DFBUGS-1214: [release-4.16] Non-linear parsing of case-insensitive content (CVE-2024-45338)

Update golang.org/x/net to 0.34.0

Signed-off-by: Niraj Yadav <niryadav@redhat.com>

vendor/golang.org/x/net/html/doc.go

@@ -78,16 +78,11 @@ example, to process each anchor node in depth-first order:
 	if err != nil {
 		// ...
 	}
-	var f func(*html.Node)
-	f = func(n *html.Node) {
+	for n := range doc.Descendants() {
 		if n.Type == html.ElementNode && n.Data == "a" {
 			// Do something with n...
 		}
-		for c := n.FirstChild; c != nil; c = c.NextSibling {
-			f(c)
-		}
 	}
-	f(doc)
 
 The relevant specifications include:
 https://html.spec.whatwg.org/multipage/syntax.html and

vendor/golang.org/x/net/html/doctype.go

@@ -87,7 +87,7 @@ func parseDoctype(s string) (n *Node, quirks bool) {
 			}
 		}
 		if lastAttr := n.Attr[len(n.Attr)-1]; lastAttr.Key == "system" &&
-			strings.ToLower(lastAttr.Val) == "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd" {
+			strings.EqualFold(lastAttr.Val, "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd") {
 			quirks = true
 		}
 	}

vendor/golang.org/x/net/html/foreign.go

@@ -40,8 +40,7 @@ func htmlIntegrationPoint(n *Node) bool {
 		if n.Data == "annotation-xml" {
 			for _, a := range n.Attr {
 				if a.Key == "encoding" {
-					val := strings.ToLower(a.Val)
-					if val == "text/html" || val == "application/xhtml+xml" {
+					if strings.EqualFold(a.Val, "text/html") || strings.EqualFold(a.Val, "application/xhtml+xml") {
 						return true
 					}
 				}

vendor/golang.org/x/net/html/iter.go

@@ -0,0 +1,56 @@
+// Copyright 2024 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+//go:build go1.23
+
+package html
+
+import "iter"
+
+// Ancestors returns an iterator over the ancestors of n, starting with n.Parent.
+//
+// Mutating a Node or its parents while iterating may have unexpected results.
+func (n *Node) Ancestors() iter.Seq[*Node] {
+	_ = n.Parent // eager nil check
+
+	return func(yield func(*Node) bool) {
+		for p := n.Parent; p != nil && yield(p); p = p.Parent {
+		}
+	}
+}
+
+// ChildNodes returns an iterator over the immediate children of n,
+// starting with n.FirstChild.
+//
+// Mutating a Node or its children while iterating may have unexpected results.
+func (n *Node) ChildNodes() iter.Seq[*Node] {
+	_ = n.FirstChild // eager nil check
+
+	return func(yield func(*Node) bool) {
+		for c := n.FirstChild; c != nil && yield(c); c = c.NextSibling {
+		}
+	}
+
+}
+
+// Descendants returns an iterator over all nodes recursively beneath
+// n, excluding n itself. Nodes are visited in depth-first preorder.
+//
+// Mutating a Node or its descendants while iterating may have unexpected results.
+func (n *Node) Descendants() iter.Seq[*Node] {
+	_ = n.FirstChild // eager nil check
+
+	return func(yield func(*Node) bool) {
+		n.descendants(yield)
+	}
+}
+
+func (n *Node) descendants(yield func(*Node) bool) bool {
+	for c := range n.ChildNodes() {
+		if !yield(c) || !c.descendants(yield) {
+			return false
+		}
+	}
+	return true
+}

vendor/golang.org/x/net/html/node.go

@@ -38,6 +38,10 @@ var scopeMarker = Node{Type: scopeMarkerNode}
 // that it looks like "a<b" rather than "a&lt;b". For element nodes, DataAtom
 // is the atom for Data, or zero if Data is not a known tag name.
 //
+// Node trees may be navigated using the link fields (Parent,
+// FirstChild, and so on) or a range loop over iterators such as
+// [Node.Descendants].
+//
 // An empty Namespace implies a "http://www.w3.org/1999/xhtml" namespace.
 // Similarly, "math" is short for "http://www.w3.org/1998/Math/MathML", and
 // "svg" is short for "http://www.w3.org/2000/svg".

vendor/golang.org/x/net/html/parse.go

@@ -840,6 +840,10 @@ func afterHeadIM(p *parser) bool {
 
 	p.parseImpliedToken(StartTagToken, a.Body, a.Body.String())
 	p.framesetOK = true
+	if p.tok.Type == ErrorToken {
+		// Stop parsing.
+		return true
+	}
 	return false
 }
 
@@ -1031,7 +1035,7 @@ func inBodyIM(p *parser) bool {
 			if p.tok.DataAtom == a.Input {
 				for _, t := range p.tok.Attr {
 					if t.Key == "type" {
-						if strings.ToLower(t.Val) == "hidden" {
+						if strings.EqualFold(t.Val, "hidden") {
 							// Skip setting framesetOK = false
 							return true
 						}
@@ -1459,7 +1463,7 @@ func inTableIM(p *parser) bool {
 			return inHeadIM(p)
 		case a.Input:
 			for _, t := range p.tok.Attr {
-				if t.Key == "type" && strings.ToLower(t.Val) == "hidden" {
+				if t.Key == "type" && strings.EqualFold(t.Val, "hidden") {
 					p.addElement()
 					p.oe.pop()
 					return true