Re: Empty flow layout diagram after execution on grid

That would be great, yes. Anyone has experience with JS libraries for time-line visualization? On Tue, Nov 18, 2014 at 8:16 PM, Henry Saputra wrote: > Gah, we should probably think of another way or framework to alllow > offline use =( > > - Henry > > On Tue, Nov 18, 2014 at 6:43 AM, Ufuk Celebi

Re: Graph traversing example

Ok thanks Stephen, I'll start from there! On Nov 26, 2014 1:07 PM, "Stephan Ewen" wrote: > This looks like a "transitive closure" or "reachability" style problem, > you can probably model it similar to that. > > Have a look at the examples, there is "ConnectedComponents" and > "TransitiveClosure"

Re: Graph traversing example

This looks like a "transitive closure" or "reachability" style problem, you can probably model it similar to that. Have a look at the examples, there is "ConnectedComponents" and "TransitiveClosure". You can start from there... On Wed, Nov 26, 2014 at 10:32 AM, Flavio Pompermaier wrote: > Hi

Graph traversing example

Hi all, I have problems in figuring out how to solve the following use case. I have a set of Tuple3 like: A, type, t1 A, hasStatus, B B, type, t2 B, atTime1, C B, atTime2, D C, type, t3 C, name, c D, type, t3 D, name, d D, someAttr, d Now I want to be able to find whether exists a path for A->(?

Re: Program crashes trying to read JSON file

There must be something simple mixed up somewhere. The string output from your previous mail suggests that the output is the name of the data reading operator object. Not sure how you achieved that, but that is pretty mixed up. My gut feeling is that you call "toString()" on a DataSet somewhere an

Re: Program crashes trying to read JSON file

Hi Anirvan, The CSVInputFormat works with two delimiters, a record delimiter and a field delimiter. The input data is split at the record delimiter into records. Each record is split along the field delimiters into several fields. The number and type of fields must be constant. There is not logic

Re: Program crashes trying to read JSON file

Ciao Stefano ! Thanks for this early morning information, very helpful. Yes, for outputting the data we are using WriteAsCSV which is stable over different versions of Flink. Our current concern is "reading" a JSON file into a dataset. As you can see, we have a simple 2-level JSON hierarchy