fix: stabilize Runtime workflows for 0.12.3

OpenCode status checks could cold-start throwaway servers, unrelated conversations could block each other, and model, Continue, or remote paths could replay or duplicate work under failure and backlog conditions. Settings could also lose edits or overwrite unrelated model fields. This update makes status checks passive, keeps per-conversation ordering while allowing concurrency, preserves native long-running work with bounded display capture and backpressure, narrows retries to known pre-dispatch failures, and aligns remote Ask permissions without reducing Execute. It also keeps Embedding updates narrow, improves settings and browser guidance, removes the unused legacy Main gateway state, and records the validated Desktop 0.12.3 and Agent 0.11.17 release metadata. Release note: 改善 OpenCode 切换、多会话长任务、模型重试、远程 Ask/Execute 与设置保存可靠性，并完成 Desktop 0.12.3 / Agent 0.11.17 发布准备。
2026-09-04 08:05:29 +08:00
parent 374968d50b
commit 4490e5fd15
60 changed files with 1639 additions and 3839 deletions
@@ -1301,6 +1301,172 @@ describe('RuntimeSettingsStore', () => {
    ).toBe(21)
  })

+  it('changes the active embedding connection without changing model profiles', async () => {
+    const { store } = await createStore()
+    const profileId = '00000000-0000-4000-8000-000000000091'
+    const connectionId = '00000000-0000-4000-8000-000000000092'
+    await store.update(
+      settings({
+        modelProfiles: [
+          {
+            id: profileId,
+            name: 'Tenant model',
+            baseUrl: 'https://gateway.example/v1',
+            modelName: 'tenant-model',
+            protocol: 'anthropic-messages',
+            authentication: 'api-key',
+            supportsImageInput: false,
+            contextWindowTokens: 200_000,
+            maximumOutputTokens: 48_000,
+            imageGenerationQuality: 'auto',
+            requestHeaders: { 'x-tenant-id': 'tenant-a' },
+            requestBody: { temperature: 0.2 },
+            apiKey: { action: 'replace', value: 'tenant-secret' }
+          }
+        ],
+        defaultModelProfileId: profileId,
+        embeddingConnections: [
+          {
+            id: builtinEmbeddingConnectionId,
+            name: 'GoodBuddy 内置向量模型',
+            kind: 'builtin'
+          },
+          {
+            id: connectionId,
+            name: '团队向量模型',
+            kind: 'openai-compatible',
+            baseUrl: 'https://team-vectors.example/v1/embeddings',
+            modelName: 'team/embed-v2',
+            authentication: 'none',
+            apiKey: { action: 'keep' }
+          }
+        ],
+        activeEmbeddingConnectionId: builtinEmbeddingConnectionId
+      })
+    )
+    const selection =
+      await store.setActiveEmbeddingConnection(connectionId)
+    expect(selection).toMatchObject({
+      changed: true,
+      previousConnectionId: builtinEmbeddingConnectionId
+    })
+
+    await expect(store.getResolvedSettings()).resolves.toMatchObject({
+      activeEmbeddingConnectionId: connectionId,
+      modelProfiles: [
+        expect.objectContaining({
+          id: profileId,
+          contextWindowTokens: 200_000,
+          maximumOutputTokens: 48_000,
+          requestHeaders: { 'x-tenant-id': 'tenant-a' },
+          requestBody: { temperature: 0.2 },
+          apiKey: 'tenant-secret'
+        })
+      ]
+    })
+  })
+
+  it('preserves configured model fields when embedding changes under an environment override', async () => {
+    const { filePath, store } = await createStore()
+    const profileId = '00000000-0000-4000-8000-000000000093'
+    await store.update(
+      settings({
+        modelProfiles: [
+          {
+            id: profileId,
+            name: 'Stored OpenAI model',
+            baseUrl: 'https://stored.example/v1',
+            modelName: 'stored-model',
+            protocol: 'openai-chat-completions',
+            authentication: 'none',
+            supportsImageInput: false,
+            imageGenerationQuality: 'auto',
+            requestHeaders: { 'x-tenant-id': 'tenant-a' },
+            requestBody: { temperature: 0.2 },
+            apiKey: { action: 'keep' }
+          }
+        ],
+        defaultModelProfileId: profileId
+      })
+    )
+    const environmentStore = new RuntimeSettingsStore(
+      filePath,
+      cipher,
+      { GOODBUDDY_BIGTOKEN_API_KEY: 'environment-key' }
+    )
+    const effective = await environmentStore.getPublicSettings()
+    expect(effective.modelProfiles[0]).toMatchObject({
+      protocol: 'anthropic-messages',
+      authentication: 'api-key',
+      credentialSource: 'environment'
+    })
+
+    await environmentStore.setActiveEmbeddingConnection(
+      legacyEmbeddingConnectionId
+    )
+
+    const withoutEnvironment = new RuntimeSettingsStore(
+      filePath,
+      cipher,
+      {}
+    )
+    await expect(
+      withoutEnvironment.getResolvedSettings()
+    ).resolves.toMatchObject({
+      activeEmbeddingConnectionId: legacyEmbeddingConnectionId,
+      modelProfiles: [
+        expect.objectContaining({
+          id: profileId,
+          baseUrl: 'https://stored.example/v1',
+          modelName: 'stored-model',
+          protocol: 'openai-chat-completions',
+          authentication: 'none',
+          requestHeaders: { 'x-tenant-id': 'tenant-a' },
+          requestBody: { temperature: 0.2 }
+        })
+      ]
+    })
+  })
+
+  it('serializes embedding selection with full settings updates', async () => {
+    const { store } = await createStore()
+    const settingsUpdate = store.update(
+      settings({ modelName: 'updated-after-selection-read' })
+    )
+    const selectionUpdate =
+      store.setActiveEmbeddingConnection(
+        legacyEmbeddingConnectionId
+      )
+
+    await Promise.all([settingsUpdate, selectionUpdate])
+
+    await expect(store.getResolvedSettings()).resolves.toMatchObject({
+      modelName: 'updated-after-selection-read',
+      activeEmbeddingConnectionId: legacyEmbeddingConnectionId
+    })
+  })
+
+  it('rolls back only the embedding selection after later settings updates', async () => {
+    const { store } = await createStore()
+    const selection =
+      await store.setActiveEmbeddingConnection(
+        legacyEmbeddingConnectionId
+      )
+    await store.update(
+      settings({
+        modelName: 'saved-while-embedding-activated',
+        activeEmbeddingConnectionId: legacyEmbeddingConnectionId
+      })
+    )
+
+    await selection.restore()
+
+    await expect(store.getResolvedSettings()).resolves.toMatchObject({
+      modelName: 'saved-while-embedding-activated',
+      activeEmbeddingConnectionId: builtinEmbeddingConnectionId
+    })
+  })
+
  it('migrates version 13 with reranking disabled by default', async () => {
    const { filePath, store } = await createStore()
    await store.update(settings())