> I've found many references to this while googling around, but
> no definite
> answers. I'm wanting to get the refering page information in
> a perl/cgi
> script and base the flow of the script on it. That is, if
> the request is
> from a refering page that is 'authorized', the script
> continues on with it's
> normal flow, but if the refering page is not authorized, the
> script will
It's best to use authentication like .htacess and have them use a user/password to get
info.
Then you can base your output on their username.
> exit. I've tried using the HTTP_REFERER variable, but in all
> cases it returns
> an empty value regardless of the browser or the address the request
> originates from. The old standard CGICount program seems to
> handle this quite well, allowing only domains/pages entered
> in it's configuration to
> access the counter, but this program is written in C and I
> don't see any
> obvious correlations in it's source code that I can make
> sense of and try to
> port over to Perl. Is there any other means of getting the referer
> information other than through the $ENV{} variables?
In my experience only $ENV{'REMOTE_ADDR'} ever gets set consistently and
that's the ip address the request is coming from. Problem is with that it can be
spoofed too although a bit harder,
It can be dynamic so it may be different everytime for the same user, and multiple
websites can share an ip address.
I'd like to know also if there's another way to get a more reliable piece of data
about where the reauest comes from but so far REMOTE_ADDR is the best I can find.
HTH
Dan
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
