traxxx/src/scrapers/whalemember.js

'use strict';

const unprint = require('unprint');

const { stripQuery } = require('../utils/url');
const slugify = require('../utils/slugify');

const dateRegex = /\d{4}-\d{2}-\d{2}T/;

function scrapeLatest(scenes, fullData, channel, parameters) {
	return scenes.map(({ query }) => {
		const release = {};

		release.url = query.url('[href*="/video"]', { origin: new URL(parameters.latest || channel.url).origin });
		release.title = query.content('a[href*="/video"] strong');

		release.entryId = release.url
			? new URL(release.url).pathname.split('/').at(-1)
			: slugify(release.title);

		// Nuxt data array does not have a predictable structure, don't rely on it more than necessary
		const dataIndex = fullData?.indexOf(release.entryId);
		const data = dataIndex > -1 ? fullData?.slice(dataIndex - 5, dataIndex + 35) : [];

		// older scenes don't have date in html
		const date = data.find((item) => dateRegex.test(item));

		if (date) {
			release.date = new Date(date);
		} else {
			release.date = query.date('a[href*="/video"] + p + p', 'MM/DD/YYYY');
		}

		release.actors = query.all('a[href*="/model"]').map((actorEl) => ({
			name: unprint.query.content(actorEl),
			url: unprint.query.url(actorEl, null, { origin: channel.url }),
		}));

		const poster = query.img('img[alt]');

		if (poster) {
			release.poster = [
				stripQuery(poster),
				poster,
			];
		}

		// photos and teasers can't be reliably extracted, MP4s include trailers and FULL SCENES

		return release;
	});
}

async function passAgeCheck(ctx) {
	const ageButton = await ctx.getByText('Continue', { exact: true });

	if (await ageButton.count() > 0) {
		await ageButton.click();
	}
}

async function fetchLatest(channel, page = 1, { parameters }) {
	// going to e.g. https://holed.com/sites/holed defined by parameter gets rid of 'top rated' section, simplifying query
	const url = `${channel.parameters?.latest || channel.url.replace('/series', '/sites')}?page=${page}`;

	// site uses Nuxt without SSR, easiest to render in browser
	const res = await unprint.browserRequest(url, {
		page: {
			timeout: 120_000, // update pages can be very slow to respond, but they usually do
		},
		async control(ctx) {
			await passAgeCheck(ctx);
		},
	});

	if (res.status === 200) {
		const scenes = unprint.initAll(res.context.query.all('.card-grid > div'));
		const data = res.context.query.json('#__NUXT_DATA__');

		return scrapeLatest(scenes, data, channel, parameters);
	}

	return res.status;
}

function scrapeScene({ query }, { url, entity }) {
	const release = {};
	const { query: infoQuery } = unprint.init(query.element('//div[./*/span[contains(text(), \'Featuring\')]]'));

	// release.entryId = query.attribute('div[data-id]', 'data-id');
	release.entryId = new URL(url).pathname.split('/').at(-1);

	release.title = infoQuery.content('h2');

	const description = infoQuery.content('h2 + p + p');

	if (!description.toLowerCase().includes('n/a')) {
		release.description = description;
	}

	// Nuxt data array does not have a predictable structure, don't rely on it more than necessary
	const fullData = query.json('#__NUXT_DATA__');
	const dataIndex = fullData?.indexOf(release.entryId);
	const data = dataIndex > -1 ? fullData?.slice(dataIndex - 5, dataIndex + 50) : [];

	const date = data.find((item) => dateRegex.test(item));

	if (date) {
		release.date = new Date(date);
	}

	release.actors = infoQuery.all('a[href*="/models"]').map((actorEl) => ({
		name: unprint.query.content(actorEl),
		url: unprint.query.url(actorEl, null, { origin: entity.url }),
	}));

	const poster = query.img('media-poster img') || query.poster('dl8-video');

	if (poster) {
		release.poster = [
			stripQuery(poster),
			poster,
		];
	}

	release.photos = Array.from(new Set(query.imgs('a img[src*="content/videos"]'))).map((src) => [
		stripQuery(src),
		src,
	]);

	release.trailer = query.video('media-player video') || query.video('dl8-video source');
	release.qualities = query.contents('//table[.//span[contains(text(), \'480p\')]]//tr').map((resolution) => Number(resolution.split('x')[1])).filter(Boolean);

	return release;
}

async function fetchScene(url, entity) {
	const res = await unprint.browserRequest(url, {
		async control(ctx) {
			await passAgeCheck(ctx);

			try {
				await ctx.locator('media-player video').hover({ trial: true, timeout: 1000 }); // wait for trailer to initialize
			} catch (__error) {
				// no trailer, that's fine
			}
		},
	});

	if (res.ok) {
		return scrapeScene(res.context, { url, entity });
	}

	return res.status;
}

module.exports = {
	fetchLatest,
	fetchScene,
};