DSXchange

No.

hemanthakumar wrote:i applied the logic, but its giving wrong result.

You'd have to show us how you "applied" it and what results you saw as it should have worked just fine. And, like most things, there are multiple solutions to most problems - dcount and field being one such example.

True, FMT would be another option. I don't recall what '%' means off the top of my head, I would usually specify a zero as the pad character:

In your Oracle select, use to_char() with the appropriate mask to select it in that format. Inside the job, lpad() would do the trick.

FTP would be one answer. Depending on what exactly you are doing, you may be able to use the FTP stage for that, and there's always command line ftp you could script for landed data.

Index() would find the position of the "-" and then you conditionally substring off everthing after it. Or not.

Ah... if you just mean you want to store them in the XMETA repository rather than the 'old Universe RT_xxx tables, then yes that's a setting. Search here for RTLooging or ORLogging for the details.

In other words you need to build a process to extract the log information you need and then load it to your target of choice. If you search here for ETLStats you'll find a free framework to do just that written by our friend Kim Duke, you can use or extend that.

For the latter, search the forums here for SetUserStatus.

$APT_STRING_PADCHAR controls that and the default is a hex zero. Change it to a hex 20 to pad with spaces.

The same query that the dssearch command runs?

How are you sorting? From the other posts you would have seen the need to declare this new sorted order in the Aggregator stage, did you?

Ah... been awhile, forgot the 'SH' rather than 'UNIX' part.

Got nothing to to with metadata, see Andy's post here.

Nope... can you post the complete, unedited error, please?