File: parse.go

package info (click to toggle)
golang-github-antchfx-xmlquery 1.3.3-1
  • links: PTS, VCS
  • area: main
  • in suites: bookworm, bullseye, bullseye-proposed-updates, sid
  • size: 156 kB
  • sloc: xml: 120; makefile: 2
file content (334 lines) | stat: -rw-r--r-- 10,950 bytes parent folder | download
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
package xmlquery

import (
	"bufio"
	"encoding/xml"
	"errors"
	"fmt"
	"io"
	"net/http"
	"regexp"
	"strings"

	"github.com/antchfx/xpath"
	"golang.org/x/net/html/charset"
)

var xmlMIMERegex = regexp.MustCompile(`(?i)((application|image|message|model)/((\w|\.|-)+\+?)?|text/)(wb)?xml`)

// LoadURL loads the XML document from the specified URL.
func LoadURL(url string) (*Node, error) {
	resp, err := http.Get(url)
	if err != nil {
		return nil, err
	}
	defer resp.Body.Close()
	// Make sure the Content-Type has a valid XML MIME type
	if xmlMIMERegex.MatchString(resp.Header.Get("Content-Type")) {
		return Parse(resp.Body)
	}
	return nil, fmt.Errorf("invalid XML document(%s)", resp.Header.Get("Content-Type"))
}

// Parse returns the parse tree for the XML from the given Reader.
func Parse(r io.Reader) (*Node, error) {
	p := createParser(r)
	for {
		_, err := p.parse()
		if err == io.EOF {
			return p.doc, nil
		}
		if err != nil {
			return nil, err
		}
	}
}

type parser struct {
	decoder             *xml.Decoder
	doc                 *Node
	space2prefix        map[string]string
	level               int
	prev                *Node
	streamElementXPath  *xpath.Expr   // Under streaming mode, this specifies the xpath to the target element node(s).
	streamElementFilter *xpath.Expr   // If specified, it provides further filtering on the target element.
	streamNode          *Node         // Need to remember the last target node So we can clean it up upon next Read() call.
	streamNodePrev      *Node         // Need to remember target node's prev so upon target node removal, we can restore correct prev.
	reader              *cachedReader // Need to maintain a reference to the reader, so we can determine whether a node contains CDATA.
}

func createParser(r io.Reader) *parser {
	reader := newCachedReader(bufio.NewReader(r))
	p := &parser{
		decoder:      xml.NewDecoder(reader),
		doc:          &Node{Type: DocumentNode},
		space2prefix: make(map[string]string),
		level:        0,
		reader:       reader,
	}
	// http://www.w3.org/XML/1998/namespace is bound by definition to the prefix xml.
	p.space2prefix["http://www.w3.org/XML/1998/namespace"] = "xml"
	p.decoder.CharsetReader = charset.NewReaderLabel
	p.prev = p.doc
	return p
}

func (p *parser) parse() (*Node, error) {
	var streamElementNodeCounter int

	for {
		tok, err := p.decoder.Token()
		if err != nil {
			return nil, err
		}

		switch tok := tok.(type) {
		case xml.StartElement:
			if p.level == 0 {
				// mising XML declaration
				node := &Node{Type: DeclarationNode, Data: "xml", level: 1}
				AddChild(p.prev, node)
				p.level = 1
				p.prev = node
			}
			// https://www.w3.org/TR/xml-names/#scoping-defaulting
			for _, att := range tok.Attr {
				if att.Name.Local == "xmlns" {
					p.space2prefix[att.Value] = ""
				} else if att.Name.Space == "xmlns" {
					p.space2prefix[att.Value] = att.Name.Local
				}
			}

			if tok.Name.Space != "" {
				if _, found := p.space2prefix[tok.Name.Space]; !found {
					return nil, errors.New("xmlquery: invalid XML document, namespace is missing")
				}
			}

			for i := 0; i < len(tok.Attr); i++ {
				att := &tok.Attr[i]
				if prefix, ok := p.space2prefix[att.Name.Space]; ok {
					att.Name.Space = prefix
				}
			}

			node := &Node{
				Type:         ElementNode,
				Data:         tok.Name.Local,
				Prefix:       p.space2prefix[tok.Name.Space],
				NamespaceURI: tok.Name.Space,
				Attr:         tok.Attr,
				level:        p.level,
			}

			if p.level == p.prev.level {
				AddSibling(p.prev, node)
			} else if p.level > p.prev.level {
				AddChild(p.prev, node)
			} else if p.level < p.prev.level {
				for i := p.prev.level - p.level; i > 1; i-- {
					p.prev = p.prev.Parent
				}
				AddSibling(p.prev.Parent, node)
			}
			// If we're in the streaming mode, we need to remember the node if it is the target node
			// so that when we finish processing the node's EndElement, we know how/what to return to
			// caller. Also we need to remove the target node from the tree upon next Read() call so
			// memory doesn't grow unbounded.
			if p.streamElementXPath != nil {
				if p.streamNode == nil {
					if QuerySelector(p.doc, p.streamElementXPath) != nil {
						p.streamNode = node
						p.streamNodePrev = p.prev
						streamElementNodeCounter = 1
					}
				} else {
					streamElementNodeCounter++
				}
			}
			p.prev = node
			p.level++
			p.reader.StartCaching()
		case xml.EndElement:
			p.level--
			// If we're in streaming mode, and we already have a potential streaming
			// target node identified (p.streamNode != nil) then we need to check if
			// this is the real one we want to return to caller.
			if p.streamNode != nil {
				streamElementNodeCounter--
				if streamElementNodeCounter == 0 {
					// Now we know this element node is the at least passing the initial
					// p.streamElementXPath check and is a potential target node candidate.
					// We need to have 1 more check with p.streamElementFilter (if given) to
					// ensure it is really the element node we want.
					// The reason we need a two-step check process is because the following
					// situation:
					//   <AAA><BBB>b1</BBB></AAA>
					// And say the p.streamElementXPath = "/AAA/BBB[. != 'b1']". Now during
					// xml.StartElement time, the <BBB> node is still empty, so it will pass
					// the p.streamElementXPath check. However, eventually we know this <BBB>
					// shouldn't be returned to the caller. Having a second more fine-grained
					// filter check ensures that. So in this case, the caller should really
					// setup the stream parser with:
					//   streamElementXPath = "/AAA/BBB["
					//   streamElementFilter = "/AAA/BBB[. != 'b1']"
					if p.streamElementFilter == nil || QuerySelector(p.doc, p.streamElementFilter) != nil {
						return p.streamNode, nil
					}
					// otherwise, this isn't our target node, clean things up.
					// note we also remove the underlying *Node from the node tree, to prevent
					// future stream node candidate selection error.
					RemoveFromTree(p.streamNode)
					p.prev = p.streamNodePrev
					p.streamNode = nil
					p.streamNodePrev = nil
				}
			}
		case xml.CharData:
			p.reader.StopCaching()
			// First, normalize the cache...
			cached := strings.ToUpper(string(p.reader.Cache()))
			nodeType := TextNode
			if strings.HasPrefix(cached, "<![CDATA[") {
				nodeType = CharDataNode
			}

			node := &Node{Type: nodeType, Data: string(tok), level: p.level}
			if p.level == p.prev.level {
				AddSibling(p.prev, node)
			} else if p.level > p.prev.level {
				AddChild(p.prev, node)
			} else if p.level < p.prev.level {
				for i := p.prev.level - p.level; i > 1; i-- {
					p.prev = p.prev.Parent
				}
				AddSibling(p.prev.Parent, node)
			}
			p.reader.StartCaching()
		case xml.Comment:
			node := &Node{Type: CommentNode, Data: string(tok), level: p.level}
			if p.level == p.prev.level {
				AddSibling(p.prev, node)
			} else if p.level > p.prev.level {
				AddChild(p.prev, node)
			} else if p.level < p.prev.level {
				for i := p.prev.level - p.level; i > 1; i-- {
					p.prev = p.prev.Parent
				}
				AddSibling(p.prev.Parent, node)
			}
		case xml.ProcInst: // Processing Instruction
			if p.prev.Type != DeclarationNode {
				p.level++
			}
			node := &Node{Type: DeclarationNode, Data: tok.Target, level: p.level}
			pairs := strings.Split(string(tok.Inst), " ")
			for _, pair := range pairs {
				pair = strings.TrimSpace(pair)
				if i := strings.Index(pair, "="); i > 0 {
					AddAttr(node, pair[:i], strings.Trim(pair[i+1:], `"`))
				}
			}
			if p.level == p.prev.level {
				AddSibling(p.prev, node)
			} else if p.level > p.prev.level {
				AddChild(p.prev, node)
			}
			p.prev = node
		case xml.Directive:
		}
	}
}

// StreamParser enables loading and parsing an XML document in a streaming
// fashion.
type StreamParser struct {
	p *parser
}

// CreateStreamParser creates a StreamParser. Argument streamElementXPath is
// required.
// Argument streamElementFilter is optional and should only be used in advanced
// scenarios.
//
// Scenario 1: simple case:
//  xml := `<AAA><BBB>b1</BBB><BBB>b2</BBB></AAA>`
//  sp, err := CreateStreamParser(strings.NewReader(xml), "/AAA/BBB")
//  if err != nil {
//      panic(err)
//  }
//  for {
//      n, err := sp.Read()
//      if err != nil {
//          break
//      }
//      fmt.Println(n.OutputXML(true))
//  }
// Output will be:
//   <BBB>b1</BBB>
//   <BBB>b2</BBB>
//
// Scenario 2: advanced case:
//  xml := `<AAA><BBB>b1</BBB><BBB>b2</BBB></AAA>`
//  sp, err := CreateStreamParser(strings.NewReader(xml), "/AAA/BBB", "/AAA/BBB[. != 'b1']")
//  if err != nil {
//      panic(err)
//  }
//  for {
//      n, err := sp.Read()
//      if err != nil {
//          break
//      }
//      fmt.Println(n.OutputXML(true))
//  }
// Output will be:
//   <BBB>b2</BBB>
//
// As the argument names indicate, streamElementXPath should be used for
// providing xpath query pointing to the target element node only, no extra
// filtering on the element itself or its children; while streamElementFilter,
// if needed, can provide additional filtering on the target element and its
// children.
//
// CreateStreamParser returns an error if either streamElementXPath or
// streamElementFilter, if provided, cannot be successfully parsed and compiled
// into a valid xpath query.
func CreateStreamParser(r io.Reader, streamElementXPath string, streamElementFilter ...string) (*StreamParser, error) {
	elemXPath, err := getQuery(streamElementXPath)
	if err != nil {
		return nil, fmt.Errorf("invalid streamElementXPath '%s', err: %s", streamElementXPath, err.Error())
	}
	elemFilter := (*xpath.Expr)(nil)
	if len(streamElementFilter) > 0 {
		elemFilter, err = getQuery(streamElementFilter[0])
		if err != nil {
			return nil, fmt.Errorf("invalid streamElementFilter '%s', err: %s", streamElementFilter[0], err.Error())
		}
	}
	sp := &StreamParser{
		p: createParser(r),
	}
	sp.p.streamElementXPath = elemXPath
	sp.p.streamElementFilter = elemFilter
	return sp, nil
}

// Read returns a target node that satisfies the XPath specified by caller at
// StreamParser creation time. If there is no more satisfying target nodes after
// reading the rest of the XML document, io.EOF will be returned. At any time,
// any XML parsing error encountered will be returned, and the stream parsing
// stopped. Calling Read() after an error is returned (including io.EOF) results
// undefined behavior. Also note, due to the streaming nature, calling Read()
// will automatically remove any previous target node(s) from the document tree.
func (sp *StreamParser) Read() (*Node, error) {
	// Because this is a streaming read, we need to release/remove last
	// target node from the node tree to free up memory.
	if sp.p.streamNode != nil {
		RemoveFromTree(sp.p.streamNode)
		sp.p.prev = sp.p.streamNodePrev
		sp.p.streamNode = nil
		sp.p.streamNodePrev = nil
	}
	return sp.p.parse()
}