Re: Is Expires fix on hem.passagen.se in 1.1.22?

Henrik Nordstrom wrote:
>
> Dancer wrote:
> > I already have code for 1.1.(20-ish) that prevents implicit or
> > explicit refreshing of objects that match certain refresh_pattern
> > rules (magic 'max' age of 1).
>
> Sounds like a handy feature if implemented in a nicer manner (i.e. use a
> explicit option keyword instead of magic values).

What I might do is exactly that. The underlying mechanics rely on the
same mechanics as refresh_pattern in any case, but a different keyword
(immortal/never_stale/whatever) would probably make things a lot
clearer.

> > 2) I subscribe to unofficial RFC -1: "Know when to break the rules. Do
> > it carefully. Do it cleanly. Do it knowing what you are letting
> > yourself in for."
>
> I agree with this, and this is why I don't like the current (1.1.22,
> 1.2b22) refresh_pattern behaviour. It was to easy to override Expires:
> without being aware that this is a protocol violation (or even that it
> is overridden)

Agreed. If we override it, we should do it because it's what we intend.

> ---
> Henrik Nordström
> Sparetime Squid Hacker

-- 
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GAT d- s++: a C++++$ UL++++B+++S+++C++H++U++V+++$ P+++$ L+++ E-
W+++(--)$ N++ w++$>--- t+ 5++ X+() R+ tv b++++ DI+++ e- h-@ 
------END GEEK CODE BLOCK------