DSXchange

Welcome aboard.

Are you using "OR logging"? This is known to cause severe performance degradation. Prefer local (non-OR) logging.

One of the attractions of REST for the more laissez-faire is that it does not require WSDL or any other form of metadata definition.

This was not properly detected in 8.0. It is now correctly detected. It shouldn't be a problem, but why would you specify nulls first when the field is not nullable?

All stage types in server jobs are "plug-in" - this is the architecture of server jobs. However, each should have its own icon. If not, it suggests that your client is running out of memory. Try closing some other windows, and removing any animated screen savers.

What were your sort keys? If you sort only by col1, you will get the answers you got. You probably need to sort also by col2 and maybe even col3. Make sure also that your data are correctly partitioned; that is, using a key-based partitioning algorithm based on col1.

Any string, but one that should be meaningful. It becomes the name of the "sequence" that the routine uses to keep track of the next available batchID. How about "batchID" ?

Only to call the routine for every column.

There are technically illegal hacks via which you can find the name of each column in the record schema, and then obtain the value of each, but these are undocumented and - under the reverse engineering prohibition in your licence - illegal.

Invoke it through a Routine activity and use the return value (activity variable $ReturnValue) to supply the value to the job parameter in a Job activity.

There doesn't appear to be, either in the stage or in the documentation.

chulett wrote:I was wondering if a target only option is what was implied by the "contains the SQL statements to be run for each input row."

I believe so, confirmed by my playing with the stage concerned (version 8.7).

Before-SQL and After-SQL can certainly be read from a file using Teradata Connector (use Before SQL File or After SQL File properties). There seems to be no way to use a file to contain a SELECT statement. Job parameter(s) may, of course, be used, so you may have a workaround there. If the Teradata ...

Trim() function with a single argument (the string on which to operate)removes leading, trailing and redundant whitespace characters.

No.

Arguments to be passed to routine must be named input columns.

Of course, RCP will take care of any column that you don't want to process, which in your case is none of them.

On AIX at least the USER environment variable uses details from /etc/environment and also (maybe) your own profile under your home directory. The content of /etc/environment is the same for all users. If USER is not updated in ~/.profile, then it will be the same for all users.

1. Use JobHandle rather than handle in the call to DSGetLinkInfo() function. 2a. Delete the line that attaches a job. You don't need to do this if you're interested in the current job (the handle for which is DSJ.ME). 2b. If you are interested in getting the link count from a job other than the one ...