Re: A question aboout Cache_HTTP_Request

> I had a closer look at Cache_HTTP_Request today - and it seems be to > quite what I've been looking for :-) > But I'm having a small question - maybe someone can help me: > I need to fetch data that are released via an official server 4 times > a year - so each 3 months. I don't know the exact dates though. And > as far as I know all cache-time are relative to the first execution > of a Cache_HTTP_Request, right? right > If so I got two proposals: > a) Would it be possible to make $expires not only being an integer > but to also allow it being an array? In there you could e.g. include > a specific date / time like. "expires-on" => "2003-08-30 01:00:00". > Or you might decide to make expires-on be a unix timestamp. > b) How about an option "re-validate"? If the content of the file has > not been modified since a certain date there is no need to download > it again. > (...) Why not cleaning the cache at given dates ? > But I got another problem: On some systems (with small internet- > connection - maybe an intranet-machine with only ISDN-dialup) getting > 5mb of data from the internet might take quite long. Therefor it > would be nice to use the currently cached data for the current > request but spawning off a separate task to download the latest > version. Do you have any idea how this might be accomplished? It's maybe too specific to be included in PEAR/Cache but IMHO, I would use a cron job called 1 time a month which will clean the cache if it's the good month. Moreover, this job would "call" the cache in background. Fabien MARTY