I know some may frown on the idea of having a base “bean” class but that is exactly what I have settled on in order to support composite objects that consist of multiple beans. Basically there are two things I always want to be able to do with my bean:

  1. dump it
  2. get a memento

When I dump a bean though I don’t want all the stuff in the var scope but instead the variables.instance structure. Also, my memento is basically just a copy of the variables.instance structure (sort of).

The reason I say sort of is because what if my variables.instance.{somevar} = another bean? Then what I really want in its place is that beans memento (variables.instance struct).

Here is an example from a recent project. I have studentenrollment object. It consists of a few things:

  • student - an instance of student.cfc
  • address - an instance of address.cfc
  • class - an instance of theclass.cfc (named that way so i don’t accidentally call getClass())
  • entryassessment - an instance of assessment.cfc
  • interimassessment - an instance of assessment.cfc
  • exitassessment - an instance of assessment.cfc
  • a bunch of other things specific to that student+class relationship

So when I go to get the memento of the studentenrollment I really want the memento that is crafted from studentenrollment.variables.instance and all the composite objects variables.instance structs. To that end I added two methods to my base bean.cfc (that all of my beans extend).

<cffunction name="getMemento" access="public" output="false" returntype="struct">
  <cfreturn copyMementoStruct(variables.instance) />

<cffunction name="copyMementoStruct" access="private" output="false" returntype="struct">
  <cfargument name="memento" type="struct">
  <cfset var newMemento = structNew() />
  <cfset var keylist = structKeyList(arguments.memento) />
  <cfset var key = 0 />

  <cfloop list="#keyList#" index="key">
    <cfif isSimpleValue(arguments.memento[key])>
      <cfset newMemento[key] = arguments.memento[key] />
    <cfelseif isQuery(arguments.memento[key])>
      <cfset newMemento[key] = arguments.memento[key] />
    <cfelseif isStruct(arguments.memento[key])>
        <cfset newMemento[key] = arguments.memento[key].getMemento() />
          <cfset newMemento[key] = copyMementoStruct(arguments.memento[key]) />
    <cfelseif isXMLDoc(arguments.memento[key])>
      <cfset newMemento[key] = arguments.memento[key] />
    <cfelseif isWDDX(arguments.memento[key])>
      <cfset newMemento[key] = arguments.memento[key] />

  <cfreturn newMemento />

This returns me a nice structure that contains all of the appropriate memento information. Then my handy dump method works in conjunction

<cffunction name="dump" access="public" output="true" return="void">
  <cfargument name="abort" type="boolean" default="0" />
  <cfdump var="#getMemento()#" />
  <cfif arguments.abort>
    <cfabort />

What do you think? I think it works out pretty well but would love to hear the opinion of others.



odd question, but I was born in 1972 - so I’m a Rat.


How can you be a rat when you are 32yr? My calculation says you are an “Ox”, or “30yr”.