[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: lynx-dev internal cache proposal
From: |
David Woolley |
Subject: |
Re: lynx-dev internal cache proposal |
Date: |
Thu, 19 Nov 1998 00:07:22 +0000 (GMT) |
> this is equivalent to "expired", sending conditional GET with "no-cache"
> is enough (was explained below).
This is actually a rather weird combination; it should either be an
unconditional request with no-cache, to force an end to end reload,
or a conditional request with max-age=0, to force end to end revalidation.
I've wondered about this in the past, but Lynx really needs at least
two and preferably three cases:
1) force revalidation (override history mechanism/refresh current page);
2) force end to end reload (an intermediate cache seems to have a corrupt
copy);
3) force end to end revalidation (option: as (1) but override any policy
by an intermediate cache to defer revalidation).
I'm not sure which of these is currently being generated.
> >> Method POST may be not cached for internal use
> >> like changing ^V etc.
>
> > Why not?
> it takes a time to understand what current code do with it.
A true cache may never cache POST as POST is expected to have side effects
and caching would prevent them. A history mechanism would normally not
resubmit a POST.