traxxx/src/scrapers/xempire.js

'use strict';

const Promise = require('bluebird');
const bhttp = require('bhttp');
const cheerio = require('cheerio');
const moment = require('moment');

const knex = require('../knex');
const { matchTags } = require('../tags');
const pluckPhotos = require('../utils/pluck-photos');

const defaultTags = {
    hardx: [],
    darkx: ['interracial'],
    eroticax: [],
    lesbianx: ['lesbian'],
};

async function fetchPhotos(url) {
    const res = await bhttp.get(url);

    return res.body.toString();
}

function scrapePhotos(html) {
    const $ = cheerio.load(html, { normalizeWhitespace: true });

    const unlockedPhotos = $('.preview .imgLink.pgUnlocked')
        .map((photoIndex, photoElement) => $(photoElement).attr('href')).toArray();

    const lockedThumbnails = $('.preview .imgLink.lockedPicture img')
        .map((photoIndex, photoElement) => $(photoElement)
            .attr('src'))
        // .replace('_tb.jpg', '.jpg')) does not always work
        .toArray();

    return unlockedPhotos.concat(lockedThumbnails);
}

async function getPhotos(albumPath, siteDomain, site) {
    const albumUrl = `https://${siteDomain}${albumPath}`;

    const html = await fetchPhotos(albumUrl);
    const $ = cheerio.load(html, { normalizeWhitespace: true });
    const photos = scrapePhotos(html);

    const pages = $('.paginatorPages a').map((pageIndex, pageElement) => $(pageElement).attr('href')).toArray();

    const otherPhotos = await Promise.map(pages, async (page) => {
        const pageUrl = `https://${siteDomain}${page}`;
        const pageHtml = await fetchPhotos(pageUrl);

        return scrapePhotos(pageHtml);
    }, {
        concurrency: 2,
    });

    const allPhotos = photos.concat(otherPhotos.flat());

    const photoLimit = (site.network.parameters && site.network.parameters.photoLimit) || 25;
    const photoIndexes = pluckPhotos(allPhotos.length - 1, photoLimit);

    const pluckedPhotos = photoIndexes.map(photoIndex => allPhotos[photoIndex]);

    return pluckedPhotos;
}

function scrape(html, site) {
    const $ = cheerio.load(html, { normalizeWhitespace: true });
    const scenesElements = $('li[data-itemtype=scene]').toArray();

    return scenesElements.map((element) => {
        const sceneLinkElement = $(element).find('.sceneTitle a');

        const url = `${site.url}${sceneLinkElement.attr('href')}`;
        const title = sceneLinkElement.attr('title');

        const entryId = $(element).attr('data-itemid');

        const date = moment
            .utc($(element).find('.sceneDate').text(), 'MM-DD-YYYY')
            .toDate();

        const actors = $(element).find('.sceneActors a')
            .map((actorIndex, actorElement) => $(actorElement).attr('title'))
            .toArray();

        const [likes, dislikes] = $(element).find('.value')
            .toArray()
            .map(value => Number($(value).text()));

        const poster = $(element).find('.imgLink img').attr('data-original');
        const trailer = `https://videothumb.gammacdn.com/307x224/${entryId}.mp4`;

        return {
            url,
            entryId,
            title,
            actors,
            director: 'Mason',
            date,
            poster,
            trailer: {
                src: trailer,
                quality: 224,
            },
            rating: {
                likes,
                dislikes,
            },
            site,
        };
    });
}

async function scrapeScene(html, url, site) {
    const $ = cheerio.load(html, { normalizeWhitespace: true });
    const json = $('script[type="application/ld+json"]').html();
    const videoJson = $('script:contains("window.ScenePlayerOptions")').html();

    const data = JSON.parse(json)[0];
    const videoData = JSON.parse(videoJson.slice(videoJson.indexOf('{"id":'), videoJson.indexOf('};') + 1));

    const entryId = new URL(url).pathname.split('/').slice(-1)[0];

    const title = $('meta[name="twitter:title"]').attr('content');
    const description = data.description || $('meta[name="twitter:description"]').attr('content');
    // date in data object is not the release date of the scene, but the date the entry was added
    const date = moment.utc($('.updatedDate').first().text(), 'MM-DD-YYYY').toDate();

    const actors = data.actor
        .sort(({ gender: genderA }, { gender: genderB }) => {
            if (genderA === 'female' && genderB === 'male') return -1;
            if (genderA === 'male' && genderB === 'female') return 1;

            return 0;
        })
        .map(actor => actor.name);

    const stars = (data.aggregateRating.ratingValue / data.aggregateRating.bestRating) * 5;

    const duration = moment.duration(data.duration.slice(2).split(':')).asSeconds();

    const siteDomain = $('meta[name="twitter:domain"]').attr('content');
    const siteId = siteDomain && siteDomain.split('.')[0].toLowerCase();
    const siteUrl = siteDomain && `https://www.${siteDomain}`;

    const poster = videoData.picPreview;
    const trailer = `${videoData.playerOptions.host}${videoData.url}`;

    const photos = await getPhotos($('.picturesItem a').attr('href'), siteDomain, site);

    const rawTags = data.keywords.split(', ');

    const [channelSite, tags] = await Promise.all([
        site.isFallback
            ? knex('sites')
                .where({ url: siteUrl })
                .orWhere({ slug: siteId })
                .first()
            : site,
        matchTags([...defaultTags[siteId], ...rawTags]),
    ]);

    return {
        url: channelSite ? `${channelSite.url}/en/video/${new URL(url).pathname.split('/').slice(-2).join('/')}` : url,
        entryId,
        title,
        date,
        actors,
        director: 'Mason',
        description,
        duration,
        poster,
        photos,
        trailer: {
            src: trailer,
            quality: parseInt(videoData.sizeOnLoad, 10),
        },
        tags,
        rating: {
            stars,
        },
        site: channelSite || site,
    };
}

async function fetchLatest(site, page = 1) {
    const res = await bhttp.get(`${site.url}/en/videos/AllCategories/0/${page}`);

    return scrape(res.body.toString(), site);
}

async function fetchUpcoming(site) {
    const res = await bhttp.get(`${site.url}/en/videos/AllCategories/0/1/upcoming`);

    return scrape(res.body.toString(), site);
}

async function fetchScene(url, site) {
    const res = await bhttp.get(url);

    return scrapeScene(res.body.toString(), url, site);
}

module.exports = {
    fetchLatest,
    fetchUpcoming,
    fetchScene,
};