add description & image for articles

api/app/archive.go

@@ -20,22 +20,31 @@ func downloadAssetFile(url string) ([]byte, error) {
 	return io.ReadAll(resp.Body)
 }
 
-// FetchArticleHTML returns page's title, html and error
-func FetchArticleHTML(urlToFetch string) (string, string, error) {
+type page struct {
+	Title       string
+	Description string
+	Body        string
+	ImageURL    string
+}
+
+// FetchArticleHTML returns page struct and error
+func FetchArticleHTML(urlToFetch string) (page, error) {
+	page := page{}
+
 	url, err := url.Parse(urlToFetch)
 	if err != nil {
-		return "", "", err
+		return page, err
 	}
 
 	resp, err := http.Get(url.String())
 	if err != nil {
-		return "", "", err
+		return page, err
 	}
 	defer resp.Body.Close()
 
 	doc, err := goquery.NewDocumentFromReader(resp.Body)
 	if err != nil {
-		return "", "", err
+		return page, err
 	}
 
 	doc.Find(`head link[rel="stylesheet"]`).Each(func(i int, s *goquery.Selection) {
@@ -65,12 +74,59 @@ func FetchArticleHTML(urlToFetch string) (string, string, error) {
 		}
 	})
 
+	doc.Find("img").Each(func(i int, s *goquery.Selection) {
+		src, ok := s.Attr("src")
+		if !ok || src == "" {
+			return
+		}
+
+		if strings.HasPrefix(src, "/") {
+			s.SetAttr("src", "https://"+url.Hostname()+src)
+		}
+	})
+
 	doc.Find("script").Each(func(i int, s *goquery.Selection) {
 		s.Remove()
 	})
 
-	title := doc.Find("head title").Text()
+	page.Title = doc.Find("head title").Text()
+	page.Body, _ = doc.Html()
 
-	html, _ := doc.Html()
-	return title, html, err
+	page.Description = doc.Find(`head meta[name="description"]`).AttrOr("content", "")
+	if page.Description == "" {
+		page.Description = doc.Find(`head meta[property="og:description"]`).AttrOr("content", "")
+	}
+
+	page.ImageURL = doc.Find(`head meta[property="og:image"]`).AttrOr("content", "")
+	if page.ImageURL == "" {
+		doc.Find("img").Each(func(i int, s *goquery.Selection) {
+			if page.ImageURL != "" {
+				return
+			}
+
+			src, exists := s.Attr("src")
+			if !exists {
+				return
+			}
+
+			u, err := url.Parse(src)
+			if err != nil {
+				return
+			}
+
+			allowedTypes := []string{"png", "jpg", "jpeg", "webp"}
+			isAllowed := false
+			for _, tp := range allowedTypes {
+				if strings.HasSuffix(u.Path, tp) {
+					isAllowed = true
+				}
+			}
+
+			if isAllowed {
+				page.ImageURL = src
+			}
+		})
+	}
+
+	return page, err
 }