next-book/pkg/storygraph/client.go

package storygraph

import (
	"fmt"
	"iter"
	"maps"
	"os"
	"slices"
	"sort"
	"strings"

	"github.com/joho/godotenv"
	"golang.org/x/net/html"
	"golang.org/x/net/html/atom"
)

var (
	c *HTTPClient
)

type Book struct {
	ID     string
	Name   string
	Link   string
	Image  string
	Rating string
	Tags   []string
}

func init() {
	godotenv.Load(os.Getenv("DOTENV_DIR") + ".env")
	cookie := os.Getenv("COOKIE")
	if cookie == "" {
		panic("COOKIE is not set")
	}

	c = New(cookie)
}

func GetLatestBooks() (map[string]*Book, error) {
	fmt.Println("Fetching latest book recommendations...")
	links := []Book{}

	page := 0
	for {
		page++
		fmt.Println("Fetching page", page)

		resp, err := c.Get(fmt.Sprintf("https://app.thestorygraph.com/to-read/averagemarcus?page=%d", page))
		if err != nil {
			fmt.Println("Error making request:", err)
			return nil, err
		}
		defer resp.Body.Close()
		doc, err := html.Parse(resp.Body)
		if err != nil {
			return nil, err
		}

		bookFound := false
		for n := range doc.Descendants() {
			if n.Type == html.ElementNode && n.DataAtom == atom.Div {
				for _, a := range n.Attr {
					if a.Key == "data-book-id" {
						bookFound = true

						bookID := a.Val
						name := getName(n.Descendants())
						link := fmt.Sprintf("https://app.thestorygraph.com/books/%s", bookID)
						tags := getTags(n.Descendants())
						image := getImage(n.Descendants())

						if !bookContains(links, bookID) {
							links = append(links, Book{
								ID:     bookID,
								Name:   name,
								Link:   link,
								Image:  image,
								Rating: getRating(bookID),
								Tags:   tags,
							})
						}
					}
				}
			}
		}

		if !bookFound {
			break
		}
	}

	sort.Slice(links, func(i, j int) bool {
		return links[i].Rating > links[j].Rating
	})

	return map[string]*Book{
		"Fiction":     nextByTag(links, "Fiction"),
		"Non-Fiction": nextByTag(links, "Non-Fiction"),
		"Health":      nextByTag(links, "Health"),
		"Art":         nextByTag(links, "Art"),
		"Business":    nextByTag(links, "Business"),
		"Technology":  nextByTag(links, "Technology"),
		"Sci-Fi":      nextByTag(links, "Sci-Fi"),
		"Fantasy":     nextByTag(links, "Fantasy"),
		"Comics":      nextByTag(links, "Comics"),
	}, nil
}

func getRating(bookID string) string {
	resp, err := c.Get(fmt.Sprintf("https://app.thestorygraph.com/books/%s/community_reviews", bookID))
	if err != nil {
		fmt.Println("Error fetching book rating:", err)
		return "0.0"
	}
	defer resp.Body.Close()

	doc, err := html.Parse(resp.Body)
	if err != nil {
		panic(err)
	}
	for n := range doc.Descendants() {
		if n.Type == html.ElementNode && n.DataAtom == atom.Span {
			for _, a := range n.Attr {
				if a.Key == "class" && strings.Contains(a.Val, "average-star-rating") {
					for t := range n.Descendants() {
						if t.Type == html.TextNode {
							return strings.TrimSpace(t.Data)
						}
					}
				}
			}
		}
	}

	fmt.Println("Error fetching book rating: no rating found for book", bookID)
	return "0.0"
}

func getName(decs iter.Seq[*html.Node]) string {
	for n := range decs {
		if n.Type == html.ElementNode && n.DataAtom == atom.Img {
			for _, a := range n.Attr {
				if a.Key == "alt" {
					return a.Val
				}
			}
		}
	}
	return ""
}

func getImage(decs iter.Seq[*html.Node]) string {
	for n := range decs {
		if n.Type == html.ElementNode && n.DataAtom == atom.Img {
			for _, a := range n.Attr {
				if a.Key == "src" {
					return a.Val
				}
			}
		}
	}
	return ""
}

func getTags(decs iter.Seq[*html.Node]) []string {
	tags := map[string]bool{}

	for n := range decs {
		if n.Type == html.ElementNode && n.DataAtom == atom.Div {
			for _, a := range n.Attr {
				if a.Key == "class" && strings.Contains(a.Val, "book-pane-tag-section") {
					for t := range n.Descendants() {
						if t.Type == html.ElementNode && (t.DataAtom == atom.Span || t.DataAtom == atom.A) {
							for b := range t.Descendants() {
								if b.Type == html.TextNode {
									switch b.Data {
									case "fiction":
										tags["Fiction"] = true
									case "nonfiction":
										tags["Non-Fiction"] = true
									case "psychology":
										fallthrough
									case "self help":
										fallthrough
									case "health":
										tags["Health"] = true
									case "art":
										tags["Art"] = true
									case "business":
										tags["Business"] = true
									case "technology":
										fallthrough
									case "computer science":
										tags["Technology"] = true
									case "science fiction":
										tags["Sci-Fi"] = true
									case "fantasy":
										tags["Fantasy"] = true
									case "comics":
										tags["Comics"] = true
									}
								}
							}
						}
					}
				}
			}
		}
	}

	// Only include comics in the "comics" category
	if tags["Comics"] {
		tags = map[string]bool{"Comics": true}
	}

	return slices.Collect(maps.Keys(tags))
}

func bookContains(links []Book, bookID string) bool {
	for _, b := range links {
		if b.ID == bookID {
			return true
		}
	}
	return false
}

func nextByTag(links []Book, tag string) *Book {
	for _, b := range links {
		if slices.Contains(b.Tags, tag) {
			return &b
		}
	}
	return nil
}