[thelist] query string manipulation

[Jeremy Ridout]

I tried it on IIS 4.0 about a year ago and it worked fine. I don't know if
IIS 5.0 changed anything or not. I havent heard that it did, so I suspect
that it still works.

The problem that I had was that IIS logfiles stopped at the index.cfm as
well. That means that the log file would show a gazillion hits to index.cfm
in the root directory, but you couldnt distinguish which REAL pages were
getting hit. I asked around on the FuseBox list but heard of no one else who
had that problem (either that or they didnt rely on IIS logs).

The other problem that we had was that some browsers would have problems
when the actual document that the fancy url pointed to was a PDF (or zip,
etc.). Most browsers handled it fine, however, some would either display
nothing or just handle the pdf as text and show the raw ascii. I don't
remember which browsers did this. The fix was to append a .htm or a .pdf to
the end of the query string depending on the type of page.

We have since gone to a method separating URL variables with commas
(www.mysite.com/1,2,3,4.html). It works SO much better and is, of course,
100% dynamic.




> > -----Original Message-----
> > From: thelist-admin at lists.evolt.org
> > [mailto:thelist-admin at lists.evolt.org]On Behalf Of Steven Gongage
> > Sent: Friday, September 07, 2001 10:42 AM
> > To: <
> > Subject: [thelist] query string manipulation
> >
> >
> >
> > I'm building a site from the ground up in Coldfusion 5 on a 2k
> > server running IIS 5.  I'm in the first stage of development
> > right now where I'm testing ideas and playing with new toys
> > before we get into the heavy stuff.  I came across an article
> > about how to "hide" query strings by replacing all "&" and "="
> > with "/", in other words making the entire URL look like a 
> long path.
> >
> > Example:
> > http://www.blah.com/index.cfm?fuseaction=querystringssuck&pageacti
> > on=addrecord&var1=123
> > http://www.blah.com/index.cfm/fuseaction/querystringssuck/pageacti
> > on/addrecord/var1/123
> >
> > Why would I be interested in this?  For one, I'm developing based
> > on the Fusebox model and using query strings to decide the page
> > to show really pisses of search engines.  Second, they're ugly to
> > the users (not that that is a real reason for the headache
> > involved).  Third, it looks interesting.
> >
> > The problem comes when I use this new URL it tries to actualy go
> > to that path.  The article I read on this explains how to get it
> > to work on Apache by saying anything that has index.cfm should be
> > read by index.cfm and therefore ignores the rest of the path,
> > which is then translated by index.cfm.  But in IIS I don't see
> > any way of keeping it from resolving the URL and going to the
> > full path other than defining all those "directories" and
> > redirecting them individualy.  I'm going to have WAY too many
> > directories to keep that strait.
> >
> > Has anyone tried this on IIS 5?  Is it even possible?  If not,
> > any alternatives?  Your feedback would be VERY helpful to me.
> >
> > Thanks,
> > SteveG
> > Webmaster
> > Osceola County IT
> >
> >
> > ---------------------------------------
> > For unsubscribe and other options, including
> > the Tip Harvester and archive of TheList go to:
> > http://lists.evolt.org Workers of the Web, evolt !
> >
> 
> 
> ---------------------------------------
> For unsubscribe and other options, including
> the Tip Harvester and archive of TheList go to:
> http://lists.evolt.org Workers of the Web, evolt ! 
>