ah yeah, it's not when i view the error in the browser, which i've not
tried as i've never been able to replicate the error.  it's when i later
view the application.log in notepad.

maybe i should just use url parameters that i know won't be interpreted by
IE incorrectly, but I don't know where i'd get a complete list of those...


Duncan Cumming
IT Manager

http://www.alienationdesign.co.uk
mailto:[EMAIL PROTECTED]
Tel: 0141 575 9700
Fax: 0141 575 9600

Creative solutions in a technical world

----------------------------------------------------------------------
Get your domain names online from:
http://www.alienationdomains.co.uk
Reseller options available!
----------------------------------------------------------------------
----------------------------------------------------------------------


                                                                                       
                             
                    "Stephen                                                           
                             
                    Moretti"              To:     <[EMAIL PROTECTED]>                  
                   
                    <[EMAIL PROTECTED]        cc:                                      
                                 
                    ter.co.uk>            Subject:     Re: [ cf-dev ] Page Parameters  
                             
                                                                                       
                             
                    08/07/2003                                                         
                             
                    11:01                                                              
                             
                    Please respond                                                     
                             
                    to dev                                                             
                             
                                                                                       
                             
                                                                                       
                             



Sounds like the browser is interpreting your query string as html when
displaying the log entry....

Its something I keep seeing over years and a quick glance at the page
source
will tell you if the query string really does have a pound sign in it or
its
just the browser turning &prod into the escaped characters that it
represents.

Officially this shouldn't happen, because these escape codes should be
completed with a semi-colon eg. &prod;, but MS being MS took it upon them
selves to ignore that fact and if it sees an escape code in some text
without the semi-colon it converts it for you even though it shouldn't....

Regards

Stephen
----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, July 08, 2003 10:43 AM
Subject: RE: [ cf-dev ] Page Parameters



on a side-note to this, has anyone ever had problems where a url parameter
seems to get interpreted as a html character entity?  e.g. you've got a
page with a parameter called Product, the url might be
page.cfm?action=jump&product=123.
then an error appears in application.log, which indicates "&prod" has been
turned into whatever that's meant to be, in the same way as &amp; or
&pound; might get turned into & and £.  And the CGI.Query_String looks like
page.cfm?action=jump.uct=123, where the . is actually some odd character
that doesn't get printed properly in the application.log, but is obviously
the 'special character' it represents.  So the page chokes as all it's got
is URL.jump[specialchar]uct, instead of URL.jump and URL.product.  Oddly I
think this happens even though the page would have <cfparam> checking for
them both.

I've never been able to duplicate this error, and it happens very rarely.

I assumed it was because of having things like <a href
="page.cfm?action=jump&product=123"> which of course should be written as
<a href="page.cfm?action=jump&amp;product=123">, as any fule kno.  However,
on a recent site I had already changed all the url's to the correct
versions, with the exception of some javascript document.location.href
= "blah" stuff, where I figured it didn't make sense to turn & into &amp;.
Still got this error on that site, once, don't know why.

Not sure if it's just &prod, but I think it could happen with others too.
it's not a big problem as it so rarely happens, but it would be nice to
know i could do something simple to prevent it ever happening in the first
place.


Duncan Cumming
IT Manager

http://www.alienationdesign.co.uk
mailto:[EMAIL PROTECTED]
Tel: 0141 575 9700
Fax: 0141 575 9600

Creative solutions in a technical world

----------------------------------------------------------------------
Get your domain names online from:
http://www.alienationdomains.co.uk
Reseller options available!
----------------------------------------------------------------------
----------------------------------------------------------------------



                    "Aidan Whitehall"
                    <[EMAIL PROTECTED]        To:
<[EMAIL PROTECTED]>
                    nks.co.uk>                    cc:
                                                  Subject:     RE: [
cf-dev ] Page Parameters
                    08/07/2003 10:30
                    Please respond to dev





> erm, use the URL scope/structure or the CGI vars

Just to clarify, if your URL looks like this:

 index.cfm?page=this&somvar=that

in the code in index.cfm, you can refer to

  #url.page# and #url.somevar#

Alternatively, you can output the entire querystring using
#cgi.query_string#

You put a question mark after the template name and then ampersands
between variable-value pairs, btw


--
Aidan Whitehall <mailto:[EMAIL PROTECTED]>
Macromedia ColdFusion Developer
Fairbanks Environmental Ltd  +44 (0)1695 51775
Queen's Awards Winner 2003 <http://www.fairbanks.co.uk/go/awards>

________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________

--
** Archive: http://www.mail-archive.com/dev%40lists.cfdeveloper.co.uk/

To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
For human help, e-mail: [EMAIL PROTECTED]






--
** Archive: http://www.mail-archive.com/dev%40lists.cfdeveloper.co.uk/

To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
For human help, e-mail: [EMAIL PROTECTED]




--
** Archive: http://www.mail-archive.com/dev%40lists.cfdeveloper.co.uk/

To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
For human help, e-mail: [EMAIL PROTECTED]






--
** Archive: http://www.mail-archive.com/dev%40lists.cfdeveloper.co.uk/

To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
For human help, e-mail: [EMAIL PROTECTED]

Reply via email to