feat: Inital commit

This commit is contained in:
2025-07-26 05:58:59 +00:00
commit 753d1c60ea
1849 changed files with 830533 additions and 0 deletions
+32
View File
@@ -0,0 +1,32 @@
# vscode
.vscode
debug
*.test
./build
# Compiled Object files, Static and Dynamic libs (Shared Objects)
*.o
*.a
*.so
# Folders
_obj
_test
# Architecture specific extensions/prefixes
*.[568vq]
[568vq].out
*.cgo1.go
*.cgo2.c
_cgo_defun.c
_cgo_gotypes.go
_cgo_export.*
_testmain.go
*.exe
*.test
*.prof
+17
View File
@@ -0,0 +1,17 @@
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
+302
View File
@@ -0,0 +1,302 @@
# xmlquery
[![Build Status](https://github.com/antchfx/xmlquery/actions/workflows/testing.yml/badge.svg)](https://github.com/antchfx/xmlquery/actions/workflows/testing.yml)
[![GoDoc](https://godoc.org/github.com/antchfx/xmlquery?status.svg)](https://godoc.org/github.com/antchfx/xmlquery)
[![Go Report Card](https://goreportcard.com/badge/github.com/antchfx/xmlquery)](https://goreportcard.com/report/github.com/antchfx/xmlquery)
# Overview
`xmlquery` is an XPath query package for XML documents, allowing you to extract
data or evaluate from XML documents with an XPath expression.
`xmlquery` has a built-in query object caching feature that caches recently used
XPATH query strings. Enabling caching can avoid recompile XPath expression for
each query.
You can visit this page to learn about the supported XPath(1.0/2.0) syntax. https://github.com/antchfx/xpath
[htmlquery](https://github.com/antchfx/htmlquery) - Package for the HTML document query.
[xmlquery](https://github.com/antchfx/xmlquery) - Package for the XML document query.
[jsonquery](https://github.com/antchfx/jsonquery) - Package for the JSON document query.
# Installation
```
$ go get github.com/antchfx/xmlquery
```
# Quick Starts
```go
import (
"github.com/antchfx/xmlquery"
)
func main(){
s := `<?xml version="1.0" encoding="UTF-8" ?>
<rss version="2.0">
<channel>
<title>W3Schools Home Page</title>
<link>https://www.w3schools.com</link>
<description>Free web building tutorials</description>
<item>
<title>RSS Tutorial</title>
<link>https://www.w3schools.com/xml/xml_rss.asp</link>
<description>New RSS tutorial on W3Schools</description>
</item>
<item>
<title>XML Tutorial</title>
<link>https://www.w3schools.com/xml</link>
<description>New XML tutorial on W3Schools</description>
</item>
</channel>
</rss>`
doc, err := xmlquery.Parse(strings.NewReader(s))
if err != nil {
panic(err)
}
channel := xmlquery.FindOne(doc, "//channel")
if n := channel.SelectElement("title"); n != nil {
fmt.Printf("title: %s\n", n.InnerText())
}
if n := channel.SelectElement("link"); n != nil {
fmt.Printf("link: %s\n", n.InnerText())
}
for i, n := range xmlquery.Find(doc, "//item/title") {
fmt.Printf("#%d %s\n", i, n.InnerText())
}
}
```
# Getting Started
### Find specified XPath query.
```go
list, err := xmlquery.QueryAll(doc, "a")
if err != nil {
panic(err)
}
```
#### Parse an XML from URL.
```go
doc, err := xmlquery.LoadURL("http://www.example.com/sitemap.xml")
```
#### Parse an XML from string.
```go
s := `<?xml version="1.0" encoding="utf-8"?><rss version="2.0"></rss>`
doc, err := xmlquery.Parse(strings.NewReader(s))
```
#### Parse an XML from io.Reader.
```go
f, err := os.Open("../books.xml")
doc, err := xmlquery.Parse(f)
```
#### Parse an XML in a stream fashion (simple case without elements filtering).
```go
f, _ := os.Open("../books.xml")
p, err := xmlquery.CreateStreamParser(f, "/bookstore/book")
for {
n, err := p.Read()
if err == io.EOF {
break
}
if err != nil {
panic(err)
}
fmt.Println(n)
}
```
Notes: `CreateStreamParser()` used for saving memory if your had a large XML file to parse.
#### Parse an XML in a stream fashion (simple case advanced element filtering).
```go
f, _ := os.Open("../books.xml")
p, err := xmlquery.CreateStreamParser(f, "/bookstore/book", "/bookstore/book[price>=10]")
for {
n, err := p.Read()
if err == io.EOF {
break
}
if err != nil {
panic(err)
}
fmt.Println(n)
}
```
#### Find authors of all books in the bookstore.
```go
list := xmlquery.Find(doc, "//book//author")
// or
list := xmlquery.Find(doc, "//author")
```
#### Find the second book.
```go
book := xmlquery.FindOne(doc, "//book[2]")
```
#### Find the last book.
```go
book := xmlquery.FindOne(doc, "//book[last()]")
```
#### Find all book elements and only get `id` attribute.
```go
list := xmlquery.Find(doc,"//book/@id")
fmt.Println(list[0].InnerText) // outout @id value
```
#### Find all books with id `bk104`.
```go
list := xmlquery.Find(doc, "//book[@id='bk104']")
```
#### Find all books with price less than 5.
```go
list := xmlquery.Find(doc, "//book[price<5]")
```
#### Evaluate total price of all books.
```go
expr, err := xpath.Compile("sum(//book/price)")
price := expr.Evaluate(xmlquery.CreateXPathNavigator(doc)).(float64)
fmt.Printf("total price: %f\n", price)
```
#### Count the number of books.
```go
expr, err := xpath.Compile("count(//book)")
count := expr.Evaluate(xmlquery.CreateXPathNavigator(doc)).(float64)
```
#### Calculate the total price of all book prices.
```go
expr, err := xpath.Compile("sum(//book/price)")
price := expr.Evaluate(xmlquery.CreateXPathNavigator(doc)).(float64)
```
# Advanced Features
### Parse `UTF-16` XML file with `ParseWithOptions()`.
```go
f, _ := os.Open(`UTF-16.XML`)
// Convert UTF-16 XML to UTF-8
utf16ToUtf8Transformer := unicode.UTF16(unicode.LittleEndian, unicode.IgnoreBOM).NewDecoder()
utf8Reader := transform.NewReader(f, utf16ToUtf8Transformer)
// Sets `CharsetReader`
options := xmlquery.ParserOptions{
Decoder: &xmlquery.DecoderOptions{
CharsetReader: func(charset string, input io.Reader) (io.Reader, error) {
return input, nil
},
},
}
doc, err := xmlquery.ParseWithOptions(utf8Reader, options)
```
### Query with custom namespace prefix.
```go
s := `<?xml version="1.0" encoding="UTF-8"?>
<pd:ProcessDefinition xmlns:pd="http://xmlns.xyz.com/process/2003" xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<pd:activity name="Invoke Request-Response Service">
<pd:type>RequestReplyActivity</pd:type>
<pd:resourceType>OpClientReqActivity</pd:resourceType>
<pd:x>300</pd:x>
<pd:y>80</pd:y>
</pd:activity>
</pd:ProcessDefinition>`
nsMap := map[string]string{
"q": "http://xmlns.xyz.com/process/2003",
"r": "http://www.w3.org/1999/XSL/Transform",
"s": "http://www.w3.org/2001/XMLSchema",
}
expr, _ := xpath.CompileWithNS("//q:activity", nsMap)
node := xmlquery.QuerySelector(doc, expr)
```
#### Create XML document without call `xml.Marshal`.
```go
doc := &xmlquery.Node{
Type: xmlquery.DeclarationNode,
Data: "xml",
Attr: []xml.Attr{
xml.Attr{Name: xml.Name{Local: "version"}, Value: "1.0"},
},
}
root := &xmlquery.Node{
Data: "rss",
Type: xmlquery.ElementNode,
}
doc.FirstChild = root
channel := &xmlquery.Node{
Data: "channel",
Type: xmlquery.ElementNode,
}
root.FirstChild = channel
title := &xmlquery.Node{
Data: "title",
Type: xmlquery.ElementNode,
}
title_text := &xmlquery.Node{
Data: "W3Schools Home Page",
Type: xmlquery.TextNode,
}
title.FirstChild = title_text
channel.FirstChild = title
fmt.Println(doc.OutputXML(true))
fmt.Println(doc.OutputXMLWithOptions(WithOutputSelf()))
```
Output:
```xml
<?xml version="1.0"?><rss><channel><title>W3Schools Home Page</title></channel></rss>
```
# FAQ
#### `Find()` vs `QueryAll()`, which is better?
`Find` and `QueryAll` both do the same thing: searches all of matched XML nodes.
`Find` panics if provided with an invalid XPath query, while `QueryAll` returns
an error.
#### Can I save my query expression object for the next query?
Yes, you can. We provide `QuerySelector` and `QuerySelectorAll` methods; they
accept your query expression object.
Caching a query expression object avoids recompiling the XPath query
expression, improving query performance.
# Questions
Please let me know if you have any questions
+43
View File
@@ -0,0 +1,43 @@
package xmlquery
import (
"sync"
"github.com/golang/groupcache/lru"
"github.com/antchfx/xpath"
)
// DisableSelectorCache will disable caching for the query selector if value is true.
var DisableSelectorCache = false
// SelectorCacheMaxEntries allows how many selector object can be caching. Default is 50.
// Will disable caching if SelectorCacheMaxEntries <= 0.
var SelectorCacheMaxEntries = 50
var (
cacheOnce sync.Once
cache *lru.Cache
cacheMutex sync.Mutex
)
func getQuery(expr string) (*xpath.Expr, error) {
if DisableSelectorCache || SelectorCacheMaxEntries <= 0 {
return xpath.Compile(expr)
}
cacheOnce.Do(func() {
cache = lru.New(SelectorCacheMaxEntries)
})
cacheMutex.Lock()
defer cacheMutex.Unlock()
if v, ok := cache.Get(expr); ok {
return v.(*xpath.Expr), nil
}
v, err := xpath.Compile(expr)
if err != nil {
return nil, err
}
cache.Add(expr, v)
return v, nil
}
+79
View File
@@ -0,0 +1,79 @@
package xmlquery
import (
"bufio"
)
type cachedReader struct {
buffer *bufio.Reader
cache []byte
caching bool
}
func newCachedReader(r *bufio.Reader) *cachedReader {
return &cachedReader{
buffer: r,
cache: make([]byte, 0, 4096),
caching: false,
}
}
func (c *cachedReader) StartCaching() {
c.cache = c.cache[:0]
c.caching = true
}
func (c *cachedReader) ReadByte() (b byte, err error) {
b, err = c.buffer.ReadByte()
if err != nil {
return
}
if c.caching {
c.cacheByte(b)
}
return
}
func (c *cachedReader) Cache() []byte {
return c.cache
}
func (c *cachedReader) CacheWithLimit(n int) []byte {
if n < 1 {
return nil
}
l := len(c.cache)
if n > l {
n = l
}
return c.cache[:n]
}
func (c *cachedReader) StopCaching() {
c.caching = false
}
func (c *cachedReader) Read(p []byte) (int, error) {
n, err := c.buffer.Read(p)
if err != nil {
return n, err
}
if c.caching {
for i := 0; i < n; i++ {
if !c.cacheByte(p[i]) {
break
}
}
}
return n, err
}
func (c *cachedReader) cacheByte(b byte) bool {
n := len(c.cache)
if n == cap(c.cache) {
return false
}
c.cache = c.cache[:n+1]
c.cache[n] = b
return true
}
+477
View File
@@ -0,0 +1,477 @@
package xmlquery
import (
"bufio"
"encoding/xml"
"fmt"
"html"
"io"
"strings"
)
// A NodeType is the type of a Node.
type NodeType uint
const (
// DocumentNode is a document object that, as the root of the document tree,
// provides access to the entire XML document.
DocumentNode NodeType = iota
// DeclarationNode is the document type declaration, indicated by the
// following tag (for example, <!DOCTYPE...> ).
DeclarationNode
// ElementNode is an element (for example, <item> ).
ElementNode
// TextNode is the text content of a node.
TextNode
// CharDataNode node <![CDATA[content]]>
CharDataNode
// CommentNode a comment (for example, <!-- my comment --> ).
CommentNode
// AttributeNode is an attribute of element.
AttributeNode
// NotationNode is a directive represents in document (for example, <!text...>).
NotationNode
)
type Attr struct {
Name xml.Name
Value string
NamespaceURI string
}
// A Node consists of a NodeType and some Data (tag name for
// element nodes, content for text) and are part of a tree of Nodes.
type Node struct {
Parent, FirstChild, LastChild, PrevSibling, NextSibling *Node
Type NodeType
Data string
Prefix string
NamespaceURI string
Attr []Attr
level int // node level in the tree
}
type outputConfiguration struct {
printSelf bool
preserveSpaces bool
emptyElementTagSupport bool
skipComments bool
useIndentation string
}
type OutputOption func(*outputConfiguration)
// WithOutputSelf configures the Node to print the root node itself
func WithOutputSelf() OutputOption {
return func(oc *outputConfiguration) {
oc.printSelf = true
}
}
// WithEmptyTagSupport empty tags should be written as <empty/> and
// not as <empty></empty>
func WithEmptyTagSupport() OutputOption {
return func(oc *outputConfiguration) {
oc.emptyElementTagSupport = true
}
}
// WithoutComments will skip comments in output
func WithoutComments() OutputOption {
return func(oc *outputConfiguration) {
oc.skipComments = true
}
}
// WithPreserveSpace will preserve spaces in output
func WithPreserveSpace() OutputOption {
return func(oc *outputConfiguration) {
oc.preserveSpaces = true
}
}
// WithoutPreserveSpace will not preserve spaces in output
func WithoutPreserveSpace() OutputOption {
return func(oc *outputConfiguration) {
oc.preserveSpaces = false
}
}
// WithIndentation sets the indentation string used for formatting the output.
func WithIndentation(indentation string) OutputOption {
return func(oc *outputConfiguration) {
oc.useIndentation = indentation
}
}
func newXMLName(name string) xml.Name {
if i := strings.IndexByte(name, ':'); i > 0 {
return xml.Name{
Space: name[:i],
Local: name[i+1:],
}
}
return xml.Name{
Local: name,
}
}
func (n *Node) Level() int {
return n.level
}
// InnerText returns the text between the start and end tags of the object.
func (n *Node) InnerText() string {
var output func(*strings.Builder, *Node)
output = func(b *strings.Builder, n *Node) {
switch n.Type {
case TextNode, CharDataNode:
b.WriteString(n.Data)
case CommentNode:
default:
for child := n.FirstChild; child != nil; child = child.NextSibling {
output(b, child)
}
}
}
var b strings.Builder
output(&b, n)
return b.String()
}
func (n *Node) sanitizedData(preserveSpaces bool) string {
if preserveSpaces {
return n.Data
}
return strings.TrimSpace(n.Data)
}
func calculatePreserveSpaces(n *Node, pastValue bool) bool {
if attr := n.SelectAttr("xml:space"); attr == "preserve" {
return true
} else if attr == "default" {
return false
}
return pastValue
}
type indentation struct {
level int
hasChild bool
indent string
w io.Writer
}
func newIndentation(indent string, w io.Writer) *indentation {
if indent == "" {
return nil
}
return &indentation{
indent: indent,
w: w,
}
}
func (i *indentation) NewLine() (err error) {
if i == nil {
return
}
_, err = io.WriteString(i.w, "\n")
return
}
func (i *indentation) Open() (err error) {
if i == nil {
return
}
if err = i.writeIndent(); err != nil {
return
}
i.level++
i.hasChild = false
return
}
func (i *indentation) Close() (err error) {
if i == nil {
return
}
i.level--
if i.hasChild {
if err = i.writeIndent(); err != nil {
return
}
}
i.hasChild = true
return
}
func (i *indentation) writeIndent() (err error) {
_, err = io.WriteString(i.w, "\n")
if err != nil {
return
}
_, err = io.WriteString(i.w, strings.Repeat(i.indent, i.level))
return
}
func outputXML(w io.Writer, n *Node, preserveSpaces bool, config *outputConfiguration, indent *indentation) (err error) {
preserveSpaces = calculatePreserveSpaces(n, preserveSpaces)
switch n.Type {
case TextNode:
_, err = io.WriteString(w, html.EscapeString(n.sanitizedData(preserveSpaces)))
return
case CharDataNode:
_, err = fmt.Fprintf(w, "<![CDATA[%v]]>", n.Data)
return
case CommentNode:
if !config.skipComments {
_, err = fmt.Fprintf(w, "<!--%v-->", n.Data)
}
return
case NotationNode:
if err = indent.NewLine(); err != nil {
return
}
_, err = fmt.Fprintf(w, "<!%s>", n.Data)
return
case DeclarationNode:
_, err = io.WriteString(w, "<?"+n.Data)
if err != nil {
return
}
default:
if err = indent.Open(); err != nil {
return
}
if n.Prefix == "" {
_, err = io.WriteString(w, "<"+n.Data)
} else {
_, err = fmt.Fprintf(w, "<%s:%s", n.Prefix, n.Data)
}
if err != nil {
return
}
}
for _, attr := range n.Attr {
if attr.Name.Space != "" {
_, err = fmt.Fprintf(w, ` %s:%s=`, attr.Name.Space, attr.Name.Local)
} else {
_, err = fmt.Fprintf(w, ` %s=`, attr.Name.Local)
}
if err != nil {
return
}
_, err = fmt.Fprintf(w, `"%v"`, html.EscapeString(attr.Value))
if err != nil {
return
}
}
if n.Type == DeclarationNode {
_, err = io.WriteString(w, "?>")
} else {
if n.FirstChild != nil || !config.emptyElementTagSupport {
_, err = io.WriteString(w, ">")
} else {
_, err = io.WriteString(w, "/>")
if err != nil {
return
}
err = indent.Close()
return
}
}
if err != nil {
return
}
for child := n.FirstChild; child != nil; child = child.NextSibling {
err = outputXML(w, child, preserveSpaces, config, indent)
if err != nil {
return
}
}
if n.Type != DeclarationNode {
if err = indent.Close(); err != nil {
return
}
if n.Prefix == "" {
_, err = fmt.Fprintf(w, "</%s>", n.Data)
} else {
_, err = fmt.Fprintf(w, "</%s:%s>", n.Prefix, n.Data)
}
}
return
}
// OutputXML returns the text that including tags name.
func (n *Node) OutputXML(self bool) string {
if self {
return n.OutputXMLWithOptions(WithOutputSelf())
}
return n.OutputXMLWithOptions()
}
// OutputXMLWithOptions returns the text that including tags name.
func (n *Node) OutputXMLWithOptions(opts ...OutputOption) string {
var b strings.Builder
n.WriteWithOptions(&b, opts...)
return b.String()
}
// Write writes xml to given writer.
func (n *Node) Write(writer io.Writer, self bool) error {
if self {
return n.WriteWithOptions(writer, WithOutputSelf())
}
return n.WriteWithOptions(writer)
}
// WriteWithOptions writes xml with given options to given writer.
func (n *Node) WriteWithOptions(writer io.Writer, opts ...OutputOption) (err error) {
config := &outputConfiguration{
preserveSpaces: true,
}
// Set the options
for _, opt := range opts {
opt(config)
}
pastPreserveSpaces := config.preserveSpaces
preserveSpaces := calculatePreserveSpaces(n, pastPreserveSpaces)
b := bufio.NewWriter(writer)
defer b.Flush()
ident := newIndentation(config.useIndentation, b)
if config.printSelf && n.Type != DocumentNode {
err = outputXML(b, n, preserveSpaces, config, ident)
} else {
for n := n.FirstChild; n != nil; n = n.NextSibling {
err = outputXML(b, n, preserveSpaces, config, ident)
if err != nil {
break
}
}
}
return
}
// AddAttr adds a new attribute specified by 'key' and 'val' to a node 'n'.
func AddAttr(n *Node, key, val string) {
attr := Attr{
Name: newXMLName(key),
Value: val,
}
n.Attr = append(n.Attr, attr)
}
// SetAttr allows an attribute value with the specified name to be changed.
// If the attribute did not previously exist, it will be created.
func (n *Node) SetAttr(key, value string) {
name := newXMLName(key)
for i, attr := range n.Attr {
if attr.Name == name {
n.Attr[i].Value = value
return
}
}
AddAttr(n, key, value)
}
// RemoveAttr removes the attribute with the specified name.
func (n *Node) RemoveAttr(key string) {
name := newXMLName(key)
for i, attr := range n.Attr {
if attr.Name == name {
n.Attr = append(n.Attr[:i], n.Attr[i+1:]...)
return
}
}
}
// AddChild adds a new node 'n' to a node 'parent' as its last child.
func AddChild(parent, n *Node) {
n.Parent = parent
n.NextSibling = nil
if parent.FirstChild == nil {
parent.FirstChild = n
n.PrevSibling = nil
} else {
parent.LastChild.NextSibling = n
n.PrevSibling = parent.LastChild
}
parent.LastChild = n
}
// AddSibling adds a new node 'n' as a last node of sibling chain for a given node 'sibling'.
func AddSibling(sibling, n *Node) {
for t := sibling.NextSibling; t != nil; t = t.NextSibling {
sibling = t
}
n.Parent = sibling.Parent
sibling.NextSibling = n
n.PrevSibling = sibling
n.NextSibling = nil
if sibling.Parent != nil {
sibling.Parent.LastChild = n
}
}
// AddImmediateSibling adds a new node 'n' as immediate sibling a given node 'sibling'.
func AddImmediateSibling(sibling, n *Node) {
n.Parent = sibling.Parent
n.NextSibling = sibling.NextSibling
sibling.NextSibling = n
n.PrevSibling = sibling
if n.NextSibling != nil {
n.NextSibling.PrevSibling = n
} else if n.Parent != nil {
sibling.Parent.LastChild = n
}
}
// RemoveFromTree removes a node and its subtree from the document
// tree it is in. If the node is the root of the tree, then it's no-op.
func RemoveFromTree(n *Node) {
if n.Parent == nil {
return
}
if n.Parent.FirstChild == n {
if n.Parent.LastChild == n {
n.Parent.FirstChild = nil
n.Parent.LastChild = nil
} else {
n.Parent.FirstChild = n.NextSibling
n.NextSibling.PrevSibling = nil
}
} else {
if n.Parent.LastChild == n {
n.Parent.LastChild = n.PrevSibling
n.PrevSibling.NextSibling = nil
} else {
n.PrevSibling.NextSibling = n.NextSibling
n.NextSibling.PrevSibling = n.PrevSibling
}
}
n.Parent = nil
n.PrevSibling = nil
n.NextSibling = nil
}
// GetRoot returns a root of the tree where 'n' is a node.
func GetRoot(n *Node) *Node {
if n == nil {
return nil
}
root := n
for root.Parent != nil {
root = root.Parent
}
return root
}
+33
View File
@@ -0,0 +1,33 @@
package xmlquery
import (
"encoding/xml"
"io"
)
type ParserOptions struct {
Decoder *DecoderOptions
}
func (options ParserOptions) apply(parser *parser) {
if options.Decoder != nil {
(*options.Decoder).apply(parser.decoder)
}
}
// DecoderOptions implement the very same options than the standard
// encoding/xml package. Please refer to this documentation:
// https://golang.org/pkg/encoding/xml/#Decoder
type DecoderOptions struct {
Strict bool
AutoClose []string
Entity map[string]string
CharsetReader func(charset string, input io.Reader) (io.Reader, error)
}
func (options DecoderOptions) apply(decoder *xml.Decoder) {
decoder.Strict = options.Strict
decoder.AutoClose = options.AutoClose
decoder.Entity = options.Entity
decoder.CharsetReader = options.CharsetReader
}
+430
View File
@@ -0,0 +1,430 @@
package xmlquery
import (
"bufio"
"bytes"
"encoding/xml"
"fmt"
"io"
"net/http"
"regexp"
"strings"
"sync"
"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) {
return ParseWithOptions(r, ParserOptions{})
}
// ParseWithOptions is like parse, but with custom options
func ParseWithOptions(r io.Reader, options ParserOptions) (*Node, error) {
p := createParser(r)
options.apply(p)
var err error
for err == nil {
_, err = p.parse()
}
if err == io.EOF {
// additional check for validity
// according to: https://www.w3.org/TR/xml
// the document MUST contain at least ONE element
valid := false
for doc := p.doc; doc != nil; doc = doc.NextSibling {
for node := doc.FirstChild; node != nil; node = node.NextSibling {
if node.Type == ElementNode {
valid = true
break
}
}
}
if !valid {
return nil, fmt.Errorf("xmlquery: invalid XML document")
}
return p.doc, nil
}
return nil, err
}
type parser struct {
decoder *xml.Decoder
doc *Node
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.
once sync.Once
space2prefix map[string]*xmlnsPrefix
}
type xmlnsPrefix struct {
name string
level int
}
func createParser(r io.Reader) *parser {
reader := newCachedReader(bufio.NewReader(r))
p := &parser{
decoder: xml.NewDecoder(reader),
doc: &Node{Type: DocumentNode},
level: 0,
reader: reader,
}
if p.decoder.CharsetReader == nil {
p.decoder.CharsetReader = charset.NewReaderLabel
}
p.prev = p.doc
return p
}
func (p *parser) parse() (*Node, error) {
p.once.Do(func() {
p.space2prefix = map[string]*xmlnsPrefix{"http://www.w3.org/XML/1998/namespace": {name: "xml", level: 0}}
})
var streamElementNodeCounter int
for {
p.reader.StartCaching()
tok, err := p.decoder.Token()
p.reader.StopCaching()
if err != nil {
return nil, err
}
switch tok := tok.(type) {
case xml.StartElement:
if p.level == 0 {
// mising XML declaration
attributes := make([]Attr, 1)
attributes[0].Name = xml.Name{Local: "version"}
attributes[0].Value = "1.0"
node := &Node{
Type: DeclarationNode,
Data: "xml",
Attr: attributes,
level: 1,
}
AddChild(p.prev, node)
p.level = 1
p.prev = node
}
for _, att := range tok.Attr {
if att.Name.Local == "xmlns" {
// https://github.com/antchfx/xmlquery/issues/67
if prefix, ok := p.space2prefix[att.Value]; !ok || (ok && prefix.level >= p.level) {
p.space2prefix[att.Value] = &xmlnsPrefix{name: "", level: p.level} // reset empty if exist the default namespace
}
} else if att.Name.Space == "xmlns" {
// maybe there are have duplicate NamespaceURL?
p.space2prefix[att.Value] = &xmlnsPrefix{name: att.Name.Local, level: p.level}
}
}
if space := tok.Name.Space; space != "" {
if _, found := p.space2prefix[space]; !found && p.decoder.Strict {
return nil, fmt.Errorf("xmlquery: invalid XML document, namespace %s is missing", space)
}
}
attributes := make([]Attr, len(tok.Attr))
for i, att := range tok.Attr {
name := att.Name
if prefix, ok := p.space2prefix[name.Space]; ok {
name.Space = prefix.name
}
attributes[i] = Attr{
Name: name,
Value: att.Value,
NamespaceURI: att.Name.Space,
}
}
node := &Node{
Type: ElementNode,
Data: tok.Name.Local,
NamespaceURI: tok.Name.Space,
Attr: attributes,
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 node.NamespaceURI != "" {
if v, ok := p.space2prefix[node.NamespaceURI]; ok {
cached := string(p.reader.CacheWithLimit(len(v.name) + len(node.Data) + 2))
if strings.HasPrefix(cached, fmt.Sprintf("%s:%s", v.name, node.Data)) || strings.HasPrefix(cached, fmt.Sprintf("<%s:%s", v.name, node.Data)) {
node.Prefix = v.name
}
}
}
// 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++
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:
// First, normalize the cache...
cached := bytes.ToUpper(p.reader.CacheWithLimit(9))
nodeType := TextNode
if bytes.HasPrefix(cached, []byte("<![CDATA[")) || bytes.HasPrefix(cached, []byte("![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)
}
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)
} 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.prev = node
case xml.Directive:
node := &Node{Type: NotationNode, 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)
}
}
}
}
// 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) {
return CreateStreamParserWithOptions(r, ParserOptions{}, streamElementXPath, streamElementFilter...)
}
// CreateStreamParserWithOptions is like CreateStreamParser, but with custom options
func CreateStreamParserWithOptions(
r io.Reader,
options ParserOptions,
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())
}
}
parser := createParser(r)
options.apply(parser)
sp := &StreamParser{
p: parser,
}
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 {
// We need to remove all siblings before the current stream node,
// because the document may contain unwanted nodes between the target
// ones (for example new line text node), which would otherwise
// accumulate as first childs, and slow down the stream over time
for sp.p.streamNode.PrevSibling != nil {
RemoveFromTree(sp.p.streamNode.PrevSibling)
}
sp.p.prev = sp.p.streamNode.Parent
RemoveFromTree(sp.p.streamNode)
sp.p.streamNode = nil
sp.p.streamNodePrev = nil
}
return sp.p.parse()
}
+304
View File
@@ -0,0 +1,304 @@
/*
Package xmlquery provides extract data from XML documents using XPath expression.
*/
package xmlquery
import (
"fmt"
"strings"
"github.com/antchfx/xpath"
)
// SelectElements finds child elements with the specified name.
func (n *Node) SelectElements(name string) []*Node {
return Find(n, name)
}
// SelectElement finds child elements with the specified name.
func (n *Node) SelectElement(name string) *Node {
return FindOne(n, name)
}
// SelectAttr returns the attribute value with the specified name.
func (n *Node) SelectAttr(name string) string {
if n.Type == AttributeNode {
if n.Data == name {
return n.InnerText()
}
return ""
}
xmlName := newXMLName(name)
for _, attr := range n.Attr {
if attr.Name == xmlName {
return attr.Value
}
}
return ""
}
var _ xpath.NodeNavigator = &NodeNavigator{}
// CreateXPathNavigator creates a new xpath.NodeNavigator for the specified
// XML Node.
func CreateXPathNavigator(top *Node) *NodeNavigator {
return &NodeNavigator{curr: top, root: top, attr: -1}
}
func getCurrentNode(it *xpath.NodeIterator) *Node {
n := it.Current().(*NodeNavigator)
if n.NodeType() == xpath.AttributeNode {
childNode := &Node{
Type: TextNode,
Data: n.Value(),
}
return &Node{
Parent: n.curr,
Type: AttributeNode,
Data: n.LocalName(),
FirstChild: childNode,
LastChild: childNode,
}
}
return n.curr
}
// Find is like QueryAll but panics if `expr` is not a valid XPath expression.
// See `QueryAll()` function.
func Find(top *Node, expr string) []*Node {
nodes, err := QueryAll(top, expr)
if err != nil {
panic(err)
}
return nodes
}
// FindOne is like Query but panics if `expr` is not a valid XPath expression.
// See `Query()` function.
func FindOne(top *Node, expr string) *Node {
node, err := Query(top, expr)
if err != nil {
panic(err)
}
return node
}
// QueryAll searches the XML Node that matches by the specified XPath expr.
// Returns an error if the expression `expr` cannot be parsed.
func QueryAll(top *Node, expr string) ([]*Node, error) {
exp, err := getQuery(expr)
if err != nil {
return nil, err
}
return QuerySelectorAll(top, exp), nil
}
// Query searches the XML Node that matches by the specified XPath expr,
// and returns first matched element.
func Query(top *Node, expr string) (*Node, error) {
exp, err := getQuery(expr)
if err != nil {
return nil, err
}
return QuerySelector(top, exp), nil
}
// QuerySelectorAll searches all of the XML Node that matches the specified
// XPath selectors.
func QuerySelectorAll(top *Node, selector *xpath.Expr) []*Node {
t := selector.Select(CreateXPathNavigator(top))
var elems []*Node
for t.MoveNext() {
elems = append(elems, getCurrentNode(t))
}
return elems
}
// QuerySelector returns the first matched XML Node by the specified XPath
// selector.
func QuerySelector(top *Node, selector *xpath.Expr) *Node {
t := selector.Select(CreateXPathNavigator(top))
if t.MoveNext() {
return getCurrentNode(t)
}
return nil
}
// FindEach searches the html.Node and calls functions cb.
// Important: this method is deprecated, instead, use for .. = range Find(){}.
func FindEach(top *Node, expr string, cb func(int, *Node)) {
for i, n := range Find(top, expr) {
cb(i, n)
}
}
// FindEachWithBreak functions the same as FindEach but allows to break the loop
// by returning false from the callback function `cb`.
// Important: this method is deprecated, instead, use .. = range Find(){}.
func FindEachWithBreak(top *Node, expr string, cb func(int, *Node) bool) {
for i, n := range Find(top, expr) {
if !cb(i, n) {
break
}
}
}
type NodeNavigator struct {
root, curr *Node
attr int
}
func (x *NodeNavigator) Current() *Node {
return x.curr
}
func (x *NodeNavigator) NodeType() xpath.NodeType {
switch x.curr.Type {
case CommentNode:
return xpath.CommentNode
case TextNode, CharDataNode, NotationNode:
return xpath.TextNode
case DeclarationNode, DocumentNode:
return xpath.RootNode
case ElementNode:
if x.attr != -1 {
return xpath.AttributeNode
}
return xpath.ElementNode
}
panic(fmt.Sprintf("unknown XML node type: %v", x.curr.Type))
}
func (x *NodeNavigator) LocalName() string {
if x.attr != -1 {
return x.curr.Attr[x.attr].Name.Local
}
return x.curr.Data
}
func (x *NodeNavigator) Prefix() string {
if x.NodeType() == xpath.AttributeNode {
if x.attr != -1 {
return x.curr.Attr[x.attr].Name.Space
}
return ""
}
return x.curr.Prefix
}
func (x *NodeNavigator) NamespaceURL() string {
if x.attr != -1 {
return x.curr.Attr[x.attr].NamespaceURI
}
return x.curr.NamespaceURI
}
func (x *NodeNavigator) Value() string {
switch x.curr.Type {
case CommentNode:
return x.curr.Data
case ElementNode:
if x.attr != -1 {
return x.curr.Attr[x.attr].Value
}
return x.curr.InnerText()
case TextNode:
return x.curr.Data
}
return ""
}
func (x *NodeNavigator) Copy() xpath.NodeNavigator {
n := *x
return &n
}
func (x *NodeNavigator) MoveToRoot() {
x.curr = x.root
}
func (x *NodeNavigator) MoveToParent() bool {
if x.attr != -1 {
x.attr = -1
return true
} else if node := x.curr.Parent; node != nil {
x.curr = node
return true
}
return false
}
func (x *NodeNavigator) MoveToNextAttribute() bool {
if x.attr >= len(x.curr.Attr)-1 {
return false
}
x.attr++
return true
}
func (x *NodeNavigator) MoveToChild() bool {
if x.attr != -1 {
return false
}
if node := x.curr.FirstChild; node != nil {
x.curr = node
return true
}
return false
}
func (x *NodeNavigator) MoveToFirst() bool {
if x.attr != -1 || x.curr.PrevSibling == nil {
return false
}
for {
node := x.curr.PrevSibling
if node == nil {
break
}
x.curr = node
}
return true
}
func (x *NodeNavigator) String() string {
return x.Value()
}
func (x *NodeNavigator) MoveToNext() bool {
if x.attr != -1 {
return false
}
for node := x.curr.NextSibling; node != nil; node = x.curr.NextSibling {
x.curr = node
if x.curr.Type != TextNode || strings.TrimSpace(x.curr.Data) != "" {
return true
}
}
return false
}
func (x *NodeNavigator) MoveToPrevious() bool {
if x.attr != -1 {
return false
}
for node := x.curr.PrevSibling; node != nil; node = x.curr.PrevSibling {
x.curr = node
if x.curr.Type != TextNode || strings.TrimSpace(x.curr.Data) != "" {
return true
}
}
return false
}
func (x *NodeNavigator) MoveTo(other xpath.NodeNavigator) bool {
node, ok := other.(*NodeNavigator)
if !ok || node.root != x.root {
return false
}
x.curr = node.curr
x.attr = node.attr
return true
}