DSXchange

Make sure your primary key values are trimmed of trailing/leading spaces if this is applicable.

It seems like the variable is one of size. If splitting the file into halves works, then I'd guess your problem might be database side. If you say you set the commit count to 1, then rollback or temp segment would be ruled out. So then it might be how your job is designed interfering with the loadin...

That brings us to the question, how can we pass a value in a transformer, for example, value contained in a stage variable to a newly created job parameter in the middle of a job and use that in the sequential file name box? Or pass the value in a stage variable to another job where it can be used?...

Your premise will NOT work. A file is a fixed resource at startup time of the job. If you have multiple due date values, would you expect multiple outfiles, each containing the relevant data and named accordingly? If all of your data is under the same due date, see the current posts about using the ...

You don't need the stage variables. Just pass an 'X' as the value for a column 'GROUPBYME' in the output link to the aggregator. Setup 3 other columns to track the numbers you want. In the aggregator stage set column GROUPBYME to have group-by checked and listed as the derivation. For the other thre...

Why don't you want to use the aggregator? To simply stream a few columns to it and group by a static column value, like 'X', will accumulate the totals for you and when the stage closes output that one row of totals. This is the simplest and most elegant solution. Stage variables don't work for what...

NO. We don't know what 'source' is: database or file. If database, what kind of query? How many columns of data --> 1000 columns vs 10 columns = big difference in runtime. We don't know what 'sort' does: is it sorting on one field, two, three, complex sort logic? Any de-duplication? We don't know wh...

Modular programming is usually a good idea. You can independently execute the script/sequencer job rather than incur the runtime of a transformation job with after/job commands.

Jeepers Craig, you won that one.

In order to use the job control APIs you must put this statement at the top of your function/routine (courtesy of Ray Wurlod):



The include statement adds the API definitions to your function.

The change does not require recompilation. When a job starts this setting is read and used.

No idea what the problem was.

Yes, I understand. Write the job and output the errant row to a sequential text file. Set the abort limit to 1 to stop on the first exception row. Or, process all rows and then look at the number of exceptions and act accordingly.

Use a series of stage variables and check every source column. Then, if any of the stage variables found a NULL value return the code associated with that column. This is a programming exercise for you and stage variables and a constraint is what you need.

If you're sure that the metadata is the problem, you can blow away the whole contents by simply doing a CLEAR.FILE DS_METADATA . Hopefully, you have all your metadata in source code control, or you can re-import it. Now, the folder structures are all logical, you can see this by doing a LIST-ITEM DS...