Unfortunately, I am passing information to a legacy CGI script written
before I came on board. It will be my task to rewrite this script, but not
for a few months. Believe me, if I could use a custom module of my own, I
would. Then I could use pnotes() or any number of other tricks... *sigh*
And as for the duration of the environment variables... From what I have
read, changes to %ENV last only as long as the current request, after which
they are replaced by a fresh set. Incidentally, I have been trying to
figure out a way that I might use Apache::Table to accomplish what I am
trying to do, but so far I haven't found a panacea for this there either.
No one has addressed my other question, so perhaps you can help. If I use
an internal redirect, I could avoid all this as the browser never "sees" the
redirection, and thus my data would remain safe. But, I need to set a
cookie at redirection as well. Now, using a standard redirect, this is no
problem, as I can set the cookie in the outgoing header. However, using
internal redirection, I cannot send a header unless I use err_header_out().
But the problem there is, I am redirectiing to a script and not a module, so
how can I then set the cookie sent through err_header_out() from the CGI
script? Is that possible?
Jason Simms
>If you don't *have* to use CGI scripts, you can stick the "sensitive"
> >stuff into $r->pnotes and do an internal redirect to another
> >PerlHandler. The browser would never see the sensitive info (it would >be
>passed among handlers on the server end), so this would go a long >way to
>protecting the sensitivity of the info.
>
>If you start adding stuff to the server's environment, how long will >it
>stay there? Will it be available until you restart the server? >That could
>get very messy...
>
>darren
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com