@lobehub/chat 1.0.0 → 1.0.2

package/src/migrations/FromV5ToV6/fixtures/session-input-v5.json

@@ -0,0 +1,81 @@
+{
+  "exportType": "sessions",
+  "state": {
+    "sessions": [
+      {
+        "config": {
+          "autoCreateTopicThreshold": 2,
+          "displayMode": "chat",
+          "enableAutoCreateTopic": true,
+          "historyCount": 1,
+          "model": "gpt-3.5-turbo",
+          "params": {
+            "frequency_penalty": 0,
+            "presence_penalty": 0,
+            "temperature": 0.6,
+            "top_p": 1
+          },
+          "plugins": [],
+          "provider": "openai",
+          "systemRole": "",
+          "tts": {
+            "showAllLocaleVoice": false,
+            "sttLocale": "auto",
+            "ttsService": "openai",
+            "voice": {
+              "openai": "alloy"
+            }
+          }
+        },
+        "group": "default",
+        "meta": {},
+        "pinned": false,
+        "type": "agent",
+        "createdAt": 1715710690494,
+        "id": "6626ebf6-fb45-4d5b-a6e8-f77b0d9b77fa",
+        "updatedAt": 1715710690494,
+        "model": "gpt-3.5-turbo"
+      },
+      {
+        "config": {
+          "autoCreateTopicThreshold": 2,
+          "displayMode": "chat",
+          "enableAutoCreateTopic": true,
+          "historyCount": 1,
+          "model": "claude-3-sonnet-20240229",
+          "params": {
+            "frequency_penalty": 0,
+            "presence_penalty": 0,
+            "temperature": 0.6,
+            "top_p": 1
+          },
+          "plugins": [],
+          "systemRole": "As the Assistant and Judge for TOEFL Writing, your primary role is to deal with tasks related to TOEFL Writing.\n\n- You can assist the user with TOEFL Writing, or give the user feedback based on the rubrics and his writing.\n- You need to know that TOFEL Writing has two parts: Integrated Writing and Writing for an Academic Discussion.\n- For Integrated Writing (IW): The writing prompt includes reading passages and listening portions (the user will send the script). The user read a brief text of at least 230 – 300 words on a general or academic topic before listening to a lecture segment on the same subject. User Usersthen asked to summarize what was stated in the lecture and how it assists or weakens the concept of the reading passage in an essay prompt.\n- For Writing for an Academic Discussion (WAD): The question simulates a discussion board on a university website. Students will see a question posted by a professor and two student responses. The student’s job is to read everything and then write their response.\n\n## Rubrics for IW:\n\n- 5 Points (Full Mark): A response at this level successfully selects the important information from the lecture and coherently and accurately presents this information in relation to the relevant information presented in the reading. The response is well organized, and occasional language errors that are present do not result in inaccurate or imprecise presentation of content or connections.\n- 4 Points: A response at this level is generally good in selecting the important information from the lecture and in coherently and accurately presenting this information in relation to the relevant information in the reading, but it may have minor omission, inaccuracy, vagueness, or imprecision of some content from the lecture or in connection to points made in the reading. A response is also scored at this level if it has more frequent or noticeable minor language errors, as long as such usage and grammatical structures do not result in anything more than an occasional lapse of clarity or in the connection of ideas.\n- 3 Points: A response at this level contains some important information from the lecture and conveys some relevant connection to the reading, but it is marked by one or more of the following:\n  - Although the overall response is definitely oriented to the task, it conveys only vague, global, unclear, or somewhat imprecise connection of the points made in the lecture to points made in the reading.\n  - The response may omit one major key point made in the lecture.\n  - Some key points in the lecture or the reading, or connections between the two, may be incomplete, inaccurate, or imprecise.\n  - Errors of usage and/or grammar may be more frequent or may result in noticeably vague expressions or obscured meanings in conveying ideas and connections.\n- 2 Points: A response at this level contains some relevant information from the lecture, but is marked by significant language difficulties or by significant omission or inaccuracy of important ideas from the lecture or in the connections between the lecture and the reading; a response at this level is marked by one or more of the following:\n  - The response significantly misrepresents or completely omits the overall connection between the lecture and the reading.\n  - The response significantly omits or significantly misrepresents important points made in the lecture.\n  - The response contains language errors or expressions that largely obscure connections or meaning at key junctures or that would likely obscure understanding of key ideas for a reader not already familiar with the reading and the lecture.\n- 1 Point: A response at this level is marked by one or more of the following:\n  - The response provides little or no meaningful or relevant coherent content from the lecture.\n  - The language level of the response is so low that it is difficult to derive meaning.\n- 0 Point: A response at this level merely copies sentences from the reading, rejects the topic or is otherwise not connected to the topic, is written in a foreign language, consists of keystroke characters, or is blank.\n\n## Rubrics for WAD:\n\n- 5 Points: A fully successful response - The response is a relevant and very clearly expressed contribution to the online discussion, and it demonstrates consistent facility in the use of language. A typical response displays the following:\n  - Relevant and well-elaborated explanations, exemplifications, and/or details\n  - Effective use of a variety of syntactic structures and precise, idiomatic word choice\n  - Almost no lexical or grammatical errors other than those expected from a competent writer writing under timed conditions (e.g., common typos or common misspellings or substitutions like there/their)\n- 4 Points: A generally successful response - The response is a relevant contribution to the online discussion, and facility in the use of language allows the writer’s ideas to be easily understood. A typical response displays the following:\n  - Relevant and adequately elaborated explanations, exemplifications, and/or details\n  - A variety of syntactic structures and appropriate word choice\n  - Few lexical or grammatical errors\n- 3 Points: A partially successful response - The response is a mostly relevant and mostly understandable contribution to the online discussion, and there is some facility in the use of language. A typical response displays the following:\n  - Elaboration in which part of an explanation, example, or detail may be missing, unclear, or irrelevant\n  - Some variety in syntactic structures and a range of vocabulary\n  - Some noticeable lexical and grammatical errors in sentence structure, word form, or use of idiomatic language\n- 2 Points: A mostly unsuccessful response - The response reflects an attempt to contribute to the online discussion, but limitations in the use of language may make ideas hard to follow. A typical response displays the following:\n  - Ideas that may be poorly elaborated or only partially relevant\n  - A limited range of syntactic structures and vocabulary\n  - An accumulation of errors in sentence structure, word forms, or use\n- 1 Point: An unsuccessful response - The response reflects an ineffective attempt to contribute to the online discussion, and limitations in the use of language may prevent the expression of ideas. A typical response may display the following:\n  - Words and phrases that indicate an attempt to address the task but with few or no coherent ideas\n  - Severely limited range of syntactic structures and vocabulary\n  - Serious and frequent errors in the use of language\n  - Minimal original language; any coherent language is mostly borrowed from the stimulus.\n- 0 Point: The response is blank, rejects the topic, is not in English, is entirely copied from the prompt, is entirely unconnected to the prompt, or consists of arbitrary keystrokes.\n\nDepending on minor factors, the score can fall between two points, and you need to take a decimal, such as: 4.25 or 4.5, depending on the other details.\n\n### Score Conversion Rules:\n\n- 4-5 Points → Scaled Score: 24-30\n- 3-4 Points → Scaled Score: 17-23\n- 2-3 Points → Scaled Score: 13-16\n- 1-2 Points → Scaled Score: 7-12\n- 0-1 Point → Scaled Score: 1-6\n- 0 Point → Scaled Score: 0\n\n// Since the scaled score is a range, you should judge more carefully about the specific score the user falls in the scaled range, like 26.\n\n## Feedback format:\n\n    ## Raw Score: [RAW SCORE]\n    ## Scaled Score: [SCALED SPECIFIC SCORE]\n    In your writing, ...[Specific problems in user's writing(Use bullet list)]\n",
+          "tts": {
+            "showAllLocaleVoice": false,
+            "sttLocale": "auto",
+            "ttsService": "openai",
+            "voice": {
+              "openai": "alloy"
+            }
+          },
+          "provider": "anthropic"
+        },
+        "group": "default",
+        "meta": {
+          "avatar": "📝",
+          "description": "Your TOEFL Writing assistant and evaluator, specializing in feedback and guidance.",
+          "tags": ["writing", "study"],
+          "title": "TOEFL Writing Tutor"
+        },
+        "type": "agent",
+        "createdAt": 1706114617725,
+        "id": "711618cf-a5a8-40c8-821f-aff8f82db584",
+        "updatedAt": 1709655104241,
+        "pinned": true,
+        "model": "claude-3-sonnet-20240229"
+      }
+    ]
+  },
+  "version": 5
+}

package/src/migrations/FromV5ToV6/fixtures/session-output-v6.json

@@ -0,0 +1,85 @@
+{
+  "exportType": "sessions",
+  "state": {
+    "sessions": [
+      {
+        "config": {
+          "chatConfig": {
+            "autoCreateTopicThreshold": 2,
+            "displayMode": "chat",
+            "enableAutoCreateTopic": true,
+            "historyCount": 1
+          },
+          "model": "gpt-3.5-turbo",
+          "params": {
+            "frequency_penalty": 0,
+            "presence_penalty": 0,
+            "temperature": 0.6,
+            "top_p": 1
+          },
+          "plugins": [],
+          "provider": "openai",
+          "systemRole": "",
+          "tts": {
+            "showAllLocaleVoice": false,
+            "sttLocale": "auto",
+            "ttsService": "openai",
+            "voice": {
+              "openai": "alloy"
+            }
+          }
+        },
+        "group": "default",
+        "pinned": false,
+        "type": "agent",
+        "createdAt": 1715710690494,
+        "meta": {},
+        "id": "6626ebf6-fb45-4d5b-a6e8-f77b0d9b77fa",
+        "updatedAt": 1715710690494,
+        "model": "gpt-3.5-turbo"
+      },
+      {
+        "config": {
+          "chatConfig": {
+            "autoCreateTopicThreshold": 2,
+            "displayMode": "chat",
+            "enableAutoCreateTopic": true,
+            "historyCount": 1
+          },
+          "model": "claude-3-sonnet-20240229",
+          "params": {
+            "frequency_penalty": 0,
+            "presence_penalty": 0,
+            "temperature": 0.6,
+            "top_p": 1
+          },
+          "plugins": [],
+          "systemRole": "As the Assistant and Judge for TOEFL Writing, your primary role is to deal with tasks related to TOEFL Writing.\n\n- You can assist the user with TOEFL Writing, or give the user feedback based on the rubrics and his writing.\n- You need to know that TOFEL Writing has two parts: Integrated Writing and Writing for an Academic Discussion.\n- For Integrated Writing (IW): The writing prompt includes reading passages and listening portions (the user will send the script). The user read a brief text of at least 230 – 300 words on a general or academic topic before listening to a lecture segment on the same subject. User Usersthen asked to summarize what was stated in the lecture and how it assists or weakens the concept of the reading passage in an essay prompt.\n- For Writing for an Academic Discussion (WAD): The question simulates a discussion board on a university website. Students will see a question posted by a professor and two student responses. The student’s job is to read everything and then write their response.\n\n## Rubrics for IW:\n\n- 5 Points (Full Mark): A response at this level successfully selects the important information from the lecture and coherently and accurately presents this information in relation to the relevant information presented in the reading. The response is well organized, and occasional language errors that are present do not result in inaccurate or imprecise presentation of content or connections.\n- 4 Points: A response at this level is generally good in selecting the important information from the lecture and in coherently and accurately presenting this information in relation to the relevant information in the reading, but it may have minor omission, inaccuracy, vagueness, or imprecision of some content from the lecture or in connection to points made in the reading. A response is also scored at this level if it has more frequent or noticeable minor language errors, as long as such usage and grammatical structures do not result in anything more than an occasional lapse of clarity or in the connection of ideas.\n- 3 Points: A response at this level contains some important information from the lecture and conveys some relevant connection to the reading, but it is marked by one or more of the following:\n  - Although the overall response is definitely oriented to the task, it conveys only vague, global, unclear, or somewhat imprecise connection of the points made in the lecture to points made in the reading.\n  - The response may omit one major key point made in the lecture.\n  - Some key points in the lecture or the reading, or connections between the two, may be incomplete, inaccurate, or imprecise.\n  - Errors of usage and/or grammar may be more frequent or may result in noticeably vague expressions or obscured meanings in conveying ideas and connections.\n- 2 Points: A response at this level contains some relevant information from the lecture, but is marked by significant language difficulties or by significant omission or inaccuracy of important ideas from the lecture or in the connections between the lecture and the reading; a response at this level is marked by one or more of the following:\n  - The response significantly misrepresents or completely omits the overall connection between the lecture and the reading.\n  - The response significantly omits or significantly misrepresents important points made in the lecture.\n  - The response contains language errors or expressions that largely obscure connections or meaning at key junctures or that would likely obscure understanding of key ideas for a reader not already familiar with the reading and the lecture.\n- 1 Point: A response at this level is marked by one or more of the following:\n  - The response provides little or no meaningful or relevant coherent content from the lecture.\n  - The language level of the response is so low that it is difficult to derive meaning.\n- 0 Point: A response at this level merely copies sentences from the reading, rejects the topic or is otherwise not connected to the topic, is written in a foreign language, consists of keystroke characters, or is blank.\n\n## Rubrics for WAD:\n\n- 5 Points: A fully successful response - The response is a relevant and very clearly expressed contribution to the online discussion, and it demonstrates consistent facility in the use of language. A typical response displays the following:\n  - Relevant and well-elaborated explanations, exemplifications, and/or details\n  - Effective use of a variety of syntactic structures and precise, idiomatic word choice\n  - Almost no lexical or grammatical errors other than those expected from a competent writer writing under timed conditions (e.g., common typos or common misspellings or substitutions like there/their)\n- 4 Points: A generally successful response - The response is a relevant contribution to the online discussion, and facility in the use of language allows the writer’s ideas to be easily understood. A typical response displays the following:\n  - Relevant and adequately elaborated explanations, exemplifications, and/or details\n  - A variety of syntactic structures and appropriate word choice\n  - Few lexical or grammatical errors\n- 3 Points: A partially successful response - The response is a mostly relevant and mostly understandable contribution to the online discussion, and there is some facility in the use of language. A typical response displays the following:\n  - Elaboration in which part of an explanation, example, or detail may be missing, unclear, or irrelevant\n  - Some variety in syntactic structures and a range of vocabulary\n  - Some noticeable lexical and grammatical errors in sentence structure, word form, or use of idiomatic language\n- 2 Points: A mostly unsuccessful response - The response reflects an attempt to contribute to the online discussion, but limitations in the use of language may make ideas hard to follow. A typical response displays the following:\n  - Ideas that may be poorly elaborated or only partially relevant\n  - A limited range of syntactic structures and vocabulary\n  - An accumulation of errors in sentence structure, word forms, or use\n- 1 Point: An unsuccessful response - The response reflects an ineffective attempt to contribute to the online discussion, and limitations in the use of language may prevent the expression of ideas. A typical response may display the following:\n  - Words and phrases that indicate an attempt to address the task but with few or no coherent ideas\n  - Severely limited range of syntactic structures and vocabulary\n  - Serious and frequent errors in the use of language\n  - Minimal original language; any coherent language is mostly borrowed from the stimulus.\n- 0 Point: The response is blank, rejects the topic, is not in English, is entirely copied from the prompt, is entirely unconnected to the prompt, or consists of arbitrary keystrokes.\n\nDepending on minor factors, the score can fall between two points, and you need to take a decimal, such as: 4.25 or 4.5, depending on the other details.\n\n### Score Conversion Rules:\n\n- 4-5 Points → Scaled Score: 24-30\n- 3-4 Points → Scaled Score: 17-23\n- 2-3 Points → Scaled Score: 13-16\n- 1-2 Points → Scaled Score: 7-12\n- 0-1 Point → Scaled Score: 1-6\n- 0 Point → Scaled Score: 0\n\n// Since the scaled score is a range, you should judge more carefully about the specific score the user falls in the scaled range, like 26.\n\n## Feedback format:\n\n    ## Raw Score: [RAW SCORE]\n    ## Scaled Score: [SCALED SPECIFIC SCORE]\n    In your writing, ...[Specific problems in user's writing(Use bullet list)]\n",
+          "tts": {
+            "showAllLocaleVoice": false,
+            "sttLocale": "auto",
+            "ttsService": "openai",
+            "voice": {
+              "openai": "alloy"
+            }
+          },
+          "provider": "anthropic"
+        },
+        "group": "default",
+        "meta": {
+          "avatar": "📝",
+          "description": "Your TOEFL Writing assistant and evaluator, specializing in feedback and guidance.",
+          "tags": ["writing", "study"],
+          "title": "TOEFL Writing Tutor"
+        },
+        "type": "agent",
+        "createdAt": 1706114617725,
+        "id": "711618cf-a5a8-40c8-821f-aff8f82db584",
+        "updatedAt": 1709655104241,
+        "pinned": true,
+        "model": "claude-3-sonnet-20240229"
+      }
+    ]
+  },
+  "version": 6
+}

package/src/migrations/FromV5ToV6/index.ts

@@ -0,0 +1,61 @@
+import type { Migration, MigrationData } from '@/migrations/VersionController';
+
+import { V5ConfigState, V5Session } from './types/v5';
+import { V6ConfigState, V6Session } from './types/v6';
+
+export class MigrationV5ToV6 implements Migration {
+  // from this version to start migration
+  version = 5;
+
+  migrate(data: MigrationData<V5ConfigState>): MigrationData<V6ConfigState> {
+    const { sessions } = data.state;
+
+    return {
+      ...data,
+      state: {
+        ...data.state,
+        sessions: MigrationV5ToV6.migrateSession(sessions),
+      },
+    };
+  }
+  static migrateChatConfig(config: V5Session['config']): V6Session['config'] {
+    const {
+      autoCreateTopicThreshold,
+      enableAutoCreateTopic,
+      compressThreshold,
+      enableCompressThreshold,
+      enableHistoryCount,
+      enableMaxTokens,
+      historyCount,
+      inputTemplate,
+      displayMode,
+      ...agentConfig
+    } = config;
+
+    return {
+      ...agentConfig,
+      chatConfig: {
+        autoCreateTopicThreshold,
+        compressThreshold,
+        displayMode,
+        enableAutoCreateTopic,
+        enableCompressThreshold,
+        enableHistoryCount,
+        enableMaxTokens,
+        historyCount,
+        inputTemplate,
+      },
+    };
+  }
+
+  static migrateSession(sessions: V5Session[]): V6Session[] {
+    return sessions.map(({ config, updateAt, updatedAt, createdAt, createAt, ...res }) => ({
+      ...res,
+      config: MigrationV5ToV6.migrateChatConfig(config),
+      createdAt: createdAt || createAt!,
+      updatedAt: updatedAt || updateAt!,
+    }));
+  }
+}
+
+export const MigrationAgentChatConfig = MigrationV5ToV6;

package/src/migrations/FromV5ToV6/migrations.test.ts

@@ -0,0 +1,50 @@
+import { describe } from 'vitest';
+
+import { MigrationData, VersionController } from '@/migrations/VersionController';
+
+import { MigrationV1ToV2 } from '../FromV1ToV2';
+import inputV1Data from '../FromV1ToV2/fixtures/input-v1-session.json';
+import { MigrationV2ToV3 } from '../FromV2ToV3';
+import { MigrationV3ToV4 } from '../FromV3ToV4';
+import { MigrationV4ToV5 } from '../FromV4ToV5';
+import outputDataFromV1ToV6 from './fixtures/from-v1-to-v6-output.json';
+import sessionInputV5 from './fixtures/session-input-v5.json';
+import sessionOutputV6 from './fixtures/session-output-v6.json';
+import { MigrationV5ToV6 } from './index';
+
+describe('MigrationV5ToV6', () => {
+  let migrations;
+  let versionController: VersionController<any>;
+
+  beforeEach(() => {
+    migrations = [MigrationV5ToV6];
+    versionController = new VersionController(migrations, 6);
+  });
+
+  describe('should migrate data correctly from previous versions', () => {
+    it('session config', () => {
+      const data: MigrationData = sessionInputV5;
+
+      const migratedData = versionController.migrate(data);
+
+      expect(migratedData.version).toEqual(sessionOutputV6.version);
+      expect(migratedData.state.sessions).toEqual(sessionOutputV6.state.sessions);
+    });
+  });
+
+  it('should work correct from v1 to v6', () => {
+    const data: MigrationData = inputV1Data;
+
+    versionController = new VersionController(
+      [MigrationV5ToV6, MigrationV4ToV5, MigrationV3ToV4, MigrationV2ToV3, MigrationV1ToV2],
+      6,
+    );
+
+    const migratedData = versionController.migrate(data);
+
+    expect(migratedData.version).toEqual(outputDataFromV1ToV6.version);
+    expect(migratedData.state.messages).toEqual(outputDataFromV1ToV6.state.messages);
+    expect(migratedData.state.sessions).toEqual(outputDataFromV1ToV6.state.sessions);
+    expect(migratedData.state.topics).toEqual(outputDataFromV1ToV6.state.topics);
+  });
+});

package/src/migrations/FromV5ToV6/types/v5.ts

@@ -0,0 +1,48 @@
+import { LobeAgentTTSConfig } from '@/types/agent';
+import { FewShots, LLMParams } from '@/types/llm';
+
+/**
+ * Lobe Agent
+ */
+export interface V5Session {
+  config: V5AgentConfig;
+  createAt?: number;
+  createdAt: number;
+  group?: string;
+  id: string;
+  meta: {
+    avatar?: string;
+    backgroundColor?: string;
+    description?: string;
+    tags?: string[];
+    title?: string;
+  };
+  model: string;
+  pinned?: boolean;
+  type: 'agent';
+  updateAt?: number;
+  updatedAt: number;
+}
+
+export interface V5AgentConfig {
+  autoCreateTopicThreshold: number;
+  compressThreshold?: number;
+  displayMode?: 'chat' | 'docs';
+  enableAutoCreateTopic: boolean;
+  enableCompressThreshold?: boolean;
+  enableHistoryCount?: boolean;
+  enableMaxTokens?: boolean;
+  fewShots?: FewShots;
+  historyCount?: number;
+  inputTemplate?: string;
+  model: string;
+  params: LLMParams;
+  plugins?: string[];
+  provider?: string;
+  systemRole: string;
+  tts: LobeAgentTTSConfig;
+}
+
+export interface V5ConfigState {
+  sessions: V5Session[];
+}

package/src/migrations/FromV5ToV6/types/v6.ts

@@ -0,0 +1,60 @@
+import { FewShots, LLMParams } from '@/types/llm';
+
+export type TTSServer = 'openai' | 'edge' | 'microsoft';
+
+export interface TTSConfig {
+  showAllLocaleVoice?: boolean;
+  sttLocale: 'auto' | string;
+  ttsService: TTSServer;
+  voice: {
+    edge?: string;
+    microsoft?: string;
+    openai: string;
+  };
+}
+
+export interface ChatConfig {
+  autoCreateTopicThreshold: number;
+  compressThreshold?: number;
+  displayMode?: 'chat' | 'docs';
+  enableAutoCreateTopic?: boolean;
+  enableCompressThreshold?: boolean;
+  enableHistoryCount?: boolean;
+  enableMaxTokens?: boolean;
+  historyCount?: number;
+  inputTemplate?: string;
+}
+
+export interface V6AgentConfig {
+  chatConfig: ChatConfig;
+  fewShots?: FewShots;
+  model: string;
+  params: LLMParams;
+  plugins?: string[];
+  provider?: string;
+  systemRole: string;
+  tts: TTSConfig;
+}
+
+export interface V6Session {
+  config: V6AgentConfig;
+  createdAt: number;
+
+  group?: string;
+  id: string;
+  meta: {
+    avatar?: string;
+    backgroundColor?: string;
+    description?: string;
+    tags?: string[];
+    title?: string;
+  };
+  model: string;
+  pinned?: boolean;
+  type: 'agent';
+  updatedAt: number;
+}
+
+export interface V6ConfigState {
+  sessions: V6Session[];
+}

package/src/migrations/FromV6ToV7/fixtures/output-v7-from-v1.json

@@ -0,0 +1,203 @@
+{
+  "exportType": "sessions",
+  "state": {
+    "sessions": [
+      {
+        "config": {
+          "displayMode": "chat",
+          "historyCount": 1,
+          "model": "gpt-3.5-turbo",
+          "params": {
+            "frequency_penalty": 0,
+            "presence_penalty": 0,
+            "temperature": 0.6,
+            "top_p": 1,
+            "max_tokens": 2200
+          },
+          "plugins": [],
+          "systemRole": "你是一名 Postgresql 的数据库专家。用户是一名数据库小白，你需要使用简单直白的方式告诉用户如何使用 ostgresql 和它相应的 orm 工具 primsa",
+          "enableCompressThreshold": false,
+          "enableMaxTokens": false,
+          "inputTemplate": ""
+        },
+        "createdAt": 1698463356269,
+        "id": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "meta": {
+          "avatar": "🐘",
+          "backgroundColor": "rgba(0,0,0,0)",
+          "title": "Postgresql 指南",
+          "description": "PostgreSQL 数据库和 Prisma ORM 工具的简单使用指南",
+          "tags": ["数据库", "PostgreSQL", "教程", "ORM", "Prisma"]
+        },
+        "type": "agent",
+        "updatedAt": 1698463356269,
+        "group": "default",
+        "pinned": true
+      }
+    ],
+    "topics": [
+      {
+        "createdAt": 1698463911747,
+        "id": "CZg3IEdy",
+        "title": "Prisma中的cuid函数作用是什么",
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "updatedAt": 1698463915716
+      },
+      {
+        "createdAt": 1698754463117,
+        "id": "MNWhBEh5",
+        "title": "启动 PostgreSQL Docker 镜像方法",
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "updatedAt": 1698754464867
+      }
+    ],
+    "messages": [
+      {
+        "content": "是的，你可以使用 Prisma 来自定义生成 `id` 的逻辑，例如使用 `nanoid` 库来生成唯一标识符。\n\n首先，你需要在项目中安装 `nanoid` 库：\n\n```\nnpm install nanoid\n```\n\n然后，你可以在 Prisma 的模型定义中使用 `@default` 指令，并提供一个自定义的默认值函数来生成 `id`。\n\n```prisma\nmodel Message {\n  id String @id @default(dbgenerated(\"nanoid(10)\"))\n\n  createdAt DateTime @default(now())\n  updatedAt DateTime @updatedAt\n}\n```\n\n在上述模型定义中，我们使用 `dbgenerated` 指令来定义一个自定义的默认值函数。函数 `nanoid(10)` 会生成一个长度为 10 的唯一标识符。\n\n请注意，使用自定义的 `id` 生成逻辑可能会对数据库性能产生一些影响，因为每次创建新记录时都需要执行自定义函数。因此，请确保你的自定义逻辑是高效的，并在实际使用中进行测试和评估性能。",
+        "createdAt": 1698463981313,
+        "id": "n2Tsl3Dm",
+        "meta": {},
+        "parentId": "JSZaJOGj",
+        "role": "assistant",
+        "updatedAt": 1698464004494,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "CZg3IEdy",
+        "fromModel": "gpt-3.5-turbo"
+      },
+      {
+        "content": "如何指定 自定义函数的引入位置？",
+        "createdAt": 1698464173224,
+        "id": "xPft8Msa",
+        "meta": {},
+        "role": "user",
+        "updatedAt": 1698464173226,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "CZg3IEdy"
+      },
+      {
+        "content": "在 Prisma 中，你可以通过在 `schema.prisma` 文件的顶部使用 `generator` 块来指定自定义函数的引入位置。\n\n例如，假设你想使用 `nanoid` 库来生成自定义的 `id`，你可以按照以下方式配置：\n\n```prisma\ngenerator client {\n  provider = \"prisma-client-js\"\n  output   = \"./node_modules/@prisma/client\"\n}\n\ngenerator dbgenerated {\n  provider = \"prisma-dbgenerator\"\n  output   = \"./node_modules/@prisma/client\"\n  binaryTargets = [\"native\"]\n  previewFeatures = [\"nativeTypes\"]\n  config = {\n    \"generator\": {\n      \"dbgenerated\": {\n        \"output\": \"./src/db/generated\",\n        \"provider\": \"./src/db/generated/provider.js\"\n      }\n    }\n  }\n}\n```\n\n在上述配置中，我们定义了两个生成器。`client` 生成器用于生成 Prisma 客户端代码，而 `dbgenerated` 生成器用于生成数据库相关的代码。\n\n注意到 `dbgenerated` 生成器的 `config` 部分中有一个 `provider` 字段，它指定了自定义函数的引入位置。在这个示例中，我们将自定义函数放在 `./src/db/generated/provider.js` 文件中。\n\n你可以根据你的项目结构和需要来调整路径和文件名，确保指定的 `provider` 路径正确引入了自定义函数。",
+        "createdAt": 1698464173228,
+        "id": "jZc2KKRs",
+        "meta": {},
+        "parentId": "xPft8Msa",
+        "role": "assistant",
+        "updatedAt": 1698464181896,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "CZg3IEdy",
+        "fromModel": "gpt-3.5-turbo"
+      },
+      {
+        "content": "如何启动一个 postgres Docker 镜像？",
+        "createdAt": 1698754460196,
+        "id": "LjG9XSuq",
+        "meta": {},
+        "role": "user",
+        "updatedAt": 1698754463118,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5"
+      },
+      {
+        "content": "要启动一个 PostgreSQL Docker 镜像，你可以按照以下步骤操作：\n\n1. 首先，确保你已经安装了 Docker，可以在终端中运行 `docker --version` 来检查。\n\n2. 在终端中运行以下命令以下载并启动一个 PostgreSQL 镜像：\n\n   ```shell\n   docker run --name my-postgres -e POSTGRES_PASSWORD=mysecretpassword -p 5432:5432 -d postgres\n   ```\n\n   这将从 Docker Hub 下载 PostgreSQL 镜像",
+        "createdAt": 1698754479751,
+        "id": "CL7E1Mc5",
+        "meta": {},
+        "parentId": "LjG9XSuq",
+        "role": "assistant",
+        "updatedAt": 1698754507690,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5",
+        "fromModel": "gpt-3.5-turbo"
+      },
+      {
+        "content": "如何指定本地存储的目录到 /www/database/postgres",
+        "createdAt": 1698754591763,
+        "id": "UIvHZr7m",
+        "meta": {},
+        "role": "user",
+        "updatedAt": 1698754591764,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5"
+      },
+      {
+        "content": "要将本地存储的目录映射到容器内的 `/var/lib/postgresql/data` 目录，你可以使用 Docker 的 `-v` 参数。\n\n以下是示例命令：\n\n```shell\ndocker run --name my-postgres -e POSTGRES_PASSWORD=mysecretpassword -p 5432:5432 -v /www/database/postgres:/var/lib/postgresql/data -d postgres\n```\n\n在这个命令中，我们使用了 `-v` 参数来指定本地目录 `/www/database/postgres` 映射到容器内的 `/var/lib/postgresql/data` 目录。这样，所有的 PostgreSQL 数据将会保存在本地指定的目录中。",
+        "createdAt": 1698754591765,
+        "id": "XLJOK44r",
+        "meta": {},
+        "parentId": "UIvHZr7m",
+        "role": "assistant",
+        "updatedAt": 1698754609275,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5",
+        "fromModel": "gpt-3.5-turbo"
+      },
+      {
+        "content": "docker run --name postgres -e POSTGRES_PASSWORD=DSLapHYdzNrai5j6 -p 33300:5432 -v /www/database/postgres:/var/lib/postgresql/data -d postgres",
+        "createdAt": 1698754700529,
+        "id": "W3J6jQjm",
+        "meta": {},
+        "role": "user",
+        "updatedAt": 1698754700530,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5"
+      },
+      {
+        "content": "如何启动 pgadmin",
+        "createdAt": 1698755749886,
+        "id": "L7mq0ehV",
+        "meta": {},
+        "role": "user",
+        "updatedAt": 1698755749888,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5"
+      },
+      {
+        "content": "要启动 pgAdmin，你可以按照以下步骤操作：\n\n1. 首先，确保你已经安装了 Docker，并且已经成功启动了 PostgreSQL Docker 容器。\n\n2. 在终端中运行以下命令以启动 pgAdmin Docker 容器：\n\n   ```shell\n   docker run --name my-pgadmin -p 5050:80 -e PGADMIN_DEFAULT_EMAIL=myemail@example.com -e PGADMIN_DEFAULT_PASSWORD=mypassword -d dpage/pgadmin4\n   ```\n\n   在这个命令中，我们使用了 `-p` 参数来将容器的端口 80 映射到本地的端口 5050。你可以根据需要修改这个端口映射。\n\n   我们还使用了 `-e` 参数来设置 pgAdmin 的默认登录邮箱和密码。你可以根据需要修改这些值。\n\n3. 在浏览器中访问 `http://localhost:5050`，你将看到 pgAdmin 的登录界面。\n\n4. 使用在上一步中设置的邮箱和密码进行登录。\n\n5. 在 pgAdmin 中，你可以添加一个新的服务器连接，以连接到运行在 Docker 容器中的 PostgreSQL 数据库。要添加一个新的服务器连接，你需要提供以下信息：\n\n   - 主机名/地址：使用 Docker 容器的 IP 地址或者 `localhost`。\n   - 端口：使用 Docker 容器中 PostgreSQL 的映射端口，例如 33300。\n   - 用户名和密码：使用在启动 PostgreSQL 容器时设置的用户名和密码。\n\n   完成上述信息的填写后，点击保存并连接到 PostgreSQL 服务器。\n\n现在，你已经成功启动了 pgAdmin 并连接到了你的 PostgreSQL 数据库。你可以使用 pgAdmin 来管理和操作你的数据库。",
+        "createdAt": 1698755749889,
+        "id": "d5XTX9EQ",
+        "meta": {},
+        "parentId": "L7mq0ehV",
+        "role": "assistant",
+        "updatedAt": 1698755786183,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "topicId": "MNWhBEh5",
+        "fromModel": "gpt-3.5-turbo"
+      },
+      {
+        "content": "abcabc",
+        "createdAt": 1690650544852,
+        "id": "KPPDiRyW",
+        "meta": {},
+        "parentId": "42k72jMi",
+        "role": "function",
+        "updatedAt": 1690650572399,
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+
+        "fromModel": "gpt-3.5-turbo-16k",
+        "plugin": {
+          "apiName": "websiteCrawler",
+          "arguments": "{\n  \"url\": \"https://mp.weixin.qq.com/s?__biz=MjM5MTA1MjAxMQ==&mid=2651264337&idx=1&sn=d7d9126578c74c912e1f0d42cb5629da&chksm=bd48ccd58a3f45c3f2cbc7d7b732c820b9e7cd6b547c06acc8170b233710b5fb5ed62f5fd94d&mpshare=1&scene=1&srcid=07294Mpw7C6JCLgtagL1cBDR&sharer_sharetime=1690622712877&sharer_shareid=0358058a42fc25387d28337fc3d22c3c#rd\"\n}",
+          "identifier": "websiteCrawler",
+          "type": "default"
+        }
+      },
+      {
+        "content": "bbbbb",
+        "createdAt": 1700065743405,
+        "id": "9cRjevRQ",
+        "meta": {},
+        "parentId": "3nDXtEKv",
+        "role": "function",
+        "sessionId": "06cc3e20-e870-4099-a619-c07a849d742d",
+        "updatedAt": 1700065751861,
+        "fromModel": "gpt-3.5-turbo-16k",
+        "plugin": {
+          "apiName": "getWebsiteContent",
+          "arguments": "{\n  \"url\": \"https://nodejs.org/api/packages.html#dual-package-hazard\"\n}",
+          "identifier": "website-crawler",
+          "type": "default"
+        }
+      }
+    ]
+  },
+  "version": 7
+}