feat: pre-request token validation, mid-stream error handling, and usage stats improvements

- Qoder: handle mid-stream errors by returning proper error Response instead of embedding in stream
- Qoder: add refreshCredentials() to validate token via quota endpoint before requests
- chatCore: validate and refresh provider tokens before sending chat requests
- chatCore: bail early with 401 on unrecoverable token refresh errors
- Usage stats: track apiKey, comboName, fallbackHistory in request details
- Dashboard: improve Combos, Endpoint, Provider, Usage, and RequestDetails pages
- API keys route: upsert logic with provider_type support
- DB repos: usageRepo query improvements, requestDetailsRepo pagination, apiKeysRepo updates

open-sse/handlers/chatCore.js

@@ -13,7 +13,7 @@ import { HTTP_STATUS } from "../config/runtimeConfig.js";
 import { handleBypassRequest } from "../utils/bypassHandler.js";
 import { trackPendingRequest, appendRequestLog, saveRequestDetail } from "@/lib/usageDb.js";
 import { getExecutor } from "../executors/index.js";
-import { buildRequestDetail, extractRequestConfig } from "./chatCore/requestDetail.js";
+import { buildRequestDetail, extractRequestConfig, saveUsageStats } from "./chatCore/requestDetail.js";
 import { handleForcedSSEToJson } from "./chatCore/sseToJsonHandler.js";
 import { handleNonStreamingResponse } from "./chatCore/nonStreamingHandler.js";
 import { handleStreamingResponse, buildOnStreamComplete } from "./chatCore/streamingHandler.js";
@@ -34,7 +34,7 @@ import { prefetchRemoteImages } from "../translator/concerns/prefetch.js";
  * @param {object} options.credentials - Provider credentials
  * @param {string} options.sourceFormatOverride - Override detected source format (e.g. "openai-responses")
  */
-export async function handleChatCore({ body, modelInfo, credentials, log, onCredentialsRefreshed, onRequestSuccess, onDisconnect, onMidStreamError, clientRawRequest, connectionId, userAgent, apiKey, ccFilterNaming, rtkEnabled, headroomEnabled, headroomUrl, headroomCompressUserMessages, cavemanEnabled, cavemanLevel, ponytailEnabled, ponytailLevel, sourceFormatOverride, providerThinking }) {
+export async function handleChatCore({ body, modelInfo, credentials, log, onCredentialsRefreshed, onRequestSuccess, onDisconnect, onMidStreamError, clientRawRequest, connectionId, userAgent, apiKey, comboName, fallbackHistory, ccFilterNaming, rtkEnabled, headroomEnabled, headroomUrl, headroomCompressUserMessages, cavemanEnabled, cavemanLevel, ponytailEnabled, ponytailLevel, sourceFormatOverride, providerThinking }) {
   const { provider, model } = modelInfo;
   const requestStartTime = Date.now();
 
@@ -185,6 +185,36 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
   const msgCount = translatedBody.messages?.length || translatedBody.input?.length || translatedBody.contents?.length || translatedBody.request?.contents?.length || 0;
   log?.debug?.("REQUEST", `${provider.toUpperCase()} | ${model} | ${msgCount} msgs`);
 
+  // Validate credentials before sending request - try refresh if needed
+  if (!executor.noAuth && credentials?.accessToken) {
+    const needsRefresh = executor.needsRefresh?.(credentials) ?? shouldRefreshCredentials(provider, credentials);
+    if (needsRefresh) {
+      try {
+        const newCredentials = await refreshWithRetry(() => executor.refreshCredentials(credentials, log), 3, log);
+        if (newCredentials?.accessToken || newCredentials?.copilotToken) {
+          log?.info?.("TOKEN", `${provider.toUpperCase()} | refreshed before request`);
+          Object.assign(credentials, newCredentials);
+          if (onCredentialsRefreshed) {
+            try { await onCredentialsRefreshed(newCredentials); } catch (e) { log?.warn?.("TOKEN", `onCredentialsRefreshed failed: ${e.message}`); }
+          }
+        } else if (newCredentials?.error === "unrecoverable_refresh_error") {
+          const msg = newCredentials.message || `${provider} token invalid; reconnect the account`;
+          log?.warn?.("TOKEN", `${provider.toUpperCase()} | unrecoverable refresh error: ${msg}`);
+          trackPendingRequest(model, provider, connectionId, false, true);
+          return createErrorResult(HTTP_STATUS.UNAUTHORIZED, msg);
+        } else {
+          log?.warn?.("TOKEN", `${provider.toUpperCase()} | refresh failed before request`);
+          trackPendingRequest(model, provider, connectionId, false, true);
+          return createErrorResult(HTTP_STATUS.UNAUTHORIZED, `${provider} token refresh failed. Please reconnect the account.`);
+        }
+      } catch (e) {
+        log?.warn?.("TOKEN", `${provider.toUpperCase()} | refresh threw before request: ${e.message}`);
+        trackPendingRequest(model, provider, connectionId, false, true);
+        return createErrorResult(HTTP_STATUS.UNAUTHORIZED, `${provider} token refresh failed: ${e.message}`);
+      }
+    }
+  }
+
   const streamController = createStreamController({
     onDisconnect: (reason) => {
       trackPendingRequest(model, provider, connectionId, false);
@@ -241,7 +271,7 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
     trackPendingRequest(model, provider, connectionId, false, true);
     appendRequestLog({ model, provider, connectionId, status: `FAILED ${error.name === "AbortError" ? 499 : HTTP_STATUS.BAD_GATEWAY}` }).catch(() => { });
     saveRequestDetail(buildRequestDetail({
-      provider, model, connectionId,
+      provider, model, connectionId, apiKey, comboName, fallbackHistory,
       latency: { ttft: 0, total: Date.now() - requestStartTime },
       tokens: { prompt_tokens: 0, completion_tokens: 0 },
       request: extractRequestConfig(body, stream),
@@ -249,6 +279,7 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
       response: { error: error.message || String(error), status: error.name === "AbortError" ? 499 : 502, thinking: null },
       status: "error"
     })).catch(() => { });
+    saveUsageStats({ provider, model, tokens: { prompt_tokens: 0, completion_tokens: 0 }, connectionId, apiKey, endpoint: clientRawRequest?.endpoint, comboName, fallbackHistory, status: "error", label: "ERROR" });
 
     if (error.name === "AbortError") {
       streamController.handleError(error);
@@ -287,7 +318,7 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
     const { statusCode, message, resetsAtMs } = await parseUpstreamError(providerResponse, executor);
     appendRequestLog({ model, provider, connectionId, status: `FAILED ${statusCode}` }).catch(() => { });
     saveRequestDetail(buildRequestDetail({
-      provider, model, connectionId,
+      provider, model, connectionId, apiKey, comboName, fallbackHistory,
       latency: { ttft: 0, total: Date.now() - requestStartTime },
       tokens: { prompt_tokens: 0, completion_tokens: 0 },
       request: extractRequestConfig(body, stream),
@@ -295,6 +326,7 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
       response: { error: message, status: statusCode, thinking: null },
       status: "error"
     })).catch(() => { });
+    saveUsageStats({ provider, model, tokens: { prompt_tokens: 0, completion_tokens: 0 }, connectionId, apiKey, endpoint: clientRawRequest?.endpoint, comboName, fallbackHistory, status: "error", label: "ERROR" });
 
     const errMsg = formatProviderError(new Error(message), provider, model, statusCode);
     console.log(`${COLORS.red}[ERROR] ${errMsg}${COLORS.reset}`);
@@ -302,7 +334,7 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
     return createErrorResult(statusCode, errMsg, resetsAtMs);
   }
 
-  const sharedCtx = { provider, model, body, stream, translatedBody, finalBody, requestStartTime, connectionId, apiKey, clientRawRequest, onRequestSuccess, midStreamError, onMidStreamError };
+  const sharedCtx = { provider, model, body, stream, translatedBody, finalBody, requestStartTime, connectionId, apiKey, comboName, fallbackHistory, clientRawRequest, onRequestSuccess, midStreamError, onMidStreamError };
   const appendLog = (extra) => appendRequestLog({ model, provider, connectionId, ...extra }).catch(() => { });
   const trackDone = () => trackPendingRequest(model, provider, connectionId, false);