html: parse the contents of <iframe> elements as raw text

Pass tests5.dat, test 4: <iframe>  " | "x" Also pass tests through test 9: <style> <!</-- </style>x R=nigeltao CC=golang-dev https://golang.org/cl/5450044

html: parse the contents of <iframe> elements as raw text
e32f4ba7 · Andrew Balholm · Nigel Tao · 72a2979e · e32f4ba7 · e32f4ba7
Commit e32f4ba7 authored Nov 30, 2011 by Andrew Balholm Committed by Nigel Tao Nov 30, 2011
Hide whitespace changes
Inline Side-by-side

Showing with 6 additions and 5 deletions

parse_test.go src/pkg/html/parse_test.go +2 -0

render.go src/pkg/html/render.go +1 -1

token.go src/pkg/html/token.go +3 -4

No files found.
--- a/src/pkg/html/parse_test.go
+++ b/src/pkg/html/parse_test.go
@@ -153,6 +153,8 @@ func TestParser(t *testing.T) {
 		{"tests1.dat", -1},
 		{"tests2.dat", -1},
 		{"tests3.dat", -1},
+		// tests4.dat is fragment cases.
+		{"tests5.dat", 10},
 	}
 	for _, tf := range testFiles {
 		f, err := os.Open("testdata/webkit/" + tf.filename)

--- a/src/pkg/html/render.go
+++ b/src/pkg/html/render.go
@@ -185,7 +185,7 @@ func render1(w writer, n *Node) error {
 	// Render any child nodes.
 	switch n.Data {
-	case "noembed", "noframes", "noscript", "plaintext", "script", "style":
+	case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style":
 		for _, c := range n.Child {
 			if c.Type != TextNode {
 				return fmt.Errorf("html: raw text element <%s> has non-text child node", n.Data)

--- a/src/pkg/html/token.go
+++ b/src/pkg/html/token.go
@@ -405,14 +405,13 @@ func (z *Tokenizer) readStartTag() TokenType {
 			break
 		}
 	}
-	// Any "<noembed>", "<noframes>", "<noscript>", "<plaintext", "<script>", "<style>",
+	// Several tags flag the tokenizer's next token as raw.
-	// "<textarea>" or "<title>" tag flags the tokenizer's next token as raw.
 	// The tag name lengths of these special cases ranges in [5, 9].
 	if x := z.data.end - z.data.start; 5 <= x && x <= 9 {
 		switch z.buf[z.data.start] {
-		case 'n', 'p', 's', 't', 'N', 'P', 'S', 'T':
+		case 'i', 'n', 'p', 's', 't', 'I', 'N', 'P', 'S', 'T':
 			switch s := strings.ToLower(string(z.buf[z.data.start:z.data.end])); s {
-			case "noembed", "noframes", "noscript", "plaintext", "script", "style", "textarea", "title":
+			case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "textarea", "title":
 				z.rawTag = s
 			}
 		}