nu-api/api.go

package nuapi

import (
	"context"
	"fmt"
	"os"
	"path"
	"strconv"
	"strings"
	"time"

	"git.milar.in/milarin/adverr"
	"git.milar.in/milarin/slices"
	"github.com/PuerkitoBio/goquery"
)

type Api struct {
	UserAgent string
	Cookie    string
}

func NewApi(cookie string) *Api {
	return &Api{
		Cookie:    cookie,
		UserAgent: "Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/110.0",
	}
}

func (api *Api) GetReadingList(ctx context.Context, listIndex int) (*ReadingList, error) {
	doc, err := api.GetWithCookie(ctx, fmt.Sprintf("https://www.novelupdates.com/reading-list/?list=%d", listIndex))
	if err != nil {
		return nil, ErrApiRequestFailed.Wrap(err)
	}

	listID := ReadingListID(doc.Find("#cssmenu ul li.active a").Text())

	selection := doc.Find("table tbody tr")
	entries := make([]ReadingListEntry, 0, selection.Length())
	selection.Each(func(i int, s *goquery.Selection) {
		link := s.Find("td:nth-child(2) a:first-child")
		href, ok := link.Attr("href")
		if !ok {
			return
		}

		novelID := NovelID(path.Base(href))
		novel := NovelEntry{
			NovelID: novelID,
			Name:    link.Text(),
		}

		currentChapterLink := s.Find("td:nth-child(3) a")
		currentChapter := ReadingListChapterEntry{
			NovelID: novelID,
			ID:      ChapterID(currentChapterLink.Text()),
			Link:    currentChapterLink.AttrOr("href", ""),
		}

		latestChapterLink := s.Find("td:nth-child(3) a")
		latestChapter := ReadingListChapterEntry{
			NovelID: novelID,
			ID:      ChapterID(latestChapterLink.Text()),
			Link:    latestChapterLink.AttrOr("href", ""),
		}

		entries = append(entries, ReadingListEntry{
			Novel:          novel,
			CurrentChapter: currentChapter,
			LatestChapter:  latestChapter,
		})
	})

	return &ReadingList{
		ID:      listID,
		Entries: entries,
	}, nil
}

func (api *Api) GetNovelByID(novelID NovelID) (*Novel, error) {
	doc, err := api.Get(fmt.Sprintf("https://www.novelupdates.com/series/%s/", novelID))
	if err != nil {
		return nil, ErrApiRequestFailed.Wrap(err)
	}

	title := doc.Find(".seriestitlenu").Text()
	description := strings.TrimSpace(doc.Find("#editdescription").Text())
	cover := doc.Find(".wpb_wrapper img").AttrOr("src", "")

	associatedNamesHtml, err := doc.Find("#editassociated").Html()
	if err != nil {
		return nil, ErrApiElementNotFound.Wrap(err, "#editassociated")
	}
	associatedNames := strings.Split(strings.TrimSpace(associatedNamesHtml), "<br/>")

	novelType := NovelType(doc.Find("#showtype a.genre.type").Text())
	originalLanguage := Language(strings.ToLower(strings.Trim(doc.Find("#showtype a.genre.type + span").Text(), "()")))

	genreElems := doc.Find("#seriesgenre a.genre")
	genres := make([]GenreID, 0, genreElems.Length())
	genreElems.Each(func(i int, s *goquery.Selection) {
		href, ok := s.Attr("href")
		if !ok {
			return
		}
		genres = append(genres, GenreID(path.Base(href)))
	})

	tagElems := doc.Find("#showtags a.genre")
	tags := make([]TagID, 0, genreElems.Length())
	tagElems.Each(func(i int, s *goquery.Selection) {
		href, ok := s.Attr("href")
		if !ok {
			return
		}
		tags = append(tags, TagID(path.Base(href)))
	})

	return &Novel{
		ID:               novelID,
		Name:             title,
		AssociatedNames:  associatedNames,
		Description:      description,
		Cover:            cover,
		Type:             novelType,
		OriginalLanguage: originalLanguage,
		Genres:           genres,
		Tags:             tags,
	}, nil
}

func (api *Api) GetChapterEntriesByNovelID(novelID NovelID) *Cursor[NovelChapterEntry] {
	ctx, cancelFunc := context.WithCancel(context.Background())
	out := make(chan *NovelChapterEntry, 15)

	go func() {
		defer close(out)

		doc, err := api.Get(fmt.Sprintf("https://www.novelupdates.com/series/%s/?pg=%d", novelID, 1))
		if err != nil {
			fmt.Fprintln(os.Stderr, err)
			return
		}

		pageCount, err := api.getPageCount(doc)
		if err != nil {
			fmt.Fprintln(os.Stderr, err)
			return
		}

		for pageIndex := pageCount; pageIndex > 0; pageIndex-- {
			if ctx.Err() != nil {
				break
			}

			entries, err := api.getChapterEntriesByPageIndex(ctx, novelID, pageIndex)
			if err != nil {
				fmt.Fprintln(os.Stderr, err)
				return
			}

			for _, entry := range entries {
				entry := entry
				out <- &entry
			}
		}
	}()

	return &Cursor[NovelChapterEntry]{
		ctx:        ctx,
		cancelFunc: cancelFunc,
		Chan:       out,
	}
}

func (api *Api) getPageCount(doc *goquery.Document) (int, error) {
	pagination := doc.Find(".digg_pagination")

	if pagination.Length() > 0 {
		// more than 1 page
		pageCount, err := strconv.ParseInt(pagination.Find("a:nth-last-child(2)").Text(), 10, 64)
		if err != nil {
			adverr.Println(ErrApiElementNotFound.Wrap(err, ".digg_pagination a:nth-child(5)"))
			return 0, err
		}

		return int(pageCount), nil
	} else {
		return 1, nil
	}
}

func (api *Api) getChapterEntriesByPageIndex(ctx context.Context, novelID NovelID, pageIndex int) ([]NovelChapterEntry, error) {
	doc, err := api.Get(fmt.Sprintf("https://www.novelupdates.com/series/%s/?pg=%d", novelID, pageIndex))
	if err != nil {
		return nil, ErrApiRequestFailed.Wrap(err)
	}

	entryElems := doc.Find("#myTable tbody tr")
	entries := make([]NovelChapterEntry, 0, entryElems.Length())
	entryElems.Each(func(i int, s *goquery.Selection) {
		td3 := s.Find("td:nth-child(3) a.chp-release")

		chapterID := strings.TrimSpace(td3.Text())
		groupID := path.Base(s.Find("td:nth-child(2) a").AttrOr("href", ""))
		link := "https:" + td3.AttrOr("href", "")

		date, err := time.Parse("01/02/06", strings.TrimSpace(s.Find("td:first-child").Text()))
		if err != nil {
			adverr.Println(ErrApiElementNotFound.Wrap(err, "td:first-child"))
			return
		}

		entries = append(entries, NovelChapterEntry{
			NovelID: novelID,
			ID:      chapterID,
			Link:    link,
			Date:    date,
			Group:   groupID,
		})
	})

	return slices.Reverse(entries), nil
}