Skip to content

AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and support for your own .pptx templates. · by Hugo He

feat(tools): nano_banana_gen Official Mode 改用 generate_content_stream

hugohe380cd5d0ba22b
diff --git a/tools/nano_banana_gen.py b/tools/nano_banana_gen.py
index ef15f8c1..13da9073 100644
--- a/tools/nano_banana_gen.py
+++ b/tools/nano_banana_gen.py
@@ -117,7 +117,12 @@ def _generate_official(api_key: str, prompt: str, negative_prompt: str = None,
output_dir: str = None, filename: str = None,
model: str = DEFAULT_MODEL) -> str:
"""
- Official Mode: 使用 Google 官方 GenAI API (非流式)。
+ Official Mode: 使用 Google 官方 GenAI API (流式)。
+
+ 使用 generate_content_stream 实现流式接收,提供实时进度反馈:
+ - 显示已等待时长
+ - 收到 chunk 时显示编号和数据大小
+ - 保留最后一个 image chunk(最高质量)
Returns:
保存的图片文件路径
@@ -153,22 +158,65 @@ def _generate_official(api_key: str, prompt: str, negative_prompt: str = None,
print(f" Image Size: {image_size}")
print()
- response = client.models.generate_content(
+ # Stream response for real-time progress feedback
+ start_time = time.time()
+ print(f" ⏳ Generating...", end="", flush=True)
+
+ # Heartbeat thread: print elapsed time every 5s while waiting
+ import threading
+ heartbeat_stop = threading.Event()
+
+ def _heartbeat():
+ while not heartbeat_stop.is_set():
+ heartbeat_stop.wait(5)
+ if not heartbeat_stop.is_set():
+ elapsed = time.time() - start_time
+ print(f" {elapsed:.0f}s...", end="", flush=True)
+
+ hb_thread = threading.Thread(target=_heartbeat, daemon=True)
+ hb_thread.start()
+
+ last_image_data = None # (PIL.Image or bytes, mime_type)
+ chunk_count = 0
+ total_bytes = 0
+
+ for chunk in client.models.generate_content_stream(
model=model,
contents=[final_prompt],
config=config,
- )
+ ):
+ elapsed = time.time() - start_time
+
+ if chunk.parts is None:
+ continue
- for part in response.parts:
- if part.text is not None:
- print(f" Model says: {part.text}")
- elif part.inline_data is not None:
- image = part.as_image()
- path = _resolve_output_path(prompt, output_dir, filename, ".png")
- image.save(path)
- print(f"File saved to: {path}")
- _report_resolution(path)
- return path
+ for part in chunk.parts:
+ if part.text is not None:
+ print(f"\n Model says: {part.text}", end="", flush=True)
+ elif part.inline_data is not None:
+ chunk_count += 1
+ data_size = len(part.inline_data.data) if part.inline_data.data else 0
+ total_bytes += data_size
+ size_str = f"{data_size / 1024:.0f}KB" if data_size < 1048576 else f"{data_size / 1048576:.1f}MB"
+ print(f"\n 📦 Chunk #{chunk_count} received ({size_str}, {elapsed:.1f}s)", end="", flush=True)
+ last_image_data = part
+
+ # Stop heartbeat
+ heartbeat_stop.set()
+ hb_thread.join(timeout=1)
+
+ elapsed = time.time() - start_time
+ print(f"\n ✅ Stream complete ({elapsed:.1f}s, {chunk_count} chunk(s), {total_bytes / 1024:.0f}KB total)")
+
+ if last_image_data is not None and last_image_data.inline_data is not None:
+ if chunk_count > 1:
+ print(f" Keeping the final chunk (highest quality).")
+ image = last_image_data.as_image()
+ path = _resolve_output_path(prompt, output_dir, filename, ".png")
+ image.save(path)
+ print(f"File saved to: {path}")
+ _report_resolution(path)
+ return path
raise RuntimeError("No image was generated. The server may have refused the request.")
@@ -280,7 +328,7 @@ def generate(prompt: str, negative_prompt: str = None,
根据环境变量 GEMINI_BASE_URL 是否存在,自动选择:
- 有 GEMINI_BASE_URL → Proxy Mode (流式)
- - 无 GEMINI_BASE_URL → Official Mode (非流式)
+ - 无 GEMINI_BASE_URL → Official Mode (流式)
遇到 429 Rate Limit 错误时自动指数退避重试。