fix: avoid panics in ParseNode()

ParseNode(node, nil) should be able to avoid panic even if node does not contain a valid absolute base path in its <head>, since there can be usable microformats with absolute paths or no paths. ParseNode(nil, base) should return nil instead of panic, because why panic?
willnorris · Jan 31, 2023 · 3f6d1e0 · 3f6d1e0
1 parent ffbbdae
commit 3f6d1e0
Show file tree

Hide file tree

Showing 2 changed files with 32 additions and 3 deletions.
diff --git a/microformats.go b/microformats.go
@@ -108,23 +108,32 @@ type parser struct {
 }
 
 // Parse the microformats found in the HTML document read from r.  baseURL is
-// the URL this document was retrieved from and is used to resolve any
-// relative URLs.
+// the URL this document was retrieved from and is used to expand any
+// relative URLs.  If baseURL is nil and the base URL is not referenced in the
+// document, relative URLs are not expanded.
 func Parse(r io.Reader, baseURL *url.URL) *Data {
 	doc, _ := html.Parse(r)
 	return ParseNode(doc, baseURL)
 }
 
 // ParseNode parses the microformats found in doc.  baseURL is the URL this
-// document was retrieved from and is used to resolve any relative URLs.
+// document was retrieved from and is used to expand any relative URLs. If
+// baseURL is nil and the base URL is not referenced in the document,
+// relative URLs are not expanded.
 func ParseNode(doc *html.Node, baseURL *url.URL) *Data {
+	if doc == nil { // makes no sense to go further
+		return nil
+	}
 	p := new(parser)
 	p.curData = &Data{
 		Items:   make([]*Microformat, 0),
 		Rels:    make(map[string][]string),
 		RelURLs: make(map[string]*RelURL),
 	}
 	p.base = baseURL
+	if p.base == nil { // can make sense if base can be inferred from contents
+		p.base = &url.URL{}
+	}
 	p.baseFound = false
 	p.root = doc
 	p.walk(doc)

diff --git a/microformats_test.go b/microformats_test.go
@@ -561,3 +561,23 @@ func Test_GetFirstPropValue(t *testing.T) {
 		}
 	}
 }
+
+func Test_ParseNodeNil(t *testing.T) {
+	tests := []struct {
+		name string
+		html string
+	}{
+		{"absolute", "<html><head><base href=\"https://example.com\"></head></html>"},
+		{"relative", "<html><head><base href=\"./something\"></head></html>"},
+		{"none", "<html><head></head></html>"}, // parseNode(tt.html) == nil
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			n, err := parseNode(tt.html)
+			if err != nil {
+				t.Fatalf("Error parsing HTML: %v", err)
+			}
+			ParseNode(n, nil) // this should not panic
+		})
+	}
+}