Multiple For Eaches or XPath Change? 2

[Genimuse]

I'm parsing through an XML document and need to get multiple values from each item. A single item in the XML looks like this:

  <Customer>
    <ID>1043</ID>
    <Name>Bob Smith</Name>
    <Provider>
      <ProviderID>860</ProviderID>
    </Provider>
  </Customer>

I'm looking for records that match a particular ProviderID (in this example, = 860), so my bit of code looks like this:

Dim xmlDoc As New System.Xml.XmlDocument()
'Load the file
xmlDoc.Load(AppPath() & "\Customers.xml")
Dim xmlSourceList As System.Xml.XmlNodeList
Dim xmlSourceItem As System.Xml.XmlNode
'List of items that match
xmlSourceList = xmlDoc.SelectNodes("//ProviderID[. = '860']/../../Name")
'Iterate through the list
For Each xmlSourceItem In xmlSourceList
  txtDisplay.Text += vbCrLf & xmlSourceItem.InnerText
Next

This works fine for retrieving the name, but I want to retrieve the Name and the ID. My xpath knowledge is pretty basic, so I don't know if there's a better way to specify it to get both. I can't use "../../*" and get everything due to... well, it's complicated, but I need to be able to specify precisely which named values to return.

I figured I could, if I wanted to, create two XmlNodeLists, one with "../../Name" and one with "../../ID", but I don't know how I could For Each iterate through both a the same time or how to keep them "synched."

Is there some other way to do it?

 

[chrissie1]

why is providerid in a separate node?



Christiaan Baes
Belgium

I just like this --> [Wiggle] [Wiggle]

 

[Genimuse]

Poor schema design. Unfortunately I don't have control over it.

 

[stsuing]

Why don't you get the customer node instead, Then you could get the name and ID nodes in the loop using SelectSingleNode method of the node object.

I have not tested this
xmlSourceList = xmlDoc.SelectNodes("//ProviderID[. = '860']/../../../Customer")


xmlSourceList.SelectSingleNode("Name").InnerText

 

[chiph]

Running another XPath query is fairly expensive. You should be able to navigate up one level via the ParentNode property, and then down again via the ChildNodes collection. This is much faster as there's no additional query run.

Or..
Just change your original query to not go down to the Name level:

"//ProviderID[. = '860']/../Customer"

and use the ChildNodes collection from there.

Chip H.


____________________________________________________________________
Donate to Katrina relief:

http://s1.amazon.com/paypage/PELYGQVJ8Q7IB/103-6821258-5919825

If you want to get the best response to a question, please read FAQ222-2244 first

 

[Genimuse]

Thanks, all.