webstreamr-github/src/extractor/Dropload.ts

51 lines
1.5 KiB
TypeScript

import bytes from 'bytes';
import * as cheerio from 'cheerio';
import { Extractor } from './types';
import { extractUrlFromPacked, Fetcher } from '../utils';
import { Context, Meta } from '../types';
import { NotFoundError } from '../error';
export class Dropload implements Extractor {
readonly id = 'dropload';
readonly label = 'Dropload';
readonly ttl = 900000; // 15m
private readonly fetcher: Fetcher;
constructor(fetcher: Fetcher) {
this.fetcher = fetcher;
}
readonly supports = (_ctx: Context, url: URL): boolean => null !== url.host.match(/dropload/);
readonly extract = async (ctx: Context, url: URL, meta: Meta) => {
const normalizedUrl = new URL(url.href.replace('/e/', '/').replace('/embed-', '/'));
const html = await this.fetcher.text(ctx, normalizedUrl);
if (html.includes('File Not Found')) {
throw new NotFoundError();
}
const heightMatch = html.match(/\d{3,}x(\d{3,}),/) as string[];
const sizeMatch = html.match(/([\d.]+ ?[GM]B)/) as string[];
const $ = cheerio.load(html);
const title = $('.videoplayer h1').text().trim();
return {
url: extractUrlFromPacked(html, [/sources:\[{file:"(.*?)"/]),
label: this.label,
sourceId: `${this.id}_${meta.countryCode.toLowerCase()}`,
ttl: this.ttl,
meta: {
bytes: bytes.parse(sizeMatch[1] as string) as number,
height: parseInt(heightMatch[1] as string) as number,
title,
...meta,
},
};
};
}