encoding/xml: correctly escape newline, carriage return, and tab

The generated encodings are those from
http://www.w3.org/TR/2000/WD-xml-c14n-20000119.html#charescaping

The change to the decoder ensures that we turn 
 in the
input into \r, not \n.

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6747043

diff --git a/src/pkg/encoding/xml/marshal_test.go b/src/pkg/encoding/xml/marshal_test.go
index e729a247af40d5110068f519c10f69c328e1b371..668fea6f85e5e5877aa270755e68858cf376ca7b 100644 (file)
--- a/src/pkg/encoding/xml/marshal_test.go
+++ b/src/pkg/encoding/xml/marshal_test.go
@@ -687,6 +687,27 @@ var marshalTests = []struct {
                Value:         &IgnoreTest{},
                UnmarshalOnly: true,
        },
+
+       // Test escaping.
+       {
+               ExpectXML: `<a><nested><value>dquote: &#34;; squote: &#39;; ampersand: &amp;; less: &lt;; greater: &gt;;</value></nested></a>`,
+               Value: &AnyTest{
+                       Nested: `dquote: "; squote: '; ampersand: &; less: <; greater: >;`,
+               },
+       },
+       {
+               ExpectXML: `<a><nested><value>newline: &#xA;; cr: &#xD;; tab: &#x9;;</value></nested></a>`,
+               Value: &AnyTest{
+                       Nested: "newline: \n; cr: \r; tab: \t;",
+               },
+       },
+       {
+               ExpectXML: "<a><nested><value>1\r2\r\n3\n\r4\n5</value></nested></a>",
+               Value: &AnyTest{
+                       Nested: "1\n2\n3\n\n4\n5",
+               },
+               UnmarshalOnly: true,
+       },
 }
 
 func TestMarshal(t *testing.T) {

diff --git a/src/pkg/encoding/xml/xml.go b/src/pkg/encoding/xml/xml.go
index fbd2208e33415da948121fbce0ef842e990562af..ab853c61a430d89019726d70a979ebecb955ce9a 100644 (file)
--- a/src/pkg/encoding/xml/xml.go
+++ b/src/pkg/encoding/xml/xml.go
@@ -964,7 +964,16 @@ Input:
                        b0, b1 = 0, 0
                        continue Input
                }
-               d.buf.WriteByte(b)
+
+               // We must rewrite unescaped \r and \r\n into \n.
+               if b == '\r' {
+                       d.buf.WriteByte('\n')
+               } else if b1 == '\r' && b == '\n' {
+                       // Skip \r\n--we already wrote \n.
+               } else {
+                       d.buf.WriteByte(b)
+               }
+
                b0, b1 = b1, b
        }
        data := d.buf.Bytes()
@@ -985,20 +994,7 @@ Input:
                }
        }
 
-       // Must rewrite \r and \r\n into \n.
-       w := 0
-       for r := 0; r < len(data); r++ {
-               b := data[r]
-               if b == '\r' {
-                       if r+1 < len(data) && data[r+1] == '\n' {
-                               continue
-                       }
-                       b = '\n'
-               }
-               data[w] = b
-               w++
-       }
-       return data[0:w]
+       return data
 }
 
 // Decide whether the given rune is in the XML Character Range, per
@@ -1689,6 +1685,9 @@ var (
        esc_amp  = []byte("&amp;")
        esc_lt   = []byte("&lt;")
        esc_gt   = []byte("&gt;")
+       esc_tab  = []byte("&#x9;")
+       esc_nl   = []byte("&#xA;")
+       esc_cr   = []byte("&#xD;")
 )
 
 // Escape writes to w the properly escaped XML equivalent
@@ -1708,6 +1707,12 @@ func Escape(w io.Writer, s []byte) {
                        esc = esc_lt
                case '>':
                        esc = esc_gt
+               case '\t':
+                       esc = esc_tab
+               case '\n':
+                       esc = esc_nl
+               case '\r':
+                       esc = esc_cr
                default:
                        continue
                }