DSXchange

No need for 2 passes.

1. Read the data, split the stream into two links, A and B
2. Link A does a aggregation on a dummy column to produce one row with the sum.
3. Link B does an inner join to the aggregator output of A

Hello rajreddy. This type of question will usually not get answered on DSXchange as it shows a disrespect to the forum. If you aren't willing to read the documentation why should someone else go to the effort of writing lots of details? In addition, searching on "dataset fileset difference" will res...

I would suggest keeping the problem simple. How are you converting the EBCDIC to ASCII in your jobs? What version of EBCDIC do you have (there are several)?

Remember that an EBCDIC "FF" converts to a different ASCII value.

I recall having had difficulties with implicit date conversions before. Is your parameter of type "date"? Can you make the data column a string and then in a downstream modify or transform change perform an explicit conversion?

Niloy,

Job sequencers don't deal with data and columns, so I am not sure what your first question means. Jobs are run using the "Job Activity" stage, and there is no direct stage to abort other jobs that are running - but you can issue a "dsjob -stop" command from the shell to stop a job.

the derivation "COUNT(CONVERT(CHAR(255),'?',In.Column),CHAR(255))" should always be 0, if you have a nonzero value you will have discovered a serious bug in DataStage string handling.

sridhar.b - few things bother the regulars here at DSXchange more than this type of post! Either post your exact question, or read up on "errors" in the documentation and then post remaining questions.

The MCP option changes all non-displayable characters into a "." (period). You might just use a CONVERT(CHAR(255),"?",In.StringColumn), if that didn't work then you don't have a char(255), try outputting a SEQ(In.ColumnName) to see which character you really have.

If you don't like the restricting of texts for a just a handful of posters, what is your recommendation for making this site pay for itself instead of costing someone (else's) money? Remember that some of these posters are paid over $100 per hour of work and are doing this DSXchange work without any...

In your complex flat file stage set the NLS map to IBM01141

With NLS enabled you would set the NLS character set to the appropriate IBM EBCDIC and not the project default ASCII

if you have dbx you could check the stack from the core file - but you might be better off submitting this to your support provider. Have you done any upgrades or changes recently?

Your data is still EBCDIC and you need to tell the CFF stage that the source is EBCDIC - I would start with the NLS settings. I don't have the stage in front of me, so you will have to look through the options yourself to find where you declare this.

I would add a filter stage to the job with the command


(I'm sure there is a better way to do it, but I'm not certain of the syntax)