Repetition Elements in XML

ankursaxena.2003 · Post by **ankursaxena.2003** » Fri Dec 14, 2012 9:08 am

Hi,
Here I have a huge XML with so many repetition elements like,

<admin>
<quoteback name="caseId">SHAWN AABROWN</quoteback>
<quoteback name="caseId">BROWN</quoteback>
<date_request_completed>08/15/2012</date_request_completed>
<date_request_completed>15/2012</date_request_completed>
</admin>
<claim number="5456767" sequence_number="1" unit_number="1">
<claim number="465476578" sequence_number="2" unit_number="1">

Is there a way apart from separating the XML chunks and placing a separate XML input stage for each of the repetition element. As I said there are 30 repetition elements in the XML and placing so many XML input stages and integrating the data is really tough.

ray.wurlod · Post by **ray.wurlod** » Fri Dec 14, 2012 3:28 pm

What exact version are you running? Is this a server job (as marked) or a parallel job (as posted)? No point our trying to guess this information, which will affect the answer we give.

ankursaxena.2003 · Post by **ankursaxena.2003** » Fri Dec 14, 2012 3:43 pm

Ray,
Currently I designed a server job(8.5), But if there is any possible solution on designing a parallel job then I will go ahead with it.

eostic · Post by **eostic** » Fri Dec 14, 2012 4:19 pm

Each repeating node needs its own "output link"....not its own stage. You can have one xmlInput Stage (or xml stage for anyone reading this that has 8.5+ and wants to use the xml stage) with as many output links as you want...... you want to parse the xml only once...but each separate repeating node needs its own link --- as though it was its own relational table...one might have (say) 3 rows and the other might have 300.

Ernie

eostic · Post by **eostic** » Fri Dec 14, 2012 4:20 pm

Each repeating node needs its own "output link"....not its own stage. You can have one xmlInput Stage (or xml stage for anyone reading this that has 8.5+ and wants to use the xml stage) with as many output links as you want...... you want to parse the xml only once...but each separate repeating node needs its own link --- as though it was its own relational table...one might have (say) 3 rows and the other might have 300.

Ernie

srinivas.g · Post by **srinivas.g** » Fri Dec 14, 2012 4:22 pm

You can do it with single XML input stage and use 30 output links(each link is having one XML chunk) and In metadata descritpion tab provide the correct path.