mirror of
https://github.com/apache/maka.git
synced 2026-10-28 08:54:38 +08:00
Wang
0b7bf39f49
* perf(desktop): defer unused WorkHub and bot SDK initialization Create the WorkHub renderer on first use rather than on enable, while retaining drafts after use. Load platform SDKs only when a configured bot starts or Slack credentials are tested. Generated-by: OpenAI Codex * fix(ui): release attachment previews when composers unmount Track Blob URL ownership before decode and stop asynchronous staging after unmount while preserving hidden drafts and StrictMode replay. Generated-by: OpenAI Codex * perf(runtime-host): compact completed message replay identities Retain canonical content digests instead of historical submit and queue edit bodies, without evicting epoch-long outcomes or changing pending operations. Generated-by: OpenAI Codex * fix(desktop): cancel completed display frame fallback callbacks Cancel the losing animation frame or timeout after a stream display batch flush, preventing suspended background frames from retaining completed handlers. Generated-by: OpenAI Codex * perf(runtime): release retired graph epoch reconciliation snapshots Fence old supervisor callbacks and preserve explicit reconcile readers across epoch handover. Retain lightweight driver metadata for historical projection repair and close diagnostics. Generated-by: OpenAI Codex * fix(runtime-host): detach transcript preparation close listeners Replace connection-lifetime Promise races with removable abort listeners so released transcript overlay buffers can be collected while the connection remains open. Generated-by: OpenAI Codex * fix(desktop): bound browser release fencing to pending acquires Drop per-session release epochs when connection attempts settle and avoid retaining tombstones for sessions that never acquired a browser, preserving reentrant teardown and single-flight fences. Generated-by: OpenAI Codex * perf(runtime-host): compact historical root admission bodies Retain canonical content digests and owned identity metadata for historical admissions while keeping full current tips and validated recovery or conflict results. Generated-by: OpenAI Codex * test(runtime-host): close admission fixture stores before cleanup Release the SQLite lease before removing temporary files so root admission tests clean up on Windows and do not retain fixture database owners. Generated-by: OpenAI Codex * perf(desktop): release activated subscription preparation roots Generated-by: OpenAI Codex * refactor(desktop): move display scheduling into conversation ownership Keep frame cleanup within the renderer architecture boundary without widening legacy budgets. Generated-by: OpenAI Codex * perf: release Zod recursive parse state after completion Keep cycle memoization scoped to each parse and restore allocation state when parsing throws. Generated-by: OpenAI Codex * docs: record Zod patch source license Generated-by: OpenAI Codex * perf(runtime): release processed WeChat SSE payloads while idle Keep iterator result and decoded message lifetimes within each completed read. Preserve reconnect cursors, parser boundaries and iterator cleanup semantics. Generated-by: OpenAI Codex * fix(peer): bind WebRTC workers to connection lifetime Cancel connection-owned channel workers and close the native peer on explicit close, unpolled muxer drop and failed or aborted upgrades. Keep successful connection lifetime separate from signaling-attempt cancellation. Generated-by: OpenAI Codex * perf(desktop): release delivery records during retry waits Keep retry callbacks outside the delivery lexical context so pending or completed attempts do not retain parsed message bodies for the five-second retry interval. Preserve durable intents, timer behavior, and replay identity. Generated-by: OpenAI Codex * perf(peer): cancel obsolete reconciliation timers Own the five-minute timer and abort listener in each reconciliation wait. Early public triggers release both immediately while normal expiry and endpoint shutdown still wake the loop. Generated-by: OpenAI Codex * test(desktop): scope POSIX mode assertion to supported platforms Keep local restart and retry assertions running on Windows without equating stat mode bits with ACL permissions. Production permission handling is unchanged. Generated-by: OpenAI Codex * fix(ui): avoid retaining messages through unknown code languages Generated-by: OpenAI Codex * fix(cli): release completed event waiters Generated-by: OpenAI Codex * fix(cli): share immutable text across editor undo snapshots Copy mutable editor state and paste maps without deep-cloning each stored paste for every typed word. Preserve all undo steps and cover public editing, paste renumbering, and retained heap growth. Generated-by: OpenAI Codex * perf(core): bound Unicode truncation allocations Collect only the code-point prefix needed for nonnegative budgets while preserving cleaning and numeric slice semantics. Cover transient allocations and Unicode behavior with regression tests. Generated-by: OpenAI Codex * perf(core): bound diagnostic truncation allocations Preserve complete redaction and byte-budget semantics while limiting temporary code-point arrays and UTF-8 encoding to the useful prefix. Generated-by: OpenAI Codex * perf(runtime-host): release consumed peer receive buffers Generated-by: OpenAI Codex * perf(ui): defer reasoning bodies until first expansion Generated-by: OpenAI Codex * perf(runtime): detach truncated web search result text Generated-by: OpenAI Codex * perf(desktop): load update attestation dependencies on demand Keep startup channel metadata free of Sigstore and TUF module graphs. Load the parser and verifier only when an update is actually verified, preserving fail-closed trust checks. Generated-by: OpenAI Codex * fix(runtime): preserve cancellation through request customization Forward the caller's effective AbortSignal instead of a temporary Request's GC-sensitive derived signal. Preserve explicit null, inherited and overriding signals on every body path, and cover prompt native HTTP cancellation after GC. Generated-by: OpenAI Codex * perf(runtime): reuse buffered JSON request bodies Decode the bytes already read for request customization instead of cloning and reading the complete body twice. Preserve the original read boundary, stream error identity, caller locks and cancellation semantics. Generated-by: OpenAI Codex * perf(computer-use): scope session cleanup metadata to live work Avoid allocating unknown-session state during ordinary cleanup. Retain cancellation generations only while queued operations own them, preserving same-turn stop authority and executor-release fencing. Generated-by: OpenAI Codex * fix(runtime-host): release uploads rejected after session removal Release only connection-owned staging in the existing missing-session rejection path. Preserve response and TTL contracts while preventing failed commit and cleanup requests from retaining completed upload buffers. Generated-by: OpenAI Codex * perf(runtime-host): release retired Goal token metadata Scope cached token totals to the existing Goal control lease so late context reads cannot repopulate retired entries or overwrite a new Goal. Delete only after committed retirement and clear during drain, retaining rollback and close fences. Generated-by: OpenAI Codex * test(runtime-host): close Goal fixture execution stores Release SQLite execution-store handles before removing temporary roots. Covers the new lifetime regressions and existing Goal coordinator tests that reproduced the same Windows EBUSY failure on pre-fix production source. Generated-by: OpenAI Codex * perf(mcp): release settled request graphs behind blocked sends Scope cancellation error observers independently and revoke the pending send observer's native rejection reference when the request settles. Preserve queued frames and late error handling in both SDK module formats. Allow the required unified-diff context prefix before upstream tabs while retaining other patch whitespace checks. Generated-by: OpenAI Codex * perf(runtime-host): share identical turn skill inventories Reuse only fully equal frozen snapshots after each fresh catalog read. Preserve per-turn historical state, the 100-entry FIFO and failed-read retry behavior, and prevent evicted late reads from adding retention. Generated-by: OpenAI Codex * perf(runtime-host): share identical completed prompt text Retain per-turn source revisions and the existing FIFO while sharing equal text only within live cache ownership. Generated-by: OpenAI Codex * test(runtime-host): make prompt heap checks portable to Electron Generated-by: Codex * perf(ui): detach bounded drafts from oversized input strings Generated-by: OpenAI Codex * perf(runtime): reuse private Responses continuation history Generated-by: OpenAI Codex * perf(ui): detach capped completed stream text Generated-by: OpenAI Codex * perf(ui): simplify bounded draft string copying Generated-by: OpenAI Codex * perf(ui): detach oversized stream seed backings Generated-by: OpenAI Codex * perf(runtime): detach bounded archive read results Generated-by: OpenAI Codex * perf(runtime): release completed tool settlement batches Generated-by: OpenAI Codex * perf(ui): detach normalized sent prompt backing Generated-by: OpenAI Codex * perf(runtime): release unused plugin isolation labels Intern named isolation symbols weakly so retired configurations do not retain historical labels. Preserve detached Context and service identity, and guard delayed finalizers against newer same-name symbols. Generated-by: OpenAI Codex * perf(runtime): materialize memory split candidates on demand Preserve split ranking and first-fit semantics while constructing only the candidate currently being tested. Verify complete model context, durable coverage, and bounded event-reference copying through the SQLite-backed Host extraction path. Generated-by: OpenAI Codex * perf(runtime): bound memory localization prefix allocation Generated-by: OpenAI Codex * perf(storage): bound session catalog preview allocation Generated-by: OpenAI Codex * perf(runtime): bound graph handoff prefix allocation Generated-by: OpenAI Codex * perf(runtime): bound child result prefix allocation Generated-by: OpenAI Codex * perf(runtime): bound workspace instruction prefix allocation Generated-by: OpenAI Codex * perf(runtime): bound PTY model tail allocation Generated-by: OpenAI Codex * test(runtime): close shell result recovery fixture store Generated-by: OpenAI Codex * perf(runtime): consume discarded OAuth error bodies once Generated-by: OpenAI Codex * perf(storage): size stable read buffers to admitted files Generated-by: OpenAI Codex * perf(runtime-host): release fully consumed frame buffers Generated-by: OpenAI Codex * test(runtime-host): isolate consumed buffers from shared pools Generated-by: OpenAI Codex * perf(runtime-host): avoid recopying empty frame prefixes Generated-by: OpenAI Codex * perf(ui): initialize persisted composer history once per mount Generated-by: OpenAI Codex * perf(runtime-host): release consumed peer byte-stream buffers Generated-by: OpenAI Codex * perf(runtime): stream live PTY code-point chunks Avoid full code-point and sliced arrays while preserving exact raw output chunk boundaries, publication sequences, and terminal flush behavior. Generated-by: Codex * perf(runtime-host): avoid length-only Base64 decoding Count validated or generated Base64 bytes without materializing temporary decoded buffers. Preserve canonical checks, artifact bounds and transcript range semantics. Generated-by: Codex * perf(runtime): bound research artifact character arrays Collect only requested code points while preserving full ledger hash validation, total length and sanitized output. Exercise the production Host tool facade against real SQLite and artifact files. Generated-by: Codex * perf(runtime): count edit lines without splitting text Reuse existing occurrence counting for match line positions. Preserve exact and fuzzy edits, trailing newline semantics and descriptor-pinned writes. Generated-by: Codex * perf(runtime): collect only localized edit diff windows Count source lines without full arrays and collect the existing bounded diff windows directly. Preserve valid edit ranges, UTF-8 budgets, hunk formatting and descriptor-pinned writes. Generated-by: Codex * perf(runtime): collect only requested file read lines Preserve split/slice window behavior while avoiding full-file line arrays in local and worker reads. Add real file regressions and allocation checks. Generated-by: OpenAI Codex * fix(desktop): scope interaction hydration fences to pending reads Replace the ever-growing per-session epoch map with one pending read identity. Preserve stale-response guards without keeping metadata for completed sessions; ratchet the architecture token ledger downward. Generated-by: OpenAI Codex * perf(runtime): avoid full line arrays in bounded tool output Count line boundaries and collect only the retained head or tail window. Preserve byte slicing, recovery markers, and removed counts, with old-source negative allocation and real pipe projection coverage. Generated-by: OpenAI Codex * perf(runtime): bound oversized tool-line UTF-8 encoding Encode only a byte-budget-sized UTF-16 window, with one extra unit to preserve surrogate boundaries. Keep legacy Buffer behavior for unusual budgets and verify exact head/tail output against old source. Generated-by: OpenAI Codex * perf(runtime): bound WebFetch preview UTF-8 encoding Encode only the model-visible prefix plus a surrogate guard unit, preserving the byte cap and truncation marker. Cover real Host HTTP fetches, privacy refusal, Unicode boundaries and old-source allocation regression. Generated-by: OpenAI Codex * perf(runtime-host): avoid duplicate resource result clones Clone each runtime resource update once before bounding its state. Preserve source isolation and query revisions, and cover get/list allocation with old-source negative controls. Generated-by: OpenAI Codex * chore: trim memory optimization test additions Remove 65 new test and fixture files plus added allocation probes from existing suites. Keep only minimal existing-test behavior and cleanup adaptations. Preserve production memory fixes and remove stale test references. Generated-by: OpenAI Codex * fix: align memory changes with CI metadata gates Re-pin the compatible byte-count optimization to main's protocol epoch, record the MCP dependency patch license, and correct the renderer token ledger to the checked source. Generated-by: OpenAI Codex * ci: include bounded line reader in Windows package inputs Generated-by: OpenAI Codex * fix(runtime): keep graph epoch handover independent of teardown Fence retired drivers synchronously, then release their snapshots after admitted work and cleanup settle without blocking the next turn. Generated-by: OpenAI Codex * fix(bots): report lazy SDK initialization failures Handle module loading and client construction inside each bridge's failure boundary so registry status reflects failed connections. Generated-by: OpenAI Codex * fix(desktop): defer cold WorkHub summons until renderer ready Reuse pending focus to keep the window hidden during initial loading, allow cancellation, and preserve synchronous warm shortcuts. Generated-by: OpenAI Codex * perf(memory): finish scoped lifecycle review follow-ups Initialize composer attachment ownership lazily, avoid empty handshake backing views, clarify line-window and Zod upgrade contracts, and cover CU observer cleanup without restoring allocation probes. Generated-by: OpenAI Codex * test(desktop): wait for automatic naming before archive * test(memory): cover cancellation and recursive lifetime contracts Generated-by: OpenAI Codex * test(peer): observe backpressure before testing teardown Generated-by: OpenAI Codex * fix(ci): repin compatible allocation change to epoch 142 Revalidated the byte-count-only protocol changes against current main after merging its Skill query epoch bump. Generated-by: OpenAI Codex