xml: add check of version in document declaration

Check that if a version is declared, for example
in '<?xml version="XX" ?>', version must be '1.0'.

Change-Id: I16ba9f78873a5f31977dcf75ac8e671fe6c08280
Reviewed-on: https://go-review.googlesource.com/8961
Reviewed-by: Russ Cox <rsc@golang.org>

2 files changed, 24 insertions, 15 deletions

diff --git a/src/encoding/xml/xml.go b/src/encoding/xml/xml.go
index 00792c4f27..3090750c48 100644
--- a/src/encoding/xml/xml.go
+++ b/src/encoding/xml/xml.go
@@ -576,7 +576,6 @@ func (d *Decoder) rawToken() (Token, error) {
 
 	case '?':
 		// <?: Processing instruction.
-		// TODO(rsc): Should parse the <?xml declaration to make sure the version is 1.0.
 		var target string
 		if target, ok = d.name(); !ok {
 			if d.err == nil {
@@ -601,7 +600,13 @@ func (d *Decoder) rawToken() (Token, error) {
 		data = data[0 : len(data)-2] // chop ?>
 
 		if target == "xml" {
-			enc := procInstEncoding(string(data))
+			content := string(data)
+			ver := procInst("version", content)
+			if ver != "" && ver != "1.0" {
+				d.err = fmt.Errorf("xml: unsupported version %q; only version 1.0 is supported", ver)
+				return nil, d.err
+			}
+			enc := procInst("encoding", content)
 			if enc != "" && enc != "utf-8" && enc != "UTF-8" {
 				if d.CharsetReader == nil {
 					d.err = fmt.Errorf("xml: encoding %q declared but Decoder.CharsetReader is nil", enc)
@@ -1962,16 +1967,17 @@ func Escape(w io.Writer, s []byte) {
 	EscapeText(w, s)
 }
 
-// procInstEncoding parses the `encoding="..."` or `encoding='...'`
+// procInst parses the `param="..."` or `param='...'`
 // value out of the provided string, returning "" if not found.
-func procInstEncoding(s string) string {
+func procInst(param, s string) string {
 	// TODO: this parsing is somewhat lame and not exact.
 	// It works for all actual cases, though.
-	idx := strings.Index(s, "encoding=")
+	param = param + "="
+	idx := strings.Index(s, param)
 	if idx == -1 {
 		return ""
 	}
-	v := s[idx+len("encoding="):]
+	v := s[idx+len(param):]
 	if v == "" {
 		return ""
 	}
diff --git a/src/encoding/xml/xml_test.go b/src/encoding/xml/xml_test.go
index be995c0d52..312a7c98a5 100644
--- a/src/encoding/xml/xml_test.go
+++ b/src/encoding/xml/xml_test.go
@@ -657,20 +657,23 @@ type procInstEncodingTest struct {
 }
 
 var procInstTests = []struct {
-	input, expect string
+	input  string
+	expect [2]string
 }{
-	{`version="1.0" encoding="utf-8"`, "utf-8"},
-	{`version="1.0" encoding='utf-8'`, "utf-8"},
-	{`version="1.0" encoding='utf-8' `, "utf-8"},
-	{`version="1.0" encoding=utf-8`, ""},
-	{`encoding="FOO" `, "FOO"},
+	{`version="1.0" encoding="utf-8"`, [2]string{"1.0", "utf-8"}},
+	{`version="1.0" encoding='utf-8'`, [2]string{"1.0", "utf-8"}},
+	{`version="1.0" encoding='utf-8' `, [2]string{"1.0", "utf-8"}},
+	{`version="1.0" encoding=utf-8`, [2]string{"1.0", ""}},
+	{`encoding="FOO" `, [2]string{"", "FOO"}},
 }
 
 func TestProcInstEncoding(t *testing.T) {
 	for _, test := range procInstTests {
-		got := procInstEncoding(test.input)
-		if got != test.expect {
-			t.Errorf("procInstEncoding(%q) = %q; want %q", test.input, got, test.expect)
+		if got := procInst("version", test.input); got != test.expect[0] {
+			t.Errorf("procInst(version, %q) = %q; want %q", test.input, got, test.expect[0])
+		}
+		if got := procInst("encoding", test.input); got != test.expect[1] {
+			t.Errorf("procInst(encoding, %q) = %q; want %q", test.input, got, test.expect[1])
 		}
 	}
 }