[6619] in www-talk@info.cern.ch
Re: Last-modified date & indexing
daemon@ATHENA.MIT.EDU (Nick Arnett)
Fri Nov 11 11:38:51 1994
Date: Fri, 11 Nov 1994 17:17:34 +0100
Errors-To: listmaster@www0.cern.ch
Reply-To: narnett@verity.com
From: narnett@verity.com (Nick Arnett)
To: Multiple recipients of list <www-talk@www0.cern.ch>
At 4:45 AM 11/11/94, Roy T. Fielding wrote:
>> I think I should know the answer, but where is an HTTP server supposed to
>> get the HTTP last-modified date? Is it from the file system (which
>> produces different results on various OSes), from the HTML header, or...?
>
>That is entirely up to the server. For the usual file objects, it is just
>the file system last-mod date -- the server should be aware of the limitations
>of the OS under which it runs. For objects with includes, it may be the
>most recent of the set of last-mod dates of its component parts. For
>database gateways, it may be the last-update timestamp of the record.
>For virtual objects, it may be the last time its internal state changed.
>In any case, the HTTP client should only know (and care) about the result --
>whatever gets stuck in the Last-modified: header -- and not worry about
>how it was obtained.
>
>However, under no circumstances should it ever come from the HTML header.
>That mechanism was never intended to provide values for metainformation
>that can be (and is) more readily obtained elsewhere.
That last part is a relief... ;-)
I guess that we really have no good choice, then, but to ask the server for
last-mod date for our indexer, if we want to be certain that it's
consistent. That rules out file-system-based indexing as a generalized
approach, unfortunately. It's also an argument for building the indexing
into the server.
Hmm.
Thanks for the explanation!
Nick
