goyomi/sources/base/mangabox/mangabox.go

// Package mangabox implements the MangaBox manga base.
// HTML scraping for lists; JSON API for chapter list; chapter pages via HTML + JS array extraction.
package mangabox

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

	"github.com/PuerkitoBio/goquery"

	"goyomi/internal/httpclient"
	"goyomi/internal/source"
	"goyomi/sources/base/util"
)

type Config struct {
	Name           string
	BaseURL        string
	Lang           string
	PopularURLPath string // default: "manga-list/hot-manga"
	LatestURLPath  string // default: "manga-list/latest-manga"
	SimpleQueryPath string // default: "search/story"
}

type Source struct {
	cfg    Config
	client *httpclient.Client
	id     int64
}

func New(cfg Config) *Source {
	if cfg.PopularURLPath == "" {
		cfg.PopularURLPath = "manga-list/hot-manga"
	}
	if cfg.LatestURLPath == "" {
		cfg.LatestURLPath = "manga-list/latest-manga"
	}
	if cfg.SimpleQueryPath == "" {
		cfg.SimpleQueryPath = "search/story"
	}
	c := httpclient.NewClient(httpclient.WithRateLimit(1, 2))
	return &Source{cfg: cfg, client: c, id: source.GenerateSourceID(cfg.Name, cfg.Lang)}
}

func (s *Source) ID() int64            { return s.id }
func (s *Source) Name() string         { return s.cfg.Name }
func (s *Source) Lang() string         { return s.cfg.Lang }
func (s *Source) SupportsLatest() bool { return true }

func (s *Source) base() string { return strings.TrimRight(s.cfg.BaseURL, "/") }

func (s *Source) get(ctx context.Context, rawURL string) (*goquery.Document, error) {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, rawURL, nil)
	if err != nil {
		return nil, err
	}
	req.Header.Set("Referer", s.cfg.BaseURL+"/")
	resp, err := s.client.Do(req)
	if err != nil {
		return nil, err
	}
	defer resp.Body.Close()
	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("mangabox: HTTP %d", resp.StatusCode)
	}
	return goquery.NewDocumentFromReader(resp.Body)
}

func (s *Source) getJSON(ctx context.Context, rawURL string, out any) error {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, rawURL, nil)
	if err != nil {
		return err
	}
	req.Header.Set("Referer", s.cfg.BaseURL+"/")
	req.Header.Set("Accept", "application/json")
	resp, err := s.client.Do(req)
	if err != nil {
		return err
	}
	defer resp.Body.Close()
	if resp.StatusCode != http.StatusOK {
		return fmt.Errorf("mangabox: HTTP %d", resp.StatusCode)
	}
	body, _ := io.ReadAll(resp.Body)
	return json.Unmarshal(body, out)
}

func mangaFromElement(el *goquery.Selection, baseURL string) source.SManga {
	m := source.SManga{}
	a := el.Find("h3 a, h2 a").First()
	if a.Length() == 0 {
		a = el.Find("a").First()
	}
	m.URL = a.AttrOr("href", "")
	m.Title = strings.TrimSpace(a.Text())
	if thumb := el.Find("img").First().AttrOr("src", ""); thumb != "" {
		m.ThumbnailURL = util.AbsURL(baseURL, thumb)
	}
	return m
}

func (s *Source) parseMangaList(doc *goquery.Document) source.MangasPage {
	var mangas []source.SManga
	sel := "div.truyen-list > div.list-truyen-item-wrap, div.comic-list > .list-comic-item-wrap"
	doc.Find(sel).Each(func(_ int, el *goquery.Selection) {
		m := mangaFromElement(el, s.cfg.BaseURL)
		if m.URL != "" && m.Title != "" {
			mangas = append(mangas, m)
		}
	})
	hasNext := doc.Find("div.group_page a:not([href]) + a:not(:contains(Last)), a.page_select + a:not(.page_last), a.page-select + a:not(.page-last)").Length() > 0
	return source.MangasPage{Mangas: mangas, HasNextPage: hasNext}
}

func (s *Source) GetPopularManga(page int) (source.MangasPage, error) {
	doc, err := s.get(context.Background(), fmt.Sprintf("%s/%s?page=%d", s.base(), s.cfg.PopularURLPath, page))
	if err != nil {
		return source.MangasPage{}, err
	}
	return s.parseMangaList(doc), nil
}

func (s *Source) GetLatestUpdates(page int) (source.MangasPage, error) {
	doc, err := s.get(context.Background(), fmt.Sprintf("%s/%s?page=%d", s.base(), s.cfg.LatestURLPath, page))
	if err != nil {
		return source.MangasPage{}, err
	}
	return s.parseMangaList(doc), nil
}

func (s *Source) GetSearchManga(page int, query string, filters []source.Filter) (source.MangasPage, error) {
	slug := normalizeSearchQuery(query)
	u := fmt.Sprintf("%s/%s/%s?page=%d", s.base(), s.cfg.SimpleQueryPath, slug, page)
	doc, err := s.get(context.Background(), u)
	if err != nil {
		return source.MangasPage{}, err
	}
	var mangas []source.SManga
	doc.Find(".panel_story_list .story_item, div.list-truyen-item-wrap, div.list-comic-item-wrap").Each(func(_ int, el *goquery.Selection) {
		m := mangaFromElement(el, s.cfg.BaseURL)
		if m.URL != "" && m.Title != "" {
			mangas = append(mangas, m)
		}
	})
	hasNext := doc.Find("a.page_select + a:not(.page_last), a.page-select + a:not(.page-last)").Length() > 0
	return source.MangasPage{Mangas: mangas, HasNextPage: hasNext}, nil
}

func (s *Source) GetMangaDetails(manga source.SManga) (source.SManga, error) {
	doc, err := s.get(context.Background(), util.AbsURL(s.cfg.BaseURL, manga.URL))
	if err != nil {
		return manga, err
	}
	result := source.SManga{URL: manga.URL}
	main := doc.Find("div.manga-info-top, div.panel-story-info").First()

	result.Title = strings.TrimSpace(main.Find("h1").Text())
	if result.Title == "" {
		result.Title = manga.Title
	}
	if thumb := doc.Find("div.manga-info-pic img, span.info-image img").First().AttrOr("src", ""); thumb != "" {
		result.ThumbnailURL = util.AbsURL(s.cfg.BaseURL, thumb)
	}
	result.Description = strings.TrimSpace(doc.Find("div#noidungm, div#panel-story-info-description, div#contentBox").First().Text())

	result.Author = strings.TrimSpace(main.Find("li:contains(author) a, td:contains(author) + td a").First().Text())

	statusText := strings.TrimSpace(main.Find("li:contains(status), td:contains(status) + td").First().Text())
	switch {
	case strings.Contains(statusText, "Ongoing"):
		result.Status = source.StatusOngoing
	case strings.Contains(statusText, "Completed"):
		result.Status = source.StatusCompleted
	default:
		result.Status = source.StatusUnknown
	}

	// Genres from Kakalot style or Nelo style
	var genres []string
	main.Find("div.manga-info-top li:contains(genres) a").Each(func(_ int, a *goquery.Selection) {
		if t := strings.TrimSpace(a.Text()); t != "" {
			genres = append(genres, t)
		}
	})
	if len(genres) == 0 {
		main.Find("td:contains(genres) + td a").Each(func(_ int, a *goquery.Selection) {
			if t := strings.TrimSpace(a.Text()); t != "" {
				genres = append(genres, t)
			}
		})
	}
	result.Genre = strings.Join(genres, ", ")

	// Alt name appended to description
	if altEl := doc.Find(".story-alternative, tr:has(.info-alternative) h2").First(); altEl.Length() > 0 {
		alt := strings.TrimSpace(altEl.Text())
		if alt != "" {
			if result.Description == "" {
				result.Description = "Alternative Name: " + alt
			} else {
				result.Description += "\n\nAlternative Name: " + alt
			}
		}
	}
	return result, nil
}

// JSON DTOs for chapter list API

type apiResponse struct {
	Data apiDataResponse `json:"data"`
}

type apiDataResponse struct {
	Chapters   []apiChapter   `json:"chapters"`
	Pagination apiPagination  `json:"pagination"`
}

type apiChapter struct {
	ChapterName string `json:"chapter_name"`
	ChapterSlug string `json:"chapter_slug"`
	ChapterNum  float32 `json:"chapter_num"`
	UpdatedAt   string  `json:"updated_at"`
}

type apiPagination struct {
	HasMore bool `json:"has_more"`
}

var dateFormats = []string{
	"2006-01-02T15:04:05.000000Z",
	"2006-01-02T15:04:05Z",
	"2006-01-02",
}

func parseChapterDate(s string) int64 {
	for _, f := range dateFormats {
		if t, err := time.Parse(f, s); err == nil {
			return t.UnixMilli()
		}
	}
	return 0
}

func (s *Source) GetChapterList(manga source.SManga) ([]source.SChapter, error) {
	// Extract slug from manga URL: last path segment
	slug := util.SlugFromURL(manga.URL)
	if slug == "" {
		// Fallback: use last non-empty path part
		parts := strings.Split(strings.TrimRight(manga.URL, "/"), "/")
		slug = parts[len(parts)-1]
	}

	offset := 0
	const limit = 500
	var chapters []source.SChapter

	for {
		u := fmt.Sprintf("%s/api/manga/%s/chapters?limit=%d&offset=%d", s.base(), slug, limit, offset)
		var apiResp apiResponse
		if err := s.getJSON(context.Background(), u, &apiResp); err != nil {
			return nil, err
		}
		for _, ch := range apiResp.Data.Chapters {
			chURL := fmt.Sprintf("%s/manga/%s/%s", s.base(), slug, ch.ChapterSlug)
			chapters = append(chapters, source.SChapter{
				URL:        chURL,
				Name:       ch.ChapterName,
				DateUpload: parseChapterDate(ch.UpdatedAt),
			})
		}
		if !apiResp.Data.Pagination.HasMore {
			break
		}
		offset += limit
	}
	return chapters, nil
}

var arrayRe = regexp.MustCompile(`(?s)(\w+)\s*=\s*\[([^\]]+)\]`)

func extractJSArray(content, name string) []string {
	re := regexp.MustCompile(`(?s)` + regexp.QuoteMeta(name) + `\s*=\s*\[([^\]]+)\]`)
	m := re.FindStringSubmatch(content)
	if len(m) < 2 {
		return nil
	}
	var result []string
	for _, part := range strings.Split(m[1], ",") {
		val := strings.TrimSpace(part)
		val = strings.Trim(val, `"'`)
		val = strings.ReplaceAll(val, `\/`, "/")
		val = strings.TrimRight(val, "/")
		if val != "" {
			result = append(result, val)
		}
	}
	return result
}

func (s *Source) GetPageList(chapter source.SChapter) ([]source.Page, error) {
	doc, err := s.get(context.Background(), util.AbsURL(s.cfg.BaseURL, chapter.URL))
	if err != nil {
		return nil, err
	}

	// Try extracting from script with cdns and chapterImages
	var scriptContent string
	doc.Find("script").Each(func(_ int, el *goquery.Selection) {
		html, _ := el.Html()
		if strings.Contains(html, "cdns") && strings.Contains(html, "chapterImages") {
			scriptContent += html + "\n"
		}
	})

	if scriptContent != "" {
		cdns := extractJSArray(scriptContent, "cdns")
		if len(cdns) == 0 {
			cdns = extractJSArray(scriptContent, "backupImage")
		}
		chapterImages := extractJSArray(scriptContent, "chapterImages")

		if len(cdns) > 0 && len(chapterImages) > 0 {
			pages := make([]source.Page, len(chapterImages))
			for i, img := range chapterImages {
				cdn := cdns[i%len(cdns)]
				var imageURL string
				if strings.HasPrefix(img, "http") {
					imageURL = img
				} else {
					imageURL = strings.TrimRight(cdn, "/") + "/" + strings.TrimLeft(img, "/")
				}
				pages[i] = source.Page{Index: i, ImageURL: imageURL}
			}
			return pages, nil
		}
	}

	// Fallback: div.container-chapter-reader > img
	var pages []source.Page
	doc.Find("div.container-chapter-reader > img").Each(func(i int, img *goquery.Selection) {
		u := img.AttrOr("src", "")
		if u != "" {
			pages = append(pages, source.Page{Index: i, ImageURL: util.AbsURL(s.cfg.BaseURL, u)})
		}
	})
	return pages, nil
}

// normalizeSearchQuery mimics the change_alias JS function from Mangakakalot.
func normalizeSearchQuery(query string) string {
	q := strings.ToLower(query)
	var b strings.Builder
	for _, r := range q {
		switch {
		case (r >= 'a' && r <= 'z') || (r >= '0' && r <= '9'):
			b.WriteRune(r)
		case r == ' ':
			b.WriteByte('_')
		}
	}
	return b.String()
}

func (s *Source) GetImageURL(page source.Page) (string, error) { return page.ImageURL, nil }
func (s *Source) GetFilterList() []source.Filter               { return nil }