Files
omnivore/packages/puppeteer-parse/apple-news-handler.js
2022-02-11 09:24:33 -08:00

40 lines
1.2 KiB
JavaScript

/* eslint-disable no-undef */
/* eslint-disable no-empty */
/* eslint-disable @typescript-eslint/explicit-function-return-type */
/* eslint-disable @typescript-eslint/no-var-requires */
/* eslint-disable @typescript-eslint/no-require-imports */
require('dotenv').config();
const Url = require('url');
const axios = require('axios');
const { promisify } = require('util');
const { DateTime } = require('luxon');
const os = require('os');
const jsdom = require("jsdom");
const { Cipher } = require('crypto');
const { JSDOM } = jsdom;
exports.appleNewsHandler = {
shouldPrehandle: (url, env) => {
const u = new URL(url);
if (u.hostname === 'apple.news') {
return true;
}
return false
},
prehandle: async (url, env) => {
const MOBILE_USER_AGENT = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/94.0.4606.61 Safari/537.36'
const response = await axios.get(url, { headers: { 'User-Agent': MOBILE_USER_AGENT } } );
const data = response.data;
const dom = new JSDOM(data);
// make sure its a valid URL by wrapping in new URL
const u = new URL(dom.window.document.querySelector('span.click-here').parentNode.href);
return { url: u.href };
}
}