DSXchange

What sort of a machine is your SAP installed on - could it have a different byte ordering from your DataStage server machine?

Again, this has been covered many times. In your case you would use stage variables. You would add the ColB values to a stage variable until such time as a new colA value of "name" occurs, then output the string from the transform stage.

Ray's suggestion gives you a structured input where the 2nd field contains the values you want. Now all you need to do is a vertical pivot (asked & answered many times here in the forum) and you have your result.

In the log file you should have an entry starting with "...Schemas.." where you can check the actual runtime schema for the input link which I still suspect contains your reference column name.

This isn't as trivial task as one might assume. When a job is compiled or imported the instance information is no longer displayed in the director, so a quick method would be to schedule an unconditional import (with binaries) when the job is not active. It isn't a pretty solution, but functional. I...

How many nodes co you have in your configuration file? do you have multiple servers in your configuration? Does the job abort immediately upon processing the first row(s) or after running and outputting data? Have you tried to see if the error persists if you use different number of nodes or turn on...

Try setting your environment variable $OSH_PRINT_SCHEMAS to "true" and check both input links to your lookup stage for "Flag1" - my first guess is that you have RCP enabled and are inadvertantly passing in more columns than you think.

You won't get an "null" values from your source sequential file unless you explicitly define default values and null replacements for each column in your source.
Or do you mean "empty" values rather than "null" - there is a big difference.

As this thraed is 3 years old and most likely there are different versions and libraries involved, it would be best to open a new thread on the topic and to include additional information in the post.

I thought you had seen the answer already. You read in your field as a character one, then convert it explicitly in your DataStage job to a timestamp field using the StringToTimestamp() function in a transform stage.

In your output sequential file stage, go to the Stage -> NLS Map tab and enter your file's map.

My mistake in the earlier post, I missed that you have a LATIN1252_0A. The MS-1252 multibyte character set is covered in the ISO-8859 and is a superset thereof, so you can read in UTF from your SQL server and write out to 8859.

If the problem is reproduceable I would submit it to your support provider as it would seem to be a bug. How big is "big"? We routinely save jobs with over 100 stage and haven't seen this problem.

Unfortunately no such flag exists. I'd also like to see the project level message handlers be included in backup settings.

No, not via RCP. RCP allows you to copy columns across stages without explicitly defining them everywhere, but doesn't perform implicit conversions. There are different approaches you can use if you really want to do this, but they require that you have a list of the column name(s) in your parameter...