xml_test.go 19.1 KB
Newer Older
1 2 3 4 5 6 7
// Copyright 2009 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.

package xml

import (
8
	"bytes"
9
	"fmt"
10 11
	"io"
	"reflect"
12
	"strings"
13
	"testing"
14
	"unicode/utf8"
15 16 17 18 19 20 21 22 23
)

const testInput = `
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<body xmlns:foo="ns1" xmlns="ns2" xmlns:tag="ns3" ` +
	"\r\n\t" + `  >
  <hello lang="en">World &lt;&gt;&apos;&quot; &#x767d;&#40300;翔</hello>
24
  <query>&何; &is-it;</query>
25 26 27 28 29 30 31 32 33
  <goodbye />
  <outer foo:attr="value" xmlns:tag="ns4">
    <inner/>
  </outer>
  <tag:name>
    <![CDATA[Some text here.]]>
  </tag:name>
</body><!-- missing final newline -->`

34 35
var testEntity = map[string]string{"何": "What", "is-it": "is it?"}

36
var rawTokens = []Token{
37
	CharData("\n"),
38
	ProcInst{"xml", []byte(`version="1.0" encoding="UTF-8"`)},
39 40
	CharData("\n"),
	Directive(`DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
41
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"`),
42
	CharData("\n"),
43
	StartElement{Name{"", "body"}, []Attr{{Name{"xmlns", "foo"}, "ns1"}, {Name{"", "xmlns"}, "ns2"}, {Name{"xmlns", "tag"}, "ns3"}}},
44
	CharData("\n  "),
45
	StartElement{Name{"", "hello"}, []Attr{{Name{"", "lang"}, "en"}}},
46
	CharData("World <>'\" 白鵬翔"),
47
	EndElement{Name{"", "hello"}},
48
	CharData("\n  "),
49 50 51 52
	StartElement{Name{"", "query"}, []Attr{}},
	CharData("What is it?"),
	EndElement{Name{"", "query"}},
	CharData("\n  "),
53
	StartElement{Name{"", "goodbye"}, []Attr{}},
54
	EndElement{Name{"", "goodbye"}},
55
	CharData("\n  "),
56
	StartElement{Name{"", "outer"}, []Attr{{Name{"foo", "attr"}, "value"}, {Name{"xmlns", "tag"}, "ns4"}}},
57
	CharData("\n    "),
58
	StartElement{Name{"", "inner"}, []Attr{}},
59
	EndElement{Name{"", "inner"}},
60
	CharData("\n  "),
61
	EndElement{Name{"", "outer"}},
62
	CharData("\n  "),
63
	StartElement{Name{"tag", "name"}, []Attr{}},
64 65 66
	CharData("\n    "),
	CharData("Some text here."),
	CharData("\n  "),
67
	EndElement{Name{"tag", "name"}},
68
	CharData("\n"),
69
	EndElement{Name{"", "body"}},
70
	Comment(" missing final newline "),
71 72 73
}

var cookedTokens = []Token{
74
	CharData("\n"),
75
	ProcInst{"xml", []byte(`version="1.0" encoding="UTF-8"`)},
76 77
	CharData("\n"),
	Directive(`DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
78
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"`),
79
	CharData("\n"),
80
	StartElement{Name{"ns2", "body"}, []Attr{{Name{"xmlns", "foo"}, "ns1"}, {Name{"", "xmlns"}, "ns2"}, {Name{"xmlns", "tag"}, "ns3"}}},
81
	CharData("\n  "),
82
	StartElement{Name{"ns2", "hello"}, []Attr{{Name{"", "lang"}, "en"}}},
83
	CharData("World <>'\" 白鵬翔"),
84
	EndElement{Name{"ns2", "hello"}},
85
	CharData("\n  "),
86 87 88 89
	StartElement{Name{"ns2", "query"}, []Attr{}},
	CharData("What is it?"),
	EndElement{Name{"ns2", "query"}},
	CharData("\n  "),
90
	StartElement{Name{"ns2", "goodbye"}, []Attr{}},
91
	EndElement{Name{"ns2", "goodbye"}},
92
	CharData("\n  "),
93
	StartElement{Name{"ns2", "outer"}, []Attr{{Name{"ns1", "attr"}, "value"}, {Name{"xmlns", "tag"}, "ns4"}}},
94
	CharData("\n    "),
95
	StartElement{Name{"ns2", "inner"}, []Attr{}},
96
	EndElement{Name{"ns2", "inner"}},
97
	CharData("\n  "),
98
	EndElement{Name{"ns2", "outer"}},
99
	CharData("\n  "),
100
	StartElement{Name{"ns3", "name"}, []Attr{}},
101 102 103
	CharData("\n    "),
	CharData("Some text here."),
	CharData("\n  "),
104
	EndElement{Name{"ns3", "name"}},
105
	CharData("\n"),
106
	EndElement{Name{"ns2", "body"}},
107
	Comment(" missing final newline "),
108 109
}

110 111 112 113 114
const testInputAltEncoding = `
<?xml version="1.0" encoding="x-testing-uppercase"?>
<TAG>VALUE</TAG>`

var rawTokensAltEncoding = []Token{
115
	CharData("\n"),
116
	ProcInst{"xml", []byte(`version="1.0" encoding="x-testing-uppercase"`)},
117
	CharData("\n"),
118
	StartElement{Name{"", "tag"}, []Attr{}},
119
	CharData("value"),
120 121 122
	EndElement{Name{"", "tag"}},
}

123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
var xmlInput = []string{
	// unexpected EOF cases
	"<",
	"<t",
	"<t ",
	"<t/",
	"<!",
	"<!-",
	"<!--",
	"<!--c-",
	"<!--c--",
	"<!d",
	"<t></",
	"<t></t",
	"<?",
	"<?p",
	"<t a",
	"<t a=",
	"<t a='",
	"<t a=''",
	"<t/><![",
	"<t/><![C",
	"<t/><![CDATA[d",
	"<t/><![CDATA[d]",
	"<t/><![CDATA[d]]",

	// other Syntax errors
	"<>",
	"<t/a",
	"<0 />",
	"<?0 >",
	//	"<!0 >",	// let the Token() caller handle
	"</0>",
	"<t 0=''>",
	"<t a='&'>",
	"<t a='<'>",
	"<t>&nbspc;</t>",
	"<t a>",
	"<t a=>",
	"<t a=v>",
	//	"<![CDATA[d]]>",	// let the Token() caller handle
	"<t></e>",
	"<t></>",
	"<t></t!",
	"<t>cdata]]></t>",
}

func TestRawToken(t *testing.T) {
171
	d := NewDecoder(strings.NewReader(testInput))
172
	d.Entity = testEntity
173
	testRawToken(t, d, testInput, rawTokens)
174 175
}

176 177 178 179 180
const nonStrictInput = `
<tag>non&entity</tag>
<tag>&unknown;entity</tag>
<tag>&#123</tag>
<tag>&#zzz;</tag>
181 182 183 184
<tag>&なまえ3;</tag>
<tag>&lt-gt;</tag>
<tag>&;</tag>
<tag>&0a;</tag>
185 186
`

187 188
var nonStringEntity = map[string]string{"": "oops!", "0a": "oops!"}

189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
var nonStrictTokens = []Token{
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("non&entity"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&unknown;entity"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&#123"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&#zzz;"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&なまえ3;"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&lt-gt;"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&;"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
	StartElement{Name{"", "tag"}, []Attr{}},
	CharData("&0a;"),
	EndElement{Name{"", "tag"}},
	CharData("\n"),
223 224 225 226 227
}

func TestNonStrictRawToken(t *testing.T) {
	d := NewDecoder(strings.NewReader(nonStrictInput))
	d.Strict = false
228
	testRawToken(t, d, nonStrictInput, nonStrictTokens)
229 230
}

231 232 233 234
type downCaser struct {
	t *testing.T
	r io.ByteReader
}
235

236
func (d *downCaser) ReadByte() (c byte, err error) {
237 238 239 240 241 242 243
	c, err = d.r.ReadByte()
	if c >= 'A' && c <= 'Z' {
		c += 'a' - 'A'
	}
	return
}

244
func (d *downCaser) Read(p []byte) (int, error) {
245
	d.t.Fatalf("unexpected Read call on downCaser reader")
246
	panic("unreachable")
247 248 249
}

func TestRawTokenAltEncoding(t *testing.T) {
250 251
	d := NewDecoder(strings.NewReader(testInputAltEncoding))
	d.CharsetReader = func(charset string, input io.Reader) (io.Reader, error) {
252 253 254 255 256
		if charset != "x-testing-uppercase" {
			t.Fatalf("unexpected charset %q", charset)
		}
		return &downCaser{t, input.(io.ByteReader)}, nil
	}
257
	testRawToken(t, d, testInputAltEncoding, rawTokensAltEncoding)
258 259 260
}

func TestRawTokenAltEncodingNoConverter(t *testing.T) {
261 262
	d := NewDecoder(strings.NewReader(testInputAltEncoding))
	token, err := d.RawToken()
263 264 265 266 267 268
	if token == nil {
		t.Fatalf("expected a token on first RawToken call")
	}
	if err != nil {
		t.Fatal(err)
	}
269
	token, err = d.RawToken()
270 271 272 273 274 275 276
	if token != nil {
		t.Errorf("expected a nil token; got %#v", token)
	}
	if err == nil {
		t.Fatalf("expected an error on second RawToken call")
	}
	const encoding = "x-testing-uppercase"
277
	if !strings.Contains(err.Error(), encoding) {
278 279 280 281 282
		t.Errorf("expected error to contain %q; got error: %v",
			encoding, err)
	}
}

283 284
func testRawToken(t *testing.T, d *Decoder, raw string, rawTokens []Token) {
	lastEnd := int64(0)
285
	for i, want := range rawTokens {
286
		start := d.InputOffset()
287
		have, err := d.RawToken()
288
		end := d.InputOffset()
289 290 291 292
		if err != nil {
			t.Fatalf("token %d: unexpected error: %s", i, err)
		}
		if !reflect.DeepEqual(have, want) {
293 294 295 296 297 298 299 300 301 302 303 304
			var shave, swant string
			if _, ok := have.(CharData); ok {
				shave = fmt.Sprintf("CharData(%q)", have)
			} else {
				shave = fmt.Sprintf("%#v", have)
			}
			if _, ok := want.(CharData); ok {
				swant = fmt.Sprintf("CharData(%q)", want)
			} else {
				swant = fmt.Sprintf("%#v", want)
			}
			t.Errorf("token %d = %s, want %s", i, shave, swant)
305
		}
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325

		// Check that InputOffset returned actual token.
		switch {
		case start < lastEnd:
			t.Errorf("token %d: position [%d,%d) for %T is before previous token", i, start, end, have)
		case start >= end:
			// Special case: EndElement can be synthesized.
			if start == end && end == lastEnd {
				break
			}
			t.Errorf("token %d: position [%d,%d) for %T is empty", i, start, end, have)
		case end > int64(len(raw)):
			t.Errorf("token %d: position [%d,%d) for %T extends beyond input", i, start, end, have)
		default:
			text := raw[start:end]
			if strings.ContainsAny(text, "<>") && (!strings.HasPrefix(text, "<") || !strings.HasSuffix(text, ">")) {
				t.Errorf("token %d: misaligned raw token %#q for %T", i, text, have)
			}
		}
		lastEnd = end
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343
	}
}

// Ensure that directives (specifically !DOCTYPE) include the complete
// text of any nested directives, noting that < and > do not change
// nesting depth if they are in single or double quotes.

var nestedDirectivesInput = `
<!DOCTYPE [<!ENTITY rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#">]>
<!DOCTYPE [<!ENTITY xlt ">">]>
<!DOCTYPE [<!ENTITY xlt "<">]>
<!DOCTYPE [<!ENTITY xlt '>'>]>
<!DOCTYPE [<!ENTITY xlt '<'>]>
<!DOCTYPE [<!ENTITY xlt '">'>]>
<!DOCTYPE [<!ENTITY xlt "'<">]>
`

var nestedDirectivesTokens = []Token{
344 345 346 347 348 349 350 351 352 353 354 355 356 357 358
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#">]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY xlt ">">]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY xlt "<">]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY xlt '>'>]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY xlt '<'>]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY xlt '">'>]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY xlt "'<">]`),
	CharData("\n"),
359 360 361
}

func TestNestedDirectives(t *testing.T) {
362
	d := NewDecoder(strings.NewReader(nestedDirectivesInput))
363 364

	for i, want := range nestedDirectivesTokens {
365
		have, err := d.Token()
366 367 368 369
		if err != nil {
			t.Fatalf("token %d: unexpected error: %s", i, err)
		}
		if !reflect.DeepEqual(have, want) {
370 371 372 373 374 375
			t.Errorf("token %d = %#v want %#v", i, have, want)
		}
	}
}

func TestToken(t *testing.T) {
376
	d := NewDecoder(strings.NewReader(testInput))
377
	d.Entity = testEntity
378 379

	for i, want := range cookedTokens {
380
		have, err := d.Token()
381 382 383 384 385 386 387 388 389 390 391
		if err != nil {
			t.Fatalf("token %d: unexpected error: %s", i, err)
		}
		if !reflect.DeepEqual(have, want) {
			t.Errorf("token %d = %#v want %#v", i, have, want)
		}
	}
}

func TestSyntax(t *testing.T) {
	for i := range xmlInput {
392
		d := NewDecoder(strings.NewReader(xmlInput[i]))
393
		var err error
394
		for _, err = d.Token(); err == nil; _, err = d.Token() {
395 396 397 398 399 400 401 402
		}
		if _, ok := err.(*SyntaxError); !ok {
			t.Fatalf(`xmlInput "%s": expected SyntaxError not received`, xmlInput[i])
		}
	}
}

type allScalars struct {
403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
	True1     bool
	True2     bool
	False1    bool
	False2    bool
	Int       int
	Int8      int8
	Int16     int16
	Int32     int32
	Int64     int64
	Uint      int
	Uint8     uint8
	Uint16    uint16
	Uint32    uint32
	Uint64    uint64
	Uintptr   uintptr
	Float32   float32
	Float64   float64
	String    string
	PtrString *string
422 423 424
}

var all = allScalars{
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
	True1:     true,
	True2:     true,
	False1:    false,
	False2:    false,
	Int:       1,
	Int8:      -2,
	Int16:     3,
	Int32:     -4,
	Int64:     5,
	Uint:      6,
	Uint8:     7,
	Uint16:    8,
	Uint32:    9,
	Uint64:    10,
	Uintptr:   11,
	Float32:   13.0,
	Float64:   14.0,
	String:    "15",
	PtrString: &sixteen,
444 445
}

446 447
var sixteen = "16"

448
const testScalarsInput = `<allscalars>
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468
	<True1>true</True1>
	<True2>1</True2>
	<False1>false</False1>
	<False2>0</False2>
	<Int>1</Int>
	<Int8>-2</Int8>
	<Int16>3</Int16>
	<Int32>-4</Int32>
	<Int64>5</Int64>
	<Uint>6</Uint>
	<Uint8>7</Uint8>
	<Uint16>8</Uint16>
	<Uint32>9</Uint32>
	<Uint64>10</Uint64>
	<Uintptr>11</Uintptr>
	<Float>12.0</Float>
	<Float32>13.0</Float32>
	<Float64>14.0</Float64>
	<String>15</String>
	<PtrString>16</PtrString>
469 470 471 472
</allscalars>`

func TestAllScalars(t *testing.T) {
	var a allScalars
473
	err := Unmarshal([]byte(testScalarsInput), &a)
474 475 476 477 478

	if err != nil {
		t.Fatal(err)
	}
	if !reflect.DeepEqual(a, all) {
479
		t.Errorf("have %+v want %+v", a, all)
480 481 482 483 484 485 486 487
	}
}

type item struct {
	Field_a string
}

func TestIssue569(t *testing.T) {
488
	data := `<item><Field_a>abcd</Field_a></item>`
489
	var i item
490
	err := Unmarshal([]byte(data), &i)
491 492

	if err != nil || i.Field_a != "abcd" {
493
		t.Fatal("Expecting abcd")
494 495 496 497 498
	}
}

func TestUnquotedAttrs(t *testing.T) {
	data := "<tag attr=azAZ09:-_\t>"
499 500 501
	d := NewDecoder(strings.NewReader(data))
	d.Strict = false
	token, err := d.Token()
502 503 504 505 506 507 508 509 510 511 512 513 514 515 516
	if _, ok := err.(*SyntaxError); ok {
		t.Errorf("Unexpected error: %v", err)
	}
	if token.(StartElement).Name.Local != "tag" {
		t.Errorf("Unexpected tag name: %v", token.(StartElement).Name.Local)
	}
	attr := token.(StartElement).Attr[0]
	if attr.Value != "azAZ09:-_" {
		t.Errorf("Unexpected attribute value: %v", attr.Value)
	}
	if attr.Name.Local != "attr" {
		t.Errorf("Unexpected attribute name: %v", attr.Name.Local)
	}
}

517 518 519 520 521 522 523 524
func TestValuelessAttrs(t *testing.T) {
	tests := [][3]string{
		{"<p nowrap>", "p", "nowrap"},
		{"<p nowrap >", "p", "nowrap"},
		{"<input checked/>", "input", "checked"},
		{"<input checked />", "input", "checked"},
	}
	for _, test := range tests {
525 526 527
		d := NewDecoder(strings.NewReader(test[0]))
		d.Strict = false
		token, err := d.Token()
528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543
		if _, ok := err.(*SyntaxError); ok {
			t.Errorf("Unexpected error: %v", err)
		}
		if token.(StartElement).Name.Local != test[1] {
			t.Errorf("Unexpected tag name: %v", token.(StartElement).Name.Local)
		}
		attr := token.(StartElement).Attr[0]
		if attr.Value != test[2] {
			t.Errorf("Unexpected attribute value: %v", attr.Value)
		}
		if attr.Name.Local != test[2] {
			t.Errorf("Unexpected attribute name: %v", attr.Name.Local)
		}
	}
}

544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560
func TestCopyTokenCharData(t *testing.T) {
	data := []byte("same data")
	var tok1 Token = CharData(data)
	tok2 := CopyToken(tok1)
	if !reflect.DeepEqual(tok1, tok2) {
		t.Error("CopyToken(CharData) != CharData")
	}
	data[1] = 'o'
	if reflect.DeepEqual(tok1, tok2) {
		t.Error("CopyToken(CharData) uses same buffer.")
	}
}

func TestCopyTokenStartElement(t *testing.T) {
	elt := StartElement{Name{"", "hello"}, []Attr{{Name{"", "lang"}, "en"}}}
	var tok1 Token = elt
	tok2 := CopyToken(tok1)
561 562 563
	if tok1.(StartElement).Attr[0].Value != "en" {
		t.Error("CopyToken overwrote Attr[0]")
	}
564 565 566
	if !reflect.DeepEqual(tok1, tok2) {
		t.Error("CopyToken(StartElement) != StartElement")
	}
567
	tok1.(StartElement).Attr[0] = Attr{Name{"", "lang"}, "de"}
568 569 570 571 572 573 574
	if reflect.DeepEqual(tok1, tok2) {
		t.Error("CopyToken(CharData) uses same buffer.")
	}
}

func TestSyntaxErrorLineNum(t *testing.T) {
	testInput := "<P>Foo<P>\n\n<P>Bar</>\n"
575
	d := NewDecoder(strings.NewReader(testInput))
576
	var err error
577
	for _, err = d.Token(); err == nil; _, err = d.Token() {
578 579 580 581 582 583 584 585 586 587 588 589
	}
	synerr, ok := err.(*SyntaxError)
	if !ok {
		t.Error("Expected SyntaxError.")
	}
	if synerr.Line != 3 {
		t.Error("SyntaxError didn't have correct line number.")
	}
}

func TestTrailingRawToken(t *testing.T) {
	input := `<FOO></FOO>  `
590
	d := NewDecoder(strings.NewReader(input))
591
	var err error
592
	for _, err = d.RawToken(); err == nil; _, err = d.RawToken() {
593
	}
594
	if err != io.EOF {
595
		t.Fatalf("d.RawToken() = _, %v, want _, io.EOF", err)
596 597 598 599 600
	}
}

func TestTrailingToken(t *testing.T) {
	input := `<FOO></FOO>  `
601
	d := NewDecoder(strings.NewReader(input))
602
	var err error
603
	for _, err = d.Token(); err == nil; _, err = d.Token() {
604
	}
605
	if err != io.EOF {
606
		t.Fatalf("d.Token() = _, %v, want _, io.EOF", err)
607 608 609 610 611
	}
}

func TestEntityInsideCDATA(t *testing.T) {
	input := `<test><![CDATA[ &val=foo ]]></test>`
612
	d := NewDecoder(strings.NewReader(input))
613
	var err error
614
	for _, err = d.Token(); err == nil; _, err = d.Token() {
615
	}
616
	if err != io.EOF {
617
		t.Fatalf("d.Token() = _, %v, want _, io.EOF", err)
618 619
	}
}
620 621 622 623 624 625 626 627 628 629

var characterTests = []struct {
	in  string
	err string
}{
	{"\x12<doc/>", "illegal character code U+0012"},
	{"<?xml version=\"1.0\"?>\x0b<doc/>", "illegal character code U+000B"},
	{"\xef\xbf\xbe<doc/>", "illegal character code U+FFFE"},
	{"<?xml version=\"1.0\"?><doc>\r\n<hiya/>\x07<toots/></doc>", "illegal character code U+0007"},
	{"<?xml version=\"1.0\"?><doc \x12='value'>what's up</doc>", "expected attribute name in element"},
630
	{"<doc>&abc\x01;</doc>", "invalid character entity &abc (no semicolon)"},
631
	{"<doc>&\x01;</doc>", "invalid character entity & (no semicolon)"},
632 633
	{"<doc>&\xef\xbf\xbe;</doc>", "invalid character entity &\uFFFE;"},
	{"<doc>&hello;</doc>", "invalid character entity &hello;"},
634 635 636 637 638
}

func TestDisallowedCharacters(t *testing.T) {

	for i, tt := range characterTests {
639
		d := NewDecoder(strings.NewReader(tt.in))
640
		var err error
641 642

		for err == nil {
643
			_, err = d.Token()
644 645 646
		}
		synerr, ok := err.(*SyntaxError)
		if !ok {
647
			t.Fatalf("input %d d.Token() = _, %v, want _, *SyntaxError", i, err)
648 649
		}
		if synerr.Msg != tt.err {
650
			t.Fatalf("input %d synerr.Msg wrong: want %q, got %q", i, tt.err, synerr.Msg)
651 652 653
		}
	}
}
654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676

type procInstEncodingTest struct {
	expect, got string
}

var procInstTests = []struct {
	input, expect 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"},
}

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)
		}
	}
}
677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709

// Ensure that directives with comments include the complete
// text of any nested directives.

var directivesWithCommentsInput = `
<!DOCTYPE [<!-- a comment --><!ENTITY rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#">]>
<!DOCTYPE [<!ENTITY go "Golang"><!-- a comment-->]>
<!DOCTYPE <!-> <!> <!----> <!-->--> <!--->--> [<!ENTITY go "Golang"><!-- a comment-->]>
`

var directivesWithCommentsTokens = []Token{
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#">]`),
	CharData("\n"),
	Directive(`DOCTYPE [<!ENTITY go "Golang">]`),
	CharData("\n"),
	Directive(`DOCTYPE <!-> <!>    [<!ENTITY go "Golang">]`),
	CharData("\n"),
}

func TestDirectivesWithComments(t *testing.T) {
	d := NewDecoder(strings.NewReader(directivesWithCommentsInput))

	for i, want := range directivesWithCommentsTokens {
		have, err := d.Token()
		if err != nil {
			t.Fatalf("token %d: unexpected error: %s", i, err)
		}
		if !reflect.DeepEqual(have, want) {
			t.Errorf("token %d = %#v want %#v", i, have, want)
		}
	}
}
710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738

// Writer whose Write method always returns an error.
type errWriter struct{}

func (errWriter) Write(p []byte) (n int, err error) { return 0, fmt.Errorf("unwritable") }

func TestEscapeTextIOErrors(t *testing.T) {
	expectErr := "unwritable"
	err := EscapeText(errWriter{}, []byte{'A'})

	if err == nil || err.Error() != expectErr {
		t.Errorf("have %v, want %v", err, expectErr)
	}
}

func TestEscapeTextInvalidChar(t *testing.T) {
	input := []byte("A \x00 terminated string.")
	expected := "A \uFFFD terminated string."

	buff := new(bytes.Buffer)
	if err := EscapeText(buff, input); err != nil {
		t.Fatalf("have %v, want nil", err)
	}
	text := buff.String()

	if text != expected {
		t.Errorf("have %v, want %v", text, expected)
	}
}
739 740 741 742 743 744 745 746 747 748 749

func TestIssue5880(t *testing.T) {
	type T []byte
	data, err := Marshal(T{192, 168, 0, 1})
	if err != nil {
		t.Errorf("Marshal error: %v", err)
	}
	if !utf8.Valid(data) {
		t.Errorf("Marshal generated invalid UTF-8: %x", data)
	}
}