DSXchange

Not sure....but carefully check your incoming data. And run with a tiny number of input records that you are intimately familiar with. I didn't have time to absorb all of your data values, but it "looks" like you want to have one set of repeating groups for the single combination of npi/ti...

hard to tell, but it "looks" like p2 is your "inner" repeating group, and p1 is the "parent". ....and the C and D cols are desired to be at the "parent" level. One would guess that the xpath for this looks something like: .../p1/C/text() .../p1/D/text() . . . ...

Hmm. Hard to say, but it's "possible" that this change in your transformer impacted the uniqueness of your element....and then altered the aggregation. You are probably going to need to break it down and determine first if the Derivation really caused it, and then what specifically about t...

Excellent work! Glad that you are at the point where you are looking for more ways to gain access to the glossary for your reporting users. Some thoughts on this.... a) The Cognos feature indeed uses javascript, exploiting a special version of the REST API that comes with Business Glossary Anywhere....

I know. The xmlOutput Stage does that for you. Multiple incoming rows are generated as multiple repeating nodes. But you must have multiple "rows" in order to trigger that action. Consider this example... three rows aaa 1 aaa 2 aaa 3 and you want this xml <myxml> <mainInfo>aaa</mainInfo> <...

If the input you describe is really coming in like that (with column names on the link of Contact_Type1 and Contact_Type2, etc.), then you need to "pivot" the data to get multiple rows. Multiple rows = multiple repeating nodes. Take a look around the forum and at DS for various ways to do ...

Confirmed. The doc is correct.

Ernie

I seem to recall finding this also, as a difference between the Stage and the Connector. I typically find myself dealing with xml payloads, so it doesn't matter, because it's always just one column. Concatenate them beforehand, or I think you could use a Column Import and get them all into one also....

There are a few kinks that continually get worked out, but here are some other points: a) They support Job Parameters FAR better than stages throughout all of their properties. Many properties in old stages weren't exposed in such a way that you could provide parameter values. b) They provide strong...

Perhaps it is a new type of candy.

The exact answer, if asked such a question is.....to ask the requestor further questions..... "That is a vague question with many possible answers..."..."What would you like to know?" Would you like to understand how and why DataStage can process huge amounts of data in parallel?...

This one looks like it has a variety of challenges. a) the need to get the multiple columns into multiple rows....the B's as a set and the C's as a set.....independent paths doing pivots will be needed for that. b) the need to treat them as independent repeating nodes of a multi-path xml document. L...

perfect!

probably best to just "make it unique". Put a new column at the detailed level in an upstream transformer. Give it a unique value, such as @INROWNUM. Put it at the lowest level of your xml document, giving it similar xpath (same as Col1 with its own element name, like "dummy"). S...

One can only guess from the details of this thread that xmla is nothing more than an executable that "receives" its arguments, commands and properties from an xml document. Perhaps xmla is "the" command from (said vendor) that looks for an xml document like the ones in this threa...