feat(kiro): 代码优化重构 + OpenAI翻译器实现

internal/api/modules/amp/response_rewriter.go

@@ -29,15 +29,71 @@ func NewResponseRewriter(w gin.ResponseWriter, originalModel string) *ResponseRe
 	}
 }
 
+const maxBufferedResponseBytes = 2 * 1024 * 1024 // 2MB safety cap
+
+func looksLikeSSEChunk(data []byte) bool {
+	// Fallback detection: some upstreams may omit/lie about Content-Type, causing SSE to be buffered.
+	// Heuristics are intentionally simple and cheap.
+	return bytes.Contains(data, []byte("data:")) ||
+		bytes.Contains(data, []byte("event:")) ||
+		bytes.Contains(data, []byte("message_start")) ||
+		bytes.Contains(data, []byte("message_delta")) ||
+		bytes.Contains(data, []byte("content_block_start")) ||
+		bytes.Contains(data, []byte("content_block_delta")) ||
+		bytes.Contains(data, []byte("content_block_stop")) ||
+		bytes.Contains(data, []byte("\n\n"))
+}
+
+func (rw *ResponseRewriter) enableStreaming(reason string) error {
+	if rw.isStreaming {
+		return nil
+	}
+	rw.isStreaming = true
+
+	// Flush any previously buffered data to avoid reordering or data loss.
+	if rw.body != nil && rw.body.Len() > 0 {
+		buf := rw.body.Bytes()
+		// Copy before Reset() to keep bytes stable.
+		toFlush := make([]byte, len(buf))
+		copy(toFlush, buf)
+		rw.body.Reset()
+
+		if _, err := rw.ResponseWriter.Write(rw.rewriteStreamChunk(toFlush)); err != nil {
+			return err
+		}
+		if flusher, ok := rw.ResponseWriter.(http.Flusher); ok {
+			flusher.Flush()
+		}
+	}
+
+	log.Debugf("amp response rewriter: switched to streaming (%s)", reason)
+	return nil
+}
+
 // Write intercepts response writes and buffers them for model name replacement
 func (rw *ResponseRewriter) Write(data []byte) (int, error) {
-	// Detect streaming on first write
-	if rw.body.Len() == 0 && !rw.isStreaming {
+	// Detect streaming on first write (header-based)
+	if !rw.isStreaming && rw.body.Len() == 0 {
 		contentType := rw.Header().Get("Content-Type")
 		rw.isStreaming = strings.Contains(contentType, "text/event-stream") ||
 			strings.Contains(contentType, "stream")
 	}
 
+	if !rw.isStreaming {
+		// Content-based fallback: detect SSE-like chunks even if Content-Type is missing/wrong.
+		if looksLikeSSEChunk(data) {
+			if err := rw.enableStreaming("sse heuristic"); err != nil {
+				return 0, err
+			}
+		} else if rw.body.Len()+len(data) > maxBufferedResponseBytes {
+			// Safety cap: avoid unbounded buffering on large responses.
+			log.Warnf("amp response rewriter: buffer exceeded %d bytes, switching to streaming", maxBufferedResponseBytes)
+			if err := rw.enableStreaming("buffer limit"); err != nil {
+				return 0, err
+			}
+		}
+	}
+
 	if rw.isStreaming {
 		return rw.ResponseWriter.Write(rw.rewriteStreamChunk(data))
 	}