refactor(chatCore): extrai runPluginOnResponseHook (#3501)

[diegosouzapw]

O quê

Decomposição do god-file chatCore.ts (#3501). Extrai o hook de plugin onResponse (fire-and-forget) da finalização do caminho streaming de handleChatCore para chatCore/pluginOnResponse.ts.

Como

runPluginOnResponseHook({ requestId, body, model, provider, apiKeyInfo }). O runOnResponse não é awaited e tanto o import dinâmico quanto o run engolem os próprios erros (fail-open); o ctx (com metadata: {}) e o { status: 200 } são preservados. Comportamento byte-idêntico.

Validação (Rule #18 — TDD)

• +3 caracterizações (hooks in-memory via registerHook): no-op sem hooks registrados, hook recebe o ctx + response status-200, fail-open com hook que lança.
• Caracterizações chatcore pré-existentes intactas → 301 → 304 verde.
• typecheck:core limpo; eslint 0.

⚠️ Base-reds pré-existentes (NÃO deste slice, provados no tip pristine)

• check:db-rules — compressionRunTelemetry.ts não re-exportado.
• check:complexity — 1919 > baseline 1916. Slice delta-0.

Impacto

chatCore.ts −7 linhas. Sem mudança de API/comportamento público.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[gemini-code-assist]

Code Review

This pull request refactors the handleChatCore handler by extracting the inline onResponse plugin hook execution into a dedicated runPluginOnResponseHook helper function, accompanied by a new unit test suite. The review feedback recommends using fallback empty strings for the model and provider parameters to avoid potential type mismatches with the strictly typed PluginContext, and implementing a beforeEach hook reset in the tests to ensure proper test isolation and prevent flaky behavior.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

The PluginContext interface in src/lib/plugins/hooks.ts strictly types model and provider as string. Since args.model and args.provider are typed as string | null | undefined, passing them directly can lead to type mismatches or runtime errors if plugins attempt to call string methods on them. Providing a fallback empty string (?? "") ensures type safety and robustness.

Suggested change

	runOnResponse(
	{
	requestId: args.requestId,
	body: args.body,
	model: args.model,
	provider: args.provider,
	apiKeyInfo: args.apiKeyInfo,
	metadata: {},
	},
	{ status: 200 }
	).catch(() => {});
	runOnResponse(
	{
	requestId: args.requestId,
	body: args.body,
	model: args.model ?? "",
	provider: args.provider ?? "",
	apiKeyInfo: args.apiKeyInfo,
	metadata: {},
	},
	{ status: 200 }
	).catch(() => {});

[gemini-code-assist]

To prevent test pollution and ensure test isolation, we should reset the hook registry before each test. Currently, multiple tests register handlers for the same 'test-onresponse-plugin' name without clearing them, which can lead to multiple handlers executing in subsequent tests and causing flaky test behavior. Using resetHooks() in a beforeEach block ensures a clean slate for each test.

import { test, after, beforeEach } from "node:test";
import assert from "node:assert/strict";

const { registerHook, unregisterHook, resetHooks } = await import("../../src/lib/plugins/hooks.ts");

beforeEach(() => {
  resetHooks();
});