DSXchange

Hi, I am following Kim's advice about having a job that reads parameters from a table (containing param names and values) and dumps them into a hash file. Then I used your routine (in the Routine Activity) to run a job. My question is, if I have multiple jobs using the same parameters, would it be o...

Great!! Works...just as I needed. Thanks Ray and Arnd.

Hi, I am creating a job so that it updates the SDKSequences file with the value MAX(my_table_key)+1 for the required sequence which is the argument to KeyMgtGetNextValue(). The SDKSequences file in my project folder only contains one field AT.ID, which contains the names of all the sequences in the ...

Since DS performs a TO_DATE() operation on the value for the date that is supplied, I performed Left(MyDate,19) to eliminate the .SSS from the time picture. It works.

What you posted is easy as it's exactly what it needs to look like for an Oracle DATE field declared as a Timestamp in DataStage: YYYY-MM-DD HH24:MI:SS So for that just send it on through - no conversion needed. :wink: For something coming in like '2005/04/15', it's just a matter of rearranging the...

vmcburney wrote: You can also load your variables in from ini or text files using DataStage routines.

Hi vmc,
Can you give me any pointers as to how the above can be done? Would be nice if there were links to some such routines posted in the past.

Thanks.
gateleys

Hi, It seems that setting of environment variable $PROJDEF is not available in DS ver 7.0. I am using that version (Server ed) on a Windows platform. Is there any work around to achieve the same? If yes, can anyone please explain in detail how it could be done?

Hi, I need to merge the content of 3 seq files into one hashed file. I used a Collector stage to stream the data into the hashed file. The job runs without any errors and the input link to the hashed file also displays the correct # of rows. But when I try to view the data (in hashed file), it says ...

Hey Ray,
Thanks again for your prompt and precise response on my 'hashed' file queries.

1. What is the maximum size of a hash file that can still give a good performance? Lets assume we have 2 to 3 fields, each of length < 10 and integer/varchar type. 2. To put from the other perspective, when would a relational stage be on par with hash file for reference? Assume the relational refere...

Ray and Shawn,
Thanks for the response. Will keep your advice in mind while designing jobs.

You can't perform any kind of lookup without a reference key. I understand that Ray. But my question was that I don't need to have any reference in the Transformer since all I need to do is just input the same value in a particular field to all the rows. Hence, my requirement de-necessitates the re...

Option 3 is what I would do since it decouples the source system from the destination. Apart from the fact that Option 3 provides flexibility in terms of recovering from the respective job failure, what merit(s) does decoupling of the source and target have? Also, how does each of the option fare i...

I have 3 options to design a simple job. For each of the circumstances, say, when the size of data is small, medium, large, very large; can you please suggest which of the designs would be good and why? option1 (direct loading to target) -------- relational source(s) ==> processing ==> relational ta...

That's what I did too, just used an OCI stage to feed the row via transformer. OCI || V ODBC ==> Transformer ==> Seq File ^ || Hash File It easily solves my problem, where I select the date where the key = 1 and set the key value of the OCI to '1' in the transformer. But the purpose of my question w...