Re: Cache control

2000-05-26 Thread Tom Mornini

On Fri, 26 May 2000, Nick Tonkin wrote:

> In our experience the only thing that guarantees no cacheing is returning
> 302. Also appending a query string (even an empty one) to the redirect URI
> seems to make it even more solid.
> 
> So in your script you would build up your destination URI, append
> '?foo=bar' to the end of it, and return it as a redirect. That way not
> even MSIE will cache it.

Though you walk the Earth, you may indeed be a God! Thank you very much.

I wish I would have thought of this myself.

-- 
-- Tom Mornini
-- InfoMania Printing and Prepress




Re: Cache control

2000-05-26 Thread Nick Tonkin


In our experience the only thing that guarantees no cacheing is returning
302. Also appending a query string (even an empty one) to the redirect URI
seems to make it even more solid.

So in your script you would build up your destination URI, append
'?foo=bar' to the end of it, and return it as a redirect. That way not
even MSIE will cache it.

HTH,

Nick


On Fri, 26 May 2000, Tom Mornini wrote:

> I have written a handler for delivering select images for a particular
> class of URIs based upon a database item that the user sets in advance.
> This allows us to "skin" some HTML allowing the user to select their look
> of choice.
> 
> The problem I have is that IE5 (and perhaps other browsers and versions)
> but NOT Netscape 4.x have absolutely no respect for cache-control.
> 
> My code does this:
> 
>   $r->no_cache(1);
> 
>   # some code to calculate a date 10 years ago (!) goes here...
> 
>   $r->header_out(
> 'Expires' => "$day_text, $day $mon_text $year $hour:$min:$sec GMT"
>   );
> 
>   $r->send_http_header($subreq->content_type);
> 
> And the HTTP headers via Lynx look like this:
> 
> HTTP/1.0 200 OK
> Date: Fri, 26 May 2000 08:58:38 GMT
> Server: Apache/1.3.12 (Unix) mod_perl/1.23
> Pragma: no-cache
> Cache-control: no-cache
> Expires: Tue, 29 May 1990 1:58:38 GMT
> Content-length: 1332
> Content-Type: image/gif
> 
> But IE5 still caches the damn things! Does anyone have a suggestion?
> Please note that I don't have the ability to modify the IMG SRC
> dynamically since the HTML resides on our customers' servers.
> 
> -- 
> -- Tom Mornini
> -- InfoMania Printing and Prepress
> 
> 


- nick





[OT] Re: Cache control

2000-05-26 Thread Drew Taylor

Tom Mornini wrote:
> 
> I have written a handler for delivering select images for a particular
> class of URIs based upon a database item that the user sets in advance.
> This allows us to "skin" some HTML allowing the user to select their look
> of choice.
> 
> The problem I have is that IE5 (and perhaps other browsers and versions)
> but NOT Netscape 4.x have absolutely no respect for cache-control.

> But IE5 still caches the damn things! Does anyone have a suggestion?
> Please note that I don't have the ability to modify the IMG SRC
> dynamically since the HTML resides on our customers' servers.
I don't yet have a suggestion, but I have the same problem with IE. I'm
building a web-based admin section, and IE keeps caching everything. One
more reason to hate IE - on IE 5 Mac they changed the default resolution
to 96 dpi (up from the std 72 dpi). So now all our style sheets for the
mac side have to be tweaked, and I have a lot of code to rework to send
the proper stylesheet! Microsoft seems to be very partial to standards -
CSS1 is completely implemented in IE5 mac, but it still caches dynamic
content!

Anyway, sorry for the OT post. I just had to vent... :-)

-- 
Drew Taylor
Vialogix Communications, Inc.
501 N. College Street
Charlotte, NC 28202
704 370 0550
http://www.vialogix.com/



RE: Cache control

2000-05-26 Thread Eric Jain

> The problem I have is that IE5 (and perhaps other browsers
> and versions)
> but NOT Netscape 4.x have absolutely no respect for cache-control.

IE5 can be set up to ignore any cache directives and keep a document
for either the duration of the session or forever. (Or reload it every
single time or check if it should be reloaded...)

This might be the cause of its respectlessness :-)


--
Eric Jain