linq to get unique a z list - Extending Umbraco

John C Scott 473 posts 1183 karma points

Feb 06, 2013 @ 07:39

linq to get unique A-Z list

I was trying to make an XSLT extension that can use linq to build an A-Z list page.

I added a document property called "navAZ"

As far as I got so far was:

IEnumerable<Node> azNodes = uQuery.GetRootNode().GetDescendantOrSelfNodes().Where(w => w.GetProperty("navAZ").Value != "");
Dictionary<string,int> azTitles = azNodes.ToDictionary(o => o.GetProperty<string>("navAZ"),o => o.Id);

Which isn't very far.

The idea is that if the node has this property and it's populated with something then it would appear in the A-Z list. I thought I could use a .distinct() to get the first letter of the property and then list each node that starts with that initial.

From the XSLT extension I wanted to return an XML navigator that looked something like

<AtoZ>
 <initial letter="A"><node name="Ade" id="1234"/><node name="Andy" id="1244"/></initial>
    <initial letter="B"><node name="Billy" id="1254"/></initial>
 <initial letter="D"><node name="Dave" id="1268"/><node name="Dell" id="1322"/></initial>
</AtoZ>

where the initial element was only there if there were matches for the documents.

I can do the last bit to generate the XML but I am really not sure what the best approach to getting the unique initial letters is and then joining that to the list of nodes (the name attribute is really very optional above and onlu really to illustrate what is going on, though I can see it could be nice).

Copy Link

John C Scott 473 posts 1183 karma points

Feb 07, 2013 @ 19:16

Well this is where I got to:

namespace bucksCC.extensions.XSLTExtension
{
    [XsltExtension(Namespace = "sa.umb.atoz")]
    public class AtoZ
    {
        public static string initials()
        {
            IEnumerable<nodeAZ> azNodes = uQuery.GetRootNode().GetDescendantOrSelfNodes()
                .Where(w => w.GetProperty<string>("navAZ") != "")
                .Select( n => new nodeAZ(n.Id.ToString(),n.GetProperty<string>("navAZ")));
            IEnumerable<char> initials = azNodes
                .Select(c => c.Initial).Distinct().OrderBy(c => c);
            var joined = initials
                .GroupJoin(azNodes, i => i, n => n.Initial, (i, n) => new { letters = i, nodes = n });
             XDocument xdoc = new XDocument(new XElement("AtoZ",
                  from l in joined
                  select new XElement("Initial", 
                      new XAttribute("letter", l.letters),
                      new XElement("Nodes",
                          from Elements in l.nodes
                          select new XElement("Node",
                              new XAttribute("ID", Elements.ID),
                              new XAttribute("Name", Elements.Name)
                          )                      )                  )));
             return xdoc.ToString();
        }
     }
public class nodeAZ {        
public string ID { get; set; }
public string Name { get; set; }
        public char Initial { get; set; }
public nodeAZ(string id, string name)
{
            ID = id;
            Name = name;
            Initial = name[0];
        }
    }
}

...will change return type to node iterator and not return just the xml string and also make the paramater passed in the doctype property rather than hard coding this

I really can't tell if this is the rantings of a mad man or really ok - i'd appreciate any criticism :)

things that would be nice to know would be
how to do the last bit as lambda if that makes any sense
and also how not to have the nodes wrapping the node elements

and generally how maybe this could be clearer and simpler

the xml i get out looks like this

<AtoZ>
  <Initial letter="A">
    <Nodes>
        <Node ID="1063" Name="Article"/>
    </Nodes>
  </Initial>
  <Initial letter="E">
      <Nodes>
        <Node ID="1058" Name="Event"/>
      </Nodes>
    </Initial>
    <Initial letter="N">
      <Nodes>
        <Node ID="1057" Name="Navigation"/>
        <Node ID="1064" Name="Navigation Two"/>
      </Nodes>
    </Initial>
</AtoZ>

any thoughts very much appreciated

Copy Link

Lee Kelleher 4026 posts 15836 karma points MVP 13x admin c-trib

Feb 08, 2013 @ 17:48

Hi John,

Take a look at the Enumerable.GroupBy() method. You could do something like this:

// get the nodes
var nodes = uQuery.GetNodesByXPath("//*[@isDoc]");

// group the nodes by first letter
var grouping = nodes.GroupBy<umbraco.NodeFactory.Node, char>(x => char.ToUpper(x.Name[0]));

// loop the grouping
foreach (var item in grouping)
{
    // item.Key - contains the letter
    foreach (var node in item)
    {
        // node.Name
    }
}

Cheers, Lee.

Copy Link

is working on a reply...

Flag this post as spam?

linq to get unique A-Z list