DSXchange

Is there any source in your job, or its just starts from TX ?

Can somebody please help with Syntax of CREATE.INDEX statement for HF with 3 key columns A,B,C.

Is it CREATE.INDEX HF A,B,C ?

Thank you

Naveen,

Every time when you try to populate the detail record, you need to check that field of header record and then populate detail record...I think you need to have a If condition.

Hey Poornima,

What is your data type ? DATE or TIMESTAMP ?

The other option you can try is converting the date and time in Internal format and then try to insert into DB2 table, via API stage.

Let us know...if it works.

Thanks.

Right on...Thanks Ray.

Man good catch Ray, How did I miss that...

please make sure the time part is in correct format...and the quickest solution I can suggest is change the datatype in API stage to Char(19).

Can you please give example of time field.

HTH

I am trying to access HF using UV Stage, but getting the following error. The metadata of UV is same as HF..just not sure what the problem is. HFTesting..UniVerse_11.DSLink7: DSD.BCIOpenR call to SQLExecDirect failed. Statement was:SELECT ID, DATE1, DATE2, DATA FROM HF_ACCT2 SQLSTATE=S1000, DBMS.COD...

Why can't you keep it like following:

#$A#\#$B#\#$C#....whats wrong in this? This takes care of your problem.

Hey Vinay,

Convert works..i have used it in past, just make sure you are giving arguments in right order..

Convert ('doublequotes','',string)

Hi Vinay,

If you are trying to remove quotes from your data then try this

Convert('"',"",String)

it will replace all occurences of " with empty string('')

HTH

Thanks Naveen, I got your point...but then won't it make it hard to debug or even maintain ??

Try Convert...

Is big parallel job, with number of stages efficient compare to a job divided in 2 or 3 different jobs ?

I always feel simpler the better and easy to debug and maintain. Is it true for parallel jobs also? or its the otherway..

Remember the hash file will have the last record of multiple key records. As Craig suggested you need to sort your data in such a fashion that the record you need is the last record for that key combination...