DSXchange

So you're saying that your target is static but your input (from a sequential file) is variable? Edit: What Prasanna said - provided that they can always provided the columns in the same position each time. ie if there is a column X in the full schema in position 3, then column X will always arrive ...

Quick question first - what is your source?

A schema file gives DS a list of columns and their definitions to a stage. By only passing 2 columns when its expecting 4 - then you are sending incomplete data - which will cause an error.

If you want to get around it, then you would have to pass different schema files.

But if you used Loop functionality, combined with some RCP, and schema files you could write the one job to do both circumstances (or any number of input columns).

Short answer: Yes - that is exactly what the Idoc Load stage is for. Long Answer: IDocs do not directly load SAP tables - direct loading of SAP tables is a big no-no. Also, I would be doubtful that the Idoc would be any quicker than the LSMW. They both are probably doing very similar processes to lo...

Which release of v8 do you have? Do you have access to the loop functionality? If yes, just read each row of the file in as a single value and then use the field function to loop through the row outputting to the required fields. eg loopvariable for description would be: if @ITERATION = 1 then CELL ...

Are you able to check for table/record locks on your target tables? I had this problem a couple of times and each time it was that the table and/or record lock.

Use a combination of the a sequential file stage and an external target stage. For the input use a sequential file stage using a file pattern. Include a column for the file name itself. Extract your required values and perform whatever transforms you need to perform. Prior to output concatenate all ...

Couldn't you just set your lookup to be caseless? (A=a in the lookup key type)

Whoops sorry totally didn't read the original post correctly. In my defence it was first thing in the morning.

When reading this file in, you are not doing anything special are you? RCP? Schemas?

A little more information on the scenario might help. Are you actively choosing to read in only 4 rows from the file? If so how are you acheiving this? If not, ie you are expecting 10 rows of data but are only getting 4, then I would expect some sort of messages in the log (look at the informational...

What version of DS are your running. 8.5? If so you can solve this with the loop functionality in the transform.

PhilHibbs wrote:Can you get the sparse lookup query to automatically supply this count?

You would need to write a custom SQL query and pass the approprate fields. Still a very basic query using either a group by or over(partition by ...) depending on the requirements.

I think that in either case - a sparse lookup or an in-memory lookup - you have to do the duplicate checking yourself. I don't think there is an automatic option to "reject if duplicates", you just get a warning in the log about duplicate rows being ignored. Agreed So, to detect duplicate...

Hi Phil,

As far as I have seen Oracle doesn't act any different.
Regarding returning multiple rows, I believe that a sparse lookup just returns as many rows for which there is a match automatically - you don't get the choice?