HtmlAgilityPack get two nodes simultaneously in c#

Question

I'm trying to parse an html page , I would get a pair of nodes from this code

 <li class="classli"> 
    <div class="element">element1</div>  
    <div class="description">description1</div> 
  </li>  
  <li class="classli"> 
    <div class="element">element2</div>  
    <div class="description">description2</div> 
  </li>  
  <li class="classli"> 
    <div class="xxxelementclass">element3</div>  
    <div class="description">description3</div> 
  </li>  
  <li class="classli"> 
    <div class="element">element4</div>  
    <div class="xxxclass">description4</div> 
  </li>

i tried this in c# :

foreach(var node in doc.SelectNodes("//li[contains(@class,classli)]"))
{
    listelement.add(node.SelectSingleNode("//div[contains(@class,element)]").InnerText);
    listdescription(node.SelectSingleNode("//div[contains(@class,description)]").InnerText);
}

in the HTML page, not all the (li) tags contain the same subtags , I would get description and element only where both present

Xi Sigma · Answer 1 · 2016-03-03T13:59:28.920

Make your xpath in your for each looks like the following

//li[contains(@class,'classli') and ./div[contains(@class,'element')] and ./div[contains(@class,'description')]]

this will only consider the elements that has both divs with the given classes as child nodes, also note that the xpaths inside your for each need to start looking for decendant nodes starting from the li node, so you need to use ./ for children or .// for decendants such as

./div[contains(@class,'element')]

./div[contains(@class,'description')]

score 1 · Answer 2 · edited May 23 '17 at 12:15

Proper XPath expression to match by CSS class is a bit complicated. Taking a moderate approach i.e the 2nd code snippet posted in this other answer, the XPath for your task would be as follow (formatted into lines for readability) :

var query = @"//li[contains(concat(' ', @class, ' '), ' classli ')]
                  [div[contains(concat(' ', @class, ' '), ' element ')]]
                  [div[contains(concat(' ', @class, ' '), ' description ')]]";

foreach(var node in doc.SelectNodes(query))
{
    var elementQuery = "div[contains(concat(' ', @class, ' '), ' element ')]";
    listelement.add(node.SelectSingleNode(elementQuery).InnerText);

    var descriptionQuery = "div[contains(concat(' ', @class, ' '), ' description ')]";
    listdescription.add(node.SelectSingleNode(descriptionQuery).InnerText);
}

score 0 · Accepted Answer · answered Mar 04 '16 at 15:58

AsEnumerableThank you all for the help I solved this way

    foreach(var node in doc.SelectNodes("//li[contains(@class,classli)]"))
    {

   List<HTMLNODE> Child = node.childnodes.where(o=> (o.getattribbutevalue(class,"") == "element") or (o.getattribbutevalue(class,"") == "description")).AsEnumerable().ToList();

    }

For(int i = 0; i <= Child.count-1;i=i+2)
{
listelement.add(Child[i].InnerHtml;
listdescription.add(Child[i+1].InnerHtml;
}

HtmlAgilityPack get two nodes simultaneously in c#

3 Answers3