I have the following javascript function (which I got from Stack Overflow) which converts XML to JSON:
function xmlToJson(xml) {
try {
var obj = {};
if (xml.nodeType == 1) {
if (xml.attributes.length > 0) {
for (var j = 0; j < xml.attributes.length; j++) {
var attribute = xml.attributes.item(j);
obj[attribute.nodeName] = attribute.nodeValue;
}
}
} else if (xml.nodeType == 3) {
obj = xml.nodeValue;
}
if (xml.hasChildNodes()) {
for (var i = 0; i < xml.childNodes.length; i++) {
var item = xml.childNodes.item(i);
var nodeName = item.nodeName;
if (typeof (obj[nodeName]) == "undefined") {
obj[nodeName] = xmlToJson(item);
} else {
if (typeof (obj[nodeName].push) == "undefined") {
var old = obj[nodeName];
obj[nodeName] = [];
obj[nodeName].push(old);
}
obj[nodeName].push(xmlToJson(item));
}
}
}
console.log(JSON.stringify(obj));
return obj;
} catch (e) {
alert(e.message);
}
}
What I want is to return it as an array ([]) when a xml node has at-least single child node and it has a parent node also. In this code it returns map ({}) if xml node has single child node but it is fine with multiple child nodes.
For example, I'd like the XML
<pnode attr1="abc">
<cnode attr2="xyz"></cnode>
</pnode>
to be transformed into the JSON
{
"pnode": {
"attr1": "abc"
},
"cnode": [
{"attr2": "xyz"}
]
}
With the clarification about what you want to achieve, here is an algorithm.
I'll leave my other answer up because I still think the wisest choice is not to play with the structure
function flattenNodes(node, isChild) {
var obj = {}, obj2, i, key, attributes = {};
if (node.attributes && node.attributes.length)
for (i = 0; i < node.attributes.length; ++i)
attributes[node.attributes[i].nodeName] = node.attributes[i].nodeValue;
if (!isChild)
obj[node.nodeName] = attributes;
else {
if (!obj.hasOwnProperty(node.nodeName))
obj[node.nodeName] = [];
else if (!(obj[node.nodeName] instanceof Array))
obj[node.nodeName] = [obj[node.nodeName]];
obj[node.nodeName].push(attributes);
}
attributes = null; // free
if (node.childNodes && node.childNodes.length)
for (i = 0; i < node.childNodes.length; ++i) {
if (node.childNodes[i].nodeType === 3) continue; // skip text node
obj2 = flattenNodes(node.childNodes[i], 1); // recurse
for (key in obj2) // merge
if (obj2.hasOwnProperty(key))
if (!obj.hasOwnProperty(key)) {
obj[key] = obj2[key];
} else {
if (!(obj[key] instanceof Array))
obj[key] = [obj[key]];
obj[key] = obj[key].concat(obj2[key]);
}
}
return obj;
}
Example usage on Node root_node
var root_node;
root_node = new DOMParser().parseFromString(
'<pnode attr1="abc"><cnode attr2="xyz"></cnode></pnode>',
'text/xml'
).documentElement;
var o = flattenNodes(root_node); // create
JSON.stringify(o); // to JSON
// {"pnode":{"attr1":"abc"},"cnode":[{"attr2":"xyz"}]}
If you have XML of the form <foo bar="baz"><foo hello="world"></foo></foo>, the first iteration will cause {foo: {bar: "baz"}}, then the second encounter will modify this to the array form of {foo: [{bar: "baz"}, {hello: "world"}]}
I would form the object representing the XML differently;
Integer nodeType
String nodeName
String nodeValue
Array childNodes
Object attributes
Now you can have the same form independent of number of child nodes/etc
function nodeToObject(node) {
var obj = {}, i;
obj.nodeType = node.nodeType;
obj.nodeName = node.nodeName;
obj.nodeValue = node.nodeValue;
obj.childNodes = [];
obj.attributes = {};
if (node.childNodes && node.childNodes.length)
for (i = 0; i < node.childNodes.length; ++i)
obj.childNodes.push(nodeToObject(node.childNodes[i]));
if (node.attributes && node.attributes.length)
for (i = 0; i < node.attributes.length; ++i)
obj.attributes[node.attributes[i].nodeName] = node.attributes[i].nodeValue;
return obj;
}
And then to transform root_node to JSON,
JSON.stringify(nodeToObject(root_node));
Going in the opposite direction is also possible in JavaScript, with some minor logic based upon nodeType to choose the creation method.
Related
I use this function to convert html element to Json file:
function converter(dom) {
if (dom.nodeType === Node.TEXT_NODE) {
return dom.nodeValue;
}
if (dom.nodeType === Node.DOCUMENT_NODE) {
dom = dom.documentElement;
}
const obj = {};
obj.nodeType = dom.nodeType;
if (dom.nodeType === Node.ELEMENT_NODE) {
obj.tagName = dom.tagName;
obj.attributes = []; // Array.from(obj.attributes) gives us a lot of things we don't want
for (let i = 0, len = dom.attributes.length; i < len; ++i) {
const attr = dom.attributes[i];
obj.attributes.push({name: attr.name, value: attr.value});
}
obj.children = [];
for (let child = dom.firstChild; child; child = child.nextSibling) {
obj.children.push(converter(child));
}
} else {
obj.nodeValue = dom.nodeValue;
}
return obj;
}
const jsonn = JSON.stringify(converter(document.getElementById("examplee")), null, 4);
var data = JSON.parse(jsonn);
console.log(data);
<div id='examplee'>text</div>
Now I want to make it accept HTML code in the conveter intstead of DOM
make it work like this:
function converter(dom) {
if (dom.nodeType === Node.TEXT_NODE) {
return dom.nodeValue;
}
if (dom.nodeType === Node.DOCUMENT_NODE) {
dom = dom.documentElement;
}
const obj = {};
obj.nodeType = dom.nodeType;
if (dom.nodeType === Node.ELEMENT_NODE) {
obj.tagName = dom.tagName;
obj.attributes = []; // Array.from(obj.attributes) gives us a lot of things we don't want
for (let i = 0, len = dom.attributes.length; i < len; ++i) {
const attr = dom.attributes[i];
obj.attributes.push({name: attr.name, value: attr.value});
}
obj.children = [];
for (let child = dom.firstChild; child; child = child.nextSibling) {
obj.children.push(converter(child));
}
} else {
obj.nodeValue = dom.nodeValue;
}
return obj;
}
const jsonn = JSON.stringify(converter("<div>12345</div>"), null, 4);
var data = JSON.parse(jsonn);
console.log(data);
But it returns empty tags. I want it to return the json when I put in the converter.
The problem is it accepts only the document.getElementById("examplee") and not accept the code like the snippet above
Not sure if this is what you're after, but in the first example you're using an HTML element as an argument, whereas in the second example you're using a string as an argument. For that, I've added the following to the beginning of your function:
if (typeof dom === 'string') {
let tmp = document.createElement('template');
tmp.innerHTML = dom;
dom = tmp.content.childNodes[0]
}
function converter(dom) {
if (typeof dom === 'string') {
let tmp = document.createElement('template');
tmp.innerHTML = dom;
dom = tmp.content.childNodes[0]
}
if (dom.nodeType === Node.TEXT_NODE) {
return dom.nodeValue;
}
if (dom.nodeType === Node.DOCUMENT_NODE) {
dom = dom.documentElement;
}
const obj = {};
obj.nodeType = dom.nodeType;
if (dom.nodeType === Node.ELEMENT_NODE) {
obj.tagName = dom.tagName;
obj.attributes = []; // Array.from(obj.attributes) gives us a lot of things we don't want
for (let i = 0, len = dom.attributes.length; i < len; ++i) {
const attr = dom.attributes[i];
obj.attributes.push({
name: attr.name,
value: attr.value
});
}
obj.children = [];
for (let child = dom.firstChild; child; child = child.nextSibling) {
obj.children.push(converter(child));
}
} else {
obj.nodeValue = dom.nodeValue;
}
return obj;
}
const jsonn = JSON.stringify(converter("<div>12345</div>"), null, 4);
var data = JSON.parse(jsonn);
console.log(data);
Bit of a weird one. Am using the following code build an array from a json object to make it easier to reference later in the code. However it would appear that when the last item of each array is created, rather than adding a new item, the Key of the item appears as the length of the array.
perfsJson = $.parseJSON(result);
var extras = new Array();
for (var i = perfsJson.length - 1; i >= 0; i--) {
var obj = perfsJson[i];
if (obj != null) {
if (obj.Extras != null) {
for (var perf_no in obj.Extras) {
if (extras[perf_no] == undefined) {
var arr = new Array();
for (var extra in obj.Extras[perf_no]) {
if (arr[extra] == undefined) {
arr[extra] = obj.Extras[perf_no][extra];
}
}
extras[perf_no] = arr;
}
}
break;
}
}
}
The resulting array appears as below:
Any ideas what's going on here?
Edit:
Sample of Json below
{"Extras":{"32516":{"24186":"Example text"},"32515":{"24186":"Example text"},"32514":{"24186":"Example text"},"32512":{"24186":"Example text"},"32513":{"24186":"Example text"},"32511":{"24186":"Example text"},"32510":{"24186":"Example text"},"32509":{"24186":"Example text"},"32507":{"24186":"Example text"},"32503":{"24186":"Example text"},"32506":{"24186":"Example text"},"32505":{"24186":"Example text"},"32508":{"24186":"Example text"},"32502":{},"32497":{}}}
What's going on hear is that you are using for..in to iterate over an array, which is a no-no because it iterates properties that are not the array elements (such as the .length property). Instead, use Array#forEach:
perfsJson = $.parseJSON(result);
var extras = new Array();
for (var i = perfsJson.length - 1; i >= 0; i--) {
var obj = perfsJson[i];
if (obj != null) {
if (obj.Extras != null) {
obj.Extras.forEach(function (item, idx) {
if (typeof extras[idx] === 'undefined') {
var arr = new Array();
item.forEach(function (item2, idx2) {
if (typeof arr[idx2] === 'undefined') {
arr[idx2] = item2;
}
});
extras[idx] = arr;
}
});
break;
}
}
}
The innermost loop is pretty pointless and can be replaced with Array#slice:
perfsJson = $.parseJSON(result);
var extras = new Array();
for (var i = perfsJson.length - 1; i >= 0; i--) {
var obj = perfsJson[i];
if (obj != null) {
if (obj.Extras != null) {
obj.Extras.forEach(function (item, idx) {
if (typeof extras[idx] === 'undefined') {
extras[idx] = item.slice();
}
});
break;
}
}
}
The next inner loop can be replaced with Array#map and two if statements can be combined:
perfsJson = $.parseJSON(result);
var extras = new Array();
for (var i = perfsJson.length - 1; i >= 0; i--) {
var obj = perfsJson[i];
if (obj != null&& obj.Extras != null) {
extras = obj.Extras.map(function (item) {
return item.slice();
});
break;
}
}
In fact, most of this code can be simplified:
function findLastElement(arr) {
for (var i = arr.length - 1; i >= 0; i -= 1) {
if (arr[i] != null && arr[i].Extras != null) { return arr[i]; }
}
}
perfsJson = $.parseJSON(result);
var lastElement = findLastElement(perfsJson);
var extras = lastElement
? lastElement.Extras.map(function (item) { return item.slice(); })
: [];
I'm using a blob of code to convert xml to json:
// Changes XML to JSON
var XmlToJson = function xmlToJson(xml) {
//console.log('called xmltojson');
//console.log(xml);
// Create the return object
var self = this;
var obj = {};
if (xml.nodeType == 1) { // element
// do attributes
if (xml.attributes.length > 0) {
obj["#attributes"] = {};
for (var j = 0; j < xml.attributes.length; j++) {
var attribute = xml.attributes.item(j);
obj["#attributes"][attribute.nodeName] = attribute.nodeValue;
}
}
} else if (xml.nodeType == 3) { // text
obj = xml.nodeValue;
}
// do children
if (xml.hasChildNodes()) {
for(var i = 0; i < xml.childNodes.length; i++) {
var item = xml.childNodes.item(i);
var nodeName = item.nodeName;
if (typeof(obj[nodeName]) == "undefined") {
obj[nodeName] = xmlToJson(item);
} else {
if (typeof(obj[nodeName].push) == "undefined") {
var old = obj[nodeName];
obj[nodeName] = [];
obj[nodeName].push(old);
}
obj[nodeName].push(xmlToJson(item));
}
}
}
return obj;
};
module.exports = XmlToJson;
Sample XML input:
<ArrayOfstring xmlns:i="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://schemas.microsoft.com/2003/10/Serialization/Arrays">
<string>asdf</string>
<string>123</string>
<string>zxcv</string>
<string>qwer</string>
<string>werty</string>
<string>dfgh</string>
<string>rytui</string>
</ArrayOfstring>
Output:
Looking at the object in Chrome Console I see:
Object {Arrayofstring: Object}
ArrayOfString: Object
#attributes: Object
string: Array[7]
0: Object
#text: "123"
1: Object
#text: "456"
I'm having problems getting to the #text data. AFAIK, hash is an illegal character in a variable name. Why is it there? How do I access the value of these #text variables?
I tried variations of:
console.log(myVariable.string[0]);
And the variations I tried all result in undefined.
The #text-variables stem from DOM text nodes (nodeType = 3) which have the shorthand type "#text".
In my console, I can do:
> var x = {};
< undefined
> x['#text']= 'abcd';
< "abcd"
> x
< Object { #text: "abcd" }
> x['#text']
< "abcd"
So in your case (I hope I get the object and it's children right:
console.log (objName.Arrayofstring.string[0]['#text'])
Can i optimize my code bit more to reduce number of line?
I am checking/passing individual array elements? Can i make it re-write in a generic way?
for (var i = 0; i < $scope.studentReport.Students.length; i++)
{
if (_isValueNan($$scope.studentReport.Students[i].Age))
$$scope.studentReport.Students[i].Age = null;
if (_isValueNan($$scope.studentReport.Students[i].Number))
$$scope.studentReport.Students[i].Number = null;
if (_isValueNan($$scope.studentReport.Students[i].Height))
$$scope.studentReport.Students[i].Height = null;
}
var _isValueNan = function (item) {
var result = false;
if (typeof item == 'number' && isNaN(item))
result = true;
return result;
}
With ref to Stumblor's answer:
for (var i = 0; i < $scope.studentReport.Students.length; i++) {
_checkValueNan($$scope.studentReport.Students[i], ["Age", "Number", "Height"]);
}
var _checkValueNan = function (item, values) {
values.forEach(function (val) {
if (typeof item[val] === 'number' && isNaN(item[val])) item[val] = null;
});
}
You can nullify the property internally in the function, and also pass the item and property value in independently. For example:
for (var i = 0; i < $scope.studentReport.Students.length; i++)
{
_checkValueNan($$scope.studentReport.Students[i], "Age");
_checkValueNan($$scope.studentReport.Students[i], "Number");
_checkValueNan($$scope.studentReport.Students[i], "Height");
}
var _checkValueNan = function (item, valueName) {
if (typeof item[valueName] == 'number' && isNaN(item[valueName]))
item[valueName] = null;
}
EDIT:
Leading on from Vicky Gonsalves answer, you could additionally check ANY of the properties of the object, which might be a more scalable solution.
var _checkAllValueNan = function (item) {
for(var key in item) { // iterates all item properties
if (!item.hasOwnProperty(key)) continue; // ensures not prop of prototype
if (typeof item[key] == 'number' && isNaN(item[key])) item[key] = null;
}
}
for (var i = 0; i < $scope.studentReport.Students.length; i++)
{
_checkAllValueNan($scope.studentReport.Students[i]);
}
i have this xml file:
<dist>
<key>keynumber1</key>
<string>value1</string>
<key>keynumber2</key>
<string>value2</string>
<key>keynumber3</key>
<string>value3</string>
<key>keynumber4</key>
<integer>value4</integer>
</dist>
how can i parse this with jquery like:
{ "dist": {"keynumber1":"value1", "keynumber2":"value2"}}
Thanks a lot for help
First step is parsing xml with jQuery using $.parseXML(str);
Then I used this pretty function, created by David Welsh
function xmlToJson(xml) {
// Create the return object
var obj = {};
if (xml.nodeType == 1) { // element
// do attributes
if (xml.attributes.length > 0) {
obj["#attributes"] = {};
for (var j = 0; j < xml.attributes.length; j++) {
var attribute = xml.attributes.item(j);
obj["#attributes"][attribute.nodeName] = attribute.nodeValue;
}
}
} else if (xml.nodeType == 3) { // text
obj = xml.nodeValue;
}
// do children
if (xml.hasChildNodes()) {
for(var i = 0; i < xml.childNodes.length; i++) {
var item = xml.childNodes.item(i);
var nodeName = item.nodeName;
if (typeof(obj[nodeName]) == "undefined") {
obj[nodeName] = xmlToJson(item);
} else {
if (typeof(obj[nodeName].length) == "undefined") {
var old = obj[nodeName];
obj[nodeName] = [];
obj[nodeName].push(old);
}
obj[nodeName].push(xmlToJson(item));
}
}
}
return obj;
};