Re: [Laszlo-dev] Expires header from LPS server

[P T Withington]

Should we be using the more modern [cache-control](http:// 
www.mnot.net/cache_docs/#CACHE-CONTROL), in particular to mark the  
LFC as publicly cacheable?  Are there any clients out there any more  
that do not use HTTP 1.1?

On 2006-11-16, at 23:12 EST, Henry Minsky wrote:
Hmm, I needed to add code to the LFC responder to check the last  
modified
header. I believe I have a patch
now that doesn't reload the lfc or app code unneccessarily.
I'll send a patch for review.

I'm using Fiddler to watch the HTTP traffic..

On 11/16/06, Henry Minsky <[EMAIL PROTECTED]> wrote:

I'm looking at http://www.openlaszlo.org/jira/browse/LPP-3053
"Server sending expires headers which prevent javascript from  
caching"

And I see this code in  LzServlet that generates an Expires header:


    public void doGet(HttpServletRequest req, HttpServletResponse  
res)
        throws IOException, ServletException
    {
        // If no encoding is specified, force it. Note that  
encoding has
to be
        // set before request parameters are accessed.
        if (req.getCharacterEncoding() == null) {
            req.setCharacterEncoding(mCharacterEncoding);
        }

        // This forces clients to talk to us. We may still return
NOT_MODIFIED.
        // TODO: [bloch 2002-12-17] turn this into an lps.property
        if (! isMacIE(req) ) { // Set only if it's not Mac IE 5.2  
(see bug
811)
            res.setHeader("Expires", "Fri, 05 Oct 2001 00:00:00  
GMT");
        }


The implication is that even if the Expires dates has passed, the  
client
will respect
the NOT_MODIFIED return code (which is being returned by the  
server) and
not request a reload
of the page. The Firefox LiveHTTPHeaders trace looks like this

http://localhost:8080/legals/test/lpp-2976.lzx? 
lzt=object&lzr=dhtml&_canvas_debug=false


GET /legals/test/lpp-2976.lzx? 
lzt=object&lzr=dhtml&_canvas_debug=false
HTTP/1.1
Host: localhost:8080
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1)
Gecko/20061010 Firefox/2.0
Accept: */*
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://localhost:8080/legals/test/lpp-2976.lzx? 
lzr=dhtml&lzt=html
Cookie: JSESSIONID=53EFE301BDBF944484A05A39E7EBDAA2;
PHPSESSID=d3f2225533a1b1d7eb6acc0c6e023edb
If-Modified-Since: Fri, 17 Nov 2006 03:22:35 GMT
Cache-Control: max-age=0

HTTP/1.x 304 Not Modified
Expires: Fri, 05 Oct 2001 00:00:00 GMT
Date: Fri, 17 Nov 2006 03:27:47 GMT
Server: Apache-Coyote/1.1


It appears that the page is not being resent, so maybe this isn't  
a bug
after all?

Maybe I'd better install a proxy monitor tool and see what bits are
actually flowing..



--
Henry Minsky
Software Architect
[EMAIL PROTECTED]





--
Henry Minsky
Software Architect
[EMAIL PROTECTED]