You can use the Catalog Input step to read CSV text file types or Parquet data formats of a Pentaho Data Catalog resource stored in a Hadoop or S3 ecosystem and output the data payload in the form of rows to be used by a transformation.
CSV files include formats generated by spreadsheets and fixed-width flat files. Parquet data formats are decoded and the fields are extracted using the schema defined in the Parquet source files.
Catalog Input can be used with the Catalog Output transformation step to gather data from various Data Catalog resources and move that data into Hadoop or S3 storage.
You must have role permissions set in Data Catalog to read the data resources.
For more information about accessing Pentaho Data Catalog in PDI, see PDI and Data Catalog.