chatgpt-plugin/utils/tools/WebsiteTool.js
ikechan8370 b7427e74c4
feat: 智能模式,添加群管、试图、联网搜索、发图、发音乐和视频等功能 (#488)
* fix: 2.7 dev start

* feat: 初步支持function call(WIP)

* fix: syntax error

* fix: syntax error

* feat: 群聊上下文

* fix: 暂时阉割掉全员禁言功能

* fix: 修改禁言时间范围

* fix: 修复一些功能易用性

* fix: 只有管理员和群主才能用jinyan和kickout

* fix: 加回来禁言和踢出

* fix: 修复管理员权限判断问题(可能吧)

* fix: 试图优化逻辑

* fix: fuck openai documents

* fix: 删掉认主不然一直禁言我烦死了

* fix: 哔哩哔哩封面损坏问题

* fix: 加个天气小工具

* fix: 天气不存在城市

* fix: website工具用浏览器

* feat: serp tool

* feat: 增加一个google搜索源

* fix: 加一句描述

* feat: 增加搜索来源选项

* feat: 搜图和发图

* fix: groupId format error

* fix: add a image caption tool

* fix: 修改一些提示。tool太多机器人开始混乱了

* fix: 一些极端的措施

* fix: 增加一些提示和一个暂时的公共接口

* fix: 收拾一下

* fix: 修改命令正则

* fix: 修改一些提示

* fix: move send avatar into send picture tool

* fix: 修复解除禁言的bug
2023-06-25 01:09:29 +08:00

86 lines
3.1 KiB
JavaScript
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

import { AbstractTool } from './AbstractTool.js'
import { ChatGPTAPI } from '../openai/chatgpt-api.js'
import { Config } from '../config.js'
import fetch from 'node-fetch'
import proxy from 'https-proxy-agent'
import { getMaxModelTokens } from '../common.js'
import { ChatGPTPuppeteer } from '../browser.js'
export class WebsiteTool extends AbstractTool {
name = 'website'
parameters = {
properties: {
url: {
type: 'string',
description: '要访问的网站网址'
}
},
required: ['url']
}
func = async function (opts) {
let { url } = opts
try {
// let res = await fetch(url, {
// headers: {
// 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/114.0.0.0 Safari/537.36'
// }
// })
// let text = await res.text()
let origin = false
if (!Config.headless) {
Config.headless = true
origin = true
}
let ppt = new ChatGPTPuppeteer()
let browser = await ppt.getBrowser()
let page = await browser.newPage()
await page.goto(url, {
waitUntil: 'networkidle2'
})
let text = await page.content()
await page.close()
if (origin) {
Config.headless = false
}
// text = text.replace(/<style\b[^<]*(?:(?!<\/style>)<[^<]*)*<\/style>/gi, '')
// .replace(/<script\b[^<]*(?:(?!<\/script>)<[^<]*)*<\/script>/gi, '')
// .replace(/<head\b[^<]*(?:(?!<\/head>)<[^<]*)*<\/head>/gi, '')
// .replace(/<!--[\s\S]*?-->/gi, '')
text = text.replace(/<style\b[^<]*(?:(?!<\/style>)<[^<]*)*<\/style>/gi, '') // 移除<style>标签及其内容
.replace(/<[^>]+style\s*=\s*(["'])(?:(?!\1).)*\1[^>]*>/gi, '') // 移除带有style属性的标签
.replace(/<[^>]+>/g, '')
let maxModelTokens = getMaxModelTokens(Config.model)
text = text.slice(0, Math.min(text.length, maxModelTokens - 1600))
let api = new ChatGPTAPI({
apiBaseUrl: Config.openAiBaseUrl,
apiKey: Config.apiKey,
debug: false,
completionParams: {
model: Config.model
},
fetch: (url, options = {}) => {
const defaultOptions = Config.proxy
? {
agent: proxy(Config.proxy)
}
: {}
const mergedOptions = {
...defaultOptions,
...options
}
return fetch(url, mergedOptions)
},
maxModelTokens
})
const htmlContentSummaryRes = await api.sendMessage(`这是一个网页html经过筛选的内容请你进一步去掉其中的标签、样式、script等无用信息并从中提取出其中的主体内容转换成自然语言告诉我不需要主观描述性的语言。${text}`)
let htmlContentSummary = htmlContentSummaryRes.text
return `this is the main content of website:\n ${htmlContentSummary}`
} catch (err) {
return `failed to visit the website, error: ${err.toString()}`
}
}
description = 'Useful when you want to browse a website by url'
}