Received: by taz.hyperreal.com (8.6.12/8.6.5) id NAA10082; Wed, 10 Jul 1996 13:05:14 -0700 Received: from fully.organic.com by taz.hyperreal.com (8.6.12/8.6.5) with ESMTP id NAA10077; Wed, 10 Jul 1996 13:05:11 -0700 Received: from localhost (brian@localhost) by fully.organic.com (8.6.12/8.6.12) with SMTP id UAA22934 for ; Wed, 10 Jul 1996 20:09:17 GMT X-Authentication-Warning: fully.organic.com: brian owned process doing -bs Date: Wed, 10 Jul 1996 13:09:17 -0700 (PDT) From: Brian Behlendorf To: new-httpd@hyperreal.com Subject: Re: SSI's In-Reply-To: <199607101818.OAA32278@hershey.ai.mit.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-new-httpd@apache.org Precedence: bulk Reply-To: new-httpd@hyperreal.com Maybe I wasn't clear, I would prefer that this be a server directive, not an SSI directive, along the lines of "CacheNegotiatedDocs". Brian On Wed, 10 Jul 1996, Robert S. Thau wrote: > Except that the headers are sent before the document is parsed. Unless > you're using chunked transfer-coding, in which case you can use footers. > But nothing does, yet. > > Umm... we actually need a reliable last-mod date before starting to > send a response to the client at all, since it may change the status > code for the response (specifically, if the client did a conditional > GET, then the value of set_last_modified may change the correct > response code). > > What might make sense is to provide for a class of directives along > the lines of which are only permitted in the > , and which we would parse for before actually starting to > send the response. (So, the sequence would be... > > parse_head(...); > if ((status = set_last_modified(...)) != OK) > return status; > send_parsed_content(...); > > or something like that...). > > rst > > --=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-- brian@organic.com www.apache.org hyperreal.com http://www.organic.com/JOBS