Node.js环境下Gemini AI集成history参数报错求助

问题现象 #

• 复制Gemini官方文档代码到Node.js服务器后，使用history参数触发错误，将history设为null时操作可正常执行
• 错误提示：{"error": "Cannot use 'in' operator to search for 'text' in H"}

出错代码片段 #

history: [
  {
    role: "user",
    parts: : "Hello, I have 2 dogs in my house.",
  },
  {
    role: "model",
    parts: "Great to meet you. What would you like to know?",
  },
],
generationConfig: {
  maxOutputTokens: 100,
},

错误栈信息 #

TypeError: Cannot use 'in' operator to search for 'text' in H
    at validateChatHistory (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\@google\generative-ai\dist\index.js:760:25)
    at new ChatSession (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\@google\generative-ai\dist\index.js:816:13)
    at GenerativeModel.startChat (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\@google\generative-ai\dist\index.js:1035:16)
    at C:\myprojects\chat_app_generativeai_dartfrog\server\routes\chat.js:21:22
    at Layer.handle [as handle_request] (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\express\lib\router\layer.js:95:5)
    at next (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\express\lib\router\route.js:149:13)
    at Route.dispatch (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\express\lib\router\route.js:119:3)
    at Layer.handle [as handle_request] (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\express\lib\router\layer.js:95:5)
    at C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\express\lib\router\index.js:284:15
    at Function.process_params (C:\myprojects\chat_app_generativeai_dartfrog\server\node_modules\express\lib\router\index.js:346:12)

解决方法 #

1. 修复语法错误：代码中parts: : "Hello..."多了一个冒号，这是JS语法错误，导致SDK无法正确解析parts字段，进而在校验聊天历史时抛出异常。修正后去掉多余的冒号即可。
2. 确认parts格式合规：Gemini SDK允许parts直接传字符串，或者传包含text字段的对象数组，两种格式都能正常识别。

修正后的代码（字符串格式） #

history: [
  {
    role: "user",
    parts: "Hello, I have 2 dogs in my house.",
  },
  {
    role: "model",
    parts: "Great to meet you. What would you like to know?",
  },
],
generationConfig: {
  maxOutputTokens: 100,
},

修正后的代码（对象数组格式） #

history: [
  {
    role: "user",
    parts: [{ text: "Hello, I have 2 dogs in my house." }],
  },
  {
    role: "model",
    parts: [{ text: "Great to meet you. What would you like to know?" }],
  },
],
generationConfig: {
  maxOutputTokens: 100,
},

内容的提问来源于stack exchange，提问作者sina zp