cmgrote / ibm-ia-rest Goto Github PK

Currently hard-coded to assume that files are always on the local engine host; could also be elsewhere (i.e. Hadoop). Need to re-work code accordingly...

data_file has a store_type attribute, which will have the following values:

• HDFS for files where the data_connection.data_connectors.name = HDFSConnector (and in which case the data_connection.connection_string = data_file.host.name)
• Local for files on the local engine host, where the data_connection.data_connectors.name = LocalFileConnector (and in which case the data_connection.data_connectors.host.name = data_file.host.name)

For example, the batchAnalysis parameter and its associated text input file:

https://www.ibm.com/support/knowledgecenter/en/SSZJPZ_11.5.0/com.ibm.swg.im.iis.ia.restapi.doc/topics/r_restapi_manage_analysistask_commands.html

Should significantly simplify the analysis process (and make more robust when running against a huge number of datasets?)