AutoLoader - process multiple files

avenu
New Contributor

I need to process files of different schema coming to different folders in ADLS using Autoloader. Do I need to start a separate read stream for each file type / folder or can this be handled using a single stream ?

When I tried using a single stream, multiple files gets clubbed in the same read and its quite difficult to process the files as the schema is different.

Wassim
Databricks Partner

As you are talking about different schemas ,perhaps schemaevolutionmode, infercolumntypes, or schemahints may help?? Check out this- 32min onward - https://youtu.be/8a38Fv9cpd8

Hope it helps, do let know how you solve it if you can.​