XML Splitter for NodeJS

It's native and full Javascript class, that provides an easy way to split huge XML data with one or more paths.

Contributors

• Nicolas Thouvenin
• Stéphane Gully
• Alison Rowland

Installation

With npm do:

$ npm install xml-splitter

Examples

Basic

	var XMLSplitter = require('xml-splitter')

	xs = new XMLSplitter('/root/item')
	xs.on('data', function(data) {
        console.log(data)
    })
    xs.on('end', function(counter) {
        console.log(counter+' slices !')
    })
    xs.parseString('<root><item><id>1</id></item><item><id>2</id></item></root>')

Output:

{ id: { '$t': '1' } }
{ id: { '$t': '2' } }
2 slices !

Multi-paths

	var XMLSplitter = require('xml-splitter')

	xs = new XMLSplitter(['/root/item', '/root/entry'])
	xs.on('data', function(data) {
        console.log(data)
    })
    xs.on('end', function(counter) {
        console.log(counter+' slices !')
    })
    xs.parseString('<root><item><id>1</id></item><entry><id>2</id></entry></root>')

Output:

{ id: { '$t': '1' } }
{ id: { '$t': '2' } }
2 slices !

Streaming

    var XMLSplitter = require('xml-splitter')

	xs = new XMLSplitter('/root/item')
	xs.on('data', function(data) {
        console.log(data)
    })
    xs.on('end', function(counter) {
        console.log(counter+' slices !')
    })
    xs.parseStream(process.stdin) // or process.stdin.pipe(xs.stream)

Tests

Use nodeunit to run the tests.

$ npm install nodeunit
$ nodeunit test

API Documentation

Methods

constructor XMLSplitter(cutter, options)

Create an new splitter, cutter is a string or an array of strings that contains path. Options are :

• regular : To indicate if the cutter is applied to not nested XML parts. By default is true (to optimize the memory consumation)
• ignoreError : To NOT emit error event when an XML Error was met . By default is false.

parseString(string, encoding)

Split XML within a string

parseStream(stream)

Split XML within a stream

Events

data

Emits three elements on each slice: the data node (object), the node's tag name (string), and the node's path (string). For example:

var xs = new XMLSplitter('//(item|unit)')
xs.on('data', function (node, tag, path) {
    console.log(node);
    console.log(tag);
    console.log(path);
})
xs.parseString('<record><item><value>X</value></item><unit><value>Y</value></unit></record>')

Output:

{ value: { '$t': 'X' } }
item
/record/item
{ value: { '$t': 'Y' } }
unit
/record/unit

close

Emit if the stream emit the close event OR if the stream is destroyed

end

Emit on the end of the XML parsing

error

Emit when something bad happened

XPath's operators

The XPath standard is not supported, only basic paths (included namespaces) and fews operotors is implemented :

• / : /record, /record/item
• // : //para, /root//item
• * : /root/*/item, /root/item/*
• | : /(record|item), /root/(item|unit)

I do not think I will implement more operators.

Also

• https://github.com/jahewson/node-big-xml
• https://github.com/DamonOehlman/xmlslicer

License

MIT/X11