DSXchange

vskr72 wrote:All the values are set to 0.000.

That shouldn't have anything to do with your validation check, all that does is either pass along the original value if it is deemed valid or replace it with a null. Look elsewhere for your "all are set to zero" issue.

What results do you get?

Could you explain for us the issue you are seeing and what things you've tried? Thanks.

Me, I'd start with this:

chandra.shekhar@tcs.com wrote:Please contact your support provider if you encounter problems.

That would be the proper choice, what "did not work" about it?

Nope... at least not that I'm aware of.

Basically, yes. However you have an "OCCURS DEPENDING ON" rather than a simple "OCCURS" clause meaning while they can repeat up to 32600 times the values in the group will repeat the number of times stored in COL7. Now, as to the CFF stage, I don't recall if it supports "DEP...

No, sorry but that's not correct. You can only read it as a sequential file if you've converted each worksheet to one first, be it csv or tab delimited or whatever you like. Before that they are more akin to a database with each individual worksheet acting like a table. To read one natively you'd ne...

If you've read the wiki, you should have found that they relate back to two operations performed by the original manual typewriter - a carriage return and a line feed. The former moved the print position back to the beginning of the 'line' and the latter moved the print line down one. We tend to cal...

But not marked as such... there. Now it's resolved.

Thing is there's no need for any "metadata" here or for much in the way of reading. For this it doesn't really matter that it is EBCDIC since all the requirement is is to bring in 250 characters and write out 240.

NOTFOUND is a "Link Variable" and is available in that category in the Expression Editor on a reference link - a lookup. I was suggesting you use it as the constraint for your output as it evaluates to true automatically when the lookup fails or 'misses' and false when it succeeds. More de...

You could issue that same SQL in the source stage so that you only get the records that you want. Alternately, stream in all records from T1 and build a hashed file reference lookup from T2. Your output constraint would be "LookupLink.NOTFOUND"... adjusted to match reality, of course.