Manta xml is a really simple library with no npm, node or browser dependencies, is a pure javascript xml parser, editor and JSON converter.
Manta_xml was developed because out there are more and simpler javascript engines than v8, spidermonkey o chakra, simpler js engines can 'understand' javascript but had no access to C/C++ bindings, node or npm modules to take advantaje of what has been already builded.
npm install manta_xml --saveIn the dist folder other formats are available, the common one for the purpouse of this library is to use the manta_xml.var.js, it will add a global variable name called mantaXML that exposes the xml2JSON, xmlParser and XMLDocument functions.
Manta_xml basically uses a Tree structure of Node Object so to create the inner representation of an XML document the only thing to do is to instantiate it with the attributes for the root node and then add new nodes to the structure.
var XMLDocument = require("manta_xml").XMLDocument;
document = new XMLDocument({nodeName: "body", attrs: {id: "main", class: "body"}}),
node = document.find({nodeName: "body"})[0],
pNode = new Node({nodeName: "p", attrs: {class: "bold"}, _text: "Hello world"});
node.addChild(pNode);
document.toXML();
/*
<?xml version="1.0" encoding="UTF-8"?>
<body id="main" class="body">
<p class="bold">
Hello World
</p>
</body>
*/
document.toJSON()
/*
{
"body":{
"_attrs":{
"id":"main",
"class":"body"
},
"p":{
"_attrs":{
"class":"bold"
},
"_text": "Hello "World"
}
}
}
*/It's possible to parse and XML document, finding and editing some nodes and after that get JSON or XML string of the current XML document with the changes.
check the Node Object to see how to change attributes, nodeName, text or CData values for a particular node.
var xmlParser = require("manta_xml").xmlParser,
document = xmlParser("<node id='23'>inner text</node>"),
node = document.find({attrs: {id: '23'}})[0],
newChild = new Node({nodeName: "child", attrs: {id: "2"}});
node.setAttr("id", "24");
node.setAttr("name", "Manta");
node.setText("New inner text");
node.addChild(newChild);
document.toXML();
// <?xml version="1.0" encoding="UTF-8"?><node id="24" name="Manta">New inner text<child id="2"></child></node>
// or a formatted xml
document.toXML(true);
/**
* <?xml version="1.0" encoding="UTF-8"?>
* <node id="24" name="Manta">
* New inner text
* <child id="2">
* </child>
* </node>
*/the xml2JSON function will take any xml string and will try to parse and convert it to an equivalent JSON object.
var xml2JSON = require("manta_xml").xml2JSON;
json = xml2JSON("<node id='23'>inner text</node>");
// {node: {attrs: {id: '23'}, _text: 'inner text'} }To search into the xml document there is a 'find' method wich can receive a query object or a callback function to go through the xml tree and find nodes that matches the params.
In both cases the result is the same, an array with objects that represent the matching xml nodes, this objects can be the node object it self or a JSON object representing the node like this one:
[{nodeName: 'node1', attrs: {id: '1'}, _text: "some text"}, {nodeName: 'node2', attrs: {id: '2', name: 'second node'}}]The JSON that represents each node will always had the 'nodeName' and 'attrs' keys, and will have '_text' or '_cdata' keys if the current node wraps some text or cdata expression.
If your intention is to edit the nodes attributes, text, CData, adding or removing child nodes you need direct access to the Node object and the find method by default will give you an Array of nodes.
Searching by query object:
Searching with a query object is pretty simple, you provide an object that indicates the attribute that you are going to search in and a value to match to, it can be a standar value (string, int, bool, etc) or a regex literal.
Syntax
var xmlParser = require("manta_xml").xmlParser,
document = xmlParser(".... some xml ...");
document.find({attrs: {some_attribute: 'value to match'}});examples:
// Find nodes with id attribute equal to 1234, something like <bill id='1234'>
results = document.find({attrs: {id: '1234'}})
// Find nodes with id attribute equal to 1234, something like <bill id='1234'>
// But return the result as a JSON.
results = document.find({attrs: {id: '1234'}}, true)
// Find all the bill type nodes, <bill id='123' total='20000'>
results = document.find({nodeName: 'bill'});
// Find all nodes wich name attribute start with 'ever' => <direction name='evergreen avenue' />
results = document.find({attrs: {name: /^ever/}});
// Find all nodes wich inner text includes the word 'world'<line id='12'>Hello world</line>
results = document.find({_text: /world/});
// Find all nodes in wich name attribute it's equal to 'line' and its inner text includes the word 'world' => <node1 name='line'>without world</node1><node2 name='line'>something else</node2>
results = document.find({_text: /mundo/, attrs: {name: 'line'}});Searching with callback function:
If you need more control over the search or perform some operations, you can pass a function to the 'find' method instead of a query object, the callback function will receive a node object and should return a boolean value to indicate if the node is a match.
examples:
// Find all nodes with id attribute equal to 1234, something like <bill id='1234'>
results = document.find(function(node) { return(node.attrs.id && node.attrs.id === '1234'); });
// Find all the bill type nodes, <bill id='123' total='20000'>
results = document.find(function(node) { return(node.nodeName === 'factura'); });
// Find all nodes wich name attribute start with 'ever' => <direction name='evergreen avenue'/>
results = document.find(function(node) { return(node.attrs.name && node.attrs.name.match(/ever/)); });The Node Object is a simple Object that holds its own data, a reference to it's parent and list of children nodes, to create a node object just instantiate it a JSON object that will be it's attributes.
check docs for detailed method description.
method list:
- setAttr
- getAttr
- setText
- getText
- setCData
- getCData
- addChild
- deleteChild
Usage Examples:
var node = new Node({id: "node", class: "inner"});
node.setAttr("name", "Hello world");
console.log(node.getAttr("name"));
node.setText("Hello this is an inner text");
node.setCData("<p>some weird html</p>");
var child = new Node({id: "node2", class: "inner"});
node.addChild(child);To build the library with the multiple outputs just run.
npm run buildto test the library the mocha framework is used
npm run testmanta_xml has be open source under the MIT license.