FetchParquet 2.3.0.4.10.0.0-147

Bundle
org.apache.nifi | nifi-parquet-nar
Description
Reads from a given Parquet file and writes records to the content of the flow file using the selected record writer. The original Parquet file will remain unchanged, and the content of the flow file will be replaced with records of the selected type. This processor can be used with ListHDFS or ListFile to obtain a listing of files to fetch.
Tags
HDFS, fetch, get, hadoop, ingest, parquet, record, source
Input Requirement
REQUIRED
Supports Sensitive Dynamic Properties
false
Properties
Restrictions
Required Permission Explanation
read distributed filesystem Provides operator the ability to retrieve any file that NiFi has access to in HDFS or the local filesystem.
Relationships
Name Description
failure FlowFiles will be routed to this relationship if the content of the file cannot be retrieved and trying again will likely not be helpful. This would occur, for instance, if the file is not found or if there is a permissions issue
retry FlowFiles will be routed to this relationship if the content of the file cannot be retrieved, but might be able to be in the future if tried again. This generally indicates that the Fetch should be tried again.
success FlowFiles will be routed to this relationship once they have been updated with the content of the file
Reads Attributes
Name Description
record.offset Gets the index of first record in the input.
record.count Gets the number of records in the input.
Writes Attributes
Name Description
fetch.failure.reason When a FlowFile is routed to 'failure', this attribute is added indicating why the file could not be fetched from the given filesystem.
record.count The number of records in the resulting flow file
hadoop.file.url The hadoop url for the file is stored in this attribute.
See Also