Looking at the sample XML below ...

<Products>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>12345</Value>
               </Identifier>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>43512</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>54321</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>12112</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
</Products>

How can I select a group of ProductsItem nodes where the count of that group's Identifier nodes is less than say four? Looking at the above example I would want my result to be only the first two ProductsItem nodes because of count of the Identifier node within those first two ProductsItem nodes is equal to three.

I've been experimenting with ...

//ProductsItem[count(Identifiers/Identifier) &lt; 4]

Unfortunately, the avove expression brings back all ProductsItem nodes in the XML. I only want the first two ProductsItem node because the combined count of the first two ProductsItem/Identifiers/Identifier nodes is three which is essentially less than four.

Any help would be greatly appreciated.

Thanks,

Ben

Hello,

Are you saying that you only want the first three ProductsItem elements, and not all ProductsItem with less than three Identifiers?

Brian

Brian,

Because the XML will be constantly changing in terms of the number of ProductsItem nodes and corresponding child Identifier nodes, I want to be able to select as many ProductsItem nodes as possible without exceeding a maximum count of three Identifier nodes in the selected ProductsItem nodes. Wheew ...

Here is another example ... If I have ...

<Products>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>43512</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>54321</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>12112</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
     <ProductsItem>
          <Identifiers>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>21213</Value>
               </Identifier>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>21334</Value>
               </Identifier>
               <Identifier>
                    <Type>ECCN Nbr</Type>
                    <Value>21456</Value>
               </Identifier>
          </Identifiers>
     </ProductsItem>
</Products>

I only want the first three ProductsItem nodes because the total count of Identifier nodes in the first three ProductsItem nodes is equal to 3 which is less than four. Then, I would wan the next (4th postion) ProductsItem node by itself because it has a total of three Identifier nodes which is less than four.

Thanks for the quick response,

Ben

Hello,

You may be able to use the position() method to do this. Maybe try:

//ProductsItem[./Identifiers/Identifier/position() &lt; 4]

The way to think of what the position() method does is it groups all the results together and numbers them 1-n, and you can use this to limit the resultset in that manner. I'm a little rusty in my XML lately, so I may be a little off, but I hope this gets you in the right direction.

Brian