Replay: 12 upstream features (routing, retries, schema fixes)

.github/workflows/release.yaml

@@ -37,3 +37,38 @@ jobs:
           VERSION: ${{ env.VERSION }}
           COMMIT: ${{ env.COMMIT }}
           BUILD_DATE: ${{ env.BUILD_DATE }}
+
+  build-termux:
+    name: Build Termux (aarch64)
+    runs-on: ubuntu-24.04-arm
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+      - name: Build in Termux Container
+        run: |
+          # Prepare metadata on host
+          VERSION=$(git describe --tags --always --dirty | sed 's/^v//')
+          COMMIT=$(git rev-parse --short HEAD)
+          BUILD_DATE=$(date -u +%Y-%m-%dT%H:%M:%SZ)
+
+          # Ensure the workspace is writable by the container
+          chmod -R 777 .
+
+          # Run the build inside Termux container
+          docker run --rm -v $(pwd):/workspace -w /workspace \
+            -e VERSION=$VERSION -e COMMIT=$COMMIT -e BUILD_DATE=$BUILD_DATE \
+            termux/termux-docker:aarch64 bash -c "
+              pkg update -y && pkg upgrade -y -o Dpkg::Options::='--force-confdef' -o Dpkg::Options::='--force-confold'
+              pkg install -y golang build-essential tar
+              CGO_ENABLED=0 go build -ldflags \"-s -w -X 'main.Version=\$VERSION' -X 'main.Commit=\$COMMIT' -X 'main.BuildDate=\$BUILD_DATE'\" -o cli-proxy-api ./cmd/server
+              tar -czf cli-proxy-api-termux-aarch64.tar.gz cli-proxy-api LICENSE README.md README_CN.md config.example.yaml
+            "
+      - name: Upload to Release
+        uses: softprops/action-gh-release@v2
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: cli-proxy-api-termux-aarch64.tar.gz
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

internal/api/handlers/management/config_basic.go

@@ -286,6 +286,8 @@ func normalizeRoutingStrategy(strategy string) (string, bool) {
 		return "round-robin", true
 	case "fill-first", "fillfirst", "ff":
 		return "fill-first", true
+	case "sticky-round-robin", "stickyroundrobin", "srr":
+		return "sticky-round-robin", true
 	default:
 		return "", false
 	}

internal/config/config.go

@@ -208,7 +208,7 @@ type QuotaExceeded struct {
 // RoutingConfig configures how credentials are selected for requests.
 type RoutingConfig struct {
 	// Strategy selects the credential selection strategy.
-	// Supported values: "round-robin" (default), "fill-first".
+	// Supported values: "round-robin" (default), "fill-first", "sticky-round-robin".
 	Strategy string `yaml:"strategy,omitempty" json:"strategy,omitempty"`
 }
 

internal/runtime/executor/antigravity_executor.go

@@ -54,8 +54,78 @@ const (
 var (
 	randSource      = rand.New(rand.NewSource(time.Now().UnixNano()))
 	randSourceMutex sync.Mutex
+	// antigravityPrimaryModelsCache keeps the latest non-empty model list fetched
+	// from any antigravity auth. Empty fetches never overwrite this cache.
+	antigravityPrimaryModelsCache struct {
+		mu     sync.RWMutex
+		models []*registry.ModelInfo
+	}
 )
 
+func cloneAntigravityModels(models []*registry.ModelInfo) []*registry.ModelInfo {
+	if len(models) == 0 {
+		return nil
+	}
+	out := make([]*registry.ModelInfo, 0, len(models))
+	for _, model := range models {
+		if model == nil || strings.TrimSpace(model.ID) == "" {
+			continue
+		}
+		out = append(out, cloneAntigravityModelInfo(model))
+	}
+	if len(out) == 0 {
+		return nil
+	}
+	return out
+}
+
+func cloneAntigravityModelInfo(model *registry.ModelInfo) *registry.ModelInfo {
+	if model == nil {
+		return nil
+	}
+	clone := *model
+	if len(model.SupportedGenerationMethods) > 0 {
+		clone.SupportedGenerationMethods = append([]string(nil), model.SupportedGenerationMethods...)
+	}
+	if len(model.SupportedParameters) > 0 {
+		clone.SupportedParameters = append([]string(nil), model.SupportedParameters...)
+	}
+	if model.Thinking != nil {
+		thinkingClone := *model.Thinking
+		if len(model.Thinking.Levels) > 0 {
+			thinkingClone.Levels = append([]string(nil), model.Thinking.Levels...)
+		}
+		clone.Thinking = &thinkingClone
+	}
+	return &clone
+}
+
+func storeAntigravityPrimaryModels(models []*registry.ModelInfo) bool {
+	cloned := cloneAntigravityModels(models)
+	if len(cloned) == 0 {
+		return false
+	}
+	antigravityPrimaryModelsCache.mu.Lock()
+	antigravityPrimaryModelsCache.models = cloned
+	antigravityPrimaryModelsCache.mu.Unlock()
+	return true
+}
+
+func loadAntigravityPrimaryModels() []*registry.ModelInfo {
+	antigravityPrimaryModelsCache.mu.RLock()
+	cloned := cloneAntigravityModels(antigravityPrimaryModelsCache.models)
+	antigravityPrimaryModelsCache.mu.RUnlock()
+	return cloned
+}
+
+func fallbackAntigravityPrimaryModels() []*registry.ModelInfo {
+	models := loadAntigravityPrimaryModels()
+	if len(models) > 0 {
+		log.Debugf("antigravity executor: using cached primary model list (%d models)", len(models))
+	}
+	return models
+}
+
 // AntigravityExecutor proxies requests to the antigravity upstream.
 type AntigravityExecutor struct {
 	cfg *config.Config
@@ -1006,13 +1076,8 @@ func (e *AntigravityExecutor) CountTokens(ctx context.Context, auth *cliproxyaut
 func FetchAntigravityModels(ctx context.Context, auth *cliproxyauth.Auth, cfg *config.Config) []*registry.ModelInfo {
 	exec := &AntigravityExecutor{cfg: cfg}
 	token, updatedAuth, errToken := exec.ensureAccessToken(ctx, auth)
-	if errToken != nil {
-		log.Warnf("antigravity executor: fetch models failed for %s: token error: %v", auth.ID, errToken)
-		return nil
-	}
-	if token == "" {
-		log.Warnf("antigravity executor: fetch models failed for %s: got empty token", auth.ID)
-		return nil
+	if errToken != nil || token == "" {
+		return fallbackAntigravityPrimaryModels()
 	}
 	if updatedAuth != nil {
 		auth = updatedAuth
@@ -1025,8 +1090,7 @@ func FetchAntigravityModels(ctx context.Context, auth *cliproxyauth.Auth, cfg *c
 		modelsURL := baseURL + antigravityModelsPath
 		httpReq, errReq := http.NewRequestWithContext(ctx, http.MethodPost, modelsURL, bytes.NewReader([]byte(`{}`)))
 		if errReq != nil {
-			log.Warnf("antigravity executor: fetch models failed for %s: create request error: %v", auth.ID, errReq)
-			return nil
+			return fallbackAntigravityPrimaryModels()
 		}
 		httpReq.Header.Set("Content-Type", "application/json")
 		httpReq.Header.Set("Authorization", "Bearer "+token)
@@ -1038,15 +1102,13 @@ func FetchAntigravityModels(ctx context.Context, auth *cliproxyauth.Auth, cfg *c
 		httpResp, errDo := httpClient.Do(httpReq)
 		if errDo != nil {
 			if errors.Is(errDo, context.Canceled) || errors.Is(errDo, context.DeadlineExceeded) {
-				log.Warnf("antigravity executor: fetch models failed for %s: context canceled: %v", auth.ID, errDo)
-				return nil
+				return fallbackAntigravityPrimaryModels()
 			}
 			if idx+1 < len(baseURLs) {
 				log.Debugf("antigravity executor: models request error on base url %s, retrying with fallback base url: %s", baseURL, baseURLs[idx+1])
 				continue
 			}
-			log.Warnf("antigravity executor: fetch models failed for %s: request error: %v", auth.ID, errDo)
-			return nil
+			return fallbackAntigravityPrimaryModels()
 		}
 
 		bodyBytes, errRead := io.ReadAll(httpResp.Body)
@@ -1058,22 +1120,27 @@ func FetchAntigravityModels(ctx context.Context, auth *cliproxyauth.Auth, cfg *c
 				log.Debugf("antigravity executor: models read error on base url %s, retrying with fallback base url: %s", baseURL, baseURLs[idx+1])
 				continue
 			}
-			log.Warnf("antigravity executor: fetch models failed for %s: read body error: %v", auth.ID, errRead)
-			return nil
+			return fallbackAntigravityPrimaryModels()
 		}
 		if httpResp.StatusCode < http.StatusOK || httpResp.StatusCode >= http.StatusMultipleChoices {
 			if httpResp.StatusCode == http.StatusTooManyRequests && idx+1 < len(baseURLs) {
 				log.Debugf("antigravity executor: models request rate limited on base url %s, retrying with fallback base url: %s", baseURL, baseURLs[idx+1])
 				continue
 			}
-			log.Warnf("antigravity executor: fetch models failed for %s: unexpected status %d, body: %s", auth.ID, httpResp.StatusCode, string(bodyBytes))
-			return nil
+			if idx+1 < len(baseURLs) {
+				log.Debugf("antigravity executor: models request failed with status %d on base url %s, retrying with fallback base url: %s", httpResp.StatusCode, baseURL, baseURLs[idx+1])
+				continue
+			}
+			return fallbackAntigravityPrimaryModels()
 		}
 
 		result := gjson.GetBytes(bodyBytes, "models")
 		if !result.Exists() {
-			log.Warnf("antigravity executor: fetch models failed for %s: no models field in response, body: %s", auth.ID, string(bodyBytes))
-			return nil
+			if idx+1 < len(baseURLs) {
+				log.Debugf("antigravity executor: models field missing on base url %s, retrying with fallback base url: %s", baseURL, baseURLs[idx+1])
+				continue
+			}
+			return fallbackAntigravityPrimaryModels()
 		}
 
 		now := time.Now().Unix()
@@ -1118,9 +1185,10 @@ func FetchAntigravityModels(ctx context.Context, auth *cliproxyauth.Auth, cfg *c
 			}
 			models = append(models, modelInfo)
 		}
+		storeAntigravityPrimaryModels(models)
 		return models
 	}
-	return nil
+	return fallbackAntigravityPrimaryModels()
 }
 
 func (e *AntigravityExecutor) ensureAccessToken(ctx context.Context, auth *cliproxyauth.Auth) (string, *cliproxyauth.Auth, error) {

internal/runtime/executor/antigravity_executor_models_cache_test.go

@@ -0,0 +1,90 @@
+package executor
+
+import (
+	"testing"
+
+	"github.com/kooshapari/cliproxyapi-plusplus/v6/internal/registry"
+)
+
+func resetAntigravityPrimaryModelsCacheForTest() {
+	antigravityPrimaryModelsCache.mu.Lock()
+	antigravityPrimaryModelsCache.models = nil
+	antigravityPrimaryModelsCache.mu.Unlock()
+}
+
+func TestStoreAntigravityPrimaryModels_EmptyDoesNotOverwrite(t *testing.T) {
+	resetAntigravityPrimaryModelsCacheForTest()
+	t.Cleanup(resetAntigravityPrimaryModelsCacheForTest)
+
+	seed := []*registry.ModelInfo{
+		{ID: "claude-sonnet-4-5"},
+		{ID: "gemini-2.5-pro"},
+	}
+	if updated := storeAntigravityPrimaryModels(seed); !updated {
+		t.Fatal("expected non-empty model list to update primary cache")
+	}
+
+	if updated := storeAntigravityPrimaryModels(nil); updated {
+		t.Fatal("expected nil model list not to overwrite primary cache")
+	}
+	if updated := storeAntigravityPrimaryModels([]*registry.ModelInfo{}); updated {
+		t.Fatal("expected empty model list not to overwrite primary cache")
+	}
+
+	got := loadAntigravityPrimaryModels()
+	if len(got) != 2 {
+		t.Fatalf("expected cached model count 2, got %d", len(got))
+	}
+	if got[0].ID != "claude-sonnet-4-5" || got[1].ID != "gemini-2.5-pro" {
+		t.Fatalf("unexpected cached model ids: %q, %q", got[0].ID, got[1].ID)
+	}
+}
+
+func TestLoadAntigravityPrimaryModels_ReturnsClone(t *testing.T) {
+	resetAntigravityPrimaryModelsCacheForTest()
+	t.Cleanup(resetAntigravityPrimaryModelsCacheForTest)
+
+	if updated := storeAntigravityPrimaryModels([]*registry.ModelInfo{{
+		ID:                         "gpt-5",
+		DisplayName:                "GPT-5",
+		SupportedGenerationMethods: []string{"generateContent"},
+		SupportedParameters:        []string{"temperature"},
+		Thinking: &registry.ThinkingSupport{
+			Levels: []string{"high"},
+		},
+	}}); !updated {
+		t.Fatal("expected model cache update")
+	}
+
+	got := loadAntigravityPrimaryModels()
+	if len(got) != 1 {
+		t.Fatalf("expected one cached model, got %d", len(got))
+	}
+	got[0].ID = "mutated-id"
+	if len(got[0].SupportedGenerationMethods) > 0 {
+		got[0].SupportedGenerationMethods[0] = "mutated-method"
+	}
+	if len(got[0].SupportedParameters) > 0 {
+		got[0].SupportedParameters[0] = "mutated-parameter"
+	}
+	if got[0].Thinking != nil && len(got[0].Thinking.Levels) > 0 {
+		got[0].Thinking.Levels[0] = "mutated-level"
+	}
+
+	again := loadAntigravityPrimaryModels()
+	if len(again) != 1 {
+		t.Fatalf("expected one cached model after mutation, got %d", len(again))
+	}
+	if again[0].ID != "gpt-5" {
+		t.Fatalf("expected cached model id to remain %q, got %q", "gpt-5", again[0].ID)
+	}
+	if len(again[0].SupportedGenerationMethods) == 0 || again[0].SupportedGenerationMethods[0] != "generateContent" {
+		t.Fatalf("expected cached generation methods to be unmutated, got %v", again[0].SupportedGenerationMethods)
+	}
+	if len(again[0].SupportedParameters) == 0 || again[0].SupportedParameters[0] != "temperature" {
+		t.Fatalf("expected cached supported parameters to be unmutated, got %v", again[0].SupportedParameters)
+	}
+	if again[0].Thinking == nil || len(again[0].Thinking.Levels) == 0 || again[0].Thinking.Levels[0] != "high" {
+		t.Fatalf("expected cached model thinking levels to be unmutated, got %v", again[0].Thinking)
+	}
+}