I creates a pipeline that takes data from csv file, handles it inside (I convert it into json and use SplitJson) and puts it into database (i.e. i input it string by string). But csv file can contain a lot of strings, and it slows down the next steps. So I desided to merge before going to the next steps. The problem is that I don't know the exact number of strings, it can be 10 or it can be 1000. And if it is 1000 (in my case it is very slow even with 470 strings), MergeContent must wait till the end of input for some time. I think about using some variable or parameter or something else that i can use as Minimum Number of Entries and change dynamicly via a processor (i.e. update it with value of fragment.count after SplitJson), but i didn't find any solution for my case. Is it possible?
Use defragment algorithm then fragment.count attribute will be used.
More in documentation: https://nifi.apache.org/docs/nifi-docs/components/org.apache.nifi/nifi-standard-nar/1.17.0/org.apache.nifi.processors.standard.MergeContent/index.html