DSXchange

How do you plan to take care of the Count in a PX job( With multiple partitions )?

Since you want to do it in one line,

DateFromDaysSince(-1, StringToDate(DateToString(DSLink2.col1, "%yyyy-%mm-%dd")[1,8] : '01', "%yyyy-%mm-%dd"))

As Craig pointed out: 1. Create first job ( on the existing server ) Hash --> Xfm --> Seq (SDKSequences) Have two columns defined in the hash file e.g.: ID VARCHAR(10), VALUE INTEGER(10) Run this job. Transfer the seq file to the 2nd server. 2. Create a 2nd job ( on the new server ) Seq --> Xfm --> ...

You could have a user-defined update statement like:

UPDATE tbl
SET col1 = NVL(:col1, col1),
col2=NVL(:col2, col2)..
WHERE pk_col=:pk_col

In the first job, you could read the dataset first and pass the date to a second job as a job parameter(vDateParam). In the second job, use a transformer and a constraint to only let those records through where incoming date is > vDateParam.

Thanks everybody for responding. I guess the final answer seems to be to somehow find out the missing columns amd update the file with default values for those columns. Once the file has been updated, Design the DS job in such a way that it reads the data section with the complete static metadata. R...

Hi, I have a requirement to load a file containing dynamic metadata. For one, the column names are included in the same file which has the data. Thus, the seq file (info.20060710 ) looks like : FILE_START FILE_NAME=info.20060710 REC_TYPE=NEW_EMPLOYEE FIELDS_START ID NAME DOB CITY #Comment # MARITAL_...

I was able to resolve the issue in a couple of jobs. I was using the SybaseOCPX stage with "Generated SQL query". However, I didn't have the "database owner" with the table name. Thus instead of using "Owner.dbo.TableName" in the "Table Names", I was using &qu...

Thanks Ray. I checked the Cross Table Analysis and found that it does have the functionality of doing cross table analysis across multiple databases.

Hi, Is it possible to perform a CrossTable Analysis across multiple Sybase databases on the same server. I need to write an extract process that extracts data from multiple databases. I am not too familiar with the data at this point and am wondering if it's even possible to perform such an analysis...

I am still unable to run the job successfully. I really need to get this resolved ASAP. Any help is greatly appreciated.
Thanks again.

It is not saving a 0 in the scale column.

I explicitly entered 0 in the scale column and saved it. However, when I reopen the columns editor, the 0 is not there.

I changed the metadata [ default : Decimal(38,10) ] to Decimal 38 [ in the DS job]. The data was written to the table. However, still got the following warnings: Oracle_Enterprise_0: Column PERSON_ID floating point decimal not fully supported. Adjusting scale. Oracle_Enterprise_0: When checking oper...

Thanks for the reply Ray.
Do we have to change the table structure in ORACLE. Isn't there a way in the datastage to overcome this. I am asking this because we have a lot of tables already defined with the datatype NUMBER.

I am new to the parallel jobs and am having trouble running a simple job. I created a job having 2 Oracle Enterprise stages joined by a link. The Source stage is reading from table aa Person_Id NUMBER first_name VARCHAR2(20) last_name VARCHAR2(20) the target table has exactly the same structure. I i...