feat: sync deployed v1.0.0-rc.38-custom.42

Sync production source including routing, billing, Astra detection and public uptime. Update custom build instructions and license packaging.

router/relay-router.go

@@ -1,6 +1,8 @@
 package router
 
 import (
+	"strings"
+
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/controller"
 	"github.com/QuantumNous/new-api/middleware"
@@ -70,6 +72,11 @@ func SetRelayRouter(router *gin.Engine) {
 	relayV1Router.Use(middleware.RouteTag("relay"))
 	relayV1Router.Use(middleware.SystemPerformanceCheck())
 	relayV1Router.Use(middleware.TokenAuth())
+	{
+		// Responses WebSocket route. Channel selection happens after the first
+		// response.create event; each event runs the ordinary request limiter.
+		relayV1Router.GET("/responses", controller.ResponsesWebSocket)
+	}
 	relayV1Router.Use(middleware.ModelRequestRateLimit())
 	{
 		// WebSocket 路由（统一到 Relay）
@@ -181,6 +188,15 @@ func SetRelayRouter(router *gin.Engine) {
 	//relayMjRouter.Use()
 
 	relayGeminiRouter := router.Group("/v1beta")
+	// :countTokens is not implemented. Answer it like an unregistered route
+	// before auth/channel selection instead of silently relaying it as
+	// generateContent (#7283).
+	relayGeminiRouter.Use(func(c *gin.Context) {
+		if strings.HasSuffix(c.Request.URL.Path, ":countTokens") {
+			controller.RelayNotFound(c)
+			c.Abort()
+		}
+	})
 	relayGeminiRouter.Use(middleware.RouteTag("relay"))
 	relayGeminiRouter.Use(middleware.SystemPerformanceCheck())
 	relayGeminiRouter.Use(middleware.TokenAuth())