I have html content which looks like
<body>Hello world</div><div>New day</div></body>
I would like to parse this html snippet and add a starting div tag before Hello. What is the approach I could follow? I tried to use HTMLCLeaner but it didnt help Basically what this means is find ending div tags without matching start div tags and add them.
If you use java try using Jsoup. Something like
Jsoup.clean("<body><div>Hello world</div><div>New day</div></body>", Whitelist.relaxed());
This will give you the proper output string.
UPDATE
You can use Jsoup.parse(html) which returns a Document on which you can call toString() to get the fixed html which will include all the html and body tags as well. It will give you the following output for you html.
<html>
<head></head>
<body>
<div>
Hello world
</div
<div>
New day
</div>
</body>
</html>
As you said most of the parser will fix the end tags but not start tags as they can not decide on where to start the start tags except just before the wrong end tag and it will be useless to add the start tag there just before the end tag.
You may need to implement you own logic for that as Trevor Hutto's suggestion (Stack based approach) bellow but it will have its own complications depends on your requirement.
You could use a stack.
Push on open tags, then when you hit a close tag, pop and compare the popped tag to the one you just ran into.
So obviously, if you have a mismatch, and it is a div, you can do something.
John Resig's HTML Parser does a pretty good job of this. It's a little old, but it's still worked for a large majority of my use cases.
Edit: Actually, it seems to only fix missing closing tags, not opening tags...though some tweaks might be able to have it do the latter.
You can use the same technique that's used in parenthesis balancing, except instead of returning True/False, you would fix the tag instead. I did this for a work project once:
Recursive method for parentheses balancing [python]
What Trevor is describing is the same thing I'm describing (used in parenthesis balancing).
I have created a Javascript/jQuery solution to add missing starting tags:
Add any HTML to the body with missing tags like:
hello</h3>
<p>hai</p>
Welcome to fiddle</span>
</div>
Javascript/jQuery
var content;
var i;
var result="";
var previousTag="";
function exeq(){
var a = content.lastIndexOf('<body>');
var z = content.lastIndexOf('</body>');
content = content.substring(a+6,z);
while(i!=-1){
var startAngle = content.indexOf('<');
var endAngle = content.indexOf('>');
i=endAngle;
var ele = content.substring(startAngle,endAngle+1);
if(ele.indexOf("/")!=-1)
{
if(previousTag != ele.replace("/",""))
result = result + ele.replace("/","");
}
result = result + content.substring(0,endAngle+1);
content = content.substring(endAngle+1);
previousTag = ele;
}
/*Below part only to append result to body*/
$('body').append('<h4>Result</h4><textarea>'+result+'</textarea>');
/******************************************/
}
$.get(window.location.href,function(data){
typeof(data)=="object"?window.location = window.location.href:
content = data;
exeq();
});
<body>
<div>Hello world</div>
<div>New day</div>
</body>
You can add a div before the hello world or you can remove the closing div after the hello world. Hello world New day
You dont need a HTMLCLeaner or any tool, if you need to work with html is very simple just remember all tag <'something'> is close with a or use a simple <'something'/> for summarize one!
