encoding/xml: bring API closer to other packages
Includes gofix module. The only case not covered should be
xml.Unmarshal, since it remains with a similar interface, and
would require introspecting the type of its first argument
better.
Fixes #2626.
R=golang-dev, rsc, gustavo
CC=golang-dev
https://golang.org/cl/5574053
diff --git a/src/pkg/encoding/xml/marshal.go b/src/pkg/encoding/xml/marshal.go
index 7724e93..77413cb 100644
--- a/src/pkg/encoding/xml/marshal.go
+++ b/src/pkg/encoding/xml/marshal.go
@@ -26,11 +26,7 @@
MarshalXML() ([]byte, error)
}
-type printer struct {
- *bufio.Writer
-}
-
-// Marshal writes an XML-formatted representation of v to w.
+// Marshal returns the XML encoding of v.
//
// If v implements Marshaler, then Marshal calls its MarshalXML method.
// Otherwise, Marshal uses the following procedure to create the XML.
@@ -76,7 +72,7 @@
// Age int `xml:"person>age"`
// }
//
-// xml.Marshal(w, &Result{Id: 13, FirstName: "John", LastName: "Doe", Age: 42})
+// xml.Marshal(&Result{Id: 13, FirstName: "John", LastName: "Doe", Age: 42})
//
// would be marshalled as:
//
@@ -91,13 +87,38 @@
// </result>
//
// Marshal will return an error if asked to marshal a channel, function, or map.
-func Marshal(w io.Writer, v interface{}) (err error) {
- p := &printer{bufio.NewWriter(w)}
- err = p.marshalValue(reflect.ValueOf(v), nil)
- p.Flush()
+func Marshal(v interface{}) ([]byte, error) {
+ var b bytes.Buffer
+ if err := NewEncoder(&b).Encode(v); err != nil {
+ return nil, err
+ }
+ return b.Bytes(), nil
+}
+
+// An Encoder writes XML data to an output stream.
+type Encoder struct {
+ printer
+}
+
+// NewEncoder returns a new encoder that writes to w.
+func NewEncoder(w io.Writer) *Encoder {
+ return &Encoder{printer{bufio.NewWriter(w)}}
+}
+
+// Encode writes the XML encoding of v to the stream.
+//
+// See the documentation for Marshal for details about the conversion
+// of Go values to XML.
+func (enc *Encoder) Encode(v interface{}) error {
+ err := enc.marshalValue(reflect.ValueOf(v), nil)
+ enc.Flush()
return err
}
+type printer struct {
+ *bufio.Writer
+}
+
func (p *printer) marshalValue(val reflect.Value, finfo *fieldInfo) error {
if !val.IsValid() {
return nil
diff --git a/src/pkg/encoding/xml/marshal_test.go b/src/pkg/encoding/xml/marshal_test.go
index 10871fd..3bdfa55 100644
--- a/src/pkg/encoding/xml/marshal_test.go
+++ b/src/pkg/encoding/xml/marshal_test.go
@@ -5,7 +5,6 @@
package xml
import (
- "bytes"
"reflect"
"strconv"
"strings"
@@ -619,13 +618,12 @@
if test.UnmarshalOnly {
continue
}
- buf := bytes.NewBuffer(nil)
- err := Marshal(buf, test.Value)
+ data, err := Marshal(test.Value)
if err != nil {
t.Errorf("#%d: Error: %s", idx, err)
continue
}
- if got, want := buf.String(), test.ExpectXML; got != want {
+ if got, want := string(data), test.ExpectXML; got != want {
if strings.Contains(want, "\n") {
t.Errorf("#%d: marshal(%#v):\nHAVE:\n%s\nWANT:\n%s", idx, test.Value, got, want)
} else {
@@ -666,8 +664,7 @@
func TestMarshalErrors(t *testing.T) {
for idx, test := range marshalErrorTests {
- buf := bytes.NewBuffer(nil)
- err := Marshal(buf, test.Value)
+ _, err := Marshal(test.Value)
if err == nil || err.Error() != test.Err {
t.Errorf("#%d: marshal(%#v) = [error] %v, want %v", idx, test.Value, err, test.Err)
}
@@ -691,8 +688,7 @@
vt := reflect.TypeOf(test.Value)
dest := reflect.New(vt.Elem()).Interface()
- buffer := bytes.NewBufferString(test.ExpectXML)
- err := Unmarshal(buffer, dest)
+ err := Unmarshal([]byte(test.ExpectXML), dest)
switch fix := dest.(type) {
case *Feed:
@@ -711,17 +707,14 @@
}
func BenchmarkMarshal(b *testing.B) {
- buf := bytes.NewBuffer(nil)
for i := 0; i < b.N; i++ {
- Marshal(buf, atomValue)
- buf.Truncate(0)
+ Marshal(atomValue)
}
}
func BenchmarkUnmarshal(b *testing.B) {
xml := []byte(atomXml)
for i := 0; i < b.N; i++ {
- buffer := bytes.NewBuffer(xml)
- Unmarshal(buffer, &Feed{})
+ Unmarshal(xml, &Feed{})
}
}
diff --git a/src/pkg/encoding/xml/read.go b/src/pkg/encoding/xml/read.go
index 78e0201..8b23fd5 100644
--- a/src/pkg/encoding/xml/read.go
+++ b/src/pkg/encoding/xml/read.go
@@ -7,7 +7,6 @@
import (
"bytes"
"errors"
- "io"
"reflect"
"strconv"
"strings"
@@ -20,10 +19,10 @@
// See package json for a textual representation more suitable
// to data structures.
-// Unmarshal parses an XML element from r and uses the
-// reflect library to fill in an arbitrary struct, slice, or string
-// pointed at by val. Well-formed data that does not fit
-// into val is discarded.
+// Unmarshal parses the XML-encoded data and stores the result in
+// the value pointed to by v, which must be an arbitrary struct,
+// slice, or string. Well-formed data that does not fit into v is
+// discarded.
//
// For example, given these definitions:
//
@@ -59,7 +58,7 @@
// <address>123 Main Street</address>
// </result>
//
-// via Unmarshal(r, &result) is equivalent to assigning
+// via Unmarshal(data, &result) is equivalent to assigning
//
// r = Result{
// xml.Name{Local: "result"},
@@ -157,18 +156,26 @@
// Unmarshal maps an XML element to a pointer by setting the pointer
// to a freshly allocated value and then mapping the element to that value.
//
-func Unmarshal(r io.Reader, val interface{}) error {
- v := reflect.ValueOf(val)
- if v.Kind() != reflect.Ptr {
+func Unmarshal(data []byte, v interface{}) error {
+ return NewDecoder(bytes.NewBuffer(data)).Decode(v)
+}
+
+// Decode works like xml.Unmarshal, except it reads the decoder
+// stream to find the start element.
+func (d *Decoder) Decode(v interface{}) error {
+ return d.DecodeElement(v, nil)
+}
+
+// DecodeElement works like xml.Unmarshal except that it takes
+// a pointer to the start XML element to decode into v.
+// It is useful when a client reads some raw XML tokens itself
+// but also wants to defer to Unmarshal for some elements.
+func (d *Decoder) DecodeElement(v interface{}, start *StartElement) error {
+ val := reflect.ValueOf(v)
+ if val.Kind() != reflect.Ptr {
return errors.New("non-pointer passed to Unmarshal")
}
- p := NewParser(r)
- elem := v.Elem()
- err := p.unmarshal(elem, nil)
- if err != nil {
- return err
- }
- return nil
+ return d.unmarshal(val.Elem(), start)
}
// An UnmarshalError represents an error in the unmarshalling process.
@@ -176,22 +183,8 @@
func (e UnmarshalError) Error() string { return string(e) }
-// The Parser's Unmarshal method is like xml.Unmarshal
-// except that it can be passed a pointer to the initial start element,
-// useful when a client reads some raw XML tokens itself
-// but also defers to Unmarshal for some elements.
-// Passing a nil start element indicates that Unmarshal should
-// read the token stream to find the start element.
-func (p *Parser) Unmarshal(val interface{}, start *StartElement) error {
- v := reflect.ValueOf(val)
- if v.Kind() != reflect.Ptr {
- return errors.New("non-pointer passed to Unmarshal")
- }
- return p.unmarshal(v.Elem(), start)
-}
-
// Unmarshal a single XML element into val.
-func (p *Parser) unmarshal(val reflect.Value, start *StartElement) error {
+func (p *Decoder) unmarshal(val reflect.Value, start *StartElement) error {
// Find start element if we need it.
if start == nil {
for {
@@ -484,9 +477,9 @@
// unmarshalPath walks down an XML structure looking for wanted
// paths, and calls unmarshal on them.
// The consumed result tells whether XML elements have been consumed
-// from the Parser until start's matching end element, or if it's
+// from the Decoder until start's matching end element, or if it's
// still untouched because start is uninteresting for sv's fields.
-func (p *Parser) unmarshalPath(tinfo *typeInfo, sv reflect.Value, parents []string, start *StartElement) (consumed bool, err error) {
+func (p *Decoder) unmarshalPath(tinfo *typeInfo, sv reflect.Value, parents []string, start *StartElement) (consumed bool, err error) {
recurse := false
Loop:
for i := range tinfo.fields {
@@ -550,7 +543,7 @@
// Read tokens until we find the end element.
// Token is taking care of making sure the
// end element matches the start element we saw.
-func (p *Parser) Skip() error {
+func (p *Decoder) Skip() error {
for {
tok, err := p.Token()
if err != nil {
diff --git a/src/pkg/encoding/xml/read_test.go b/src/pkg/encoding/xml/read_test.go
index ff61bd7..833eafc 100644
--- a/src/pkg/encoding/xml/read_test.go
+++ b/src/pkg/encoding/xml/read_test.go
@@ -6,7 +6,6 @@
import (
"reflect"
- "strings"
"testing"
)
@@ -14,7 +13,7 @@
func TestUnmarshalFeed(t *testing.T) {
var f Feed
- if err := Unmarshal(strings.NewReader(atomFeedString), &f); err != nil {
+ if err := Unmarshal([]byte(atomFeedString), &f); err != nil {
t.Fatalf("Unmarshal: %s", err)
}
if !reflect.DeepEqual(f, atomFeed) {
@@ -281,7 +280,7 @@
func TestUnmarshalPaths(t *testing.T) {
for _, pt := range pathTests {
v := reflect.New(reflect.TypeOf(pt).Elem()).Interface()
- if err := Unmarshal(strings.NewReader(pathTestString), v); err != nil {
+ if err := Unmarshal([]byte(pathTestString), v); err != nil {
t.Fatalf("Unmarshal: %s", err)
}
if !reflect.DeepEqual(v, pt) {
@@ -331,7 +330,7 @@
func TestUnmarshalBadPaths(t *testing.T) {
for _, tt := range badPathTests {
- err := Unmarshal(strings.NewReader(pathTestString), tt.v)
+ err := Unmarshal([]byte(pathTestString), tt.v)
if !reflect.DeepEqual(err, tt.e) {
t.Fatalf("Unmarshal with %#v didn't fail properly:\nhave %#v,\nwant %#v", tt.v, err, tt.e)
}
@@ -350,7 +349,7 @@
func TestUnmarshalWithoutNameType(t *testing.T) {
var x TestThree
- if err := Unmarshal(strings.NewReader(withoutNameTypeData), &x); err != nil {
+ if err := Unmarshal([]byte(withoutNameTypeData), &x); err != nil {
t.Fatalf("Unmarshal: %s", err)
}
if x.Attr != OK {
diff --git a/src/pkg/encoding/xml/xml.go b/src/pkg/encoding/xml/xml.go
index d001c40..5066f5c 100644
--- a/src/pkg/encoding/xml/xml.go
+++ b/src/pkg/encoding/xml/xml.go
@@ -36,7 +36,7 @@
// A Name represents an XML name (Local) annotated
// with a name space identifier (Space).
-// In tokens returned by Parser.Token, the Space identifier
+// In tokens returned by Decoder.Token, the Space identifier
// is given as a canonical URL, not the short prefix used
// in the document being parsed.
type Name struct {
@@ -124,9 +124,9 @@
return t
}
-// A Parser represents an XML parser reading a particular input stream.
+// A Decoder represents an XML parser reading a particular input stream.
// The parser assumes that its input is encoded in UTF-8.
-type Parser struct {
+type Decoder struct {
// Strict defaults to true, enforcing the requirements
// of the XML specification.
// If set to false, the parser allows input containing common
@@ -139,9 +139,9 @@
//
// Setting:
//
- // p.Strict = false;
- // p.AutoClose = HTMLAutoClose;
- // p.Entity = HTMLEntity
+ // d.Strict = false;
+ // d.AutoClose = HTMLAutoClose;
+ // d.Entity = HTMLEntity
//
// creates a parser that can handle typical HTML.
Strict bool
@@ -184,16 +184,16 @@
tmp [32]byte
}
-// NewParser creates a new XML parser reading from r.
-func NewParser(r io.Reader) *Parser {
- p := &Parser{
+// NewDecoder creates a new XML parser reading from r.
+func NewDecoder(r io.Reader) *Decoder {
+ d := &Decoder{
ns: make(map[string]string),
nextByte: -1,
line: 1,
Strict: true,
}
- p.switchToReader(r)
- return p
+ d.switchToReader(r)
+ return d
}
// Token returns the next XML token in the input stream.
@@ -218,17 +218,17 @@
// set to the URL identifying its name space when known.
// If Token encounters an unrecognized name space prefix,
// it uses the prefix as the Space rather than report an error.
-func (p *Parser) Token() (t Token, err error) {
- if p.nextToken != nil {
- t = p.nextToken
- p.nextToken = nil
- } else if t, err = p.RawToken(); err != nil {
+func (d *Decoder) Token() (t Token, err error) {
+ if d.nextToken != nil {
+ t = d.nextToken
+ d.nextToken = nil
+ } else if t, err = d.RawToken(); err != nil {
return
}
- if !p.Strict {
- if t1, ok := p.autoClose(t); ok {
- p.nextToken = t
+ if !d.Strict {
+ if t1, ok := d.autoClose(t); ok {
+ d.nextToken = t
t = t1
}
}
@@ -240,29 +240,29 @@
// the translations first.
for _, a := range t1.Attr {
if a.Name.Space == "xmlns" {
- v, ok := p.ns[a.Name.Local]
- p.pushNs(a.Name.Local, v, ok)
- p.ns[a.Name.Local] = a.Value
+ v, ok := d.ns[a.Name.Local]
+ d.pushNs(a.Name.Local, v, ok)
+ d.ns[a.Name.Local] = a.Value
}
if a.Name.Space == "" && a.Name.Local == "xmlns" {
// Default space for untagged names
- v, ok := p.ns[""]
- p.pushNs("", v, ok)
- p.ns[""] = a.Value
+ v, ok := d.ns[""]
+ d.pushNs("", v, ok)
+ d.ns[""] = a.Value
}
}
- p.translate(&t1.Name, true)
+ d.translate(&t1.Name, true)
for i := range t1.Attr {
- p.translate(&t1.Attr[i].Name, false)
+ d.translate(&t1.Attr[i].Name, false)
}
- p.pushElement(t1.Name)
+ d.pushElement(t1.Name)
t = t1
case EndElement:
- p.translate(&t1.Name, true)
- if !p.popElement(&t1) {
- return nil, p.err
+ d.translate(&t1.Name, true)
+ if !d.popElement(&t1) {
+ return nil, d.err
}
t = t1
}
@@ -272,7 +272,7 @@
// Apply name space translation to name n.
// The default name space (for Space=="")
// applies only to element names, not to attribute names.
-func (p *Parser) translate(n *Name, isElementName bool) {
+func (d *Decoder) translate(n *Name, isElementName bool) {
switch {
case n.Space == "xmlns":
return
@@ -281,20 +281,20 @@
case n.Space == "" && n.Local == "xmlns":
return
}
- if v, ok := p.ns[n.Space]; ok {
+ if v, ok := d.ns[n.Space]; ok {
n.Space = v
}
}
-func (p *Parser) switchToReader(r io.Reader) {
+func (d *Decoder) switchToReader(r io.Reader) {
// Get efficient byte at a time reader.
// Assume that if reader has its own
// ReadByte, it's efficient enough.
// Otherwise, use bufio.
if rb, ok := r.(io.ByteReader); ok {
- p.r = rb
+ d.r = rb
} else {
- p.r = bufio.NewReader(r)
+ d.r = bufio.NewReader(r)
}
}
@@ -314,47 +314,47 @@
stkNs
)
-func (p *Parser) push(kind int) *stack {
- s := p.free
+func (d *Decoder) push(kind int) *stack {
+ s := d.free
if s != nil {
- p.free = s.next
+ d.free = s.next
} else {
s = new(stack)
}
- s.next = p.stk
+ s.next = d.stk
s.kind = kind
- p.stk = s
+ d.stk = s
return s
}
-func (p *Parser) pop() *stack {
- s := p.stk
+func (d *Decoder) pop() *stack {
+ s := d.stk
if s != nil {
- p.stk = s.next
- s.next = p.free
- p.free = s
+ d.stk = s.next
+ s.next = d.free
+ d.free = s
}
return s
}
// Record that we are starting an element with the given name.
-func (p *Parser) pushElement(name Name) {
- s := p.push(stkStart)
+func (d *Decoder) pushElement(name Name) {
+ s := d.push(stkStart)
s.name = name
}
// Record that we are changing the value of ns[local].
// The old value is url, ok.
-func (p *Parser) pushNs(local string, url string, ok bool) {
- s := p.push(stkNs)
+func (d *Decoder) pushNs(local string, url string, ok bool) {
+ s := d.push(stkNs)
s.name.Local = local
s.name.Space = url
s.ok = ok
}
// Creates a SyntaxError with the current line number.
-func (p *Parser) syntaxError(msg string) error {
- return &SyntaxError{Msg: msg, Line: p.line}
+func (d *Decoder) syntaxError(msg string) error {
+ return &SyntaxError{Msg: msg, Line: d.line}
}
// Record that we are ending an element with the given name.
@@ -363,36 +363,36 @@
// After popping the element, apply any undo records from
// the stack to restore the name translations that existed
// before we saw this element.
-func (p *Parser) popElement(t *EndElement) bool {
- s := p.pop()
+func (d *Decoder) popElement(t *EndElement) bool {
+ s := d.pop()
name := t.Name
switch {
case s == nil || s.kind != stkStart:
- p.err = p.syntaxError("unexpected end element </" + name.Local + ">")
+ d.err = d.syntaxError("unexpected end element </" + name.Local + ">")
return false
case s.name.Local != name.Local:
- if !p.Strict {
- p.needClose = true
- p.toClose = t.Name
+ if !d.Strict {
+ d.needClose = true
+ d.toClose = t.Name
t.Name = s.name
return true
}
- p.err = p.syntaxError("element <" + s.name.Local + "> closed by </" + name.Local + ">")
+ d.err = d.syntaxError("element <" + s.name.Local + "> closed by </" + name.Local + ">")
return false
case s.name.Space != name.Space:
- p.err = p.syntaxError("element <" + s.name.Local + "> in space " + s.name.Space +
+ d.err = d.syntaxError("element <" + s.name.Local + "> in space " + s.name.Space +
"closed by </" + name.Local + "> in space " + name.Space)
return false
}
// Pop stack until a Start is on the top, undoing the
// translations that were associated with the element we just closed.
- for p.stk != nil && p.stk.kind != stkStart {
- s := p.pop()
+ for d.stk != nil && d.stk.kind != stkStart {
+ s := d.pop()
if s.ok {
- p.ns[s.name.Local] = s.name.Space
+ d.ns[s.name.Local] = s.name.Space
} else {
- delete(p.ns, s.name.Local)
+ delete(d.ns, s.name.Local)
}
}
@@ -401,17 +401,17 @@
// If the top element on the stack is autoclosing and
// t is not the end tag, invent the end tag.
-func (p *Parser) autoClose(t Token) (Token, bool) {
- if p.stk == nil || p.stk.kind != stkStart {
+func (d *Decoder) autoClose(t Token) (Token, bool) {
+ if d.stk == nil || d.stk.kind != stkStart {
return nil, false
}
- name := strings.ToLower(p.stk.name.Local)
- for _, s := range p.AutoClose {
+ name := strings.ToLower(d.stk.name.Local)
+ for _, s := range d.AutoClose {
if strings.ToLower(s) == name {
// This one should be auto closed if t doesn't close it.
et, ok := t.(EndElement)
if !ok || et.Name.Local != name {
- return EndElement{p.stk.name}, true
+ return EndElement{d.stk.name}, true
}
break
}
@@ -422,53 +422,53 @@
// RawToken is like Token but does not verify that
// start and end elements match and does not translate
// name space prefixes to their corresponding URLs.
-func (p *Parser) RawToken() (Token, error) {
- if p.err != nil {
- return nil, p.err
+func (d *Decoder) RawToken() (Token, error) {
+ if d.err != nil {
+ return nil, d.err
}
- if p.needClose {
+ if d.needClose {
// The last element we read was self-closing and
// we returned just the StartElement half.
// Return the EndElement half now.
- p.needClose = false
- return EndElement{p.toClose}, nil
+ d.needClose = false
+ return EndElement{d.toClose}, nil
}
- b, ok := p.getc()
+ b, ok := d.getc()
if !ok {
- return nil, p.err
+ return nil, d.err
}
if b != '<' {
// Text section.
- p.ungetc(b)
- data := p.text(-1, false)
+ d.ungetc(b)
+ data := d.text(-1, false)
if data == nil {
- return nil, p.err
+ return nil, d.err
}
return CharData(data), nil
}
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
switch b {
case '/':
// </: End element
var name Name
- if name, ok = p.nsname(); !ok {
- if p.err == nil {
- p.err = p.syntaxError("expected element name after </")
+ if name, ok = d.nsname(); !ok {
+ if d.err == nil {
+ d.err = d.syntaxError("expected element name after </")
}
- return nil, p.err
+ return nil, d.err
}
- p.space()
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ d.space()
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if b != '>' {
- p.err = p.syntaxError("invalid characters between </" + name.Local + " and >")
- return nil, p.err
+ d.err = d.syntaxError("invalid characters between </" + name.Local + " and >")
+ return nil, d.err
}
return EndElement{name}, nil
@@ -477,95 +477,95 @@
// TODO(rsc): Should parse the <?xml declaration to make sure
// the version is 1.0 and the encoding is UTF-8.
var target string
- if target, ok = p.name(); !ok {
- if p.err == nil {
- p.err = p.syntaxError("expected target name after <?")
+ if target, ok = d.name(); !ok {
+ if d.err == nil {
+ d.err = d.syntaxError("expected target name after <?")
}
- return nil, p.err
+ return nil, d.err
}
- p.space()
- p.buf.Reset()
+ d.space()
+ d.buf.Reset()
var b0 byte
for {
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
- p.buf.WriteByte(b)
+ d.buf.WriteByte(b)
if b0 == '?' && b == '>' {
break
}
b0 = b
}
- data := p.buf.Bytes()
+ data := d.buf.Bytes()
data = data[0 : len(data)-2] // chop ?>
if target == "xml" {
enc := procInstEncoding(string(data))
if enc != "" && enc != "utf-8" && enc != "UTF-8" {
- if p.CharsetReader == nil {
- p.err = fmt.Errorf("xml: encoding %q declared but Parser.CharsetReader is nil", enc)
- return nil, p.err
+ if d.CharsetReader == nil {
+ d.err = fmt.Errorf("xml: encoding %q declared but Decoder.CharsetReader is nil", enc)
+ return nil, d.err
}
- newr, err := p.CharsetReader(enc, p.r.(io.Reader))
+ newr, err := d.CharsetReader(enc, d.r.(io.Reader))
if err != nil {
- p.err = fmt.Errorf("xml: opening charset %q: %v", enc, err)
- return nil, p.err
+ d.err = fmt.Errorf("xml: opening charset %q: %v", enc, err)
+ return nil, d.err
}
if newr == nil {
panic("CharsetReader returned a nil Reader for charset " + enc)
}
- p.switchToReader(newr)
+ d.switchToReader(newr)
}
}
return ProcInst{target, data}, nil
case '!':
// <!: Maybe comment, maybe CDATA.
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
switch b {
case '-': // <!-
// Probably <!-- for a comment.
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if b != '-' {
- p.err = p.syntaxError("invalid sequence <!- not part of <!--")
- return nil, p.err
+ d.err = d.syntaxError("invalid sequence <!- not part of <!--")
+ return nil, d.err
}
// Look for terminator.
- p.buf.Reset()
+ d.buf.Reset()
var b0, b1 byte
for {
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
- p.buf.WriteByte(b)
+ d.buf.WriteByte(b)
if b0 == '-' && b1 == '-' && b == '>' {
break
}
b0, b1 = b1, b
}
- data := p.buf.Bytes()
+ data := d.buf.Bytes()
data = data[0 : len(data)-3] // chop -->
return Comment(data), nil
case '[': // <![
// Probably <![CDATA[.
for i := 0; i < 6; i++ {
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if b != "CDATA["[i] {
- p.err = p.syntaxError("invalid <![ sequence")
- return nil, p.err
+ d.err = d.syntaxError("invalid <![ sequence")
+ return nil, d.err
}
}
// Have <![CDATA[. Read text until ]]>.
- data := p.text(-1, true)
+ data := d.text(-1, true)
if data == nil {
- return nil, p.err
+ return nil, d.err
}
return CharData(data), nil
}
@@ -573,18 +573,18 @@
// Probably a directive: <!DOCTYPE ...>, <!ENTITY ...>, etc.
// We don't care, but accumulate for caller. Quoted angle
// brackets do not count for nesting.
- p.buf.Reset()
- p.buf.WriteByte(b)
+ d.buf.Reset()
+ d.buf.WriteByte(b)
inquote := uint8(0)
depth := 0
for {
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if inquote == 0 && b == '>' && depth == 0 {
break
}
- p.buf.WriteByte(b)
+ d.buf.WriteByte(b)
switch {
case b == inquote:
inquote = 0
@@ -602,45 +602,45 @@
depth++
}
}
- return Directive(p.buf.Bytes()), nil
+ return Directive(d.buf.Bytes()), nil
}
// Must be an open element like <a href="foo">
- p.ungetc(b)
+ d.ungetc(b)
var (
name Name
empty bool
attr []Attr
)
- if name, ok = p.nsname(); !ok {
- if p.err == nil {
- p.err = p.syntaxError("expected element name after <")
+ if name, ok = d.nsname(); !ok {
+ if d.err == nil {
+ d.err = d.syntaxError("expected element name after <")
}
- return nil, p.err
+ return nil, d.err
}
attr = make([]Attr, 0, 4)
for {
- p.space()
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ d.space()
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if b == '/' {
empty = true
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if b != '>' {
- p.err = p.syntaxError("expected /> in element")
- return nil, p.err
+ d.err = d.syntaxError("expected /> in element")
+ return nil, d.err
}
break
}
if b == '>' {
break
}
- p.ungetc(b)
+ d.ungetc(b)
n := len(attr)
if n >= cap(attr) {
@@ -650,85 +650,85 @@
}
attr = attr[0 : n+1]
a := &attr[n]
- if a.Name, ok = p.nsname(); !ok {
- if p.err == nil {
- p.err = p.syntaxError("expected attribute name in element")
+ if a.Name, ok = d.nsname(); !ok {
+ if d.err == nil {
+ d.err = d.syntaxError("expected attribute name in element")
}
- return nil, p.err
+ return nil, d.err
}
- p.space()
- if b, ok = p.mustgetc(); !ok {
- return nil, p.err
+ d.space()
+ if b, ok = d.mustgetc(); !ok {
+ return nil, d.err
}
if b != '=' {
- if p.Strict {
- p.err = p.syntaxError("attribute name without = in element")
- return nil, p.err
+ if d.Strict {
+ d.err = d.syntaxError("attribute name without = in element")
+ return nil, d.err
} else {
- p.ungetc(b)
+ d.ungetc(b)
a.Value = a.Name.Local
}
} else {
- p.space()
- data := p.attrval()
+ d.space()
+ data := d.attrval()
if data == nil {
- return nil, p.err
+ return nil, d.err
}
a.Value = string(data)
}
}
if empty {
- p.needClose = true
- p.toClose = name
+ d.needClose = true
+ d.toClose = name
}
return StartElement{name, attr}, nil
}
-func (p *Parser) attrval() []byte {
- b, ok := p.mustgetc()
+func (d *Decoder) attrval() []byte {
+ b, ok := d.mustgetc()
if !ok {
return nil
}
// Handle quoted attribute values
if b == '"' || b == '\'' {
- return p.text(int(b), false)
+ return d.text(int(b), false)
}
// Handle unquoted attribute values for strict parsers
- if p.Strict {
- p.err = p.syntaxError("unquoted or missing attribute value in element")
+ if d.Strict {
+ d.err = d.syntaxError("unquoted or missing attribute value in element")
return nil
}
// Handle unquoted attribute values for unstrict parsers
- p.ungetc(b)
- p.buf.Reset()
+ d.ungetc(b)
+ d.buf.Reset()
for {
- b, ok = p.mustgetc()
+ b, ok = d.mustgetc()
if !ok {
return nil
}
// http://www.w3.org/TR/REC-html40/intro/sgmltut.html#h-3.2.2
if 'a' <= b && b <= 'z' || 'A' <= b && b <= 'Z' ||
'0' <= b && b <= '9' || b == '_' || b == ':' || b == '-' {
- p.buf.WriteByte(b)
+ d.buf.WriteByte(b)
} else {
- p.ungetc(b)
+ d.ungetc(b)
break
}
}
- return p.buf.Bytes()
+ return d.buf.Bytes()
}
// Skip spaces if any
-func (p *Parser) space() {
+func (d *Decoder) space() {
for {
- b, ok := p.getc()
+ b, ok := d.getc()
if !ok {
return
}
switch b {
case ' ', '\r', '\n', '\t':
default:
- p.ungetc(b)
+ d.ungetc(b)
return
}
}
@@ -736,35 +736,35 @@
// Read a single byte.
// If there is no byte to read, return ok==false
-// and leave the error in p.err.
+// and leave the error in d.err.
// Maintain line number.
-func (p *Parser) getc() (b byte, ok bool) {
- if p.err != nil {
+func (d *Decoder) getc() (b byte, ok bool) {
+ if d.err != nil {
return 0, false
}
- if p.nextByte >= 0 {
- b = byte(p.nextByte)
- p.nextByte = -1
+ if d.nextByte >= 0 {
+ b = byte(d.nextByte)
+ d.nextByte = -1
} else {
- b, p.err = p.r.ReadByte()
- if p.err != nil {
+ b, d.err = d.r.ReadByte()
+ if d.err != nil {
return 0, false
}
- if p.saved != nil {
- p.saved.WriteByte(b)
+ if d.saved != nil {
+ d.saved.WriteByte(b)
}
}
if b == '\n' {
- p.line++
+ d.line++
}
return b, true
}
// Return saved offset.
// If we did ungetc (nextByte >= 0), have to back up one.
-func (p *Parser) savedOffset() int {
- n := p.saved.Len()
- if p.nextByte >= 0 {
+func (d *Decoder) savedOffset() int {
+ n := d.saved.Len()
+ if d.nextByte >= 0 {
n--
}
return n
@@ -772,23 +772,23 @@
// Must read a single byte.
// If there is no byte to read,
-// set p.err to SyntaxError("unexpected EOF")
+// set d.err to SyntaxError("unexpected EOF")
// and return ok==false
-func (p *Parser) mustgetc() (b byte, ok bool) {
- if b, ok = p.getc(); !ok {
- if p.err == io.EOF {
- p.err = p.syntaxError("unexpected EOF")
+func (d *Decoder) mustgetc() (b byte, ok bool) {
+ if b, ok = d.getc(); !ok {
+ if d.err == io.EOF {
+ d.err = d.syntaxError("unexpected EOF")
}
}
return
}
// Unread a single byte.
-func (p *Parser) ungetc(b byte) {
+func (d *Decoder) ungetc(b byte) {
if b == '\n' {
- p.line--
+ d.line--
}
- p.nextByte = int(b)
+ d.nextByte = int(b)
}
var entity = map[string]int{
@@ -802,18 +802,18 @@
// Read plain text section (XML calls it character data).
// If quote >= 0, we are in a quoted string and need to find the matching quote.
// If cdata == true, we are in a <![CDATA[ section and need to find ]]>.
-// On failure return nil and leave the error in p.err.
-func (p *Parser) text(quote int, cdata bool) []byte {
+// On failure return nil and leave the error in d.err.
+func (d *Decoder) text(quote int, cdata bool) []byte {
var b0, b1 byte
var trunc int
- p.buf.Reset()
+ d.buf.Reset()
Input:
for {
- b, ok := p.getc()
+ b, ok := d.getc()
if !ok {
if cdata {
- if p.err == io.EOF {
- p.err = p.syntaxError("unexpected EOF in CDATA section")
+ if d.err == io.EOF {
+ d.err = d.syntaxError("unexpected EOF in CDATA section")
}
return nil
}
@@ -827,17 +827,17 @@
trunc = 2
break Input
}
- p.err = p.syntaxError("unescaped ]]> not in CDATA section")
+ d.err = d.syntaxError("unescaped ]]> not in CDATA section")
return nil
}
// Stop reading text if we see a <.
if b == '<' && !cdata {
if quote >= 0 {
- p.err = p.syntaxError("unescaped < inside quoted string")
+ d.err = d.syntaxError("unescaped < inside quoted string")
return nil
}
- p.ungetc('<')
+ d.ungetc('<')
break Input
}
if quote >= 0 && b == byte(quote) {
@@ -850,16 +850,16 @@
// Parsers are required to recognize lt, gt, amp, apos, and quot
// even if they have not been declared. That's all we allow.
var i int
- for i = 0; i < len(p.tmp); i++ {
+ for i = 0; i < len(d.tmp); i++ {
var ok bool
- p.tmp[i], ok = p.getc()
+ d.tmp[i], ok = d.getc()
if !ok {
- if p.err == io.EOF {
- p.err = p.syntaxError("unexpected EOF")
+ if d.err == io.EOF {
+ d.err = d.syntaxError("unexpected EOF")
}
return nil
}
- c := p.tmp[i]
+ c := d.tmp[i]
if c == ';' {
break
}
@@ -869,18 +869,18 @@
c == '_' || c == '#' {
continue
}
- p.ungetc(c)
+ d.ungetc(c)
break
}
- s := string(p.tmp[0:i])
- if i >= len(p.tmp) {
- if !p.Strict {
+ s := string(d.tmp[0:i])
+ if i >= len(d.tmp) {
+ if !d.Strict {
b0, b1 = 0, 0
- p.buf.WriteByte('&')
- p.buf.Write(p.tmp[0:i])
+ d.buf.WriteByte('&')
+ d.buf.Write(d.tmp[0:i])
continue Input
}
- p.err = p.syntaxError("character entity expression &" + s + "... too long")
+ d.err = d.syntaxError("character entity expression &" + s + "... too long")
return nil
}
var haveText bool
@@ -901,28 +901,28 @@
if r, ok := entity[s]; ok {
text = string(r)
haveText = true
- } else if p.Entity != nil {
- text, haveText = p.Entity[s]
+ } else if d.Entity != nil {
+ text, haveText = d.Entity[s]
}
}
if !haveText {
- if !p.Strict {
+ if !d.Strict {
b0, b1 = 0, 0
- p.buf.WriteByte('&')
- p.buf.Write(p.tmp[0:i])
+ d.buf.WriteByte('&')
+ d.buf.Write(d.tmp[0:i])
continue Input
}
- p.err = p.syntaxError("invalid character entity &" + s + ";")
+ d.err = d.syntaxError("invalid character entity &" + s + ";")
return nil
}
- p.buf.Write([]byte(text))
+ d.buf.Write([]byte(text))
b0, b1 = 0, 0
continue Input
}
- p.buf.WriteByte(b)
+ d.buf.WriteByte(b)
b0, b1 = b1, b
}
- data := p.buf.Bytes()
+ data := d.buf.Bytes()
data = data[0 : len(data)-trunc]
// Inspect each rune for being a disallowed character.
@@ -930,12 +930,12 @@
for len(buf) > 0 {
r, size := utf8.DecodeRune(buf)
if r == utf8.RuneError && size == 1 {
- p.err = p.syntaxError("invalid UTF-8")
+ d.err = d.syntaxError("invalid UTF-8")
return nil
}
buf = buf[size:]
if !isInCharacterRange(r) {
- p.err = p.syntaxError(fmt.Sprintf("illegal character code %U", r))
+ d.err = d.syntaxError(fmt.Sprintf("illegal character code %U", r))
return nil
}
}
@@ -970,8 +970,8 @@
// Get name space name: name with a : stuck in the middle.
// The part before the : is the name space identifier.
-func (p *Parser) nsname() (name Name, ok bool) {
- s, ok := p.name()
+func (d *Decoder) nsname() (name Name, ok bool) {
+ s, ok := d.name()
if !ok {
return
}
@@ -986,37 +986,37 @@
}
// Get name: /first(first|second)*/
-// Do not set p.err if the name is missing (unless unexpected EOF is received):
+// Do not set d.err if the name is missing (unless unexpected EOF is received):
// let the caller provide better context.
-func (p *Parser) name() (s string, ok bool) {
+func (d *Decoder) name() (s string, ok bool) {
var b byte
- if b, ok = p.mustgetc(); !ok {
+ if b, ok = d.mustgetc(); !ok {
return
}
// As a first approximation, we gather the bytes [A-Za-z_:.-\x80-\xFF]*
if b < utf8.RuneSelf && !isNameByte(b) {
- p.ungetc(b)
+ d.ungetc(b)
return "", false
}
- p.buf.Reset()
- p.buf.WriteByte(b)
+ d.buf.Reset()
+ d.buf.WriteByte(b)
for {
- if b, ok = p.mustgetc(); !ok {
+ if b, ok = d.mustgetc(); !ok {
return
}
if b < utf8.RuneSelf && !isNameByte(b) {
- p.ungetc(b)
+ d.ungetc(b)
break
}
- p.buf.WriteByte(b)
+ d.buf.WriteByte(b)
}
// Then we check the characters.
- s = p.buf.String()
+ s = d.buf.String()
for i, c := range s {
if !unicode.Is(first, c) && (i == 0 || !unicode.Is(second, c)) {
- p.err = p.syntaxError("invalid XML name: " + s)
+ d.err = d.syntaxError("invalid XML name: " + s)
return "", false
}
}
diff --git a/src/pkg/encoding/xml/xml_test.go b/src/pkg/encoding/xml/xml_test.go
index 524d4dd..1d0696c 100644
--- a/src/pkg/encoding/xml/xml_test.go
+++ b/src/pkg/encoding/xml/xml_test.go
@@ -5,7 +5,6 @@
package xml
import (
- "bytes"
"io"
"reflect"
"strings"
@@ -155,8 +154,8 @@
}
func TestRawToken(t *testing.T) {
- p := NewParser(strings.NewReader(testInput))
- testRawToken(t, p, rawTokens)
+ d := NewDecoder(strings.NewReader(testInput))
+ testRawToken(t, d, rawTokens)
}
type downCaser struct {
@@ -179,27 +178,27 @@
func TestRawTokenAltEncoding(t *testing.T) {
sawEncoding := ""
- p := NewParser(strings.NewReader(testInputAltEncoding))
- p.CharsetReader = func(charset string, input io.Reader) (io.Reader, error) {
+ d := NewDecoder(strings.NewReader(testInputAltEncoding))
+ d.CharsetReader = func(charset string, input io.Reader) (io.Reader, error) {
sawEncoding = charset
if charset != "x-testing-uppercase" {
t.Fatalf("unexpected charset %q", charset)
}
return &downCaser{t, input.(io.ByteReader)}, nil
}
- testRawToken(t, p, rawTokensAltEncoding)
+ testRawToken(t, d, rawTokensAltEncoding)
}
func TestRawTokenAltEncodingNoConverter(t *testing.T) {
- p := NewParser(strings.NewReader(testInputAltEncoding))
- token, err := p.RawToken()
+ d := NewDecoder(strings.NewReader(testInputAltEncoding))
+ token, err := d.RawToken()
if token == nil {
t.Fatalf("expected a token on first RawToken call")
}
if err != nil {
t.Fatal(err)
}
- token, err = p.RawToken()
+ token, err = d.RawToken()
if token != nil {
t.Errorf("expected a nil token; got %#v", token)
}
@@ -213,9 +212,9 @@
}
}
-func testRawToken(t *testing.T, p *Parser, rawTokens []Token) {
+func testRawToken(t *testing.T, d *Decoder, rawTokens []Token) {
for i, want := range rawTokens {
- have, err := p.RawToken()
+ have, err := d.RawToken()
if err != nil {
t.Fatalf("token %d: unexpected error: %s", i, err)
}
@@ -258,10 +257,10 @@
}
func TestNestedDirectives(t *testing.T) {
- p := NewParser(strings.NewReader(nestedDirectivesInput))
+ d := NewDecoder(strings.NewReader(nestedDirectivesInput))
for i, want := range nestedDirectivesTokens {
- have, err := p.Token()
+ have, err := d.Token()
if err != nil {
t.Fatalf("token %d: unexpected error: %s", i, err)
}
@@ -272,10 +271,10 @@
}
func TestToken(t *testing.T) {
- p := NewParser(strings.NewReader(testInput))
+ d := NewDecoder(strings.NewReader(testInput))
for i, want := range cookedTokens {
- have, err := p.Token()
+ have, err := d.Token()
if err != nil {
t.Fatalf("token %d: unexpected error: %s", i, err)
}
@@ -287,9 +286,9 @@
func TestSyntax(t *testing.T) {
for i := range xmlInput {
- p := NewParser(strings.NewReader(xmlInput[i]))
+ d := NewDecoder(strings.NewReader(xmlInput[i]))
var err error
- for _, err = p.Token(); err == nil; _, err = p.Token() {
+ for _, err = d.Token(); err == nil; _, err = d.Token() {
}
if _, ok := err.(*SyntaxError); !ok {
t.Fatalf(`xmlInput "%s": expected SyntaxError not received`, xmlInput[i])
@@ -368,8 +367,7 @@
func TestAllScalars(t *testing.T) {
var a allScalars
- buf := bytes.NewBufferString(testScalarsInput)
- err := Unmarshal(buf, &a)
+ err := Unmarshal([]byte(testScalarsInput), &a)
if err != nil {
t.Fatal(err)
@@ -386,8 +384,7 @@
func TestIssue569(t *testing.T) {
data := `<item><Field_a>abcd</Field_a></item>`
var i item
- buf := bytes.NewBufferString(data)
- err := Unmarshal(buf, &i)
+ err := Unmarshal([]byte(data), &i)
if err != nil || i.Field_a != "abcd" {
t.Fatal("Expecting abcd")
@@ -396,9 +393,9 @@
func TestUnquotedAttrs(t *testing.T) {
data := "<tag attr=azAZ09:-_\t>"
- p := NewParser(strings.NewReader(data))
- p.Strict = false
- token, err := p.Token()
+ d := NewDecoder(strings.NewReader(data))
+ d.Strict = false
+ token, err := d.Token()
if _, ok := err.(*SyntaxError); ok {
t.Errorf("Unexpected error: %v", err)
}
@@ -422,9 +419,9 @@
{"<input checked />", "input", "checked"},
}
for _, test := range tests {
- p := NewParser(strings.NewReader(test[0]))
- p.Strict = false
- token, err := p.Token()
+ d := NewDecoder(strings.NewReader(test[0]))
+ d.Strict = false
+ token, err := d.Token()
if _, ok := err.(*SyntaxError); ok {
t.Errorf("Unexpected error: %v", err)
}
@@ -472,9 +469,9 @@
func TestSyntaxErrorLineNum(t *testing.T) {
testInput := "<P>Foo<P>\n\n<P>Bar</>\n"
- p := NewParser(strings.NewReader(testInput))
+ d := NewDecoder(strings.NewReader(testInput))
var err error
- for _, err = p.Token(); err == nil; _, err = p.Token() {
+ for _, err = d.Token(); err == nil; _, err = d.Token() {
}
synerr, ok := err.(*SyntaxError)
if !ok {
@@ -487,41 +484,41 @@
func TestTrailingRawToken(t *testing.T) {
input := `<FOO></FOO> `
- p := NewParser(strings.NewReader(input))
+ d := NewDecoder(strings.NewReader(input))
var err error
- for _, err = p.RawToken(); err == nil; _, err = p.RawToken() {
+ for _, err = d.RawToken(); err == nil; _, err = d.RawToken() {
}
if err != io.EOF {
- t.Fatalf("p.RawToken() = _, %v, want _, io.EOF", err)
+ t.Fatalf("d.RawToken() = _, %v, want _, io.EOF", err)
}
}
func TestTrailingToken(t *testing.T) {
input := `<FOO></FOO> `
- p := NewParser(strings.NewReader(input))
+ d := NewDecoder(strings.NewReader(input))
var err error
- for _, err = p.Token(); err == nil; _, err = p.Token() {
+ for _, err = d.Token(); err == nil; _, err = d.Token() {
}
if err != io.EOF {
- t.Fatalf("p.Token() = _, %v, want _, io.EOF", err)
+ t.Fatalf("d.Token() = _, %v, want _, io.EOF", err)
}
}
func TestEntityInsideCDATA(t *testing.T) {
input := `<test><![CDATA[ &val=foo ]]></test>`
- p := NewParser(strings.NewReader(input))
+ d := NewDecoder(strings.NewReader(input))
var err error
- for _, err = p.Token(); err == nil; _, err = p.Token() {
+ for _, err = d.Token(); err == nil; _, err = d.Token() {
}
if err != io.EOF {
- t.Fatalf("p.Token() = _, %v, want _, io.EOF", err)
+ t.Fatalf("d.Token() = _, %v, want _, io.EOF", err)
}
}
// The last three tests (respectively one for characters in attribute
// names and two for character entities) pass not because of code
// changed for issue 1259, but instead pass with the given messages
-// from other parts of xml.Parser. I provide these to note the
+// from other parts of xml.Decoder. I provide these to note the
// current behavior of situations where one might think that character
// range checking would detect the error, but it does not in fact.
@@ -541,15 +538,15 @@
func TestDisallowedCharacters(t *testing.T) {
for i, tt := range characterTests {
- p := NewParser(strings.NewReader(tt.in))
+ d := NewDecoder(strings.NewReader(tt.in))
var err error
for err == nil {
- _, err = p.Token()
+ _, err = d.Token()
}
synerr, ok := err.(*SyntaxError)
if !ok {
- t.Fatalf("input %d p.Token() = _, %v, want _, *SyntaxError", i, err)
+ t.Fatalf("input %d d.Token() = _, %v, want _, *SyntaxError", i, err)
}
if synerr.Msg != tt.err {
t.Fatalf("input %d synerr.Msg wrong: want '%s', got '%s'", i, tt.err, synerr.Msg)
