{"id":89708,"date":"2026-08-03T14:56:32","date_gmt":"2026-08-03T06:56:32","guid":{"rendered":"https:\/\/www.wsisp.com\/helps\/89708.html"},"modified":"2026-08-03T14:56:32","modified_gmt":"2026-08-03T06:56:32","slug":"langgraph-%e5%a4%9a%e6%99%ba%e8%83%bd%e4%bd%93%e5%ae%9e%e6%88%98%ef%bc%9a%e6%90%ad%e5%bb%ba%e7%94%b5%e5%8a%9b%e6%94%bf%e7%ad%96%e8%b7%9f%e8%b8%aa-agent%ef%bc%8c%e5%ae%9e%e7%8e%b0%e7%bd%91%e9%a1%b5","status":"publish","type":"post","link":"https:\/\/www.wsisp.com\/helps\/89708.html","title":{"rendered":"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790"},"content":{"rendered":"<h2><\/h2>\n<h3>\u6587\u7ae0\u524d\u8a00<\/h3>\n<p>\u968f\u7740\u56fd\u5185\u7535\u529b\u5e02\u573a\u5316\u6539\u9769\u6301\u7eed\u63a8\u8fdb&#xff0c;\u5404\u7701\u7535\u529b\u4ea4\u6613\u4e2d\u5fc3\u6301\u7eed\u53d1\u5e03\u4ea4\u6613\u89c4\u5219\u3001\u653f\u7b56\u6587\u4ef6\u3001\u5e02\u573a\u901a\u77e5\u3002\u7535\u529b\u4ece\u4e1a\u8005\u9700\u8981\u6301\u7eed\u8ddf\u8e2a\u6d77\u91cfPDF\u516c\u544a\u3001\u7f51\u9875\u653f\u7b56\u6587\u672c&#xff0c;\u4eba\u5de5\u6574\u7406\u3001\u5bf9\u6bd4\u653f\u7b56\u5dee\u5f02\u3001\u63d0\u53d6\u6838\u5fc3\u6761\u6b3e\u6210\u672c\u6781\u9ad8\u3002\u4f20\u7edf\u5355\u4e00\u5927\u6a21\u578b\u5e94\u7528\u65e0\u6cd5\u5b8c\u6210\u7f51\u9875\u91c7\u96c6\u2192\u6587\u6863\u89e3\u6790\u2192\u6587\u672c\u62bd\u53d6\u2192\u77e5\u8bc6\u5165\u5e93\u2192\u667a\u80fd\u5206\u6790\u5b8c\u6574\u6d41\u6c34\u7ebf\u3002<\/p>\n<p>\u672c\u6587\u57fa\u4e8eLangGraph\u6784\u5efa\u591a\u667a\u80fd\u4f53\u534f\u540c\u7cfb\u7edf&#xff0c;\u62c6\u5206\u722c\u866b\u667a\u80fd\u4f53\u3001\u6587\u6863\u89e3\u6790\u667a\u80fd\u4f53\u3001\u89c4\u5219\u62bd\u53d6\u667a\u80fd\u4f53\u3001\u77e5\u8bc6\u5e93\u66f4\u65b0\u667a\u80fd\u4f53\u3002\u6574\u5957\u5de5\u4f5c\u6d41\u81ea\u52a8\u5b8c\u6210\u7f51\u9875\u6293\u53d6\u3001PDF\u89e3\u6790\u3001\u56fe\u7247OCR\u8bc6\u522b\u3001\u653f\u7b56\u6761\u6b3e\u7ed3\u6784\u5316\u62bd\u53d6&#xff0c;\u7ed3\u5408RAG\u5411\u91cf\u5b58\u50a8\u5b9e\u73b0\u7535\u529b\u653f\u7b56\u6301\u7eed\u8ddf\u8e2a\u3002\u6574\u5957\u4ee3\u7801\u53ef\u76f4\u63a5\u5de5\u7a0b\u6539\u9020&#xff0c;\u9002\u914d\u80fd\u6e90\u3001\u7535\u529b\u4ea4\u6613\u4e1a\u52a1\u573a\u666f\u3002<\/p>\n<p>\u6280\u672f\u6808&#xff1a;Python3.10&#043; | LangGraph | LangChain | Pytesseract(OCR) | PyPDF2 | BeautifulSoup | Chroma\u5411\u91cf\u5e93 | OpenAI\u517c\u5bb9\u5927\u6a21\u578b\u63a5\u53e3<\/p>\n<h3>\u4e00\u3001\u7cfb\u7edf\u6574\u4f53\u67b6\u6784\u8bbe\u8ba1<\/h3>\n<h4>1.1 \u591a\u667a\u80fd\u4f53\u89d2\u8272\u5212\u5206<\/h4>\n<p>\u6211\u4eec\u5c06\u6574\u4e2a\u4e1a\u52a1\u6d41\u7a0b\u62c6\u89e3\u4e3a4\u4e2a\u72ec\u7acbAgent&#xff0c;\u7531LangGraph\u8d1f\u8d23\u72b6\u6001\u6d41\u8f6c\u3001\u8def\u7531\u8c03\u5ea6&#xff1a;<\/p>\n<li>\u722c\u866bAgent&#xff1a;\u76ee\u6807\u7f51\u9875\u53d1\u8d77\u8bf7\u6c42&#xff0c;\u6293\u53d6\u7535\u529b\u653f\u7b56\u516c\u544a\u94fe\u63a5&#xff0c;\u8fc7\u6ee4\u91cd\u590d\u6587\u6863\u5730\u5740<\/li>\n<li>\u6587\u6863\u89e3\u6790Agent&#xff1a;\u533a\u5206\u7f51\u9875\u6587\u672c\u3001PDF\u6587\u4ef6\u3001\u56fe\u7247\u9644\u4ef6&#xff1b;PDF\u6587\u672c\u63d0\u53d6&#xff0c;\u56fe\u7247\u6267\u884cOCR\u8bc6\u522b<\/li>\n<li>\u89c4\u5219\u62bd\u53d6Agent&#xff1a;\u8c03\u7528\u5927\u6a21\u578b&#xff0c;\u4ece\u539f\u59cb\u6587\u6863\u4e2d\u7ed3\u6784\u5316\u63d0\u53d6\u653f\u7b56\u751f\u6548\u65f6\u95f4\u3001\u4ea4\u6613\u89c4\u5219\u3001\u653f\u7b56\u8981\u70b9\u3001\u53d8\u66f4\u6761\u6b3e<\/li>\n<li>\u77e5\u8bc6\u5e93Agent&#xff1a;\u5c06\u7ed3\u6784\u5316\u6570\u636e\u5411\u91cf\u5316&#xff0c;\u5b58\u5165\u5411\u91cf\u6570\u636e\u5e93&#xff0c;\u5b8c\u6210\u7535\u529b\u4ea4\u6613\u77e5\u8bc6\u5e93\u66f4\u65b0<\/li>\n<h4>1.2 \u5de5\u4f5c\u6d41\u6d41\u8f6c\u903b\u8f91<\/h4>\n<p>\u8d77\u59cb\u8282\u70b9 \u2192 \u722c\u866bAgent \u2192 \u6587\u6863\u4e0b\u8f7d\u5206\u53d1 \u2192 \u6587\u6863\u89e3\u6790Agent&#xff08;\u5206\u652f&#xff1a;\u7f51\u9875\/PDF\/OCR\u56fe\u7247&#xff09;<br \/>\n\u2192 \u89c4\u5219\u62bd\u53d6Agent \u2192 \u77e5\u8bc6\u5e93\u5199\u5165Agent \u2192 \u4efb\u52a1\u7ed3\u675f<br \/>\n\u51fa\u73b0\u5f02\u5e38\u81ea\u52a8\u91cd\u8bd5&#xff0c;\u89e3\u6790\u5931\u8d25\u6587\u6863\u6807\u8bb0&#xff0c;\u5b58\u5165\u5f02\u5e38\u961f\u5217\u3002<\/p>\n<h4>1.3 \u72b6\u6001\u5b9a\u4e49&#xff08;State&#xff09;<\/h4>\n<p>LangGraph\u4f9d\u9760State\u5728\u5404\u4e2a\u8282\u70b9\u4e4b\u95f4\u4f20\u9012\u5168\u5c40\u6570\u636e&#xff0c;\u5305\u542b&#xff1a;\u5f85\u6293\u53d6URL\u5217\u8868\u3001\u539f\u59cb\u6587\u6863\u5185\u5bb9\u3001\u7ed3\u6784\u5316\u653f\u7b56\u4fe1\u606f\u3001\u9519\u8bef\u65e5\u5fd7\u3002<\/p>\n<h3>\u4e8c\u3001\u73af\u5883\u4f9d\u8d56\u5b89\u88c5<\/h3>\n<p>\u5148\u6267\u884c\u4f9d\u8d56\u5305\u5b89\u88c5<\/p>\n<p>pip <span class=\"token function\">install<\/span> langgraph langchain langchain-openai chromadb beautifulsoup4 requests pypdf2 pytesseract python-dotenv<\/p>\n<p>\u6ce8\u610f&#xff1a;OCR\u8bc6\u522b\u9700\u8981\u672c\u5730\u5b89\u88c5Tesseract-OCR\u7a0b\u5e8f&#xff0c;\u5e76\u914d\u7f6e\u73af\u5883\u53d8\u91cf\u3002Windows\u7528\u6237\u4e0b\u8f7d\u5b89\u88c5\u5305&#xff0c;Linux\u6267\u884csudo apt install tesseract-ocr\u3002<\/p>\n<h3>\u4e09\u3001\u5b8c\u6574\u4ee3\u7801\u5b9e\u73b0<\/h3>\n<h4>3.1 \u5168\u5c40\u72b6\u6001\u5b9a\u4e49\u4e0e\u57fa\u7840\u521d\u59cb\u5316<\/h4>\n<p>\u65b0\u5efapower_policy_agent.py<\/p>\n<p><span class=\"token keyword\">import<\/span> os<br \/>\n<span class=\"token keyword\">import<\/span> requests<br \/>\n<span class=\"token keyword\">import<\/span> pytesseract<br \/>\n<span class=\"token keyword\">from<\/span> PyPDF2 <span class=\"token keyword\">import<\/span> PdfReader<br \/>\n<span class=\"token keyword\">from<\/span> bs4 <span class=\"token keyword\">import<\/span> BeautifulSoup<br \/>\n<span class=\"token keyword\">from<\/span> typing <span class=\"token keyword\">import<\/span> TypedDict<span class=\"token punctuation\">,<\/span> Annotated<span class=\"token punctuation\">,<\/span> Sequence<span class=\"token punctuation\">,<\/span> Any<br \/>\n<span class=\"token keyword\">import<\/span> operator<br \/>\n<span class=\"token keyword\">from<\/span> langgraph<span class=\"token punctuation\">.<\/span>graph <span class=\"token keyword\">import<\/span> StateGraph<span class=\"token punctuation\">,<\/span> END<br \/>\n<span class=\"token keyword\">from<\/span> langchain_openai <span class=\"token keyword\">import<\/span> ChatOpenAI<br \/>\n<span class=\"token keyword\">from<\/span> langchain<span class=\"token punctuation\">.<\/span>vectorstores <span class=\"token keyword\">import<\/span> Chroma<br \/>\n<span class=\"token keyword\">from<\/span> langchain<span class=\"token punctuation\">.<\/span>embeddings <span class=\"token keyword\">import<\/span> OpenAIEmbeddings<br \/>\n<span class=\"token keyword\">from<\/span> langchain<span class=\"token punctuation\">.<\/span>prompts <span class=\"token keyword\">import<\/span> PromptTemplate<br \/>\n<span class=\"token keyword\">from<\/span> dotenv <span class=\"token keyword\">import<\/span> load_dotenv<\/p>\n<p>load_dotenv<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><\/p>\n<p><span class=\"token comment\"># \u914d\u7f6e\u5927\u6a21\u578b&#xff0c;\u652f\u6301\u901a\u4e49\u5343\u95ee\u3001DeepSeek\u7b49\u517c\u5bb9OpenAI\u63a5\u53e3\u6a21\u578b<\/span><br \/>\nllm <span class=\"token operator\">&#061;<\/span> ChatOpenAI<span class=\"token punctuation\">(<\/span><br \/>\n    model<span class=\"token operator\">&#061;<\/span><span class=\"token string\">&#034;deepseek-chat&#034;<\/span><span class=\"token punctuation\">,<\/span><br \/>\n    api_key<span class=\"token operator\">&#061;<\/span>os<span class=\"token punctuation\">.<\/span>getenv<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;LLM_API_KEY&#034;<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span><br \/>\n    base_url<span class=\"token operator\">&#061;<\/span>os<span class=\"token punctuation\">.<\/span>getenv<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;LLM_BASE_URL&#034;<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span><br \/>\n    temperature<span class=\"token operator\">&#061;<\/span><span class=\"token number\">0.1<\/span><br \/>\n<span class=\"token punctuation\">)<\/span><\/p>\n<p><span class=\"token comment\"># \u5411\u91cf\u6570\u636e\u5e93\u521d\u59cb\u5316<\/span><br \/>\nembedding <span class=\"token operator\">&#061;<\/span> OpenAIEmbeddings<span class=\"token punctuation\">(<\/span><br \/>\n    model<span class=\"token operator\">&#061;<\/span><span class=\"token string\">&#034;text-embedding&#034;<\/span><span class=\"token punctuation\">,<\/span><br \/>\n    api_key<span class=\"token operator\">&#061;<\/span>os<span class=\"token punctuation\">.<\/span>getenv<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;EMBED_KEY&#034;<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span><br \/>\n    base_url<span class=\"token operator\">&#061;<\/span>os<span class=\"token punctuation\">.<\/span>getenv<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;EMBED_BASE&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n<span class=\"token punctuation\">)<\/span><br \/>\nvector_db <span class=\"token operator\">&#061;<\/span> Chroma<span class=\"token punctuation\">(<\/span>persist_directory<span class=\"token operator\">&#061;<\/span><span class=\"token string\">&#034;.\/power_policy_db&#034;<\/span><span class=\"token punctuation\">,<\/span> embedding_function<span class=\"token operator\">&#061;<\/span>embedding<span class=\"token punctuation\">)<\/span><\/p>\n<p><span class=\"token comment\"># &#061;&#061;&#061;&#061;&#061;&#061;&#061;&#061;&#061;&#061; \u5b9a\u4e49\u5168\u5c40\u72b6\u6001 &#061;&#061;&#061;&#061;&#061;&#061;&#061;&#061;&#061;&#061;<\/span><br \/>\n<span class=\"token keyword\">class<\/span> <span class=\"token class-name\">AgentState<\/span><span class=\"token punctuation\">(<\/span>TypedDict<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">:<\/span><br \/>\n    url_list<span class=\"token punctuation\">:<\/span> Annotated<span class=\"token punctuation\">[<\/span><span class=\"token builtin\">list<\/span><span class=\"token punctuation\">[<\/span><span class=\"token builtin\">str<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span> operator<span class=\"token punctuation\">.<\/span>add<span class=\"token punctuation\">]<\/span>       <span class=\"token comment\"># \u9700\u8981\u6293\u53d6\u7684\u653f\u7b56\u7f51\u9875\u94fe\u63a5<\/span><br \/>\n    raw_documents<span class=\"token punctuation\">:<\/span> Annotated<span class=\"token punctuation\">[<\/span><span class=\"token builtin\">list<\/span><span class=\"token punctuation\">[<\/span><span class=\"token builtin\">dict<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span> operator<span class=\"token punctuation\">.<\/span>add<span class=\"token punctuation\">]<\/span> <span class=\"token comment\"># \u539f\u59cb\u6587\u6863\u6570\u636e {url, content, doc_type}<\/span><br \/>\n    structured_policy<span class=\"token punctuation\">:<\/span> Annotated<span class=\"token punctuation\">[<\/span><span class=\"token builtin\">list<\/span><span class=\"token punctuation\">[<\/span><span class=\"token builtin\">dict<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span> operator<span class=\"token punctuation\">.<\/span>add<span class=\"token punctuation\">]<\/span> <span class=\"token comment\"># \u5927\u6a21\u578b\u62bd\u53d6\u7ed3\u6784\u5316\u653f\u7b56<\/span><br \/>\n    error_log<span class=\"token punctuation\">:<\/span> Annotated<span class=\"token punctuation\">[<\/span><span class=\"token builtin\">list<\/span><span class=\"token punctuation\">[<\/span><span class=\"token builtin\">str<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span> operator<span class=\"token punctuation\">.<\/span>add<span class=\"token punctuation\">]<\/span>      <span class=\"token comment\"># \u9519\u8bef\u65e5\u5fd7<\/span><\/p>\n<h4>3.2 Agent\u8282\u70b91&#xff1a;\u722c\u866b\u667a\u80fd\u4f53&#xff08;CrawlerAgent&#xff09;<\/h4>\n<p>\u8d1f\u8d23\u8bbf\u95ee\u7535\u529b\u4ea4\u6613\u7f51\u7ad9&#xff0c;\u6293\u53d6\u653f\u7b56\u9875\u9762\u5185\u5bb9&#xff0c;\u533a\u5206\u666e\u901a\u7f51\u9875\u3001PDF\u9644\u4ef6\u94fe\u63a5\u3002<\/p>\n<p><span class=\"token keyword\">def<\/span> <span class=\"token function\">crawler_agent<\/span><span class=\"token punctuation\">(<\/span>state<span class=\"token punctuation\">:<\/span> AgentState<span class=\"token punctuation\">)<\/span> <span class=\"token operator\">&#8211;<\/span><span class=\"token operator\">&gt;<\/span> AgentState<span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token triple-quoted-string string\">&#034;&#034;&#034;\u722c\u866b\u667a\u80fd\u4f53&#xff1a;\u904d\u5386url&#xff0c;\u83b7\u53d6\u7f51\u9875\u5185\u5bb9&#xff0c;\u8bc6\u522bPDF\u9644\u4ef6&#034;&#034;&#034;<\/span><br \/>\n    new_raw_docs <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    error <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    headers <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">{<\/span><br \/>\n        <span class=\"token string\">&#034;User-Agent&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#034;Mozilla\/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit\/537.36&#034;<\/span><br \/>\n    <span class=\"token punctuation\">}<\/span><br \/>\n    <span class=\"token keyword\">for<\/span> url <span class=\"token keyword\">in<\/span> state<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;url_list&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">:<\/span><br \/>\n        <span class=\"token keyword\">try<\/span><span class=\"token punctuation\">:<\/span><br \/>\n            resp <span class=\"token operator\">&#061;<\/span> requests<span class=\"token punctuation\">.<\/span>get<span class=\"token punctuation\">(<\/span>url<span class=\"token punctuation\">,<\/span> headers<span class=\"token operator\">&#061;<\/span>headers<span class=\"token punctuation\">,<\/span> timeout<span class=\"token operator\">&#061;<\/span><span class=\"token number\">15<\/span><span class=\"token punctuation\">)<\/span><br \/>\n            resp<span class=\"token punctuation\">.<\/span>raise_for_status<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><br \/>\n            <span class=\"token comment\"># \u5224\u65ad\u662f\u5426\u4e3aPDF\u6587\u4ef6<\/span><br \/>\n            <span class=\"token keyword\">if<\/span> url<span class=\"token punctuation\">.<\/span>endswith<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;.pdf&#034;<\/span><span class=\"token punctuation\">)<\/span> <span class=\"token keyword\">or<\/span> <span class=\"token string\">&#034;application\/pdf&#034;<\/span> <span class=\"token keyword\">in<\/span> resp<span class=\"token punctuation\">.<\/span>headers<span class=\"token punctuation\">.<\/span>get<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;Content-Type&#034;<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;&#034;<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">:<\/span><br \/>\n                new_raw_docs<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">{<\/span><br \/>\n                    <span class=\"token string\">&#034;source_url&#034;<\/span><span class=\"token punctuation\">:<\/span> url<span class=\"token punctuation\">,<\/span><br \/>\n                    <span class=\"token string\">&#034;doc_type&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#034;pdf&#034;<\/span><span class=\"token punctuation\">,<\/span><br \/>\n                    <span class=\"token string\">&#034;binary_data&#034;<\/span><span class=\"token punctuation\">:<\/span> resp<span class=\"token punctuation\">.<\/span>content<span class=\"token punctuation\">,<\/span><br \/>\n                    <span class=\"token string\">&#034;text&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#034;&#034;<\/span><br \/>\n                <span class=\"token punctuation\">}<\/span><span class=\"token punctuation\">)<\/span><br \/>\n            <span class=\"token keyword\">else<\/span><span class=\"token punctuation\">:<\/span><br \/>\n                soup <span class=\"token operator\">&#061;<\/span> BeautifulSoup<span class=\"token punctuation\">(<\/span>resp<span class=\"token punctuation\">.<\/span>text<span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;html.parser&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n                <span class=\"token comment\"># \u6e05\u9664\u7f51\u9875\u6807\u7b7e&#xff0c;\u63d0\u53d6\u6b63\u6587<\/span><br \/>\n                text <span class=\"token operator\">&#061;<\/span> soup<span class=\"token punctuation\">.<\/span>get_text<span class=\"token punctuation\">(<\/span>strip<span class=\"token operator\">&#061;<\/span><span class=\"token boolean\">True<\/span><span class=\"token punctuation\">,<\/span> separator<span class=\"token operator\">&#061;<\/span><span class=\"token string\">&#034;\\\\n&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n                new_raw_docs<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">{<\/span><br \/>\n                    <span class=\"token string\">&#034;source_url&#034;<\/span><span class=\"token punctuation\">:<\/span> url<span class=\"token punctuation\">,<\/span><br \/>\n                    <span class=\"token string\">&#034;doc_type&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token string\">&#034;webpage&#034;<\/span><span class=\"token punctuation\">,<\/span><br \/>\n                    <span class=\"token string\">&#034;binary_data&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token boolean\">None<\/span><span class=\"token punctuation\">,<\/span><br \/>\n                    <span class=\"token string\">&#034;text&#034;<\/span><span class=\"token punctuation\">:<\/span> text<br \/>\n                <span class=\"token punctuation\">}<\/span><span class=\"token punctuation\">)<\/span><br \/>\n        <span class=\"token keyword\">except<\/span> Exception <span class=\"token keyword\">as<\/span> e<span class=\"token punctuation\">:<\/span><br \/>\n            error<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token string-interpolation\"><span class=\"token string\">f&#034;\u6293\u53d6\u5931\u8d25 <\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>url<span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">: <\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span><span class=\"token builtin\">str<\/span><span class=\"token punctuation\">(<\/span>e<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">&#034;<\/span><\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">return<\/span> <span class=\"token punctuation\">{<\/span><span class=\"token string\">&#034;raw_documents&#034;<\/span><span class=\"token punctuation\">:<\/span> new_raw_docs<span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;error_log&#034;<\/span><span class=\"token punctuation\">:<\/span> error<span class=\"token punctuation\">}<\/span><\/p>\n<h4>3.3 Agent\u8282\u70b92&#xff1a;\u6587\u6863\u89e3\u6790\u667a\u80fd\u4f53&#xff08;DocumentParserAgent&#xff09;<\/h4>\n<p>\u5206\u6d41\u5904\u7406\u4e09\u7c7b\u6570\u636e&#xff1a;\u7f51\u9875\u6587\u672c\u76f4\u63a5\u4f7f\u7528&#xff1b;PDF\u6587\u672c\u63d0\u53d6&#xff1b;\u56fe\u7247\u9644\u4ef6\u8c03\u7528OCR\u8bc6\u522b\u3002<\/p>\n<p><span class=\"token keyword\">def<\/span> <span class=\"token function\">document_parser_agent<\/span><span class=\"token punctuation\">(<\/span>state<span class=\"token punctuation\">:<\/span> AgentState<span class=\"token punctuation\">)<\/span> <span class=\"token operator\">&#8211;<\/span><span class=\"token operator\">&gt;<\/span> AgentState<span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token triple-quoted-string string\">&#034;&#034;&#034;\u6587\u6863\u89e3\u6790Agent&#xff1a;PDF\u89e3\u6790 &#043; OCR\u56fe\u7247\u8bc6\u522b&#034;&#034;&#034;<\/span><br \/>\n    processed_docs <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    error <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    <span class=\"token keyword\">for<\/span> doc <span class=\"token keyword\">in<\/span> state<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;raw_documents&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">:<\/span><br \/>\n        <span class=\"token keyword\">try<\/span><span class=\"token punctuation\">:<\/span><br \/>\n            <span class=\"token keyword\">if<\/span> doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;doc_type&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;&#061;<\/span> <span class=\"token string\">&#034;webpage&#034;<\/span><span class=\"token punctuation\">:<\/span><br \/>\n                processed_docs<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span>doc<span class=\"token punctuation\">)<\/span><br \/>\n            <span class=\"token keyword\">elif<\/span> doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;doc_type&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;&#061;<\/span> <span class=\"token string\">&#034;pdf&#034;<\/span><span class=\"token punctuation\">:<\/span><br \/>\n                <span class=\"token comment\"># PDF\u89e3\u6790<\/span><br \/>\n                <span class=\"token keyword\">from<\/span> io <span class=\"token keyword\">import<\/span> BytesIO<br \/>\n                pdf_stream <span class=\"token operator\">&#061;<\/span> BytesIO<span class=\"token punctuation\">(<\/span>doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;binary_data&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">)<\/span><br \/>\n                reader <span class=\"token operator\">&#061;<\/span> PdfReader<span class=\"token punctuation\">(<\/span>pdf_stream<span class=\"token punctuation\">)<\/span><br \/>\n                full_text <span class=\"token operator\">&#061;<\/span> <span class=\"token string\">&#034;&#034;<\/span><br \/>\n                <span class=\"token keyword\">for<\/span> page <span class=\"token keyword\">in<\/span> reader<span class=\"token punctuation\">.<\/span>pages<span class=\"token punctuation\">:<\/span><br \/>\n                    page_text <span class=\"token operator\">&#061;<\/span> page<span class=\"token punctuation\">.<\/span>extract_text<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><br \/>\n                    <span class=\"token keyword\">if<\/span> page_text<span class=\"token punctuation\">:<\/span><br \/>\n                        full_text <span class=\"token operator\">&#043;&#061;<\/span> page_text <span class=\"token operator\">&#043;<\/span> <span class=\"token string\">&#034;\\\\n&#034;<\/span><br \/>\n                doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;text&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;<\/span> full_text<br \/>\n                processed_docs<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span>doc<span class=\"token punctuation\">)<\/span><br \/>\n            <span class=\"token keyword\">elif<\/span> doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;doc_type&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;&#061;<\/span> <span class=\"token string\">&#034;image&#034;<\/span><span class=\"token punctuation\">:<\/span><br \/>\n                <span class=\"token comment\"># OCR\u56fe\u7247\u8bc6\u522b&#xff0c;\u9002\u7528\u4e8e\u626b\u63cf\u7248\u653f\u7b56\u6587\u4ef6<\/span><br \/>\n                <span class=\"token keyword\">from<\/span> PIL <span class=\"token keyword\">import<\/span> Image<br \/>\n                <span class=\"token keyword\">from<\/span> io <span class=\"token keyword\">import<\/span> BytesIO<br \/>\n                img <span class=\"token operator\">&#061;<\/span> Image<span class=\"token punctuation\">.<\/span><span class=\"token builtin\">open<\/span><span class=\"token punctuation\">(<\/span>BytesIO<span class=\"token punctuation\">(<\/span>doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;binary_data&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><br \/>\n                ocr_text <span class=\"token operator\">&#061;<\/span> pytesseract<span class=\"token punctuation\">.<\/span>image_to_string<span class=\"token punctuation\">(<\/span>img<span class=\"token punctuation\">,<\/span> lang<span class=\"token operator\">&#061;<\/span><span class=\"token string\">&#034;chi_sim&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n                doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;text&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;<\/span> ocr_text<br \/>\n                processed_docs<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span>doc<span class=\"token punctuation\">)<\/span><br \/>\n        <span class=\"token keyword\">except<\/span> Exception <span class=\"token keyword\">as<\/span> e<span class=\"token punctuation\">:<\/span><br \/>\n            error<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token string-interpolation\"><span class=\"token string\">f&#034;\u6587\u6863\u89e3\u6790\u5931\u8d25 <\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;source_url&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">: <\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span><span class=\"token builtin\">str<\/span><span class=\"token punctuation\">(<\/span>e<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">&#034;<\/span><\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">return<\/span> <span class=\"token punctuation\">{<\/span><span class=\"token string\">&#034;raw_documents&#034;<\/span><span class=\"token punctuation\">:<\/span> processed_docs<span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;error_log&#034;<\/span><span class=\"token punctuation\">:<\/span> error<span class=\"token punctuation\">}<\/span><\/p>\n<h4>3.4 Agent\u8282\u70b93&#xff1a;\u653f\u7b56\u62bd\u53d6\u667a\u80fd\u4f53&#xff08;PolicyExtractAgent&#xff09;<\/h4>\n<p>\u5229\u7528\u5927\u6a21\u578b\u4ece\u975e\u7ed3\u6784\u5316\u6587\u672c\u63d0\u53d6\u7535\u529b\u653f\u7b56\u7ed3\u6784\u5316\u4fe1\u606f&#xff0c;\u4e5f\u662f\u672c\u9879\u76ee\u6838\u5fc3\u4e1a\u52a1\u8282\u70b9\u3002<\/p>\n<p>extract_prompt <span class=\"token operator\">&#061;<\/span> PromptTemplate<span class=\"token punctuation\">(<\/span><br \/>\n    input_variables<span class=\"token operator\">&#061;<\/span><span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;document_text&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span><br \/>\n    template<span class=\"token operator\">&#061;<\/span><span class=\"token triple-quoted-string string\">&#034;&#034;&#034;<br \/>\n\u4f60\u662f\u7535\u529b\u5e02\u573a\u653f\u7b56\u5206\u6790\u5e08&#xff0c;\u8bf7\u9605\u8bfb\u4e0b\u9762\u7535\u529b\u4ea4\u6613\u653f\u7b56\u6587\u6863&#xff0c;\u4e25\u683c\u8f93\u51faJSON\u683c\u5f0f&#xff0c;\u4e0d\u8981\u989d\u5916\u89e3\u91ca\u3002<br \/>\n\u6587\u6863\u5185\u5bb9&#xff1a;<br \/>\n{document_text}<\/p>\n<p>\u8f93\u51faJSON\u5b57\u6bb5\u8981\u6c42&#xff1a;<br \/>\n{{<br \/>\n&#034;province&#034;: &#034;\u653f\u7b56\u6240\u5c5e\u7701\u4efd&#034;,<br \/>\n&#034;policy_name&#034;: &#034;\u653f\u7b56\u6587\u4ef6\u5168\u79f0&#034;,<br \/>\n&#034;effective_date&#034;: &#034;\u751f\u6548\u65f6\u95f4&#034;,<br \/>\n&#034;publish_date&#034;: &#034;\u53d1\u5e03\u65f6\u95f4&#034;,<br \/>\n&#034;core_rules&#034;: [&#034;\u89c4\u5219\u8981\u70b91&#034;,&#034;\u89c4\u5219\u8981\u70b92&#034;],<br \/>\n&#034;target_subject&#034;: &#034;\u9002\u7528\u5e02\u573a\u4e3b\u4f53&#xff08;\u53d1\u7535\u4f01\u4e1a\/\u552e\u7535\u516c\u53f8\/\u7535\u529b\u7528\u6237&#xff09;&#034;,<br \/>\n&#034;change_content&#034;: &#034;\u76f8\u8f83\u4e8e\u65e7\u653f\u7b56\u7684\u53d8\u66f4\u5185\u5bb9&#xff0c;\u5982\u679c\u65e0\u5bf9\u6bd4\u586b\u6682\u65e0&#034;,<br \/>\n&#034;impact_analysis&#034;: &#034;\u8be5\u653f\u7b56\u5bf9\u7535\u529b\u4ea4\u6613\u6536\u76ca\u3001\u62a5\u4ef7\u7b56\u7565\u5e26\u6765\u7684\u5f71\u54cd\u7b80\u8ff0&#034;<br \/>\n}}<br \/>\n\u53ea\u8fd4\u56de\u6807\u51c6JSON&#xff0c;\u7981\u6b62\u589e\u52a0markdown\u3001\u6ce8\u91ca\u3002<br \/>\n&#034;&#034;&#034;<\/span><br \/>\n<span class=\"token punctuation\">)<\/span><\/p>\n<p><span class=\"token keyword\">def<\/span> <span class=\"token function\">policy_extract_agent<\/span><span class=\"token punctuation\">(<\/span>state<span class=\"token punctuation\">:<\/span> AgentState<span class=\"token punctuation\">)<\/span> <span class=\"token operator\">&#8211;<\/span><span class=\"token operator\">&gt;<\/span> AgentState<span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token triple-quoted-string string\">&#034;&#034;&#034;\u5927\u6a21\u578b\u62bd\u53d6\u653f\u7b56\u7ed3\u6784\u5316\u4fe1\u606f&#034;&#034;&#034;<\/span><br \/>\n    structured_result <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    error <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    <span class=\"token keyword\">for<\/span> doc <span class=\"token keyword\">in<\/span> state<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;raw_documents&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">:<\/span><br \/>\n        text <span class=\"token operator\">&#061;<\/span> doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;text&#034;<\/span><span class=\"token punctuation\">]<\/span><br \/>\n        <span class=\"token keyword\">if<\/span> <span class=\"token builtin\">len<\/span><span class=\"token punctuation\">(<\/span>text<span class=\"token punctuation\">.<\/span>strip<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span> <span class=\"token operator\">&lt;<\/span> <span class=\"token number\">20<\/span><span class=\"token punctuation\">:<\/span><br \/>\n            error<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token string-interpolation\"><span class=\"token string\">f&#034;\u6587\u6863<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;source_url&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">\u6709\u6548\u6587\u672c\u8fc7\u77ed&#xff0c;\u8df3\u8fc7\u62bd\u53d6&#034;<\/span><\/span><span class=\"token punctuation\">)<\/span><br \/>\n            <span class=\"token keyword\">continue<\/span><br \/>\n        <span class=\"token keyword\">try<\/span><span class=\"token punctuation\">:<\/span><br \/>\n            chain <span class=\"token operator\">&#061;<\/span> extract_prompt <span class=\"token operator\">|<\/span> llm<br \/>\n            resp <span class=\"token operator\">&#061;<\/span> chain<span class=\"token punctuation\">.<\/span>invoke<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">{<\/span><span class=\"token string\">&#034;document_text&#034;<\/span><span class=\"token punctuation\">:<\/span> text<span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">:<\/span><span class=\"token number\">12000<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><span class=\"token punctuation\">)<\/span> <span class=\"token comment\"># \u622a\u65ad\u8d85\u957f\u6587\u672c<\/span><br \/>\n            json_str <span class=\"token operator\">&#061;<\/span> resp<span class=\"token punctuation\">.<\/span>content<span class=\"token punctuation\">.<\/span>strip<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><br \/>\n            <span class=\"token keyword\">import<\/span> json<br \/>\n            policy_data <span class=\"token operator\">&#061;<\/span> json<span class=\"token punctuation\">.<\/span>loads<span class=\"token punctuation\">(<\/span>json_str<span class=\"token punctuation\">)<\/span><br \/>\n            policy_data<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;source_url&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;<\/span> doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;source_url&#034;<\/span><span class=\"token punctuation\">]<\/span><br \/>\n            policy_data<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;original_text&#034;<\/span><span class=\"token punctuation\">]<\/span> <span class=\"token operator\">&#061;<\/span> text<span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">:<\/span><span class=\"token number\">5000<\/span><span class=\"token punctuation\">]<\/span><br \/>\n            structured_result<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span>policy_data<span class=\"token punctuation\">)<\/span><br \/>\n        <span class=\"token keyword\">except<\/span> Exception <span class=\"token keyword\">as<\/span> e<span class=\"token punctuation\">:<\/span><br \/>\n            error<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token string-interpolation\"><span class=\"token string\">f&#034;\u653f\u7b56\u62bd\u53d6\u5931\u8d25 <\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>doc<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;source_url&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">: <\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span><span class=\"token builtin\">str<\/span><span class=\"token punctuation\">(<\/span>e<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">&#034;<\/span><\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">return<\/span> <span class=\"token punctuation\">{<\/span><span class=\"token string\">&#034;structured_policy&#034;<\/span><span class=\"token punctuation\">:<\/span> structured_result<span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;error_log&#034;<\/span><span class=\"token punctuation\">:<\/span> error<span class=\"token punctuation\">}<\/span><\/p>\n<h4>3.5 Agent\u8282\u70b94&#xff1a;\u77e5\u8bc6\u5e93\u5165\u5e93\u667a\u80fd\u4f53&#xff08;KnowledgeBaseAgent&#xff09;<\/h4>\n<p>\u5c06\u7ed3\u6784\u5316\u653f\u7b56\u5b58\u5165\u5411\u91cf\u5e93&#xff0c;\u6784\u5efa\u7535\u529b\u4ea4\u6613\u77e5\u8bc6\u5e93&#xff0c;\u652f\u6301\u540e\u7eedRAG\u68c0\u7d22\u3002<\/p>\n<p><span class=\"token keyword\">def<\/span> <span class=\"token function\">knowledge_base_agent<\/span><span class=\"token punctuation\">(<\/span>state<span class=\"token punctuation\">:<\/span> AgentState<span class=\"token punctuation\">)<\/span> <span class=\"token operator\">&#8211;<\/span><span class=\"token operator\">&gt;<\/span> AgentState<span class=\"token punctuation\">:<\/span><br \/>\n    <span class=\"token triple-quoted-string string\">&#034;&#034;&#034;\u5c06\u7ed3\u6784\u5316\u653f\u7b56\u5199\u5165\u5411\u91cf\u77e5\u8bc6\u5e93&#034;&#034;&#034;<\/span><br \/>\n    error <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    texts <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    metadatas <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    <span class=\"token keyword\">for<\/span> policy <span class=\"token keyword\">in<\/span> state<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;structured_policy&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">:<\/span><br \/>\n        content <span class=\"token operator\">&#061;<\/span> <span class=\"token string-interpolation\"><span class=\"token string\">f&#034;&#034;&#034;<br \/>\n\u7701\u4efd&#xff1a;<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;province&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\"><br \/>\n\u653f\u7b56\u540d\u79f0&#xff1a;<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;policy_name&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\"><br \/>\n\u751f\u6548\u65f6\u95f4&#xff1a;<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;effective_date&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\"><br \/>\n\u6838\u5fc3\u89c4\u5219&#xff1a;<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span><span class=\"token string\">&#034;;&#034;<\/span><span class=\"token punctuation\">.<\/span>join<span class=\"token punctuation\">(<\/span>policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;core_rules&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\"><br \/>\n\u5f71\u54cd\u8bc4\u4f30&#xff1a;<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span>policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;impact_analysis&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\"><br \/>\n        &#034;&#034;&#034;<\/span><\/span><br \/>\n        texts<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span>content<span class=\"token punctuation\">)<\/span><br \/>\n        metadatas<span class=\"token punctuation\">.<\/span>append<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">{<\/span><br \/>\n            <span class=\"token string\">&#034;source_url&#034;<\/span><span class=\"token punctuation\">:<\/span> policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;source_url&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span><br \/>\n            <span class=\"token string\">&#034;province&#034;<\/span><span class=\"token punctuation\">:<\/span> policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;province&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span><br \/>\n            <span class=\"token string\">&#034;publish_date&#034;<\/span><span class=\"token punctuation\">:<\/span> policy<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;publish_date&#034;<\/span><span class=\"token punctuation\">]<\/span><br \/>\n        <span class=\"token punctuation\">}<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">if<\/span> texts<span class=\"token punctuation\">:<\/span><br \/>\n        vector_db<span class=\"token punctuation\">.<\/span>add_texts<span class=\"token punctuation\">(<\/span>texts<span class=\"token operator\">&#061;<\/span>texts<span class=\"token punctuation\">,<\/span> metadatas<span class=\"token operator\">&#061;<\/span>metadatas<span class=\"token punctuation\">)<\/span><br \/>\n        vector_db<span class=\"token punctuation\">.<\/span>persist<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">return<\/span> <span class=\"token punctuation\">{<\/span><span class=\"token string\">&#034;error_log&#034;<\/span><span class=\"token punctuation\">:<\/span> error<span class=\"token punctuation\">}<\/span><\/p>\n<h4>3.6 LangGraph \u6784\u5efa\u5de5\u4f5c\u6d41\u56fe&#xff0c;\u542f\u52a8\u4efb\u52a1<\/h4>\n<p><span class=\"token keyword\">def<\/span> <span class=\"token function\">build_graph<\/span><span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">:<\/span><br \/>\n    graph <span class=\"token operator\">&#061;<\/span> StateGraph<span class=\"token punctuation\">(<\/span>AgentState<span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token comment\"># \u6ce8\u518c\u8282\u70b9<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_node<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;crawler&#034;<\/span><span class=\"token punctuation\">,<\/span> crawler_agent<span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_node<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;parser&#034;<\/span><span class=\"token punctuation\">,<\/span> document_parser_agent<span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_node<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;extractor&#034;<\/span><span class=\"token punctuation\">,<\/span> policy_extract_agent<span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_node<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;knowledge_store&#034;<\/span><span class=\"token punctuation\">,<\/span> knowledge_base_agent<span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token comment\"># \u8bbe\u7f6e\u6d41\u8f6c\u987a\u5e8f<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>set_entry_point<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;crawler&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_edge<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;crawler&#034;<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;parser&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_edge<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;parser&#034;<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;extractor&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_edge<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;extractor&#034;<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token string\">&#034;knowledge_store&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    graph<span class=\"token punctuation\">.<\/span>add_edge<span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;knowledge_store&#034;<\/span><span class=\"token punctuation\">,<\/span> END<span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">return<\/span> graph<span class=\"token punctuation\">.<\/span><span class=\"token builtin\">compile<\/span><span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><\/p>\n<p><span class=\"token keyword\">if<\/span> __name__ <span class=\"token operator\">&#061;&#061;<\/span> <span class=\"token string\">&#034;__main__&#034;<\/span><span class=\"token punctuation\">:<\/span><br \/>\n    app <span class=\"token operator\">&#061;<\/span> build_graph<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token comment\"># \u8f93\u5165\u5f85\u6293\u53d6\u7535\u529b\u653f\u7b56\u94fe\u63a5&#xff0c;\u66ff\u6362\u4e3a\u5404\u7701\u7535\u529b\u4ea4\u6613\u4e2d\u5fc3\u516c\u544a\u5730\u5740<\/span><br \/>\n    init_input <span class=\"token operator\">&#061;<\/span> <span class=\"token punctuation\">{<\/span><br \/>\n        <span class=\"token string\">&#034;url_list&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><br \/>\n            <span class=\"token string\">&#034;https:\/\/example.com\/jiangsu_power_policy.pdf&#034;<\/span><span class=\"token punctuation\">,<\/span><br \/>\n            <span class=\"token string\">&#034;https:\/\/example.com\/henan_trade_notice.html&#034;<\/span><br \/>\n        <span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span><br \/>\n        <span class=\"token string\">&#034;raw_documents&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span><br \/>\n        <span class=\"token string\">&#034;structured_policy&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">,<\/span><br \/>\n        <span class=\"token string\">&#034;error_log&#034;<\/span><span class=\"token punctuation\">:<\/span> <span class=\"token punctuation\">[<\/span><span class=\"token punctuation\">]<\/span><br \/>\n    <span class=\"token punctuation\">}<\/span><br \/>\n    result <span class=\"token operator\">&#061;<\/span> app<span class=\"token punctuation\">.<\/span>invoke<span class=\"token punctuation\">(<\/span>init_input<span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;&#061;&#061;&#061;&#061;\u4efb\u52a1\u6267\u884c\u5b8c\u6210&#061;&#061;&#061;&#061;&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">&#034;\u9519\u8bef\u65e5\u5fd7&#xff1a;&#034;<\/span><span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">for<\/span> err <span class=\"token keyword\">in<\/span> result<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#034;error_log&#034;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">:<\/span><br \/>\n        <span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span>err<span class=\"token punctuation\">)<\/span><br \/>\n    <span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string-interpolation\"><span class=\"token string\">f&#034;\u6210\u529f\u89e3\u6790\u653f\u7b56\u6570\u91cf&#xff1a;<\/span><span class=\"token interpolation\"><span class=\"token punctuation\">{<\/span><span class=\"token builtin\">len<\/span><span class=\"token punctuation\">(<\/span>result<span class=\"token punctuation\">[<\/span><span class=\"token string\">&#039;structured_policy&#039;<\/span><span class=\"token punctuation\">]<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">}<\/span><\/span><span class=\"token string\">&#034;<\/span><\/span><span class=\"token punctuation\">)<\/span><\/p>\n<h3>\u56db\u3001\u4ee3\u7801\u8fd0\u884c\u8bf4\u660e\u4e0e\u5de5\u7a0b\u4f18\u5316\u65b9\u6848<\/h3>\n<h4>4.1 \u57fa\u7840\u8fd0\u884c\u95ee\u9898\u89e3\u51b3<\/h4>\n<li>Tesseract OCR\u62a5\u9519&#xff1a;\u5fc5\u987b\u5b89\u88c5\u672c\u5730\u7a0b\u5e8f&#xff0c;\u4ee3\u7801\u4e2d\u6dfb\u52a0\u8def\u5f84 pytesseract.pytesseract.tesseract_cmd &#061; r&#039;C:\\\\Program Files\\\\Tesseract-OCR\\\\tesseract.exe&#039;<\/li>\n<li>\u8d85\u957f\u6587\u6863\u62a5\u9519&#xff1a;\u589e\u52a0\u6587\u672c\u6ed1\u52a8\u7a97\u53e3\u5206\u6bb5\u62bd\u53d6&#xff0c;\u907f\u514d\u5927\u6a21\u578b\u4e0a\u4e0b\u6587\u6ea2\u51fa<\/li>\n<li>\u722c\u866b\u5c01\u7981&#xff1a;\u589e\u52a0\u4ee3\u7406\u6c60\u3001\u8bf7\u6c42\u95f4\u9694&#xff0c;\u968f\u673aUA<\/li>\n<h4>4.2 \u751f\u4ea7\u73af\u5883\u4f18\u5316\u65b9\u5411<\/h4>\n<li>\u8def\u7531\u5206\u652f\u4f18\u5316&#xff1a;\u5f53\u524d\u4e3a\u7ebf\u6027\u6d41\u7a0b&#xff0c;\u53ef\u6539\u9020LangGraph\u6761\u4ef6\u5206\u652f&#xff0c;PDF\u3001\u7f51\u9875\u3001\u56fe\u7247\u5e76\u884c\u89e3\u6790&#xff0c;\u63d0\u5347\u901f\u5ea6<\/li>\n<li>\u53bb\u91cd\u673a\u5236&#xff1a;\u589e\u52a0URL\u54c8\u5e0c\u3001\u6587\u6863\u6587\u672c\u76f8\u4f3c\u5ea6\u6bd4\u5bf9&#xff0c;\u907f\u514d\u91cd\u590d\u5165\u5e93\u76f8\u540c\u653f\u7b56<\/li>\n<li>\u589e\u91cf\u66f4\u65b0&#xff1a;\u5b9a\u65f6\u722c\u866b\u8f6e\u8be2\u7535\u529b\u4ea4\u6613\u7f51\u7ad9&#xff0c;\u53ea\u6293\u53d6\u6700\u65b0\u53d1\u5e03\u653f\u7b56&#xff0c;\u5b9e\u73b0\u6301\u7eed\u8ddf\u8e2a<\/li>\n<li>Function Calling\u589e\u5f3a&#xff1a;\u6269\u5c55Agent\u80fd\u529b&#xff0c;\u65b0\u589e\u653f\u7b56\u5dee\u5f02\u5bf9\u6bd4Agent&#xff0c;\u81ea\u52a8\u5bf9\u6bd4\u65b0\u65e7\u7248\u672c\u89c4\u5219\u53d8\u5316<\/li>\n<li>\u6301\u4e45\u5316\u72b6\u6001&#xff1a;\u7ed3\u5408Redis\u5b58\u50a8LangGraph\u8fd0\u884c\u72b6\u6001&#xff0c;\u652f\u6301\u4e2d\u65ad\u4efb\u52a1\u7eed\u8dd1<\/li>\n<h3>\u4e94\u3001\u4e1a\u52a1\u843d\u5730\u601d\u8003&#xff1a;\u7535\u529b\u653f\u7b56Agent\u7684\u4ef7\u503c<\/h3>\n<p>\u4f20\u7edf\u65b9\u6848\u4f7f\u7528\u56fa\u5b9a\u722c\u866b&#043;\u6b63\u5219\u8868\u8fbe\u5f0f\u63d0\u53d6\u653f\u7b56\u6761\u6b3e&#xff0c;\u7f3a\u70b9\u5341\u5206\u660e\u663e&#xff1a;\u7535\u529b\u653f\u7b56\u6587\u4ef6\u683c\u5f0f\u4e0d\u7edf\u4e00\u3001\u626b\u63cf\u7248PDF\u65e0\u6cd5\u63d0\u53d6\u6587\u672c\u3001\u5404\u7701\u6587\u4ef6\u884c\u6587\u89c4\u8303\u5dee\u5f02\u5de8\u5927&#xff0c;\u6b63\u5219\u5f88\u96be\u9002\u914d\u3002<\/p>\n<p>\u57fa\u4e8eLangGraph\u591a\u667a\u80fd\u4f53\u65b9\u6848\u4f18\u52bf&#xff1a;<\/p>\n<li>\u6a21\u5757\u5316\u89e3\u8026&#xff1a;\u722c\u866b\u3001\u89e3\u6790\u3001\u62bd\u53d6\u3001\u77e5\u8bc6\u5e93\u6a21\u5757\u72ec\u7acb&#xff0c;\u5355\u72ec\u8fed\u4ee3\u7ef4\u62a4<\/li>\n<li>\u81ea\u9002\u5e94\u975e\u7ed3\u6784\u5316\u6587\u6863&#xff1a;OCR&#043;PDF\u53cc\u65b9\u6848\u517c\u5bb9\u626b\u63cf\u4ef6\u3001\u7535\u5b50\u7248\u653f\u7b56<\/li>\n<li>\u81ea\u7136\u8bed\u8a00\u7406\u89e3&#xff1a;\u4f9d\u9760\u5927\u6a21\u578b\u7406\u89e3\u653f\u7b56\u8bed\u4e49&#xff0c;\u4e0d\u9700\u8981\u7f16\u5199\u5927\u91cf\u89c4\u5219\u8868\u8fbe\u5f0f<\/li>\n<li>\u53ef\u6269\u5c55&#xff1a;\u540e\u7eed\u53ef\u4ee5\u65b0\u589e\u7535\u4ef7\u6570\u636e\u91c7\u96c6Agent\u3001\u6536\u76ca\u6d4b\u7b97Agent&#xff0c;\u5b9e\u73b0\u591a\u667a\u80fd\u4f53\u534f\u540c\u51b3\u7b56&#xff0c;\u5bf9\u63a5\u7535\u529b\u4ea4\u6613\u4e1a\u52a1\u7cfb\u7edf<\/li>\n<h3>\u516d\u3001\u73b0\u5b58\u75db\u70b9\u4e0e\u907f\u5751\u6307\u5357<\/h3>\n<li>\u62bd\u53d6\u7a33\u5b9a\u6027\u95ee\u9898&#xff1a;\u5927\u6a21\u578b\u5076\u5c14\u8f93\u51fa\u975e\u6cd5JSON&#xff0c;\u5de5\u7a0b\u4e2d\u589e\u52a0\u91cd\u8bd5\u8282\u70b9\u3001JSON\u6e05\u6d17\u5de5\u5177&#xff1b;<\/li>\n<li>\u5411\u91cf\u5e93\u68c0\u7d22\u7cbe\u5ea6&#xff1a;\u5355\u7eaf\u6bb5\u843d\u5207\u7247\u6548\u679c\u5dee&#xff0c;\u540e\u7eed\u53ef\u5f15\u5165Hierarchical\u5207\u7247\u3001\u7236\u6587\u6863\u68c0\u7d22\u63d0\u5347RAG\u6548\u679c&#xff1b;<\/li>\n<li>\u957fPDF\u6587\u4ef6&#xff1a;\u6570\u767e\u9875\u653f\u7b56\u6587\u4ef6\u4e00\u6b21\u6027\u4f20\u5165LLM\u4f1a\u8d85\u9650&#xff0c;\u9700\u8981\u5b9e\u73b0\u6587\u6863\u5206\u5757\u3001\u5206\u6279\u62bd\u53d6&#xff1b;<\/li>\n<li>\u626b\u63cf\u4ef6\u8bc6\u522b&#xff1a;\u6a21\u7cca\u56fe\u7247OCR\u51c6\u786e\u7387\u4e0b\u964d&#xff0c;\u53ef\u5f15\u5165OpenCV\u56fe\u7247\u9884\u5904\u7406&#xff08;\u964d\u566a\u3001\u4e8c\u503c\u5316&#xff09;\u63d0\u5347\u8bc6\u522b\u6548\u679c\u3002<\/li>\n<h3>\u4e03\u3001\u62d3\u5c55\u5ef6\u4f38<\/h3>\n<p>\u672c\u5957\u591a\u667a\u80fd\u4f53\u6846\u67b6\u4e0d\u5c40\u9650\u4e8e\u7535\u529b\u653f\u7b56\u573a\u666f&#xff0c;\u53ef\u4ee5\u5feb\u901f\u8fc1\u79fb\u81f3&#xff1a;\u65b0\u80fd\u6e90\u653f\u7b56\u8ddf\u8e2a\u3001\u50a8\u80fd\u884c\u4e1a\u6587\u4ef6\u89e3\u6790\u3001\u62db\u6295\u6807\u516c\u544a\u6293\u53d6\u3002\u7ed3\u5408LangGraph\u7684\u6761\u4ef6\u5206\u652f\u3001\u5faa\u73af\u80fd\u529b&#xff0c;\u8fd8\u53ef\u4ee5\u5b9e\u73b0\u81ea\u4e3b\u641c\u7d22Agent&#xff1a;\u5f53\u653f\u7b56\u4fe1\u606f\u7f3a\u5931\u65f6&#xff0c;\u667a\u80fd\u4f53\u4e3b\u52a8\u751f\u6210\u5173\u952e\u8bcd\u518d\u6b21\u68c0\u7d22\u7f51\u9875&#xff0c;\u5f62\u6210\u95ed\u73af\u81ea\u4e3b\u8c03\u7814\u3002<\/p>\n<hr \/>\n","protected":false},"excerpt":{"rendered":"<p>\u6587\u7ae0\u524d\u8a00<br \/>\n\u968f\u7740\u56fd\u5185\u7535\u529b\u5e02\u573a\u5316\u6539\u9769\u6301\u7eed\u63a8\u8fdb&#xff0c;\u5404\u7701\u7535\u529b\u4ea4\u6613\u4e2d\u5fc3\u6301\u7eed\u53d1\u5e03\u4ea4\u6613\u89c4\u5219\u3001\u653f\u7b56\u6587\u4ef6\u3001\u5e02\u573a\u901a\u77e5\u3002\u7535\u529b\u4ece\u4e1a\u8005\u9700\u8981\u6301\u7eed\u8ddf\u8e2a\u6d77\u91cfPDF\u516c\u544a\u3001\u7f51\u9875\u653f\u7b56\u6587\u672c&#xff0c;\u4eba\u5de5\u6574\u7406\u3001\u5bf9\u6bd4\u653f\u7b56\u5dee\u5f02\u3001\u63d0\u53d6\u6838\u5fc3\u6761\u6b3e\u6210\u672c\u6781\u9ad8\u3002\u4f20\u7edf\u5355\u4e00\u5927\u6a21\u578b\u5e94\u7528\u65e0\u6cd5\u5b8c\u6210\u7f51\u9875\u91c7\u96c6\u2192\u6587\u6863\u89e3\u6790\u2192\u6587\u672c\u62bd\u53d6\u2192\u77e5\u8bc6\u5165\u5e93\u2192\u667a\u80fd\u5206\u6790\u5b8c\u6574\u6d41\u6c34\u7ebf\u3002<br \/>\n\u672c\u6587\u57fa\u4e8eLangGraph\u6784\u5efa\u591a\u667a\u80fd\u4f53\u534f\u540c\u7cfb\u7edf&#xff0c;\u62c6\u5206\u722c\u866b\u667a\u80fd\u4f53\u3001\u6587\u6863\u89e3\u6790\u667a\u80fd\u4f53\u3001\u89c4\u5219\u62bd\u53d6\u667a\u80fd\u4f53\u3001\u77e5\u8bc6\u5e93\u66f4\u65b0\u667a\u80fd\u4f53\u3002\u6574\u5957\u5de5\u4f5c\u6d41\u81ea\u52a8\u5b8c\u6210\u7f51\u9875\u6293\u53d6\u3001PDF\u89e3\u6790\u3001\u56fe<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[756,2150,677],"topic":[],"class_list":["post-89708","post","type-post","status-publish","format-standard","hentry","category-server","tag-ocr","tag-pdf","tag-677"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v20.3 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.wsisp.com\/helps\/89708.html\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"og:description\" content=\"\u6587\u7ae0\u524d\u8a00 \u968f\u7740\u56fd\u5185\u7535\u529b\u5e02\u573a\u5316\u6539\u9769\u6301\u7eed\u63a8\u8fdb&#xff0c;\u5404\u7701\u7535\u529b\u4ea4\u6613\u4e2d\u5fc3\u6301\u7eed\u53d1\u5e03\u4ea4\u6613\u89c4\u5219\u3001\u653f\u7b56\u6587\u4ef6\u3001\u5e02\u573a\u901a\u77e5\u3002\u7535\u529b\u4ece\u4e1a\u8005\u9700\u8981\u6301\u7eed\u8ddf\u8e2a\u6d77\u91cfPDF\u516c\u544a\u3001\u7f51\u9875\u653f\u7b56\u6587\u672c&#xff0c;\u4eba\u5de5\u6574\u7406\u3001\u5bf9\u6bd4\u653f\u7b56\u5dee\u5f02\u3001\u63d0\u53d6\u6838\u5fc3\u6761\u6b3e\u6210\u672c\u6781\u9ad8\u3002\u4f20\u7edf\u5355\u4e00\u5927\u6a21\u578b\u5e94\u7528\u65e0\u6cd5\u5b8c\u6210\u7f51\u9875\u91c7\u96c6\u2192\u6587\u6863\u89e3\u6790\u2192\u6587\u672c\u62bd\u53d6\u2192\u77e5\u8bc6\u5165\u5e93\u2192\u667a\u80fd\u5206\u6790\u5b8c\u6574\u6d41\u6c34\u7ebf\u3002 \u672c\u6587\u57fa\u4e8eLangGraph\u6784\u5efa\u591a\u667a\u80fd\u4f53\u534f\u540c\u7cfb\u7edf&#xff0c;\u62c6\u5206\u722c\u866b\u667a\u80fd\u4f53\u3001\u6587\u6863\u89e3\u6790\u667a\u80fd\u4f53\u3001\u89c4\u5219\u62bd\u53d6\u667a\u80fd\u4f53\u3001\u77e5\u8bc6\u5e93\u66f4\u65b0\u667a\u80fd\u4f53\u3002\u6574\u5957\u5de5\u4f5c\u6d41\u81ea\u52a8\u5b8c\u6210\u7f51\u9875\u6293\u53d6\u3001PDF\u89e3\u6790\u3001\u56fe\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wsisp.com\/helps\/89708.html\" \/>\n<meta property=\"og:site_name\" content=\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"article:published_time\" content=\"2026-08-03T06:56:32+00:00\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"5 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/89708.html\",\"url\":\"https:\/\/www.wsisp.com\/helps\/89708.html\",\"name\":\"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"isPartOf\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\"},\"datePublished\":\"2026-08-03T06:56:32+00:00\",\"dateModified\":\"2026-08-03T06:56:32+00:00\",\"author\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/89708.html#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.wsisp.com\/helps\/89708.html\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/89708.html#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wsisp.com\/helps\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\",\"url\":\"https:\/\/www.wsisp.com\/helps\/\",\"name\":\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"description\":\"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"contentUrl\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"caption\":\"admin\"},\"sameAs\":[\"http:\/\/wp.wsisp.com\"],\"url\":\"https:\/\/www.wsisp.com\/helps\/author\/admin\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.wsisp.com\/helps\/89708.html","og_locale":"zh_CN","og_type":"article","og_title":"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","og_description":"\u6587\u7ae0\u524d\u8a00 \u968f\u7740\u56fd\u5185\u7535\u529b\u5e02\u573a\u5316\u6539\u9769\u6301\u7eed\u63a8\u8fdb&#xff0c;\u5404\u7701\u7535\u529b\u4ea4\u6613\u4e2d\u5fc3\u6301\u7eed\u53d1\u5e03\u4ea4\u6613\u89c4\u5219\u3001\u653f\u7b56\u6587\u4ef6\u3001\u5e02\u573a\u901a\u77e5\u3002\u7535\u529b\u4ece\u4e1a\u8005\u9700\u8981\u6301\u7eed\u8ddf\u8e2a\u6d77\u91cfPDF\u516c\u544a\u3001\u7f51\u9875\u653f\u7b56\u6587\u672c&#xff0c;\u4eba\u5de5\u6574\u7406\u3001\u5bf9\u6bd4\u653f\u7b56\u5dee\u5f02\u3001\u63d0\u53d6\u6838\u5fc3\u6761\u6b3e\u6210\u672c\u6781\u9ad8\u3002\u4f20\u7edf\u5355\u4e00\u5927\u6a21\u578b\u5e94\u7528\u65e0\u6cd5\u5b8c\u6210\u7f51\u9875\u91c7\u96c6\u2192\u6587\u6863\u89e3\u6790\u2192\u6587\u672c\u62bd\u53d6\u2192\u77e5\u8bc6\u5165\u5e93\u2192\u667a\u80fd\u5206\u6790\u5b8c\u6574\u6d41\u6c34\u7ebf\u3002 \u672c\u6587\u57fa\u4e8eLangGraph\u6784\u5efa\u591a\u667a\u80fd\u4f53\u534f\u540c\u7cfb\u7edf&#xff0c;\u62c6\u5206\u722c\u866b\u667a\u80fd\u4f53\u3001\u6587\u6863\u89e3\u6790\u667a\u80fd\u4f53\u3001\u89c4\u5219\u62bd\u53d6\u667a\u80fd\u4f53\u3001\u77e5\u8bc6\u5e93\u66f4\u65b0\u667a\u80fd\u4f53\u3002\u6574\u5957\u5de5\u4f5c\u6d41\u81ea\u52a8\u5b8c\u6210\u7f51\u9875\u6293\u53d6\u3001PDF\u89e3\u6790\u3001\u56fe","og_url":"https:\/\/www.wsisp.com\/helps\/89708.html","og_site_name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","article_published_time":"2026-08-03T06:56:32+00:00","author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"5 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wsisp.com\/helps\/89708.html","url":"https:\/\/www.wsisp.com\/helps\/89708.html","name":"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","isPartOf":{"@id":"https:\/\/www.wsisp.com\/helps\/#website"},"datePublished":"2026-08-03T06:56:32+00:00","dateModified":"2026-08-03T06:56:32+00:00","author":{"@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41"},"breadcrumb":{"@id":"https:\/\/www.wsisp.com\/helps\/89708.html#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wsisp.com\/helps\/89708.html"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wsisp.com\/helps\/89708.html#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wsisp.com\/helps"},{"@type":"ListItem","position":2,"name":"LangGraph \u591a\u667a\u80fd\u4f53\u5b9e\u6218\uff1a\u642d\u5efa\u7535\u529b\u653f\u7b56\u8ddf\u8e2a Agent\uff0c\u5b9e\u73b0\u7f51\u9875\u722c\u866b\u3001OCR \u4e0e PDF \u6587\u6863\u81ea\u52a8\u5316\u89e3\u6790"}]},{"@type":"WebSite","@id":"https:\/\/www.wsisp.com\/helps\/#website","url":"https:\/\/www.wsisp.com\/helps\/","name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","description":"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41","name":"admin","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/","url":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","contentUrl":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","caption":"admin"},"sameAs":["http:\/\/wp.wsisp.com"],"url":"https:\/\/www.wsisp.com\/helps\/author\/admin"}]}},"_links":{"self":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/89708","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/comments?post=89708"}],"version-history":[{"count":0,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/89708\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media?parent=89708"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/categories?post=89708"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/tags?post=89708"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/topic?post=89708"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}