diff --git a/src/utils/common.js b/src/utils/common.js index 8ec8fed..84f12a4 100644 --- a/src/utils/common.js +++ b/src/utils/common.js @@ -297,6 +297,7 @@ export async function handleStreamRequest(res, service, model, requestBody, from let fullResponseJson = ''; let fullOldResponseJson = ''; let responseClosed = false; + let anyDataSent = retryContext?.anyDataSent || false; // 跟踪是否已向客户端发送过任何数据 // 重试上下文：包含 CONFIG 和重试计数 // maxRetries: 凭证切换最大次数（跨凭证），默认 5 次 @@ -439,6 +440,7 @@ export async function handleStreamRequest(res, service, model, requestBody, from if (!clientDisconnected.value && !res.writableEnded) { try { res.write(`event: ${chunk.type}\n`); + anyDataSent = true; } catch (writeErr) { logger.error('[Stream] Failed to write event:', writeErr.message); clientDisconnected.value = true; @@ -453,6 +455,7 @@ export async function handleStreamRequest(res, service, model, requestBody, from if (!clientDisconnected.value && !res.writableEnded) { try { res.write(`data: ${JSON.stringify(chunk)}\n\n`); + anyDataSent = true; } catch (writeErr) { logger.error('[Stream] Failed to write data:', writeErr.message); clientDisconnected.value = true; @@ -482,9 +485,9 @@ export async function handleStreamRequest(res, service, model, requestBody, from return; } - // 如果已经发送了内容，不进行重试（避免响应数据损坏） - if (fullResponseText.length > 0) { - logger.info(`[Stream Retry] Cannot retry: ${fullResponseText.length} bytes already sent to client`); + // 如果已经发送了数据（包括 metadata），不进行重试（避免响应数据损坏或顺序错误） + if (anyDataSent) { + logger.info(`[Stream Retry] Cannot retry: data already sent to client`); // 直接发送错误并结束 const errorPayload = createStreamErrorResponse(error, fromProvider); if (!res.writableEnded) { @@ -553,7 +556,8 @@ export async function handleStreamRequest(res, service, model, requestBody, from CONFIG, currentRetry: currentRetry + 1, maxRetries, - clientDisconnected // 传递断开状态 + clientDisconnected, // 传递断开状态 + anyDataSent // 传递数据发送状态 }; // 递归调用，使用新的服务 @@ -821,12 +825,23 @@ export async function handleModelListRequest(req, res, service, endpointType, CO logger.info(`[ModelList] Aggregating models for 'auto' mode...`); clientModelList = await providerPoolManager.getAllAvailableModels(endpointType); } else { - // --- 原有的单提供商逻辑 --- + // --- 单提供商逻辑 --- const toProvider = CONFIG.MODEL_PROVIDER; - + + // service 可能未在上层预先注入（例如仅改了路径 provider 前缀），这里兜底获取 + let resolvedService = service; + if (!resolvedService) { + const { getApiService } = await import('../services/service-manager.js'); + resolvedService = await getApiService(CONFIG, null, { skipUsageCount: true }); + } + + if (!resolvedService || typeof resolvedService.listModels !== 'function') { + throw new Error(`[ModelList] Service adapter is unavailable or does not implement listModels() for provider: ${toProvider}`); + } + // 1. Get the model list in the backend's native format. - const nativeModelList = await service.listModels(); - + const nativeModelList = await resolvedService.listModels(); + // 2. Convert the model list to the client's expected format, if necessary. clientModelList = nativeModelList; if (!getProtocolPrefix(toProvider).includes(getProtocolPrefix(fromProvider))) { @@ -896,17 +911,20 @@ export async function handleContentGenerationRequest(req, res, service, endpoint let actualCustomName = CONFIG.customName; - // 2.5. 如果使用了提供商池，根据模型重新选择提供商（支持 Fallback） - // 注意：这里开启 acquireSlot: true，会占用并发名额或进入队列 - if (providerPoolManager && (CONFIG.MODEL_PROVIDER === MODEL_PROVIDER.AUTO || (CONFIG.providerPools && CONFIG.providerPools[CONFIG.MODEL_PROVIDER]))) { + // 2.5. 根据模型选择服务适配器： + // - service 缺失时（例如上游未预先注入）进行兜底选择 + // - 使用号池/AUTO 时按模型重选并支持 fallback + // 注意：仅在号池场景开启 acquireSlot，占用并发名额或进入队列 + const shouldSelectByPool = providerPoolManager && (CONFIG.MODEL_PROVIDER === MODEL_PROVIDER.AUTO || (CONFIG.providerPools && CONFIG.providerPools[CONFIG.MODEL_PROVIDER])); + if (!service || shouldSelectByPool) { const { getApiServiceWithFallback } = await import('../services/service-manager.js'); - const result = await getApiServiceWithFallback(CONFIG, model, { acquireSlot: true }); - + const result = await getApiServiceWithFallback(CONFIG, model, { acquireSlot: shouldSelectByPool }); + service = result.service; toProvider = result.actualProviderType; actualUuid = result.uuid || pooluuid; actualCustomName = result.serviceConfig?.customName || CONFIG.customName; - + // 如果发生了模型级别的 fallback，需要更新请求使用的模型 if (result.actualModel && result.actualModel !== model) { logger.info(`[Content Generation] Model Fallback: ${model} -> ${result.actualModel}`); @@ -916,7 +934,7 @@ export async function handleContentGenerationRequest(req, res, service, endpoint if (result.isFallback) { logger.info(`[Content Generation] Fallback activated: ${CONFIG.MODEL_PROVIDER} -> ${toProvider} (uuid: ${actualUuid})`); } else { - logger.info(`[Content Generation] Re-selected service adapter based on model: ${model}`); + logger.info(`[Content Generation] Selected service adapter based on model: ${model}`); } } diff --git a/static/app/provider-manager.js b/static/app/provider-manager.js index 634f7d3..3f82a13 100644 --- a/static/app/provider-manager.js +++ b/static/app/provider-manager.js @@ -16,6 +16,8 @@ import { setServiceMode } from './event-handlers.js'; let initialServerTime = null; let initialUptime = null; let initialLoadTime = null; +let isStaticProviderConfigsUpdated = false; +let cachedSupportedProviders = null; /** * 加载系统信息 @@ -186,11 +188,35 @@ function updateTimeDisplay() { */ async function loadProviders() { try { - const [providers, supportedProviders] = await Promise.all([ - window.apiClient.get('/providers'), - window.apiClient.get('/providers/supported') - ]); - renderProviders(providers, supportedProviders); + const providers = await window.apiClient.get('/providers'); + + // 动态更新其他模块的提供商信息，只需更新一次 + if (!isStaticProviderConfigsUpdated) { + cachedSupportedProviders = await window.apiClient.get('/providers/supported'); + const providerConfigs = getProviderConfigs(cachedSupportedProviders); + + // 动态更新凭据文件管理的提供商类型筛选项 + updateProviderFilterOptions(providerConfigs); + + // 动态更新仪表盘页面的路径路由调用示例 + renderRoutingExamples(providerConfigs); + + // 动态更新仪表盘页面的可用模型列表提供商信息 + updateModelsProviderConfigs(providerConfigs); + + // 动态更新配置教程页面的提供商信息 + updateTutorialProviderConfigs(providerConfigs); + + // 动态更新用量查询页面的提供商信息 + updateUsageProviderConfigs(providerConfigs); + + // 动态更新配置管理页面的提供商选择标签 + updateConfigProviderConfigs(providerConfigs); + + isStaticProviderConfigsUpdated = true; + } + + renderProviders(providers, cachedSupportedProviders); } catch (error) { console.error('Failed to load providers:', error); } @@ -349,24 +375,6 @@ function renderProviders(providers, supportedProviders = []) { } }); - // 动态更新凭据文件管理的提供商类型筛选项 - updateProviderFilterOptions(providerConfigs); - - // 动态更新仪表盘页面的路径路由调用示例 - renderRoutingExamples(providerConfigs); - - // 动态更新仪表盘页面的可用模型列表提供商信息 - updateModelsProviderConfigs(providerConfigs); - - // 动态更新配置教程页面的提供商信息 - updateTutorialProviderConfigs(providerConfigs); - - // 动态更新用量查询页面的提供商信息 - updateUsageProviderConfigs(providerConfigs); - - // 动态更新配置管理页面的提供商选择标签 - updateConfigProviderConfigs(providerConfigs); - // 更新统计卡片数据 const activeProviders = hasProviders ? Object.keys(providers).length : 0; updateProviderStatsDisplay(activeProviders, totalHealthy, totalAccounts);