On 7/9/12 1:35 PM, Sławomir Skowron wrote:
If the object is in cache it's work ok, but when i don't have object
in cache i wan't Get this object.
When object is not in cache i need full direct head operation into
mapped backend, and then if necessary get object into cache.
Is this possible ??
I think it would be difficult, and potentially dangerous, for the
general use case. For example, if you convert the HEAD request to a GET
request to origin, how do you know that the response is going to be
cacheable? If it's not cacheable, you have now wasted a lot of resources
fetching the full object (with a GET) instead of the less resource
intensive HEAD request.
Also, we would need to convert the response from the GET to the origin
to just a "HEAD" response. This would be doable, but I don't think we
have anything in the code in there for that. And I suspect it's not a
priority, because there's no general use case where you can (and should)
do this sort of conversion?
Maybe someone else of our protocol jockeys have a different opinion?
-- Leif
