torrent-indexer/api/bludv.go

package handler

import (
	"context"
	"encoding/json"
	"fmt"
	"net/http"
	"net/url"
	"strings"
	"time"

	"github.com/PuerkitoBio/goquery"

	"github.com/felipemarinho97/torrent-indexer/magnet"
	"github.com/felipemarinho97/torrent-indexer/schema"
	goscrape "github.com/felipemarinho97/torrent-indexer/scrape"
	"github.com/felipemarinho97/torrent-indexer/utils"
)

var bludv = IndexerMeta{
	Label:       "bludv",
	URL:         "https://bludv.xyz/",
	SearchURL:   "?s=",
	PagePattern: "page/%s",
}

func (i *Indexer) HandlerBluDVIndexer(w http.ResponseWriter, r *http.Request) {
	start := time.Now()
	metadata := bludv

	defer func() {
		i.metrics.IndexerDuration.WithLabelValues(metadata.Label).Observe(time.Since(start).Seconds())
		i.metrics.IndexerRequests.WithLabelValues(metadata.Label).Inc()
	}()

	ctx := r.Context()
	// supported query params: q, season, episode, page, filter_results
	q := r.URL.Query().Get("q")
	page := r.URL.Query().Get("page")

	// URL encode query param
	q = url.QueryEscape(q)
	url := metadata.URL
	if page != "" {
		url = fmt.Sprintf(fmt.Sprintf("%s%s", url, metadata.PagePattern), page)
	} else {
		url = fmt.Sprintf("%s%s%s", url, metadata.SearchURL, q)
	}

	fmt.Println("URL:>", url)
	resp, err := i.requester.GetDocument(ctx, url)
	if err != nil {
		w.WriteHeader(http.StatusInternalServerError)
		err = json.NewEncoder(w).Encode(map[string]string{"error": err.Error()})
		if err != nil {
			fmt.Println(err)
		}
		i.metrics.IndexerErrors.WithLabelValues(metadata.Label).Inc()
		return
	}
	defer resp.Close()

	doc, err := goquery.NewDocumentFromReader(resp)
	if err != nil {
		w.WriteHeader(http.StatusInternalServerError)
		err = json.NewEncoder(w).Encode(map[string]string{"error": err.Error()})
		if err != nil {
			fmt.Println(err)
		}

		i.metrics.IndexerErrors.WithLabelValues(metadata.Label).Inc()
		return
	}

	var links []string
	doc.Find(".post").Each(func(i int, s *goquery.Selection) {
		// get link from h2.entry-title > a
		link, _ := s.Find("div.title > a").Attr("href")
		links = append(links, link)
	})

	// extract each torrent link
	indexedTorrents := utils.ParallelMap(links, func(link string) ([]schema.IndexedTorrent, error) {
		return getTorrentsBluDV(ctx, i, link)
	})

	// Apply post-processors
	postProcessedTorrents := indexedTorrents
	for _, processor := range i.postProcessors {
		postProcessedTorrents = processor(i, r, postProcessedTorrents)
	}

	w.Header().Set("Content-Type", "application/json")
	err = json.NewEncoder(w).Encode(Response{
		Results: postProcessedTorrents,
		Count:   len(postProcessedTorrents),
	})
	if err != nil {
		fmt.Println(err)
	}
}

func getTorrentsBluDV(ctx context.Context, i *Indexer, link string) ([]schema.IndexedTorrent, error) {
	var indexedTorrents []schema.IndexedTorrent
	doc, err := getDocument(ctx, i, link)
	if err != nil {
		return nil, err
	}

	article := doc.Find(".post")
	title := strings.Replace(article.Find(".title > h1").Text(), " - Download", "", -1)
	textContent := article.Find("div.content")
	date := getPublishedDate(doc)
	magnets := textContent.Find("a[href^=\"magnet\"]")
	var magnetLinks []string
	magnets.Each(func(i int, s *goquery.Selection) {
		magnetLink, _ := s.Attr("href")
		magnetLinks = append(magnetLinks, magnetLink)
	})

	adwareLinks := textContent.Find("a[href^=\"https://www.seuvideo.xyz\"]")
	adwareLinks.Each(func(_ int, s *goquery.Selection) {
		href, _ := s.Attr("href")
		// extract querysting "id" from url
		parsedUrl, err := url.Parse(href)
		if err != nil {
			fmt.Println(err)
			return
		}
		magnetLink := parsedUrl.Query().Get("id")
		magnetLinkDecoded, err := utils.DecodeAdLink(magnetLink)
		if err != nil {
			fmt.Printf("failed to decode ad link \"%s\": %v\n", href, err)
			return
		}

		// if decoded magnet link is indeed a magnet link, append it
		if strings.HasPrefix(magnetLinkDecoded, "magnet:") {
			magnetLinks = append(magnetLinks, magnetLinkDecoded)
		} else {
			fmt.Printf("WARN: link \"%s\" decoding resulted in non-magnet link: %s\n", href, magnetLinkDecoded)
		}
	})

	var audio []schema.Audio
	var year string
	var size []string
	article.Find("div.content p").Each(func(i int, s *goquery.Selection) {
		// pattern:
		// Título Traduzido: Fundação
		// Título Original: Foundation
		// IMDb: 7,5
		// Ano de Lançamento: 2023
		// Gênero: Ação | Aventura | Ficção
		// Formato: MKV
		// Qualidade: WEB-DL
		// Áudio: Português | Inglês
		// Idioma: Português | Inglês
		// Legenda: Português
		// Tamanho: –
		// Qualidade de Áudio: 10
		// Qualidade de Vídeo: 10
		// Duração: 59 Min.
		// Servidor: Torrent
		text := s.Text()

		audio = append(audio, findAudioFromText(text)...)
		y := findYearFromText(text, title)
		if y != "" {
			year = y
		}
		size = append(size, findSizesFromText(text)...)
	})

	// find any link from imdb
	imdbLink := ""
	article.Find("div.content a").Each(func(i int, s *goquery.Selection) {
		link, _ := s.Attr("href")
		_imdbLink, err := getIMDBLink(link)
		if err == nil {
			imdbLink = _imdbLink
		}
	})

	size = utils.StableUniq(size)

	var chanIndexedTorrent = make(chan schema.IndexedTorrent)

	// for each magnet link, create a new indexed torrent
	for it, magnetLink := range magnetLinks {
		it := it
		go func(it int, magnetLink string) {
			magnet, err := magnet.ParseMagnetUri(magnetLink)
			if err != nil {
				fmt.Println(err)
			}
			releaseTitle := magnet.DisplayName
			infoHash := magnet.InfoHash.String()
			trackers := magnet.Trackers
			magnetAudio := getAudioFromTitle(releaseTitle, audio)

			peer, seed, err := goscrape.GetLeechsAndSeeds(ctx, i.redis, i.metrics, infoHash, trackers)
			if err != nil {
				fmt.Println(err)
			}

			title := processTitle(title, magnetAudio)

			// if the number of sizes is equal to the number of magnets, then assign the size to each indexed torrent in order
			var mySize string
			if len(size) == len(magnetLinks) {
				mySize = size[it]
			}

			ixt := schema.IndexedTorrent{
				Title:         releaseTitle,
				OriginalTitle: title,
				Details:       link,
				Year:          year,
				IMDB:          imdbLink,
				Audio:         magnetAudio,
				MagnetLink:    magnetLink,
				Date:          date,
				InfoHash:      infoHash,
				Trackers:      trackers,
				LeechCount:    peer,
				SeedCount:     seed,
				Size:          mySize,
			}
			chanIndexedTorrent <- ixt
		}(it, magnetLink)
	}

	for i := 0; i < len(magnetLinks); i++ {
		it := <-chanIndexedTorrent
		indexedTorrents = append(indexedTorrents, it)
	}

	return indexedTorrents, nil
}

func getPublishedDate(document *goquery.Document) time.Time {
	var date time.Time
	//<meta property="article:published_time" content="2019-08-23T13:20:57+00:00">
	datePublished := strings.TrimSpace(document.Find("meta[property=\"article:published_time\"]").AttrOr("content", ""))

	if datePublished != "" {
		date, _ = time.Parse(time.RFC3339, datePublished)
	}

	return date
}