html: add "in head noscript" im support
In the spec 12.2.6.4.5, the "in head noscript" insertion mode is defined.
However, this package and its parser doesn't have the insertion mode,
because the scripting=false case is not considered currently.
This commit adds a test and a support for the "in head noscript"
insertion mode. This change has no effect on the actual behavior.
Updates golang/go#16318
Change-Id: I9314c3342bea27fa2acf2fa7d980a127ee0fbf91
Reviewed-on: https://go-review.googlesource.com/c/net/+/172557
Reviewed-by: Nigel Tao <nigeltao@golang.org>
diff --git a/html/parse.go b/html/parse.go
index 1d3c198..2aaf2ff 100644
--- a/html/parse.go
+++ b/html/parse.go
@@ -630,7 +630,16 @@
p.oe.pop()
p.acknowledgeSelfClosingTag()
return true
- case a.Script, a.Title, a.Noscript, a.Noframes, a.Style:
+ case a.Noscript:
+ p.addElement()
+ if p.scripting {
+ p.setOriginalIM()
+ p.im = textIM
+ } else {
+ p.im = inHeadNoscriptIM
+ }
+ return true
+ case a.Script, a.Title, a.Noframes, a.Style:
p.addElement()
p.setOriginalIM()
p.im = textIM
@@ -692,6 +701,49 @@
return false
}
+// 12.2.6.4.5.
+func inHeadNoscriptIM(p *parser) bool {
+ switch p.tok.Type {
+ case DoctypeToken:
+ // Ignore the token.
+ return true
+ case StartTagToken:
+ switch p.tok.DataAtom {
+ case a.Html:
+ return inBodyIM(p)
+ case a.Basefont, a.Bgsound, a.Link, a.Meta, a.Noframes, a.Style:
+ return inHeadIM(p)
+ case a.Head, a.Noscript:
+ // Ignore the token.
+ return true
+ }
+ case EndTagToken:
+ switch p.tok.DataAtom {
+ case a.Noscript, a.Br:
+ default:
+ // Ignore the token.
+ return true
+ }
+ case TextToken:
+ s := strings.TrimLeft(p.tok.Data, whitespace)
+ if len(s) == 0 {
+ // It was all whitespace.
+ return inHeadIM(p)
+ }
+ case CommentToken:
+ return inHeadIM(p)
+ }
+ p.oe.pop()
+ if p.top().DataAtom != a.Head {
+ panic("html: the new current node will be a head element.")
+ }
+ p.im = inHeadIM
+ if p.tok.DataAtom == a.Noscript {
+ return true
+ }
+ return false
+}
+
// Section 12.2.6.4.6.
func afterHeadIM(p *parser) bool {
switch p.tok.Type {
diff --git a/html/parse_test.go b/html/parse_test.go
index 9bba918..b49181c 100644
--- a/html/parse_test.go
+++ b/html/parse_test.go
@@ -228,7 +228,7 @@
t.Fatal(err)
}
- err = testParseCase(text, want, context)
+ err = testParseCase(text, want, context, Parse)
if err != nil {
t.Errorf("%s test #%d %q, %s", tf, i, text, err)
@@ -238,11 +238,45 @@
}
}
+// Issue 16318
+func TestParserWithoutScripting(t *testing.T) {
+ text := `<noscript><img src='https://golang.org/doc/gopher/frontpage.png' /></noscript><p><img src='https://golang.org/doc/gopher/doc.png' /></p>`
+ want := `| <html>
+| <head>
+| <noscript>
+| <body>
+| "<img src='https://golang.org/doc/gopher/frontpage.png' />"
+| <p>
+| <img>
+| src="https://golang.org/doc/gopher/doc.png"
+`
+ err := testParseCase(text, want, "", func(r io.Reader) (*Node, error) {
+ p := &parser{
+ tokenizer: NewTokenizer(r),
+ doc: &Node{
+ Type: DocumentNode,
+ },
+ scripting: false,
+ framesetOK: true,
+ im: initialIM,
+ }
+ err := p.parse()
+ if err != nil {
+ return nil, err
+ }
+ return p.doc, nil
+ })
+
+ if err != nil {
+ t.Errorf("test with scripting is disabled, %q, %s", text, err)
+ }
+}
+
// testParseCase tests one test case from the test files. If the test does not
// pass, it returns an error that explains the failure.
// text is the HTML to be parsed, want is a dump of the correct parse tree,
// and context is the name of the context node, if any.
-func testParseCase(text, want, context string) (err error) {
+func testParseCase(text, want, context string, parseFunc func(r io.Reader) (*Node, error)) (err error) {
defer func() {
if x := recover(); x != nil {
switch e := x.(type) {
@@ -256,7 +290,7 @@
var doc *Node
if context == "" {
- doc, err = Parse(strings.NewReader(text))
+ doc, err = parseFunc(strings.NewReader(text))
if err != nil {
return err
}
