XML Parse question

[Jfish 126]

Hello all. I have a question about XML parsing and xpath. I am extremely new to both so apologies if this is a basic question. I did search the forum and this link on w3schools as well as the MSDN documentation. I was working from (which was extemely helpful). Here is my problem: I can find a specific named node and then get some info from it which I need. However, I am having trouble finding that node with an alternative approach using "contains" to search for any node that contains a subset of the tag name. Here is the problem in code-speak:

#Include
#Include
$oXML=ObjCreate("Microsoft.XMLDOM")
$DirInput =@ScriptDir
$FileInput = "in-20120401.xml"
$sXmlFile =$DirInput & "\" & $FileInput
$oXML.load($sXmlFile)
$searchTerm="us-gaap:DeferredRevenueCurrent";
$XmlRootPath = "//"&$searchTerm
;
;LOOKING TO REPLICATE THIS RESULT USING CONTAINS FUNCTION:
;$searchTerm="us-gaap:DeferredRevenueCurrent";
;$XmlRootPath = "//"&$searchTerm
;dim $result2=$oXML.selectNodes($XmlRootPath)
;THE ABOVE WORKS ... BUT I CAN'T "FIND" IT USING CONTAINS as below
dim $result2=$oXML.selectNodes("//*[contains(.,'DeferredRevenueCurrent')]")
func _advancedQuery()
dim $xpathresult
For $xpathresult In $result2
ConsoleWrite($xpathresult.nodeName&@CRLF)
Next
EndFunc
_advancedQuery()

I am also attaching the XML file that I am using. Any help anyone could offer would be greatly appreciated.

in-20120401.xml

[jdelaney 313]

Right now, you are looking for a node like this <*>DeferredRevenueCurrent</*>

[Jfish 126]

Thank you. That helps me understand what the query is doing (I am way off). Could you also help me understand what it should be? I am trying to find nodes that contain "DeferredRevenueCurrent" not nodes whose value is "DeferredRevenueCurrent" - does that make sense? I am trying to simulate this: "//us-gaap:DeferredRevenueCurrent" but looking based on just the term DeferredRevenueCurrent. Any additional help you could offer would be much appreciated.

[jdelaney 313]

I'm not aware of any wild card usage on the node name, except for .nodename...but that would require you grab the collection of all possible parents, and looping through each of their children.nodename, and do a string comaprison (like StringInStr)

Edited March 19, 2013 by jdelaney

[Jfish 126]

So you always need to know the node names before you search unless you do as you suggest .... bummer. Thanks for the help. I really appreciate it.

[jdelaney 313]

It's probably possible to do node-name searches with wild cards, i've just never needed to find a way

Edited March 19, 2013 by jdelaney

[Jfish 126]

Actually the stringinstr worked pretty well - it was very fast.

if StringInStr($xpathresult.nodeName,"DeferredRevenue",1)<>0 then
ConsoleWrite($xpathresult.nodeName&@CRLF)

Thanks again for your help.

[Jfish 126]

@jdelaney - I have one more question for you. Is there a way to loadwait an XML doc? In other words, can I wait for this to finish somehow: $oXML.load($interactiveFilingURL)? Thanks for any help you can offer.

[jdelaney 313]

Not sure. I'd save the XML to a file, and use .loadxml

or, get the xml text, and use that in .load

[Jfish 126]

Thanks. I can do it locally but I need to add a sleep to get it off the net - it can be extremely short: sleep(100) - but it won't work without that. I am also trying to test for "true" on return with a loop - but it seems to return true before it is ready to parse by the smallest fraction of a second ...