html: parse <xmp> tags

Pass tests5.dat, test 10: <p><xmp></xmp> | <html> | <head> | <body> | <p> | <xmp> Also pass the remaining tests in tests5.dat. R=nigeltao CC=golang-dev https://golang.org/cl/5440062

html: parse <xmp> tags
3b392277 · Andrew Balholm · Nigel Tao · e32f4ba7 · 3b392277 · 3b392277
Commit 3b392277 authored Nov 30, 2011 by Andrew Balholm Committed by Nigel Tao Nov 30, 2011
Showing with 11 additions and 6 deletions

src/pkg/html/parse.go src/pkg/html/parse.go +5 -0

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

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

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

No files found.
--- a/src/pkg/html/parse.go
+++ b/src/pkg/html/parse.go
@@ -770,6 +770,11 @@ func inBodyIM(p *parser) bool {
 			p.oe.pop()
 			p.oe.pop()
 			p.form = nil
+		case "xmp":
+			p.popUntil(buttonScopeStopTags, "p")
+			p.reconstructActiveFormattingElements()
+			p.framesetOK = false
+			p.addElement(p.tok.Data, p.tok.Attr)
 		case "caption", "col", "colgroup", "frame", "head", "tbody", "td", "tfoot", "th", "thead", "tr":
 			// Ignore the token.
 		default:

--- a/src/pkg/html/parse_test.go
+++ b/src/pkg/html/parse_test.go
@@ -154,7 +154,7 @@ func TestParser(t *testing.T) {
 		{"tests2.dat", -1},
 		{"tests3.dat", -1},
 		// tests4.dat is fragment cases.
-		{"tests5.dat", 10},
+		{"tests5.dat", -1},
 	}
 	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 "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style":
+	case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "xmp":
 		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
@@ -406,12 +406,12 @@ func (z *Tokenizer) readStartTag() TokenType {
 		}
 	}
 	// Several tags flag 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 {
+	// The tag name lengths of these special cases ranges in [3, 9].
+	if x := z.data.end - z.data.start; 3 <= x && x <= 9 {
 		switch z.buf[z.data.start] {
-		case 'i', 'n', 'p', 's', 't', 'I', 'N', 'P', 'S', 'T':
+		case 'i', 'n', 'p', 's', 't', 'x', 'I', 'N', 'P', 'S', 'T', 'X':
 			switch s := strings.ToLower(string(z.buf[z.data.start:z.data.end])); s {
-			case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "textarea", "title":
+			case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "textarea", "title", "xmp":
 				z.rawTag = s
 			}
 		}