jsonquery

README

jsonquery

Overview

jsonquery is an XPath query package for JSON document, lets you extract data from JSON documents through an XPath expression. Built-in XPath expression cache avoid re-compile XPath expression each query.

Getting Started

Install Package

go get github.com/antchfx/jsonquery

Load JSON document from URL.

doc, err := jsonquery.LoadURL("http://www.example.com/feed?json")

Load JSON document from string.

s :=`{
    "name":"John",
    "age":31, 
    "city":"New York" 
    }`
doc, err := jsonquery.Parse(strings.NewReader(s))

Load JSON document from io.Reader.

f, err := os.Open("./books.json")
doc, err := jsonquery.Parse(f)

Find authors of all books in the store.

list := jsonquery.Find(doc, "store/book/*/author")
// or equal to
list := jsonquery.Find(doc, "//author")
// or by QueryAll()
nodes, err := jsonquery.QueryAll(doc, "//a")

Find the third book.

book := jsonquery.Find(doc, "//book/*[3]")

Find the last book.

book := jsonquery.Find(doc, "//book/*[last()]")

Find all books that have an isbn number.

list := jsonquery.Find(doc, "//book/*[isbn]")

Find all books priced less than 10.

list := jsonquery.Find(doc, "//book/*[price<10]")

Examples

func main() {
	s := `{
		"name": "John",
		"age"      : 26,
		"address"  : {
		  "streetAddress": "naist street",
		  "city"         : "Nara",
		  "postalCode"   : "630-0192"
		},
		"phoneNumbers": [
		  {
			"type"  : "iPhone",
			"number": "0123-4567-8888"
		  },
		  {
			"type"  : "home",
			"number": "0123-4567-8910"
		  }
		]
	}`
	doc, err := jsonquery.Parse(strings.NewReader(s))
	if err != nil {
		panic(err)
	}
	name := jsonquery.FindOne(doc, "name")
	fmt.Printf("name: %s\n", name.InnerText())
	var a []string
	for _, n := range jsonquery.Find(doc, "phoneNumbers/*/number") {
		a = append(a, n.InnerText())
	}
	fmt.Printf("phone number: %s\n", strings.Join(a, ","))
	if n := jsonquery.FindOne(doc, "address/streetAddress"); n != nil {
		fmt.Printf("address: %s\n", n.InnerText())
	}
}

Implement Principle

If you are familiar with XPath and XML, you can easily figure out how to write your XPath expression.

{
"name":"John",
"age":30,
"cars": [
	{ "name":"Ford", "models":[ "Fiesta", "Focus", "Mustang" ] },
	{ "name":"BMW", "models":[ "320", "X3", "X5" ] },
	{ "name":"Fiat", "models":[ "500", "Panda" ] }
]
}

The above JSON document will be convert to similar to XML document by the JSONQuery, like below:

<name>John</name>
<age>30</age>
<cars>
	<element>
		<name>Ford</name>
		<models>
			<element>Fiesta</element>
			<element>Focus</element>
			<element>Mustang</element>
		</models>		
	</element>
	<element>
		<name>BMW</name>
		<models>
			<element>320</element>
			<element>X3</element>
			<element>X5</element>
		</models>		
	</element>
	<element>
		<name>Fiat</name>
		<models>
			<element>500</element>
			<element>Panda</element>
		</models>		
	</element>
</cars>

Notes: element is empty element that have no any name.

List of XPath query packages

Name	Description
htmlquery	XPath query package for the HTML document
xmlquery	XPath query package for the XML document
jsonquery	XPath query package for the JSON document

Documentation

Index

• Variables
• type Node
  • func Find(top *Node, expr string) []*Node
  • func FindOne(top *Node, expr string) *Node
  • func LoadURL(url string) (*Node, error)
  • func Parse(r io.Reader) (*Node, error)
  • func Query(top *Node, expr string) (*Node, error)
  • func QueryAll(top *Node, expr string) ([]*Node, error)
  • func QuerySelector(top *Node, selector *xpath.Expr) *Node
  • func QuerySelectorAll(top *Node, selector *xpath.Expr) []*Node
  • func (n *Node) ChildNodes() []*Node
  • func (n *Node) InnerText() string
  • func (n *Node) OutputXML() string
  • func (n *Node) SelectElement(name string) *Node
• type NodeNavigator
  • func CreateXPathNavigator(top *Node) *NodeNavigator
  • func (a *NodeNavigator) Copy() xpath.NodeNavigator
  • func (a *NodeNavigator) Current() *Node
  • func (a *NodeNavigator) LocalName() string
  • func (a *NodeNavigator) MoveTo(other xpath.NodeNavigator) bool
  • func (a *NodeNavigator) MoveToChild() bool
  • func (a *NodeNavigator) MoveToFirst() bool
  • func (a *NodeNavigator) MoveToNext() bool
  • func (x *NodeNavigator) MoveToNextAttribute() bool
  • func (a *NodeNavigator) MoveToParent() bool
  • func (a *NodeNavigator) MoveToPrevious() bool
  • func (a *NodeNavigator) MoveToRoot()
  • func (a *NodeNavigator) NodeType() xpath.NodeType
  • func (a *NodeNavigator) Prefix() string
  • func (a *NodeNavigator) String() string
  • func (a *NodeNavigator) Value() string
• type NodeType

Variables

var DisableSelectorCache = false

DisableSelectorCache will disable caching for the query selector if value is true.

var SelectorCacheMaxEntries = 50

SelectorCacheMaxEntries allows how many selector object can be caching. Default is 50. Will disable caching if SelectorCacheMaxEntries <= 0.

Types

type Node

type Node struct {
	Parent, PrevSibling, NextSibling, FirstChild, LastChild *Node

	Type NodeType
	Data string
	// contains filtered or unexported fields
}

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.

func Find

func Find(top *Node, expr string) []*Node

Find is like QueryAll but will panics if `expr` cannot be parsed.

func FindOne

func FindOne(top *Node, expr string) *Node

FindOne is like Query but will panics if `expr` cannot be parsed.

func LoadURL

func LoadURL(url string) (*Node, error)

LoadURL loads the JSON document from the specified URL.

func Parse

func Parse(r io.Reader) (*Node, error)

Parse JSON document.

func Query

func Query(top *Node, expr string) (*Node, error)

Query searches the Node that matches by the specified XPath expr, and returns first element of matched.

func QueryAll

func QueryAll(top *Node, expr string) ([]*Node, error)

QueryAll searches the Node that matches by the specified XPath expr. Return an error if the expression `expr` cannot be parsed.

func QuerySelector

func QuerySelector(top *Node, selector *xpath.Expr) *Node

QuerySelector returns the first matched XML Node by the specified XPath selector.

func QuerySelectorAll

func QuerySelectorAll(top *Node, selector *xpath.Expr) []*Node

QuerySelectorAll searches all of the Node that matches the specified XPath selectors.

func (*Node) ChildNodes

func (n *Node) ChildNodes() []*Node

ChildNodes gets all child nodes of the node.

func (*Node) InnerText

func (n *Node) InnerText() string

InnerText gets the value of the node and all its child nodes.

func (*Node) OutputXML

func (n *Node) OutputXML() string

OutputXML prints the XML string.

func (*Node) SelectElement

func (n *Node) SelectElement(name string) *Node

SelectElement finds the first of child elements with the specified name.

type NodeNavigator

type NodeNavigator struct {
	// contains filtered or unexported fields
}

NodeNavigator is for navigating JSON document.

func CreateXPathNavigator

func CreateXPathNavigator(top *Node) *NodeNavigator

CreateXPathNavigator creates a new xpath.NodeNavigator for the specified html.Node.

func (*NodeNavigator) Copy

func (a *NodeNavigator) Copy() xpath.NodeNavigator

func (*NodeNavigator) Current

func (a *NodeNavigator) Current() *Node

func (*NodeNavigator) LocalName

func (a *NodeNavigator) LocalName() string

func (*NodeNavigator) MoveTo

func (a *NodeNavigator) MoveTo(other xpath.NodeNavigator) bool

func (*NodeNavigator) MoveToChild

func (a *NodeNavigator) MoveToChild() bool

func (*NodeNavigator) MoveToFirst

func (a *NodeNavigator) MoveToFirst() bool

func (*NodeNavigator) MoveToNext

func (a *NodeNavigator) MoveToNext() bool

func (*NodeNavigator) MoveToNextAttribute

func (x *NodeNavigator) MoveToNextAttribute() bool

func (*NodeNavigator) MoveToParent

func (a *NodeNavigator) MoveToParent() bool

func (*NodeNavigator) MoveToPrevious

func (a *NodeNavigator) MoveToPrevious() bool

func (*NodeNavigator) MoveToRoot

func (a *NodeNavigator) MoveToRoot()

func (*NodeNavigator) NodeType

func (a *NodeNavigator) NodeType() xpath.NodeType

func (*NodeNavigator) Prefix

func (a *NodeNavigator) Prefix() string

func (*NodeNavigator) String

func (a *NodeNavigator) String() string

func (*NodeNavigator) Value

func (a *NodeNavigator) Value() string

type NodeType

type NodeType uint

A NodeType is the type of a Node.

const (
	// DocumentNode is a document object that, as the root of the document tree,
	// provides access to the entire XML document.
	DocumentNode NodeType = iota
	// ElementNode is an element.
	ElementNode
	// TextNode is the text content of a node.
	TextNode
)