Question:

Read xlsx from AWS S bucket, then write it back to txt file in another local path. There would be three types of excel retrieved, and one of the excel will have a json in the column. So i will have to read the json data as well the perticular file and then create the txt file according to the data.

what would be the best approach, 

Can we have dynamic column headers, if dynamic headers is not possible that would be fine.

Complete requirement:

Will be doing some nlp processing using Data Flows. By xlsx file stored in S3 bucket, which will have data which needs to be formated / filtered. 

Approach : 

Read the xlsx file using java and then create a txt file. 

Then do the processing of DataFlow using txt file.

Please provide me with solution, and if there is OOTB way to achieve, if java code extc.