It's native javascript implementation of a bidirectional converter between XML and JS data structure (aka JSON). You can convert any type of XML documents in an Javascript data structure. You can also do the reverse, converting a Javascript data structure in XML String. XML is still valid.
This is a fork of https://github.com/lindory-project/node-xml-mapping with improvements made to allow optional throwing of parser errors rather than the default of tossing them away.
With npm do:
$ npm install git+https://github.com/zacronos/node-xml-mapping.git
var xm = require('xml-mapping');
// will not throw errors -- ignores them instead
var json = xm.load('<key>value</key>');
var xml = xm.dump(json);
console.log(xml);
console.log(json);
try {
// throws parsing errors with line and column numbers
json = xm.load('<key>value', {throwErrors: true});
} catch (err) {
console.log(err.toString())
}Output:
<key>value</key>
{ key: { '$t': 'value' } }
Error: Unexpected end of input
Line: 0
Column: 10
Char: null
The rules for converting XML to JSON are those used by Google in its GData protocol. More information here : http://code.google.com/apis/gdata/docs/json.html
Use nodeunit to run the tests.
$ npm install nodeunit
$ nodeunit test
Transform a string with XML in Javascript data structure (JSON). Return Object.
Transform a Javascript data structure (JSON) in XML string. Return String.
Alias of load.
Alias of dump.
- It would be nice allow passthrough of other options to SAX and XMLWriter.
- The code could be more efficient by instantiating a single parser/writer and allowing it to be reused, rather than always starting from scratch.
- It would be nice to expose SAX's incremental parsing / data buffering mechanism.
- The existing code is not reversible for html-escaped characters; this should be fixed, though the issue may lie in the underlying XMLWriter code.