summaryrefslogtreecommitdiff
path: root/src/cmd/gofix/testdata/reflect.read.go.in
blob: 9ae3bb8eee91a08dcd121a4d6fc15ba6d4102a10 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
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
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
// 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 (
	"bytes"
	"fmt"
	"io"
	"os"
	"reflect"
	"strconv"
	"strings"
	"unicode"
	"utf8"
)

// BUG(rsc): Mapping between XML elements and data structures is inherently flawed:
// an XML element is an order-dependent collection of anonymous
// values, while a data structure is an order-independent collection
// of named values.
// 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.
//
// For example, given these definitions:
//
//	type Email struct {
//		Where string "attr"
//		Addr  string
//	}
//
//	type Result struct {
//		XMLName xml.Name "result"
//		Name	string
//		Phone	string
//		Email	[]Email
//		Groups  []string "group>value"
//	}
//
//	result := Result{Name: "name", Phone: "phone", Email: nil}
//
// unmarshalling the XML input
//
//	<result>
//		<email where="home">
//			<addr>gre@example.com</addr>
//		</email>
//		<email where='work'>
//			<addr>gre@work.com</addr>
//		</email>
//		<name>Grace R. Emlin</name>
// 		<group>
// 			<value>Friends</value>
// 			<value>Squash</value>
// 		</group>
//		<address>123 Main Street</address>
//	</result>
//
// via Unmarshal(r, &result) is equivalent to assigning
//
//	r = Result{xml.Name{"", "result"},
//		"Grace R. Emlin", // name
//		"phone",	  // no phone given
//		[]Email{
//			Email{"home", "gre@example.com"},
//			Email{"work", "gre@work.com"},
//		},
//		[]string{"Friends", "Squash"},
//	}
//
// Note that the field r.Phone has not been modified and
// that the XML <address> element was discarded. Also, the field
// Groups was assigned considering the element path provided in the
// field tag.
//
// Because Unmarshal uses the reflect package, it can only
// assign to upper case fields.  Unmarshal uses a case-insensitive
// comparison to match XML element names to struct field names.
//
// Unmarshal maps an XML element to a struct using the following rules:
//
//   * If the struct has a field of type []byte or string with tag "innerxml",
//      Unmarshal accumulates the raw XML nested inside the element
//      in that field.  The rest of the rules still apply.
//
//   * If the struct has a field named XMLName of type xml.Name,
//      Unmarshal records the element name in that field.
//
//   * If the XMLName field has an associated tag string of the form
//      "tag" or "namespace-URL tag", the XML element must have
//      the given tag (and, optionally, name space) or else Unmarshal
//      returns an error.
//
//   * If the XML element has an attribute whose name matches a
//      struct field of type string with tag "attr", Unmarshal records
//      the attribute value in that field.
//
//   * If the XML element contains character data, that data is
//      accumulated in the first struct field that has tag "chardata".
//      The struct field may have type []byte or string.
//      If there is no such field, the character data is discarded.
//
//   * If the XML element contains a sub-element whose name matches
//      the prefix of a struct field tag formatted as "a>b>c", unmarshal
//      will descend into the XML structure looking for elements with the
//      given names, and will map the innermost elements to that struct field.
//      A struct field tag starting with ">" is equivalent to one starting
//      with the field name followed by ">".
//
//   * If the XML element contains a sub-element whose name
//      matches a struct field whose tag is neither "attr" nor "chardata",
//      Unmarshal maps the sub-element to that struct field.
//      Otherwise, if the struct has a field named Any, unmarshal
//      maps the sub-element to that struct field.
//
// Unmarshal maps an XML element to a string or []byte by saving the
// concatenation of that element's character data in the string or []byte.
//
// Unmarshal maps an XML element to a slice by extending the length
// of the slice and mapping the element to the newly created value.
//
// Unmarshal maps an XML element to a bool by setting it to the boolean
// value represented by the string.
//
// Unmarshal maps an XML element to an integer or floating-point
// field by setting the field to the result of interpreting the string
// value in decimal.  There is no check for overflow.
//
// Unmarshal maps an XML element to an xml.Name by recording the
// element name.
//
// 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{}) os.Error {
	v, ok := reflect.NewValue(val).(*reflect.PtrValue)
	if !ok {
		return os.NewError("non-pointer passed to Unmarshal")
	}
	p := NewParser(r)
	elem := v.Elem()
	err := p.unmarshal(elem, nil)
	if err != nil {
		return err
	}
	return nil
}

// An UnmarshalError represents an error in the unmarshalling process.
type UnmarshalError string

func (e UnmarshalError) String() string { return string(e) }

// A TagPathError represents an error in the unmarshalling process
// caused by the use of field tags with conflicting paths.
type TagPathError struct {
	Struct       reflect.Type
	Field1, Tag1 string
	Field2, Tag2 string
}

func (e *TagPathError) String() string {
	return fmt.Sprintf("%s field %q with tag %q conflicts with field %q with tag %q", e.Struct, e.Field1, e.Tag1, e.Field2, e.Tag2)
}

// 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) os.Error {
	v, ok := reflect.NewValue(val).(*reflect.PtrValue)
	if !ok {
		return os.NewError("non-pointer passed to Unmarshal")
	}
	return p.unmarshal(v.Elem(), start)
}

// fieldName strips invalid characters from an XML name
// to create a valid Go struct name.  It also converts the
// name to lower case letters.
func fieldName(original string) string {

	var i int
	//remove leading underscores
	for i = 0; i < len(original) && original[i] == '_'; i++ {
	}

	return strings.Map(
		func(x int) int {
			if x == '_' || unicode.IsDigit(x) || unicode.IsLetter(x) {
				return unicode.ToLower(x)
			}
			return -1
		},
		original[i:])
}

// Unmarshal a single XML element into val.
func (p *Parser) unmarshal(val reflect.Value, start *StartElement) os.Error {
	// Find start element if we need it.
	if start == nil {
		for {
			tok, err := p.Token()
			if err != nil {
				return err
			}
			if t, ok := tok.(StartElement); ok {
				start = &t
				break
			}
		}
	}

	if pv, ok := val.(*reflect.PtrValue); ok {
		if pv.Get() == 0 {
			zv := reflect.MakeZero(pv.Type().(*reflect.PtrType).Elem())
			pv.PointTo(zv)
			val = zv
		} else {
			val = pv.Elem()
		}
	}

	var (
		data         []byte
		saveData     reflect.Value
		comment      []byte
		saveComment  reflect.Value
		saveXML      reflect.Value
		saveXMLIndex int
		saveXMLData  []byte
		sv           *reflect.StructValue
		styp         *reflect.StructType
		fieldPaths   map[string]pathInfo
	)

	switch v := val.(type) {
	default:
		return os.ErrorString("unknown type " + v.Type().String())

	case *reflect.SliceValue:
		typ := v.Type().(*reflect.SliceType)
		if typ.Elem().Kind() == reflect.Uint8 {
			// []byte
			saveData = v
			break
		}

		// Slice of element values.
		// Grow slice.
		n := v.Len()
		if n >= v.Cap() {
			ncap := 2 * n
			if ncap < 4 {
				ncap = 4
			}
			new := reflect.MakeSlice(typ, n, ncap)
			reflect.Copy(new, v)
			v.Set(new)
		}
		v.SetLen(n + 1)

		// Recur to read element into slice.
		if err := p.unmarshal(v.Elem(n), start); err != nil {
			v.SetLen(n)
			return err
		}
		return nil

	case *reflect.BoolValue, *reflect.FloatValue, *reflect.IntValue, *reflect.UintValue, *reflect.StringValue:
		saveData = v

	case *reflect.StructValue:
		if _, ok := v.Interface().(Name); ok {
			v.Set(reflect.NewValue(start.Name).(*reflect.StructValue))
			break
		}

		sv = v
		typ := sv.Type().(*reflect.StructType)
		styp = typ
		// Assign name.
		if f, ok := typ.FieldByName("XMLName"); ok {
			// Validate element name.
			if f.Tag != "" {
				tag := f.Tag
				ns := ""
				i := strings.LastIndex(tag, " ")
				if i >= 0 {
					ns, tag = tag[0:i], tag[i+1:]
				}
				if tag != start.Name.Local {
					return UnmarshalError("expected element type <" + tag + "> but have <" + start.Name.Local + ">")
				}
				if ns != "" && ns != start.Name.Space {
					e := "expected element <" + tag + "> in name space " + ns + " but have "
					if start.Name.Space == "" {
						e += "no name space"
					} else {
						e += start.Name.Space
					}
					return UnmarshalError(e)
				}
			}

			// Save
			v := sv.FieldByIndex(f.Index)
			if _, ok := v.Interface().(Name); !ok {
				return UnmarshalError(sv.Type().String() + " field XMLName does not have type xml.Name")
			}
			v.(*reflect.StructValue).Set(reflect.NewValue(start.Name).(*reflect.StructValue))
		}

		// Assign attributes.
		// Also, determine whether we need to save character data or comments.
		for i, n := 0, typ.NumField(); i < n; i++ {
			f := typ.Field(i)
			switch f.Tag {
			case "attr":
				strv, ok := sv.FieldByIndex(f.Index).(*reflect.StringValue)
				if !ok {
					return UnmarshalError(sv.Type().String() + " field " + f.Name + " has attr tag but is not type string")
				}
				// Look for attribute.
				val := ""
				k := strings.ToLower(f.Name)
				for _, a := range start.Attr {
					if fieldName(a.Name.Local) == k {
						val = a.Value
						break
					}
				}
				strv.Set(val)

			case "comment":
				if saveComment == nil {
					saveComment = sv.FieldByIndex(f.Index)
				}

			case "chardata":
				if saveData == nil {
					saveData = sv.FieldByIndex(f.Index)
				}

			case "innerxml":
				if saveXML == nil {
					saveXML = sv.FieldByIndex(f.Index)
					if p.saved == nil {
						saveXMLIndex = 0
						p.saved = new(bytes.Buffer)
					} else {
						saveXMLIndex = p.savedOffset()
					}
				}

			default:
				if strings.Contains(f.Tag, ">") {
					if fieldPaths == nil {
						fieldPaths = make(map[string]pathInfo)
					}
					path := strings.ToLower(f.Tag)
					if strings.HasPrefix(f.Tag, ">") {
						path = strings.ToLower(f.Name) + path
					}
					if strings.HasSuffix(f.Tag, ">") {
						path = path[:len(path)-1]
					}
					err := addFieldPath(sv, fieldPaths, path, f.Index)
					if err != nil {
						return err
					}
				}
			}
		}
	}

	// Find end element.
	// Process sub-elements along the way.
Loop:
	for {
		var savedOffset int
		if saveXML != nil {
			savedOffset = p.savedOffset()
		}
		tok, err := p.Token()
		if err != nil {
			return err
		}
		switch t := tok.(type) {
		case StartElement:
			// Sub-element.
			// Look up by tag name.
			if sv != nil {
				k := fieldName(t.Name.Local)

				if fieldPaths != nil {
					if _, found := fieldPaths[k]; found {
						if err := p.unmarshalPaths(sv, fieldPaths, k, &t); err != nil {
							return err
						}
						continue Loop
					}
				}

				match := func(s string) bool {
					// check if the name matches ignoring case
					if strings.ToLower(s) != k {
						return false
					}
					// now check that it's public
					c, _ := utf8.DecodeRuneInString(s)
					return unicode.IsUpper(c)
				}

				f, found := styp.FieldByNameFunc(match)
				if !found { // fall back to mop-up field named "Any"
					f, found = styp.FieldByName("Any")
				}
				if found {
					if err := p.unmarshal(sv.FieldByIndex(f.Index), &t); err != nil {
						return err
					}
					continue Loop
				}
			}
			// Not saving sub-element but still have to skip over it.
			if err := p.Skip(); err != nil {
				return err
			}

		case EndElement:
			if saveXML != nil {
				saveXMLData = p.saved.Bytes()[saveXMLIndex:savedOffset]
				if saveXMLIndex == 0 {
					p.saved = nil
				}
			}
			break Loop

		case CharData:
			if saveData != nil {
				data = append(data, t...)
			}

		case Comment:
			if saveComment != nil {
				comment = append(comment, t...)
			}
		}
	}

	var err os.Error
	// Helper functions for integer and unsigned integer conversions
	var itmp int64
	getInt64 := func() bool {
		itmp, err = strconv.Atoi64(string(data))
		// TODO: should check sizes
		return err == nil
	}
	var utmp uint64
	getUint64 := func() bool {
		utmp, err = strconv.Atoui64(string(data))
		// TODO: check for overflow?
		return err == nil
	}
	var ftmp float64
	getFloat64 := func() bool {
		ftmp, err = strconv.Atof64(string(data))
		// TODO: check for overflow?
		return err == nil
	}

	// Save accumulated data and comments
	switch t := saveData.(type) {
	case nil:
		// Probably a comment, handled below
	default:
		return os.ErrorString("cannot happen: unknown type " + t.Type().String())
	case *reflect.IntValue:
		if !getInt64() {
			return err
		}
		t.Set(itmp)
	case *reflect.UintValue:
		if !getUint64() {
			return err
		}
		t.Set(utmp)
	case *reflect.FloatValue:
		if !getFloat64() {
			return err
		}
		t.Set(ftmp)
	case *reflect.BoolValue:
		value, err := strconv.Atob(strings.TrimSpace(string(data)))
		if err != nil {
			return err
		}
		t.Set(value)
	case *reflect.StringValue:
		t.Set(string(data))
	case *reflect.SliceValue:
		t.Set(reflect.NewValue(data).(*reflect.SliceValue))
	}

	switch t := saveComment.(type) {
	case *reflect.StringValue:
		t.Set(string(comment))
	case *reflect.SliceValue:
		t.Set(reflect.NewValue(comment).(*reflect.SliceValue))
	}

	switch t := saveXML.(type) {
	case *reflect.StringValue:
		t.Set(string(saveXMLData))
	case *reflect.SliceValue:
		t.Set(reflect.NewValue(saveXMLData).(*reflect.SliceValue))
	}

	return nil
}

type pathInfo struct {
	fieldIdx []int
	complete bool
}

// addFieldPath takes an element path such as "a>b>c" and fills the
// paths map with all paths leading to it ("a", "a>b", and "a>b>c").
// It is okay for paths to share a common, shorter prefix but not ok
// for one path to itself be a prefix of another.
func addFieldPath(sv *reflect.StructValue, paths map[string]pathInfo, path string, fieldIdx []int) os.Error {
	if info, found := paths[path]; found {
		return tagError(sv, info.fieldIdx, fieldIdx)
	}
	paths[path] = pathInfo{fieldIdx, true}
	for {
		i := strings.LastIndex(path, ">")
		if i < 0 {
			break
		}
		path = path[:i]
		if info, found := paths[path]; found {
			if info.complete {
				return tagError(sv, info.fieldIdx, fieldIdx)
			}
		} else {
			paths[path] = pathInfo{fieldIdx, false}
		}
	}
	return nil

}

func tagError(sv *reflect.StructValue, idx1 []int, idx2 []int) os.Error {
	t := sv.Type().(*reflect.StructType)
	f1 := t.FieldByIndex(idx1)
	f2 := t.FieldByIndex(idx2)
	return &TagPathError{t, f1.Name, f1.Tag, f2.Name, f2.Tag}
}

// unmarshalPaths walks down an XML structure looking for
// wanted paths, and calls unmarshal on them.
func (p *Parser) unmarshalPaths(sv *reflect.StructValue, paths map[string]pathInfo, path string, start *StartElement) os.Error {
	if info, _ := paths[path]; info.complete {
		return p.unmarshal(sv.FieldByIndex(info.fieldIdx), start)
	}
	for {
		tok, err := p.Token()
		if err != nil {
			return err
		}
		switch t := tok.(type) {
		case StartElement:
			k := path + ">" + fieldName(t.Name.Local)
			if _, found := paths[k]; found {
				if err := p.unmarshalPaths(sv, paths, k, &t); err != nil {
					return err
				}
				continue
			}
			if err := p.Skip(); err != nil {
				return err
			}
		case EndElement:
			return nil
		}
	}
	panic("unreachable")
}

// Have already read a start element.
// 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() os.Error {
	for {
		tok, err := p.Token()
		if err != nil {
			return err
		}
		switch t := tok.(type) {
		case StartElement:
			if err := p.Skip(); err != nil {
				return err
			}
		case EndElement:
			return nil
		}
	}
	panic("unreachable")
}