I have a string that represents a non indented XML that I would like to pretty-print. For example:
<root><node/></root>
should become:
<root>
<node/>
</root>
Syntax highlighting is not a requirement. To tackle the problem I first transform the XML to add carriage returns and white spaces and then use a pre tag to output the XML. To add new lines and white spaces I wrote the following function:
function formatXml(xml) {
var formatted = '';
var reg = /(>)(<)(\/*)/g;
xml = xml.replace(reg, '$1\r\n$2$3');
var pad = 0;
jQuery.each(xml.split('\r\n'), function(index, node) {
var indent = 0;
if (node.match( /.+<\/\w[^>]*>$/ )) {
indent = 0;
} else if (node.match( /^<\/\w/ )) {
if (pad != 0) {
pad -= 1;
}
} else if (node.match( /^<\w[^>]*[^\/]>.*$/ )) {
indent = 1;
} else {
indent = 0;
}
var padding = '';
for (var i = 0; i < pad; i++) {
padding += ' ';
}
formatted += padding + node + '\r\n';
pad += indent;
});
return formatted;
}
I then call the function like this:
jQuery('pre.formatted-xml').text(formatXml('<root><node1/></root>'));
This works perfectly fine for me but while I was writing the previous function I thought that there must be a better way. So my question is do you know of any better way given an XML string to pretty-print it in an html page? Any javascript frameworks and/or plugins that could do the job are welcome. My only requirement is this to be done on the client side.
From the text of the question I get the impression that a string result is expected, as opposed to an HTML-formatted result.
If this is so, the simplest way to achieve this is to process the XML document with the identity transformation and with an <xsl:output indent="yes"/> instruction:
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output omit-xml-declaration="yes" indent="yes"/>
<xsl:template match="node()|@*">
<xsl:copy>
<xsl:apply-templates select="node()|@*"/>
</xsl:copy>
</xsl:template>
</xsl:stylesheet>
When applying this transformation on the provided XML document:
<root><node/></root>
most XSLT processors (.NET XslCompiledTransform, Saxon 6.5.4 and Saxon 9.0.0.2, AltovaXML) produce the wanted result:
<root> <node /> </root>
consider to use vkBeautify plugin
http://www.eslinstructor.net/vkbeautify/
it's written in plain javascript, very small: less then 1.5K if minified, very fast: less then 5 msec. to process 50K XML text.
Slight modification of efnx clckclcks's javascript function. I changed the formatting from spaces to tab, but most importantly I allowed text to remain on one line:
var formatXml = this.formatXml = function (xml) {
var reg = /(>)\s*(<)(\/*)/g; // updated Mar 30, 2015
var wsexp = / *(.*) +\n/g;
var contexp = /(<.+>)(.+\n)/g;
xml = xml.replace(reg, '$1\n$2$3').replace(wsexp, '$1\n').replace(contexp, '$1\n$2');
var pad = 0;
var formatted = '';
var lines = xml.split('\n');
var indent = 0;
var lastType = 'other';
// 4 types of tags - single, closing, opening, other (text, doctype, comment) - 4*4 = 16 transitions
var transitions = {
'single->single': 0,
'single->closing': -1,
'single->opening': 0,
'single->other': 0,
'closing->single': 0,
'closing->closing': -1,
'closing->opening': 0,
'closing->other': 0,
'opening->single': 1,
'opening->closing': 0,
'opening->opening': 1,
'opening->other': 1,
'other->single': 0,
'other->closing': -1,
'other->opening': 0,
'other->other': 0
};
for (var i = 0; i < lines.length; i++) {
var ln = lines[i];
// Luca Viggiani 2017-07-03: handle optional <?xml ... ?> declaration
if (ln.match(/\s*<\?xml/)) {
formatted += ln + "\n";
continue;
}
// ---
var single = Boolean(ln.match(/<.+\/>/)); // is this line a single tag? ex. <br />
var closing = Boolean(ln.match(/<\/.+>/)); // is this a closing tag? ex. </a>
var opening = Boolean(ln.match(/<[^!].*>/)); // is this even a tag (that's not <!something>)
var type = single ? 'single' : closing ? 'closing' : opening ? 'opening' : 'other';
var fromTo = lastType + '->' + type;
lastType = type;
var padding = '';
indent += transitions[fromTo];
for (var j = 0; j < indent; j++) {
padding += '\t';
}
if (fromTo == 'opening->closing')
formatted = formatted.substr(0, formatted.length - 1) + ln + '\n'; // substr removes line break (\n) from prev loop
else
formatted += padding + ln + '\n';
}
return formatted;
};
This can be done using native javascript tools, without 3rd party libs, extending the @Dimitre Novatchev's answer:
var prettifyXml = function(sourceXml)
{
var xmlDoc = new DOMParser().parseFromString(sourceXml, 'application/xml');
var xsltDoc = new DOMParser().parseFromString([
// describes how we want to modify the XML - indent everything
'<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform">',
' <xsl:strip-space elements="*"/>',
' <xsl:template match="para[content-style][not(text())]">', // change to just text() to strip space in text nodes
' <xsl:value-of select="normalize-space(.)"/>',
' </xsl:template>',
' <xsl:template match="node()|@*">',
' <xsl:copy><xsl:apply-templates select="node()|@*"/></xsl:copy>',
' </xsl:template>',
' <xsl:output indent="yes"/>',
'</xsl:stylesheet>',
].join('\n'), 'application/xml');
var xsltProcessor = new XSLTProcessor();
xsltProcessor.importStylesheet(xsltDoc);
var resultDoc = xsltProcessor.transformToDocument(xmlDoc);
var resultXml = new XMLSerializer().serializeToString(resultDoc);
return resultXml;
};
console.log(prettifyXml('<root><node/></root>'));
Outputs:
<root>
<node/>
</root>
Personnaly, I use google-code-prettify with this function :
prettyPrintOne('<root><node1><root>', 'xml')
Found this thread when I had a similar requirement but I simplified OP's code as follows:
function formatXml(xml, tab) { // tab = optional indent value, default is tab (\t)
var formatted = '', indent= '';
tab = tab || '\t';
xml.split(/>\s*</).forEach(function(node) {
if (node.match( /^\/\w/ )) indent = indent.substring(tab.length); // decrease indent by one 'tab'
formatted += indent + '<' + node + '>\r\n';
if (node.match( /^<?\w[^>]*[^\/]$/ )) indent += tab; // increase indent
});
return formatted.substring(1, formatted.length-3);
}
works for me!
Or if you'd just like another js function to do it, I've modified Darin's (a lot):
var formatXml = this.formatXml = function (xml) {
var reg = /(>)(<)(\/*)/g;
var wsexp = / *(.*) +\n/g;
var contexp = /(<.+>)(.+\n)/g;
xml = xml.replace(reg, '$1\n$2$3').replace(wsexp, '$1\n').replace(contexp, '$1\n$2');
var pad = 0;
var formatted = '';
var lines = xml.split('\n');
var indent = 0;
var lastType = 'other';
// 4 types of tags - single, closing, opening, other (text, doctype, comment) - 4*4 = 16 transitions
var transitions = {
'single->single' : 0,
'single->closing' : -1,
'single->opening' : 0,
'single->other' : 0,
'closing->single' : 0,
'closing->closing' : -1,
'closing->opening' : 0,
'closing->other' : 0,
'opening->single' : 1,
'opening->closing' : 0,
'opening->opening' : 1,
'opening->other' : 1,
'other->single' : 0,
'other->closing' : -1,
'other->opening' : 0,
'other->other' : 0
};
for (var i=0; i < lines.length; i++) {
var ln = lines[i];
var single = Boolean(ln.match(/<.+\/>/)); // is this line a single tag? ex. <br />
var closing = Boolean(ln.match(/<\/.+>/)); // is this a closing tag? ex. </a>
var opening = Boolean(ln.match(/<[^!].*>/)); // is this even a tag (that's not <!something>)
var type = single ? 'single' : closing ? 'closing' : opening ? 'opening' : 'other';
var fromTo = lastType + '->' + type;
lastType = type;
var padding = '';
indent += transitions[fromTo];
for (var j = 0; j < indent; j++) {
padding += ' ';
}
formatted += padding + ln + '\n';
}
return formatted;
};
This library does exactly what you want!
All of the javascript functions given here won't work for an xml document having unspecified white spaces between the end tag '>' and the start tag '<'. To fix them, you just need to replace the first line in the functions
var reg = /(>)(<)(\/*)/g;
by
var reg = /(>)\s*(<)(\/*)/g;
what about creating a stub node (document.createElement('div') - or using your library equivalent), filling it with the xml string (via innerHTML) and calling simple recursive function for the root element/or the stub element in case you don't have a root. The function would call itself for all the child nodes.
You could then syntax-highlight along the way, be certain the markup is well-formed (done automatically by browser when appending via innerHTML) etc. It wouldn't be that much code and probably fast enough.
If you are looking for a JavaScript solution just take the code from the Pretty Diff tool at http://prettydiff.com/?m=beautify
You can also send files to the tool using the s parameter, such as: http://prettydiff.com/?m=beautify&s=https://stackoverflow.com/
var formatXml = this.formatXml = function (xml) {
var reg = /(>)(<)(\/*)/g;
var wsexp = / *(.*) +\n/g;
var contexp = /(<.+>)(.+\n)/g;
xml = xml.replace(reg, '$1\n$2$3').replace(wsexp, '$1\n').replace(contexp, '$1\n$2');
var pad = 0;
var formatted = '';
var lines = xml.split('\n');
var indent = 0;
var lastType = 'other';
Or just print out the special HTML characters?
Ex: <xmlstuff> 	<node /> </xmlstuff>
	 Horizontal tab
Line feed
XMLSpectrum formats XML, supports attribute indentation and also does syntax-highlighting for XML and any embedded XPath expressions:
XMLSpectrum is an open source project, coded in XSLT 2.0 - so you can run this server-side with a processor such as Saxon-HE (recommended) or client-side using Saxon-CE.
XMLSpectrum is not yet optimised to run in the browser - hence the recommendation to run this server-side.
Use above method for pretty print and then add this in any div by using jquery text() method. for example id of div is xmldiv then use :
$("#xmldiv").text(formatXml(youXmlString));
here is another function to format xml
function formatXml(xml){
var out = "";
var tab = " ";
var indent = 0;
var inClosingTag=false;
var dent=function(no){
out += "\n";
for(var i=0; i < no; i++)
out+=tab;
}
for (var i=0; i < xml.length; i++) {
var c = xml.charAt(i);
if(c=='<'){
// handle </
if(xml.charAt(i+1) == '/'){
inClosingTag = true;
dent(--indent);
}
out+=c;
}else if(c=='>'){
out+=c;
// handle />
if(xml.charAt(i-1) == '/'){
out+="\n";
//dent(--indent)
}else{
if(!inClosingTag)
dent(++indent);
else{
out+="\n";
inClosingTag=false;
}
}
}else{
out+=c;
}
}
return out;
}
You can get pretty formatted xml with xml-beautify
var prettyXmlText = new XmlBeautify().beautify(xmlText,
{indent: " ",useSelfClosingElement: true});
indent:indent pattern like white spaces
useSelfClosingElement: true=>use self-closing element when empty element.
Original(Before)
<?xml version="1.0" encoding="utf-8"?><example version="2.0">
<head><title>Original aTitle</title></head>
<body info="none" ></body>
</example>
Beautified(After)
<?xml version="1.0" encoding="utf-8"?>
<example version="2.0">
<head>
<title>Original aTitle</title>
</head>
<body info="none" />
</example>
var reg = /(>)\s*(<)(\/*)/g;
xml = xml.replace(/\r|\n/g, ''); //deleting already existing whitespaces
xml = xml.replace(reg, '$1\r\n$2$3');
https://www.npmjs.com/package/js-beautify
This library works for me. Supports tab, supports web and node version. Also supports JS, HTML, CSS. Also available as CDN.
来源：https://stackoverflow.com/questions/376373/pretty-printing-xml-with-javascript