RE: [CFCDev] Page context bug workaround

Tue, 25 Mar 2003 21:26:07 -0800

Title: Message
Steve:
 
Sorry for my abrupt reply -- seemed like a complete answer to your question ;)
 
As far as I know MACR engineering is very aware of the "page context bug" in CFCs and has been looking into a fix for some time (at one point I feel like there was a public announcement by someone at MACR on one of the lists that it was going to be fixed in U3, but alas...).
 
The issue is that a CFC instance holds onto the page context from the request in which it was instantiated.  The result is that you cannot reference application/session/request variables properly from within a CFC instance if you are using it in a request different from the one in which it was instantiated.
 
More interestingly, IMO, is the bug that I think is somehow related to the page context bug that the "variables" scope inside a CFC does not refer to a private scope as many would expect, but the variables scope of the calling page.  I think this is a different but related issue, but I'm not sure.
 
Is that the info you were looking for?
 
 
 
 
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]On Behalf Of Steven Erat
Sent: Tuesday, March 25, 2003 8:54 PM
To: '[EMAIL PROTECTED]'
Subject: RE: [CFCDev] Page context bug workaround

Ah... I should never try to compose an email after a coding session without first decompressing.  ;-)
 
I was asking the list if the pagecontext bug that had been mentioned often, but not elaborated on, was in fact the same bug that you pointed out was fixed in U3.
 
Nathan replied just "no" ... my fault for asking a closed-ended question. ( and a poorly written one)
 
Brandon Purcell elaborated a bit off list, but we still didn't clear it up, yet.
 
Either way, I believe they agree that this sore spot is not the getPageContext().forward("page.cfm") issue.  I'm just looking for either a bug number or a good explanation so I can either followup on the bug status or log one if necessary.  Then, if needed, I can try to advocate for a bug fix as soon as possible on behalf of the list.
 
-Steve
 
 
-----Original Message-----
From: Charlie Arehart [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 25, 2003 9:57 PM
To: [EMAIL PROTECTED]
Subject: RE: [CFCDev] Page context bug workaround

Steve, did you leave off a keystroke or two? Is that a question or a statement?
 
I had pointed out earlier today that the forward() method was now working. Just not clear what you're trying to say here. :-)
 
/charlie
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]On Behalf Of Steven Erat
Sent: Tuesday, March 25, 2003 5:32 PM
To: '[EMAIL PROTECTED]'
Subject: RE: [CFCDev] Page context bug workaround

Is the page context problem that has been widely discussed today?
 
 
CFMX Updater 3 Release notes: Issues fixed in this release
 
Using getpageContext().forward('pagename.cfm') to do a server side redirect threw an err.io.short_read error if there is any data in the Form scope.
 
Steven Erat
Macromedia
-----Original Message-----
From: Benoit Hediard [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 25, 2003 5:15 PM
To: [EMAIL PROTECTED]
Subject: RE: [CFCDev] Page context bug workaround

You're right Nathan.
Your method during the constructor is more elegant than cf_reinit and generates less overhead than cf_reinit.
 
Hope that MM will solve this page context problem in an updater...
Till then, we might have to replace the "temporarly" cf_reinit bug fix by this workaround approach for production use!
 
Benoit Hediard

Reply via email to