moves scraper to a subpackage

scraper/scrapers/kickass/kickass_api.js

@@ -0,0 +1,187 @@
+const cheerio = require('cheerio');
+const needle = require('needle');
+const moment = require('moment');
+const decode = require('magnet-uri');
+
+const defaultProxies = [
+  'https://katcr.co'
+];
+const defaultTimeout = 10000;
+
+const Categories = {
+  MOVIE: 'movies',
+  TV: 'tv',
+  ANIME: 'anime',
+  APPS: 'applications',
+  GAMES: 'games',
+  MUSIC: 'music',
+  BOOKS: 'books',
+  PORN: 'xxx',
+  OTHER: 'other',
+};
+
+function torrent(torrentId, config = {}, retries = 2) {
+  if (!torrentId || retries === 0) {
+    return Promise.reject(new Error(`Failed ${torrentId} search`));
+  }
+  const proxyList = config.proxyList || defaultProxies;
+
+  return raceFirstSuccessful(proxyList
+      .map((proxyUrl) => singleRequest(`${proxyUrl}/torrent/${torrentId}`, config)))
+      .then((body) => parseTorrentPage(body))
+      .then((torrent) => ({ torrentId, ...torrent }))
+      .catch((err) => torrent(torrentId, config, retries - 1));
+}
+
+function search(keyword, config = {}, retries = 2) {
+  if (!keyword || retries === 0) {
+    return Promise.reject(new Error(`Failed ${keyword} search`));
+  }
+  const proxyList = config.proxyList || defaultProxies;
+  const page = config.page || 1;
+  const category = config.category;
+
+  return raceFirstSuccessful(proxyList
+      .map((proxyUrl) => singleRequest(`${proxyUrl}/search/${keyword}/${page}/99/${category}`, config)))
+      .then((body) => parseTableBody(body))
+      .catch((err) => search(keyword, config, retries - 1));
+}
+
+function browse(config = {}, retries = 2) {
+  if (retries === 0) {
+    return Promise.reject(new Error(`Failed browse request`));
+  }
+  const proxyList = config.proxyList || defaultProxies;
+  const page = config.page || 1;
+  const category = config.category;
+
+  return raceFirstSuccessful(proxyList
+      .map((proxyUrl) => singleRequest(`${proxyUrl}/category/${category}/page/${page}`, config)))
+      .then((body) => parseTableBody(body))
+      .catch((err) => browse(config, retries - 1));
+}
+
+function singleRequest(requestUrl, config = {}) {
+  const timeout = config.timeout || defaultTimeout;
+
+  return needle('get', requestUrl, { open_timeout: timeout, follow: 2 })
+      .then((response) => {
+        const body = response.body;
+        if (!body) {
+          throw new Error(`No body: ${requestUrl}`);
+        } else if (body.includes('Access Denied')) {
+          console.log(`Access Denied: ${requestUrl}`);
+          throw new Error(`Access Denied: ${requestUrl}`);
+        } else if (body.includes('502: Bad gateway') ||
+            body.includes('403 Forbidden') ||
+            body.includes('Origin DNS error') ||
+            !body.includes('Kickass Torrents</title>')) {
+          throw new Error(`Invalid body contents: ${requestUrl}`);
+        }
+        return body;
+      });
+}
+
+function parseTableBody(body) {
+  return new Promise((resolve, reject) => {
+    const $ = cheerio.load(body);
+
+    if (!$) {
+      reject(new Error('Failed loading body'));
+    }
+
+    const torrents = [];
+
+    $('.table > tbody > tr').each((i, element) => {
+      const row = $(element);
+      torrents.push({
+        torrentId: row.find('a[class="torrents_table__torrent_title"]').first().attr('href').replace('/torrent/', ''),
+        name: row.find('a[class="torrents_table__torrent_title"]').first().children('b').text(),
+        category: row.find('span[class="torrents_table__upload_info"]').first().children('a').first().attr('href')
+            .match(/category\/([^\/]+)/)[1],
+        seeders: parseInt(row.find('td[data-title="Seed"]').first().text()),
+        leechers: parseInt(row.find('td[data-title="Leech"]').first().text()),
+        size: parseSize(row.find('td[data-title="Size"]').first().text()),
+        uploadDate: moment(row.find('td[data-title="Age"]').first().attr('title')).toDate()
+      });
+    });
+
+    resolve(torrents);
+  });
+}
+
+function parseTorrentPage(body) {
+  return new Promise((resolve, reject) => {
+    const $ = cheerio.load(body);
+
+    if (!$) {
+      reject(new Error('Failed loading body'));
+    }
+    const content = $('div[class="col"]').first();
+    const info = content.find('div[class="torrent_stats"]').parent();
+    const description = content.find('div[id="main"]');
+    const magnetLink = info.find('a[title="Download verified Magnet"]').attr('href');
+    const imdbIdMatch = description.html().match(/imdb\.com\/title\/tt(\d+)/i);
+
+    const torrent = {
+      name: info.find('h1').first().text(),
+      infoHash: decode(magnetLink).infoHash,
+      magnetLink: magnetLink,
+      seeders: parseInt(info.find('span[class="torrent_stats__seed_count mr-2"]').first().text().match(/\d+/)[0], 10),
+      leechers: parseInt(info.find('span[class="torrent_stats__leech_count mr-2"]').first().text().match(/\d+/)[0], 10),
+      category: info.find('small').first().children('a').first().attr('href').match(/\/category\/([^\/]+)/)[1],
+      language: description.find('span:contains(\'Audio\')').next().children().eq(0).text(),
+      size: parseSize(description.find('ul[class="file_list"]').first().find('li').first().contents().eq(2).text()
+          .match(/\(Size: (.+)\)/)[1]),
+      uploadDate: moment(info.find('time').first().text()).toDate(),
+      imdbId: imdbIdMatch && `tt${imdbIdMatch[1].padStart(7, '0')}`,
+      files: content.find('ul[class="file_list"]').first().find('li > ul > li[class="file_list__file"]')
+          .map((i, elem) => $(elem))
+          .map((i, ele) => ({
+            fileIndex: i,
+            name: ele.find('span > ul > li').contents().eq(1).text().trim().replace(/^.+\//g, ''),
+            path: ele.find('span > ul > li').contents().eq(1).text().trim(),
+            size: parseSize(ele.contents().eq(2).text())
+          })).get()
+    };
+    if (torrent.files.length >= 50) {
+      // a max of 50 files are displayed on the page
+      delete torrent.files;
+    }
+    resolve(torrent);
+  });
+}
+
+function parseSize(sizeText) {
+  if (!sizeText) {
+    return undefined;
+  }
+  let scale = 1;
+  if (sizeText.includes('GB')) {
+    scale = 1024 * 1024 * 1024
+  } else if (sizeText.includes('MB')) {
+    scale = 1024 * 1024;
+  } else if (sizeText.includes('KB') || sizeText.includes('kB')) {
+    scale = 1024;
+  }
+  return Math.floor(parseFloat(sizeText.replace(/[',]/g, '')) * scale);
+}
+
+function raceFirstSuccessful(promises) {
+  return Promise.all(promises.map((p) => {
+    // If a request fails, count that as a resolution so it will keep
+    // waiting for other possible successes. If a request succeeds,
+    // treat it as a rejection so Promise.all immediately bails out.
+    return p.then(
+        (val) => Promise.reject(val),
+        (err) => Promise.resolve(err)
+    );
+  })).then(
+      // If '.all' resolved, we've just got an array of errors.
+      (errors) => Promise.reject(errors),
+      // If '.all' rejected, we've got the result we wanted.
+      (val) => Promise.resolve(val)
+  );
+}
+
+module.exports = { torrent, search, browse, Categories };

scraper/scrapers/kickass/kickass_dump_scraper.js

@@ -0,0 +1,98 @@
+const moment = require('moment');
+const Bottleneck = require('bottleneck');
+const LineByLineReader = require('line-by-line');
+const fs = require('fs');
+const { Type } = require('../../lib/types');
+const { createTorrentEntry, createSkipTorrentEntry, getStoredTorrentEntry } = require('../../lib/torrentEntries');
+
+const NAME = 'KickassTorrents';
+const CSV_FILE_PATH = '/tmp/kickass.csv';
+
+const limiter = new Bottleneck({ maxConcurrent: 40 });
+
+async function scrape() {
+  console.log(`starting to scrape KAT dump: ${JSON.stringify(lastDump)}`);
+
+  let entriesProcessed = 0;
+  const lr = new LineByLineReader(CSV_FILE_PATH);
+  lr.on('line', (line) => {
+    if (entriesProcessed % 1000 === 0) {
+      console.log(`Processed ${entriesProcessed} entries`);
+    }
+    const row = line.match(/(?<=^|\|)(".*"|[^|]+)(?=\||$)/g);
+    if (row.length !== 11) {
+      console.log(`Invalid row: ${line}`);
+      return;
+    }
+    const torrent = {
+      infoHash: row[0].toLowerCase(),
+      title: row[1]
+          .replace(/^"|"$/g, '')
+          .replace(/&amp;/g, '&')
+          .replace(/&\w{2,6};/g, ' ')
+          .replace(/\s+/g, ' ')
+          .trim(),
+      category: row[2],
+      size: parseInt(row[5], 10),
+      seeders: parseInt(row[8], 10),
+      uploadDate: moment.unix(parseInt(row[10], 10)).toDate(),
+    };
+
+    if (!limiter.empty()) {
+      lr.pause()
+    }
+
+    limiter.schedule(() => processTorrentRecord(torrent)
+        .catch((error) => console.log(`failed ${torrent.title} due: ${error}`)))
+        .then(() => limiter.empty())
+        .then((empty) => empty && lr.resume())
+        .then(() => entriesProcessed++);
+  });
+  lr.on('error', (err) => {
+    console.log(err);
+  });
+  lr.on('end', () => {
+    fs.unlink(CSV_FILE_PATH);
+    console.log(`finished to scrape KAT dump: ${JSON.stringify(lastDump)}!`);
+  });
+}
+
+const categoryMapping = {
+  "Movies": Type.MOVIE,
+  "TV": Type.SERIES,
+  "Anime": Type.ANIME
+};
+
+async function processTorrentRecord(record) {
+  if (!categoryMapping[record.category] || record.seeders === 0) {
+    return createSkipTorrentEntry(record);
+  }
+  if (await getStoredTorrentEntry(record)) {
+    return;
+  }
+
+  const torrentFound = await findTorrent(record).catch(() => undefined);
+
+  if (!torrentFound) {
+    return createSkipTorrentEntry(record);
+  }
+
+  const torrent = {
+    infoHash: record.infoHash,
+    provider: NAME,
+    title: torrentFound.name,
+    size: record.size,
+    type: categoryMapping[record.category],
+    imdbId: torrentFound.imdbId,
+    uploadDate: record.uploadDate,
+    seeders: torrentFound.seeders,
+  };
+
+  return createTorrentEntry(torrent);
+}
+
+async function findTorrent(record) {
+  return Promise.reject("not found");
+}
+
+module.exports = { scrape };

scraper/scrapers/kickass/kickass_scraper.js

@@ -0,0 +1,87 @@
+const moment = require('moment');
+const Bottleneck = require('bottleneck');
+const kickass = require('./kickass_api');
+const { Type } = require('../../lib/types');
+const repository = require('../../lib/repository');
+const {
+  createTorrentEntry,
+  createSkipTorrentEntry,
+  getStoredTorrentEntry,
+  updateTorrentSeeders
+} = require('../../lib/torrentEntries');
+
+const NAME = 'KickassTorrents';
+const UNTIL_PAGE = 1;
+const TYPE_MAPPING = typeMapping();
+
+const limiter = new Bottleneck({ maxConcurrent: 40 });
+
+async function scrape() {
+  const scrapeStart = moment();
+  const lastScrape = await repository.getProvider({ name: NAME });
+  console.log(`[${scrapeStart}] starting ${NAME} scrape...`);
+
+  const latestTorrents = await getLatestTorrents();
+  return Promise.all(latestTorrents.map(torrent => limiter.schedule(() => processTorrentRecord(torrent))))
+      .then(() => {
+        lastScrape.lastScraped = scrapeStart;
+        lastScrape.lastScrapedId = latestTorrents.length && latestTorrents[latestTorrents.length - 1].torrentId;
+        return repository.updateProvider(lastScrape);
+      })
+      .then(() => console.log(`[${moment()}] finished ${NAME} scrape`));
+}
+
+async function getLatestTorrents() {
+  const allowedCategories = [
+    kickass.Categories.MOVIE,
+    kickass.Categories.TV,
+    kickass.Categories.ANIME,
+  ];
+
+  return Promise.all(allowedCategories.map(category => getLatestTorrentsForCategory(category)))
+      .then(entries => entries.reduce((a, b) => a.concat(b), []));
+}
+
+async function getLatestTorrentsForCategory(category, page = 1) {
+  return kickass.browse(({ category, page }))
+      .then(torrents => torrents.length && page < UNTIL_PAGE
+          ? getLatestTorrents(category, page + 1).then(nextTorrents => torrents.concat(nextTorrents))
+          : torrents)
+      .catch(() => []);
+}
+
+async function processTorrentRecord(record) {
+  if (await getStoredTorrentEntry(record)) {
+    return updateTorrentSeeders(record);
+  }
+
+  const torrentFound = await kickass.torrent(record.torrentId).catch(() => undefined);
+
+  if (!torrentFound || !TYPE_MAPPING[torrentFound.category]) {
+    return createSkipTorrentEntry(record);
+  }
+
+  const torrent = {
+    infoHash: torrentFound.infoHash,
+    provider: NAME,
+    torrentId: torrentFound.torrentId,
+    title: torrentFound.name.replace(/\t|\s+/g, ' '),
+    size: torrentFound.size,
+    type: TYPE_MAPPING[torrentFound.category],
+    imdbId: torrentFound.imdbId,
+    uploadDate: torrentFound.uploadDate,
+    seeders: torrentFound.seeders,
+  };
+
+  return createTorrentEntry(torrent);
+}
+
+function typeMapping() {
+  const mapping = {};
+  mapping[kickass.Categories.MOVIE] = Type.MOVIE;
+  mapping[kickass.Categories.TV] = Type.SERIES;
+  mapping[kickass.Categories.ANIME] = Type.ANIME;
+  return mapping;
+}
+
+module.exports = { scrape };