Hey @SAMSAL Amazing response - thank you! I'll go over a couple of things you have mentioned, firstly the overwriting... I was wrong about that one. When I converted it to CSV instead using the 2 sheets, it actually appends the information from sheet 2 to the end of sheet 1 (leaving 2 rows as a gap first) so in your example above you would actually end up with something like this in CSV (when reimported into XL): I'm not sure if I can work with that, as I don't see any way in any of the readers to be able to "split" that (or only reading from row 2 to X for example) whilst keeping the flow etc and just using some of the values The other issue is with the Avro you have set up, because we don't know what the column headers will be (field names) I'm unable to define a schema in Avro for this as they can add and remove these columns at will - the only thing I will be able to rely on is that both the sheets will have identical columns. When I convert direct to JSON, I get something more like this: [ {
"column_0" : "ID",
"column_1" : "NAME",
"column_2" : "ADDRESS"
}, {
"column_0" : "1",
"column_1" : "Sam",
"column_2" : "TX"
}, {
"column_0" : "2",
"column_1" : "Ali",
"column_2" : "FL"
}] Again, not sure this is something I can use in this format but I'm just not familiar enough with the NiFi elements at this stage I guess I'll try the enrichment fork method though and see if there is some mileage in that and then I can feed back
... View more
