httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Roy T. Fielding" <field...@gbiv.com>
Subject Re: thoughts on ETags and mod_dav
Date Sat, 29 Dec 2007 20:44:41 GMT
This is getting ridiculous.  ETags are important to caching.
Caching is important to HTTP.  If you sum up all of the instances
of webdav in the world (including the ones my company sells), they
still amount to nothing compared to the rest of the uses of HTTP.

If you want strong etags for mod_dav, then do the right thing and
generate strong etags in mod_dav.  Replace its backend with a
versioned filestore (like subversion) and use the revision
number as the strong etag.

All this other nonsense about removing the weakness indicator
on weak etags, or just not supplying etags at all on weak
filesystem writes, is not going to happen.  We have weak etags
because our generic core filesystem backend doesn't need strong
etags anywhere near as much as it needs high performance.  That's
why modules that do authoring should not use the generic core
filesystem backend.  Until mod_dav learns that, just apply the
patch that Werner proposed

   http://issues.apache.org/bugzilla/attachment.cgi?id=20552

(and perhaps make it a hook that can be replaced by modules).
That is sufficient to fix the bug for dav.

If you want a better meets_conditions in general, then write a new
one with a new interface that can pass in the metadata for the
chosen representation (if any) and does not stink of stupid DAVisms
(like "resource exists" where it should say "representation exists").
Do not make patches to the old interface because those won't be
distributed until Apache 3.x.

If the weak etags are not being matched to the string etags on
GET, then that is another bug that must be fixed.  It is not an
excuse to ignore the HTTP design.

....Roy

Mime
View raw message