Thanks Alan! I see what part of the problem here is. When you dumped the variables on the index page, it is coming through the call of the scanner.LoadTemplate, which has access to the full cfc. So, to test this. I created another function in the scanner.cfc which just returned "Hi" and then I called it from the index.cfm page with <cfoutput>#testing()#</cfoutput> and I get "Hi" on my page, so the index.cfm has full access to the scanner.cfc......
I guess that I'm going to have to think this through a little bit more and come up with a better solution. The strange thing still though, is that shouldn't each call to the function reset its local variables? For example, if I do a dump of the variables scope inside of my LoadTemplate function, each call to the function has the exact same variables. Shouldn't this be reset with each call? Thanks for the name info. I didn't realize that I had done that as I was throwing this together. On Tue, Apr 3, 2012 at 9:29 PM, Alan Holden <[email protected]> wrote: > It would be nice if we could post the salient parts of this code right in > the mailing list, if you don't mind. You might get a broader response that > way. > > NOTE: I get the same results from ACF9 as from OpenBD 2.02... *G G G > Ghost Vars!! RUN! AAAAyyyyeeeeee!!* > > I put this line <cfdump var="#variables#"> at the end of your index.cfm > file, and I can see all the things that CF considers "un-scoped". For > example, while you DO scope *application.Scanner* on startup, this object > has 2 methods within (LoadTemplate & LoadPage) that have popped up in the > "variables" scope at the time the dump is called. > > I think the issue might come from using the OnRequest() method as the > "code shell" for the whole page; also perhaps that the methods in * > scanner.cfc* are not well encapsulated. They read from external scopes > and write directly to the page, as opposed to [*returning values back - > from input arguments passed in*], like most well-behaved functions. > > Here are some quasi-related conversations I found - that might help: > http://forums.adobe.com/message/2285269 > > http://stackoverflow.com/questions/8188015/do-variables-inside-application-cfc-functions-need-scoping > http://forums.adobe.com/thread/604831 > > http://livedocs.adobe.com/coldfusion/8/htmldocs/help.html?content=appFramework_04.html > > Finally, your application is named "Scanner", and you've also created an > object named "Scanner" within it. Despite the referential separation by > tiny dots, some people would cringe and shudder at such a naming practice. > > Al > > > > On 4/3/2012 8:29 AM, Benjamin Davis wrote: > > Hey Alan, > > Thanks for the info. Attached is a zip file with the basic of basics. > In this example, the index.cfm file sets a variable and then when you to > go page2.cfm, the variable is displayed, even though that page never > specifically sets the variable. Let me know what you think. > > Thanks! > > Ben > > > On Mon, Apr 2, 2012 at 5:54 PM, Alan Holden <[email protected]> wrote: > >> It certainly seems like something (perhaps that function) is definitely >> causing the variables scope to go zombie on you. But you may need to >> whittle your code down to a core sample and actually post that - to get the >> greatest qty of help. >> >> Alan Holden >> (at work) >> >> >> >> On 4/2/2012 2:35 PM, Ben wrote: >> >>> So, I have an interesting case that needs some external thought. I've >>> created a "framework" that handles all of my security as well as combining >>> the page that is requested into a template file. I've made this into a cfc >>> that is loaded into the Application Scope. In my OnRequest function, I >>> call another function of my framework that loads any external files, the >>> requested file, and then the template file where everything gets combined. >>> This has been working great, but then I realized that if I set a new >>> variable on one page I can output it on another page even though I haven't >>> declare it on the second page. For example: >>> >>> index.cfm: >>> <cfset Tempvar = 'Whatup'> >>> >>> about.cfm >>> <cfoutput>#Tempvar#</cfoutput> >>> >>> On the about I get Whatup even though I haven't declared the Tempvar on >>> the about.cfm page. It appears like it is caching the variable somewhere. >>> I have debugging turned on and if I search the debugging output for >>> Tempvar there is nothing. I'm not sure how or where it is caching this. >>> Any ideas? Is this because I am loading the page through a function that >>> is stored in the application scope? >>> >>> Thanks! >>> -- >>> online documentation: http://openbd.org/manual/ >>> google+ hints/tips: https://plus.google.com/115990347459711259462 >>> http://groups.google.com/group/openbd?hl=en >>> >> >> -- >> online documentation: http://openbd.org/manual/ >> google+ hints/tips: https://plus.google.com/115990347459711259462 >> http://groups.google.com/group/openbd?hl=en >> > > -- > online documentation: http://openbd.org/manual/ > google+ hints/tips: https://plus.google.com/115990347459711259462 > http://groups.google.com/group/openbd?hl=en > > -- > online documentation: http://openbd.org/manual/ > google+ hints/tips: https://plus.google.com/115990347459711259462 > http://groups.google.com/group/openbd?hl=en > -- online documentation: http://openbd.org/manual/ google+ hints/tips: https://plus.google.com/115990347459711259462 http://groups.google.com/group/openbd?hl=en
