diff options
Diffstat (limited to 'libgo/go/encoding/xml/xml.go')
| -rw-r--r-- | libgo/go/encoding/xml/xml.go | 54 |
1 files changed, 43 insertions, 11 deletions
diff --git a/libgo/go/encoding/xml/xml.go b/libgo/go/encoding/xml/xml.go index decb2be..021f7e4 100644 --- a/libgo/go/encoding/xml/xml.go +++ b/libgo/go/encoding/xml/xml.go @@ -169,6 +169,11 @@ type Decoder struct { // the CharsetReader's result values must be non-nil. CharsetReader func(charset string, input io.Reader) (io.Reader, error) + // DefaultSpace sets the default name space used for unadorned tags, + // as if the entire XML stream were wrapped in an element containing + // the attribute xmlns="DefaultSpace". + DefaultSpace string + r io.ByteReader buf bytes.Buffer saved *bytes.Buffer @@ -268,6 +273,8 @@ func (d *Decoder) Token() (t Token, err error) { return } +const xmlURL = "http://www.w3.org/XML/1998/namespace" + // Apply name space translation to name n. // The default name space (for Space=="") // applies only to element names, not to attribute names. @@ -277,11 +284,15 @@ func (d *Decoder) translate(n *Name, isElementName bool) { return case n.Space == "" && !isElementName: return + case n.Space == "xml": + n.Space = xmlURL case n.Space == "" && n.Local == "xmlns": return } if v, ok := d.ns[n.Space]; ok { n.Space = v + } else if n.Space == "" { + n.Space = d.DefaultSpace } } @@ -956,7 +967,7 @@ Input: b0, b1 = 0, 0 continue Input } - ent := string(d.buf.Bytes()[before]) + ent := string(d.buf.Bytes()[before:]) if ent[len(ent)-1] != ';' { ent += " (no semicolon)" } @@ -1692,7 +1703,7 @@ var HTMLAutoClose = htmlAutoClose var htmlAutoClose = []string{ /* hget http://www.w3.org/TR/html4/loose.dtd | - 9 sed -n 's/<!ELEMENT (.*) - O EMPTY.+/ "\1",/p' | tr A-Z a-z + 9 sed -n 's/<!ELEMENT ([^ ]*) +- O EMPTY.+/ "\1",/p' | tr A-Z a-z */ "basefont", "br", @@ -1702,7 +1713,7 @@ var htmlAutoClose = []string{ "param", "hr", "input", - "col ", + "col", "frame", "isindex", "base", @@ -1718,15 +1729,18 @@ var ( esc_tab = []byte("	") esc_nl = []byte("
") esc_cr = []byte("
") + esc_fffd = []byte("\uFFFD") // Unicode replacement character ) -// Escape writes to w the properly escaped XML equivalent +// EscapeText writes to w the properly escaped XML equivalent // of the plain text data s. -func Escape(w io.Writer, s []byte) { +func EscapeText(w io.Writer, s []byte) error { var esc []byte last := 0 - for i, c := range s { - switch c { + for i := 0; i < len(s); { + r, width := utf8.DecodeRune(s[i:]) + i += width + switch r { case '"': esc = esc_quot case '\'': @@ -1744,13 +1758,31 @@ func Escape(w io.Writer, s []byte) { case '\r': esc = esc_cr default: + if !isInCharacterRange(r) { + esc = esc_fffd + break + } continue } - w.Write(s[last:i]) - w.Write(esc) - last = i + 1 + if _, err := w.Write(s[last : i-width]); err != nil { + return err + } + if _, err := w.Write(esc); err != nil { + return err + } + last = i + } + if _, err := w.Write(s[last:]); err != nil { + return err } - w.Write(s[last:]) + return nil +} + +// Escape is like EscapeText but omits the error return value. +// It is provided for backwards compatibility with Go 1.0. +// Code targeting Go 1.1 or later should use EscapeText. +func Escape(w io.Writer, s []byte) { + EscapeText(w, s) } // procInstEncoding parses the `encoding="..."` or `encoding='...'` |