refactor: 更新依赖并迁移至新的LangChain模块结构

- 将langchain依赖替换为@langchain/core和@langchain/textsplitters - 更新FileLoader.ts以使用新的模块导入路径 - 调整eslint规则配置 - 在项目规则文档中添加类型检查规范
2025-12-23 16:11:10 +08:00
1 parent 08e1b84361
commit dc4ecfb0fa
@@ -62,6 +62,7 @@ NCurator-Desktop 是一个基于 Electron-Vite 构建的本地知识库搜索应
 - **IPC**: 使用 `ipcMain` and `ipcRenderer` (通过 `preload`) 在 UI 和主进程中的繁重处理逻辑之间进行通信。
 - **模块化**: 将解析、索引和搜索逻辑与 UI 组件分离。
 - **注释**: 注释复杂的逻辑，特别是 RRF 算法和模型加载步骤。
+- **类型**: 禁止使用 `any` 类型，必须使用明确的类型注解。

 ## 6. AI 交互规则
 - 生成代码时，优先考虑技术栈中列出的库。
@@ -25,7 +25,11 @@ export default defineConfig(
    },
    rules: {
      ...eslintPluginReactHooks.configs.recommended.rules,
-      ...eslintPluginReactRefresh.configs.vite.rules
+      ...eslintPluginReactRefresh.configs.vite.rules,
+      '@typescript-eslint/explicit-function-return-type': 0,
+      'prettier/prettier': 0,
+      '@typescript-eslint/no-explicit-any': 0,
+      '@typescript-eslint/no-empty-function': 0
    }
  },
  eslintConfigPrettier
@@ -26,10 +26,11 @@
    "@huggingface/transformers": "^3.8.1",
    "@lancedb/lancedb": "^0.23.0",
    "@langchain/community": "^1.1.1",
+    "@langchain/core": "^1.1.8",
+    "@langchain/textsplitters": "^1.0.1",
    "antd": "5.22.1",
    "electron-updater": "^6.3.9",
    "flexsearch": "^0.8.212",
-    "langchain": "^1.2.3",
    "mammoth": "^1.11.0",
    "pdf-parse": "^2.4.5"
  },
@@ -23,6 +23,12 @@ importers:
      '@langchain/community':
        specifier: ^1.1.1
        version: 1.1.1(@browserbasehq/sdk@2.6.0(encoding@0.1.13))(@browserbasehq/stagehand@1.14.0(@playwright/test@1.57.0)(deepmerge@4.3.1)(dotenv@16.6.1)(encoding@0.1.13)(openai@6.15.0(ws@8.18.3)(zod@4.2.1))(zod@4.2.1))(@huggingface/transformers@3.8.1)(@ibm-cloud/watsonx-ai@1.7.5)(@lancedb/lancedb@0.23.0(apache-arrow@18.1.0))(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(ibm-cloud-sdk-core@5.4.5)(ignore@5.3.2)(jsonwebtoken@9.0.3)(lodash@4.17.21)(mammoth@1.11.0)(openai@6.15.0(ws@8.18.3)(zod@4.2.1))(pdf-parse@2.4.5)(playwright@1.57.0)(ws@8.18.3)
+      '@langchain/core':
+        specifier: ^1.1.8
+        version: 1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
+      '@langchain/textsplitters':
+        specifier: ^1.0.1
+        version: 1.0.1(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))
      antd:
        specifier: 5.22.1
        version: 5.22.1(react-dom@19.2.3(react@19.2.3))(react@19.2.3)
@@ -32,9 +38,6 @@ importers:
      flexsearch:
        specifier: ^0.8.212
        version: 0.8.212
-      langchain:
-        specifier: ^1.2.3
-        version: 1.2.3(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(openai@6.15.0(ws@8.18.3)(zod@4.2.1))(react-dom@19.2.3(react@19.2.3))(react@19.2.3)(zod-to-json-schema@3.25.0(zod@4.2.1))
      mammoth:
        specifier: ^1.11.0
        version: 1.11.0
@@ -921,7 +924,6 @@ packages:
  '@lancedb/lancedb@0.23.0':
    resolution: {integrity: sha512-aYrIoEG24AC+wILCL57Ius/Y4yU+xFHDPKLvmjzzN4byAjzeIGF0TC86S5RBt4Ji+dxS7yIWV5Q/gE5/fybIFQ==}
    engines: {node: '>= 18'}
-    cpu: [x64, arm64]
    os: [darwin, linux, win32]
    peerDependencies:
      apache-arrow: '>=15.0.0 <=18.1.0'
@@ -1308,37 +1310,6 @@ packages:
    resolution: {integrity: sha512-kIUidOgc0ZdyXo4Ahn9Zas+OayqOfk4ZoKPi7XaDipNSWSApc2+QK5BVcjvwtzxstsNOrmXJiJWEN6WPF/MvAw==}
    engines: {node: '>=20'}

-  '@langchain/langgraph-checkpoint@1.0.0':
-    resolution: {integrity: sha512-xrclBGvNCXDmi0Nz28t3vjpxSH6UYx6w5XAXSiiB1WEdc2xD2iY/a913I3x3a31XpInUW/GGfXXfePfaghV54A==}
-    engines: {node: '>=18'}
-    peerDependencies:
-      '@langchain/core': ^1.0.1
-
-  '@langchain/langgraph-sdk@1.3.1':
-    resolution: {integrity: sha512-zTi7DZHwqtMEzapvm3I1FL4Q7OZsxtq9tTXy6s2gcCxyIU3sphqRboqytqVN7dNHLdTCLb8nXy49QKurs2MIBg==}
-    peerDependencies:
-      '@langchain/core': ^1.0.1
-      react: ^18 || ^19
-      react-dom: ^18 || ^19
-    peerDependenciesMeta:
-      '@langchain/core':
-        optional: true
-      react:
-        optional: true
-      react-dom:
-        optional: true
-
-  '@langchain/langgraph@1.0.7':
-    resolution: {integrity: sha512-EBGqNOWoRiEoLUaeuiXRpUM8/DE6QcwiirNyd97XhezStebBoTTilWH8CUt6S94JRGl5zwfBBRHfzotDnZS/eA==}
-    engines: {node: '>=18'}
-    peerDependencies:
-      '@langchain/core': ^1.0.1
-      zod: ^3.25.32 || ^4.1.0
-      zod-to-json-schema: ^3.x
-    peerDependenciesMeta:
-      zod-to-json-schema:
-        optional: true
-
  '@langchain/openai@1.2.0':
    resolution: {integrity: sha512-r2g5Be3Sygw7VTJ89WVM/M94RzYToNTwXf8me1v+kgKxzdHbd/8XPYDFxpXEp3REyPgUrtJs+Oplba9pkTH5ug==}
    engines: {node: '>=20'}
@@ -1839,9 +1810,6 @@ packages:
  '@types/responselike@1.0.3':
    resolution: {integrity: sha512-H/+L+UkTV33uf49PH5pCAUBVPNj2nDBXTN+qS1dOwyyg24l3CcicicCA7ca+HMvJBZcFgl5r8e+RR6elsb4Lyw==}

-  '@types/retry@0.12.0':
-    resolution: {integrity: sha512-wWKOClTTiizcZhXnPY4wikVAwmdYHp8q6DmC+EJUzAMsycb7HB32Kh9RN4+0gExjmPmZSAQjgURXIGATPegAvA==}
-
  '@types/tough-cookie@4.0.5':
    resolution: {integrity: sha512-/Ad8+nIOV7Rl++6f1BdKxFSMgmoqEoYbHRpPcx3JEfv8VRsQe9Z4mCXeJBzxs7mbHY/XOZZuXlRNfhpVPbs6ZA==}

@@ -3194,12 +3162,6 @@ packages:
  keyv@4.5.4:
    resolution: {integrity: sha512-oxVHkHR/EJf2CNXnWxRLW6mg7JyCCUcG0DtEGmL2ctUo1PNTin1PUil+r/+4r5MpVgC/fn1kjsx7mjSujKqIpw==}

-  langchain@1.2.3:
-    resolution: {integrity: sha512-3k986xJuqg4az53JxV5LnGlOzIXF1d9Kq6Y9s7XjitvzhpsbFuTDV5/kiF4cx3pkNGyw0mUXC4tLz9RxucO0hw==}
-    engines: {node: '>=20'}
-    peerDependencies:
-      '@langchain/core': 1.1.8
-
  langsmith@0.4.2:
    resolution: {integrity: sha512-BvBeFgSmR9esl8x5wsiDlALiHKKPybw2wE2Hh6x1tgSZki46H9c9KI9/06LARbPhyyDu/TZU7exfg6fnhdj1Qg==}
    peerDependencies:
@@ -3668,10 +3630,6 @@ packages:
    resolution: {integrity: sha512-RwFpb72c/BhQLEXIZ5K2e+AhgNVmIejGlTgiB9MzZ0e93GRvqZ7uSi0dvRF7/XIXDeNkra2fNHBxTyPDGySpjQ==}
    engines: {node: '>=8'}

-  p-retry@4.6.2:
-    resolution: {integrity: sha512-312Id396EbJdvRONlngUx0NydfrIQ5lsYu0znKVUzVvArzEIt08V1qhtyESbGVd1FGX7UKtiFp5uwKZdM8wIuQ==}
-    engines: {node: '>=8'}
-
  p-timeout@3.2.0:
    resolution: {integrity: sha512-rhIwUycgwwKcP9yTOOFK/AKsAopjjCakVqLHePO3CC6Mir1Z99xT+R63jZxAT5lFZLa2inS5h+ZS2GvR99/FBg==}
    engines: {node: '>=8'}
@@ -4159,10 +4117,6 @@ packages:
    resolution: {integrity: sha512-9LkiTwjUh6rT555DtE9rTX+BKByPfrMzEAtnlEtdEwr3Nkffwiihqe2bWADg+OQRjt9gl6ICdmB/ZFDCGAtSow==}
    engines: {node: '>= 4'}

-  retry@0.13.1:
-    resolution: {integrity: sha512-XQBQ3I8W1Cge0Seh+6gjj03LbmRFWuoszgK9ooCpwYIrhhoO80pfq4cUkU5DkknwfOfFteRwlZ56PYOGYyFWdg==}
-    engines: {node: '>= 4'}
-
  rimraf@2.6.3:
    resolution: {integrity: sha512-mwqeW5XsA2qAejG46gYdENaxXjx9onRNCfn7L0duuP4hCuTIi/QO7PDK07KJfp1d+izWPrzEJDcSqBa0OZQriA==}
    deprecated: Rimraf versions prior to v4 are no longer supported
@@ -4585,10 +4539,6 @@ packages:
    resolution: {integrity: sha512-8XkAphELsDnEGrDxUOHB3RGvXz6TeuYSGEZBOjtTtPm2lwhGBjLgOzLHB63IUWfBpNucQjND6d3AOudO+H3RWQ==}
    hasBin: true

-  uuid@9.0.1:
-    resolution: {integrity: sha512-b+1eJOlsR9K8HJpow9Ok3fiWOWSIcIzXodvv0rQjVoOVNpWMpxf1wZNpt4y9h10odCNrqnYp1OBzRktckBe3sA==}
-    hasBin: true
-
  verror@1.10.1:
    resolution: {integrity: sha512-veufcmxri4e3XSrT0xwfUR7kguIkaxBeosDg00yDWhk49wdwkSUrvvsm7nc75e1PUyvIeZj6nS8VQRYz2/S4Xg==}
    engines: {node: '>=0.6.0'}
@@ -5572,34 +5522,6 @@ snapshots:
      - '@opentelemetry/sdk-trace-base'
      - openai

-  '@langchain/langgraph-checkpoint@1.0.0(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))':
-    dependencies:
-      '@langchain/core': 1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
-      uuid: 10.0.0
-
-  '@langchain/langgraph-sdk@1.3.1(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(react-dom@19.2.3(react@19.2.3))(react@19.2.3)':
-    dependencies:
-      p-queue: 6.6.2
-      p-retry: 4.6.2
-      uuid: 9.0.1
-    optionalDependencies:
-      '@langchain/core': 1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
-      react: 19.2.3
-      react-dom: 19.2.3(react@19.2.3)
-
-  '@langchain/langgraph@1.0.7(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(react-dom@19.2.3(react@19.2.3))(react@19.2.3)(zod-to-json-schema@3.25.0(zod@4.2.1))(zod@4.2.1)':
-    dependencies:
-      '@langchain/core': 1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
-      '@langchain/langgraph-checkpoint': 1.0.0(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))
-      '@langchain/langgraph-sdk': 1.3.1(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(react-dom@19.2.3(react@19.2.3))(react@19.2.3)
-      uuid: 10.0.0
-      zod: 4.2.1
-    optionalDependencies:
-      zod-to-json-schema: 3.25.0(zod@4.2.1)
-    transitivePeerDependencies:
-      - react
-      - react-dom
-
  '@langchain/openai@1.2.0(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(ws@8.18.3)':
    dependencies:
      '@langchain/core': 1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
@@ -6017,8 +5939,6 @@ snapshots:
    dependencies:
      '@types/node': 22.19.3

-  '@types/retry@0.12.0': {}
-
  '@types/tough-cookie@4.0.5': {}

  '@types/uuid@10.0.0': {}
@@ -7531,7 +7451,7 @@ snapshots:
      isstream: 0.1.2
      jsonwebtoken: 9.0.3
      mime-types: 2.1.35
-      retry-axios: 2.6.0(axios@1.13.2)
+      retry-axios: 2.6.0(axios@1.13.2(debug@4.4.3))
      tough-cookie: 4.1.4
    transitivePeerDependencies:
      - supports-color
@@ -7820,23 +7740,6 @@ snapshots:
    dependencies:
      json-buffer: 3.0.1

-  langchain@1.2.3(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(openai@6.15.0(ws@8.18.3)(zod@4.2.1))(react-dom@19.2.3(react@19.2.3))(react@19.2.3)(zod-to-json-schema@3.25.0(zod@4.2.1)):
-    dependencies:
-      '@langchain/core': 1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
-      '@langchain/langgraph': 1.0.7(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))(react-dom@19.2.3(react@19.2.3))(react@19.2.3)(zod-to-json-schema@3.25.0(zod@4.2.1))(zod@4.2.1)
-      '@langchain/langgraph-checkpoint': 1.0.0(@langchain/core@1.1.8(openai@6.15.0(ws@8.18.3)(zod@4.2.1)))
-      langsmith: 0.4.2(openai@6.15.0(ws@8.18.3)(zod@4.2.1))
-      uuid: 10.0.0
-      zod: 4.2.1
-    transitivePeerDependencies:
-      - '@opentelemetry/api'
-      - '@opentelemetry/exporter-trace-otlp-proto'
-      - '@opentelemetry/sdk-trace-base'
-      - openai
-      - react
-      - react-dom
-      - zod-to-json-schema
-
  langsmith@0.4.2(openai@6.15.0(ws@8.18.3)(zod@4.2.1)):
    dependencies:
      '@types/uuid': 10.0.0
@@ -8281,11 +8184,6 @@ snapshots:
      eventemitter3: 4.0.7
      p-timeout: 3.2.0

-  p-retry@4.6.2:
-    dependencies:
-      '@types/retry': 0.12.0
-      retry: 0.13.1
-
  p-timeout@3.2.0:
    dependencies:
      p-finally: 1.0.0
@@ -8844,14 +8742,12 @@ snapshots:
      onetime: 5.1.2
      signal-exit: 3.0.7

-  retry-axios@2.6.0(axios@1.13.2):
+  retry-axios@2.6.0(axios@1.13.2(debug@4.4.3)):
    dependencies:
      axios: 1.13.2(debug@4.4.3)

  retry@0.12.0: {}

-  retry@0.13.1: {}
-
  rimraf@2.6.3:
    dependencies:
      glob: 7.2.3
@@ -9378,8 +9274,6 @@ snapshots:

  uuid@10.0.0: {}

-  uuid@9.0.1: {}
-
  verror@1.10.1:
    dependencies:
      assert-plus: 1.0.0
@@ -1,55 +1,55 @@
-import { PDFLoader } from '@langchain/community/document_loaders/fs/pdf';
-import { DocxLoader } from '@langchain/community/document_loaders/fs/docx';
-import { RecursiveCharacterTextSplitter } from 'langchain/text_splitter';
-import { Document } from 'langchain/document';
+import { PDFLoader } from '@langchain/community/document_loaders/fs/pdf'
+import { DocxLoader } from '@langchain/community/document_loaders/fs/docx'
+import { RecursiveCharacterTextSplitter } from '@langchain/textsplitters'
+import { Document } from '@langchain/core/documents'

 export class IngestionService {
-  private static instance: IngestionService;
+  private static instance: IngestionService

  private constructor() {}

  public static getInstance(): IngestionService {
    if (!IngestionService.instance) {
-      IngestionService.instance = new IngestionService();
+      IngestionService.instance = new IngestionService()
    }
-    return IngestionService.instance;
+    return IngestionService.instance
  }

  public async loadFile(filePath: string): Promise<Document[]> {
-    let loader;
+    let loader
    if (filePath.endsWith('.pdf')) {
      loader = new PDFLoader(filePath, {
-        splitPages: false, // Combine pages or keep separate? Usually separate is better for granular chunks, but let's see. 
+        splitPages: false // Combine pages or keep separate? Usually separate is better for granular chunks, but let's see.
        // Langchain default is one doc per page.
-      });
+      })
    } else if (filePath.endsWith('.docx')) {
-      loader = new DocxLoader(filePath);
+      loader = new DocxLoader(filePath)
    } else {
-      throw new Error('Unsupported file type');
+      throw new Error('Unsupported file type')
    }

-    const docs = await loader.load();
-    return docs;
+    const docs = await loader.load()
+    return docs
  }

  public async splitDocuments(docs: Document[]): Promise<Document[]> {
    const splitter = new RecursiveCharacterTextSplitter({
      chunkSize: 500,
-      chunkOverlap: 50,
-    });
+      chunkOverlap: 50
+    })

-    const splitDocs = await splitter.splitDocuments(docs);
-    return splitDocs;
+    const splitDocs = await splitter.splitDocuments(docs)
+    return splitDocs
  }

  public async processFile(filePath: string) {
-    const docs = await this.loadFile(filePath);
-    const chunks = await this.splitDocuments(docs);
+    const docs = await this.loadFile(filePath)
+    const chunks = await this.splitDocuments(docs)
    // Add metadata
-    chunks.forEach(chunk => {
-        chunk.metadata.source = filePath;
-        chunk.metadata.filename = filePath.split('/').pop();
-    });
-    return chunks;
+    chunks.forEach((chunk) => {
+      chunk.metadata.source = filePath
+      chunk.metadata.filename = filePath.split('/').pop()
+    })
+    return chunks
  }
 }
@@ -1,50 +1,50 @@
-import React, { useState } from 'react';
-import { Input, Button, Upload, message, List, Card, Typography } from 'antd';
-import { UploadOutlined } from '@ant-design/icons';
+import React, { useState } from 'react'
+import { Input, Button, Upload, message, List, Card, Typography } from 'antd'
+import { UploadOutlined } from '@ant-design/icons'

-const { Paragraph } = Typography;
+const { Paragraph } = Typography

 function App(): React.JSX.Element {
-  const [searchResults, setSearchResults] = useState<any[]>([]);
-  const [loading, setLoading] = useState(false);
+  const [searchResults, setSearchResults] = useState<any[]>([])
+  const [loading, setLoading] = useState(false)

  const handleSearch = async (value: string) => {
-    if (!value.trim()) return;
-    setLoading(true);
+    if (!value.trim()) return
+    setLoading(true)
    try {
-      const results = await window.api.search(value);
-      setSearchResults(results);
+      const results = await window.api.search(value)
+      setSearchResults(results)
    } catch (error) {
-      console.error(error);
-      message.error('Search failed');
+      console.error(error)
+      message.error('Search failed')
    } finally {
-      setLoading(false);
+      setLoading(false)
    }
-  };
+  }

  const handleUpload = async (file: any) => {
    // In Electron, File object has a 'path' property
-    const filePath = file.path;
+    const filePath = file.path

    if (!filePath) {
-      message.error('Could not get file path');
-      return Upload.LIST_IGNORE;
+      message.error('Could not get file path')
+      return Upload.LIST_IGNORE
    }

    try {
-      message.loading({ content: 'Ingesting file...', key: 'ingest', duration: 0 });
-      const result = await window.api.ingestFile(filePath);
+      message.loading({ content: 'Ingesting file...', key: 'ingest', duration: 0 })
+      const result = await window.api.ingestFile(filePath)
      if (result.success) {
-        message.success({ content: `Ingested ${result.count} chunks`, key: 'ingest' });
+        message.success({ content: `Ingested ${result.count} chunks`, key: 'ingest' })
      } else {
-        message.error({ content: `Ingestion failed: ${result.error}`, key: 'ingest' });
+        message.error({ content: `Ingestion failed: ${result.error}`, key: 'ingest' })
      }
    } catch (error) {
-      console.error(error);
-      message.error({ content: 'Ingestion error', key: 'ingest' });
+      console.error(error)
+      message.error({ content: 'Ingestion error', key: 'ingest' })
    }
-    return Upload.LIST_IGNORE; // Prevent upload
-  };
+    return Upload.LIST_IGNORE // Prevent upload
+  }

  return (
    <div className="p-8 max-w-4xl mx-auto h-screen flex flex-col">
@@ -71,7 +71,9 @@ function App(): React.JSX.Element {
          renderItem={(item) => (
            <List.Item>
              <Card
-                title={<span className="text-sm text-gray-500">Score: {item.score?.toFixed(4)}</span>}
+                title={
+                  <span className="text-sm text-gray-500">Score: {item.score?.toFixed(4)}</span>
+                }
                className="w-full shadow-sm hover:shadow-md transition-shadow"
              >
                <Paragraph ellipsis={{ rows: 3, expandable: true, symbol: 'more' }}>
@@ -88,7 +90,7 @@ function App(): React.JSX.Element {
        />
      </div>
    </div>
-  );
+  )
 }

-export default App;
+export default App