I would imagine that attributes.message is a local copy of caller.message. This
means that a custom tag is free to play around with attributes.message and it
will not affect the calling script.
However, the caller scope is designed for communicating information back to the
calling script, and can in fact be used to access variables that weren't passed
into the custom tag at all.
The other thing to be aware of is that if you invoke a custom tag as follows:
<cf_mytag variable="#a_variable#"> then there are two distinct variables you can
access from "mytag" - attributes.variable and caller.a_variable.
If my understanding of this system is incorrect, I'm sure somebody will let me
know... ;)
David Cummins
Carol Chandler wrote:
>
> You're the only person who has taken a stab at this, so thank you very much!
>
> It is my understanding that caller.message and attributes.message point to the same
>thing, but that any modifications that refer to it as "attributes.message" will be
>strictly local to that invocation of the tag. Is anyone willing to help with my
>understanding of caller vs. attribute, without necessarily getting too involved in
>the headache-y recursion? *grin*
>
> Thanks,
> Carol
>
> >>> [EMAIL PROTECTED] 08/21/00 04:53PM >>>
> Hi, Carol.
>
> Perhaps your recursive invocation of cf_output_unknown needs to pass
> caller.message as the message= argument rather than the attributes.message.
> I may be getting wrapped around the recursion here, but it seems to me that
> within a given invocation of cf_output_unknown you modify caller.message
> with by appending the value of the argument test_object (or an element of
> it) to the value of the message argument. But then you pass the unmodified
> attributes.message to the next level of recursion. (Of course, if in the
> final analysis the caller and attributes scopes actually point to the same
> value because of the recursion then this advice doesn't mean much.)
>
> HTH,
> -- Tim Dempsey
>
> -----Original Message-----
> From: Carol Chandler [mailto:[EMAIL PROTECTED]]
> Sent: Monday, August 21, 2000 4:48 PM
> To: [EMAIL PROTECTED]
> Subject: Help with custom tag
>
> Hi list,
>
> I'm working on my very first custom tag. I was writing a specific routine
> to collect the contents of the CFCATCH structure when it occurred to me that
> a more general solution might be useful. So, what this tag is intended to
> do is to take a variable - simple, array, or structure, or any combination
> (array of structures of arrays, etc.) - and append all of the component
> names and values into one string. What I have here does correctly traverse
> and output a nest of structures/arrays, but I can't seem to collect it all
> into one variable. There must be more to "caller.variable" than meets the
> eye. What happens in the calling page is that "message" has the name of the
> last component of the outer structure.
>
> Can anyone tell me what I am missing? Thanks very much. (And I hope the
> recursion doesn't give anyone a headache!)
>
> <!--- CF_output_unknown [input] test_object [output] message ->
>
> <cfif IsDefined("attributes.test_object") IS "No">
> <cfset #caller.message# = "">
> <cfelseif #IsSimpleValue(attributes.test_object)#>
> <cfoutput>#attributes.test_object#</cfoutput>
> <cfset #caller.message# = #attributes.message# &
> #attributes.test_object#>
> <cfelseif #IsStruct(attributes.test_object)#>
> <CFLOOP COLLECTION=#attributes.test_object# item="struct">
> <cfset #caller.message# = #attributes.message# & #struct# &
> " = ">
> <cfoutput><BR>#struct# = </cfoutput>
> <CF_output_unknown
> test_object=#attributes.test_object[struct]# message=#attributes.message#>
> </CFLOOP>
> <cfelseif #IsArray(attributes.test_object)#>
> <CFLOOP index=i from=1 to= #arrayLen(attributes.test_object)#>
> <CF_output_unknown test_object=#attributes.test_object[i]#
> message= #attributes.message#>
> </cfloop>
> <cfelse>
> <Cfset #caller.message# = #attributes.message# & "Unable to proceed
> with CF_output_unknown.">
> </cfif>
>
> ----------------------------------------------------------------------------
> --
> Archives: http://www.mail-archive.com/[email protected]/
> To Unsubscribe visit
> http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or
> send a message to [EMAIL PROTECTED] with 'unsubscribe' in
> the body.
> ------------------------------------------------------------------------------
> Archives: http://www.mail-archive.com/[email protected]/
> To Unsubscribe visit
>http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
>message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
>
> ------------------------------------------------------------------------------
> Archives: http://www.mail-archive.com/[email protected]/
> To Unsubscribe visit
>http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
>message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
------------------------------------------------------------------------------
Archives: http://www.mail-archive.com/[email protected]/
To Unsubscribe visit
http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
