Copy / Import XML Nodes Into A ColdFusion XML Document

Posted March 18, 2009 at 10:04 AM

Tags: ColdFusion

A long time ago, I described how to append XML nodes from one ColdFusion XML document into another ColdFusion XML document. My solution, at the time, required you to use the underlying Java methods of the ColdFusion XML document. As much as I like the concept of leveraging the underlying Java power, I do like my solutions to be as ColdFusion friendly as possible. As such, I wanted to see if I could come with a ColdFusion-only solution to the problem of importing XML nodes from one ColdFusion XML document into another ColdFusion XML document. Also, while the previous solution imported and appended in one call, I wanted my new solution to be only the import part such that the resultant nodes could be manipulated independently.

I thought about the kind of use cases in which this would be useful and I figured that I would most likely be passing in a given XML node tree (or sub-tree) or an array of nodes. A given node seemed like it would be the easiest since you can duplicate it and then recursively import its children. An array of nodes on the other hand is a bit more complicated because each array index could be a sub-tree of another index in the same array. For example, if you did an XmlSearch() for:

//*

The first node returned might be the parent node of the second node returned, in which case the first node would contain references to the second node. As you can see, when dealing with arrays of node references, there can be a lot of overlapping. Unfortunately, I could not come up with a way to handle this overlapping well. As such, when you import an array of nodes, each array index is treated as a completely separate tree.

The resultant ColdFusion user defined function, XmlImport(), handles the above two use cases:

 Launch code in new window » Download code as text file »

In the above UDF, I have the namespace functionality commented out because I never use XML namespaces. However, if you wanted to add it, you would just need to uncomment the two properties and change the XmlElemNew() method call. Unfortunately, there was no way that I could find to directly import an XML node, so, if you look at the recursive nature of the XmlImport() function, you will see that it is actually building a mirror copy of the node, not technically importing it. While less than elegant, the end result in the same.

To test this, I set up two ColdFusion XML documents:

 Launch code in new window » Download code as text file »

Now, that we have two XML documents set up, let's try to copy the nodes directly (to make sure the wrong way still doesn't work):

 Launch code in new window » Download code as text file »

As expected, when we run this, ColdFusion will not allow use to copy one node directly into another XML document and throws the following error:

WRONG_DOCUMENT_ERR: A node is used in a different document than the one that created it. null

To get around this, we need to import the nodes from the second XML document into the first XML document before we can use them:

 Launch code in new window » Download code as text file »

As you can see, importing the XML nodes from one document into another doesn't inherently do anything. The newly imported nodes are not automatically inserted into the target XML DOM; they are simply copied into the target XML document context. Once they have been imported, they can then be inserted into the new XML DOM. Running the above code, we get the following CFDump output:

As you can see, the girl nodes from the second XML document have been successfully imported into the first ColdFusion XML document and then inserted into the XML tree.

Download Code Snippet ZIP File

Post Comment  |  Ask Ben  |  Other Searches  |  Print Page

• Add To Del.icio.us
• Add To Digg
• Add to Spurl.net
• Add to Wists
• Add to Simpy
• Add to Newsvine
• Add to BlinkList
• Add to Furl
• Add to reddit
• Add to FARK
• Add to BlogMarks
• Add to Yahoo! My Web
• Add to smarking
• Add to Ma.gnolia
• Add to Segnalo

Reader Comments

Hi Ben,

I was trying to use your previous code just a week or two ago without much success. It may be because I was trying to do something it's not designed to do. Will your script allow you to import one set of nodes into another child? For example, if I have:

groups
--group
----people
------persons

Could I use your function to add more persons to the people node above?

Thanks for all you do.

Dan

@Dan,

Yeah, no problem. If you wanted to, you could build up the new children in a new CFXML tag (or something like that) and then append them.

Pseudo example:

<cfxml variable="xmlNewPerson">
. . . . <person>
. . . . . . . . <name>Ben Nadel</name>
. . . . </person>
</cfxml>

<cfset xmlPerson = XmlImport( xmlExistingTree, xmlNewPerson ) />

<cfset ArrayAppend(
. . . . xmlExistingTree.groups.group.people.XmlChildren,
. . . . xmlPerson
) />

Something like that.

Awesome Ben, I'll give it a try.

Thanks,

Dan

Ben,

I may be missing one or two details, but would this version work?

<cffunction name="XmlImport" access="public" returntype="struct" output="false" hint="I import the given XML data into the given XML document so that it can inserted into the node tree.">
<cfargument name="ParentDocument" type="xml" required="true" />
<cfargument name="Nodes" type="any" required="true" />

<cfset var ImportedNodes = arrayNew(1) />
<cfset var NewNode = arrayNew(1) />
<cfset var returnStruct = structNew() />

<cfif NOT IsArray(arguments.Nodes)>
<cfreturn Arguments.Nodes />
<cfelse>
<cfloop array="#ARGUMENTS.Nodes#" index="Node">
<cfset ArrayAppend(ImportedNodes, XmlImport(ARGUMENTS.ParentDocument, Node)) />
</cfloop>
</cfif>

<cfset returnStruct.importedXML = ImportedNodes />

<cfreturn returnStruct />
</cffunction>

Makes it a bit shorter and cleaner, methinks -- I could be missing something with the recursion though -- I seem to remember something about recursion being most efficient when the recursive call is the last thing that gets done, but it's been a while since Comp 15 at Halligan.

@Joe,

Ahhh, Halligan hall. That makes me sentimental :)

The problem with your solution is that you never actually import the node into the target XML document. If I passed in an XML node reference, your code would do this:

<cfif NOT IsArray(arguments.Nodes)>
. . . . <cfreturn Arguments.Nodes />
<cfelse>

This simply returns the reference back to the calling code. The node reference is still in the context of it's original XML parent document. The trick is that we have to create a mirrored element in the context of the target XML document, hence the XmlElemNew() call.

Great stuff Ben - your older solution saved me oodles of work a few days ago. Cheers.

Waow! That's good!!!

Post Comment  |  Ask Ben

Post A Comment

Comment Etiquette: Please do not post spam. Please keep the comments on-topic. Please do not post unrelated questions or large chunks of code. And, above all, please be nice to each other - we're trying to have a good conversation here.

Please review the following issues:

Author Name:
Author Email:
Author Website:
Comment:	Formatting: <strong>bold</strong> <em>italic<em>
	Remember my information
	Subscribe to entry *
	Send me a copy of this comment (as part of the comment thread)

 

* If you subscribe, any new posts to this thread will be sent to your email address.