index.js

const path = require('path');
const got = require('got');
const pMap = require('p-map');
const streamz = require('streamz');
const unzipper = require('unzipper').Parse;
const srt2vtt = require('srt-to-vtt');
const cheerio = require('cheerio');
const langsFormat = require('./langs');

const yifysubtitlesWithGoogleStorage = ({
  storage, bucketName
}) => {
  // Down const apiUri = 'http://api.yifysubtitles.com/subs';
  const uri = 'https://www.yifysubtitles.com/movie-imdb';
  const downloadUri = 'https://yifysubtitles.com';
  const langK = Object.keys(langsFormat);
  const langV = langK.map(i => langsFormat[i]);

  const formatLangLong = lang => langV[langK.indexOf(lang)];
  const formatLangShort = lang => langK[langV.indexOf(lang)];

  // Since yifysubtitle api is not working anymore we scrape the site instead
  const scrape = imdbId => {
    return got(`${uri}/${imdbId}`)
      .then(res => cheerio.load(res.body))
      .then($ => {
        return $('tbody tr')
          .map((i, el) => {
            const $el = $(el);
            return {
              rating: $el.find('.rating-cell').text(),
              language: $el
                .find('.flag-cell .sub-lang')
                .text()
                .toLowerCase(),
              url:
                $el
                  .find('.download-cell a')
                  .attr('href')
                  .replace('subtitles/', 'subtitle/') + '.zip'
            };
          })
          .get();
      });
  };

  const langFilter = (subs, langs) => {
    const data = langs.reduce((acc, l) => {
      const lang = subs
        .filter(s => s.language === l)
        .sort((a, b) => b.rating - a.rating);
      if (lang.length > 0) {
        acc[l] = lang[0];
      }

      return acc;
    }, {});
    return data;
  };

  const downloadFormat = () => (lang, url) => {
    let writed = '';
    const bucket = storage.bucket(bucketName);
    let blob = null;
    return got
      .stream(downloadUri + url)
      .pipe(unzipper())
      .pipe(
        streamz(entry => {
          const parsedPath = path.parse(entry.path);
          // Add Language to subtitle name and deete spaces
          const escapedLang = lang.replace('/', '-');
          entry.path = entry.path
            .replace(/\s+/g, '.')
            .replace(parsedPath.ext, `_${escapedLang}_${parsedPath.ext}`);
          if (parsedPath.dir === '' && parsedPath.ext === '.srt') {
            writed = entry.path.replace('srt', 'vtt');
            blob = bucket.file(writed);
            return entry.pipe(srt2vtt()).pipe(blob.createWriteStream());
          }
          entry.autodrain();
        })
      )
      .promise()
      .then(() => ({
        lang,
        langShort: formatLangShort(lang),
        path: `https://${bucket.name}.storage.googleapis.com/${blob.name}`,
        fileName: writed
      }));
  };

  const downloads = (res, opts) => {
    const download = downloadFormat();
    const { concurrency } = opts;

    return pMap(
      Object.keys(res),
      lang => download(lang, res[lang].url),
      concurrency
    );
  };

  const runConditional = (imdbId, opts, res) => {
    return Promise.resolve(
      langFilter(res, opts.langs.map(formatLangLong))
    ).then(res => downloads(res, opts));
  };

  const yifysubtitles = (imdbId, opts) => {
    opts = Object.assign(
      {
        langs: ['en'],
        concurrency: Infinity,
        format: 'vtt'
      },
      opts
    );

    if (opts.langs.constructor !== Array) {
      throw new TypeError('Expected `langs` to be an array');
    } else if (opts.langs.some(lang => langK.indexOf(lang) === -1)) {
      throw new TypeError(`Expected \`langs\` members to be in ${langK}`);
    }

    return scrape(imdbId).then(res =>
      (res.length > 0) ? runConditional(imdbId, opts, res) : []
    );
  };

  return yifysubtitles;
}

module.exports = yifysubtitlesWithGoogleStorage;