feat(openai): add responses/compact support

sdk/api/handlers/openai/openai_responses_compact_test.go

@@ -0,0 +1,120 @@
+package openai
+
+import (
+	"context"
+	"errors"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+
+	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers"
+	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
+	coreexecutor "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+	sdkconfig "github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
+)
+
+type compactCaptureExecutor struct {
+	alt          string
+	sourceFormat string
+	calls        int
+}
+
+func (e *compactCaptureExecutor) Identifier() string { return "test-provider" }
+
+func (e *compactCaptureExecutor) Execute(ctx context.Context, auth *coreauth.Auth, req coreexecutor.Request, opts coreexecutor.Options) (coreexecutor.Response, error) {
+	e.calls++
+	e.alt = opts.Alt
+	e.sourceFormat = opts.SourceFormat.String()
+	return coreexecutor.Response{Payload: []byte(`{"ok":true}`)}, nil
+}
+
+func (e *compactCaptureExecutor) ExecuteStream(context.Context, *coreauth.Auth, coreexecutor.Request, coreexecutor.Options) (<-chan coreexecutor.StreamChunk, error) {
+	return nil, errors.New("not implemented")
+}
+
+func (e *compactCaptureExecutor) Refresh(ctx context.Context, auth *coreauth.Auth) (*coreauth.Auth, error) {
+	return auth, nil
+}
+
+func (e *compactCaptureExecutor) CountTokens(context.Context, *coreauth.Auth, coreexecutor.Request, coreexecutor.Options) (coreexecutor.Response, error) {
+	return coreexecutor.Response{}, errors.New("not implemented")
+}
+
+func (e *compactCaptureExecutor) HttpRequest(context.Context, *coreauth.Auth, *http.Request) (*http.Response, error) {
+	return nil, errors.New("not implemented")
+}
+
+func TestOpenAIResponsesCompactRejectsStream(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	executor := &compactCaptureExecutor{}
+	manager := coreauth.NewManager(nil, nil, nil)
+	manager.RegisterExecutor(executor)
+
+	auth := &coreauth.Auth{ID: "auth1", Provider: executor.Identifier(), Status: coreauth.StatusActive}
+	if _, err := manager.Register(context.Background(), auth); err != nil {
+		t.Fatalf("Register auth: %v", err)
+	}
+	registry.GetGlobalRegistry().RegisterClient(auth.ID, auth.Provider, []*registry.ModelInfo{{ID: "test-model"}})
+	t.Cleanup(func() {
+		registry.GetGlobalRegistry().UnregisterClient(auth.ID)
+	})
+
+	base := handlers.NewBaseAPIHandlers(&sdkconfig.SDKConfig{}, manager)
+	h := NewOpenAIResponsesAPIHandler(base)
+	router := gin.New()
+	router.POST("/v1/responses/compact", h.Compact)
+
+	req := httptest.NewRequest(http.MethodPost, "/v1/responses/compact", strings.NewReader(`{"model":"test-model","stream":true}`))
+	req.Header.Set("Content-Type", "application/json")
+	resp := httptest.NewRecorder()
+	router.ServeHTTP(resp, req)
+
+	if resp.Code != http.StatusBadRequest {
+		t.Fatalf("status = %d, want %d", resp.Code, http.StatusBadRequest)
+	}
+	if executor.calls != 0 {
+		t.Fatalf("executor calls = %d, want 0", executor.calls)
+	}
+}
+
+func TestOpenAIResponsesCompactExecute(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	executor := &compactCaptureExecutor{}
+	manager := coreauth.NewManager(nil, nil, nil)
+	manager.RegisterExecutor(executor)
+
+	auth := &coreauth.Auth{ID: "auth2", Provider: executor.Identifier(), Status: coreauth.StatusActive}
+	if _, err := manager.Register(context.Background(), auth); err != nil {
+		t.Fatalf("Register auth: %v", err)
+	}
+	registry.GetGlobalRegistry().RegisterClient(auth.ID, auth.Provider, []*registry.ModelInfo{{ID: "test-model"}})
+	t.Cleanup(func() {
+		registry.GetGlobalRegistry().UnregisterClient(auth.ID)
+	})
+
+	base := handlers.NewBaseAPIHandlers(&sdkconfig.SDKConfig{}, manager)
+	h := NewOpenAIResponsesAPIHandler(base)
+	router := gin.New()
+	router.POST("/v1/responses/compact", h.Compact)
+
+	req := httptest.NewRequest(http.MethodPost, "/v1/responses/compact", strings.NewReader(`{"model":"test-model","input":"hello"}`))
+	req.Header.Set("Content-Type", "application/json")
+	resp := httptest.NewRecorder()
+	router.ServeHTTP(resp, req)
+
+	if resp.Code != http.StatusOK {
+		t.Fatalf("status = %d, want %d", resp.Code, http.StatusOK)
+	}
+	if executor.alt != "responses/compact" {
+		t.Fatalf("alt = %q, want %q", executor.alt, "responses/compact")
+	}
+	if executor.sourceFormat != "openai-response" {
+		t.Fatalf("source format = %q, want %q", executor.sourceFormat, "openai-response")
+	}
+	if strings.TrimSpace(resp.Body.String()) != `{"ok":true}` {
+		t.Fatalf("body = %s", resp.Body.String())
+	}
+}

sdk/api/handlers/openai/openai_responses_handlers.go

@@ -91,6 +91,44 @@ func (h *OpenAIResponsesAPIHandler) Responses(c *gin.Context) {
 
 }
 
+func (h *OpenAIResponsesAPIHandler) Compact(c *gin.Context) {
+	rawJSON, err := c.GetRawData()
+	if err != nil {
+		c.JSON(http.StatusBadRequest, handlers.ErrorResponse{
+			Error: handlers.ErrorDetail{
+				Message: fmt.Sprintf("Invalid request: %v", err),
+				Type:    "invalid_request_error",
+			},
+		})
+		return
+	}
+
+	streamResult := gjson.GetBytes(rawJSON, "stream")
+	if streamResult.Type == gjson.True {
+		c.JSON(http.StatusBadRequest, handlers.ErrorResponse{
+			Error: handlers.ErrorDetail{
+				Message: "Streaming not supported for compact responses",
+				Type:    "invalid_request_error",
+			},
+		})
+		return
+	}
+
+	c.Header("Content-Type", "application/json")
+	modelName := gjson.GetBytes(rawJSON, "model").String()
+	cliCtx, cliCancel := h.GetContextWithCancel(h, c, context.Background())
+	stopKeepAlive := h.StartNonStreamingKeepAlive(c, cliCtx)
+	resp, errMsg := h.ExecuteWithAuthManager(cliCtx, h.HandlerType(), modelName, rawJSON, "responses/compact")
+	stopKeepAlive()
+	if errMsg != nil {
+		h.WriteErrorResponse(c, errMsg)
+		cliCancel(errMsg.Error)
+		return
+	}
+	_, _ = c.Writer.Write(resp)
+	cliCancel()
+}
+
 // handleNonStreamingResponse handles non-streaming chat completion responses
 // for Gemini models. It selects a client from the pool, sends the request, and
 // aggregates the response before sending it back to the client in OpenAIResponses format.