTranscriptMessages/loadUnprocessedTranscriptSlice. Replace with a zero-stringify cursor scan on the unprocessed slice only. - manager.ts: add isUnderMemoryPressure() using existing MemoryPressureMonitor. Gates in runExtract(), scheduleDream(), and scheduleSkillReview(). - client.ts: add shutdownRequested flag + requestShutdown(). - AppContainer.tsx: call requestShutdown() in /quit callback. - Tests: 46 passed (11 extract + 34 manager + 2 client). - Docs: update memory-system.md flowchart to cursor-first flow. Co-authored-by: 俊良 <zzj542558@alibaba-inc.com> |
||
Yan Shen
|
8d2fe0a798
|
feat(stats): expose token usage for cost visibility (#4564)
* feat(stats): expose token usage for cost visibility Persist content-free API token counters and surface daily/monthly summaries plus CSV/JSON export through /stats. Constraint: Issue #4479 requested CLI token visibility with monthly/model breakdowns and export while coordinating with #4252/#4182.\nRejected: Add a separate top-level token command | /stats keeps related statistics in one surface.\nConfidence: high\nScope-risk: moderate\nDirective: Keep TTFT/TPS generation timing and memory diagnostics outside this token-usage surface unless their issues explicitly broaden scope.\nTested: npx vitest run src/services/tokenUsageService.test.ts src/telemetry/loggers.test.ts; npx vitest run src/ui/commands/statsCommand.test.ts src/ui/hooks/useAutoAcceptIndicator.test.ts src/ui/components/AutoAcceptIndicator.test.tsx; npm run check-i18n --workspace=packages/cli; npm run lint --workspace=packages/cli; npm run lint --workspace=packages/core; npm run typecheck; npm run build; git diff --check\nNot-tested: full integration suite * fix: Address token usage review feedback Tighten persisted token usage so internal prompt traffic and disabled usage statistics do not write history, while surfacing non-ENOENT write failures outside debug logs. Complete the reviewer-requested i18n coverage and regression tests around auto mode notices and best-effort writes. Constraint: Follow-up to wenshao review comments on PR #4564. Rejected: Keeping token usage recording outside the internal-prompt gate | It would inflate daily and monthly stats with background prompts. Confidence: high Scope-risk: narrow Directive: Keep /stats token usage scoped to user-visible API responses unless future requirements explicitly include background traffic. Tested: npx vitest run src/services/tokenUsageService.test.ts src/telemetry/loggers.test.ts; npx vitest run src/ui/hooks/useAutoAcceptIndicator.test.ts src/ui/commands/statsCommand.test.ts; npm run typecheck; npm run lint --workspace=packages/core; npm run lint --workspace=packages/cli; npm run check-i18n --workspace=packages/cli; npm run build; git diff --check Not-tested: Full repository test suite * fix(stats): satisfy token usage review contract Constraint: wenshao review required consistent token stats, exports, i18n, and best-effort logging behavior. Rejected: Change cached-token labeling | keeping cached tokens included in input preserves the accepted /stats display contract. Confidence: high Scope-risk: narrow Directive: Keep cached tokens included in input whenever cached-only metadata is used in total fallback. Tested: cd packages/core && npx vitest run src/services/tokenUsageService.test.ts src/telemetry/loggers.test.ts Tested: cd packages/cli && npx vitest run src/ui/commands/statsCommand.test.ts src/i18n/mustTranslateKeys.test.ts Tested: npm run check-i18n --workspace=packages/cli; npm run typecheck; git diff --check Not-tested: full integration suite * Refine token usage recording after review Constraint: Address wenshao's latest PR #4564 review suggestions without expanding the /stats command surface. Rejected: Keeping synchronous token-usage writes | sync I/O remains on the API response hot path. Confidence: high Scope-risk: narrow Directive: Keep token usage persistence best-effort and gated by explicit usage-statistics enablement. Tested: cd packages/core; npx vitest run src/services/tokenUsageService.test.ts src/telemetry/loggers.test.ts; cd packages/cli; npx vitest run src/ui/commands/statsCommand.test.ts; npm run typecheck; npm run build; npm run lint --workspace=packages/core; npm run lint --workspace=packages/cli; git diff --check Not-tested: Full repository test suite * fix(stats): avoid silent zero usage on read failures Propagate token usage read failures through the existing /stats error path while keeping missing usage files empty, and remove the unreachable telemetry wrapper catch. Constraint: PR #4564 review requested user-visible read failures, full i18n for export errors, and removal of dead telemetry catch code. Rejected: Adding warning fields to TokenUsageSummary | It would expand the JSON/export schema when the existing command error path already fits read failures. Confidence: high Scope-risk: narrow Directive: Keep jsonl.read default swallowing behavior for existing session/history callers unless a user-visible caller opts into rethrowing non-ENOENT errors. Tested: npx vitest run src/utils/jsonl-utils.test.ts src/services/tokenUsageService.test.ts src/telemetry/loggers.test.ts Tested: npx vitest run src/ui/commands/statsCommand.test.ts Tested: npm run check-i18n --workspace=packages/cli Tested: npx prettier --check changed files Tested: npm run typecheck Tested: npm run lint --workspace=packages/core Tested: npm run lint --workspace=packages/cli Tested: git diff --check Tested: npm run build Not-tested: Full integration test suite * fix: close token usage review gaps Keep the review follow-ups local to token usage accounting and stats export without adding new abstractions. Constraint: Address PR #4564 reviewer requests on token usage export/query reuse, write-failure stderr noise, and invalid-record diagnostics. Confidence: high Scope-risk: narrow Directive: Keep token usage writes best-effort and avoid noisy stderr loops for repeated local failures. Tested: git diff --check; prior targeted core/cli tests, typecheck, and lint passed for this working tree. Not-tested: Full repository test suite. * fix(stats): address token usage review feedback * Update packages/core/src/services/tokenUsageService.ts Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> * test(core): keep token usage stderr assertion current Keep the repeated write-failure regression test aligned with the runtime wording that the PR now emits. Constraint: PR #4564 CI failed after the implementation wording changed to "since last log". Rejected: Reverting the implementation wording | it is the latest PR behavior and the failure is test-only. Confidence: high Scope-risk: narrow Tested: cd packages/core && npx vitest run src/services/tokenUsageService.test.ts Not-tested: full repository test suite * fix(stats): clarify export review edge cases Address the remaining PR review polish without changing token accounting, export formats, or path containment behavior. Constraint: Review 4452925552 requested narrow documentation, ENOENT wording, and NOTICES cleanup only. Rejected: Broader merge-conflict rework | GitHub currently reports the PR as mergeable, and the requested fixes are review polish. Confidence: high Scope-risk: narrow Directive: Keep token usage records content-free and preserve export path validation semantics except for the final ENOENT message. Tested: cd packages/core && npx vitest run src/services/tokenUsageService.test.ts; cd packages/cli && npx vitest run src/ui/commands/statsCommand.test.ts; npm run check-i18n --workspace=packages/cli; npm run typecheck; git diff --check on changed code and i18n files Not-tested: Full test suite not run. * fix: address token usage review feedback * fix(stats): harden token usage CSV export * fix(stats): remove unrelated auto mode noise --------- Co-authored-by: Shaojin Wen <shaojin.wensj@alibaba-inc.com> |
||
|
qqqys
|
f4d405ca4a
|
feat(loop): wire prompt-only /loop to self-paced wakeups (#5197)
* feat(loop): add second-resolution session wakeup engine
Add a session-scoped wakeup primitive for self-paced /loop, aligned with
Claude Code's ScheduleWakeup. An independent, second-resolution channel in
CronScheduler — separate from cron jobs (never durable, not counted against
MAX_JOBS, fired at an exact time, not minute-rounded):
- scheduleWakeup(delaySeconds, prompt): clamps to [60, 3600]s (1200s default
for non-finite input); returns {scheduledFor, clampedDelaySeconds, wasClamped}.
- Fires through the existing onFire channel and counts toward sessionSize, so
there are no cli delivery-path changes and a pending wakeup holds a headless
run open — re-arm keeps the loop alive, omitting the call ends it.
- cancelWakeup / cancelAllWakeups primitives (for loop-scoped cancellation).
- loop_wakeup tool: delaySeconds schema, structured clamp output, cache-window
picking guidance, verbatim /loop prompt, reason shown to the user, and the
"call to keep alive / omit to end" contract — all mirroring ScheduleWakeup.
getDefaultPermission stays 'ask' (out of SAFE_TOOL_ALLOWLIST) so AUTO still
routes scheduling future model input through the classifier, like CronCreate.
Closes #5156
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(loop): tighten session wakeup lifecycle
* fix(loop): address wakeup review nits
* feat(loop): wire prompt-only /loop to self-paced wakeups
Make `/loop <prompt>` (no interval) a self-paced loop in the bundled loop
skill: run the prompt immediately, then schedule at most one future
continuation via loop_wakeup (delaySeconds) — no recurring cron.
- Three explicit paths: prompt-only self-paced (LoopWakeup), fixed-interval
recurring (CronCreate), and list/clear management (CronList/CronDelete).
- The continuation uses delaySeconds (aligned with the second-resolution
wakeup engine) and re-feeds `/loop ${original prompt}` verbatim to re-enter
the skill; the model re-arms only when a further check is useful.
- Adds loop_wakeup to the skill's allowedTools.
- Static SKILL contract tests, including delaySeconds (not delayMinutes).
Closes #5184
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(loop): clarify wakeup delay reporting
* fix(loop): make wakeups manageable
* fix(loop): address wakeup review feedback
* fix(loop): clarify wakeup management wording
* fix(loop): clarify wakeup continuation tooling
* fix(loop): bound self-paced wakeup chains
* fix(loop): distinguish wakeup fires in lists and UI
* test(loop): cover wakeup labels in cli paths
* fix(loop): enforce session wakeup chain limit
* fix(loop): align wakeup delay metadata
* fix(loop): handle stopped wakeup scheduling
* fix(loop): make the wakeup chain limit a true session-level budget
The 24h chain limit reset `wakeupChainStartedAt` whenever `wakeups`
emptied — on every fire and on cancel. Because a self-paced loop leaves
at most one pending wakeup, each fire emptied the map and restarted the
clock, so a continuous re-arming loop never reached the cap (and a
cancel-then-reschedule could reset it too).
Reset the chain clock only on stop()/destroy() (a new session): the 24h
budget now spans the whole session, bounds continuous re-arming, and
closes the cancel bypass. Tests cover the clock persisting across fires,
cancel not resetting it, and stop starting a fresh budget.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(loop): correct CronList durability wording and bound wakeup prompt
Address two review suggestions on #5197:
- CronList tool description said cron jobs and loop wakeups are "both
session-only and durable", implying wakeups can be durable. Loop
wakeups are always session-only; only cron jobs can be durable.
Reword so the model isn't misled into expecting durable wakeups.
- LoopWakeup `prompt` had no maxLength, unlike sibling tools
(task-create, send-message). Add maxLength: 10000 to bound the
model-generated continuation prompt.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(loop): let the first wakeup arm before the scheduler starts
Critical (wenshao, #5197): LoopWakeup hard-rejected when
`scheduler.running` was false, but on the first self-paced /loop in a
session with no cron jobs the scheduler hasn't started yet
(#startCronSchedulerIfNeeded bails on !hasPendingWork). The post-prompt
hook starts the tick *after* the turn, once a wakeup exists — so the
guard rejected the very call that makes the loop possible, breaking the
primary use case.
The `running` check was a proxy for "cron is alive", added to reject
re-arms after the token-limit breaker. Replace it with an explicit,
permanent `disabled` state so the two cases are distinguishable:
- CronScheduler gains `disabled` + `disable()` (sets the flag, stops).
- LoopWakeup rejects only when `scheduler.disabled`, not when merely
stopped — a stopped-but-restartable scheduler still accepts wakeups.
- The token-limit breaker calls `disable()` instead of `stop()`, so its
rejection (the original intent) is preserved.
Also attribute cron-prompt errors by source: `[loop error]` vs
`[cron error]` (item.source was already in scope).
Tests: reject-when-disabled, schedule-when-stopped (the regression),
and a disable() unit test.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(loop): clear the pending wakeup when a re-arm exceeds the 24h budget
Critical (round-4 review, #5197): scheduleWakeup() threw the 24h-limit
error *before* clearing the prior wakeup, so a rejected re-arm left the
previous wakeup in the map. Its fireAtMs is now in the past, so the next
tick fires it — one iteration past the budget it's meant to cap. A test
even codified this (sessionSize === 1 after a rejected re-arm).
Production can't actually reach it (the 1s tick fires the wakeup at
~+3600s, and a stopped scheduler clears wakeups), but the safety budget
should hold unconditionally. Clear the pending wakeup up front, before
the budget check, so a rejected re-arm leaves nothing behind. Update the
test to assert no wakeup remains.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(loop): update Session token-limit test for the disable() breaker
CI regression from the disable() refactor (
|
||
jifeng
|
858c900af9
|
feat(serve): add daemon idle detection to GET /health?deep=true (#4934)
* docs(design): add daemon idle detection API design for machine reclamation When qwen daemon is deployed across multiple machines, an external scheduler needs a reliable signal to determine if a daemon is idle and the machine can be reclaimed. This design proposes enhancing GET /health?deep=true with activePrompts, connectedClients, channelAlive, lastActivityAt, and idleSinceMs fields. * feat(serve): add idle-detection fields to GET /health?deep=true * fix(test): add activePromptCount and lastActivityAt to fakeBridge Update the test helper to satisfy the expanded AcpSessionBridge interface so /health?deep=true tests pass with the new fields. * test(idle-detection): add unit tests for activePromptCount, lastActivityAt and /health?deep=true new fields * refactor(bridge): replace activePromptCount iteration with O(1) counter * fix(bridge): move idleSinceMs computation into bridge to eliminate race window * fix(idle-detection): keep daemon health state consistent Guard prompt teardown so activePromptCount only decrements once and compute deep health idle fields from the same activity snapshot. * fix(acp): prevent active prompt leak after channel crash Reject queued prompts after their session entry has been torn down so daemon health does not report phantom active prompts. --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> |
||
|
ChiGao
|
ce4b0cf629
|
feat(sdk,serve): DaemonTransport abstraction + ACP standard compliance (#5040)
* feat(sdk): DaemonTransport abstraction — pluggable transport for REST/ACP-HTTP/ACP-WS
- DaemonTransport interface with fetch + subscribeEvents
- RestSseTransport: extract current SSE logic from DaemonClient
- AcpWsTransport: WebSocket multiplexer + URL-to-JSON-RPC mapping
- AcpHttpTransport: POST /acp + session-scoped SSE
- AcpEventDenormalizer: JSON-RPC notification -> DaemonEvent
- AutoReconnectTransport: opt-in reconnect + fallback wrapper
- negotiateTransport(): auto-detect best transport via GET /capabilities
- Provider: DaemonWorkspaceProvider gains transport prop
- Server: GET /capabilities advertises supported transports
- Zero breaking changes: no transport = current REST behavior
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* docs(design): include DaemonTransport design doc in implementation PR
* fix(sdk): address 6 verification findings — bundle size, WS hang, error types, ACP compat
- Remove ACP transport class re-exports from barrel (index.ts) to avoid
~19.7KB browser bundle bloat; keep type-only exports
- Fix WS dial hang: reject connect promise in onerror when not yet
connected (Node WebSocket may only fire error, not close)
- Fix parked generators: maintain _activeGenerators set, abort all on
WS close so generators throw DaemonTransportClosedError
- Forward abort signal through AcpHttpTransport.sendRequest to fetch
- Restore DaemonHttpError in RestSseTransport (was plain Error)
- ACP endpoint compat: extract connectionId from initialize, send
Acp-Connection-Id header, add _qwen/ prefix for vendor methods,
preserve real HTTP status in error mapping, fetch /capabilities
from REST endpoint for correct shape
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): address 16 review findings + CI bundle size
- CI: move negotiateTransport to separate file, extract DaemonHttpError
to break static import chain from barrel -> DaemonClient. Browser
bundle drops from 136KB to 115KB, under the 116KB budget.
- Route table: extract shared acpRouteTable.ts, used by both transports.
Unify method naming (remove _qwen/ prefix inconsistency).
- Token: move from URL query to Authorization header on WS upgrade
- Error type: DaemonHttpError extracted to DaemonHttpError.ts; import
in RestSseTransport no longer pulls in DaemonClient.
- Init retry: reset failed initPromise so next call retries
- Reconnect mutex: prevent concurrent reconnect storms
- Generator queue: cap at 256, drop-oldest
- WS init timeout: 30s default
- negotiate: clear timer on all paths, catch dispose rejection
- Headers: forward init.headers in ACP transports via mergeHeaders()
- Dead code: remove unused pendingRequests/sseAbort fields
- Provider: dispose client on unmount
- Helpers: extract matchRoute/synthesizeResponse/jsonRpcErrorToHttpStatus/
isRecord/composeAbortSignals to shared acpTransportUtils.ts
- Package exports: add deep import paths for ACP transports
- Tests: add AcpEventDenormalizer unit tests (17 cases)
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): ESLint array-type rule — ReadonlyArray<T> → readonly T[]
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): fix 3 ACP wire bugs + bundle size + npm exports
Wire bugs (verified broken against real daemon):
1. AcpHttpTransport: read connectionId from response header + correct JSON path
2. AcpWsTransport: send token via Authorization header, not URL query
3. AcpEventDenormalizer: read params.update.sessionUpdate, not params.type
Bundle: remove negotiateTransport from barrel-reachable imports
Exports: add package.json deep import paths for ACP transports
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* test(sdk): comprehensive ACP transport test suite (~175 tests)
- RestSseTransport: fetch delegation, SSE subscribe, auth, timeout, signal
- AcpWsTransport: route mapping, token auth, event filtering, queue cap
- AcpHttpTransport: connectionId extraction, header injection, init retry
- AutoReconnectTransport: reconnect mutex, fallback, delegation
- negotiateTransport: capability probing, timeout, fallback
- acpRouteTable: URL→method mapping, param extraction
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): route table coverage, browser WS auth, header forwarding, capabilities type
- Route table: add file/stat/list/glob/write/edit paths (all DaemonClient URLs)
- Route table: add session diagnostic routes (context, tasks, stats, rewind, language)
- Route table: add bulk sessions/delete
- WS auth: document browser limitation, Node uses headers, browser needs proxy
- Headers: forward X-Qwen-Client-Id via JSON-RPC _meta in WS transport
- DaemonCapabilities: add transports field to SDK type
- Package exports: remove unreachable deep exports, document monorepo usage
- Provider bypass: document limitation for glob/stat/list in workspace actions
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): add missing detach + hooks routes per QA doc
Cross-referenced with daemon-acp-integration-qa.md route table.
Added POST /session/:id/detach and GET /session/:id/hooks.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve,sdk): enforce ACP standard session/new — always isolated session
ACP standard mandates session/new MUST create a new isolated session.
Server-side (dispatch.ts):
- Force sessionScope='thread' on /acp session/new, ignoring client params
- REST POST /session retains 'single' default for backward compat
SDK-side (acpRouteTable.ts):
- Strip sessionScope from session/new params in ACP transports
- Document that ACP follows the standard (no extensions)
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve): ACP session/new returns standard models/modes fields
ACP standard NewSessionResponse includes optional `models` and `modes`
top-level fields alongside `configOptions`. Extract model/mode state
from configOptions and surface them as standard-shaped objects:
- models: { currentModelId, availableModels: [{id}] }
- modes: { currentModeId, availableModes: [{id}] }
Also update test to verify sessionScope is always forced to 'thread'
(ACP standard compliance — session/new always creates isolated session).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* feat(serve): add standard ACP methods session/set_mode, session/set_model, session/fork
Align /acp endpoint with ACP standard protocol:
- session/set_mode: dedicated method for mode changes (standard)
Maps to bridge.setSessionApprovalMode(). Params: {modeId, sessionId}
- session/set_model: dedicated method for model changes (unstable)
Maps to bridge.setSessionModel(). Params: {modelId, sessionId}
- session/fork: create a branched copy of an existing session
Maps to bridge.branchSession(). Response includes configOptions,
models, modes per ACP standard.
- session/load, session/resume: responses now include configOptions,
models, modes (per ACP LoadSessionResponse/ResumeSessionResponse)
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve): TS2345 — pass persist: false to setSessionApprovalMode
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(webui): add dispose() to MockDaemonClient in provider tests
DaemonClient now has dispose() (called in provider cleanup effect).
Mock clients in test files need to implement it.
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(serve): add sessionId pre-validation + remove type assertion
- session/set_mode, session/set_model: add explicit sessionId empty
check before requireOwned (consistent with session/fork)
- session/set_model: remove `as unknown as` type assertion, pass
proper {modelId, sessionId} matching SetSessionModelRequest
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk,serve): align route table with dispatcher + AcpHttp SSE response correlation
Route table:
- Add _qwen/ prefix to all vendor session/workspace methods
- Split workspace catch-all into granular dispatcher methods
- Fix session/branch → session/fork, model → session/set_model
- Remove routes with no dispatcher handler
AcpHttpTransport:
- Implement conn-scoped SSE stream for response correlation
- POST returns 202 (ack), real response rides SSE stream
- Map<id, {resolve, reject}> for pending request correlation
dispatch.ts:
- Remove session/set_mode, session/set_model from CONN_ROUTED_METHODS
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): bump browser bundle budget 116KB→118KB for transport abstraction
Main uses 117,753 bytes (99.1% of 116KB budget). The transport
abstraction adds ~1.5KB (DaemonTransport interface + RestSseTransport
default constructor in DaemonClient). Bump to 118KB (120,832 bytes).
Also change RestSseTransport to type-only export from barrel (class
is constructed internally by DaemonClient, not needed as a value
export for consumers).
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): fix 2 test failures — SSE error message + workspace catch-all route
- RestSseTransport: error message 'SSE response has no body' → 'No SSE body'
(matches existing DaemonClient.test.ts assertion)
- acpRouteTable: re-add GET/POST /workspace/* catch-all after granular routes
(AcpWsTransport.test.ts expects generic workspace path to resolve)
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(sdk): align RestSseTransport test with updated error message
Generated with AI
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
---------
Co-authored-by: 秦奇 <gary.gq@alibaba-inc.com>
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
|
||
tanzhenxin
|
2ba4ca90ad
|
feat(core): durable cron jobs — /loop tasks that survive restarts (#5004)
Some checks are pending
Qwen Code CI / Lint (push) Blocked by required conditions
Qwen Code CI / Classify PR (push) Waiting to run
Qwen Code CI / Test (macos-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (ubuntu-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (windows-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Post Coverage Comment (push) Blocked by required conditions
Qwen Code CI / CodeQL (push) Blocked by required conditions
E2E Tests / E2E Test (Linux) - sandbox:docker (push) Waiting to run
E2E Tests / E2E Test (Linux) - sandbox:none (push) Waiting to run
E2E Tests / E2E Test - macOS (push) Waiting to run
Persist /loop tasks per-project under ~/.qwen/tmp/<project-hash>/ so they survive restarts; the default stays session-only. Missed one-shots are surfaced at startup confirm-first; overdue recurring jobs catch up once then resume. A per-project lock elects a single firing session across concurrent sessions, with takeover on owner exit. Recurring jobs expire after 7 days (final fire), and never-matching cron expressions are rejected at creation. Durable storage lives in the user runtime dir, not the working tree, so it is never committed or shared via the repo. |
||
yao
|
8000342667
|
fix(core): remove unused debugResponses array and dead extractUsageFromGeminiClient (#4982) | ||
|
jinye
|
531a15dd93
|
feat(daemon): merge daemon-mode feature batch into main (#4490)
Some checks are pending
Qwen Code CI / Classify PR (push) Waiting to run
Qwen Code CI / Lint (push) Blocked by required conditions
Qwen Code CI / Test (macos-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (ubuntu-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (windows-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Post Coverage Comment (push) Blocked by required conditions
Qwen Code CI / CodeQL (push) Blocked by required conditions
E2E Tests / E2E Test (Linux) - sandbox:docker (push) Waiting to run
E2E Tests / E2E Test (Linux) - sandbox:none (push) Waiting to run
E2E Tests / E2E Test - macOS (push) Waiting to run
* perf(core): F2 cleanup PR A — R9/W11/W12/R10 (post-merge follow-ups) (#4411) * refactor(core): F2 PR A R9 — McpClientManager options-object ctor R9 (filed as F2 follow-up from #4336 review): 7 positional ctor args collapse to (config, toolRegistry, options?: McpClientManagerOptions). The trailing 5 (eventEmitter, sendSdkMcpMessage, healthConfig, budgetConfig, pool) become named fields on `McpClientManagerOptions`. Test factory `mkManager(overrides?)` introduced at the top of `mcp-client-manager.test.ts` so each of the prior 80 inline constructions becomes a single line naming only the field(s) the test overrides; the 4 `undefined` sentinels each test threaded through to reach the trailing `pool` arg are gone. Net: 113 LOC removed (test) + 35 LOC added (src exposes interface + mkManager factory + tool-registry call site update). Behavior unchanged — same field assignments, same downgrade-enforce-without- budget breadcrumb, same budget event wiring. Filed bucket: F2 perf / cleanup PR A (R9 + W11 + W12 + R10/R23 T7), see issue #4175 item 7 "F2 post-merge cleanup PRs". This is the first of the 4 fixes in PR A; W11/W12/R10 follow as separate commits. Test sweep: 84/84 mcp-client-manager.test.ts pass; typecheck clean. * refactor(core): F2 PR A W11 — extract attachPooledSession + rollbackReservationOnSpawnFailure W11 (filed as F2 follow-up from #4336 review): two private helpers on `McpTransportPool` to eliminate inline duplication in `acquire()`: - `attachPooledSession(entry, id, serverName, cfg, sessionId, toolReg, promptReg)`: builds `SessionMcpView` + `entry.attach` with the standard pool release callback. Used by both the fast-path attach (existing entry) and the post-spawn attach (after `await inFlight`). NOT used by `createUnpooledConnection` — its release callback runs `entry.forceShutdown('manual')` + `indexDetach` directly (no pool refcount accounting since unpooled entries are per-session). - `rollbackReservationOnSpawnFailure(reservationResult, serverName)`: R24 T17 contract — only release the budget slot if THIS acquire actually reserved a new slot (`'reserved'`); `'already_held'` skips because the sibling owns it. Used by both the unpooled catch and the pooled spawn-in-flight catch. Race-window invariants (W10 / W77 / W90 / W111 / W125 / R24 T17) stay at the call sites because they describe the SURROUNDING ordering, not the helpers themselves. Helpers are documented to defer those decisions back to callers. Behavior unchanged. Filed bucket: F2 perf cleanup PR A (R9 done / W11 this commit / W12 + R10 to follow). Test sweep: 28/28 mcp-transport-pool.test.ts pass; typecheck clean. * refactor(core): F2 PR A W12 — SessionMcpView precompute filter Sets W12 (filed as F2 follow-up from #4336 review): `applyTools` / `applyPrompts` precompute `excludeSet` + `includeSet` once per pass instead of scanning `cfg.includeTools` / `cfg.excludeTools` arrays inside every per-tool iteration. Pre-fix the per-tool predicate (`passesSessionFilter`) walked both arrays for every snapshot entry → O(M × N) per `applyTools` call. With M tools × N filter entries, typical M=5-20 / N=2-5 case finishes in microseconds either way; the win is data-structure correctness and code clarity, not perceived perf. `passesSessionFilter` / `passesSessionPromptFilter` (the array- based predicates) stay exported and unchanged for unit tests + any caller wanting to test a single name without paying Set construction. The bulk path uses two new private helpers `compileNameFilter` + `compiledFilterAccepts` whose Sets live on the `applyTools` / `applyPrompts` stack frame. Same semantics: `excludeTools` is direct-equality match (no parens strip — pre-F2 behavior preserved); `includeTools` strips the first `(...)` suffix so `toolName(args)` matches `toolName`. Filed bucket: F2 perf cleanup PR A (R9 + W11 done / W12 this commit / R10 to follow). Test sweep: 13/13 session-mcp-view.test.ts pass; typecheck clean. * perf(core): F2 PR A R10 / R23 T7 — pid-descendants ps snapshot + pgrep fallback R10 / R23 T7 (filed as F2 follow-up from #4336 review): the Linux / macOS pid-descendant enumeration moves from per-pid `pgrep -P <pid>` BFS (one subprocess fork per node visited) to a single `ps -A -o pid=,ppid=` snapshot followed by an in-memory tree walk over `Map<ppid, pid[]>`. Windows analog: single `Get-CimInstance Win32_Process | ConvertTo-Csv` snapshot of all `(ProcessId, ParentProcessId)` rows replaces per-pid `Get-CimInstance -Filter "ParentProcessId=$p"` BFS. Two motivations: 1. **Fork count**: typical `npx → tool` / `uvx → tool` wrapper trees are 2-3 levels deep with B=1-3 children per node → pre-fix BFS forked ~5-10 subprocesses per pool-shutdown call. Post-fix: exactly 1 fork regardless of tree depth. 2. **Snapshot consistency**: pre-fix BFS walked the table level by level; a child that forked between two adjacent BFS levels could be missed (we'd see the child but query its descendants AFTER the new fork). The snapshot path captures the table at one instant; new descendants forked after the snapshot are tolerated by the existing ESRCH-tolerant SIGTERM loop. Caveats: - `ps -A -o pid=,ppid=` is POSIX standard (macOS / Linux / *BSD), but BusyBox `ps` <v1.28 (2018) doesn't support `-o`. Distroless containers may not have `ps` at all. To preserve behavior on those edge platforms, the legacy per-pid `pgrep` BFS is retained as a fallback (`listDescendantPidsUnixPgrepFallback`). Same retention on Windows for the per-pid filter path. - Snapshot path uses `maxBuffer: 8MB` to cover ~250k-process pathological hosts. Default 1MB would clip at ~30k processes. - `MAX_DESCENDANTS = 256` / `MAX_DEPTH = 8` caps preserved on both snapshot + fallback paths. - Snapshot scans the entire host process table (not just the target subtree). On the typical 200-500 process developer machine this parses in <10ms; the win over BFS is real but not order-of-magnitude — ~2x improvement, not 100x. PR A's motivation framing is "fork hygiene + consistency", not raw perf. Empty-result detection: snapshot path tracks `parsedRows`. If the ps/CIM tool runs successfully but produces 0 parseable rows (BusyBox without `-o` echoing usage, AppLocker truncating CIM output, etc.), we throw — the outer catch falls back to the per-pid path. A genuine "root has no children" case parses many rows and just returns empty from the walk. So the "no-children-found" semantics are preserved across both paths. Test gate update: pre-fix `integration: spawn-and-enumerate` test skipped on `CI === '1'` because pgrep wasn't available on minimal CI runners. Post-fix `ps -A` is universally available on non-distroless Linux/macOS — only the Windows skip remains. 6/6 pid-descendants tests pass including the now-active integration spawn test. Design doc (`docs/design/f2-mcp-transport-pool.md` §6.4 + the F2 follow-up table at lines 82-85) updated to reflect the snapshot + fallback shape, and to mark W11 / W12 / R9 / R10 as ✅ Done in PR A with the per-fix commit refs. This commit completes F2 cleanup PR A. Filed bucket order: R9 (commit |
||
|
jinye
|
509ad4a5bb
|
feat(telemetry): Phase 3 — qwen-code.subagent span with concurrent isolation (#3731) (#4410) | ||
|
jinye
|
b3fa1350f7
|
feat(telemetry): Phase 4b — retry visibility for qwen-code.llm_request (#3731) (#4432)
* feat(telemetry): Phase 4b — retry visibility for qwen-code.llm_request (#3731) Adds per-attempt retry telemetry for HTTP-status retries (429/5xx) emitted by retryWithBackoff at the 4 LLM call sites. Second slice of Phase 4 (sub-issue Architectural discovery (mid-planning) -------------------------------------- The Phase 4 design doc assumed claude-code's "one LLM span owns the retry loop" pattern. Reading the 4 retryWithBackoff call sites revealed qwen-code inverts that: retryWithBackoff sits ABOVE LoggingContentGenerator. Each attempt creates a fresh LLM span. The original "in-LCG accumulator" plan wouldn't work. Resolution: propagate retry state via AsyncLocalStorage (`retryContext`). retryWithBackoff wraps each `await fn()` in `retryContext.run(...)`, and LoggingContentGenerator reads the ALS in its synchronous prelude (before the first await) and threads the snapshot into all endLLMRequestSpan callsites — success / error / idle-timeout / abort. Matches existing patterns (promptIdContext, subagentNameContext, agent-context). Plan went through 3 review rounds (Plan-agent reviews) finding 22 issues total — all addressed before implementation. Changes ------- - New retryContext.ts (AsyncLocalStorage<RetryAttemptContext>) with attempt + requestSetupMs + retryTotalDelayMs fields. Computed in retry.ts immediately before `await fn()` so values are anchored to the attempt's actual start, not derived downstream. - retry.ts: - New `onRetry?: (info: RetryAttemptInfo) => void` option on RetryOptions. Opt-in per caller: non-LLM callers stay silent. - Monotonic `iterationCount` decoupled from `attempt` (which is clamped at `maxAttempts - 1` in persistent mode). Always reflects "this is the Nth fn() call" — no flip-flopping for mixed-error sequences. - retryContext.run wrap around fn() so LCG can read the ALS. - onRetry invocations wrapped in try/catch: telemetry exceptions never break the retry loop (logged via debugLogger). - logRetryAttempt debug log line KEPT — useful when OTel SDK isn't wired up (local CLI debugging, integration tests, early-startup errors). - ApiRetryEvent telemetry event class (types.ts) with model + promptId + attempt_number + error fields + subagent_name. JSDoc cross-references ContentRetryEvent (they cover different retry budgets — HTTP-status vs invalid-stream — and can both fire for one prompt). - logApiRetry function in loggers.ts — three-sink fan-out matching logContentRetry: QwenLogger RUM, OTel log signal (bridged via LogToSpanProcessor), recordApiRetry metric counter. - recordApiRetry metric (metrics.ts) — `qwen-code.api.retry.count` Counter tagged with {model}. Full COUNTER_DEFINITIONS entry + initialization + recording function + index.ts export. - qwen-logger.ts adds logApiRetryEvent for RUM consistency. - 4 LLM caller wiring sites (client.ts, baseLlmClient.ts x2, geminiChat.ts) opt in with onRetry callback that emits ApiRetryEvent with subagentName from subagentNameContext.getStore(). - LoggingContentGenerator: snapshotRetryMetadata() helper called in the SYNCHRONOUS prelude of generateContent / generateContentStream — only point where retryContext is guaranteed active for the streaming path (the returned AsyncGenerator is iterated AFTER retryWithBackoff resolves). Snapshot threaded as parameter to loggingStreamWrapper so every endLLMRequestSpan callsite (success / error / idle-timeout / abort) sees the same values. `attempt` defaults to 1 when no retry context is present (warmup, side-queries, direct calls) so dashboards filtering WHERE attempt=1 include those. Bundled Phase 4a bug fix (sampling_ms formula) ----------------------------------------------- Phase 4a's `sampling_ms = duration_ms - ttft_ms - (requestSetupMs ?? 0)` was silently wrong. `duration_ms` only covers `ttft + sampling` for the span (startTime is captured when startLLMRequestSpan runs, AFTER any setup phase). Subtracting setup again is double-counting. Phase 4a masked the bug because requestSetupMs was always undefined → 0. Phase 4b populates requestSetupMs with cumulative retry overhead — without this fix, sampling_ms would clamp to 0 for every retried request, wiping output-throughput data exactly when operators need it most. Fix: `sampling_ms = duration_ms - ttft_ms` (drop the setup subtraction). Phase 4a tests updated accordingly: 1 test rewritten to use inputs that actually exercise the clamp under the new formula (ttft > duration = clock skew); 1 test renamed to assert the FIX (setup is NOT subtracted). Out of scope (deferred, noted in PR description) ------------------------------------------------ - Persistent retry mode emission cap (50+ events under QWEN_CODE_UNATTENDED_RETRY). Aggregated attempt/retry_total_delay_ms remain accurate regardless. - SDK-internal retries (openai/google-genai maxRetries=3) remain invisible — operator awareness only. - Stream-iteration errors (mid-stream network drop during for-await) bypass retryWithBackoff entirely. Pre-existing behavior, not a Phase 4b regression. - shouldRetryOnContent content-retry path (retry.ts:184-193) skips onRetry. No caller uses this path today — code path is dead. Tests ----- - retry.test.ts: 9 new cases (monotonic counter, requestSetupMs growth, first-try success, onRetry callback contract, absent-callback silence, callback-throws resilience, shouldRetryOnError mid-loop giveup, parallel-call ALS isolation, nested-retry inner-frame read). - loggers.test.ts: 3 new cases (3-sink fan-out, subagent_name propagation, SDK-not-initialized path). - loggingContentGenerator.test.ts: 4 new cases (non-stream ALS propagation, non-stream default attempt=1, stream ALS propagation through wrapper closure, stream default attempt=1). - session-tracing.test.ts: 1 test rewritten + 1 renamed for the sampling_ms fix. All 580 telemetry + retry + LCG tests pass. tsc --noEmit clean. eslint clean. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): address Phase 4b review comments (#4432) Fixes 6 of 9 inline review comments from wenshao + Copilot. The remaining 3 are pushback (duration_ms semantic = design intent per D5; persistent retry cap = explicitly deferred in PR description). 1. Fix JSDoc inaccuracy on `onRetry` contract (#1+#2): the comment incorrectly said "synchronous throws inside fn execute OUTSIDE the ALS frame." In fact fn() runs inside retryContext.run() so throws ARE inside the frame. What's outside the frame is the onRetry callback itself (it fires from the catch block). Rewritten per wenshao's suggestion: tells callers not to read retryContext.getStore() inside onRetry — all data comes via the RetryAttemptInfo parameter. 2. Add doc comment on content-retry delay inflation (#3): retryTotalDelayMs accumulator includes content-retry delays (shouldRetryOnContent path) which don't fire onRetry. This is intentional — the LLM span attribute reports total user-perceived backoff time — but was undocumented. 3. Add signal?.aborted guard before onRetry invocations (#6): if the abort signal fires between the catch and onRetry execution point, we now skip the callback to avoid phantom retry events that inflate the counter for retries that never actually proceeded. Applied to both persistent and normal retry paths. 4. Add persistent retry path test (status=429 + persistentMode) (#4): the highest-volume production retry path had zero Phase 4b test coverage. Now verifies onRetry fires with monotonic attempt counter and that persistent-mode exponential backoff produces increasing delayMs. 5. Add Retry-After header path test (status=429 + retry-after: 2) (#7): verifies that when the error carries a Retry-After header, onRetry.delayMs reflects the parsed header value (2000ms) instead of the exponential backoff calculation. 6. Add stream idle-timeout retry-attr propagation test (#8): verifies that the closure-captured retrySnapshot reaches the setTimeout-fired endLLMRequestSpan call with correct retry context values (attempt=4, requestSetupMs=3000, retryTotalDelayMs=2500). All 186 affected tests pass (retry 68 + LCG 48 + session-tracing 70). tsc --noEmit clean. eslint clean. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): R3 review fixes — idle-timeout test guard + prompt_id in RUM (#4432) Addresses 2 of 5 R3 review comments from wenshao (2026-05-26): 1. loggingContentGenerator.test.ts:2290 — replace `if (timeoutRecord)` guard with `expect(timeoutRecord).toBeDefined()` so the idle-timeout retry-attr test fails loudly instead of passing with 0 assertions when setTimeout doesn't fire. Also rewrote the test to use fake timers from the START (so the 5-min idle timeout is created under fake clock and can be advanced via vi.advanceTimersByTimeAsync), fixing the underlying reason it wasn't firing. 2. qwen-logger.ts:963 — add `prompt_id: event.prompt_id` to logApiRetryEvent RUM properties. Without this, RUM dashboards cannot correlate api_retry events with specific prompts, unlike the analogous logApiErrorEvent which already includes prompt_id. 165 affected tests pass. Remaining 3 R3 items (#9 onRetry helper, #10 error-path test coverage, #11 caller integration assertions) deferred to follow-up PR — non-blocking refactor/test-hardening. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) |
||
|
ChiGao
|
1285214d10
|
feat(cli): virtual viewport for long conversations on ink 7 (#4146)
* chore(deps): re-upgrade ink 6 → 7.0.3 (upstream Static remount fix landed) PR #3860 first upgraded ink 6 → 7.0.2. PR #4083 reverted because of a TUI regression: `<Static>` did not re-emit items when its `key` prop was bumped, so `/clear` / Ctrl+O / refreshStatic left the history area blank under ink 7.0.2. ink 7.0.3 (released after #4083) contains the exact fixes: - be9f44cda Fix: <Static> remount via key change drops new items (#948) - 669c4386c Fix: Drop stale <Static> output from fullStaticOutput on identity change (#950) - 7c2267c01 Fix `useBoxMetrics` not accepting ref objects with an initial null value (#945) Changes: - `ink` ^6.2.3 → ^7.0.3 (root hoist + cli direct) - `react` ^19.1.0 → ^19.2.4 (cli direct; ink 7.0.3 peerDeps requires >=19.2.0) - `react`/`react-dom` overrides ^19.2.4 added so the transitive graph stays deduped to a single instance (avoids `Invalid hook call` from multiple React copies, the classic ink-upgrade hazard) - `wrap-ansi` already on ^10.0.0 from #4083's partial-revert (no change) Verified: - `npm ls ink` → single `ink@7.0.3` across all peer deps - `npm ls react` → single `react@19.2.4` - `npm run typecheck --workspace=@qwen-code/qwen-code` clean - `npm run typecheck --workspace=@qwen-code/qwen-code-core` clean - Composer.test.tsx 20/20, MainContent.test.tsx 6/6, TableRenderer.test.tsx 59/59 + 1 skipped — all key UI components green on the new ink The Static-remount regression is upstream-fixed in 7.0.3, so the runtime path is restored without needing #3941's overflowY-self-managed viewport. #3941 (virtual viewport) remains an opt-in performance feature on top. * fix(deps,cli): add @types/react overrides + move refreshStatic out of setCurrentModel updater Two follow-ups from the multi-round audit of the ink 7.0.3 re-upgrade: 1. @types/react / @types/react-dom now pinned to ^19.2.0 in root overrides. packages/web-templates still declares @types/react ^18.2.0 in its devDeps. Today the CLI build is unaffected (web-templates's 18.x types are nested in its own node_modules and the React-using src/insight and src/export-html files are excluded from its tsconfig build), but a future reincludes-or-hoist accident would land conflicting global JSX namespaces in the CLI compile graph. Match the dep dedup we already enforce for `react` and `react-dom` so the type graph stays as deduped as the runtime graph. 2. AppContainer's onModelChange handler was calling refreshStatic() as a side-effect inside the setCurrentModel updater. React.StrictMode double-invokes state updaters in dev, so model swaps fired two clearTerminal writes + two <Static> key bumps. The double work was masked under ink 6 (key changes were no-ops on <Static>), but ink 7.0.3 honors key changes — the doubled work is now potentially visible as a faster flash-flash on every model switch. Refactor: setCurrentModel becomes a pure setter; refreshStatic moves into a useEffect keyed on currentModel with a ref-comparison guard so the first render doesn't fire. Single clearTerminal write per real model change, even under StrictMode. Verified: npm ls ink → single 7.0.3, npm ls react → single 19.2.4, npm ls @types/react → 19.2.10 hoisted (npm flags web-templates's 18.x constraint as overridden, which is the intended behavior). Typecheck clean across cli + core workspaces. * docs(design): virtual viewport on ink 7 — analysis + PR sequence Captures the architectural analysis of how to thoroughly close the flicker / refresh-storm class of issues (#2950, #3118, #3007, #3838 UI side, #3899 follow-on) using a virtualized history viewport. - Surveys claude-code (forked ink) and gemini-cli (@jrichman/ink + ScrollableList + VirtualizedList) reference implementations. - Confirms ink 7 already exposes the primitives needed (`useBoxMetrics`, `measureElement`, `useWindowSize`, `useAnimation`) — no fork swap required. - Picks porting gemini-cli's virtualized list components to ink 7 with `ResizeObserver` -> `useBoxMetrics` and a custom `StaticRender`. - Splits the work into V.0..V.4 PRs with scope, dependencies, risk. - Lists open questions + 11-item approval checklist that must clear before V.0 implementation begins. This is a docs-only PR per the project's design-first workflow. No runtime code changes. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * feat(cli): virtual viewport for long conversations on ink 7 Port gemini-cli's VirtualizedList + ScrollableList to stock ink 7, adapting for ink 7's available primitives: - `overflowY="hidden"` + `marginTop={-scrollTop}` instead of ink-fork's `overflowY="scroll"` (ink 7 has proper clip/unclip in render-node-to-output) - `useBoxMetrics` inside each VirtualizedListItem (Option A) instead of a single ResizeObserver WeakMap; reports height changes via onHeightChange callback so the parent can update its heights record - Custom `StaticRender` as `React.memo` with a reference-equality comparator, keyed on `itemKey-static-{width}` to freeze completed conversation items - Character scrollbar column (`│` track / `█` thumb) since ink 7 has no native scrollbar prop - No ScrollProvider / mouse drag (deferred to a follow-up PR) Wire into MainContent.tsx behind `ui.useTerminalBuffer` setting (Settings dialog → UI → Virtualized History; default false — opt-in). Key bindings: Shift+↑/↓ (line), PgUp/PgDn (page), Ctrl+Home/End (top/bottom). Re-render optimisations: - renderItem wrapped in useCallback so renderedItems useMemo only recomputes when actual deps change (not on every streaming tick) - Completed history items passed by original object reference so VirtualHistoryItem = memo(HistoryItemDisplay) can bail out on stable props - estimatedItemHeight / keyExtractor / isStaticItem defined as module-level constants with no closure deps Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(cli): add test coverage for virtual viewport scroll bindings and settings - keyMatchers.test.ts: 6 new test cases for SCROLL_UP/DOWN, PAGE_UP/DOWN, SCROLL_HOME/END commands (41 tests total) - settingsSchema.test.ts: assert ui.useTerminalBuffer is boolean, default false, showInDialog true, requiresRestart false Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * feat(cli): use ink 7 native overflow for VP pending items In VP mode, pending items are rendered inside VirtualizedList's overflowY="hidden" container, which uses ink 7's native clipping as the viewport guard. Remove the availableTerminalHeight JS- truncation bound from pending items in renderVirtualItem: - JS truncation at terminal height would silently cut off content the user could scroll to read within the virtual viewport. - ink 7 overflowY="hidden" on the VirtualizedList container is the correct clip guard — no JS line-counting workaround needed. - Remove uiState.constrainHeight from renderVirtualItem deps (no longer referenced in the VP rendering path). The legacy <Static> path is unchanged. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * perf(cli): binary-search offsets in virtualized list hot path Replace linear findLastIndex / findIndex scans on the offsets array with upperBound. Offsets are monotonic by construction, so the lookups inside the render body and getAnchorForScrollTop drop from O(n) to O(log n). Material for thousand-turn sessions where the lookup runs on every frame. * fix(cli): wire ShowMoreLines + skip clearTerminal in VP mode Two audit-found bugs in the VP path: 1. `<ShowMoreLines>` was outside the `<OverflowProvider>` that wraps `<ScrollableList>` in VP mode. `useOverflowState()` returns `undefined` outside the provider, so the component returned `null` and the "press ctrl-s to show more lines" affordance silently disappeared. Move `<ShowMoreLines>` inside the provider so the hook sees the live overflow state, matching the legacy path. 2. `refreshStatic()` and `repaintStaticViewport()` wrote `clearTerminal` / `cursorTo+eraseDown` to the host terminal unconditionally. In VP mode the React tree owns the visible region via ink 7's native `overflowY="hidden"` clipping — the physical write is a wasted flash on Ctrl+O / Alt+M / model change / resize. Guard both writes on `useTerminalBuffer === false`. The `historyRemountKey` bump still fires so the legacy `<Static>` fallback would still remount if someone toggled the setting mid- session. Extends the targeted-repaint pattern introduced in #3967 to all refreshStatic call sites, gated by the VP setting instead of by event type. * fix(cli): VP renderItem stability + source-copy offsets + heights GC Three audit-found regressions tightened, in order of severity: 1. **Source-copy index offsets missing in VP** — legacy `<Static>` path threads per-item `sourceCopyIndexOffsets` so `/copy mermaid N` / `/copy latex N` hints stay stable across continuation messages. VP `renderVirtualItem` was not passing this prop, so the copy hints shown under each diagram drifted on every `gemini_content` chunk (the clipboard mechanism itself still worked from raw history; only the displayed number was wrong). Add two lookup tables — identity-keyed for static items, index-keyed for pending — without changing the VirtualizedList data signature, and thread offsets in both render branches. 2. **`renderVirtualItem` callback invalidated on every streaming tick** — its deps included `activePtyId` / `embeddedShellFocused` / `isEditorDialogOpen`, all of which flip mid-stream when a shell tool runs or a dialog opens. Each flip rebuilt the callback, invalidated `VirtualizedList.renderedItems`'s useMemo, and forced every static item to re-render through `<StaticRender>` — defeating the very memoization the design relies on. Move the three pending- only fields into a ref read inside the callback. Static-item closure now depends only on inputs that legitimately affect static output (terminalWidth, slashCommands, getCompactLabel, …). Pending items still re-render correctly because their item identity changes per tick, so the callback is called fresh each time and reads the latest ref. 3. **`pending` items now honour `constrainHeight`** in VP, matching the legacy path. Previously VP unconditionally passed `undefined` for `availableTerminalHeight` on pending, relying on the viewport `overflowY="hidden"` clip to limit visible size — but that hid the `<ShowMoreLines>` affordance from the user. Now that ShowMoreLines is correctly wired (previous commit), restore parity. 4. **Heights map memory leak** in `VirtualizedList` — `setHeights` only grew. Each `/clear` left orphan `h-N` keys; each pending → completed transition left orphan `p-N` keys. Add a `useLayoutEffect` that prunes entries whose keys are not in the current `data`. Runs in layout phase so the prune commits in the same paint as the data change — no stale-offsets frame. * test+fix(cli): VP path coverage + stabilize absorbedCallIds empty Set Completion-pass artifacts driven by the multi-agent audit: - Settings description rewritten to enumerate the symptoms VP fixes so users with active flicker reports can find the toggle without reading the design doc. - `absorbedCallIds` returns a module-level constant Set when compact mode is off, instead of a fresh `new Set()` per render. Fixes a hidden cascade: `activePtyId` flip mid-stream → useMemo runs → returns a new empty Set → `isSummaryAbsorbed` rebuilds → `renderVirtualItem` rebuilds → `VirtualizedList.renderedItems` recomputes → every static item re-renders. With the constant, the cascade dies at the source. Helps both VP and legacy paths. - VP-path unit tests for MainContent (4 cases): ScrollableList mounts and Static does not when `useTerminalBuffer: true`; ShowMoreLines is reachable in VP mode (regression of the OverflowProvider mis-wrap); source-copy index offsets thread into renderItem for static items; renderItem callback identity is stable across `activePtyId` flips (proves the ref-based read keeps StaticRender memo effective). * fix(cli): stabilize absorbedCallIds in compact mode + gate heights prune + tighten ShowMoreLines test Round-2 audit follow-ups. Three real findings addressed; one flagged false positive documented separately. 1. **absorbedCallIds Set identity now content-stable when compact mode is on.** The earlier EMPTY constant only short-circuited the compactMode= false path; when compact mode is enabled (some users default-on it), activePtyId / embeddedShellFocused flips during streaming still produced fresh Sets per render even when membership was unchanged, restarting the same cascade the pendingStateRef fix was meant to avoid. Compare-and-reuse via a ref: if the new Set has identical membership to the previous one, return the previous reference. 2. **`heights` map prune in `VirtualizedList` is gated.** Previously every streaming tick rebuilt an N-key Set and walked all heights, even on the steady-state path where nothing changes. Now only fires when the heights record has clearly outpaced live data (`size > max(8, 2 × data.length)`) — covers `/clear` and accumulated pending → completed transitions, skips the 30-Hz hot path entirely. 3. **VP ShowMoreLines test now actually verifies overflow connectivity.** Previous mock unconditionally rendered "SHOW_MORE", so the test only proved the JSX mounted — it would still pass if a future refactor moved `<OverflowProvider>` out of the VP tree again. The mock now reads `useOverflowState()` and emits "OVERFLOW_DISCONNECTED" when the context is missing. The VP test asserts both presence of "SHOW_MORE" and absence of the disconnected marker, so the regression is now caught. Not addressed: - Audit P0-1 claim that `renderMode` (Alt+M) / model-change updates don't reach VP static items: false positive. `renderMode` is a React Context (`RenderModeContext`), and Context propagation traverses the tree past `memo` boundaries — MarkdownDisplay's `useRenderMode()` consumer re-renders on context change regardless of whether `StaticRender` bails out. Verified by reading `packages/cli/src/ui/contexts/RenderModeContext.tsx` and `MarkdownDisplay.tsx:172`. No code change. - Audit P1-2 pendingStateRef write-during-render race: speculative, relies on a multi-pass render path React 18+ does not currently use. Documented assumption in the existing inline comment. * fix(cli): isolate renderItem errors + defensive height coerce + compact-mode mergedHistory stability Round-3 audit follow-ups. Three real findings; the rest verified clean. 1. **`renderItem` errors no longer crash the CLI.** Previously a throw inside a per-item render propagated through `VirtualizedList`'s useMemo into React's commit phase, tearing down the whole Ink tree — one bad history record could nuke the session. Wrap each call in a try/catch and substitute a small red `[render error] …` text box on failure. The row stays in the viewport so the user can scroll past it. 2. **Defensive height coerce in offset accumulation.** A buggy `estimatedItemHeight` returning NaN / negative / Infinity would poison every downstream offset and break the `upperBound` / `findLastLE` binary search (which assumes monotonic offsets). Clamp to `Number.isFinite(raw) && raw > 0 ? raw : 0`. No-op for the in-tree estimators that return 3; insurance against future consumers. 3. **`mergedHistory` is content-stable when compact mode is on.** The Round-2 absorbedCallIds stability fix didn't reach this path: `mergeCompactToolGroups` always allocates a fresh array, and `mergedHistory`'s useMemo lists `activePtyId` / `embeddedShellFocused` as deps, so every streaming tick mid-shell-tool produced a new array even when items aligned. Cascade went `mergedHistory` → offsets map → `renderVirtualItem` → every static item re-rendered. Pair-wise compare new vs previous and return the previous reference when items align. Restores StaticRender memo effectiveness for compact-mode users. Not addressed (audit findings deemed not worth fixing in this PR): - `scrollToItem` silently no-ops when item is not in data — no current caller checks the return value, low impact. - `allVirtualItems` array spread is O(n) per streaming tick — real but not a crash; revisit in a perf-focused follow-up. - `itemRefs.current` is dead surface (never read) — cosmetic. - StrictMode-only-in-DEBUG double-invoke paths verified safe. * test+chore(cli): VP review round 4 — VirtualizedList/useBatchedScroll coverage + cleanups Addresses wenshao's CHANGES_REQUESTED review on PR #3941. - Add focused unit tests for `VirtualizedList` (9 cases) covering empty data, `renderStatic` full-render, `initialScrollIndex` with `SCROLL_TO_ITEM_END`, `targetScrollIndex` anchoring, imperative `scrollToEnd` / `scrollToIndex`, per-item `renderItem` error isolation, NaN/negative estimator coercion, and out-of-range `initialScrollIndex` clamping. - Add `useBatchedScroll` unit tests (4 cases) covering initial reads, pending-value reads in the same tick, post-commit pending reset, and callback identity stability across rerenders. - Remove dead `itemRefs` / `onSetRef` plumbing (declared, written, never read; `useCallback` with empty deps was also a stale-closure trap). - Remove unused `isStatic?: boolean` from `VirtualizedListProps` (only `isStaticItem` is actually consumed). - Tighten the render-phase setState block: each setter is now guarded by an equality check so React bails out of redundant updates, and a comment documents that this is the React-endorsed "adjusting state while rendering" pattern (the synchronous update avoids a one-frame flash at the previous position when `targetScrollIndex` changes). Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * chore(cli): remove dead `dataRef` from VirtualizedList (round-4 followup) Declared and written in a `useLayoutEffect` on every `data` change but never read anywhere in the component. Flagged in wenshao's round-4 review of PR #3941. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(cli): collapse model-change effect back into one batched handler wenshao's PR #4119 review correctly flagged that splitting the onModelChange flow into two effects ( |
||
gwinthis
|
7bed56b9b6
|
feat(telemetry): foundation for skill-based RT optimization (P0+P1) (#4565)
Some checks are pending
Qwen Code CI / Classify PR (push) Waiting to run
Qwen Code CI / Lint (push) Blocked by required conditions
Qwen Code CI / Test (macos-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (ubuntu-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (windows-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Post Coverage Comment (push) Blocked by required conditions
Qwen Code CI / CodeQL (push) Blocked by required conditions
E2E Tests / E2E Test (Linux) - sandbox:docker (push) Waiting to run
E2E Tests / E2E Test (Linux) - sandbox:none (push) Waiting to run
E2E Tests / E2E Test - macOS (push) Waiting to run
* docs(design): add RT optimization design doc
Two-round review trail documenting the analysis path: original D1-D4
proposal, code-level verification in §6 that recanted the cost estimates,
and §7 ROI reordering after DashScope ephemeral cache implementation was
confirmed already in place — which collapsed D2's net benefit and led to
deferring D2 and D4 as won't-fix.
The doc is preserved as the canonical record of why the obvious-looking
directions (fast-model routing, prevalidate scheduling) turn out to be
dead ends, so future work doesn't relitigate the same conclusions.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* docs(design): add reduce-rounds-via-skill-design with spec-first gating
Companion design to rt-optimization-design.md. The core argument:
the real lever for reducing agent loop rounds is at the skill/tool
design layer, not the agent framework. Round 2 in §1.2's baseline
exists because Round 1's skill didn't return a complete answer —
fixing that per-skill collapses 3 rounds into 2, an angle the
original framework-centric proposal completely missed.
Layout:
- §0 acceptance spec is the front-loaded gate: engineering specs
lock at P-1, statistical thresholds lock at P1.5 (after baseline),
per-skill specs are data-driven and live in PR descriptions
- §3-§4 three-layer plan: telemetry → per-skill rewrites → prompt
guidance for concurrent tool calls; each layer is independently
measurable and reversible
- §5.3 stop-loss lines split into result + process metrics to catch
the "looks like progress, no actual ROI" failure mode early
The doc was reviewed by codex twice — once on initial draft (caught
qwen-logger dead-code path, batch_size state-passing cost, prompts.ts
line drift) and once after §0 was added (caught spec rigidity, missing
per-skill template, framework boundary case). Both rounds' findings
were either applied or explicitly recorded as not-adopted with reasons
inline.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(telemetry): connect logSkillLaunch to QwenLogger
logSkillLaunch in loggers.ts went only through the OTLP path, while
QwenLogger.logSkillLaunchEvent in qwen-logger.ts had no callers anywhere
in the repo — leaving the skill_launch event invisible to any backend
that consumes from the qwen-logger pipeline rather than OTLP.
Mirror the logToolCall pattern at loggers.ts:230: forward the event to
QwenLogger before the OTLP path so the call still reaches QwenLogger when
the OTEL SDK is not initialized.
This is P0 of docs/design/rt-optimization/reduce-rounds-via-skill-design.md
§4.1.1b — a prerequisite for the prompt_id propagation in P1 so the
SkillLaunchEvent / ToolCallEvent join in §4.1.2 has data to query against.
Tests: 2 new cases under describe('logSkillLaunch') covering forwarding
to QwenLogger plus the OTLP-uninitialized branch; loggers.test.ts now
47/47 pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* feat(telemetry): thread prompt_id through SkillLaunchEvent
To join skill_launch events with the subsequent tool_call events they
trigger, SkillLaunchEvent now carries the prompt_id of the user turn
that fired the skill. The scheduler already holds the request and its
prompt_id; the missing piece was getting that id into the invocation
that does the actual logSkillLaunch call.
Wiring:
- SkillLaunchEvent constructor adds a required prompt_id parameter so
the field can never be silently undefined in a backend join.
- SkillToolInvocation exposes setPromptId(id) and stores the value;
the four logSkillLaunch sites in execute() pass this.promptId through.
- CoreToolScheduler.buildInvocation grew an optional fourth promptId
argument and duck-types setPromptId on the freshly-built invocation,
mirroring the existing setCallId hook. The two callers (setArgs path
at L1036 and the main schedule path at L1497) pass
request.prompt_id / reqInfo.prompt_id.
- qwen-logger.logSkillLaunchEvent forwards prompt_id in the RUM event
properties so the join works on the qwen-logger pipeline too.
The empty-string default on SkillToolInvocation.promptId is deliberate:
direct invocations (e.g. buildAndExecute in tests) that skip the
scheduler still log a valid event, and downstream queries can filter
prompt_id != '' to exclude non-scheduled launches from joins.
Implements P1 of docs/design/rt-optimization/reduce-rounds-via-skill-design.md
§4.1.1 — required prerequisite for the SkillFollowupRecord SQL in §4.1.2.
Tests: 2 new cases in skill.test.ts cover the setPromptId path and the
empty-default path; loggers.test.ts updated for the new 3-arg signature.
256 tests pass across loggers / skill / coreToolScheduler suites.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* test(scheduler): cover prompt_id propagation through buildInvocation
The duck-typed setPromptId hook added in the previous commit went
through unit tests on each side independently — SkillToolInvocation
tests verified that setting the field changes the logged event, and
the loggers tests verified the SkillLaunchEvent shape — but the
integration point in CoreToolScheduler.buildInvocation that wires the
two together was only exercised indirectly. Same is true of the older
setCallId hook it mirrors, which had no test at all.
Two cases here close that gap on the scheduler side:
- A purpose-built PromptIdAwareTool whose invocation records every
setPromptId call; the test schedules a request with a known
prompt_id and asserts the invocation captured it. This is the
positive contract.
- The existing TestApprovalTool (no setPromptId) scheduled through
the same path to confirm the duck-type guard does not throw when
the method is absent. This is the backward-compatibility contract
that lets every existing tool keep working unchanged.
The two cases together pin both branches of the typeof check in
buildInvocation, so future refactors of that hook cannot regress
silently. 165 tests in the suite still pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* test(telemetry,scheduler,skill): close P0/P1 coverage gaps
Three blind spots remained after the initial P0+P1 work — each one
was a path that the production code change had already touched
mechanically but no test was pinning it against future regressions.
qwen-logger.ts logSkillLaunchEvent now has two cases asserting that
prompt_id reaches the RUM event properties, on both the success and
failure branch. Previously the loggers.test.ts spy stopped at "method
was called" and never inspected the payload qwen-logger built.
skill.ts had four logSkillLaunch sites, but only the happy path and
the empty-default path were tested. The commandExecutor-success
branch (L386), not-found branch (L399), and thrown-exception branch
(L482) now each have a test that sets promptId, drives execute()
through that specific path, and asserts the emitted event carries
both the right success flag and the right prompt_id. This catches
the failure mode where someone later edits one of those branches
and forgets the promptId argument — replace_all guaranteed today's
correctness but no test would catch a regression tomorrow.
CoreToolScheduler.buildInvocation now has two direct unit tests
that exercise the method through a type-assertion cast. Reaching
the L1036 setArgs path through the public API would require mocking
modifyWithEditor + the filesystem + an editor type, which would
dwarf the change under test. The direct call covers both L1036 and
L1497 simultaneously: when promptId is supplied the duck-typed
setPromptId is invoked; when it is omitted, the captured field
stays undefined and no throw happens.
298 tests pass across loggers / qwen-logger / skill / scheduler suites.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* review(PR #4565): address Copilot + github-actions feedback
Five spots flagged by the automated review on #4565. Three were real
and worth fixing; two were comment-quality touch-ups that traveled
along with the same patch.
- SkillLaunchEvent.prompt_id is now optional with a default of '',
removing the breaking-change footprint on the exported telemetry API.
All current internal callers still pass the value explicitly through
the SkillToolInvocation.promptId field, so the §0.1 spec ("prompt_id
串联") is still enforced in production paths — type-level enforcement
just steps aside in favor of API stability, with §0.5 治理 covering
the discipline at the process layer.
- The skill-design doc §4.1.1 used to claim "BaseToolInvocation 已有
request.prompt_id" which is wrong: BaseToolInvocation only holds
params, and the prompt_id flows through CoreToolScheduler's duck-typed
setPromptId hook (mirroring setCallId). The doc now reflects the
actual implementation and notes that the earlier text was the bug.
- CoreToolScheduler.buildInvocation gained a short JSDoc explaining
why the two extra args (callId, promptId) are optional — they
match the existing duck-type pattern that lets older tools and
non-scheduler call sites work without implementing the setters.
- skill.test.ts adds a one-comment note next to the first setPromptId
cast explaining that setPromptId is a scheduler-only hook, not part
of the public ToolInvocation interface.
- SkillToolInvocation.promptId field comment shrank from 8 lines to 2
with a pointer to the design doc so the inline noise drops without
losing the empty-string semantics.
Pre-existing scope-creep findings (Chinese-only doc, mock-config
duplication in scheduler tests, redundant optional-chain comment,
prompt_id sanitization for an internally-generated UUID) are
deliberately not addressed here — see the reply on PR #4565 for
disposition per item.
298 tests still pass across loggers / qwen-logger / skill / scheduler.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
顾盼
|
5ad5301805
|
feat(worktree): Phase D — startup --worktree flag + symlinkDirectories + PR refs (#4381)
* feat(worktree): Phase D — startup --worktree flag + symlinkDirectories + PR refs
Three cross-cutting capabilities on top of the Phase A-C worktree
foundation (PRs #4073, #4174).
D-1: --worktree [name] CLI flag creates a worktree (or re-attaches to
one that already exists) before any model turn runs. Supports bare,
plain-slug, `=`, and PR-reference forms; --worktree + --acp rejected
with a clear error; --worktree + --resume overrides the resumed
session's saved sidecar and emits a stderr line.
D-2: worktree.symlinkDirectories: string[] settings key opts into
symlinking main-repo directories (e.g. node_modules) into every
newly-created general-purpose worktree. Applies to all three creation
paths: --worktree flag, EnterWorktreeTool, AgentTool isolation. Path
traversal, absolute paths, and existing destinations all guarded;
missing source dirs and EEXIST silently skipped (fail-open).
D-3: --worktree=#<N> / --worktree <github-url> resolves a PR number,
runs `git fetch origin pull/<N>/head` (30s timeout, no `gh` CLI
dependency, LANG=C for stable error-taxonomy matching), and creates
the worktree off FETCH_HEAD. URL regex tolerates /files, /commits,
/checks sub-paths so users can paste any GitHub PR URL.
Phase 6 verification fixes also included:
- Re-attach to an existing worktree instead of failing with "Worktree
already exists" — the common `qwen --resume <sid> --worktree foo`
workflow now succeeds. The session ownership marker is preserved on
re-attach so cross-session exit_worktree action="remove" still fails
for non-owners.
- Normalize path-taking argv fields (mcpConfig, jsonSchema @<path>,
openaiLoggingDir, jsonFile, inputFile, telemetryOutfile,
includeDirectories) to absolute paths against the launch cwd BEFORE
the worktree chdir. Otherwise downstream fs.existsSync('./mcp.json')
resolves into the worktree, where the file doesn't exist.
Phase 7 code-review fixes:
- buildStartupWorktreeNotice differentiates "Active worktree" (fresh
create) from "Re-attached to worktree" (re-attach path).
- Notice survives sidecar persist failure: set before the try block,
refreshed inside with override addendum if persist succeeded.
- getRegisteredWorktreeBranch verifies the candidate path's git
common-dir matches the source repo's — rejects sibling `git init`
directories that happen to be on a worktree-<slug> branch.
Three-mode parity for the startup notice: TUI consumes via
AppContainer effect, headless prepends a <system-reminder> + emits a
worktree_started JSON event. ACP path is mutually exclusive with
--worktree (ACP hosts supply per-session cwd separately).
Tests (66 + 15 new):
- 15 cli/src/startup/worktreeStartup.test.ts (slug forms, PR fetch
against local fake remote, re-attach happy + wrong-branch guard)
- 8 core/src/services/gitWorktreeService.test.ts (parsePRReference:
#N, URLs, malformed, traversal, leading zeros, non-string)
- 10 core/src/services/gitWorktreeService.symlinks.integ.test.ts
(symlink loop + fetchPullRequestRef error taxonomy)
Known limitations (documented in docs/users/features/worktree.md):
- Cross-slug --resume <sid> --worktree <different-new-slug> is
unsupported by design (sessions are bound to projectHash(cwd));
future Config refactor anchoring storage at repo root would lift this.
- Mid-session enter_worktree still does NOT switch cwd/targetDir
(Phase A's simplification); only the startup --worktree flag does.
- yargs ambiguity: `qwen --worktree "say hi"` consumes the prompt as
the slug. Quick Start shows the `=` form and reordering workarounds.
Docs:
- docs/users/features/worktree.md (new): Quick Start with --worktree
flag, CLI Reference table for all four input forms + error codes,
settings table, Limitations.
- docs/design/worktree.md: Phase D section expanded into D-1/D-2/D-3
with open questions resolved; capability table updated.
- docs/e2e-tests/worktree-phase-d.md (new): full E2E plan with Phase 4
dry-run baseline + Phase 6 post-impl reproduction tables.
Refs #4056
* refactor(worktree): apply self-review feedback on Phase D
Self-review pass over the Phase D commit (2636f59273) catching one real
typecheck regression plus a batch of small quality + efficiency
improvements. No user-visible behavior change beyond fixing the build.
Build fix:
- worktreeStartup.ts imports — pre-commit prettier had reorganized
`writeWorktreeSession` and `readWorktreeSession` under an
`import type { ... }` block, erasing them at compile time
(verbatimModuleSyntax). `tsc --noEmit` was failing with TS1361.
Bundle path still worked (esbuild is lenient) so this only surfaced
when running typecheck.
Startup-path efficiency (~10-25 ms saved per --worktree invocation on
macOS; more on Windows):
- Drop redundant `isGitRepository()` probe — `getRepoTopLevel()`
returns null on non-git paths and covers both gates in one
subprocess.
- Run `getCurrentBranch()` + `getCurrentCommitHash()` in parallel via
Promise.all (independent calls).
- Combine the two `git rev-parse` probes inside
`getRegisteredWorktreeBranch` into a single multi-arg call, and run
it in parallel with the source-repo common-dir lookup. Saves one
fork+exec on the re-attach path.
Quality:
- Extract `withReminder()` local helper in nonInteractiveCli.ts so the
startup-notice and resume-restore branches share the system-reminder
wrapping.
- Log `readWorktreeSession` failures in `persistStartupWorktreeSidecar`
with the sidecar path so operators can recover the previous slug
from a backup. Silent swallow was making "where did my worktree
binding go?" undebuggable.
- Drop the dead `Config.getWorktreeSettings()` accessor (only
`getWorktreeSymlinkDirectories()` has callers); keep the underlying
`WorktreeSettings` interface for future fields.
- Document the `pendingStartupWorktreeNotice` invariant: at most one
consumer per process; ACP path is gated out earlier so only TUI XOR
headless reads it.
- Add a maintainer note in the gemini.tsx path-normalization block:
the argv path-field allowlist is hand-maintained, register new
path-bearing flags there or `--worktree` silently breaks for them.
- Drop `Phase 6 fix (G1)/(G2)` parenthetical labels from inline
comments — internal review-cycle identifiers that decay to noise
post-merge. Substantive prose retained.
Tests: cli 15/15 (unchanged) + core 66/66 (unchanged); bundle smoke
verified fresh / re-attach / invalid slug / non-git cases.
Findings deliberately left for follow-up:
- Larger refactor extracting a shared `provisionUserWorktree` helper
for the EnterWorktreeTool / startup overlap (~80% duplicate).
- Splitting the re-attach branch out of `setupStartupWorktree` into
its own function.
- `isPathWithinRoot` / `isInsideManagedWorktree` shared utils.
- `symlinkConfiguredDirectories` loop concurrency (saves 5-15 ms on a
cold path that runs only when symlinkDirectories is configured).
* docs(worktree): refresh stale docstring in worktreeStartup
Top-of-file docstring still said `{adj}-{noun}-{4hex}` (actual format
is 6 hex chars) and described the PR form as "detected and rejected
with a clear 'coming in D-3' message" — but D-3 shipped in the same
PR. Tighten to reflect what the code actually does.
* fix(worktree): address findings from dual-reviewer self-check
Two real bugs surfaced by an independent dual-reviewer pass (Claude +
Codex) on the Phase D commits. Both correctness-affecting; both
escaped the earlier internal reviews.
P0 — re-attach captured the wrong baseline for the exit dialog
(Codex):
setupStartupWorktree captured `originalHeadCommit` from the launch
cwd (main checkout) before any chdir. On the re-attach path the
WorktreeExitDialog later runs `git rev-list <originalHeadCommit>..HEAD`
inside the worktree to count "new commits this session". With the
main-checkout baseline this counted every commit ever made in the
kept worktree as new work from the current session — misleading the
keep/remove prompt. Re-capture HEAD from inside the worktree after
chdir so the count means what the dialog text says it means.
P0 — getRegisteredWorktreeBranch mis-identified plain directories as
registered worktrees (Claude):
A plain directory at `<repo>/.qwen/worktrees/<slug>/` (e.g. a stale
artifact from a previous tool) had no `.git` file of its own, so
`git rev-parse --git-common-dir` walked up to the outer repo and
returned the outer common-dir — matching the source repo's
common-dir check and impersonating a registered worktree. If the
outer repo happened to be on `worktree-<slug>`, setupStartupWorktree
would silently chdir into the plain directory and treat it as
attached; subsequent `exit_worktree action="remove"` would then
delete a directory that was never registered.
Fix: also probe `--show-toplevel` and require it to equal the
candidate path (canonicalised via `realpath` so macOS /var → /private/var
doesn't break the equality check). A plain dir under the main repo
gets the outer repo's toplevel and is correctly rejected.
Smaller polish from the same review:
- Normalize the literal string `'HEAD'` returned by `getCurrentBranch`
on detached HEAD to `undefined`, so the `baseRef` handed to
`git worktree add -b … HEAD` does not implicitly anchor against
the loose commit when the launch cwd is detached.
- `symlinkConfiguredDirectories`: blocklist `.git` (any nested
ancestor) and `.qwen/worktrees` (any nested ancestor). Linking
`.git` would silently break commits inside the worktree; linking
`.qwen/worktrees` would create a worktrees-inside-worktrees loop
that confuses the startup sweep.
- `WorktreeSettings.symlinkDirectories` typed `readonly string[]` to
match the `createUserWorktree(options.symlinkDirectories)` contract
and the immutable-config convention elsewhere. `Config.getWorktreeSymlinkDirectories()`
return type updated to match.
Docs:
- design/worktree.md precedence table rewritten. The previous
`--worktree` 赢 row was unreachable in practice (sessions are bound
to `projectHash(cwd)`, and the chdir happens before session lookup).
New table reflects what actually happens for each combination of
`--resume` × `--worktree`, including the documented
cross-projectHash limitation. The `persistStartupWorktreeSidecar`
override branch is now annotated as dead-on-the-current-architecture
but kept so a future Config refactor (anchor storage at repo root)
picks it up for free.
Tests: cli 15/15 + core 66/66 unchanged. Bundle smoke confirms both
P0 fixes end-to-end (re-attach captures worktree HEAD = run-1 tip,
plain-dir attempt errors out without clobbering existing content).
* refactor(worktree): consolidate probe + name detached-HEAD sentinel
Second /simplify pass on the dual-reviewer fixes. Three convergent
findings; net effect is one fewer subprocess on the re-attach path
and clearer intent on string handling / blocklist guards.
Efficiency + quality:
- Fold the worktree HEAD SHA into `getRegisteredWorktreeBranch`'s
combined rev-parse. The probe already requests common-dir,
toplevel, and abbrev-ref HEAD in a single subprocess; adding a
leading `HEAD` positional (which must come BEFORE `--abbrev-ref` so
the flag doesn't apply to it) returns the SHA on its own line.
Return type widened to `{ branch, headCommit } | null`. Removes
the second `GitWorktreeService` instantiation and `getCurrentCommitHash`
call that `setupStartupWorktree`'s re-attach branch used to do.
Quality:
- Hoist `'HEAD'` to a module-level `DETACHED_HEAD` constant in
`worktreeStartup.ts`. Three uses, two meanings (input filter when
normalizing `getCurrentBranch` output, fallback metadata for the
sidecar's `originalBranch` field on detached state). Naming the
sentinel makes intent self-documenting and pre-empts the "why is
the value we just stripped re-appearing as a fallback?" reader stall
flagged by the round-3 quality review.
Reuse + quality:
- `symlinkConfiguredDirectories`: replace two hand-rolled containment
checks (`startsWith(prefix + sep)` for `.qwen/worktrees`; `path.relative(...).split(sep)[0]`
for `.git`) with `isWithinRoot` from `utils/fileUtils.ts`, which is
already imported in this file. Replace the hardcoded
`path.join(repoRootAbs, '.qwen', 'worktrees')` with `this.getUserWorktreesDir()`
so the layout lives in one place (the exported `WORKTREES_DIR`
constant). Split the misleading `sourceAbs === repoRootAbs` clause
out of the `.git` branch into its own dedicated "empty / repo-root
path" rejection with a clearer warn message.
Tests: cli 15/15 + core 66/66 unchanged. Bundle smoke verified the
folded probe still captures the worktree's HEAD on re-attach (not
the launch-cwd HEAD).
Skipped from this review pass:
- Moving `'HEAD'` normalization into `GitWorktreeService.getCurrentBranch()`
itself — would ripple through `enter-worktree.ts` and `agent.ts`
callers that hand the result verbatim to `git worktree add -b ...`.
Out of scope for a polish pass; the local const is enough.
* fix(worktree): broaden symlink blocklist from .qwen/worktrees to all of .qwen
Caught by a second pr-tracker dual-reviewer pass (Codex). The previous
guard at `symlinkConfiguredDirectories` only refused paths inside
`<repoRoot>/.qwen/worktrees/` — `.qwen` itself (the parent) sailed
through because `isWithinRoot` is a strict descendant check. A user
setting `symlinkDirectories: ['.qwen']` would therefore symlink the
entire CLI metadata tree into the new worktree, recursively pulling
in `.qwen/worktrees` and recreating the loop the guard was meant to
prevent. Other `.qwen/*` subtrees (`projects`, `tmp`, …) are CLI
state with no legitimate cross-worktree sharing use case either.
Fix: broaden the guard to reject the whole `<repoRoot>/.qwen` tree.
Both `.qwen` itself and any descendant fail closed.
Also synced the user-facing settings schema description (the in-IDE
help text and the published JSON schema) so it mentions the `.git`
and `.qwen` rejection rules. The `WorktreeSettings` interface JSDoc
already mentioned them; the schema description had not been updated.
Tests: cli 15/15 + core 66/66 unchanged. Smoke confirms `--worktree foo`
with `symlinkDirectories: ['.qwen']` configured leaves the worktree
free of any `.qwen` symlink (only the legitimate per-worktree
`.qwen-session` marker file appears).
* fix(worktree): guard fetchPullRequestRef against CodeQL command-injection alert
CodeQL flagged a "Second order command injection" finding (rule 235) on
the `git fetch origin pull/<N>/head` call in `fetchPullRequestRef`. The
taint analyzer doesn't see the type-narrowing at the function entry
(`Number.isSafeInteger(prNumber) && prNumber > 0 && prNumber <= 1e9`),
so it considers `prNumber` library input that could in principle reach
a `--upload-pack=…`-shaped flag and thereby execute an arbitrary
program. In practice the entry guard already prevents that, but the
alert blocks the CodeQL CI check.
Add `--end-of-options` between `origin` and the refspec — git's
canonical "stop parsing flags" marker (git ≥ 2.24). Tells git
definitively that every subsequent argv element is a positional, not
a flag, which (a) satisfies the analyzer, (b) adds defense-in-depth
against a future regression that might relax the entry guard, and
(c) has zero behavior change for any well-formed PR number.
Verified locally: `git fetch --end-of-options origin pull/<N>/head`
against a local bare-remote with a seeded `refs/pull/42/head` still
fetches the ref correctly; the `--worktree=#42` smoke test reads back
the PR content from the materialized worktree.
Tests: cli 15/15 + core 66/66 unchanged.
* fix(worktree): lexical sanitizer for CodeQL + missing test mock entry
Two fixes from the third CI round on PR #4381:
1. CodeQL re-fires (round 2 of the same finding).
`--end-of-options` is a git-runtime defense, not a lexical sanitizer
that CodeQL's `js/second-order-command-line-injection` taint tracker
recognises. The alert re-fired against the same call after the
previous fix.
Switch to a CodeQL-recognised sanitizer: validate the numeric
component against `/^[1-9][0-9]*$/` immediately at the sink. The
regex digit-only check is one of the documented sanitizer patterns
the rule looks for, and proves at the analyzer level that the
resulting argv element cannot resemble a flag (`--foo`). The entry
guard at the top of the function still establishes the same fact
at runtime; this layer makes the proof visible to static analysis.
Keep `--end-of-options` as a runtime fallback against any future
regression that loosens the entry guard.
2. `nonInteractiveCli.test.ts` mock was missing the new
`consumePendingStartupWorktreeNotice` Config method.
Phase D-1 added the method on `Config` and `nonInteractiveCli`
calls it on every prompt to pick up the one-shot startup-worktree
notice. The test file's `mockConfig` literal was not updated, so
all 19 `runNonInteractive` tests threw
`TypeError: config.consumePendingStartupWorktreeNotice is not a
function` on Ubuntu / macOS CI.
Add a stub returning `null` so the helper short-circuits, matching
the equivalent Phase C stub for `getResumedSessionData`.
Local: cli (worktreeStartup + nonInteractiveCli) 60 passed + 1
skipped; core (gitWorktreeService + symlinks + hooks +
enter-worktree) 66 passed.
* test(worktree): mock getWorktreeSymlinkDirectories in three more test files
Round 4 of the same Phase D-2 mock-drift class. CI surfaced 9 test
failures across three files whose `Config` mocks construct
`EnterWorktreeTool` for setup but lack the new
`getWorktreeSymlinkDirectories` method `createUserWorktree` now
calls:
- enter-worktree.session.integ.test.ts (2 tests)
- exit-worktree.session.integ.test.ts (3 tests) — provisions
worktrees via EnterWorktreeTool before exercising exit paths
- exit-worktree.test.ts (4 tests) — same provisioning pattern via
`provisionWorktree()` and the `makeMockConfig` helper
Add a `getWorktreeSymlinkDirectories: () => []` stub to each so
the symlink loop is a no-op in tests.
`enter-worktree.test.ts` and `agent/agent.test.ts` intentionally
skipped — they mock `GitWorktreeService.createUserWorktree` outright,
so the method call never fires in their code paths. Adding the stub
there would be defensive speculation. If a future test exercises
the real path, it'll surface there too and we'll add it then.
Local: core tools tests now 123 passed (was 9 failed / 114 passed
on CI run 26213122427 against commit
|
||
|
jinye
|
62ed44e1f3
|
feat(telemetry): client-side HTTP span + opt-in W3C traceparent propagation (#4384) (#4390)
Some checks are pending
Qwen Code CI / Classify PR (push) Waiting to run
Qwen Code CI / Lint (push) Blocked by required conditions
Qwen Code CI / Test (macos-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (ubuntu-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Test (windows-latest, Node 22.x) (push) Blocked by required conditions
Qwen Code CI / Post Coverage Comment (push) Blocked by required conditions
Qwen Code CI / CodeQL (push) Blocked by required conditions
E2E Tests / E2E Test (Linux) - sandbox:docker (push) Waiting to run
E2E Tests / E2E Test (Linux) - sandbox:none (push) Waiting to run
E2E Tests / E2E Test - macOS (push) Waiting to run
* feat(telemetry): propagate W3C traceparent on outbound LLM requests Part 1 of #4384 (sub-issue of #3731 P3 deeper observability). Today qwen-code's only OTel instrumentation is `HttpInstrumentation`, which only patches Node's `http`/`https` modules. The `openai` and `@google/genai` SDKs use `globalThis.fetch` (undici), so outbound LLM requests carry no `traceparent` header and trace context dies at the qwen-code process boundary. Adds `@opentelemetry/instrumentation-undici@0.14.0` (peer-compatible with the installed `@opentelemetry/instrumentation@0.203.0`) and wires it into `initializeTelemetry()` next to the existing `HttpInstrumentation`. Default propagator (W3C tracecontext + baggage) remains unchanged — no explicit `textMapPropagator` needed. `ignoreRequestHook` skips OTLP exporter endpoints to avoid the classic feedback loop (OTel SDK uses fetch to upload OTLP data; without the hook each upload would create a span that gets uploaded, infinitely). Configured `otlpEndpoint` / per-signal endpoints are stripped of trailing slash and query string for robust prefix matching against undici's `request.origin + request.path`. Outbound LLM calls now also produce a client-side HTTP span (separating network TTFB / transfer time from the existing `api.generateContent` total-duration span). Design doc: docs/design/telemetry-outbound-propagation-design.md (Part A — traceparent; Part B — session id header — lands in a follow-up PR per the design's split rationale.) 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): harden OTLP feedback-loop guard + slim lockfile diff Review feedback on #4390: 1. CI was failing on npm ci because the lockfile was generated with npm 11 locally (it sprinkles `peer: true` annotations npm 10 reads differently and rejects). Regenerated with npm 10 (matching CI's Node 22.x default), so the diff vs main is now 18 lines (the actual instrumentation-undici entry) instead of 105 lines of npm-version drift noise. 2. (Copilot inline at sdk.ts:330) `otlpUrlPrefixes` was derived from raw Config strings, so a settings.json `"otlpEndpoint": "\"http://...\""` (quoted) or trailing `#fragment` would silently miss the prefix match and reintroduce the feedback loop the hook exists to prevent. Replaced the regex-based suffix trim with a WHATWG URL parser: - strips ?query, #fragment, trailing slash - trims symmetric ASCII quotes a user may have placed in settings.json - falls back to safe suffix trimming if URL parsing fails (misconfigured endpoint still gets SOME protection) 3. (CodeQL inline) Replaced the `/\?.*$/` regex in ignoreRequestHook with `indexOf('?')`/`indexOf('#')` slicing for ReDoS hygiene. The regex was linear in practice but flagged as polynomial — using indexOf removes the ambiguity and is arguably simpler. Added 3 tests in sdk.test.ts covering the new normalizations (#fragment on incoming path, quoted endpoint, #fragment on configured endpoint). 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * feat(telemetry): propagate X-Qwen-Code-Session-Id on outbound LLM requests Part 2 of #4384. Stacks on top of PR #4390 (traceparent via undici). Adds a product-namespaced HTTP header X-Qwen-Code-Session-Id to every outbound LLM request when telemetry is enabled, so server-side ingestion can correlate observed requests with qwen-code session metric/log records. Pattern matched from claude-code (X-Claude-Code-Session-Id, verified at src/services/api/client.ts:108 in their open-source repo). Critical design decision (design doc section 4.3): the OpenAI / Anthropic providers use a per-request fetch wrapper rather than the SDK defaultHeaders option, because content-generator SDK clients are constructed once and NOT recreated on /clear-triggered session resets (Config.resetSession updates this.sessionId but the contentGenerator keeps using the stale header value). Reading config.getSessionId() from inside the wrapper at request time gives the live value. Gemini provider uses static httpOptions.headers — @google/genai HttpOptions interface does not expose a fetch hook (only headers, baseUrl, apiVersion, timeout, extraParams). This is a known limitation: after session reset, Gemini X-Qwen-Code-Session-Id stays stale until the contentGenerator is recreated. Documented in telemetry.md and the design doc section 8.6; spans/logs continue to carry the live session id for trace/log correlation. Lazy-invalidate fix is a follow-up sub-issue. Header is omitted when telemetry is disabled OR when getSessionId returns an empty string (some HTTP middleware rejects empty header values). Integration sites: - packages/core/src/core/openaiContentGenerator/provider/default.ts (base class — automatically covered by deepseek/minimax/mistral/ modelscope/openrouter subclasses; openrouter calls super.buildHeaders) - packages/core/src/core/openaiContentGenerator/provider/dashscope.ts (overrides buildClient — must be touched separately; QwenContentGenerator inherits via this provider) - packages/core/src/core/anthropicContentGenerator/anthropicContentGenerator.ts - packages/core/src/core/geminiContentGenerator/index.ts (factory function, not the GeminiContentGenerator class — no signature change) End-to-end verification (local HTTP server in tmux): PASS: traceparent + X-Qwen-Code-Session-Id on every LLM request PASS: session id refreshes after simulated /clear (staleness regression guarded by llm-correlation-fetch.test.ts) PASS: OTLP upload traffic not traced (no feedback loop — PR A ignoreRequestHook working) Robot generated with Qwen Code https://github.com/QwenLM/qwen-code * fix(telemetry): R2 review fixes — critical correctness + tsc + boundary safety Adopts 7 review findings from wenshao on #4390 (+ duplicates from now-closed #4393). Critical bugs first, polish second. CRITICAL: 1. tsc TS2322 — wrapper return type incompatible with Anthropic SDK Fetch. `typeof fetch` (Node WHATWG, 2 overloads) is not structurally assignable to Anthropic's narrower `Fetch = (input: RequestInfo, init?) => ...`, even though they're call-compatible at runtime. Make wrapper generic `<TFetch extends FetchLikeLoose>` so callers preserve their exact fetch signature; cast the Anthropic call site through `unknown` with a comment explaining why. 2. tsc TS2352 / TS2493 — `baseFetch.mock.calls[0]![1] as RequestInit` was out-of-bounds when wrapped was called with no init arg. Replaced with a `makeFetchMock()` helper returning typed accessors. 3. normalizeOtlpPrefix catch fallback was DANGEROUS — a config of `"http"` produced prefix `"http"` which `startsWith`-matched every outbound HTTP request → silently disabled ALL instrumentation (no client spans, no correlation header — defeats the entire feature). Fixed: catch returns undefined + diag.warn. Misconfigured endpoint loses its feedback-loop guard (acceptable) instead of disabling all guards (catastrophic). 4. `url.startsWith(prefix)` matching was NOT boundary-safe — port collision (`:4318` matches `:43180`), hostname suffix collision (`otlp.example.com` matches `otlp.example.com.evil.net`), path-segment collision (`/v1` matches `/v1foo/x`). Replaced with origin-equality + path-prefix + boundary-char check (next char must be `/`, `?`, `#`, or end-of-string). 5. HttpInstrumentation also lacked the OTLP feedback-loop guard. The OTLP HTTP exporter (`@opentelemetry/exporter-trace-otlp-http`) uses node:http (patched by HttpInstrumentation, NOT undici). Without this, every OTLP upload batch creates a parasitic client span → feedback loop. Added `ignoreOutgoingRequestHook` that reuses the same `matchesOtlpPrefix` / `stripPathSuffix` helpers as the undici instrumentation. SAFETY: 6. Request input + undefined init dropped the Request's own headers (Authorization etc.) because `new Headers(undefined)` → `{...init, headers}` replaced them with just our session header. Fix: when input is a Request and init.headers is unset, seed from input.headers before adding ours. 7. Wrapped fetch had no try/catch — a throwing Config getter or Headers constructor would propagate as TypeError and break the LLM request path. Wrapped header construction in try/catch; on failure, fall through to baseFetch with original init (no header) + diag.warn. Telemetry must never break the model call. COVERAGE: - 3 new sdk.test.ts boundary tests (port/host/path) - 1 new sdk.test.ts normalizeOtlpPrefix catch-branch coverage - 1 new sdk.test.ts HttpInstrumentation OTLP guard test - 1 new sdk.test.ts proxy-mode wrapped-fetch test (default.test.ts) - 1 new anthropic test asserting wrapped fetch installed on Anthropic SDK - 2 new llm-correlation-fetch.test.ts (Request-headers preservation + try/catch fall-through) All 668 tests pass (1 pre-existing Anthropic User-Agent failure on main is unrelated). tsc clean. Declined: #10 DRY-refactor of baseFetch extraction across 3 sites — the duplication was pre-existing (default/dashscope buildClient was already near-identical), refactoring is a separate cleanup PR not gated by this feature. Will reply on the thread. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * chore(deps): allow patch updates for @opentelemetry/instrumentation-undici Switch from exact pin `0.14.0` to `^0.14.0` for consistency with the rest of the `@opentelemetry/*` deps in this block (all carated). For 0.x semver, npm treats `^0.14.0` as `>=0.14.0 <0.15.0`, so patch updates within the 0.14.x line — which are tied to the same `@opentelemetry/instrumentation@0.203.x` peer — flow in via `npm update` without requiring a manual package.json edit. A bump across the 0.x minor (e.g. 0.15.x) would shift the instrumentation peer compatibility and still requires explicit attention, which the caret correctly blocks. Per review feedback on #4390 (wenshao). 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * test(telemetry): stub getTelemetryEnabled + getSessionId in Gemini factory tests The X-Qwen-Code-Session-Id commit added a `staticCorrelationHeaders(gcConfig)` call inside the Gemini content generator factory. That helper reads `gcConfig.getTelemetryEnabled()` and `gcConfig.getSessionId()` per request. Both pre-existing Gemini tests in `contentGenerator.test.ts` build a minimal partial Config stub via `as unknown as Config` and only stub the methods the factory used to need. The new call path now hits the unstubbed methods at runtime, surfacing as `TypeError: config.getTelemetryEnabled is not a function` on all three CI platforms. Add the two missing stubs to both test cases. The Gemini factory continues to ignore the values when telemetry is off — these stubs only have to exist, not return anything in particular. Local check ran the full test suite for the four directories `/loop` covers plus `src/core/contentGenerator.test.ts` itself; all green. Also re-ran the other test files that build partial Config mocks via the same idiom (`client.test.ts`, `config.test.ts`, `nextSpeakerChecker.test.ts`, `content-generator-config.test.ts`) — none exercise the new code path. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): R3 review fixes — port + protocol + quote + safety Four issues found by wenshao reviewing the R2 boundary-safety pass on PR #4390. All four close gaps where the OTLP feedback-loop guard or the correlation-header path could fail silently. 1. **Port normalization mismatch** (sdk.ts ignoreOutgoingRequestHook): `normalizeOtlpPrefix` builds prefixes via `URL.origin`, which strips default ports (`:80` for http, `:443` for https). The hook reconstructed request origin manually as `${proto}://${host}${portPart}`, keeping the port. Result: prefix `http://collector` (no explicit port) didn't match a request to `http://collector:80/v1/traces` because their `.origin` differed → guard bypassed → feedback loop. Now the reconstructed origin is also routed through `URL` so both sides apply the same default-port stripping. 2. **HTTPS proto silent fallback** (sdk.ts ignoreOutgoingRequestHook): The `(req.protocol && ...) || 'http'` fallback would silently mis-bucket HTTPS requests as HTTP when `req.protocol` was unset, so HTTPS OTLP endpoints couldn't match their prefix. Changed to fail open: when proto can't be determined, return false (request gets instrumented). Worst case is a parasitic client span — observable, recoverable — versus the previous unbounded silent feedback loop. Picked fail-open over the bot's port-based heuristic because non-standard HTTPS ports break the heuristic but not fail-open. 3. **Quote-stripping divergence** (sdk.ts normalizeOtlpPrefix): `parseOtlpEndpoint` (line 109) uses `/^["']|["']$/g` which strips asymmetric leading/trailing quotes; `normalizeOtlpPrefix` previously only stripped symmetric pairs. A settings.json typo like `"value'` would let the exporter connect (parseOtlpEndpoint trims) but leave the guard returning `undefined` (normalizeOtlpPrefix rejected) → parasitic loop. Aligned `normalizeOtlpPrefix` to the same lenient regex. 4. **`staticCorrelationHeaders` missing try/catch** (llm-correlation-fetch.ts): `wrapFetchWithCorrelation` already catches all internal exceptions and falls through to baseFetch — same "telemetry must never break LLM path" contract was missing on the static-headers helper. A throw here would propagate up to the Gemini content-generator factory and crash content-generator init for the whole session. Wrapped the body in try/catch with `diag.warn` fall-through to `{}`. Tests: added 4 regression tests covering each scenario: - default-port HTTP request matched against portless prefix (1) - hook returns false when req.protocol missing on https endpoint (2) - asymmetric-quoted endpoint normalizes for guard parity (3) - staticCorrelationHeaders returns {} when config getter throws (4) 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * docs(telemetry): fix misleading "BOTH" wording in wrapFetchWithCorrelation The comment described the header-seeding logic as merging "BOTH the init.headers AND the Request's own headers", but the two branches are mutually exclusive — `new Headers(init?.headers)` runs unconditionally (empty Headers when init.headers is undefined), and the Request-headers copy only runs when init.headers is undefined. So in practice it's either-or, not BOTH. Reworded to match the actual logic per #4390 review feedback (wenshao). Behavior unchanged. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): strip port from req.host fallback + document undici scope Two issues found by wenshao reviewing the R3 boundary-safety fixes on PR #4390. 1. **`req.host` may already include `:port`** (sdk.ts ignoreOutgoingRequestHook): When `req.hostname` is absent and `req.host` is the fallback, the value may already be `"collector:4318"`. Naively appending `:${req.port}` produced `"http://collector:4318:4318"` → `new URL()` rejects → catch returns false → silent guard bypass for that request. Currently unreachable because `@opentelemetry/otlp-exporter-base` always sets `hostname` from WHATWG URL parsing, but the fallback exists in the code and must be correct — a future OTLP transport that emits `host` without `hostname` would silently trigger the feedback loop. Strip the port when falling back; bracketed IPv6 literals like `"[::1]:443"` keep their bracketed host intact. 2. **Undici scope honesty** (telemetry.md): Previous docs framed the propagation as "outbound LLM requests", but `UndiciInstrumentation` actually patches `globalThis.fetch` for the whole process — `WebFetch`, MCP clients, IDE extension calls all get spans + `traceparent` injection too. Added a "Scope: all fetch() calls, not just LLM" subsection covering: (a) trace ID leakage to third-party URLs (the user-supplied destinations of `WebFetch` see our trace ID; not secret per W3C but worth knowing); (b) non-LLM span volume inflating OTLP batches with a workaround tip. Per-destination scoping toggle deferred as a follow-up — out of scope for this PR. Added regression test for the host:port-fallback path. Test exercises the previously broken combination (hostname absent, host carries port) through the existing test harness. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * feat(telemetry): scope X-Qwen-Code-Session-Id to first-party hosts by default Address LaZzyMan's REQUEST_CHANGES review of PR #4390. The original design injected `X-Qwen-Code-Session-Id` on every outbound LLM request gated only by `telemetry.enabled`. Review caught that this broadcasts a stable cross-request client identifier to every configured third-party provider (OpenAI, Anthropic, OpenRouter, MiniMax, ModelScope, Mistral, vanilla Gemini, ...), which the claude-code precedent does NOT justify — claude-code is a first-party Anthropic→Anthropic flow; qwen-code is an open-source CLI connecting to many providers. Fix: add a host allowlist with a deliberately narrow default. The header is now only attached to destinations whose hostname matches: dashscope.aliyuncs.com dashscope-intl.aliyuncs.com *.dashscope.aliyuncs.com *.dashscope-intl.aliyuncs.com *.alibaba-inc.com *.aliyun-inc.com This is exactly the set where the LLM provider, the upstream telemetry backend (ARMS Tracing), and qwen-code itself are the same legal entity — mirroring the first-party claude-code pattern and preserving the real product value (server-side trace stitching against DashScope) without exposing the session id to third parties. Operators with broader correlation requirements override via: "telemetry": { "sessionIdHeaderHosts": ["*"] // restore broadcast "sessionIdHeaderHosts": [] // fully disable "sessionIdHeaderHosts": ["api.example.com", "*.foo"] // custom allowlist } Implementation: - NEW `telemetry/trusted-llm-hosts.ts`: `DEFAULT_SESSION_ID_HEADER_HOSTS` + `matchesTrustedHost(hostname, patterns)` + `extractRequestHost(input)`. Pattern syntax is intentionally tiny (bare hostname OR `*.suffix`, dot-anchored to reject `evil-alibaba-inc.com` style attacks). Unit-tested in dedicated test file including TLD/sub-domain attack vectors. - `wrapFetchWithCorrelation` (openai + anthropic providers): resolves the allowlist at wrap time (Config snapshot), inspects each request's destination URL inside `correlationFetch`, falls through to baseFetch for non-trusted destinations. Wildcard escape hatch via `["*"]`. - `staticCorrelationHeaders` (Gemini factory): now takes an optional `destinationUrl` and applies the same host gate. The Gemini SDK default endpoint `generativelanguage.googleapis.com` is NOT on the default allowlist, so vanilla Gemini calls receive no header — matching the "first-party only" scope. Operators who put the Gemini SDK on a DashScope-compatible endpoint via `baseUrl` get the header naturally. - `Config.getTelemetrySessionIdHeaderHosts()` getter + `TelemetrySettings.sessionIdHeaderHosts` interface field + JSON schema entry in `settingsSchema.ts`. Wired through `resolveTelemetrySettings`. - Defensive optional-chaining + try/catch on the Config getter call at wrap time so partial test mocks (or pre-getter Config implementations) fall back to the default allowlist rather than crashing buildClient. Tests: 12 new cases covering host match/skip on default allowlist, sub-domain handling, TLD-suffix attack rejection, `["*"]` broadcast override, `[]` full-disable, custom operator allowlist, unparseable destination (fail closed), and the three Gemini factory paths (googleapis.com default → omit; DashScope `baseUrl` → inject; custom allowlist → inject). Docs updated in `docs/developers/development/telemetry.md` Session correlation header section, including override examples and the new Gemini host-gate semantics. Closes the LaZzyMan REQUEST_CHANGES blocker. The cross-vendor fingerprint-broadcast failure mode is now opt-in rather than default, restoring the first-party-only semantics that make the claude-code precedent applicable. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): R5 review fixups — Vertex destination + ["*"] trim + docs Self-review pass on commit |
||
|
顾盼
|
a8a6ad2d06
|
feat(core)!: redesign auto-compaction thresholds with three-tier ladder (#4345)
* feat(core)!: redesign auto-compaction thresholds with three-tier ladder
Replaces the single 70% proportional threshold with a three-tier ladder
(warn/auto/hard) that combines proportional fallback with absolute
reservation. Large-window models (>=128K) now reserve ~33K instead of
30% of the window, freeing tens of thousands of context tokens that the
old formula wasted.
Other improvements bundled in the same redesign:
- Compression sideQuery now disables thinking and caps maxOutputTokens
at 20K, matching claude-code so the buffer math is predictable across
providers (Anthropic/OpenAI/Gemini handle thinking budgets
inconsistently)
- Failure handling upgraded from one-shot permanent lock to a 3-strike
circuit breaker; reactive overflow still latches immediately
- New estimatePromptTokens helper closes the lag-by-one-turn and
first-send-is-0 gaps in lastPromptTokenCount
- Hard-tier rescue pulls reactive overflow recovery forward to before
the API call, saving an oversized round-trip
- /context command displays the three-tier ladder + current tier
- tipRegistry's context-* tips track the new thresholds instead of
fixed 50/80/95 percentages
BREAKING CHANGE: chatCompression.contextPercentageThreshold setting is
removed. Settings files containing the field log a one-line deprecation
warning at startup and the value is ignored; behaviour is now controlled
by built-in thresholds via the new computeThresholds() function.
Design: docs/design/auto-compaction-threshold-redesign.md
Plan: docs/plans/2026-05-14-auto-compaction-threshold-redesign.md
* test(core): fix leftover hasFailedCompressionAttempt option in compress test
A pre-existing test case at chatCompressionService.test.ts:678 still
passed `hasFailedCompressionAttempt: false` in the CompressOptions
shape; rebasing onto current main surfaced this as a typecheck error
because the field was renamed to `consecutiveFailures` (Task 7 of the
three-tier ladder migration). Update to `consecutiveFailures: 0` —
semantically equivalent, the test asserts the side-query is called
when `force: true`, no other behaviour change.
* fix(core): drop compaction summary when output hits maxOutputTokens cap
Adds a defensive guard in ChatCompressionService.compress() that detects
when the side-query summary hit COMPACT_MAX_OUTPUT_TOKENS (20K). In that
case the summary is likely truncated mid-content, so we drop it and
return NOOP rather than persist a half-summary. The next send re-tries;
reactive overflow still catches the catastrophic case where the API
rejects the next request as too large.
Documented in the design doc as risk #2; the bot reviewer on PR #4168
correctly pushed for it to land alongside the threshold redesign rather
than as a follow-up since the new 20K cap is what makes truncation
likely in the first place.
* fix(cli): render three-tier thresholds in /context TUI view
The Task 11 redesign updated the non-interactive text formatter
(formatContextUsageText) but left ContextUsage.tsx — the interactive
React component that real /context users see — unchanged. As a result
the TUI still showed the old single "Autocompact buffer" line and none
of the new warn/auto/hard ladder.
Adds a "Compaction thresholds" section after the per-category breakdown:
- Effective window
- Warn / Auto / Hard threshold rows with a ▶ marker on the row the
current usage has crossed
- Current tier label coloured by severity (safe→green, warn/auto→
yellow, hard→red)
The existing progress bar legend (Used / Free / Autocompact buffer)
is preserved because it's tied to the three-segment progress bar
visualisation; the new section adds the absolute numbers + tier badge
on top of that.
Caught by the tmux e2e test (PR #4168 ci-monitor follow-up). Pre-fix
the assertion 'Compaction thresholds' missed completely from the TUI;
post-fix the new section renders correctly for fresh and live sessions
on 1M / 200K / 128K windows.
* fix(core,cli): address PR #4168 review batch 4
Behavior fixes:
- MAX_TOKENS truncation guard now returns COMPRESSION_FAILED_EMPTY_SUMMARY
instead of NOOP so the consecutive-failure breaker actually trips after
repeated max-length summaries (R1.1).
- Reactive overflow failure increments consecutiveFailures by 1 instead
of latching to MAX in one shot, so a transient network blip doesn't
permanently disable auto-compaction. The hard-tier rescue resets the
counter, which remains the designated recovery path (R1.2).
- /context current-tier classification uses rawOverhead (system + tools +
memory + skills) as the tier input when API data is not yet available,
rather than 0 — large inherited contexts no longer silently show 'safe'
(R2.2).
Performance:
- sendMessageStream computes effectiveTokens ONCE and passes it through
TryCompressOptions.precomputedEffectiveTokens, so the cheap-gate inside
service.compress doesn't redo the estimation. Also fixes the
imageTokenEstimate inconsistency between the rescue and cheap-gate
paths (R1.3 + R1.4).
- Steady-state path (lastPromptTokenCount > 0) skips the costly
getHistory(true) clone — estimatePromptTokens only needs the user
message in that branch.
Code hygiene:
- BYTES_PER_TOKEN → CHARS_PER_TOKEN (inputs are char counts, not byte
counts; CJK text would mislead under the old name) (R3.1).
- Drop dead getContextUsagePercent helper + index re-export — no callers
in source after the threshold rewire (R1.5).
- Add a comment on estimatePromptTokens' first-send fallback documenting
the ~15-20K under-estimate (system prompt + tools + skills) and that
reactive overflow is the safety net (R3.3).
Tests:
- New CLI ContextUsage.test.tsx exercises the React renderer for the
three-tier section: section presence, ▶ marker placement per tier,
current-tier label coloring (R1.6).
- New chatCompressionService.test.ts case pins that a stale
contextPercentageThreshold: 0 value in user settings no longer
short-circuits compaction (R2.1).
- New tokenEstimation.test.ts case covers functionResponse (distinct
nested-parts branch from functionCall) (R3.5).
- New geminiChat.test.ts integration test exercises the real
ChatCompressionService — not a mock — for the first-send-after-
inherited-history scenario where lastPromptTokenCount=0 and only the
full-history estimate can cross the auto threshold (R3.4).
Declined: R3.2 (change `>=` to `>` on the MAX_TOKENS guard). The current
operator catches the at-cap case as suspicious, which is intentional —
landing exactly at the output cap is far more likely truncation than
clean stop given p99.99 ≈ 17K. With R1.1 in place, persistent truncations
trip the breaker after MAX_CONSECUTIVE_FAILURES so the worst case is
bounded.
* fix(core,cli): address PR #4168 review batch 5
- R5.1: tighten /context tier comment + TODO. The rawOverhead-based fix
doesn't cover `--continue` restores with many history messages (since
rawOverhead excludes messagesTokens). UI may still show 'safe' for one
render until the first send. Documented inline and added a TODO to plumb
chat history into collectContextData for same-source-of-truth as the
cheap-gate.
- R5.2a: add TODO(finish_reason) at the truncation guard. The `>= cap`
heuristic false-positives on legitimate at-cap summaries; the proper
signal is finish_reason which runSideQuery doesn't surface today.
- R5.2b: split telemetry — new CompressionStatus.COMPRESSION_FAILED_OUTPUT_TRUNCATED
enum value. Distinct from EMPTY_SUMMARY so logs/telemetry can tell
prompt-quality failures (tune prompt / splitter) from capacity failures
(raise cap / shrink splitter input). isCompressionFailureStatus()
treats both as failures so the breaker behavior is unchanged.
- R5.3: expand consecutiveFailures JSDoc to clarify it tracks
"non-force, non-hard-rescue consecutive failures" — hard-rescue resets
the counter and force=true skips increments, so the counter is the
"regular path" health signal only; reactive overflow is the real
safety net for the force-only paths.
- R5.4: document the CompressOptions field rename
(hasFailedCompressionAttempt: boolean → consecutiveFailures: number)
as an SDK breaking change in the design doc with migration guide.
* fix(core): disambiguate hard-rescue from manual /compress orphan-strip
Self-review (dual reviewer / pr-triage round 1) caught a correctness
regression in the hard-rescue path:
`sendMessageStream` calls `tryCompress(force=true)` from inside the
pre-push window when `effectiveTokens >= hard`. The service's
orphan-strip predicate at `chatCompressionService.ts:426-429` gated on
`force` alone, which conflated two distinct call shapes:
- manual `/compress` (force=true, trigger='manual'): user-initiated
between turns; trailing model funcCall IS orphaned because no
funcResponse is coming
- hard-rescue (force=true, trigger='auto'): automatic mid-turn;
trailing model funcCall is ACTIVE because its matching funcResponse
is sitting in the pending `userContent` waiting to be pushed
The strip fired for both, so a hard-rescue triggered mid tool-use loop
would drop the active funcCall. After compression returned and
`userContent` (the funcResponse) was pushed, the next API request
carried tool_result with no matching tool_use → provider validation
error.
The in-code comment at L422-424 already documented this exact
constraint for the auto-compress case (`force=false`), but reusing
`force=true` for hard-rescue silently violated the same constraint.
Fix:
- Gate `hasOrphanedFuncCall` on `compactTrigger === 'manual'` instead
of `force`. The trigger field already disambiguates intent.
- `sendMessageStream` hard-rescue now passes `trigger: 'auto'`
explicitly (without it, `force=true` defaults to `trigger='manual'`
via the `?? (force ? 'manual' : 'auto')` resolver).
Sibling audit for "force=true non-manual callsites":
- `GeminiClient.tryCompressChat` (manual /compress): correct — manual
- `sendMessageStream` hard-rescue: fixed in this commit
- `sendMessageStream` reactive overflow catch: already passes
trigger='auto'; runs AFTER API call (userContent in history), so if
it observes a trailing funcCall it IS orphaned but findCompressSplitPoint
handles the case without needing the strip
RED-first regression test added:
`preserves trailing model+funcCall under hard-rescue (force=true + trigger=auto)`
in `chatCompressionService.test.ts`. Failed against pre-fix code (the
strip dropped the funcCall); passes against the fix.
Adjacent fixes from the same triage round:
- `docs/users/configuration/settings.md`: the
`chatCompression.contextPercentageThreshold` row still said "use 0
to disable compression entirely" — code has ignored the value since
the removal commit. Marked the row REMOVED with migration guidance
pointing at the design doc.
- `packages/core/src/config/config.ts`: the deprecation warning now
tells users how to silence it (remove the key) and where to read
current behavior, instead of just announcing the removal.
- `docs/design/auto-compaction-threshold-redesign.md`: closed Open
Question 2 (small-window hard/auto collapse) — decision is to NOT
annotate `/context`, with rationale on file.
Tests: 2395 core tests passing, typecheck clean.
* docs(core): fix tier-collapse direction in auto-compaction design doc
Self-review on the
|
||
|
jinye
|
fd75f77e19
|
feat(telemetry): Phase 4a — TTFT capture + GenAI semconv dual-emit (#3731) (#4417)
Some checks failed
Qwen Code CI / Classify PR (push) Has been cancelled
E2E Tests / E2E Test (Linux) - sandbox:docker (push) Has been cancelled
E2E Tests / E2E Test (Linux) - sandbox:none (push) Has been cancelled
E2E Tests / E2E Test - macOS (push) Has been cancelled
Qwen Code CI / Lint (push) Has been cancelled
Qwen Code CI / Test (macos-latest, Node 22.x) (push) Has been cancelled
Qwen Code CI / Test (ubuntu-latest, Node 22.x) (push) Has been cancelled
Qwen Code CI / Test (windows-latest, Node 22.x) (push) Has been cancelled
Qwen Code CI / Post Coverage Comment (push) Has been cancelled
Qwen Code CI / CodeQL (push) Has been cancelled
|
||
|
jinye
|
64401e1d17
|
feat(telemetry): support custom resource attributes and add metric cardinality controls (#4367)
* feat(telemetry): support custom resource attributes and add metric cardinality controls Resolves #4365. Adds two coupled OpenTelemetry capabilities to make qwen-code's telemetry production-ready in multi-team / multi-tenant deployments: 1. Custom resource attributes via standard `OTEL_RESOURCE_ATTRIBUTES` and `OTEL_SERVICE_NAME` env vars and a new `telemetry.resourceAttributes` setting. Operators can now tag every span / log / metric with `team`, `env`, `cost_center`, or anything else their backend needs. 2. Metric cardinality controls. `session.id` is moved off the OpenTelemetry Resource (where it auto-attached to every metric data point and caused unbounded time-series fan-out on Prometheus / ARMS Metric / etc.) and gated behind a new opt-in `telemetry.metrics.includeSessionId` toggle. Spans and logs still carry `session.id` for trace and log correlation. Reserved keys (`service.version`, `session.id`) are stripped from both env and settings sources with a `diag.warn`. `OTEL_SERVICE_NAME` follows the OTel spec precedence (highest priority for `service.name`). Settings JSON values are runtime-coerced to strings as defense against hand-edited non-conforming JSON. Breaking change: metrics no longer carry `session.id` by default. Operators who need it can restore the previous behavior with `QWEN_TELEMETRY_METRICS_INCLUDE_SESSION_ID=true` or `telemetry.metrics.includeSessionId: true` in settings.json; recommended only for short-term debugging since it re-introduces the cardinality problem. For long-term session-level analysis, prefer trace and log backends which handle per-event data without cardinality pressure. Design doc: docs/design/telemetry-resource-attributes-design.md 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * docs(telemetry): align reserved-key descriptions with implementation Round 1 review fixes (#4367). After session.id was added to RESERVED_RESOURCE_ATTRIBUTE_KEYS in Codex review, four user-facing descriptions still claimed only service.version was reserved: - packages/core/src/telemetry/config.ts (merge comment) - packages/core/src/config/config.ts (TelemetrySettings JSDoc) - packages/cli/src/config/settingsSchema.ts (schema description) - packages/vscode-ide-companion/schemas/settings.schema.json (regenerated) Also corrects scope claim: resource attributes apply to every signal the SDK exports (OTLP and file outfile share the same Resource), not just OTLP. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * docs(telemetry): clarify warning destination and surface percent-encoding hint Round 2 self-review fixes (#4367). Two small but real UX gaps: 1. Reserved-key / malformed-pair / coerce warnings route to the debug log (per #3986), not the console — so a user who types `OTEL_RESOURCE_ATTRIBUTES=service.version=2.0` sees no feedback that the value was silently dropped. Adds a "Troubleshooting" section in telemetry.md telling users where to look, and a note in the parser docstring documenting where warns go. 2. A literal (unencoded) comma in an env var value is a common foot-gun: the parser splits on it, producing a malformed second half that is silently dropped. Updates the warn text to include a "hint: percent-encode literal commas as %2C" callout, and adds the same guidance to the docs. Deferred to a follow-up: startup-time stderr summary of dropped attributes. Stderr during TUI render could break Ink rendering, so the right surface needs separate design. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * test(telemetry): cover first-`=` split contract in OTEL_RESOURCE_ATTRIBUTES parser Per review feedback on #4367. The parser uses `indexOf('=')` so the first `=` separates key and value while subsequent `=` stay in the value. The behavior was correct but untested; a future refactor to `split('=')` would silently break base64-padded, JWT, or connection-string values. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * feat(telemetry): tighten resource-attribute input validation + startup summary Adopts review feedback from #4367 (wenshao via Qwen Code /review). Five accepted suggestions, bundled because they all touch the same parse/coerce/strip pipeline: 1. Key percent-decoding (CRITICAL). `parseOtelResourceAttributes` now percent-decodes both keys and values per the OTel / W3C Baggage spec. Without this, `OTEL_RESOURCE_ATTRIBUTES=service%2Eversion=99` lands on Resource as the literal key `service%2Eversion`, bypassing the reserved-key filter; a collector that decodes keys downstream could then resurrect `service.version` and spoof the version label. 2. Startup summary of dropped attributes. Every `diag.warn` in resource-attributes.ts routes only to the OTel debug log (per #3986), giving operators zero feedback when their attributes are silently dropped. Helpers now optionally accumulate diagnostics into a `ResourceAttributeWarnings` array; the resolver collects them and the SDK emits a one-time console summary at init (before Ink renders, so no TUI conflict). 3. `||` instead of `??` for service.name fallback. Settings can put an empty string through `??`, producing a blank `service.name` that some backends reject. `||` falls through to the default. 4. `coerceStringResourceAttributes` now trims keys and skips empty/whitespace-only keys, matching `parseOtelResourceAttributes`. Previously `{" ": "x"}` or `{"team ": "y"}` from settings.json would land as malformed Resource attributes. 5. `OTEL_SERVICE_NAME` is trimmed before the truthy check, so values like `' '` or `'\t'` are treated as unset rather than producing a whitespace-only service name on Resource. One suggestion declined (in-thread reply on PR): - "Redundant `?? {}` in sdk.ts:160" — intentional defense-in-depth for `vi.mock('../config/config.js')` callers in `telemetry.test.ts` where auto-stub returns undefined. The reviewer is right that production code paths never hit it, but tests do. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) * fix(telemetry): trim whitespace-only service.name + add invalid-key-encoding test Adopts two review suggestions on #4367 (wenshao via Qwen Code /review): 1. `service.name` fallback uses `.trim() || SERVICE_NAME` instead of plain `||`. Plain `||` lets whitespace-only values (`" "`, `"\t"`) through as truthy, producing a blank service name on Resource that some backends reject. Both settings (no value trimming) and env (`%20` decodes to `" "`) can deliver such values. Test added. 2. Adds `key%ZZ=val` to the parameterized parser test to cover the invalid-percent-encoding-on-key catch branch. Previously only the value-side catch was tested. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) |
||
|
易良
|
a3037889a6
|
fix(core): replace structuredClone with shallow copy to prevent OOM in long sessions (#4286)
* docs: add OOM investigation reports and auto-compaction redesign proposal
- Runtime memory investigation plan
- Non-interactive memory benchmark report
- OOM reproduction report with 2GiB/4GiB synthetic tests
- Runtime diagnostics benchmark report
- Auto-compaction threshold redesign proposal
* fix(core): replace structuredClone with shallow copy to prevent OOM
Replace `structuredClone(this.history)` (called up to 4x per turn on the
send path) with a lightweight shallow copy via `copyContentContainer()`.
This eliminates the OOM root cause in long tool-heavy sessions where the
full deep clone exceeded remaining V8 heap headroom.
Key changes:
- Add `copyContentContainer()` helper ({...content, parts: [...parts]})
- Add `getRequestHistory()` private method for the send path
- Add `getHistoryShallow()`, `getHistoryTailShallow()`,
`peekLastHistoryEntry()`, `getLastModelMessageText()`,
`getHistoryLength()` for read-only callers
- Remove HEAP_PRESSURE_COMPRESSION_RATIO safety net (no longer needed
now that the underlying OOM cause is fixed)
- Update chatCompressionService to use getHistoryShallow(true)
- Update nextSpeakerChecker to send only lastMessage (not full history)
- Update memoryDiagnostics with process-tree RSS measurement
* feat(core): add runtimeDiagnostics utility for heap/memory instrumentation
Required by content generators (anthropic, openai, logging) which import
runtimeDiagnostics for optional heap-pressure telemetry during streaming.
Gated by QWEN_CODE_PROFILE_RUNTIME=1 environment variable.
* fix(cli): update doctorCommand test mocks for new MemoryDiagnostics interface
Add missing maxRSSRaw, maxRSSUnit, and processTree fields to test fixtures
to match the updated MemoryResourceUsage and MemoryDiagnostics interfaces.
* fix(vscode-ide-companion): use public core imports
* fix: address review comments — type guards, dead fallbacks, and doc accuracy
Code:
- Fix unsound type guard: `'text' in part` → `typeof part.text === 'string'`
in geminiChat.ts and client.ts (Copilot + wenshao feedback)
- Remove unnecessary optional chaining and dead fallback chains in client.ts
(getHistoryShallow, peekLastHistoryEntry, getHistoryLength, etc. now call
GeminiChat methods directly)
- Add 5s timeout to `execFileAsync('ps', ...)` in memoryDiagnostics.ts
Docs:
- Fix GiB conversion accuracy and add single-run caveat to summary
- Add Node.js version to test environment table
- Fix auto-compaction attempt count (5→4) in OOM report
- Soften root-cause attribution certainty
- Add MCP child process context to investigation plan
- Clarify "Codex" reference (→ OpenAI Codex)
- Fix truncated MCP server name (chrome → chrome-devtools)
- Remove duplicate verification commands in benchmark table
- Clarify thread exhaustion vs V8 heap OOM distinction
- Add workload confound caveat to before/after comparison
- Fix SUMMARY_RESERVE "hard relationship" vs thinking budget contradiction
* fix(core): restore fallback chains in client.ts for mock compatibility
The previous commit removed optional chaining from client.ts wrapper
methods, but client.test.ts mocks getChat() with partial objects that
lack the new shallow methods. Restore ?. fallback chains so both
production (GeminiChat) and test (mock) paths work correctly.
* docs: clarify memory review follow-ups
* docs: fix runtime benchmark unit conversion
* docs: add default-heap OOM stress report
* fix: update copyright year to 2026 in new files [skip ci]
New files added in this PR had 2025 copyright headers. Updated to 2026
to reflect the current year.
|
||
|
顾盼
|
a7e05302e6
|
feat(worktree): Phase C — session persistence, hooksPath, Footer + WorktreeExitDialog, three-mode --resume restore (#4174)
* docs(worktree): update design doc — split Phase C/D, add Future section
- Phase C: session persistence + hooksPath + StatusLine + WorktreeExitDialog
- Phase D: --worktree CLI flag + symlinkDirectories
- Future: sparse checkout, .worktreeinclude, tmux, PR reference parsing
- Feature comparison table updated with Phase A/B completion status
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* docs(worktree): add Phase C implementation plan
8 tasks: WorktreeSession sidecar storage, hooksPath setup,
EnterWorktree/ExitWorktree session wiring, useWorktreeSession hook,
Footer display, --resume context injection, WorktreeExitDialog.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* docs(worktree): update Phase C plan after claude-code comparison
- WorktreeSession: add originalHeadCommit field
- hooksPath: add .husky/ detection + skip-if-already-set logic
- StatusLine payload: expand worktree field to match claude-code schema
- WorktreeExitDialog: load dirty state on mount, display counts in dialog
- UIState.activeWorktree: add originalCwd, originalBranch, originalHeadCommit
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(worktree): add WorktreeSession sidecar storage
New worktreeSessionService.ts exposes read/write/clear functions for the
sidecar JSON file at <chatsDir>/<sessionId>.worktree.json. SessionService
gains getWorktreeSessionPath() so callers don't need to know the layout.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): configure core.hooksPath after worktree creation
createUserWorktree() now sets `core.hooksPath` inside the new worktree to
the main repo's hooks directory (.husky preferred, .git/hooks fallback) so
commits inside the worktree run the same pre-commit checks as the main
repo. Mirrors claude-code's performPostCreationSetup logic — skips the
subprocess when the value already matches to avoid ~14ms spawn overhead.
Failures are non-fatal: the worktree is still usable without hooks.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): persist WorktreeSession sidecar in EnterWorktreeTool
After creating a worktree, EnterWorktreeTool now writes a sidecar JSON
file at <chatsDir>/<sessionId>.worktree.json with the full session state
(slug, paths, branches, original HEAD SHA). --resume reads this in Phase
C task 7 to restore worktree context. Best-effort: write failures don't
abort the creation.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): clear WorktreeSession sidecar in ExitWorktreeTool
After successful keep or remove, ExitWorktreeTool now clears the sidecar
JSON file iff its slug matches the worktree being exited. The slug check
prevents wiping the sidecar when the user exits a worktree that isn't
currently tracked (multiple worktrees on disk, sidecar tracks one).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): expose active worktree via useWorktreeSession + UIState
New useWorktreeSession hook watches the sidecar JSON file (created by
EnterWorktreeTool, deleted by ExitWorktreeTool) and returns the current
WorktreeSession or null. AppContainer wires it into a new
UIState.activeWorktree field consumed by Footer (Task 6) and
WorktreeExitDialog (Task 8).
A showWorktreeExitDialog state placeholder is added too, hardcoded false
until Task 8 wires the dialog trigger.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): show active worktree in Footer + StatusLine payload
Footer renders `⎇ <branch> (<slug>)` when activeWorktree != null, but
only when the user has no custom statusline (their script likely
handles it from the stdin payload itself).
useStatusLine's StatusLineCommandInput gains a `worktree` field with
{name, path, branch, original_cwd, original_branch} — matches claude-code's
schema so statusline scripts can be shared across both CLIs.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): inject context hint on --resume when worktree is active
On --resume, if the session has a WorktreeSession sidecar, append an
INFO history item pointing the model at the worktree path so it
continues using it for file operations. Stale sidecars (worktree dir
deleted out-of-band) are cleaned up so the Footer indicator doesn't
go stale.
qwen-code can't process.chdir() the way claude-code does because
Config.targetDir is immutable; the context hint is the equivalent
behavioral cue.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): add WorktreeExitDialog with dirty-state inspection
WorktreeExitDialog renders when the user double-presses Ctrl+C inside a
worktree. On mount it runs `git status --porcelain` and
`git rev-list --count <originalHeadCommit>..HEAD` to show how many
uncommitted files and new commits the user would discard by choosing
"Remove". The dialog never auto-removes — every exit goes through
explicit user confirmation per requirements.
handleExit in AppContainer intercepts the second-press quit when
activeWorktree is set and shows the dialog instead. A new UIAction
handleWorktreeExit(choice) routes the user's choice through removal
(via GitWorktreeService.removeUserWorktree) + sidecar cleanup + /quit.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* docs(worktree): add Phase C E2E test plan
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* docs(worktree): fix E2E test plan sidecar path + jq selector
- sidecar lives at ~/.qwen/projects/<sanitized-cwd>/chats/, not ~/.qwen/tmp/<hash>/
- qwen --output-format json emits a JSON array, not NDJSON — jq needs .[]
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* fix(worktree): add showWorktreeExitDialog to dialogsVisible
Phase C task 8 introduced showWorktreeExitDialog state and the dialog
render in DialogManager, but missed adding the flag to the dialogsVisible
OR expression. DefaultAppLayout only renders DialogManager when
dialogsVisible is true, so the dialog was never shown — second Ctrl+C
in a worktree silently absorbed instead of triggering the prompt.
Caught by Group E E2E tests.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* feat(worktree): extend --resume context restore to headless + ACP modes
Phase C task 7 originally placed the worktree-restore logic in
AppContainer.tsx (TUI only). E2E Group C exposed that headless and ACP
modes never run AppContainer, so stale sidecars accumulate and the model
loses worktree context after --resume.
Refactor to a shared `restoreWorktreeContext` helper in core, then wire
the three entry points:
- TUI (AppContainer): keep historyManager.addItem(INFO) UX, route via
the helper.
- Headless (nonInteractiveCli): prepend the notice as a system-reminder
block on the user prompt; emit a `worktree_restored` system message to
the JSON adapter so SDK consumers can react.
- ACP (Session.pendingWorktreeNotice): set by acpAgent.loadSession on
resume, consumed and cleared exactly once on the next #executePrompt.
All three modes call the same helper, so stale-sidecar cleanup is
consistent. Helper covers: missing sidecar, live worktree dir,
deleted worktree dir, regular file at worktreePath, malformed JSON.
5 new unit tests for restoreWorktreeContext (13/13 pass total).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* test(worktree): add ACP-mode integration tests for --resume context
Covers:
- acpAgent.worktree.test.ts (3 tests): loadSession sets
pendingWorktreeNotice only when worktree dir is live, clears
stale sidecar otherwise, swallows restoreWorktreeContext errors.
- Session.worktree.test.ts (4 tests): #executePrompt prepends the
system-reminder block exactly once on first prompt, clears the
pending notice, second prompt sees no leakage, no-op when nothing
was set.
E2E via real ACP protocol is impractical without a Zed client; these
tests cover the integration boundaries directly.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* docs(worktree): clarify hooksPath comment + pendingWorktreeNotice one-shot rationale
Two doc-only fixes from PR #4174 review:
- gitWorktreeService.ts: previous hooksPath comment overstated the
optimization (claimed claude-code's ~14ms saving but we still do a
read subprocess). Rewrite to be explicit: write-skip only, read
retained, parseGitConfigValue's full optimization deliberately not
ported because the read happens once per worktree creation.
- Session.ts: pendingWorktreeNotice doc now explains why it's one-shot
(after the first prompt the worktree path is already in conversation
context; re-injecting would clutter history without adding signal).
No behavior change.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* fix(test): add getResumedSessionData to nonInteractiveCli mock Config
CI surfaced TypeError: config.getResumedSessionData is not a function
across 12 tests in nonInteractiveCli.test.ts. The Phase C
|
||
|
顾盼
|
b0ea9f4849
|
fix(core): decouple auto-memory recall from main-agent request path (#4172)
* docs: add async memory recall design spec and implementation plan
* refactor(core): introduce MemoryPrefetchHandle, replace pendingRecallAbortController field
* refactor(core): fire memory recall as non-blocking prefetch with settledAt flag
* refactor(core): replace blocking await with zero-wait settledAt poll at UserQuery consume point
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(core): inject recalled memory on first ToolResult when UserQuery consume point misses
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(core): replace pendingRecallAbortController with pendingMemoryPrefetch in all cleanup paths
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(memory): remove 1s AbortSignal.timeout from relevanceSelector — caller controls lifetime
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* test(core): update auto-memory tests for async prefetch pattern — drop fake timers and deadline references
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* test(core): add ToolResult inject test — memory injected on first ToolResult when recall settles after UserQuery
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(core): address codex review findings on async memory recall
Three findings fixed:
1. Abort previous prefetch before installing a new one (line 1059):
A new UserQuery/Cron used to overwrite pendingMemoryPrefetch without
aborting the old controller, leaking an unbounded background recall now
that the 1s side-query timeout is gone.
2. Move the UserQuery consume poll AFTER the async reminder setup:
ensureTool + listSubagents are awaited between the old poll location and
the final assembly, so recalls that settled during those awaits used to
be missed (and a tool-less turn never got a ToolResult retry). The poll
now runs immediately before requestToSend assembly, and unshifts memory
to the front of systemReminders to preserve ordering.
3. Append memory after functionResponse on ToolResult turns:
The Qwen API requires the functionResponse part to immediately follow
the model's functionCall (see lines 1209-1213). Prepending memory text
risked breaking that pairing on the native Gemini path. Appending keeps
the pair intact on Gemini and produces the same OpenAI output (text
becomes a separate user message after the tool messages).
Tests:
- Updated ToolResult inject test to assert memory index > functionResponse
- Added abort-previous-prefetch test (mid-flight UserQuery aborts old handle)
224/224 tests pass; tsc clean on changed files.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* docs(core): add JSDoc + clarifying comments per review feedback
Annotations only, no behavior change:
- MemoryPrefetchHandle: full JSDoc covering lifecycle (create → consume → discard)
- UserQuery consume site: explain why we unshift (front of systemReminders)
- ToolResult inject site: reference hasPendingToolCall pattern instead of
brittle line numbers when citing the Qwen functionCall/Response constraint
- relevanceSelector.ts: explain why the side-query has no inline timeout
(caller controls lifetime via MemoryPrefetchHandle.controller)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(core): bridge caller abort signal into memory prefetch + doc accuracy fixes
Behavior fix (addresses copilot review on client.ts:1071):
- When the parent sendMessageStream signal aborts (user Ctrl-C / Esc),
the prefetch controller now aborts too. Previously the recall side-query
would keep running until a later cleanup (next UserQuery / /clear / etc),
wasting fast-model tokens on work whose result no one would consume.
- Listener uses { once: true } and is also removed in the promise's
finally() so a long-lived parent signal doesn't accumulate listeners
across many turns under normal completion.
- Edge case: if signal is already aborted when fire runs, abort the
controller synchronously instead of attaching a listener.
Test:
- New regression guard: "should abort the pending prefetch when the caller
signal aborts" — verifies the abort handler installed on the recall side
fires once the parent signal aborts.
Doc accuracy (addresses copilot review on the design spec):
- ToolResult inject: was documented as "prepend", actual implementation
appends to preserve functionCall/functionResponse pairing. Updated both
the prose summary and the code sample.
- Cleanup section: was documented as 6 abort-locations including the
"post-consume clear"; the consume sites don't actually abort (the promise
has already settled). Reorganized as 5 abort-and-clear sites + 2
clear-only sites with the distinction made explicit.
- Fire path snippet: added the abort-previous-prefetch line and the
caller-signal bridge so the spec matches the current implementation.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(core): consolidate memory-prefetch lifecycle + safety nets per round-3 review
Architectural (root-cause fix for cleanup-path sibling drift):
- New private cancelPendingMemoryPrefetch() consolidates the abort+clear
idiom (was duplicated across 6 sites). Logs at debug when discarding a
settled-but-unconsumed handle so missing-memory scenarios are diagnosable.
- New private tryConsumeMemoryPrefetch() consolidates the
consume-and-mark-consumed dance (was duplicated UserQuery + ToolResult).
- All existing cleanup sites + the two newly-flagged early-return sites
(LoopDetected, Error) now use the helper; future early-returns can rely
on the finally-block safety net.
- sendMessageStream try-finally now uses a `normalCompletion` flag:
only the bottom-of-try return path preserves the prefetch (intentional
— next ToolResult turn may consume it); every other exit (uncaught
exception, abnormal early-return) goes through cancelPendingMemoryPrefetch
in finally.
Diagnostics:
- Restored AbortError debug log in fire-path catch (was silent after
removing the deadline mechanism; aborts now come from 4+ sources so a
trace is valuable).
- Updated stale "deadline" log in recall.ts to reflect current abort
sources (caller signal / new UserQuery / cleanup / 30 s safety timeout).
Safety net:
- Added 30 s ceiling in relevanceSelector via AbortSignal.any(...).
Generous enough that normal ~1 s recalls don't trip it; bounds zombie
side-queries if the model API hangs and the caller never aborts.
Replaces the uncancellable `new AbortController().signal` fallback that
would have left callerless invocations running indefinitely.
Doc sync:
- Design doc updated: UserQuery consume code sample now shows `unshift`
(matches implementation) with an inline note on the prepend-vs-append
contrast.
Tests:
- New regression guard: resetChat aborts pending prefetch and clears the
handle.
- New regression guard: LoopDetected mid-stream aborts pending prefetch
and clears the handle (catches the sibling-drift bug this round caught).
227/227 tests pass; tsc clean on changed files.
Declined from this round:
- `await Promise.resolve()` after fire path: defensive — current code has
multiple natural microtask drains before consume point. Added comment
documenting the dependency instead.
- Renaming `settledAt: number | null` to `settled: boolean`: timestamp
has diagnostic value for future instrumentation; current consumers'
null-check usage is documented in the JSDoc.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(test): correct getLastLoopType mock return type — null, not undefined
CI tsc --build (stricter than --noEmit) caught:
src/core/client.test.ts(2996,65): error TS2345: Argument of type
'undefined' is not assignable to parameter of type 'LoopType | null'.
getLastLoopType()'s contract returns LoopType | null; the test mock was
returning undefined. Switched to null to match the type.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(core): preserve memory prefetch across hook/next-speaker continuations + accurate recall abort log
Round-4 review findings (self-inflicted regression from round-3):
1. Preserve pending prefetch on `return hookTurn` (Stop-hook continuation)
and `return continueTurn` (next-speaker continuation). The round-3
`normalCompletion = true` was only set at the bottom-of-try `return turn`,
leaving these two recursive-yield paths to trip the finally cleanup.
When the inner Hook turn produced tool calls, the subsequent ToolResult
turn found `pendingMemoryPrefetch === undefined` and memory was silently
dropped.
2. recall.ts catch log distinguishes caller-driven aborts (heuristic
genuinely skipped below) from the 30s safety-net timeout in
relevanceSelector (the caller's signal is NOT aborted by that path,
so the heuristic fallback actually runs).
Regression guard added:
- "should PRESERVE the pending prefetch when next-speaker continueTurn
returns" — was red before this commit, green after.
258/258 tests pass; tsc --build clean.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
|
||
Shaojin Wen
|
ad23c7ab34
|
docs: user + design docs for --json-schema structured output (#4051)
* docs: add user + design docs for --json-schema structured output Follows up #3598 (cli/core feature shipped to main, no docs). **User doc** `docs/users/features/structured-output.md` — covers quick-start, schema input forms (inline + `@path`), output shapes per `--output-format`, parse-time restrictions, retry/failure modes, privacy redaction, permission gating, MCP shadow-tool handling, and a worked `jq`-piped pipeline example. Registered under the existing `features/_meta.ts` so it shows up in the docs sidebar between "Headless Mode" and "Dual Output". **Design doc** `docs/design/structured-output/structured-output.md` — why the synthetic-tool-whose-param-schema-is-the-user-schema approach, the four-stage parse-time validation pipeline, `schemaRootAcceptsObject`'s decided-vs-deferred boundaries, main-turn vs drain-turn parity via `processToolCallBatch`, the structured- success terminal block, the cross-surface privacy redaction sharing `STRUCTURED_OUTPUT_REDACTED_ARGS`, subagent context handling (`forSubAgent`), MCP shadow-tool guard, the compatibility surface, alternatives considered (and why rejected), and a file-by-file index. Both docs are English-only — repo convention is English-only for both `docs/users/features/` (zero zh-CN siblings) and `docs/design/` (only `customize-banner-area/` has a zh-CN twin). Open to adding zh-CN translations as a separate PR if there's demand. * docs(structured-output): address PR review feedback User doc: - explicit stdout-vs-stderr contract and `{}`-schema behavior. - 500 ms shutdown-holdback latency note. - ReDoS warning for user-supplied `pattern` keywords. - root `$ref` rejection + `allOf` workaround. - per-retry token cost note. - sibling-suppression success vs retry paths split out. - numeric exit codes (1 / 53 / 130) for every failure mode. - new "Session resumption" section for --continue / --resume. Design doc: - gloss the ToolSearch on-demand-loading reference. - `not` row: drop the array-indexing-lookalike `[…]`. - 500 ms holdback is best-effort, not guaranteed. - redaction rationale extends to validation-failure retries. - `CORE_TOOLS` phrasing: structured_output is excluded FROM the set; skill is in a separate dynamically-discovered category. - subagent suppression maintainer note (single brittle call path). - `--bare` parenthetical lists the three retained core tools. - PR #4001 status (closed 2026-05-11, superseded). * docs(structured-output): correct empty-schema / holdback / SIGINT claims Three doc claims were stronger than the actual code behaviour: - **Empty schema produces `{}`, not `null`.** `turn.ts` normalises the tool args via `(fnCall.args || {})` before they land in `structuredSubmission`, so a zero-arg call against `{}` is emitted as `{}` on stdout. The `?? null` in the adapter is defence-in-depth for the strictly-undefined case, which the upstream path doesn't produce. - **Holdback is a cap, not a fixed wait.** The loop guard is `Date.now() < deadline && registry.hasUnfinalizedTasks()`, so it exits immediately when nothing is in flight. Reword as "capped at ~500 ms" with an early-exit note. - **SIGINT can still flush a captured result.** The holdback loop does not poll the abort signal, so a SIGINT after the structured call is captured but before `adapter.emitResult` finishes may still land on stdout. Treat exit code 130 as the source of truth. Also addresses the new auto-review summary suggestion about per-turn schema cost: pull the cost callout up out of the bullet list (so it covers both retry cost and schema-embedded-every-turn cost), since the schema-embedding cost isn't retry-specific. * docs(structured-output): correct stdout/stderr + json-mode envelope claims Two doc claims didn't match `JsonOutputAdapter.emitResult`: - **Model prose doesn't go to stderr in text mode.** Only error messages and log lines do. Successful runs emit just the JSON-stringified payload on stdout; accumulated assistant prose is discarded entirely (not mirrored to stderr). Point users at `--output-format json` / `stream-json` when they need the prose. - **`--output-format json` emits a JSON array, not a single document with top-level fields.** The adapter calls `JSON.stringify(this.messages)` where `messages` is an array of message objects. `structured_result` lives on the final `type: "result"` element of that array, not at the document root, so consumers must read `.[-1].structured_result` rather than `.structured_result`. * docs(structured-output): note schema-itself reaches the provider The Privacy section so far only described `structured_output` *args* being redacted from local on-device surfaces (telemetry + chat recording). The schema body is a separate exposure surface — it ships as the function declaration's `parameters` block on every model request, so `enum`, `const`, `default`, `examples`, `description`, `$comment`, etc. travel to the provider in cleartext. Users defaulting to "redaction covers everything" could legitimately leak secrets via schema-literal fields. Add a callout in the user doc, plus a parallel paragraph in the design doc explaining why the redaction stops at on-device surfaces (the model needs the schema to satisfy the tool-call contract, so provider-side redaction isn't possible). * docs(structured-output): correct stdout-on-failure / ReDoS example / hooks / --bare deny / typo Five issues from the latest /qreview pass: - **stdout-vs-stderr is text-mode only.** In `--output-format json` and `stream-json`, the failure result message is emitted on stdout (final element of the JSON array, or the terminating `result` line on the JSONL stream). Wrappers in those modes must switch on `is_error`, not on whether stdout is empty. - **ReDoS example didn't actually demonstrate the threat.** JSON Schema `pattern` only fires on string instances, and tool args are always objects, so the bare `{"pattern": "(a+)+b"}` schema doesn't constrain anything the model can supply. Move the pattern inside a string-typed property. - **Hooks see raw `tool_input`.** `PreToolUse` / `PostToolUse` / `PostToolUseFailure` receive the unredacted args — including HTTP hooks that can forward off-device. Call this out explicitly so users with audit-style catch-all hooks know to filter or add hook-side redaction. - **`--bare` drops settings-level deny.** Bare mode builds `mergedDeny` as `[...(bareMode ? [] : settings.permissions.deny), …]` — settings-level denies are skipped while the synthetic tool stays registered. Argv-level `--exclude-tools` still applies. Document this exception in the user doc and the design doc. - **`maxSessionTurns` hint typo.** The hint points at "schema is unsatisfiable" — the original text inverted the polarity. * feat(core): PR-2.5 — post-promote stream redirect + natural-exit registry settle Closes the two limitations PR-2 (#3894) deferred for the Phase D part (b) Ctrl+B promote flow (#3831): 1. **Post-promote stream redirect**: today the `bg_xxx.output` file is frozen at promote time because `ShellExecutionService` detaches its data listener as part of PR-1's ownership-transfer contract. PR-2.5 wires a caller-side `onPostPromoteData` callback so bytes from the still-running child append to the file via an `fs.createWriteStream` opened in `handlePromotedForeground`. 2. **Natural-exit registry settle**: today the registry entry stays `'running'` until `task_stop` / session-end `abortAll` fires its abort listener. PR-2.5 wires `onPostPromoteSettle` so natural child exit transitions the entry to `'completed'` / `'failed'` with the right exitCode / signal / error message. ## Service (`shellExecutionService.ts`) - New exported types: `ShellExecuteOptions`, `ShellPostPromoteHandlers`, `ShellPostPromoteSettleInfo`. - `execute()` options bag now accepts `postPromote?: { onData, onSettle }`. Threaded through to both `executeWithPty` and `childProcessFallback`. - PTY's `performBackgroundPromote` (line ~1159): after disposing the foreground data + exit + error listeners, RE-ATTACH minimal forwarders that call `postPromote.onData` / `postPromote.onSettle` when the caller opted in. Backwards compat: when `postPromote` is unset the PR-2 detach-everything contract is preserved (the re-attach is gated on each callback being defined). - `childProcessFallback`'s `performBackgroundPromote` (line ~706): same pattern — re-attach `stdout.on('data', ...)`, `stderr.on('data', ...)`, `child.once('exit', ...)`, `child.once('error', ...)` when the caller opted in. `error` listener routes through `onSettle` with `error` populated, so spawn-side errors after the foreground errorHandler detached don't crash the daemon via the default unhandled `'error'` event. - Both paths wrap caller callbacks in try/catch so a thrown handler doesn't crash the child's data loop / unhandled-rejection the service. ## Shell tool (`shell.ts`) - New `PromoteArtifacts` type — slots shared between the foreground `execute()` postPromote handlers (which fire on the service side as soon as promote happens) and the post-resolve `handlePromotedForeground` finalizer (which runs after `await resultPromise` returns). The two race; the buffer + settle-queue absorb that race so neither chunks nor the eventual exit info are lost. - `executeForeground` wires `postPromote` handlers that route data to either `promoteArtifacts.stream` (if open) or `promoteArtifacts.buffer` (drained when the stream opens), and queue settle info if the wired handler isn't yet installed. - `handlePromotedForeground` opens `fs.createWriteStream(outputPath, { flags: 'w' })`, writes the initial snapshot first, drains the buffer, then registers the entry and wires `onSettleWired` with the full registry decision table: - `error` set → `registry.fail(shellId, error.message, endTime)` - `exitCode === 0` → `registry.complete(shellId, 0, endTime)` - non-zero exitCode → `registry.fail(shellId, "Exited with code N", endTime)` - signal !== null → `registry.fail(shellId, "Terminated by signal N", endTime)` - all-null fallback → `registry.fail(shellId, "Exited with unknown status", endTime)` - Fires queued settle synchronously after wiring so a fast command that exits between promote and finalizer doesn't get lost. - Self-audit catch: closes the output stream on the `registry.register` throw path so the FD doesn't leak past the orphan-child kill. ## Tests - 3 new in `shellExecutionService.test.ts`: - `post-promote bytes route to postPromote.onData when callback provided` - `postPromote.onSettle fires on natural child exit after promote` - `backwards compat: without postPromote, listeners stay fully detached` - 3 new in `shell.test.ts` under a `foreground → background promote PR-2.5` describe block: - `post-promote bytes APPEND to bg_xxx.output via write stream` - `natural child exit transitions registry entry to "completed"` - `non-zero exit / signal / error → "failed" with descriptive message` - Bulk-replaced 50 prior `{},` (empty 6th-arg shellExecutionConfig) with `expect.objectContaining({}),` + added `expect.objectContaining({ postPromote: expect.any(Object) }),` as the 7th-arg expectation for the foreground execute call. - Updated the existing `registers a bg_xxx entry on result.promoted` test to assert on `fs.createWriteStream` + `stream.write` instead of the now-removed `fs.writeFileSync` snapshot path. 182/182 shell.test.ts pass + 73/73 shellExecutionService.test.ts pass + 111/111 coreToolScheduler.test.ts pass + 60/60 AppContainer.test.tsx pass; tsc + ESLint clean. Self-audit: 3 rounds (positive / reverse / cross-file) found one issue — output stream FD leak on `registry.register` throw — and fixed it before flagging complete. All flagged edge cases (stream errors, child-exits-before-wire-up race, task_stop during natural- exit window, promote-never-happens cleanup, backwards compat without callbacks) have explicit handling and / or test pinning. * fix(core): #4102 review wave — 3 Critical + UTF-8 + tests 3 Critical race/correctness issues + 1 multibyte-corruption suggestion + 3 test coverage gaps addressed: **Critical 1 — child_process late-chunk drop (service)** Settle was fired on 'exit', but stdout/stderr can emit buffered data between 'exit' and 'close'. Late chunks landed in `promoteArtifacts.buffer` after shell.ts had already closed the stream + transitioned the registry → silently dropped → truncated `bg_xxx.output`. Switched to listening on 'close' which guarantees all stdio is fully drained. (code, signal) payload is identical to 'exit', just with proper ordering. **Critical 2 — stream-flush wait before registry transition (shell)** `stream.end()` is asynchronous; pending writes can still be in the libuv queue when it returns. The old code transitioned the registry immediately after `.end()`, so a /tasks consumer could observe a `completed` entry and read the output file BEFORE the trailing bytes were on disk. Fixed: wired settle now `stream.once('finish', ...)` BEFORE calling `registry.complete/fail`. `error` event also short-circuits to the transition so a late ENOSPC doesn't hang the settle path forever. **Critical 3 — stream-open-fail buffer leak (shell)** If `fs.createWriteStream` threw, the catch path set `stream = null` but the foreground `onData` handler would still take the `stream === null` branch and push chunks into `promoteArtifacts.buffer` — unbounded growth under a sustained child whose output file couldn't be opened. Added a `streamFailed: boolean` latch on `PromoteArtifacts`. When set, `onData` drops chunks (with a debug log) instead of buffering. The catch branch sets the latch. **Suggestion — shared TextDecoder corrupts multibyte UTF-8 (service)** child_process post-promote used ONE TextDecoder for both stdout AND stderr. The decoder's continuation-byte state machine assumes one byte source; interleaved multibyte chunks corrupted. Now uses separate decoders + flushes both with `decode()` (no `stream: true`) on settle so trailing bytes surface as their final characters. **Suggestion — llmContent reflects already-settled status (shell)** When the queued-settle drain transitions the registry synchronously (fast-exit race), the model-facing copy was still saying "Status: running. … task_stop({...})". Updated to branch on `postPromoteAlreadySettled` / `postPromoteFinalStatus` — when the process is already gone, the copy says "Status: completed/failed" and replaces the `task_stop` suggestion with "Process has already exited; no `task_stop` needed". **Suggestion — test coverage gaps** Added: (a) `queued-settle race: onSettle BEFORE handlePromotedForeground completes` — custom service impl fires onSettle synchronously before resolving the promote promise, pins the drain path. (b) child_process post-promote tests for stdout/stderr forwarding + 'close'-not-'exit' settle + spawn-error settle. **Self-audit**: Round 1 + reverse audit. Stream.once mock added to fire 'finish' synchronously so existing tests don't hang on the new flush wait. 76/76 shellExecutionService.test.ts (+3) + 183/183 shell.test.ts (+1) pass; tsc + ESLint clean. * fix(core): #4102 review wave-2 — 3 more from gpt-5.5 C1 (shell.ts:2227): the WriteStream `'error'` event handler only logged. `fs.createWriteStream` reports common open failures (ENOENT / EACCES / ENOSPC) asynchronously via that event rather than throwing. Result: `promoteArtifacts.stream` kept pointing at the failed stream; `onSettleWired` attached a `.once('finish')` listener that would never fire → registry stuck on `running` forever. Latch the failure (null the shared `stream` slot, set `streamFailed`); `onSettleWired`'s existing `if (!stream)` branch then transitions the registry immediately. C2 (shellExecutionService.ts:1468): the promote handoff removes the foreground `ptyErrorHandler` and only re-attaches data + exit listeners. A subsequent PTY `error` event had no listener — Node treats an unhandled `error` from an EventEmitter as a fatal exception that takes the whole CLI down. Attach a post-promote forwarder that ignores expected PTY read-exit codes (EIO / EAGAIN, same filter the foreground handler uses) and routes unexpected errors through `postPromote.onSettle` with `error` populated. Single-fire latch shared with `onExit` so settle never fires twice. C3 (shell.ts:2503): `onSettleWired` waits for the stream's asynchronous `'finish'` event before flipping `postPromoteAlreadySettled`, but the model-facing `statusLine` was built immediately after invoking `onSettleWired` on the queued settle. A fast-exited promoted command could therefore land "Status: running" + a `task_stop` instruction in production even though settle was already observed. Split into two flags: `postPromoteSettleObserved` (set synchronously when settle is classified) drives the model copy; the registry transition stays behind the stream flush. Tests: +1 PR-2.5 wave-2 PTY error-routing test; +2 shell.ts tests (stream open async error → registry still transitions; async `'finish'` after queued-settle drain → llmContent says 'completed' before registry transition fires). * fix(core): #4102 review wave-3 — 4 actionable from deepseek-v4-pro T2 (shell.ts:2456) — Critical buffer-leak race `onSettleWired` previously set `promoteArtifacts.stream = null` BEFORE calling `stream.end()`. Any `postPromote.onData` chunk that landed between that null assignment and the actual flush completing saw `stream === null && streamFailed === false` and pushed into `promoteArtifacts.buffer` — a buffer that has no further drain path (the foreground finalizer has already returned). Result: chunks stranded indefinitely; PTY mode in particular hits this because `onExit` can fire while kernel buffers still hold data. Fix drains the pre-settle buffer to the stream BEFORE nulling AND latches `streamFailed = true` so any subsequent chunk drops via the existing `else if (streamFailed)` arm in `onData` instead of leaking. Updates the `streamFailed` doc to cover both setters (open-fail and settle-done) so the dual semantic is explicit. T3 (shell.ts:2262) — silent chunk-drop in catch path When `fs.createWriteStream` throws synchronously (rare: ENOENT on a vanished tmpdir), chunks already in `promoteArtifacts.buffer` were silently lost with no observability — oncall reading a truncated `bg_xxx.output` had no way to distinguish "stream open failed" from "child produced nothing." Logs the dropped chunk count and empties the buffer. T5 (shell.ts:2443) — opaque all-null fallback The "Exited with unknown status" fallback fired the registry to 'failed' without any context about which fields were null. This branch is meant to be unreachable; hitting it indicates the service emitted a defective settle info object. Includes the field values in both the fail message and a warn log so the oncall engineer can tell this path apart from the other "failed" branches. T6 (shellExecutionService.ts:1452) — leaked PTY post-promote listeners `ptyProcess.onData(...)` returns an `IDisposable` that was being discarded; same for `onExit`. The `'error'` listener function was also not captured (no way to `removeListener` it). EventEmitter holds refs to listener closures, which transitively hold refs to `onPostData` / `onPostSettle` / the caller's `promoteArtifacts`. While bounded by the PTY's lifetime, the closures keep the caller's state pinned for the post-settle delay window. Captures all three handles into `postPromoteDataDisposable` / `postPromoteExitDisposable` / `postPromoteErrorListener`, then releases them via a shared `disposePostPromoteListeners()` call from `firePostSettle` (idempotent — each slot null-checked and nulled after disposal). Tests: +1 service test for IDisposable + error-listener cleanup; +2 shell.ts tests for buffer drain race and catch-path snapshot fallback. Existing tests stay green (262 → 265 in the touched suites; 7819 → 7822 across the core package). * fix(core/test): drop unused 'registry' in wave-3 T2 test (TS6133) CI build failed across all platforms with src/tools/shell.test.ts(4395,15): error TS6133. The variable was a leftover from copying the queued-settle test pattern; the wave-3 T2 test inspects writeStreamMock.write call history directly and never reads the registry, so the assignment is dead code. Drop it. * fix(core): #4102 review wave-4 — 6 actionable from gpt-5.5 + deepseek-v4-pro T1 (Critical, shellExecutionService.ts:860 child_process onSettle exactly-once) The PTY path used a `firePostSettle` latch but child_process wired `close` and `error` independently to `onPostSettle`. A spawn-side error followed by Node's auto-emitted `'close'` would call the caller's settle TWICE, racing the registry transition. Added the same single-fire latch on the child_process path. T2 (Critical, shell.ts:2264 handoff race reorder) Original order was `write(snapshot) -> drain buffer -> assign stream`. Synchronous today (no race in current code), but assign-after-drain leaves a hazard for any future refactor that adds an `await` inside the drain loop — a chunk arriving in that window would land in `promoteArtifacts.buffer`, then post-assign chunks would write to the stream first, producing out-of-order bytes until the settle drain. Reordered to `write(snapshot) -> assign stream -> drain buffer`, which closes the hazard regardless of future async additions. T3 (Suggestion, shellExecutionService.ts:816 decoder flush gated on onSettle) The trailing-multibyte flush ran inside the `child.once('close', ...)` handler, which was only installed when `onSettle` was set. An `onData`-only caller (no onSettle) lost trailing continuation bytes silently. Hoisted flush into `flushPostPromoteDecoders` called from `firePostSettle`, and made `firePostSettle` available on the `'close'` path independent of onSettle (T6 install). T4 (Suggestion, shell.ts:1700 promoted ANSI passthrough) The regular `executeBackground` path strips ANSI before writing to `bg_xxx.output`; the promoted-foreground onData path appended raw chunks. Reading `bg_xxx.output` after Ctrl+B showed plain text up to the snapshot then raw `\x1b[31m` / cursor-move / clear-screen sequences for the post-promote tail — unreadable. Apply `stripAnsi(rawChunk)` before write/buffer, matching the executeBackground contract. T5 (Suggestion, shellExecutionService.ts:786 UTF-8 hardcoded) The post-promote child_process decoders were hard-coded to `new TextDecoder('utf-8')`, but the foreground decoder runs encoding detection via `getCachedEncodingForBuffer`. On a non-UTF-8 child (e.g. GBK on a Chinese Windows shell), the snapshot decoded correctly but the post-promote tail was mojibake. Capture the foreground decoder's `.encoding` property and reuse it for post-promote (with utf-8 fallback if foreground hadn't seen any bytes yet, and a try/catch around `new TextDecoder` for the rare unsupported-encoding case). T6 (Suggestion, shellExecutionService.ts:1540 `error` listener gated on onSettle) The post-promote `error` listener was attached only when `onSettle` was set. An `onData`-only caller still had the foreground errorHandler detached; a post-promote spawn error would then crash the CLI via Node's unhandled-error default. Hoisted the close + error listeners into `if (postPromote)` so any caller opting into post-promote gets crash protection; if `onSettle` is absent the listeners log + drop instead of routing. T7 (Suggestion, shellExecutionService.ts:791 onSettle-only pipe-block deadlock) Same root cause as T6: when only `onSettle` is set, the foreground `stdout`/`stderr` 'data' listeners are detached and no post-promote listener replaces them. The Readables stay paused, the OS pipe buffer fills (~64KB on Linux), the child blocks on `stdout.write`, 'close' never fires, onSettle never fires. Added `child.stdout?.resume()` and `child.stderr?.resume()` in the no-onData branch so the child can drain its pipes and reach exit. T8 (Suggestion, shell.ts:2614 dead inspectLine ternary) `inspectLine`'s ternary returned the same string on both sides — copy-paste leftover from when the other two adjacent ternaries (statusLine / stopLine) were correctly varied. Collapsed to a single string assignment. Tests: +5 regression tests (4 child_process: T1 double-fire latch, T3 onData-only flush, T6 onData-only error survives, T7 onSettle- only resume; +1 shell.ts: T4 ANSI strip). 265 -> 270 in the touched suites; 7822 -> 7827 across the core package; full suite green. * fix(core/test): use ShellOutputEvent type in wave-4 onData callbacks (TS2345) CI lint failed on the wave-4 (T3 / T6) tests with TS2345: pushing ShellOutputEvent into Array<{type:string;chunk:unknown}> narrows incompatibly. Switch to ShellOutputEvent[] (matches earlier helpers at lines 758/966) and discriminate the union via .type === 'data' when reading .chunk so the narrowed multibyte assertion still type-checks. * docs(structured-output): address doudouOUC's four review findings - Tighten JSON/stream-json paragraph: not all failures emit a result to stdout (exit 53 / exit 130 are stderr-only); check exit code first - Fix suppressed-sibling retry guidance: re-issue in a separate turn that does not include structured_output (avoids re-suppression) - Distinguish settings-deny (exit 53) from --exclude-tools (exit 1) in Permission gating section - Replace <projectDir> placeholder with actual path ~/.qwen/projects/<sanitized-cwd>/chats/<sessionId>.jsonl in both docs Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * docs(structured-output): fix Permission gating — both deny paths strip registration Forward audit against source code found that the Permission gating section incorrectly distinguished settings.permissions.deny (claiming tool stays visible, exit 53) from --exclude-tools (claiming declaration stripped, exit 1). Both go through the same mergedDeny → isToolEnabled path and both prevent registration — the model never sees the tool. Corrected both docs to reflect the actual mechanism: typical outcome is plain text (exit 1), with maxSessionTurns (exit 53) as the fallback if the model loops through other tools. * docs(structured-output): address doudouOUC's May 17 review (5 items) - Clarify validation is client-side Ajv, not provider-side - Qualify "same way" with DeclarativeTool abstraction parenthetical - Match symptom→cause structure for maxSessionTurns hint - Expand $ref workaround with concrete $defs example - Clarify Dual Output See Also doesn't require --json-schema * docs(structured-output): address 2 unresolved design-doc suggestions 1. Privacy/redaction section: note hooks as intentionally non-redacted surface (matches user-doc "Hooks see raw args" callout). 2. Dual call-site section: clarify differing post-helper termination flow between main-turn (direct return) and drain-turn (sentinel hop). * docs(structured-output): address doudouOUC's May 17 review (2 nits) 1. Failure-paths table: align "three common causes" cell with the symptom→cause framing already used at parse-time validation pipeline section ("common stuck-run symptom and its two likely causes"). 2. Dual call-site section: fix factual inaccuracy from prior commit — `drainOneItem` is `async (): Promise<void>` and returns nothing. The two-hop termination is via closure-mutated `structuredSubmission` (set by `processToolCallBatch`, checked by `drainLocalQueue` and the holdback loop), not a return-value sentinel. --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> |
||
|
jinye
|
8dfbdaa5d4
|
feat(telemetry): unify span creation paths for hierarchical trace tree (#4126)
* feat(telemetry): unify span creation paths for hierarchical trace tree (#3731 P3 Phase 1) Replace disconnected withSpan/startSpanWithContext calls in runtime with session-tracing typed helpers so LLM and tool spans become children of the interaction span instead of siblings under the session root. - Add toolContext ALS with runInToolSpanContext() for concurrent-safe tool span scoping (uses AsyncLocalStorage.run, not enterWith) - Wire startLLMRequestSpan/endLLMRequestSpan in loggingContentGenerator for both streaming and non-streaming paths - Wire startToolSpan/endToolSpan + startToolExecutionSpan/endToolExecutionSpan in coreToolScheduler with proper try/finally lifecycle - Remove redundant withSpan('client.generateContent') wrapper from client.ts - Fix endToolSpan to not override pre-set status when metadata is omitted - Change startToolExecutionSpan to read parent from toolContext ALS - Update tests for new span creation APIs and remove dead test infrastructure * fix(telemetry): address CI build errors in session-tracing tests - Remove unused _toolSpan variable (TS6133) - Use bracket notation for index signature property access (TS4111) * fix(telemetry): update coreToolScheduler and loggingContentGenerator test mocks - coreToolScheduler.test.ts: mock startToolSpan/endToolSpan/runInToolSpanContext instead of withSpan; update cancellation tests for restored safeSetStatus call - loggingContentGenerator.test.ts: fix attribute keys in mock, add try/catch in endLLMRequestSpan mock to match production best-effort behavior * fix(telemetry): address review feedback from wenshao - Add debugLogger.warn in catch blocks of endLLMRequestSpan/endToolSpan/ endToolExecutionSpan instead of silent swallowing - Add JSDoc on endToolSpan documenting intentional no-metadata-no-status contract with setToolSpanFailure/setToolSpanCancelled - Add warning in startToolExecutionSpan when called outside runInToolSpanContext (no active toolContext) - Sanitize error message in endToolExecutionSpan: use constant TOOL_SPAN_STATUS_TOOL_EXCEPTION instead of raw error message * fix(telemetry): use partial mock for telemetry/index.js in coreToolScheduler tests The full mock shadowed all re-exports (logToolCall, etc.) causing 49 test failures. Use importActual to preserve other exports, only override span functions. * fix(telemetry): getLastToolSpan must skip tool.execution sub-spans startToolExecutionSpan mock also pushes to toolSpanRecords, so at(-1) returns the execution sub-span instead of the tool span. Use findLast to filter by name. * fix(telemetry): address second round review feedback - Remove redundant safeSetStatus(span, OK) on success path — endToolSpan in finally already sets OK via metadata - Add llm_request.stream attribute (true/false) to distinguish streaming vs non-streaming LLM requests in trace backends * fix(telemetry): endToolSpan mock writes to record directly Bypass span.setStatus() in mock to avoid potential interference from vitest module resolution. Write to statusCalls/ended directly on the ToolSpanRecord. * fix(telemetry): mock session-tracing.js directly instead of telemetry/index.js Mocking the barrel re-export (telemetry/index.js) with importActual was unreliable — vitest's module resolution could bind production code to the real endToolSpan before the mock override took effect. Mock the source module (session-tracing.js) directly to guarantee interception. * fix(telemetry): fix endToolSpan status on success — toolCalls is empty in finally Root cause: checkAndNotifyCompletion clears this.toolCalls before the finally block in executeSingleToolCall runs, so the tc lookup always returns undefined. Fix: set OK status explicitly in _executeToolCallBody's success path via safeSetStatus(span, OK), and call endToolSpan() without metadata in finally (just ends the span, preserves pre-set status from any path). * fix(telemetry): address Codex review — activate OTel context, end span on failure - Wrap non-stream generateContent API call + logging in context.with(spanContext) so nested OTel spans (HTTP instrumentation, log-bridge spans) parent to qwen-code.llm_request instead of session root (matches streaming path). - runInToolSpanContext now also activates OTel context via otelContext.with, not just the custom toolContext ALS. Hooks/HTTP/IO during tool execution now correctly parent to qwen-code.tool span. - Split end*Span helpers: span.end() runs in its own try/catch so a throwing setAttributes/setStatus can't leak unended spans. * fix(telemetry): address Codex review v2 — session-root fallback + execution span timing - start{LLMRequest,Tool,ToolExecution}Span now fall back to getSessionContext() when no parent context, instead of otelContext.active(). Side-query LLM calls (auto-title, recap) now stay in the session trace instead of starting a new detached trace. - Move startToolExecutionSpan() to BEFORE invocation.execute(), matching claude-code. Previously the synchronous setup inside execute (shell command preprocessing, child_process.spawn) ran outside the execution span. * fix(telemetry): address Codex review v3 — sync throw, idle timeout race, test coverage - coreToolScheduler.executeSingleToolCall: move try-block to wrap invocation.execute() so synchronous throws (e.g. shell setup failure) flow into the same catch path as async rejections. Previously a sync throw would leak the execution span and skip failure hooks. - loggingStreamWrapper: track spanEndedByTimeout flag so a stream that resumes after the 5-min idle timeout does not run the final endLLMRequestSpan (which would no-op anyway, but the flag also stops resetSpanTimeout from queuing further timer callbacks). - coreToolScheduler.test: add execution sub-span assertions for success, ToolResult.error, thrown invocation exceptions, and pre-hook denial. - loggingContentGenerator.test: capture setAttribute calls into the mock span attributes record; assert llm_request.stream is false for non-stream and true for stream paths. * fix(telemetry): address Codex review v4 — consistency + test coverage gaps - endLLMRequestSpan now uses spanCtx.span for mutations (matches endToolSpan/endToolExecutionSpan pattern). Same object, but consistent lookup pattern prevents future drift. - Mocks capture endLLMRequestSpan and endToolSpan/endToolExecutionSpan metadata so tests can assert token counts, durationMs, success, error are forwarded correctly. Add assertions on: * Non-stream LLM: inputTokens, outputTokens, success on response path * Non-stream LLM: success: false + sanitized error on rejection * Stream LLM: final lastUsageMetadata reaches endLLMRequestSpan * Tool execution sub-span: success: true on happy path * Tool execution sub-span: success: false on ToolResult.error * Tool execution sub-span: success: false + sanitized error on throw - Add OTel error resilience tests: when setAttributes or setStatus throws, span.end() must still run and the span must be removed from activeSpans. Covered for endLLMRequestSpan, endToolSpan, endToolExecutionSpan. * fix(telemetry): address Codex review v5 — abort distinction + API symmetry - session-tracing.ts SpanContext.type: comment 'tool.blocked_on_user' | 'hook' as Phase 2 forward-declarations (no helpers wired yet). - endToolExecutionSpan: align no-metadata-no-status behavior with endToolSpan. Currently no caller omits metadata, but the asymmetric default (OK vs preserve-pre-set) was a maintenance trap. - loggingContentGenerator generateContent (non-stream) catch block: call endLLMRequestSpan BEFORE the logging block, mirroring the streaming path. Defense-in-depth against logging-side rejections. - loggingContentGenerator: restore abort-specific span status message. All three LLM error paths (non-stream catch, stream eager-error catch, stream loggingStreamWrapper finally) now use API_CALL_ABORTED_SPAN_STATUS_MESSAGE when req.config.abortSignal.aborted, matching the original withSpan('client.generateContent') behavior. Trace backends can now distinguish cancellations from real failures. - coreToolScheduler _executeToolCallBody catch: distinguish abort vs exception in execSpan error message. New constant TOOL_SPAN_STATUS_TOOL_CANCELLED prevents operators filtering exec spans for errors from seeing cancellation false positives. - New test asserting exec span uses cancelled-by-user message when the invocation throws after abort. * fix(telemetry): always write 'success' attribute on tool spans E2E review found qwen-code.tool spans never carry the `success` boolean attribute (the helper only writes it when metadata is passed, and the finally block calls endToolSpan(toolSpan) without metadata). This breaks the most common observability query — filtering tool failures with `success = false` — because tool spans don't have that field at all. Fix: setToolSpanFailure / setToolSpanCancelled now also call span.setAttribute('success', false); the success path in _executeToolCallBody adds span.setAttribute('success', true) after safeSetStatus(span, OK). Mirrors the unconditional `success` attribute on llm_request spans, so backends can use one query for both span types. Add 4 scheduler-level tests asserting the success attribute on: - success path - ToolResult.error path - thrown invocation path - cancellation path |
||
|
顾盼
|
609e05baee
|
feat(tools): add generic worktree support — EnterWorktree/ExitWorktree + Agent isolation (#4073)
* feat(tools): add generic worktree support (Phase A + B of #4056) Adds first-class git worktree as a general-purpose capability: Phase A — User-facing tools - enter_worktree: creates `<projectRoot>/.qwen/worktrees/<slug>` on a `worktree-<slug>` branch and returns the absolute path. Slug auto-generated when omitted; validated against path traversal and disallowed characters. - exit_worktree: keeps or removes the worktree (and its branch). Refuses to remove a worktree with uncommitted tracked changes or untracked files unless `discard_changes: true` is set. Phase B — Agent isolation - Agent tool gains an `isolation: 'worktree'` parameter that provisions a temporary `agent-<7hex>` worktree, prepends a worktree notice to the task prompt, and on completion either removes the worktree (no changes) or preserves it and reports its path/branch in the result. Background and foreground execution paths both wired up; rejected for fork agents. - worktreeCleanup.cleanupStaleAgentWorktrees: fail-closed sweep for ephemeral `agent-<7hex>` worktrees older than 30 days with no tracked changes and no unpushed commits. User-named worktrees are never swept. - buildWorktreeNotice helper for fork subagents (parity with claude-code). Arena compatibility - The existing Arena worktree implementation (GitWorktreeService.setupWorktrees, ArenaManager, agents.arena.worktreeBaseDir) is untouched. Arena uses its own batch APIs and `~/.qwen/arena` base dir; the new general-purpose APIs live alongside under `<projectRoot>/.qwen/worktrees/`. Subagent safety - enter_worktree / exit_worktree are added to EXCLUDED_TOOLS_FOR_SUBAGENTS so a subagent cannot mutate the parent session's worktree state. Refs #4056 Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * test(worktree): use path.join in expected paths so the test passes on Windows The Windows CI run reported `enter-worktree.test.ts` failing because the expected string was hardcoded with `/` while `getUserWorktreesDir()` uses `path.join`, which returns `\\` on Windows. Build the expected path via `path.join` so the platform-correct separator is compared. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(enter-worktree): treat empty name as auto-generate Some models pass `{ "name": "" }` when calling EnterWorktree, because the schema marks `name` as optional and they emit an empty placeholder. The previous validation rejected the empty string with "Worktree name must be a non-empty string", which surprised users running the auto-slug path. Now both `validateToolParams` and `execute` treat `name: ""` as equivalent to `name: undefined` and fall back to the auto-generated `{adj}-{noun}-{4hex}` slug. Explicit invalid slugs (`'../etc'`, `'a/b'`, etc.) are still rejected as before. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): address review findings 1-6 from PR #4073 Six issues raised on the initial review; each addressed with a verifiable guarantee. 1. Real isolation for `agent isolation: 'worktree'` Before: subagent's Config still resolved `getTargetDir()` to the parent project root, so Edit/Write/Read workspace checks and Shell's default cwd silently operated on the parent tree. The cleanup helper then saw a "clean" worktree and removed it — destroying the evidence. After: the worktree is provisioned BEFORE `createApprovalModeOverride`, and the resulting agent Config has `getTargetDir`/`getCwd`/`getWorkingDir` rebound to the worktree path. Relative paths, unqualified shell commands, and glob/grep roots all confine to the worktree. 2. `exit_worktree action='remove'` now prompts in default/auto-edit modes Added `getDefaultPermission()` on the invocation: `'ask'` when action is `remove`, `'allow'` when `keep`. Brings it in line with edit, write_file, and run_shell_command. 3. Force-delete no longer silently destroys unpushed commits `removeUserWorktree` now uses `git branch -d` (refuses unmerged) by default and surfaces `branchPreserved: true` when git refuses. Added `hasUnmergedWorktreeCommits` (checks if branch tip is reachable from any other local branch or remote ref). Both the agent isolation cleanup and `exit_worktree action='remove'` use this check: if the branch has work not covered elsewhere, the worktree+branch are preserved even when `discard_changes: true` is set (there is no `discard_commits` flag — committed work is rarely what `remove` means to discard). 4. Both new tools are now deferred behind ToolSearch `shouldDefer: true` + `searchHint` on both. Verified via openai-logging: `enter_worktree` and `exit_worktree` no longer appear in the function- declaration list sent on every API request. 5. Stale-worktree cleanup is wired in `Config.initialize()` fires `cleanupStaleAgentWorktrees(targetDir)` as a non-awaited startup sweep (skipped in bare mode). Picks up orphaned `agent-<7hex>` worktrees left by crashed runs. 6. Foreground isolation no longer leaks on uncaught throw The foreground try block tracks whether the cleanup helper ran on the success path; the finally block invokes it as a fallback when the try bailed early. Mirrors the background path's pattern. Verification: - Unit tests: 83 passed (16 worktree + 64 existing agent + 3 cleanup) — no regressions. - E2E #1: agent told to write `hello.txt` via RELATIVE path — file landed at `.qwen/worktrees/agent-XXXXXXX/hello.txt`, NOT at the parent root. - E2E #3: created worktree, committed work inside it, called exit_worktree with `discard_changes=true` — refused with clear message; worktree and branch both preserved. - E2E #4: openai-logging confirms worktree tools absent from API tool list (7 tools sent instead of 9). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): address review round 2 findings (1 from tanzhenxin, 7+8 from wenshao) The first round closed the data-loss-class issues. This round addresses follow-ups from a deeper audit: 1. Stale-worktree sweep was inert on common-case repos `cleanupStaleAgentWorktrees` previously ran `git log --branches --not --remotes --oneline` from each worktree's directory — that lists unpushed commits across EVERY local branch, not just the worktree's own branch. On any repo with no remote configured (or with stray unpushed branches), the sweep refused to remove every candidate. Replaced with `service.hasUnmergedWorktreeCommits(slug)` which scopes the check to the worktree branch via `for-each-ref --contains <tip>`. Also added the `branchPreserved` warn log requested in M7 and an `fs.access` shortcut for the empty-worktrees-dir case (M8). 2. `cleanupWorktreeIsolation` and `worktreeIsolation` were inside the inner try (~660 lines from the outer catch). Hoisted both to the top of `execute()` so the outer catch can reap or preserve the worktree when anything between provisioning and the inner try throws (e.g. `createApprovalModeOverride`, agent creation). Closure carries the resolved `repoRoot` so cleanup never has to re-resolve. 3. Background error path discarded the cleanup result. Now captures `formatWorktreeSuffix(...)` and appends it to the registry's failure /cancel message, so users see the preserved path/branch even when the agent crashed before reporting. 4. `cleanupWorktreeIsolation` now treats `result.success === false` as "worktree still on disk" and surfaces it as preserved instead of silently dropping it from the result. 5. Override was incomplete. Several Config methods read `this.targetDir` directly (`getProjectRoot`, `getFileService`, etc.) — own-property getter overrides did not redirect them. Now also shadows `targetDir` and `cwd` as own properties on the agent's Config override, swaps in a `FileDiscoveryService` rooted at the worktree, and rebuilds `WorkspaceContext` to point at the worktree only. Verified end-to-end: shell `pwd > pwd-record.txt` (no directory arg) lands at `.qwen/worktrees/agent-<7hex>/pwd-record.txt`, not the parent root. 6. monorepo subdir issue. Both `enter_worktree` and the agent isolation path now resolve `git rev-parse --show-toplevel` first and anchor `.qwen/worktrees/<slug>` at the repo root. Worktrees created from any subdirectory now end up where the startup sweep can find them. 7. Replaced `git worktree add -B` (silent force-reset of pre-existing branches) with `git worktree add -b` plus an explicit existence check via `git for-each-ref` (NOT `show-ref --quiet`, which simple-git swallows). Pre-existing `worktree-<slug>` branches now trigger a clear error instead of clobbering committed work. 8. First worktree creation in a repo writes `<projectRoot>/.qwen/.gitignore` with `worktrees/` so worktree contents stay out of the parent's `git status`, glob/grep results, and bundle tools. Idempotent: never overwrites an existing file. 9. Logging across the failure paths (`enter_worktree` errors, `agent.ts:failWorktreeProvisioning`, `cleanupWorktreeIsolation`, `hasUnmergedWorktreeCommits` swallowed errors, `cleanupStaleAgentWorktrees`'s `branchPreserved` race). 10. `exit_worktree` no longer suggests `discard_changes: true` when the git status check itself fails — that would be advising the user to bypass a safety check whose precondition is unknown. Now points at the underlying repo problem. 11. `generateAutoSlug` switched from `Math.random()` (4 hex, weak RNG, one-in-65k collision) to `randomBytes` (6 hex, ~16M combinations). Two RNG sources in this file collapsed to one. Pushed back: the TOCTOU swap in `removeUserWorktree` (S6 round 1) is left as-is — `git branch -d` is the real safety, and reordering does not eliminate the window. Windows reserved-name validation (M5 round 2) deferred to a follow-up; the current allowlist already rejects path separators, `..`, leading dot/dash, and the >64-char case. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): use randomInt to silence CodeQL biased-modulo finding CodeQL's `js/biased-cryptographic-random` flagged `randomBytes(4)[i] % ARRAY.length` in `generateAutoSlug`. The math is actually exact for the current word-list lengths (256 % 8 == 0), but the lint rule does not know that — and a future contributor changing the list to a non-power-of-two length would silently introduce bias. Switched the index lookups to `crypto.randomInt(0, length)`, which uses rejection sampling and is uniform by construction. Suffix still uses `randomBytes(3).toString('hex')` since hex encoding is unbiased. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): address review round 3 findings 1-6 from PR #4073 The previous round added `getRepoTopLevel` for `enter_worktree`'s provisioning, but missed three sibling call sites that still used the raw cwd. The double-cleanup race in the foreground path also leaked stale `[worktree preserved]` suffixes on rejected promises. All six findings from the deeper audit are addressed: 1. exit_worktree now resolves through `getRepoTopLevel()` before building its `GitWorktreeService`, mirroring `enter_worktree`. Without this, launching `qwen` from a monorepo subdirectory created the worktree under the repo root but exit_worktree looked under the subdir's `.qwen/worktrees/` and always returned "Worktree not found". Verified end-to-end: enter + exit from `packages/core/` works. 2. agent.ts cleanup helper now nulls `worktreeIsolation` immediately after capturing the closure value. The previous structure could reach the helper twice — once in the foreground try's success path and once in the foreground finally fallback (or once in the inner try and once in the outer catch on a thrown rejection). The second call would `hasWorktreeChanges()` against a directory the first call already removed, fail-closed, and emit a bogus `[worktree preserved: <missing path>]` suffix. 3. Config.initialize's startup sweep now resolves `getRepoTopLevel()` before invoking `cleanupStaleAgentWorktrees`. Without this, every subdir launch scanned a non-existent `<subdir>/.qwen/worktrees/` and the 30-day expiry sweep was permanently a no-op. 4. agent.ts's `buildWorktreeNotice` now passes `worktreeIsolation.repoRoot` as `parentCwd` instead of `this.config.getTargetDir()`. The notice's path-translation guidance (≈ "translate paths from <parent> to <worktree>") would otherwise misdirect the subagent in a monorepo subdir launch. 5. Removed dead method `GitWorktreeService.listUserWorktrees`. It had no callers anywhere in the codebase and used `execSync` in a loop (would have blocked the event loop if anyone wired it up). 6. `localBranchExists` no longer swallows git failures silently. The defensive `false` default is preserved (so `git worktree add -b` itself surfaces the conflict if the check missed an existing branch), but the catch now logs via `debugLogger.warn` so disk-full / permission / ref-store-corruption cases are visible in debug output instead of being invisible. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): address review round 4 findings (data-loss + visibility) Seven actionable findings from a deeper audit, all closed: 1. User worktree slugs could collide with ephemeral-agent shape `validateUserWorktreeSlug` did not reject names starting with `agent-`, so a user-named `agent-1234567` matched the cleanup regex `/^agent-[0-9a-f]{7}$/` and would be silently swept after 30 days along with whatever work was in it. Now reserved — clear error message points users at the cause. 2. Slug producer and consumer were string-coupled across files `agent.ts` hardcoded `agent-${hex(7)}` and `worktreeCleanup.ts` independently hardcoded `/^agent-[0-9a-f]{7}$/`. Future change to hex length on one side would silently break the other. Lifted `AGENT_WORKTREE_PREFIX`, `AGENT_WORKTREE_HEX_LENGTH`, `AGENT_WORKTREE_SLUG_PATTERN`, and `generateAgentWorktreeSlug()` to `gitWorktreeService.ts`; both call sites import them. 3. Startup sweep was invisible at default log level Fire-and-forget sweep used `debug` for errors and discarded the success count. A leak-chasing operator had no log breadcrumb. Errors promoted to `warn`; successful removals (count > 0) logged at `info`. 4. `getRepoTopLevel()` silent catch Returned `null` on any git failure with no log. Combined with `?? cwd` fallback in callers, a flaky git would have made worktree creators and the startup sweep disagree silently about which dir to use. Now logs the underlying error. 5. `hasTrackedChanges()` silent catch Cleanup's fail-closed `return true` had no log. Couldn't tell "has real changes — leave alone" from "git index unreadable — repo may be corrupt". Now logs. 6. `cleanupWorktreeIsolation` claimed `preservedPath` for a removed dir When `removeUserWorktree` returns `{ success: true, branchPreserved: true }` it has already deleted the directory and failed only on `git branch -d`. The helper still reported the (now non-existent) path as preserved. Now returns only `preservedBranch` for that case; `formatWorktreeSuffix` emits a distinct message instructing recovery via `git worktree add <new-path> <branch>`. 7. `removeUserWorktree` swallowed branch-delete failures Both `-d` and `-D` catch blocks were empty. Locked refs, perms, disk full all looked identical to "unmerged commits". Both now `debugLogger.warn` with the underlying error. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * refactor(worktree): self-review pass — reuse, parallelism, dead code Self-review caught a handful of issues across three categories: Reuse: - `pathExists` in the new code now uses the existing `fileExists` from `utils/fileUtils.ts` instead of duplicating an `fs.access` wrapper. - `worktree-` branch prefix was string-literalled in five places. Added `WORKTREE_BRANCH_PREFIX` and `worktreeBranchForSlug(slug)` exports in `gitWorktreeService.ts`; updated `gitWorktreeService.ts`, `worktreeCleanup.ts`, and `exit-worktree.ts` to use them. Future prefix changes are a single edit. Efficiency: - `Config.initialize` used two `await import(...)` calls inside the startup-sweep IIFE, paying that cost on every CLI start. Switched to static imports at the top of `config.ts` — the modules are tiny and the dynamic indirection bought nothing. - `cleanupWorktreeIsolation` in `agent.ts` ran `hasWorktreeChanges` and `hasUnmergedWorktreeCommits` sequentially. They have no data dependency on each other and each spawns its own `git` invocation; `Promise.all` halves the cleanup wall-clock on the common path. Same fix in `worktreeCleanup.ts`'s per-entry loop. - `ensureWorktreesGitignored` used `fs.access` then `fs.writeFile`, a TOCTOU race when two agent invocations created worktrees concurrently (both could pass the `access` check and the second would clobber the first's `.gitignore`). Now writes with `flag: 'wx'` and treats `EEXIST` as the no-op case — atomic in one syscall. Quality: - Dropped the `worktreeCleanupRan` boolean in the foreground execution path. `cleanupWorktreeIsolation` already nulls its closure variable at the top of every call (see the comment at its definition), so re-entries are no-ops. The boolean and its tracking were dead weight that obscured the real guard. - Trimmed the Phase-2 override comment block to drop the WHAT-stating enumerations (items 3 and 4 just narrated the lines below) and removed a navigation comment about hoisted helpers — the helpers are visible at the top of the same method. 84 unit tests pass; typecheck clean. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): address review round 5 — design-doc commitments + correctness Five critical findings + four suggestions, all closed. Critical: 1. Wrong base branch for agent isolation. `createUserWorktree(slug)` with no `baseBranch` arg fell back to `getCurrentBranch()` on the **main** working tree, returning `main` regardless of which branch the user was actually on. A subagent invoked from `feature-x` would silently start from `main` and produce diffs against the wrong baseline. `enter_worktree` had the same bug. Both now resolve the parent's current branch first and pass it explicitly. Verified end-to-end: `git checkout feature-x` → `enter_worktree` → worktree HEAD includes the feature-x commit. 2. `countWorktreeChanges` (used by `exit_worktree`'s dirty-state guard) missed `status.conflicted[]`. In simple-git that array is mutually exclusive with the staged/modified/etc. arrays, so a worktree mid-merge with only conflicts looked `{tracked: 0, untracked: 0}` to the guard and `action='remove'` would proceed without `discard_changes: true`. Added `+ status.conflicted.length`. 3. `exit_worktree` had no session-ownership check, contradicting the design doc's "only operates on worktrees created by THIS session". In yolo mode a prompt injection could enumerate `.qwen/worktrees/` and pass any name to drop another session's work. Now: `enter_worktree` and agent isolation write a `.qwen-session` marker into the worktree at provisioning time; `exit_worktree action='remove'` reads it and refuses if it does not match the current `Config.getSessionId()`. Worktrees from before this guard (no marker file) are treated as "owner unknown" — allowed with a warn log so the change is observable. 4. `enter_worktree` did not refuse nested invocations from inside an existing worktree, contradicting the design doc. Now rejects any cwd containing `.qwen/worktrees/` as a path component, with a clear "Already inside a git worktree…" message. Verified: enter from inside a worktree returns is_error with that text. 6. `hasTrackedChanges` (cleanup sweep) had the same `conflicted[]` gap. Rewrote to use raw `git status --porcelain --untracked-files=no` which lists every tracked change including `UU` conflict markers in a single git call and explicitly skips the untracked walk (the prior comment claimed to skip it, but `status()` always does the scan). Suggestion: 7. `buildWorktreeNotice` now receives the parent agent's actual `getTargetDir()` again (was switched to `repoRoot` in round 3 on a different reviewer's suggestion; round-5 caught that the model's inherited paths reference the parent's cwd, not necessarily the repo root, so the prior behaviour was correct). 8. Startup sweep now does `fs.access(<targetDir>/.qwen/worktrees)` *before* importing GitWorktreeService and spawning `git rev-parse --show-toplevel`. The git probe is reserved for users who actually have a worktrees directory locally — 99% of users pay only one syscall on startup. 9. Tests: - New `exit-worktree.test.ts` covers metadata, validation, `getDefaultPermission` (ask vs allow), and getDescription. - `agent.test.ts` adds three `validateToolParams` cases for the `isolation` parameter (accepted with subagent_type, rejected without, rejected for non-"worktree" values). - `enter-worktree.test.ts` adds round-trip tests for `writeWorktreeSessionMarker` / `readWorktreeSessionMarker` plus a `worktreeBranchForSlug` sanity check. - Total: 101 tests pass (was 86 → +15). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(test): drop unused @ts-expect-error in exit-worktree.test.ts Empty string `''` is a valid `string` type, so the @ts-expect-error directive on `validateToolParams({ name: '', action: 'keep' })` did nothing — TypeScript correctly accepted the line, and `tsc --build` in CI reported TS2578 ("Unused '@ts-expect-error' directive"). The runtime assertion already covers the case; the directive was leftover from an earlier draft. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(test): use importActual in ArenaManager mock to preserve new exports The Arena test mocks `gitWorktreeService.js` with a factory that returns only `{ GitWorktreeService }`. PR #4073 added several other exports to that module (`AGENT_WORKTREE_SLUG_PATTERN`, `WORKTREE_BRANCH_PREFIX`, `worktreeBranchForSlug`, `generateAgentWorktreeSlug`, `writeWorktreeSessionMarker`, `readWorktreeSessionMarker`, `WORKTREE_SESSION_FILE`). Other modules in the dep graph reach the mocked surface — most notably `worktreeCleanup.ts` imports `AGENT_WORKTREE_SLUG_PATTERN` and `worktreeBranchForSlug`, and now reaches the mock via the static `config.ts` → `worktreeCleanup.ts` import chain added in the self-review pass. The Arena test failed at module-load with: Caused by: Error: [vitest] No "AGENT_WORKTREE_SLUG_PATTERN" export is defined on the "../../services/gitWorktreeService.js" mock. Did you forget to return it from "vi.mock"? Use `importOriginal` to capture every real export, spread it into the return object, and only replace `GitWorktreeService` (the class the test actually needs to mock). The class-level mock keeps its existing static-method shims. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): address review round 6 (5 critical + 6 suggestions) The biggest item — #1 — is a self-inflicted regression from round 5: the new agent- prefix reservation in `validateUserWorktreeSlug` rejected EVERY slug that `generateAgentWorktreeSlug` produces, since that helper emits exactly `agent-<7hex>`. Net effect: every `AgentTool isolation: 'worktree'` invocation failed at validation. The reservation now allows the canonical pattern through (everything the helper can produce) and only rejects user-chosen `agent-*` names that don't match it. Added a round-trip regression guard: 50 `generateAgentWorktreeSlug()` outputs are fed back through `validateUserWorktreeSlug` and must all pass. Other critical fixes: 2. `hasWorktreeChanges` (used by agent isolation cleanup) was the one remaining caller relying solely on `status.isClean()`. Defensive `|| status.conflicted.length > 0` so a future simple-git bookkeeping change can't let a mid-merge worktree appear clean and get auto-deleted. 3. `readWorktreeSessionMarker` swallowed every I/O error as "marker missing", which let a disk error / EACCES silently bypass the session-ownership guard. ENOENT is still treated as missing (legitimate); every other code now logs. 4. `exit_worktree` `fs.stat` catch was the same shape — every error collapsed to "Worktree not found". ENOENT → not found; everything else logs and returns a distinct "cannot access" error. 5. `cleanupStaleAgentWorktrees` `fs.stat` catch was again the same. ENOENT → silently skip (entry vanished between readdir and stat); everything else logs. Suggestions: 6. Startup sweep fast-bail was running BEFORE resolving the repo top-level. For monorepo subdir launches, `targetDir/.qwen/worktrees` never exists and the sweep early-returned — permanently a no-op. Now resolves the root first, then fast-bails against the resolved `<root>/.qwen/worktrees`. Also logs the skip case so operators can tell "skipped" from "ran, found nothing". 7. `.qwen-session` marker was visible to `git add -A` inside the worktree. Now writes a `.git/info/exclude` rule (resolved via `git rev-parse --git-dir`, since worktree `.git` is a file pointing at the parent repo's `.git/worktrees/<name>/`). Best-effort: failure to write the rule does not abort provisioning. 8. Agent isolation now refuses to provision when the parent's cwd is already inside a worktree — same regex guard as `enter_worktree`. 9. `exit_worktree`'s wrapper around `hasUnmergedWorktreeCommits` now logs at the call site so the chain (caller → reason it asked → underlying git error) is complete in operator logs. 10. Sweep now logs unconditionally at `info`. Three distinct messages: "skipped (no worktrees dir)", "ran, nothing to remove", "removed N". Tests: 11. New `execute()` coverage: • exit-worktree: session-ownership refusal, keep happy path, legacy/no-marker fallthrough with warn log, missing-worktree error, unmerged-commits guard with `discard_changes: true`, `writeWorktreeSessionMarker` round-trip. • enter-worktree: nested-guard rejection, non-git-repo error. These spin up real temp git repos (no filesystem mocking) and drive the actual tool invocation pipeline. Total: 135 tests pass (was 101 → +34). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * refactor(worktree): demote noise startup-sweep logs to debug Self-review pass applying the round-6 review-triage framework (filter #5: "If a log only fires on the happy path, it's noise.") to my own round-6 changes: - "Stale worktree sweep skipped: <dir> does not exist" — fires on every CLI start for ~99% of users who never use worktrees. - "Stale worktree sweep ran under <root>: nothing to remove" — fires on every CLI start for users who have any worktrees but no stale ones at the moment. Both are happy-path noise at `info`. Demoted to `debug` so an operator can opt in via `--debug` when they want to confirm the sweep is wired up, but normal output stays clean. Only the actually-actionable case ("removed N worktrees") stays at `info` — that's the signal someone chasing a worktree leak would grep for. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(worktree): close AUTO_EDIT bypass + parent-dirty stale-code hazard Round-7 review caught two correctness gaps: 1. exit_worktree action='remove' was still auto-approved in AUTO_EDIT `getDefaultPermission` returning 'ask' is necessary but not sufficient. `permissionFlow.isAutoEditApproved` auto-approves any tool whose `confirmationDetails.type` is 'edit' OR 'info', and `BaseToolInvocation` returns 'info' by default. So a session in AUTO_EDIT could silently destroy a worktree (with branch deletion) without a confirmation prompt — the data-loss path the round-1 `'ask'` switch was meant to close. Now overrides `getConfirmationDetails` to return `type: 'exec'` for action=remove, which keeps the prompt in AUTO_EDIT. The `keep` action still falls through to the base info-type since it is non-destructive. Regression-guard test asserts the type is 'exec' (not 'info') for remove and that the command field describes both the worktree-remove and branch-delete operations. 2. Agent isolation worktrees ran against parent's HEAD, not its working tree `git worktree add -b <branch> <path> <base>` only checks out the base ref's tip — uncommitted edits in the parent's working tree do NOT propagate. The "edit code → ask review/test agent before committing" workflow silently ran the subagent against the pre-edit HEAD and returned results that looked authoritative but reflected stale code. Reviewer offered two options: overlay parent's dirty state à la Arena (~50 LOC, edge cases), or refuse isolation when parent is dirty (~10 LOC, clear UX). Chose the latter for Phase B scope — simpler, decisive, and matches the design-doc's explicit commitment that dirty-state overlay is Arena-specific. Users can commit/stash before re-invoking agent isolation; overlay can be a follow-up if users complain about the friction. Fail-closed on the dirty-check itself (assume dirty rather than silently launch on a possibly-stale tree). Test exercises both "dirty parent → guard fires" and "clean parent → guard passes" against real temp git repos. 139 unit tests pass (was 135, +4 regression guards). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com> |
||
|
顾盼
|
c512427f93
|
feat(core): strip inline media before chat compaction summary (#4101)
* feat(core): strip inline media before chat compaction summary
Compaction's side-query previously shipped historyToCompress verbatim.
Two related issues degraded summary quality and accuracy:
- Inline image / document bytes (from MCP tool results) leaked into the
summary model's prompt where they could not be interpreted and merely
inflated payload.
- findCompressSplitPoint apportioned chars via JSON.stringify(content),
so a single 1 MB base64 image looked like ~350K tokens and biased
the split point. Real Qwen-VL token cost is at most a few thousand.
This change adds a new compactionInputSlimming module that replaces
inlineData / fileData parts with short [image: <mime>] / [document:
<mime>] placeholders before the side-query, leaving live history
unchanged. The same constant feeds estimateContentChars so the
split-point algorithm sees the budget the summary model actually
consumes downstream. Microcompact is also extended to clear stale
inline images alongside old tool results.
A previous draft of the design also externalized large pastes to a
content-addressable on-disk cache, but it was withdrawn after surveying
claude-code's 2026-03 to 2026-05 releases - upstream consensus is to
keep user input visible to the model and amortize cost via prompt
caching rather than externalize. See the Out-of-scope section of the
design doc for the full rationale.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): recurse into functionResponse.parts when stripping media
E2E exposed that `read_file` (and any tool that surfaces an image)
wraps the result in `functionResponse.parts` via
`coreToolScheduler.createFunctionResponsePart`. The slimming module
only walked top-level `part.inlineData` / `part.fileData`, so the
nested base64 bytes leaked into the compaction side-query payload.
The previous design doc incorrectly claimed that no recursive walk
was needed.
Three changes:
- `slimCompactionInput.transformPart` recurses into the nested
`functionResponse.parts` array and replaces each entry via the
same image/document placeholder logic.
- `estimatePartChars` walks the nested array too, so the split-point
algorithm doesn't fall back to `JSON.stringify` and over-count the
base64 bytes.
- `microcompactHistory` drops `functionResponse.parts` when clearing
an old tool result; the previous spread of `...part.functionResponse`
silently carried the original media through.
New unit tests cover (a) nested image / document stripping, (b) the
estimator no longer being skewed by nested base64. The previously
failing E2E now PASSES: side-query payload contains zero `data:image/`
occurrences, zero long base64 runs, and exactly one
`[image: image/png]` placeholder.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): address review findings on compaction image stripping
Addresses 8 valid findings from PR review:
- [Critical] estimatePartTokens now handles `fileData` parts (both
top-level and nested under functionResponse.parts). Without this,
microcompact's `tokensSaved === 0` short-circuit silently discarded
every fileData clear.
- estimatePartTokens for binary parts now uses a fixed
MEDIA_PART_TOKEN_ESTIMATE constant (1,600) instead of base64-length
divided by 4. The old formula billed a 1 MB image as ~250K tokens
rather than its actual ~1,280 visual tokens on Qwen-VL, inflating
the saved-token metric by orders of magnitude.
- mimeType values from MCP tool servers are now run through
sanitizeMimeForPlaceholder before being embedded in `[image: …]` /
`[document: …]` placeholders. An adversarial server could otherwise
craft `image/png]\n\n[SYSTEM: …` and inject instructions into the
summary side-query.
- collectCompactablePartRefs now recognizes a third 'nested-media'
kind: functionResponse parts from non-compactable tools (e.g. MCP
screenshots whose names aren't in COMPACTABLE_TOOLS) that carry
images on functionResponse.parts. The nested media is dropped while
the tool's text output is preserved. Previously such media
accumulated forever in live history.
- keepRecent budgets are now per-kind (tool / media / nested-media).
Setting `toolResultsNumToKeep: 1` keeps 1 of each kind rather than 1
entry total across the merged list — matches the natural reading of
the setting name.
- findCompressSplitPoint's `precomputedCharCounts` fallback path is
now documented as test-only; production callers MUST pass the
precomputed array.
- The text-based branch of isAlreadyCleared is gone: with the new
nested-media handling (drops `parts`) and existing media handling
(replaces with `{ text: … }` that is no longer collected) it was
unreachable.
- OpenAI converter (createToolMessage) now passes text parts inside
functionResponse.parts through as text content. The slimmer writes
`{ text: '[image: image/png]' }` placeholders into the nested array;
without this fix the converter dropped them when serializing to the
OpenAI wire format, leaving the summary model with empty tool
responses instead of the placeholder.
Two findings deferred with rationale (see design doc Open Questions):
MIN_COMPRESSION_FRACTION still uses pre-slim counts (acceptable —
"user shared an image" is itself worth summarizing); SlimResult is not
re-exported (round-3 simplify decided to keep core's public surface
minimal).
E2E re-verified end-to-end: side-query payload contains 0 data:image/
occurrences, 0 long base64 runs, and 1 `[image: image/png]` placeholder
in the expected position. 185/185 collocated unit tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* chore(core): tidy compaction slimming after self-review
Three small polishes from a follow-up code review pass:
- `stripNestedMedia` no longer re-casts its return value: after
destructuring `parts` out of the widened input type, TypeScript
infers the original `FunctionResponse` shape without help.
- `isAlreadyCleared` shed a 10-line comment block — the body is now
one line, so one descriptive line above it is enough.
- OpenAI converter's nested-part text check switched from
`(part as { text?: unknown }).text` to
`'text' in part && typeof part.text === 'string'`, dropping the
cast and letting `in` narrow the type.
No behavior change. 185/185 unit tests still pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): wire slim stats to debug log; split MicrocompactMeta tools vs media
Addresses two follow-up review suggestions:
- `slimCompactionInput` returned `stats.imagesStripped` and
`stats.documentsStripped` but the orchestrator never consumed them.
Now logged at debug level whenever non-zero so operators can confirm
the slimming pipeline actually fires on image-heavy compactions.
- `MicrocompactMeta.toolsCleared` lost meaning after the recent
refactor: it had grown to count both tool-result clears AND
inline-media / nested-media clears. Renamed:
- `toolsCleared` → only `tool`-kind clears (compactable tool output)
- `mediaCleared` → `media` + `nested-media` clears (new)
- `toolsKept` / `mediaKept` mirror the split, replacing the prior
`toolsKept` that was actually a combined count.
The single non-test consumer (`client.ts` debug log) updated to use
both fields.
185/185 unit tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
||
pomelo
|
faf646b081
|
docs(auth): add custom API key wizard PRD (#3583)
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> |
||
|
pomelo
|
7e11428545
|
refactor(cli): remove legacy qwen auth CLI subcommand, redirect to /auth TUI dialog (#3959)
The `qwen auth` CLI subcommand (with subcommands like qwen-oauth, coding-plan, api-key, openrouter, status) has been superseded by the richer /auth TUI dialog introduced in the provider-first auth registry (#3864). Running `qwen auth` now prints a deprecation notice pointing users to the /auth TUI dialog (interactive), env vars (CI/headless), or /doctor (status check). Changes: - Replace auth.ts with a stub that prints a removal notice and exits - Delete handler.ts (734 lines), interactiveSelector.ts, and their tests (interactiveSelector.test.ts, openrouter.test.ts, status.test.ts) - Update /auth slash command to handle non-interactive/ACP modes gracefully - Enrich /doctor auth check with provider-aware diagnostics using findProviderByCredentials - Mark `auth` as a subcommand that handles its own exit in config.ts Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> |
||
|
顾盼
|
b55b52543a
|
feat(cli): improve slash command discovery (#3736)
* feat(cli): improve slash command discovery Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * test(cli): update input prompt completion expectations Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(cli): address review feedback on slash command phase 3 - Fix getBestSlashCommandMatch sort order: completionPriority first, recentOrder second — consistent with compareRankedCommandMatches in useSlashCompletion.ts so ghost text and dropdown agree on best match - Fix findSlashCommandTokens to index altNames into commandMap so alias tokens (e.g. /usage for /stats) are highlighted as valid instead of being marked invalid - Fix getRecentScore decay formula: 10 * Math.max(0, 1 - ageMs / RECENT_DECAY_MS) so the recent boost truly decays to 0 within the 10-minute window named by RECENT_DECAY_MS - Fix Help.tsx CommandsHelp scroll indicator to show command count range (e.g. 1-12/49) instead of raw line count (18/108), which was confusing because each command expands into 2-3 render lines - Fix Help.tsx CommandLine key prop: use stable type:text:index key instead of scrollOffset-index to avoid remounting every line on scroll - Internationalize Help.tsx tab labels via t() instead of hardcoded English strings - Add Tab/Shift+Tab to switch tabs hint in Help footer alongside Esc - Add commandMetadata.test.ts with full branch coverage for all 6 exported functions (getCommandSourceBadge, getCommandSourceGroup, formatSupportedModes, getCommandDisplayName, getCommandSubcommandNames, formatCommandSourceLabel) - Add direct unit tests for getBestSlashCommandMatch covering: null on empty input, null on no match, null for non-modelInvocable commands, completionPriority ordering, recentCommands tie-breaking, argumentHint return path, exact-match exclusion without hint, inclusion with hint - Update Session.test.ts expectation for sendAvailableCommandsUpdate to include _meta field that was added in this PR * test(cli): add missing test coverage for slash completion - Add test: midInputGhostText is null when only non-modelInvocable commands match - Add test: recentCommands boosts non-root prefix suggestions via recentScore Both tests address coverage gaps identified in code review. --------- Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> |
||
|
顾盼
|
f5af7fbf95
|
feat(memory): add autoSkill background project skill extraction (#3673)
* feat(memory): add autoSkill background project skill extraction
* fix(test): add missing mock methods for autoSkill (getAutoSkillEnabled, recordCompletedToolCall, consumePendingMemoryTaskPromises)
* fix(test): fix cross-platform path comparison in skillReviewNudge integration test
* fix(autoSkill): address critical review comments
- Fix merged_with_extract silent drop: remove broken merge optimization
in scheduleSkillReview(). When an extract task is pending/running,
skill review is now scheduled independently instead of recording
metadata that no production code ever reads.
- Fix SKILL_MANAGE blocked from skill review agent: prepareTools() now
only enforces the recursion guard (AGENT tool) when the agent has an
explicit tools list. Wildcard/inherit subagents still get the full
EXCLUDED_TOOLS_FOR_SUBAGENTS filter, preventing task subagents from
calling skill_manage. The dedicated skill review agent can now receive
the skill_manage tool it requires.
- Update manager.test.ts: replace merged_with_extract tests with
concurrent-extract independent-scheduling tests.
- Update skill-manage.test.ts: clarify test description to reflect
wildcard-only exclusion semantics.
* fix(autoSkill): reject symlink traversal in skill_manage path guard
assertProjectSkillPath() uses path.resolve() which is purely lexical and
does not dereference symlinks. If any path component inside .qwen/skills/
is a symlink pointing outside the project, fs.writeFile/readFile/rm would
follow the link and mutate files outside the advertised write boundary.
Add assertRealProjectSkillPath() (async) in skill-paths.ts that:
- Resolves the real path of the skills root via fs.realpath()
- Walks up from targetPath to find the nearest existing ancestor
- Resolves that ancestor to its real filesystem path
- Rejects if the real path falls outside the real skills root
skill-manage.ts execute() now calls both the cheap lexical check (fast
fail for obviously wrong paths) and the async real-path check before any
fs.writeFile / fs.rm mutation.
Add three symlink-specific tests in skill-paths.test.ts covering:
- Legitimate path accepted
- Symlinked directory pointing outside skills root rejected
- Skills root itself being a symlink (safe target) accepted
* refactor(autoSkill): remove skill_manage tool, use path-based skill write detection
Address reviewer feedback: instead of keeping skill_manage as the sole
write gate (which still had symlink bypass risk via generic tools), remove
the dedicated tool entirely and replace with a two-layer protection:
1. skillsModifiedInSession (client.ts): detects writes to .qwen/skills/
by inspecting the file_path arg of every completed tool call, replacing
the fragile historyCallsSkillManage() history scan.
2. hasAutoSkillSource + evaluateScopedDecision (skillReviewAgentPlanner.ts):
the review agent's permission sandbox now verifies BOTH that the target
path is inside the skills directory AND that the existing file already
contains 'source: auto-skill' in its frontmatter before allowing edits,
preventing the agent from overwriting user-managed skills.
Changes:
- Delete skill-manage.ts and skill-manage.test.ts
- Remove SKILL_MANAGE from ToolNames, ToolDisplayNames, config registerLazy,
agent-core EXCLUDED_TOOLS comment, and agent.ts comment
- Replace historyCallsSkillManage() with skillsModified: boolean param in
scheduleSkillReview; skip reason renamed skills_modified_in_session
- recordCompletedToolCall(name, filePath?) detects .qwen/skills/ writes;
CLI layers pass file_path arg from tool call request
- Fix buildTaskPrompt frontmatter template to use top-level source: auto-skill
- Update skill-paths.ts error messages to remove skill_manage references
- Update all unit/integration tests accordingly
* fix(autoSkill): deduplicate concurrent skill-review tasks per projectRoot
scheduleSkillReview() was launching a new background task every time the
threshold was reached for the same project, with no guard against multiple
in-flight reviews running concurrently.
Fix: add skillReviewInFlightByProject Map that tracks the taskId of any
running review per projectRoot. A second call while one is in-flight returns
{ status: 'skipped', skippedReason: 'already_running', taskId: <existing> }.
The map entry is cleared in a finally block inside runSkillReview() so the
next session can schedule a fresh review after the current one completes.
Also extend SkillReviewScheduleResult.skippedReason union to include
'already_running', and add a unit test covering the full lifecycle:
first call schedules, second call is skipped with existing taskId, and a
third call after completion schedules a new task.
* fix(autoSkill): address all critical review comments
1. hasAutoSkillSource: narrow catch to ENOENT only (EISDIR/EACCES etc.
return false to deny); tighten frontmatter regex to match opening block only.
2. evaluateScopedDecision: add explicit allow for READ_FILE and LS so they
don't fall to 'default' which the base PermissionManager might widen;
EDIT/WRITE_FILE now call assertRealProjectSkillPath() (async realpath guard)
in addition to the lexical check, closing the symlink traversal hole.
3. isScopedTool / getScopedDenyRule: cover READ_FILE and LS so hasRelevantRules
returns true and findMatchingDenyRule is correctly consulted for them.
4. recordCompletedToolCall (client.ts): broaden tool name set to match
WRITE_TOOL_NAMES in manager.ts (write_file, edit, replace, create_file) and
inspect all three arg keys (file_path, path, target_file). Signature changed
from (name, filePath?) to (name, args?) to carry all args through.
5. client.ts hardcoded literals: replace threshold/maxTurns/timeoutMs with the
named constants AUTO_SKILL_THRESHOLD / DEFAULT_AUTO_SKILL_MAX_TURNS /
DEFAULT_AUTO_SKILL_TIMEOUT_MS imported from manager.ts and
skillReviewAgentPlanner.ts.
6. toolCallCount / skillsModifiedInSession reset: only reset when skill review
is actually scheduled (status === 'scheduled'), not every turn, so the
counter correctly accumulates across turns within a session as per design doc.
7. runSkillReview (manager.ts): rethrow after marking record failed, consistent
with runExtract behavior.
8. skillReviewNudge.integration.test.ts test 5: rewrite to reflect the
in-flight dedup contract (second same-project call returns already_running
with existing taskId; third call after completion gets a new task). Add
vi.mock for runSkillReviewByAgent so the test does not need a full Config.
* fix(autoSkill): address all review comments
- skill-paths: detect dangling symlinks with lstat before treating ENOENT as safe
- skill-paths: fix isProjectSkillPath relative path resolution to use projectRoot
- skillReviewAgentPlanner: restrict READ_FILE/LS to project root only
- skillReviewAgentPlanner: remove SHELL tool from review agent tool list
- skillReviewAgentPlanner: add path import; remove unused shell imports
- skillReviewAgentPlanner: add comment for buildAgentHistory trailing user message
- client: fix runManagedAutoMemoryBackgroundTasks gate widening
- client: fix skillsModifiedInSession deadlock
- client: add .catch() to skill review promise
- client: hoist SKILL_WRITE_TOOL_NAMES to module-level ReadonlySet
- agent-core: use full EXCLUDED_TOOLS_FOR_SUBAGENTS for explicit tool list subagents
- manager: extend notify() signature to accept 'skill-review' taskType
- config: fix JSDoc default value comment (false, not true)
* fix(autoSkill): address second round review comments
- client: reset toolCallCount when scheduleSkillReview returns already_running
and count >= threshold, preventing immediate cascade after in-flight review
- client.test: add autoSkill branch tests (scheduled/already_running/skills_modified)
- client.test: add full recordCompletedToolCall unit tests (skillsModifiedInSession,
toolCallCount increment, skill path detection for write_file/edit/read_file)
- client.test: add scheduleSkillReview mock to mockMemoryManager
- nonInteractiveCli.test: add assertions for recordCompletedToolCall and
consumePendingMemoryTaskPromises in tool-call integration test
|
||
|
pomelo
|
997796f532
|
refactor(cli): provider-first auth registry with unified install pipeline (#3864)
* fix(cli): refresh static header on model switch
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* feat(cli): simplify api key provider registry
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(cli): split Alibaba auth providers
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* polish(cli): refine auth provider onboarding
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(cli): update OpenRouter free defaults
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(cli): restrict token plan models
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* chore(cli): remove unused third-party providers
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* feat(cli): add regional third-party providers
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(cli): simplify api key provider endpoints
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(cli): split auth dialog flows
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(cli): unify auth around declarative provider config
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Introduce ProviderConfig abstraction (providerConfig.ts) and a central provider registry (allProviders.ts), replacing the per-flow UI components (AlibabaModelStudioFlow, CustomProviderFlow, OAuthFlow, ThirdPartyProvidersFlow, etc.) with unified ProviderSetupSteps and useProviderSetupFlow.
Key changes:
- Remove setupMethods/apiKey/ directory entirely
- Collapse flow-specific hooks/components into a single generic provider setup flow
- Simplify each provider file to export only a ProviderConfig descriptor
- Add alibabaStandard provider alongside codingPlan/tokenPlan
- Move all baseUrl resolution, install plan building, and settings writing into providerConfig
- Update useAuth, AuthDialog, command handler, and upstream consumers to use the new registry
* refactor(cli): simplify provider setup input flow
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(cli): remove toLlmProvider and legacy auth wrappers
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(cli): flatten auth flow files and simplify ProviderSetupSteps props
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* feat(cli): prefill API key from existing env settings in provider setup flow
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(cli): correct third-party provider context windows
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(cli): harden provider auth setup
* feat(cli): support provider modality and context settings
* feat: eable modelsEditable for coding plan
* refactor(cli): auto-derive provider metadata key and state
Move metadataKey and getProviderState from per-provider config to
auto-derived helpers (resolveMetadataKey, resolveProviderState) in
providerConfig.ts. This centralizes version tracking logic and reduces
boilerplate in individual provider definitions.
Add useProviderUpdates hook that detects model template changes across
all version-tracked providers and surfaces update/ignore choices.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Closes: OSS-1730, OSS-1729
* refactor(cli): namespace provider metadata under providerMetadata key
Introduce PROVIDER_METADATA_NS ('providerMetadata') to avoid top-level
settings key collisions. Provider metadata now lives under
e.g. providerMetadata.coding-plan.version instead of codingPlan.version.
Add migration logic (migrateProviderMetadata) to automatically move
legacy top-level keys (codingPlan, tokenPlan) into the new namespace
on first run.
Update auth handler, useProviderUpdates hook, and all related tests
to use the new namespace structure.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
[skip ci]
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(cli): polish ProviderUpdatePrompt styling and test coverage [skip ci]
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(auth): simplify auth flows around provider abstraction [skip ci]
- Rewrite motivation.md to document provider-centric architecture
- Remove Alibaba Standard API Key and Coding Plan UI flows from handler
- Update status tests to use providerMetadata instead of codingPlan settings
- Streamline API key auth to show docs link only
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(auth): update provider models and refine auth infrastructure
- Bump model versions (qwen3.6-plus, glm-5.1) and add deepseek-v4-pro/flash
with modalities to Alibaba Standard provider
- Reorder DeepSeek models, add thinking+image/video modalities to v4-pro,
fix v4-flash context window
- Enhance auth tests with provider metadata setValue assertions
- Switch env key generation from hash-based to URL-based with
trailing-slash normalization
- Remove deprecated codingPlan section from settings schema
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(i18n): add missing zh-TW translations for token plan and subscription providers
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* refactor(auth): improve provider install error recovery and AuthDialog state init
- Restore settings from backup on provider install plan failure
- Fix AuthDialog mainIndex state to null (was 0), preventing stale selection
- Remove ownsModel from customProvider; fall back to id-based filtering
- Change provider migration log from console.error to console.log
- Add sync reminder comments between CLI and VSCode subscription models
- Expand handleApiKeyAuth JSDoc explaining its role as lightweight fallback
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
* fix(auth): i18n for step labels, lazy preview JSON, and accurate header label
- Wrap getStepLabel() strings and PROTOCOL_ITEMS in t() for i18n
- Only compute previewJson when on the review step
- Return matched provider's own label in getAuthDisplayType instead
of hardcoding CODING_PLAN for all managed providers
* fix(auth): address round-3 review blockers
- Fix CI: add missing useProviderUpdates mock in AppContainer.test.tsx
that caused TypeError breaking React effects (title/height tests)
- Fix half-rollback: snapshot settings + modelProviders before install,
restore in-memory state (not just disk) on refreshAuth failure
- Fix .orig backup reuse: always create fresh backup (overwrite stale),
cleanup on success, unlink after restore to prevent data loss
- Fix cross-package key consistency: VS Code settingsWriter now writes
to providerMetadata namespace matching CLI's new structure
- Fix validateApiKey: remove baseUrl guard so sk-sp- prefix check
applies to both China and Global Coding Plan endpoints
* fix(cli): stabilize AuthDialog tests for slower CI environments
Increase vi.waitFor timeouts from default 1000ms to 5000ms and replace
unreliable fixed-delay waits with proper render-completion assertions,
preventing flaky failures on Linux/Windows CI runners with Node 22/24.
* fix(core): use id+baseUrl composite key for model identity
Custom provider installs previously used model id alone to determine
ownership, causing the second install to remove the first backend's
model entry when both expose the same model id (e.g. gpt-4o) with
different baseUrls. Use id+baseUrl as the composite identity key
throughout the model registry, ModelDialog, and modelsConfig to
prevent cross-provider model collisions.
* fix(cli): update ModelDialog tests for composite-key model identity
Add missing getModelsConfig and getActiveRuntimeModelSnapshot mocks,
and update switchModel assertion to expect the new { baseUrl } options
object introduced in
|
||
|
ChiGao
|
7f0c9791b7
|
feat(cli): expand TUI markdown rendering (#3680)
Some checks are pending
Qwen Code CI / Lint (push) Waiting to run
Qwen Code CI / Test (push) Blocked by required conditions
Qwen Code CI / Test-1 (push) Blocked by required conditions
Qwen Code CI / Test-2 (push) Blocked by required conditions
Qwen Code CI / Test-3 (push) Blocked by required conditions
Qwen Code CI / Test-4 (push) Blocked by required conditions
Qwen Code CI / Test-5 (push) Blocked by required conditions
Qwen Code CI / Test-6 (push) Blocked by required conditions
Qwen Code CI / Test-7 (push) Blocked by required conditions
Qwen Code CI / Test-8 (push) Blocked by required conditions
Qwen Code CI / Post Coverage Comment (push) Blocked by required conditions
Qwen Code CI / CodeQL (push) Waiting to run
E2E Tests / E2E Test (Linux) - sandbox:docker (push) Waiting to run
E2E Tests / E2E Test (Linux) - sandbox:none (push) Waiting to run
E2E Tests / E2E Test - macOS (push) Waiting to run
SDK Python / SDK Python (3.10) (push) Waiting to run
SDK Python / SDK Python (3.11) (push) Waiting to run
SDK Python / SDK Python (3.12) (push) Waiting to run
* feat(cli): expand markdown rendering in tui * fix(cli): render finalized mermaid images synchronously * fix(cli): harden markdown rendering paths * fix(cli): preserve mermaid source fallbacks * test(cli): make mermaid image renderer mock cross-platform * fix(cli): run windows mmdc shims through shell * fix(cli): address markdown rendering review comments * fix(cli): validate mermaid render timeout * feat(cli): expose rendered markdown source blocks * fix(cli): align mermaid source copy controls * fix(cli): make markdown render toggle visible * fix(cli): keep markdown render toggle quiet * fix(cli): generalize markdown render mode * fix(cli): broaden render mode and copy latex blocks * fix(cli): align rendered copy source indices * fix(cli): support copying inline latex expressions * feat(cli): document markdown render controls * test(cli): cover markdown render controls * fix(cli): tighten markdown render fallbacks * fix(cli): bound mermaid renderer cache * fix(cli): address markdown render review feedback * fix(cli): address markdown render review comments * test(cli): strengthen render mode shortcut coverage * fix(cli): address mermaid image review comments * fix(cli): stabilize renderer output truncation * test(cli): flush fake renderer stderr before exit * fix(cli): address markdown renderer review feedback * docs(cli): clarify mermaid image limits * fix(cli): refresh mermaid images on height resize * fix(cli): address markdown render review * chore: revert unrelated review formatting churn * fix(cli): avoid mermaid regex codeql alert * fix(cli): silence mermaid operator codeql alert * fix(cli): render inline math in markdown tables * fix(cli): harden markdown visual renderers * fix(cli): strip c1 controls from mermaid previews --------- Co-authored-by: 秦奇 <gary.gq@alibaba-inc.com> |
||
|
ChiGao
|
4f084352f4
|
feat(cli): customize banner area (logo, title, hide) (#3710)
* docs(design): add banner customization design (#3005) Document the design for issue #3005 (customize CLI banner area). Covers the banner region taxonomy and what is replaceable vs. locked, the three proposed settings (`ui.hideBanner`, `ui.customBannerTitle`, `ui.customAsciiArt`) and their resolution pipeline, the schema additions and wiring touch points, five alternative shapes considered, and the security / failure-handling guards. Mirrored EN + zh-CN under `docs/design/customize-banner-area/`. No code changes in this commit; implementation lands in a follow-up PR. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * feat(cli): customize banner area (logo, title, hide) Adds three opt-in `ui.*` settings that let users replace brand chrome on startup while keeping the operational lines (version, auth, model, path) locked: `hideBanner`, `customBannerTitle`, `customAsciiArt` (string, {path}, or {small,large}). A new resolver in `packages/cli/src/ui/utils/customBanner.ts` walks the loaded settings, normalizes each tier per scope (so {path} resolves against the file that declared it), reads the file with O_NOFOLLOW and a 64 KB cap on POSIX, sanitizes via a banner-specific stripper that drops OSC/CSI/SS2/SS3 sequences while preserving newlines, and caps art at 200 lines × 200 cols and titles at 80 chars. Every soft failure logs a `[BANNER]` warn and falls through to the bundled QWEN logo or default brand title — banner config can never crash the CLI. `<Header />` now picks the widest custom tier that fits via a shared `pickAsciiArtTier` helper and falls back to `shortAsciiLogo` otherwise; `<AppHeader />` extends the existing `showBanner` gate to honor `hideBanner` alongside the screen-reader fallback. Tracks #3005 and the design merged in #3671. * docs(design): apply prettier to banner customization design Reformats the EN and zh-CN design docs in `docs/design/customize-banner-area/` to satisfy `npx prettier --check`: table column alignment and trailing commas in `jsonc` examples. No content changes — the words, tables, and code blocks all say the same thing as before. Carries forward the only actionable feedback from the now-closed docs-only PR #3671, where the prettier check was the sole change requested. * fix(cli): address banner audit findings Three audit-driven fixes for the banner customization feature: 1. **VSCode JSON schema accepts every documented shape.** The `ui.customAsciiArt` entry in `packages/vscode-ide-companion/schemas/settings.schema.json` was declared as `type: object`, which made VSCode flag the inline-string form (`"customAsciiArt": " ___"`) — a shape the runtime accepts and the design doc recommends — as a schema violation. Replaced with a `oneOf` covering string, `{path}`, and `{small,large}` (with each tier itself string-or-`{path}`). 2. **Narrow terminals no longer leak the QWEN logo over a white-label deployment.** When a user supplied custom ASCII art but neither tier fit the terminal, `Header.tsx` previously fell back to the bundled `shortAsciiLogo` — silently undoing the white-label intent on small windows. The fallback now distinguishes "user supplied custom art" from "no custom art at all": in the first case the logo column is hidden entirely (info panel still renders); in the second case the default logo shows as before. Soft-failure paths (missing file, sanitization rejection) still fall through to `shortAsciiLogo`. 3. **Sanitizer strips C1 control bytes (0x80-0x9F).** The art and title strippers previously stopped at 0x7F, leaving single-byte CSI (`0x9B`), DCS (`0x90`), ST (`0x9C`) and other C1 controls intact — which legacy 8-bit terminals would still interpret. Aligned the ranges with the repo's existing `stripUnsafeCharacters` (in `textUtils.ts`) so banner content can't carry interpreted control bytes through. New tests cover: C1 strip in art and title, absolute path reads, symlink rejection on POSIX, narrow-terminal hide-on-custom-art, and end-to-end `<AppHeader />` rendering through `resolveCustomBanner`. The full banner suite is 48 tests (was 42). * docs(design): clarify cross-scope tier merge and white-label fallback Two clarifications surfaced by the audit on the implementation PR: 1. The design said `customAsciiArt` follows standard merge precedence, but the resolver actually walks scopes per-tier so workspace can override only `large` while user keeps `small`. Document that this per-tier walk is intentional — both because each `{path}` has to resolve against the file that declared it (the merged view loses that information) and because it lets users keep a personal default tier and override the other one per-workspace. 2. The render-time tier-selection step now distinguishes "user supplied custom art but neither tier fits" (hide the logo column entirely; falling back to `shortAsciiLogo` would silently undo a white-label deployment on narrow terminals) from "user supplied no custom art at all" (fall through to `shortAsciiLogo` and let the default-logo width gate decide). Step 5's pure soft-failure fallback (missing file, sanitization rejection) is unchanged — still `shortAsciiLogo`. Mirrored both edits in the zh-CN translation. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * docs(design): add size budget section to banner customization Question raised on the implementation PR: "why is the test logo `CCA` instead of the full `Custom Code Agent` — is there a character limit?" There is no character-count limit on titles or art. There is a **width budget** driven by terminal columns, plus an absolute hard cap (200×200 art, 80-char title) to keep malformed input from freezing layout. The existing user-facing guide didn't quantify the budget anywhere, so users were guessing why long inline names didn't render. Add a "How wide can the logo be? — the size budget" subsection that spells out the formula (`availableLogoWidth = terminalCols − 4 − 2 − 44`), tabulates it at 80 / 100 / 120 / 200 cols, calls out that a 17-char brand like "Custom Code Agent" can't render as a single ANSI Shadow line on most terminals (~120 cols of art), and shows the stacked-words `{ small, large }` recipe — including the `figlet` one-liner that generates the corresponding `banner-large.txt`. Mirrored in the zh-CN translation. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * docs(design): add limits-at-a-glance table; switch demo to Custom Agent The banner-customization design now has the size budget written down, but the per-cap limits (80-char title, 200×200 art, 64 KB file) were buried inside the size-budget formula table. Surface them as their own "Limits at a glance" subsection at the top of the user-configuration guide so users see the hard caps before they start hand-crafting art. Also switch the running example from "Custom Code Agent" (17 chars, ~120 cols of ANSI Shadow art on one line — too wide for any common terminal) to "Custom Agent" (12 chars, two-word stack at ~54 cols × 12 lines, fits any terminal ≥ 104 cols). The figlet recipe is now a two-word pipeline so a copy-paste run produces art the size the doc claims. Mirrored both changes in the zh-CN translation. The implementation itself is unchanged. Generated with AI Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * fix(cli): address PR review + CI Lint failure Two reviewer findings on PR #3710 (and the Lint job that fails for the same root cause): 1. **Schema regen now reproduces the committed JSON Schema.** The CI Lint step runs `npm run generate:settings-schema` and fails when the worktree dirties — my earlier hand-authored `oneOf` got blown away because `customAsciiArt` is `type: 'object'` in the source schema and the generator had no way to emit a union. Add a `jsonSchemaOverride` escape-hatch field on `SettingDefinition`: when set, the generator emits the override verbatim (description carried forward) instead of the type-driven shape. Set it on `customAsciiArt` to express the runtime union (string | {path} | {small,large} where each tier is itself string-or-{path}). The committed schema is now regenerated from source and CI's regenerate-and-diff check passes; two back-to-back regens produce identical output. 2. **Untrusted workspace settings no longer influence the banner.** `collectScopedTiers()` walked `settings.workspace` directly because per-scope file paths are needed to resolve relative `{path}` entries — but that bypassed the trust gate that `settings.merged` enforces. An untrusted checkout could therefore render its own ASCII art and trigger local file reads through a `{path}` entry before the user trusts the folder. Skip `settings.workspace` entirely when `settings.isTrusted` is false. Two regression tests cover the gate (untrusted = workspace silenced, falls through to user; trusted = workspace honored). Test suite for the banner is now 30 resolver tests + the existing Header / AppHeader / settingsSchema tests = 66 total, all green. * feat(cli): add ui.customBannerSubtitle for the spacer row Adds a fourth opt-in setting to the banner customization surface. The info panel renders four rows (title, subtitle/spacer, status, path); the second row was a hard-coded single-space spacer up to now. With this change a fork or white-label deployment can set `ui.customBannerSubtitle` to a one-line subtitle (e.g. "Built-in DataWorks Official Skills") and have it render in the secondary text color in place of the spacer. Empty/unset preserves the previous blank-spacer layout, so the change is back-compat. The subtitle is sanitized through the same `sanitizeSingleLine` helper as the title (now factored out): OSC / CSI / SS2 / SS3 leaders dropped, every other C0/C1 control byte replaced with a space, internal whitespace collapsed, ends trimmed. Capped at 160 characters — looser than the title's 80 because tagline / "powered by" copy commonly runs longer — with the same `[BANNER]` warn on truncation. Wiring: - `settingsSchema.ts` — new `customBannerSubtitle` entry next to `customBannerTitle`, `showInDialog: false` (free-form text in the TUI dialog isn't worth its own picker). - `customBanner.ts` — `ResolvedBanner.subtitle` field; `resolveCustomBanner` populates it; `sanitizeTitle` and the new `sanitizeSubtitle` share the same helper. - `Header.tsx` — when `customBannerSubtitle` is truthy the spacer row renders the string (secondary color, single line) instead of `<Text> </Text>`. Auth/model and path still sit at their usual positions. - `AppHeader.tsx` — pipes `resolvedBanner.subtitle` through. - VSCode JSON schema regenerated from source (idempotent). Tests: 5 new resolver tests (default, sanitize, length cap, empty, newline + C1 strip), 2 new Header tests (renders subtitle between title and auth; spacer preserved when unset), 1 new AppHeader integration test (end-to-end through resolver). Banner suite is now 35 + 17 + 6 + 16 = 74 tests, all green. Design docs (EN + zh-CN) updated: region taxonomy now lists four B-rows; "Limits at a glance" table grows a subtitle row; "Customization rules" matrix and "How to modify" section gain a "Add a brand subtitle" example with a rendered four-row preview. * docs(design): sweep stale 3-setting references after subtitle add Self-review found several sections of the banner customization design doc still framed for the original three settings; bring them in line with the four-setting reality landed in |
||
|
jinye
|
f0e8601982
|
fix(cli): add API Key option to qwen auth interactive menu (#3624)
* fix(cli): add "API Key" option to `qwen auth` interactive menu The `qwen auth` CLI command only showed 2 options (Coding Plan, Qwen OAuth), while the interactive `/auth` dialog showed 3 (Coding Plan, API Key, Qwen OAuth). Users following the README instructions to configure OpenRouter/Fireworks via `qwen auth` had no API Key entry point. - Add "API Key" option to the `runInteractiveAuth` menu with two sub-paths: "Alibaba Cloud ModelStudio Standard API Key" (guided flow) and "Custom API Key" (prints docs link) - Add `qwen auth api-key` yargs subcommand for direct access - Extract `createMinimalArgv` / `loadAuthConfig` helpers to eliminate duplicated CliArgs boilerplate - Extract `promptForInput` to share raw-mode stdin logic between `promptForKey` and `promptForModelIds` - Improve `showAuthStatus` to distinguish Coding Plan, Standard API Key, and generic OpenAI-compatible configurations - Align menu labels and descriptions with the interactive `/auth` dialog Closes #3413 Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> * docs: add `qwen auth api-key` to auth subcommand tables Update documentation to reflect the new `qwen auth api-key` subcommand: - auth.md: add to subcommands table, examples, and interactive menu display - commands.md: add to CLI Auth Subcommands table - quickstart.md: add to quick-reference command table * fix(cli): restore incomplete Coding Plan warning in showAuthStatus When selectedType is USE_OPENAI and Coding Plan metadata exists but the API key is missing, show the incomplete warning instead of falling through to the generic "OpenAI-compatible" status. * refactor(cli): use endpoint constants in region selector and fix status formatting - Use ALIBABA_STANDARD_API_KEY_ENDPOINTS constants for region descriptions instead of hardcoded URLs - Restore trailing newline in showAuthStatus "no auth" command list for consistent spacing * fix(cli): determine active auth method from model config in showAuthStatus Previously showAuthStatus checked which env keys exist to determine the auth method, causing false reports when users switch providers (e.g., Coding Plan key still present after switching to Standard API Key). Now it inspects the active model's provider config (baseUrl/envKey) to determine the actual method, and validates the corresponding key exists: - Coding Plan: check via isCodingPlanConfig + CODING_PLAN_ENV_KEY - Standard API Key: check via DASHSCOPE_STANDARD_API_KEY_ENV_KEY + endpoints - Generic OpenAI-compatible: check if the model's envKey is set Also clear stale Coding Plan metadata (codingPlan.region/version and process.env) when switching to Standard API Key. * fix(cli): add legacy fallback in showAuthStatus and clear persisted Coding Plan env - When no active model config is found (legacy setups without modelProviders), fall back to env key / metadata checks for Coding Plan status detection. Fixes CI test failures. - When activeConfig exists but has no envKey, report incomplete status instead of false positive "Configured". - Clear persisted env.BAILIAN_CODING_PLAN_API_KEY from settings when switching to Standard API Key, not just process.env. * fix(cli): also remove Coding Plan model entries when switching to Standard API Key When switching to Standard API Key, filter out existing Coding Plan model entries from modelProviders.openai in addition to old Standard entries. Previously these were preserved but their credential source (BAILIAN_CODING_PLAN_API_KEY) was cleared, leaving broken model entries visible in /model. --------- Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com> |