http - (Weak) ETags and Last-Modified -

As far as I understand the specs, Itgong, which was introduced in RFC 2616 (HTTP / 1.1), a The successor is (finally for revised-header, which is proposed to give software-architect more control over the cash-modifying process.

If both cache-verification-headers (if-any-match And if-modified-since) are present in accordance with RFC 2616, the customer (i.e. browser) Should use, if a resource has changed, according to Section 14.26 of RFC 2616, if an ETA has been changed in an act-non-match header, then the server should not be modified in 304, and the server If the matches are presented, it should not be requested to do so, unless the date in the last modified headers does not say so. (If it matches in matches The server should answer with 304- GET- or HEAD- in the case of a request ...)

This section leaves the place for some speculation:

• A strong itag is considered to change "every", resource changes. Therefore, responding with something else has not been modified to request with an unchanged itgge and a modified-rat-header as 304, which is a bit of a paradox because it does not match the dose, because Strong eatag says that the resource was not changed, (However, this is not fatal because the server can only send the same unchanged resource.)
• ...

... o.k. When I was writing this, the answer to this question was boiling:

The above mentioned (small) paradoxes were made due to weak eartoges. A resource marked with weak istag can change, though not the tag. So, in the case of a weak stake it would be wrong, 304 to answer, not modified, when ETG has not changed, but if the date presented in the revised-does not match, okay?

In the scenario you exist, the file on the server can be new to the copy of the cache in the client - but the ETC matches are the same as the cached copy, so it would be acceptable to give 304 responses.