traxxx/src/scrapers/radical.js

209 lines
6.1 KiB
JavaScript
Executable File

'use strict';
const unprint = require('unprint');
const mime = require('mime');
const http = require('../utils/http');
const slugify = require('../utils/slugify');
const { convert } = require('../utils/convert');
const teaserOrder = ['large', 'small', 'mobile', 'mp4', 'jpg'];
function getVideoPath(data, parameters) {
if (data.is_published === 0 && parameters.upcoming) {
return parameters.upcoming;
}
if (parameters.videos) {
return parameters.videos;
}
return 'videos';
}
function scrapeScene(data, channel, parameters) {
const release = {};
release.entryId = data.id;
release.url = `${new URL(channel.url).origin}/${getVideoPath(data, parameters)}/${data.slug}`;
release.title = data.title;
release.description = data.description;
release.date = unprint.extractDate(data.publish_date, 'YYYY/MM/DD HH:mm:ss') || unprint.extractDate(data.formatted_date, 'Do MMM YYYY');
release.duration = data.seconds_duration || unprint.extractDuration(data.videos_duration);
// TWM in particular has a habit of putting two names in a single link https://tour.2girls1camera.com/scenes/richelle-ryan-ariella-ferrara
release.actors = (data.models_thumbs || data.models)?.flatMap((actor) => {
const actorNames = actor.name.split('&').map((actorName) => actorName.trim());
if (actorNames.length === 1) {
return {
name: actor.name,
avatar: actor.thumb,
url: actor.slug && `${channel.url}/models/${actor.slug}`,
};
}
return actorNames;
});
release.poster = data.trailer_screencap || data.thumb;
if (mime.getType(data.thumb) === 'image/gif') {
release.poster = [
data.trailer_screencap,
data.thumb,
];
} else {
release.poster = data.thumb;
// release.teaser = data.thumb;
}
release.photos = [
...data.previews?.full ? Object.values(data.previews?.full) : [], // sometimes it's an array, sometimes an object { '1': 'url' }
...data.extra_thumbnails?.filter((thumbnail) => !thumbnail.includes('mobile') // mobile is the cropped photo of a photo already in the set
&& !(thumbnail.includes('_scene') && release.poster?.includes('_scene')) // likely the same photo, filename may differ so cannot compare full path
&& !(thumbnail.includes('_player') && release.poster?.includes('_player'))
&& !(thumbnail.includes('1920') && release.poster?.includes('1920'))) || [],
];
release.caps = data.thumbs;
release.trailer = data.trailer_url;
release.teaser = data.special_thumbnails
?.filter((teaserUrl) => new URL(teaserUrl).pathname !== '/') // on Top Web Models, https://z7n5n3m8.ssl.hwcdn.net/ is listed as a teaser
.sort((teaserA, teaserB) => teaserOrder.findIndex((label) => teaserA.includes(label)) - teaserOrder.findIndex((label) => teaserB.includes(label)));
release.tags = data.tags;
release.channel = slugify(data.site, '');
release.qualities = Object.values(data.videos || []).map((video) => video.height);
release.photoCount = Number(data.photos_duration) || null;
return release;
}
function scrapeAll(scenes, channel, parameters) {
return scenes.map((data) => scrapeScene(data, channel, parameters));
}
async function fetchEndpoint(channel, parameters) {
const res = await unprint.get(channel.url);
if (res.ok) {
const data = res.context.query.json('#__NEXT_DATA__');
if (data?.buildId) {
return data.buildId;
}
}
// still allow manual configuration as a back-up
return parameters.endpoint;
}
async function fetchLatest(channel, page, { parameters }) {
const endpoint = await fetchEndpoint(channel, parameters);
if (!endpoint) {
return null;
}
const url = parameters.site
? `${channel.parent.url}/_next/data/${endpoint}/sites/${parameters.site}.json?sitename=${parameters.site}&order_by=publish_date&sort_by=desc&per_page=30&page=${page}`
: `${channel.url}/_next/data/${endpoint}/${parameters.videos || 'videos'}.json?order_by=publish_date&sort_by=desc&per_page=30&page=${page}`;
const res = await http.get(url);
if (res.ok && res.body.pageProps?.contents?.data) {
return scrapeAll(res.body.pageProps.contents.data, channel, parameters);
}
return res.status;
}
async function fetchUpcoming(channel, _page, { parameters }) {
const res = await unprint.get(channel.url, {
parser: {
runScripts: 'dangerously',
},
});
if (res.ok) {
const data = res.context.query.json('#__NEXT_DATA__');
const scene = data?.props.pageProps.upcoming_scene;
if (scene) {
return scrapeScene(scene, channel, parameters);
}
return null;
}
return res.status;
}
async function fetchScene(url, channel, _baseScene, { parameters }) {
const slug = new URL(url).pathname.split('/').at(-1);
const endpoint = await fetchEndpoint(channel);
const res = await http.get(`${channel.url}/_next/data/${endpoint}/${parameters.videos || 'videos'}/${slug}.json?slug=${slug}`);
if (res.ok && res.body.pageProps?.content) {
return scrapeScene(res.body.pageProps.content, channel, parameters);
}
return res.status;
}
function scrapeProfile(data, channel, scenes, parameters) {
const profile = {};
const bio = Object.fromEntries(Object.entries(data).map(([key, value]) => [slugify(key, '_'), value])); // keys are mixed upper and lowercase
profile.entryId = bio.id;
profile.description = bio.bio;
profile.gender = bio.gender;
profile.dateOfBirth = unprint.extractDate(bio.birthdate, 'YYYY-MM-DD');
profile.birthPlace = bio.born || bio.birthplace;
profile.age = bio.age;
profile.measurements = bio.measurements;
profile.height = convert(bio.height, 'cm');
profile.weight = convert(bio.weight, 'lb', 'kg');
profile.eyes = bio.eyes || bio.eye_color;
profile.hairColor = bio.hair || bio.hair_color;
profile.avatar = data.thumb;
if (scenes) {
profile.scenes = scrapeAll(scenes, channel, parameters);
}
return profile;
}
async function fetchProfile(actor, { channel, parameters }) {
const endpoint = await fetchEndpoint(channel);
const res = await http.get(`${channel.url}/_next/data/${endpoint}/models/${actor.slug}.json?slug=${actor.slug}`);
if (res.ok && res.body.pageProps?.model) {
return scrapeProfile(res.body.pageProps.model, channel, res.body.pageProps.model_contents, parameters);
}
return res.status;
}
module.exports = {
fetchLatest,
fetchUpcoming,
fetchScene,
fetchProfile,
};