DataStage stages are designed so that they execute one row at a time, so your requested approach poses some difficulties so that it doesn't get executed once per row.
The best was has alread been suggested, to get the count in a job sequence and then pass that value as a parameter to your job.
In your case you might just read your file in a sequential stage, then perform an aggregation to get a row count and then pass that value on to the next stage.
You would send a single non-null field down another link to the Aggregator and tell it to count that. Doesn't even have to be a "real" field, you could hard-code a value.
-craig
"You can never have too many knives" -- Logan Nine Fingers
i need this count in a same link, lets mf src file have two fields A and B, then tgt contain three fileds A and B And C, aditional field contains Count of records passing on that link.
'Count of records passing on that link'? That would always be 1 unless you are doing aggregation based on the two fields... are you?
If so, then an inline Aggregator can do the grouping and counting for you. If you need to do something else, then please be more specific in your problem description.
-craig
"You can never have too many knives" -- Logan Nine Fingers