goyomi/sources/base/mangataro/mangataro.go

// Package mangataro implements the MangaTaro manga base.
// WP JSON API (browse/details) + custom auth endpoints with MD5 token (chapters/pages); CF-protected.
package mangataro

import (
	"bytes"
	"context"
	"crypto/md5"
	"encoding/json"
	"fmt"
	"io"
	"net/http"
	"net/url"
	"strings"
	"time"

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

type Config struct {
	Name    string
	BaseURL string
	Lang    string
}

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

func New(cfg Config) *Source {
	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) doGet(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("mangataro: HTTP %d for %s", resp.StatusCode, rawURL)
	}
	body, _ := io.ReadAll(resp.Body)
	return json.Unmarshal(body, out)
}

func (s *Source) doPost(ctx context.Context, rawURL string, payload any, out any) error {
	body, err := json.Marshal(payload)
	if err != nil {
		return err
	}
	req, err := http.NewRequestWithContext(ctx, http.MethodPost, rawURL, bytes.NewReader(body))
	if err != nil {
		return err
	}
	req.Header.Set("Referer", s.cfg.BaseURL+"/")
	req.Header.Set("Content-Type", "application/json")
	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("mangataro: HTTP %d for %s", resp.StatusCode, rawURL)
	}
	respBody, _ := io.ReadAll(resp.Body)
	return json.Unmarshal(respBody, out)
}

// mangaURLDTO is stored as the manga.URL field (JSON-encoded).
type mangaURLDTO struct {
	ID   string `json:"id"`
	Slug string `json:"slug"`
}

func encodeMangaURL(id, slug string) string {
	b, _ := json.Marshal(mangaURLDTO{ID: id, Slug: slug})
	return string(b)
}

func decodeMangaURL(raw string) mangaURLDTO {
	var dto mangaURLDTO
	_ = json.Unmarshal([]byte(raw), &dto)
	return dto
}

// Search/browse DTOs

type searchPayload struct {
	Page          int    `json:"page"`
	Search        string `json:"search"`
	Years         string `json:"years"`
	Genres        string `json:"genres"`
	Types         string `json:"types"`
	Statuses      string `json:"statuses"`
	Sort          string `json:"sort"`
	GenreMatchMode string `json:"genreMatchMode"`
}

type browseManga struct {
	ID          string `json:"id"`
	URL         string `json:"url"` // slug
	Title       string `json:"title"`
	Cover       string `json:"cover"`
	Type        string `json:"type"`
	Description string `json:"description"`
	Status      string `json:"status"`
}

func (s *Source) browse(ctx context.Context, page int, search, sort string) (source.MangasPage, error) {
	payload := searchPayload{
		Page:          page,
		Search:        search,
		Years:         "[]",
		Genres:        "[]",
		Types:         "[]",
		Statuses:      "[]",
		Sort:          sort,
		GenreMatchMode: "and",
	}
	var items []browseManga
	if err := s.doPost(ctx, s.base()+"/wp-json/manga/v1/load", payload, &items); err != nil {
		return source.MangasPage{}, err
	}
	var mangas []source.SManga
	for _, item := range items {
		if item.Type == "Novel" || item.URL == "" {
			continue
		}
		mangas = append(mangas, source.SManga{
			URL:          encodeMangaURL(item.ID, item.URL),
			Title:        unescapeHTML(item.Title),
			ThumbnailURL: item.Cover,
			Description:  unescapeHTML(item.Description),
			Status:       parseStatus(item.Status),
		})
	}
	// hasNextPage: Kotlin checks data.size == 24
	hasNext := len(items) == 24
	return source.MangasPage{Mangas: mangas, HasNextPage: hasNext}, nil
}

func (s *Source) GetPopularManga(page int) (source.MangasPage, error) {
	return s.browse(context.Background(), page, "", "popular_desc")
}

func (s *Source) GetLatestUpdates(page int) (source.MangasPage, error) {
	return s.browse(context.Background(), page, "", "post_desc")
}

func (s *Source) GetSearchManga(page int, query string, filters []source.Filter) (source.MangasPage, error) {
	return s.browse(context.Background(), page, query, "popular_desc")
}

// Manga details DTOs (WP JSON API)

type mangaDetails struct {
	ID       int      `json:"id"`
	Slug     string   `json:"slug"`
	Title    rendered `json:"title"`
	Content  rendered `json:"content"`
	Type     string   `json:"type"`
	Embedded embedded `json:"_embedded"`
}

type rendered struct {
	Rendered string `json:"rendered"`
}

type embedded struct {
	FeaturedMedia []thumbnail    `json:"wp:featuredmedia"`
	Terms         [][]term       `json:"wp:term"`
}

func (e embedded) getTerms(taxonomy string) []string {
	for _, group := range e.Terms {
		if len(group) > 0 && group[0].Taxonomy == taxonomy {
			names := make([]string, len(group))
			for i, t := range group {
				names[i] = t.Name
			}
			return names
		}
	}
	return nil
}

type thumbnail struct {
	URL string `json:"source_url"`
}

type term struct {
	Name     string `json:"name"`
	Taxonomy string `json:"taxonomy"`
}

func (s *Source) GetMangaDetails(manga source.SManga) (source.SManga, error) {
	dto := decodeMangaURL(manga.URL)
	if dto.ID == "" {
		return manga, fmt.Errorf("mangataro: cannot decode manga URL: %s", manga.URL)
	}
	u := fmt.Sprintf("%s/wp-json/wp/v2/manga/%s?_embed", s.base(), dto.ID)
	var data mangaDetails
	if err := s.doGet(context.Background(), u, &data); err != nil {
		return manga, err
	}

	result := source.SManga{URL: manga.URL}
	result.URL = encodeMangaURL(fmt.Sprint(data.ID), data.Slug)
	result.Title = unescapeHTML(data.Title.Rendered)
	result.Description = plainText(data.Content.Rendered)

	tags := data.Embedded.getTerms("post_tag")
	genreSet := make(map[string]bool)
	for _, t := range tags {
		genreSet[t] = true
	}
	knownTypes := []string{"Manhwa", "Manhua", "Manga"}
	hasKnown := false
	for _, kt := range knownTypes {
		if genreSet[kt] {
			hasKnown = true
			break
		}
	}
	if !hasKnown && data.Type != "" {
		genreSet[data.Type] = true
	}
	var genres []string
	for g := range genreSet {
		genres = append(genres, g)
	}
	result.Genre = strings.Join(genres, ", ")
	result.Author = strings.Join(data.Embedded.getTerms("manga_author"), ", ")

	if len(data.Embedded.FeaturedMedia) > 0 {
		result.ThumbnailURL = data.Embedded.FeaturedMedia[0].URL
	}
	result.Status = manga.Status // preserved from browse
	return result, nil
}

// Chapter list DTOs

type chapterList struct {
	Chapters []chapter `json:"chapters"`
}

type chapter struct {
	URL       string  `json:"url"`
	Chapter   string  `json:"chapter"`
	Title     *string `json:"title"`
	Date      string  `json:"date"`
	GroupName *string `json:"group_name"`
	Language  string  `json:"language"`
}

func md5Token(timestamp int64) string {
	date := time.Unix(timestamp, 0).UTC().Format("2006-01-02")
	input := fmt.Sprintf("%dmng_ch_%s", timestamp, date)
	sum := md5.Sum([]byte(input))
	return fmt.Sprintf("%x", sum)[:16]
}

func (s *Source) GetChapterList(manga source.SManga) ([]source.SChapter, error) {
	dto := decodeMangaURL(manga.URL)
	if dto.ID == "" {
		return nil, fmt.Errorf("mangataro: cannot decode manga URL: %s", manga.URL)
	}

	ts := time.Now().Unix()
	token := md5Token(ts)

	u, _ := url.Parse(s.base() + "/auth/manga-chapters")
	q := u.Query()
	q.Set("manga_id", dto.ID)
	q.Set("offset", "0")
	q.Set("limit", "9999")
	q.Set("order", "DESC")
	q.Set("_t", token)
	u.RawQuery = q.Encode()

	var data chapterList
	if err := s.doGet(context.Background(), u.String(), &data); err != nil {
		return nil, err
	}

	placeholders := map[string]bool{"": true, "N/A": true, "—": true}
	var chapters []source.SChapter
	for _, ch := range data.Chapters {
		if !strings.EqualFold(ch.Language, s.cfg.Lang) {
			continue
		}
		name := "Chapter " + ch.Chapter
		if ch.Title != nil && !placeholders[*ch.Title] {
			name += ": " + unescapeHTML(*ch.Title)
		}
		chURL := ch.URL
		if !strings.HasPrefix(chURL, "http") {
			chURL = s.base() + chURL
		}
		chapters = append(chapters, source.SChapter{
			URL:        chURL,
			Name:       name,
			DateUpload: util.ParseRelativeDate(ch.Date),
		})
	}
	return chapters, nil
}

// Pages DTO

type pagesDTO struct {
	Images []string `json:"images"`
}

func (s *Source) GetPageList(chapter source.SChapter) ([]source.Page, error) {
	// chapterId = last path segment of chapter URL, after last "-"
	chapterURL := chapter.URL
	if !strings.HasPrefix(chapterURL, "http") {
		chapterURL = s.base() + chapterURL
	}
	parsed, err := url.Parse(chapterURL)
	if err != nil {
		return nil, err
	}
	segs := strings.Split(strings.TrimRight(parsed.Path, "/"), "/")
	lastSeg := segs[len(segs)-1]
	chapterID := lastSeg
	if idx := strings.LastIndex(lastSeg, "-"); idx >= 0 {
		chapterID = lastSeg[idx+1:]
	}

	u := fmt.Sprintf("%s/auth/chapter-content?chapter_id=%s", s.base(), chapterID)
	var data pagesDTO
	if err := s.doGet(context.Background(), u, &data); err != nil {
		return nil, err
	}

	pages := make([]source.Page, len(data.Images))
	for i, img := range data.Images {
		pages[i] = source.Page{Index: i, ImageURL: img}
	}
	return pages, nil
}

func parseStatus(s string) int {
	switch strings.ToLower(s) {
	case "ongoing":
		return source.StatusOngoing
	case "completed", "complete":
		return source.StatusCompleted
	case "hiatus", "on hold", "on-hold":
		return source.StatusHiatus
	case "cancelled", "canceled":
		return source.StatusCancelled
	}
	return source.StatusUnknown
}

func unescapeHTML(s string) string {
	// Basic HTML entity unescaping
	r := strings.NewReplacer(
		"&", "&",
		"&lt;", "<",
		"&gt;", ">",
		"&quot;", `"`,
		"&#039;", "'",
		"&apos;", "'",
	)
	prev := ""
	for prev != s {
		prev = s
		s = r.Replace(s)
	}
	return s
}

// plainText strips HTML tags from a string.
func plainText(html string) string {
	// Quick approximation: remove tags
	var b strings.Builder
	inTag := false
	for _, r := range html {
		switch {
		case r == '<':
			inTag = true
		case r == '>':
			inTag = false
		case !inTag:
			b.WriteRune(r)
		}
	}
	return strings.TrimSpace(b.String())
}

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