What is run time column propagation ?

rhaddur · Post by **rhaddur** » Mon Mar 19, 2007 6:38 am

chulett · Post by **chulett** » Mon Mar 19, 2007 7:26 am

An interview question. And a trick one if used in the context of a Server job.

DSguru2B · Post by **DSguru2B** » Mon Mar 19, 2007 8:32 am

I'd be interested in 'its' implemention in server jobs

mehimadri · Post by **mehimadri** » Mon Mar 19, 2007 12:32 pm

U can define part of ur schema and specify that if ur job encounters extra columns that are are not defined in the meta data, it will adopt these columns and propagate them thru the rest of the job.

I have seen them in sequential, file sets and almost every other data source stage in parallel jobs, but not in server ones.

ray.wurlod · Post by **ray.wurlod** » Mon Mar 19, 2007 6:54 pm

DSguru2B wrote:I'd be interested in 'its' implemention in server jobs

That's an easy one. 's impossible.

DSguru2B · Post by **DSguru2B** » Mon Mar 19, 2007 7:56 pm

Me know that. Was just playing along the 'tricky part'

pravin1581 · Post by **pravin1581** » Fri Mar 23, 2007 7:10 am

DSguru2B wrote:Me know that. Was just playing along the 'tricky part'

Do we need to define the metadata in the colums tab of the file stage?

DSguru2B · Post by **DSguru2B** » Fri Mar 23, 2007 7:32 am

Always, for server jobs. For px jobs, if any transformation is required with rcp turned on, then yes, else no.