Question

I have an xml file that looks like this

<step>
    <id>3</id>
    <backid>1</backid>
    <question>Are you having a good day</question>
    <yesid>4</yesid>
    <noid>5</noid>
       </step>

It’s set up to be an application that asks the user yes and no questions to assist in guiding the user through a complicated business process. The yesid and noid are the id numbers of the next step of the process. In the past (and in other languages) I would have loaded the info from the xml file into a multidimensional array and gone from there.

However I am attempting to use linq to xml in this application and I am wondering if there is a more efficient way of doing this than the xpath to array that I have been doing in the past

Answer 1

I would create a class that represents a step, read the XML file into it using LINQ-to-XML, and then create a dictionary of steps for easy lookup:

var doc = XDocument.Load("xmlfile1.xml");

var steps = from step in doc.Root.Elements("step")
            select new Step
                   {
                       Id       = (int)step.Element("id"),
                       BackId   = (int)step.Element("backid"),
                       Question = (string)step.Element("question"),
                       YesId    = (int)step.Element("yesid"),
                       NoId     = (int)step.Element("noid"),
                   };

var dict = steps.ToDictionary(step => step.Id);

var currentStep = dict[3];

while (true)
{
    switch (Ask(currentStep.Question))
    {
        case Yes:  currentStep = dict[currentStep.YesId];  break;
        case No:   currentStep = dict[currentStep.NoId];   break;
        case Back: currentStep = dict[currentStep.BackId]; break;
        default:   return;
    }
}

(Assuming the file contains several <step> elements under a common root.)