linkedin / data-integration-library Goto Github PK

Many times content type for CSV type may be text/plain instead of text/csv. Because currently CSV extractor does not expect text/plain, the job will fail.

Currently DIL supports many structured format like CSV, Json, Avro and also many compression formats. Unstructured text format is supported only through FileDumpExtractor, which dumps output to HDFS. With FileDumpExtractor, output cannot be passed to any converter. Text Extractor should be supported, which can extract output in any format and pass it to some converter for further ETL rather than directly pushing this to HDFS. This is useful in cases where we want to get some URL output and then apply some custom parsing to get the required output.