That is where caching comes into play. Fetch the data from Twitter once every [n] minutes. Store it locally (in some database on your server). When a request for a certain category comes, you pull it all out of your local DB. User requests never wait for Twitter. Twitter never gets hammered. Everyone goes home happy.
-- Patrick Burrows http://www.CleverHumans.com On Thu, Apr 30, 2009 at 1:21 PM, Borja Martín <[email protected]> wrote: > the possible hammering could come when I select a certain category and then > I send a request to fetch the status for each account I have assigned to > this category... > > regards > > > On Thu, Apr 30, 2009 at 6:24 PM, P Burrows <[email protected]> wrote: > >> If you are just pulling all this information from your own account's >> friends timeline, how would you be hammering the twitter API? I can't >> imagine you'd need to do that more than once a minute (tops!). That still >> leaves you with 40+ additional calls / hr you can make. >> >> Cache all the data you get back locally, of course. >> >> As far as splitting up the categories -- just grab everything you haven't >> already grabbed from your timeline and do the categorization locally. >> (twitter doesn't actually have categories, unless you use different accounts >> -- but that would be kinda silly.) >> >> -- >> Patrick Burrows >> http://www.CleverHumans.com >> >> >> >> On Thu, Apr 30, 2009 at 5:05 AM, Borja Martín <[email protected]> wrote: >> >>> Hi, >>> I would like to develop some kind of mashup by categorizing a few twitter >>> profiles(the idea would be something like >>> http://tweetminster.co.uk/mps/party/LD) >>> the problem is that I'm not sure how to focus it without hammering the >>> twitter api servers. the first idea was just to perform a search for each >>> profile my application had but this could eventually make a lot of requests. >>> then, the obvious approach would be to use the friends timeline of my own >>> account(we plan to update it, not just leech the content) but then I'm not >>> sure how to search for statuses within a certain category I would have >>> defined before >>> any idea how to approach this? maybe I should organize the content >>> fetched from the friends timeline as far as the api doesn't offer a direct >>> solution? >>> >>> thanks in advance >>> >>> -- >>> def dagi3d(me) >>> case me >>> when :web then "http://dagi3d.net" >>> when :twitter then "http://twitter.com/dagi3d" >>> end >>> end >>> >> >> > > > -- > def dagi3d(me) > case me > when :web then "http://dagi3d.net" > when :twitter then "http://twitter.com/dagi3d" > end > end >
