Parsing Error in data flow source.

Question

Hi,

I created a data flow and the source file is json with an array of document.

and its a valid json file as you can see here

Now i want to flatten this file and only want to extract and save ID column of the data as.csv

I tried all the three options of in Json settings

but still i am getting this error

Can anyone tell me how can i get rid of this error ?

Answer 1

Hi @Amar Agnihotri ,

Thankyou for using Microsoft Q&A platform and thanks for posting your query here.

As I understand your issue , the source transformation in dataflow is giving error : "Malformed records are detected" even though the json is a valid json.

I tried to repro your scenario, I tried with a dummy json having similar format like yours and was able to get expected result and didn't experience any error. For seeking better help, I would recommend you to kindly share your complete json content . Here is the video of the dataflow:

---

Hope it helps. Please let me know in case of additional queries, Thanks

Answer 2

@AnnuKumari-MSFT

The issue is resolved now . Actually we need to change the encoding scheme to UTF-8 without BOM in the sink data set of the pipeline which is creating the TempoTeamId.json file . Like here -

I did this and ow dataflow is working fine