DSXchange

However, I have created a pdf of the design which basically shows the logic inside the transformer of the job. In this design, DSLink4 is your source link, DSLink5 is your Rules Table Link (remember no hashed file used, looking up the database with Source_ID as the key) and DSLink6 is your target l...

Amey Vaidya, That is not what kaps is looking for. And the previous posts make this clear . The suggestion made by gateleys is also the same. Kaps doesn't want to do multiple look-ups. Sorry Naveen; But I disagree! A simple way would be to have 3 hashed files, each with two columns , both of which ...

Hi Kaps, I believe this is what you are looking for: Issue: Do multiple simultaneous lookups on a single translation table. Multiple source columns need to be mapped to the same reference column and the translated(standard) value needs to be passed to target. The job design can be: DB2 Src for rule ...

HI Kiritkumar,
IF the number of comma-seperated names is known and is fixed, a transformer can seperate the names into different columns using the field() function.

IHTH

Hi Sengs, what you are looking for are called BCI functions. This <a href="http://66.34.129.47/viewtopic.php?t=97848&highlight=bci+functions"> link</a> leads to an example post. You can search for others. From what I've read in DSXchange, its usually too complex to use these functions....

Try this: Keep the Metadata of all output columns from both transformers the same: OutRecI->OutRec OutRecD->OutRec I assume that RCP is on. (RCP Off) Trans1----------| | Funnel-------> RemDuplicates----->Dataset (RCP Off)| Trans2----------| or if duplicates are to be removed from each link seperatel...

You may need to double-click the results section in the test grid to see the output of the print statements though.

From what I remember from the Manuals: RCP is a feature of the base orchestrate framework. It cannot be switched "off". What actually happens when RCP is switched off is an implicit Modify operator is added after every stage that changes column definitions. This may mean potentially doubli...

Jim, Viewing of data through Datastage is not possible through Director (Atleast I havent figured out a way of doing it). As far as securing the file locations and database connections, they can be stored as encrypted user-defined environment variables. So if only director is installed on the operat...

I interested in knowing if there is another way around this entire issue, when the job is set up as I outlined in my previous post. Hi Keith, Have no access to DS at the moment but this is what we used to do: Hash Partition before the sort stage Run the sort stage in Parallel mode Do not repartitio...

If the requirement is that the last record of the stream be identified for special processing, one way is to execute a "count()" on the table or "wc -l" on the file and save the result with a dummy lookup key (value) in a hashed file. During processing, read this value in through...

Hi Mark,

I'd recommend a filter stage instead of the transformer.. Unless the transformer already exists as a necessary part of your job design. The filter should give you better performance.

Hi Nitin, If RCp is disabled at the project level... it usually is disabled at the job level.. Anyways the RCP off check box is usually found at the output of any stage at the left bottom corner of the column definitions page. For transformers and lookups, the RCP is in the output(/?general?) page o...

Thanks Raj, but as I mentioned earlier that we are pretty sure that all the mappings are correct and most of the times the jobs run without any warning. As far as I know, this warning should be consistent(atleast for the same input data). Anyways the below has always workd for me: If RCP is on for ...

Unlike a dataset, in a sequential file , only the columns defined as being written to the file are written. The RCP'ed columns are silently dropped.