fix: Google updated DOM of search result page

google/search.go

@@ -157,8 +157,8 @@ func (gogl *Google) Search(query core.Query) ([]core.SearchResult, error) {
 		gogl.acceptCookies(page)
 	}
 
-	// Find all results
-	results, err := page.Timeout(gogl.Timeout).Search("div[data-hveid]")
+	// Find all results using stable attributes
+	results, err := page.Timeout(gogl.Timeout).Search("div[data-hveid][data-ved]")
 	if err != nil {
 		logrus.Errorf("Cannot parse search results: %s", err)
 		return nil, core.ErrSearchTimeout
@@ -265,35 +265,51 @@ func (gogl *Google) Search(query core.Query) ([]core.SearchResult, error) {
 				searchResults = append(searchResults, srchRes)
 			}
 			continue
-		} else if strings.Contains(attrs, "data-ved") && strings.Contains(attrs, "lang") {
-			// 3. Parse regular search results
-			// Get URL
-			link, err := resEl.Element("a")
+		} else if strings.Contains(attrs, "data-ved") {
+			// Parse regular search results
+			// Get title from h3
+			titleTag, err := resEl.Element("h3")
 			if err != nil {
 				continue
 			}
-			href, err := link.Property("href")
-			if err != nil {
-				logrus.Debug("No `href` tag found")
+			srchRes.Title, _ = titleTag.Text()
+
+			// Get URL from parent link of h3
+			link, err := titleTag.Parent()
+			if err == nil && link.MustMatches("a") {
+				href, _ := link.Property("href")
+				srchRes.URL = href.String()
 			}
-			srchRes.URL = href.String()
+
+			// Get description using multiple fallback strategies
+			desc := ""
+			if descTag, err := resEl.Element("div[data-sncf='1'] div"); err == nil {
+				desc = descTag.MustText()
+			} else if descTag, err := resEl.Element("div.VwiC3b"); err == nil {
+				desc = descTag.MustText()
+			} else {
+				// Structural fallback
+				parent, err := titleTag.Parent()
+				if err == nil {
+					parent, err = parent.Parent()
+					if err == nil {
+						parent, err = parent.Parent()
+						if err == nil {
+							if descTag, err := parent.Next(); err == nil {
+								if descDiv, err := descTag.Element("div"); err == nil {
+									desc = descDiv.MustText()
+								}
+							}
+						}
+					}
+				}
+			}
+			srchRes.Description = desc
+
 			rank += 1
-
-			// Get title
-			titleTag, err := link.Element("h3")
-			if err != nil {
-				continue
-			}
-
-			srchRes.Title, err = titleTag.Text()
-			if err != nil {
-				logrus.Debug("Cannot extract text from title")
-			}
-
-			// Get description
-			text := resEl.MustText()
-			textSliced := strings.Split(text, "\n")
-			srchRes.Description = strings.Join(textSliced[:], "\n")
+			srchRes.Rank = rank
+			searchResults = append(searchResults, srchRes)
+			continue
 
 		} else {
 			//fmt.Println(i, attrs)
@@ -324,7 +340,7 @@ func (gogl *Google) SearchImage(query core.Query) ([]core.SearchResult, error) {
 		page.Mouse.Scroll(0, 1000000, 1)
 		page.WaitLoad()
 
-		results, err := page.Timeout(gogl.Timeout).Search("div[data-hveid][data-ved][jsaction][jsdata]")
+		results, err := page.Timeout(gogl.Timeout).Search("div[data-hveid][data-ved][jsaction]")
 		if err != nil {
 			logrus.Errorf("Cannot parse search results: %s", err)
 			return *core.ConvertSearchResultsMap(searchResultsMap), core.ErrSearchTimeout

google/search_raw.go

@@ -33,6 +33,7 @@ func googleRequest(searchURL string, query core.Query) (*http.Response, error) {
 		Transport: transport,
 		Timeout:   time.Second * 10,
 	}
+
 	req, err := http.NewRequest("GET", searchURL, nil)
 	if err != nil {
 		return nil, err
@@ -55,23 +56,49 @@ func googleResultParser(response *http.Response) ([]core.SearchResult, error) {
 	results := []core.SearchResult{}
 	rank := 1
 
-	// Get individual results
-	sel := doc.Find("div.g")
+	// Use data attributes instead of class names to find results
+	// Both old and new DOM have data-hveid and data-ved attributes
+	sel := doc.Find("div[data-hveid][data-ved]")
 
 	for i := range sel.Nodes {
 		item := sel.Eq(i)
 
-		// Find URL
-		linkTag := item.Find("a")
-		link, _ := linkTag.Attr("href")
+		// Skip items without an h3 element (which indicates a search result)
+		if item.Find("h3").Length() == 0 {
+			continue
+		}
+
+		// Find URL - look for the anchor that contains the h3 title
+		linkTag := item.Find("h3").Parent()
+		if linkTag.Is("a") == false {
+			linkTag = item.Find("h3").Closest("a")
+		}
+
+		link, exists := linkTag.Attr("href")
+		if !exists || link == "" || link == "#" {
+			continue
+		}
 		link = strings.Trim(link, " ")
 
-		// Find title
+		// Find title - this is inside the h3 element
 		titleTag := item.Find("h3")
 		title := titleTag.Text()
 
-		// Find description
-		descTag := item.Find(`div[data-sncf~="1"]`)
+		// Find description - find div with text content after the heading
+		// Using attribute selectors that match the description container
+		descTag := item.Find("div[data-sncf='1']").Find("div").First()
+		if descTag.Length() == 0 {
+			// Try another selector approach if the first one fails
+			descTag = item.Find("div.VwiC3b")
+			if descTag.Length() == 0 {
+				// As a last resort, look for any div after the title that might contain description
+				titleParent := titleTag.Parent()
+				if titleParent.Is("a") {
+					titleParent = titleParent.Parent().Parent()
+				}
+				descTag = titleParent.NextAll().First().Find("div").First()
+			}
+		}
 		desc := descTag.Text()
 
 		if link != "" && link != "#" {