I have a job which i have to use for multiple interfaces. But the metadata isn't similar for each interface or rather every interface has unique metadata for itself. Hence rather than constructing the same job multiple times for every interface is there any way we can give the required metadata as input in the runtime so that we can avoid constructing multiple jobs and reduce time consumption in construction and use the same job for every interface.
run time column propogation................... RPC or use schema files to input metadata definition to the jobs. it worked for me
Is it that for each source file metadata structure, should we have a separate schema file?If so, how could we determine in runtime that a specific schema file has to be used?
RCP is NOT applicable to all PX stages. Specifically, it cannot be used with the Teradata API stage. I'd like to make this generalization; wherever a column is needed to generate the SQL at compile time, RCP cannot be used. The teradata API stage is such a place.