Yesterday, in the comments on my post about using XML with ColdFusion 8's new array iteration using CFLoop, Scott Bennett posted an example of using an XmlSearch() as the value being passed to the Array attribute in the CFLoop tag. I don't know why exactly, but this just struck me as quite cool (cool enough to get its own ColdFusion 8 post). Here is an example of what he was talking about:
- <!--- Create xml data. --->
- <cfxml variable="xmlGirls">
- name="Maura Tierney"
- name="Christina Cox"
- name="Sarah Vivenzio"
- name="Jodie Foster"
- name="Ashley Thomas"
- <strong>Brunette Girls</strong>
- Loop over all the returned xml nodes that are
- found at the following XPath value.
- array="#XmlSearch( xmlGirls, '//brunette' )#">
- <!--- Check to see if this girl is an actress. --->
- <cfif (
- StructKeyExists( xmlGirl.XmlAttributes, "isactress" ) AND
- <br />
As you can see, for the Array attribute of the CFLoop tag, we are directly passing in the node set returned by the XPath search, XmlSearch( xmlGirls, "//brunette" ). Running the code, we output only the array consisting of the brunette child nodes:
Maura Tierney [Actress]
This just feels like a much more elegant solution than trying to use the xmlGirls.girls.brunette pseudo array. Also, I think part of why I like it is that it feels much more parallel to the equivalent XSL Transformation:
- <xsl:for-each select="//brunette">
I like the symmetry. As a trade-off, you can't refer to the array itself without an intermediary variable, but for these kind of one-offs, it is perfect.
Thanks Scott Bennett!
I have been working on trying looping over data out of an XML file and choosing a particular node through XmlSearch.
XML File Structure:
<propertyList date="2008-01-25-10:02:00" >
<residential modTime="2008-01-25-10:02:01" status="current">
<img id="m" file="HI5160_1.jpg"/>
<img id="a" file="HI5160_2.jpg"/>
<img id="b" file="HI5160_3.jpg"/>
<img id="c" file="HI5160_4.jpg"/>
<img id="d" file="HI5160_5.jpg"/>
<img id="e" file="HI5160_6.jpg"/>
<img id="f" file="HI5160_7.jpg"/>
<img id="g" file="HI5160_8.jpg"/>
<img id="h" file="HI5160_9.jpg"/>
<img id="i" file="HI5160_10.jpg"/>
<img id="j" file="HI5160_11.jpg"/>
<img id="k" file="HI5160_12.jpg"/>
The only way i can get the desired results is this way:
<CFSET propID = getlist.listing_id>
<!--- Loop over the info-main nodes. --->
to="#ArrayLen( Listings.PropertyList.residential )#"
<!--- Get a short hand for the current info-main node. --->
<cfset xmlInfoMain = Listings.PropertyList[ "residential" ][ intMainIndex ] />
<cfif xmlInfoMain.uniqueID.XmlText EQ propID>
uniqueID = [<cfoutput>#xmlInfoMain.uniqueID.XmlText#</cfoutput>]<br />
<!--- Get the image-list children. --->
<cfset arrImageList = xmlInfoMain[ "images" ] />
<!--- Loop over the image list. --->
to="#ArrayLen( arrImageList )#"
<!--- Get a short hand for the current image-list node. --->
<cfset xmlImageList = arrImageList[ intImageListIndex ] />
<!--- Loop over image nodes. --->
to="#ArrayLen( xmlImageList.img )#"
. . . . img[ file = [<cfoutput>#xmlImageList.img[ intImageIndex ].XmlAttributes.file#</cfoutput>] [<cfoutput>#xmlImageList.img[ intImageIndex ].XmlAttributes.id#</cfoutput>]<br />
I'm sure there is a quicker way to do this using XmlSearch.
I have successfully grabbed the record out of the XML i need via this code:
<cfset arrResNodes = XmlSearch(Listings,"//residential[ uniqueID[ text() = '#propID#' ]]/") />
The only problem is i cannot loop over the results of XmlSearch?? Could you point me in the right direction? I can dump the output, just not loop over the img nodes.
I think you are really close here. To get directly at the array of IMG nodes, you can use this XPath:
//residential[ uniqueID[ text() = 'HI5160' ] ]//img/
where I put in the hard-coded ID for testing. Once you have that array, you just loop over it the same way you are already looping over the image nodes array.
After your excellent advice which worked like a treat with images, i need your advice on the following:
I have 1 main xml file another xml file with updates only.
I need to append or copy over the main xml with the update data.
I was using the XmlSearch theory: //residential[ uniqueID[ text() = 'HI5160' ] ]
and then trying to append the main file:
<cfset ArrayAppend(mydoc.PropertyList.residential.XmlChildren, XmlElemNew(myupdate,myupdate.PropertyList.residential))>
I would then write the new xml file.
Any further help with this would be appreciated.
Can i use XmlSearch within 1 file and append another file at the same time?
You can't just copy a node from one tree to another. You have to import it first. Check out this UDF that I mad to do that:
CF8 allows you to do the
<cfloop index="x" array="#myarray#">