docs: add comments explaining the logic

1 files changed, 11 insertions, 0 deletions

diff --git a/internal/findlinks/findlinks.go b/internal/findlinks/findlinks.go
index a4f29d7..a281a49 100644
--- a/internal/findlinks/findlinks.go
+++ b/internal/findlinks/findlinks.go
@@ -29,7 +29,16 @@ func FindLinks(r io.Reader) ([]Link, error) {
 	return links, nil
 }
 
+// iterHTML recursively scans the HTML tree n for link data.
 func iterHTML(n *html.Node, buffer []Link) []Link {
+	// Return if n doesn't contain the right kind of data, since
+	// we could potentially iterate twice over things like text
+	// nodes when calling extractText.
+	if n.Type != html.ElementNode && n.Type != html.DocumentNode {
+		return buffer
+	}
+
+	// If we've hit a link, go for it.
 	if n.Type == html.ElementNode && n.DataAtom == atom.A {
 		var link Link
 
@@ -42,6 +51,8 @@ func iterHTML(n *html.Node, buffer []Link) []Link {
 
 		buffer = append(buffer, link)
 	} else {
+		// If not a link, just dive down the tree looking for
+		// more links.
 		for c := n.FirstChild; c != nil; c = c.NextSibling {
 			buffer = iterHTML(c, buffer)
 		}