SAS Data Integration Studio, DataFlux Data Management Studio, SAS/ACCESS, SAS Data Loader for Hadoop and others

propagating lengths from source to target

Reply
Regular Contributor
Posts: 163

propagating lengths from source to target

Hi:

With DIS 4.2 is there a way to propagate column lengths? For example, I have a target with a length that is larger than the source since this was updated. We have about 200 jobs that need for the target to get updated. Without having to go into each job and delete target columns and propagate them from source, is there another way of propagating this without doing the latter manual way?

Contributor
Posts: 38

propagating lengths from source to target

Updating table metadata in the Library should propagate into jobs automatically.

TEST:

1. Just create a simple job with two tables with extract>load transformation then run and save.

2. Close the job

3. Change the length of target table column in database and come back to DIS and go to table > right click > update metadata.

4. open the job. Then go to the table properties and properties of mapping columns in Loader (or any other transformation you like to test) and check to see the column length has changed to new value.

If your test is successfull then no harm applying to production.

Hope this helps.

Regular Contributor
Posts: 163

propagating lengths from source to target

Halaku, thanks for the answer. What if you had hundreds of jobs where the source changed the lengths; do I have to go to each target table then open the job to update the job for each job? 

Contributor
Posts: 38

propagating lengths from source to target

As you can see from the test that updating the metadata in the Lib cascades to the job(s) automatically. the only thing that remains is to redeploy those jobs, if they are scheduled, or you can run manually.

SAS DIS will try to strip the values to fit the target tables. I think you should do a test for the numerical values to be sure.

SAS Employee
Posts: 36

propagating lengths from source to target

In 9.3 there is a new tool in the tools menu called the column propagation tool.  It will support modifying/standarizing one or more metadata columns downstream of a source column for all metadata attributes including formats, length, and other attributes.  Lineage is also available from the tool to help identify where a column is used.  It is still a visual tool, but it does help automate the mapping process when columns change.  We have captured your request for additional automated tools to help with this task and will consider more features in this area in a future release.  Thanks for the feedback. 

Ask a Question
Discussion stats
  • 4 replies
  • 225 views
  • 8 likes
  • 3 in conversation