Simple parsing and building of XML documents, powered by new features only available in ES6.
The goals of this project:
- To make XML parsing require little-to-no error checking. Exploring the node tree should be as forgiving as possible.
- To provide an easy way to open, modify and save an XML document in a single, unified library.
- To be easily installed without any native code compilation.
- To provide a novel and terse API by taking advantage of the latest ES6 metaprogramming features.
- Useful for quick and dirty jobs which don't need fine grained control over every intricacy of XML.
- Prioritising ease-of-use over speed.
- To make you re-think your approach to XML parsing.
Requires a recent version of Node which has Proxy support.
var trixml = require('trixml');
var xml = trixml.parseSync(`
<?xml version="1.0"?>
<Root an="attribute" another="one">
<Envelope>
<Message id="msg1">Hello</Message>
<Message id="msg2">World</Message>
<Message id="msg3">!</Message>
</Envelope>
</Root>
`);
xml.attr() // {an: "attribute", another: "one"}
xml.name; // "Root"
xml.value; // ""
xml.Envelope.Message.value; // "Hello"
xml.Envelope.Message.map(msg => msg.value).join(' ') // "Hello World !"
xml.Envelope.Message.attr('id') // msg1
xml.Envelope.Message.map(msg => msg.attr('id')).join(',') // "msg1,msg2,msg3"
var trixml = require('trixml');
var xml = trixml.newDoc('root');
xml.Header("Hello ...");
xml.Body.Message("World", {length: 5});
xml.Body.Message("World!", {length: 6, planet: "Earth"}); // overwrites value & merges attributes
xml.Footer.wibble({a: "b", c: "d"});
xml.Footer.wobble("Easy!");
xml.Footer.addChild("wubble", {e: "f"}, "Simple!");
xml.Footer.addChild("wubble");
xml.Footer.flub("fluub");
xml.Trailer;
console.log(xml.toXML());
// <?xml version="1.0"?>
// <root>
// <Header>Hello ...</Header>
// <Body>
// <Message length="6" planet="Earth">World!</Message>
// </Body>
// <Footer>
// <wibble a="b" c="d" />
// <wobble>Easy!</wobble>
// <wubble e="f">Simple</wubble>
// <wubble />
// <flub>fluub</flub>
// </Footer>
// <Trailer />
// </root>
var trixml = require('trixml');
var https = require('https');
function getRSSXML(callback) {
var xmlString = "";
https.request({
hostname: 'github.com',
path: '/mikuso.atom'
}, res => {
res.on('data', (d) => {
xmlString += d;
});
res.once('end', () => callback(xmlString));
}).end();
}
getRSSXML((xmlString) => {
var doc = trixml.parseSync(xmlString);
var items = doc.entry.map(e => `${e.published}: ${e.title}`).join("\n");
console.log(`Title: ${doc.title}\nUpdated: ${doc.updated}\n------\n${items}`);
// Title: mikuso’s Activity
// Updated: 2016-07-22T17:21:22Z
// ------
// 2016-07-22T17:21:22Z: mikuso pushed to master at mikuso/trixml
// 2016-07-22T17:19:48Z: mikuso pushed to master at mikuso/trixml
// 2016-07-22T17:18:00Z: mikuso pushed to master at mikuso/trixml
// 2016-07-22T17:18:00Z: mikuso created tag v0.0.3 at mikuso/trixml
// ...
});
It's possible for nodes in your XML document to have names which conflict with the names of members of the XMLNode
or XMLNodeCollection
objects. When this happens, you cannot use the shorthand method for accessing child nodes.
The following XML nodes are affected:
<inspect>
, <toJSON>
, <toString>
, <value>
, <children>
, <comments>
, <attr>
, <addChild>
, <get>
, <cloneSync>
, <remove>
, <empty>
In these cases, you must use the longhand method like this:
var root = trixml.parseSync(`
<?xml version="1.0"?>
<root>
<inspect>
<addChild>gotcha</addChild>
</inspect>
</root>
`);
root.get('inspect').get('addChild').value === "gotcha"; // true
// don't do this... (it won't work)
root.inspect.addChild.value; // TypeError: Cannot read property 'addChild' of null
- Add support for streams
- Async parsing, returning Promises
- Ability to remove attributes
- Add support for more Array methods on XMLNodeCollection
- More tests