Main consideration is if the APIs aren't identical, this won't work out of the box. At this point, it is not so much a GetTwitter as it is Get<Data Provider> that has their own schema. Anecdotally, I have a bit more than a foggy recollection that DataSift's API was accessed and possessed a different "schema" than Twitter's HBC.
Code wise, there would be some work needed to integrate custom endpoints. Again, this is in view of, and secondary to the previous consideration of resource matching. On Tue, Dec 8, 2015 at 9:10 PM, Joe Witt <[email protected]> wrote: > Is there any reason we'd not just let the user enter an override URL > to pull from? > > On Tue, Dec 8, 2015 at 9:02 PM, Bob Zhao <[email protected]> wrote: > > Thanks for the quick response, Aldrin! > > > > I know we are not NSA. But there are at least 2 3rd parties who provide > the > > Firehose stream(paid). > > Does NIFI have any plan to add their API later? After playing with the > > GetTwitter on the sample stream, I really like the way that NIFI handles > > these? > > > > Thanks, > > Bob > > > > ________________________________ > > From: [email protected] > > Date: Tue, 8 Dec 2015 20:51:21 -0500 > > Subject: Re: The EndPoints of GetTwitter > > To: [email protected] > > > > > > Bob, > > > > The Firehose stream is not publicly available via Twitter and requires > > special authorization to utilize [1]. As a result, this endpoint was not > > incorporated into the GetTwitter processor's properties and there is no > way > > to provide the totality of the public feed in NiFi. > > > > [1] https://dev.twitter.com/streaming/firehose > > > > On Tue, Dec 8, 2015 at 8:24 PM, Bob Zhao <[email protected]> wrote: > > > > Hello, > > > > I want to know which provider is the Firehose Endpoint pointing to? > > Or how could I get the 100% public content through NIFI? > > > > > > Thanks, > > Bob > > > > >
