DSXchange

Scenario dates in mm/dd/yyyy p_startdate p_end_date q_start_date q_end_date 01/06/2011 06/06/2011 01/06/2011 07/06/2011 --- 1 row 04/07/2011 10/07/2011 04/07/2011 11/07/2011 --- 2 row overlapping dates 1 row ---- 01/06/2011 02/06/2011 03/06/2011 04/06/2011 05/06/2011 06/06/2011 2 row ---- 04/07/2011...

IBM Information Server 8.1 (with Fix Pack 1)

on Redhat Linux

with Oracle as the XMETA

official support provider contacted them, waiting for the response.

Yes i agree thanks ernie a lot. have a gr8 weekend.

ahh right 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...

tried both ernie,

may be you want to have a look at the xpath expressions, posted here.

duplicate post

Ernie,

i was able to get this before, but with the new change, i am loosing this.

i made the last column as transformation to produce as
<performance-rate xsi:nil="true" />

duplicate post

And also

made the last element as XML

<performance-rate xsi:nil="true" />

Hi Folks, - <measure-group ID="X"> - <p1> <C>Y</C> <D /> <E /> - <p2> <p2-number>ABC</p2-number> <eligible-instances>16</eligible-instances> <meets-performance-instances>28</meets-performance-instances> <performance-rate xsi:nil="true" /> </p2> </p1> - <p1> <C>Y</C> <D>2011-04-01...

after making this change in transformer, IF ISNULL(read_ora.middle_name) OR read_ora.middle_name <= 0 THEN '<': 'middle_name':' xsi:nil="true"':'/>' ELSE read_ora.middle_name i had some elements which were not repeating, now it seems like they are repating. i have checked one column from w...

this should work, i did the same thing it works.

those are the default values i had put for null values, but it seems like the question has been answered by this statement. All parsers whether it is custom made / product based should follow W3C xml 1.0 standard. According to the standard the parser should read <empty/> and <empty> </empty> in the ...

Thanks chullet and ernie, got the secong question,

still working on the first one.