Download latest torrents from onejav

onejavdl.go

package main

import (
	"context"
	"fmt"
	"io"
	"log/slog"
	"net/http"
	"os"
	"path/filepath"
	"slices"
	"strings"

	"github.com/PuerkitoBio/goquery"
)

const (
	baseURL = "https://onejav.com"
	mainTag = "FC2" // they upload uncensored videos
	maxPage = 5
	maxItem = 10
)

var additionalTags = []string{
	"Uncensored",
}

func main() {
	if err := mainE(); err != nil {
		panic(err)
	}
}

func mainE() error  {
	ctx, cancel := context.WithCancel(context.Background())
	defer cancel()

	ch := make(chan string, 10)
	go getUrls(ctx, ch)

	length := 0
	for url := range ch {
		if length > maxItem {
			break
		}
		err := downloadFile(ctx, url)
		if err != nil {
			slog.Warn("failed to download file", "err", err)
			continue
		}
	}

	return nil
}

func downloadFile(ctx context.Context, url string) error {
	fileName := filepath.Base(url)

	_, err := os.Stat(fileName)
	if err == nil || !os.IsNotExist(err) {
		return fmt.Errorf("file already exists: %s", fileName)
	}

	req, err := http.NewRequestWithContext(ctx, "GET", url, nil)
	if err != nil {
		return fmt.Errorf("failed to create request: %w", err)
	}
	resp, err := http.DefaultClient.Do(req)
	if err != nil {
		return fmt.Errorf("failed to get response: %w", err)
	}
	defer resp.Body.Close()

	file, err := os.Create(fileName)
	if err != nil {
		return fmt.Errorf("failed to create file: %w", err)
	}
	defer file.Close()

	_, err = io.Copy(file, resp.Body)
	if err != nil {
		return fmt.Errorf("failed to write file: %w", err)
	}

	return nil
}

func getUrls(ctx context.Context, ch chan string) {
	defer close(ch)

	page := 1
	for {
		req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://onejav.com/tag/FC2?page=%d", page), nil)
		if err != nil {
			slog.Error("failed to create request", "err", err)
			return
		}
		resp, err := http.DefaultClient.Do(req)
		if err != nil {
			slog.Error("failed to get response", "err", err)
			return
		}
		
		parseDocument(resp.Body, ch)
		if page > maxPage {
			break
		}
		page++
	}
}

func parseDocument(r io.ReadCloser, ch chan string) {
	defer r.Close()

	doc, err := goquery.NewDocumentFromReader(r)
	if err != nil {
		slog.Error("failed to parse document", "err", err)
		return
	}

	nodes := doc.Find("div.card.mb-3")
	nodes.Each(func(i int, s *goquery.Selection) {
		shouldDownload := len(additionalTags) == 0
		if !shouldDownload {
			s.Find("div.tags a").Each(func(i int, s *goquery.Selection) {
				tag := strings.TrimSpace(s.Text())
				if slices.Contains(additionalTags, tag) {
					shouldDownload = true
				}
			})
		}
		if !shouldDownload {
			return
		}
		link := s.Find("a.button.is-primary").AttrOr("href", "")
		ch <- baseURL + link
	})
}