DSXchange

The PX definition is more rigidly enforced. Server jobs don't attempt to data type verify or check widths, etc. So, PX jobs will have issues with non-conforming data. Start small, take the first row from the sequential file and create a new one-row file. Try viewing that file and see if it works. If...

Change your output from the job to sequential text files for debugging purposes. It's easy to see the results and what is happening, rather than deal with the job breaking of the time when you have a DB issue. Spool it to files and check the results there. That's just a handy development tip. Then, ...

Ken, I kind of knew this was coming and indeed it was a piece of cake. I was just hoping there may be a function lurking around somewhere that i am unaware of. There will be if you share yours! So, paste the code and enlighten us all!!! Make sure you give yourself credit for creating it!!

We requested an enhancement that would prompt when viewing data from an OCI stage for a set of values corresponding to the where clause. That would be nice from a troubleshooting point of view. Hey, make sure you ask for a "CANCEL" button for when the query takes too long. It's great fun when you v...

Are you sure about that, Ken? The server part, that is? Hey man, didn't you see the slideshow on the release? Apparently, because 5.x is still sooo much Universe that release 7.x is almost a pure DSEngine. They even changed the port number being used, as well as the shared memory segment identifier...

Yes. Only release 7 has the ability to do so. You can co-exist them, but you run into an issue with the client piece. Check the cd regarding a utility to "switch" you between clients.

This is an uncertified DS BASIC subroutine that my team is working on, but you might be able to get what you need from it. It is meant to be called from a DS job either before/after stage/job routine. It uses a delimited argument that contains comma separated list of packed arguments. FunctionName =...

Sorry for the late reply, my first read indicated this was a well traveled topic so I let it ride, hoping you'd search the forum and get the answer yourself. As I understand it, you don't have a single fixed output file, you have a dynamic number of output files based on a column of data, and that c...

gdean wrote:I would really appreciate any tips for getting started with the process.

Get Tom inhouse and have him give you a bootcamp....He's getting too comfortable in Key West.

Where did you get the definitions? Do they match the file? PX is very picky about definitions. Are you absolutely positive that the file definitions are correct, you have the right delimiter, quote character, number of columns defined, data types correct, widths and scaling, etc.

StringDecode does not work like you desire, as you noted it does not have the ability to do anything "else". You're going to have to write your own routine. You could look at modifying StringDecode to have the necessary logic. aravindkStringDecode(%LookupKey%,'A=1|B=2|C=3|DEFAULT=50') Copy t...

If the first job has only the responsibility of getting that max value and writing it somewhere, why not encapsulate that SQL logic into a shell script call or a DS routine wrapped to that shell script call? Then, you could fetch that value either in job control and pass it as a value, or, during a ...

In a source OCI stage, viewing the data actually runs the SQL statement. In a reference OCI/ODBC stage, viewing the data runs an altered SQL statement. Think about it a minute, the query for a reference is used to being handed specific values to use in the select. For example, your reference link is...

Your DS BASIC syntax is wrong. To compare two variables, you do X = Y. If you have three sets that have to compare, you do A=X AND B=Y AND C=Z, or the appropriate expression. However, your insert constraint should be: ISNULL(referencelink.primarykey) and your update constraint should be NOT&...

Another option is to read your source data and pass it into a transformer stage. Put an output link of the data into a hash file using your defined primary key. Setup a reference link to that same hash file. In your constraint put something like ISNULL(referencelink.primary) so that only rows not wr...