On 05/30/03(22:38) you "James R. Blaney" <firstname.lastname@example.org> wrote
|Thanks for the reply. If I use the following:
|it should handle the Expires tag, butI don't think that will work for the
|following headers, will it?
|> |2. <meta http-equiv="refresh" content="1800">
|> |3. <meta http-equiv="Pragma" content="no-cache">
|> |4. <meta http-equiv="Expires" content="Fri, 30 May 2003, 04:42:15
These tags contained in a HTML document to be used to generate HTTP headers
on server, or used in a client instead of header, affects only the header
of the HTML document which contains the tags. And you seemed not to care
the expiration of text like text/html: "Subject: expire text only?"
|I will probably give the external filter or CFI mechanism a try, because in
|our case, the overhead should still be far less than having to reload lots
|of images over a 56K connection.
I don't think removing these tags is not necessary nor affect the expiration
|Here's a suggested feature: When the proxy receives a request for any image
|in cache from an expired page, DeleGate issues a HEAD (information only)
|request about each image, to obtain size and date/time, and if these haven't
|changed, it reads image from the cache. In this way, many "expired" pages
|would only reload the html text, not the images. I believe this would make a
|huge difference in speed of page loading, and without requiring any changes
|to html tags.
There is If-Modified-Since header or so which are used with GET method to
omitt transmission of body data (header information only) if the original
data is not modified since it is cached. I think any client program
including browsers and proxies including DeleGate implement this feature.
See RFC1945(HTTP/1.0) or RFC2616(HTTP/1.1)
@ @ Yutaka Sato <email@example.com> http://www.delegate.org/y.sato/
( - ) National Institute of Advanced Industrial Science and Technology (AIST)
_< >_ 1-1-4 Umezono, Tsukuba, Ibaraki, 305-8568 Japan
Do the more with the less -- B. Fuller