David Buchmann

i like where this is heading. we can provide a StringSerializer or something and use that as default implementation when no BodySerializer is specified (to stay BC, and to keep...

ups, i was mixing up things. you are right, there is no explicit purging. sorry for that confusion. then the only thing we could do is pass the expires after...

i am surprised why we then see the request at all, but not deep enough into the profiling to tell. is there a reason why we don't treat a cached...

is this still relevant? has it meanwhile been solved, or is it important to have this?

i wonder if we don't want to add the upper bound anyways. then at least in the future, we are more robust.

when this is released, update the changes from https://github.com/php-http/cache-plugin/pull/56

in https://github.com/FriendsOfSymfony/FOSHttpCacheBundle/issues/283 we found that we should be able to use the `--resolve` option of curl [http://serverfault.com/questions/443949/how-to-test-a-https-url-with-a-given-ip-address](to force the IP of a domain), which translates to [CURLOPT_RESOLVE in php](http://stackoverflow.com/questions/10433016/curl-resolve-equivalent-in-php-curl-library) ....

but that only delegates the problem to the matching client. we need to send the same request to various IPs, and we found that if HTTPS is used, the certificate...

in FOSHttpCache, this is what we currently do: https://github.com/FriendsOfSymfony/FOSHttpCache/blob/master/src/ProxyClient/HttpDispatcher.php#L188 its a very specific scenario, however. and i think all scenarios that require sending a request to multiple hosts are very...

i would vote that we implement overwriting then. if you use a different cache instance per service you can already use different paths, and ifnot the PR @Nyholm did will...