Thanks for the answer, i'll be considering your suggestion for further documentation.
For your second question, there no policy for doing this, i want just to that from a given list of ID (from a text file maybe with an ID for each line), the structure of the link is the same (i.e: graphs.facebook.com/user+id only the id changes). Here is the comments example. i have many people who commented the post, i need to do an HTTP request for each one. So i need to invoke not one HTTP request but many after each retrieval of one post comment. Kindly. On 1 December 2015 at 17:07, Joe Percivall <[email protected]> wrote: > Hello Kacem, > > For your first question you can set up a relatively simple flow to achieve > this. Assuming the first retrieval is a GET method call you can do: GetHTTP > -> ExtractText -> InvokeHTTP. What this does is, it first gets the JSON > files from the social network using GetHTTP. Then with ExtractText you > extract the values you are interested in using in the last Invoke (in other > words add the values as attributes of the FlowFile). Then in InvokeHTTP use > the attribute expression language [1] to form your request. The docs for > the processors can be found here[2]. > > For the second scenario, are you specifically asking how to create calls > that hit the ids 1-1000 in a round robin fashion or is there a specific way > you determine the ids/scheduling? > > [1] > https://nifi.apache.org/docs/nifi-docs/html/expression-language-guide.html > [2] https://nifi.apache.org/docs.html > > Welcome to NiFi! > Joe > > > - - - - - - Joseph Percivall > linkedin.com/in/Percivall > e: [email protected] > > > > > On Tuesday, December 1, 2015 10:35 AM, "BOUKRAA, Kacem" <[email protected]> > wrote: > > > > > > Hello everyone, > So my case is as the following, i'm trying to retrieve JSON files from a > social network (Facebook), and based on some values in the file of the HTTP > response i want to invoke a new HTTP requests dynamically (in function of > the total number of values count). > Ex: Retrieve a JSON file for a page publication in Facebook Page (with > HTTP request), and then retrieve the info about each profil of each one who > commented on that publication. > So is there any way to achieve this? > Another question, Is there any Loop possibility for creating a new > instances of a HTTP processor with different parameter for each (the > profile ID for each one who commented)? > (Ex: graphs.facebook.com/ID1, graphs.facebook.com/ID2 .. > graphs.facebook.com/ID1000 without creating each processor separately), > If this not possible do you recommend me to use Flume for this purpose? > Thanks in advance :)-- > > > > Kacem BOUKRAA > 5thyear student at ESI | Higher National School Of Computer Science > (Information Systems) > Google Student Ambassador in Algeria > Kouba - Alger > > mobile: +213 559 859 858 | email: [email protected] > twitter: @kacem4dz | website: www.kacemb.com > -- [image: Image] *Kacem BOUKRAA**5**th**year student at ESI | Higher National School Of Computer Science** (Information Systems)* *Google Student Ambassador in Algeria* *Kouba - Alger* *mobile: +213 559 859 858 | **email: [email protected] <[email protected]>* *twitter: @kacem4dz | website: www.kacemb.com <http://www.kacemb.com>* <https://www.facebook.com/KacemRostomBoukraa> <https://plus.google.com/u/0/+KacemBOUKRAA> <https://www.linkedin.com/profile/view?id=177866682> <https://twitter.com/kacem4dz>
