Improved HTTP client with rate limiting and retries

Signed-off-by: Marcus Noble <github@marcusnoble.co.uk>

pkg/storygraph/client.go

@@ -4,7 +4,6 @@ import (
 	"fmt"
 	"iter"
 	"maps"
-	"net/http"
 	"os"
 	"slices"
 	"sort"
@@ -16,8 +15,7 @@ import (
 )
 
 var (
-	COOKIE string
-	c      http.Client
+	c *HTTPClient
 )
 
 type Book struct {
@@ -31,12 +29,12 @@ type Book struct {
 
 func init() {
 	godotenv.Load(os.Getenv("DOTENV_DIR") + ".env")
-	COOKIE = os.Getenv("COOKIE")
-	if COOKIE == "" {
+	cookie := os.Getenv("COOKIE")
+	if cookie == "" {
 		panic("COOKIE is not set")
 	}
 
-	c = http.Client{}
+	c = New(cookie)
 }
 
 func GetLatestBooks() (map[string]*Book, error) {
@@ -46,21 +44,12 @@ func GetLatestBooks() (map[string]*Book, error) {
 	for {
 		page++
 
-		req, err := http.NewRequest("GET", fmt.Sprintf("https://app.thestorygraph.com/to-read/averagemarcus?page=%d", page), nil)
-		if err != nil {
-			return nil, err
-		}
-		req.Header.Set("Cookie", COOKIE)
-
-		resp, err := c.Do(req)
+		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()
-		if resp.StatusCode != 200 {
-			break
-		}
-
 		doc, err := html.Parse(resp.Body)
 		if err != nil {
 			return nil, err
@@ -116,21 +105,12 @@ func GetLatestBooks() (map[string]*Book, error) {
 }
 
 func getRating(bookID string) string {
-	req, err := http.NewRequest("GET", fmt.Sprintf("https://app.thestorygraph.com/books/%s/community_reviews", bookID), nil)
+	resp, err := c.Get(fmt.Sprintf("https://app.thestorygraph.com/books/%s/community_reviews", bookID))
 	if err != nil {
-		panic(err)
-	}
-	req.Header.Set("Cookie", COOKIE)
-
-	resp, err := c.Do(req)
-	if err != nil {
-		panic(err)
+		fmt.Println("Error fetching book rating:", resp.StatusCode)
+		return "0.0"
 	}
 	defer resp.Body.Close()
-	if resp.StatusCode != 200 {
-		fmt.Println("Error fetching book rating:", resp.StatusCode)
-		return ""
-	}
 
 	doc, err := html.Parse(resp.Body)
 	if err != nil {
@@ -145,14 +125,13 @@ func getRating(bookID string) string {
 							return strings.TrimSpace(t.Data)
 						}
 					}
-
 				}
 			}
 		}
 	}
 
-	fmt.Println("Error fetching book rating: no rating found")
-	return "0"
+	fmt.Println("Error fetching book rating: no rating found for book", bookID)
+	return "0.0"
 }
 
 func getName(decs iter.Seq[*html.Node]) string {
@@ -189,7 +168,7 @@ func getTags(decs iter.Seq[*html.Node]) []string {
 			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 {
+						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 {

pkg/storygraph/http.go

@@ -0,0 +1,68 @@
+package storygraph
+
+import (
+	"context"
+	"fmt"
+	"net/http"
+	"time"
+
+	"golang.org/x/time/rate"
+)
+
+type HTTPClient struct {
+	Cookie string
+
+	client  *http.Client
+	rl      *rate.Limiter
+	ctx     context.Context
+	retries int
+}
+
+func New(cookie string) *HTTPClient {
+	return &HTTPClient{
+		Cookie: cookie,
+
+		client:  &http.Client{},
+		rl:      rate.NewLimiter(rate.Every(1*time.Second), 15),
+		ctx:     context.Background(),
+		retries: 3,
+	}
+}
+
+func (h *HTTPClient) Get(url string) (*http.Response, error) {
+	for h.retries > 0 {
+		if err := h.rl.Wait(h.ctx); err != nil {
+			fmt.Println("Error waiting for rate limiter:", err)
+			return nil, err
+		}
+
+		req, err := http.NewRequest("GET", url, nil)
+		if err != nil {
+			fmt.Println("Error creating request:", err)
+			return nil, err
+		}
+
+		req.Header.Set("Cookie", h.Cookie)
+		resp, err := h.client.Do(req)
+		if err != nil {
+			fmt.Println("Error making request:", err)
+			return nil, err
+		}
+
+		if resp.StatusCode == 429 {
+			fmt.Println("Rate limit exceeded, retrying...")
+			h.retries--
+			continue
+		}
+
+		if resp.StatusCode != 200 {
+			fmt.Println("Error fetching page:", resp.StatusCode)
+			return resp, err
+		}
+
+		return resp, nil
+	}
+
+	fmt.Println("Max retries exceeded")
+	return nil, fmt.Errorf("max retries exceeded")
+}