Mapping problem.

[Former Member]

Hi all.

Have the following situation.

The source message can look something like this.

<?xml version="1.0" encoding="UTF-8"?>

<ns0:sourcerecord xmlns:ns0="http:">
   <s_record>
      <ponum>1</ponum>
      <polinenum>1</polinenum>
   </s_record>
   <s_record>
      <ponum>2</ponum>
      <polinenum>1</polinenum>
   </s_record>
   <s_record>
      <ponum>1</ponum>
      <polinenum>4</polinenum>
   </s_record>
   <s_record>
      <ponum>1</ponum>
      <polinenum>3</polinenum>
   </s_record>
   <s_record>
      <ponum>2</ponum>
      <polinenum>3</polinenum>
   </s_record>
   <s_record>
      <ponum>2</ponum>
      <polinenum>2</polinenum>
   </s_record>
</ns0:sourcerecord>

And i want the following result at the target.

<?xml version="1.0" encoding="UTF-8"?>

<ns0:targetrecord xmlns:ns0="http:">
   <t_record>
      <ponum>1</ponum>
      <linerec>
         <polinenum>1</polinenum>
      <linerec>
      </linerec>
         <polinenum>3</polinenum>
      <linerec>
      </linerec>
         <polinenum>4</polinenum>
      </linerec>
   </t_record>
   <t_record>
      <ponum>2</ponum>
      <linerec>
         <polinenum>1</polinenum>
      <linerec>
      </linerec>
         <polinenum>2</polinenum>
      <linerec>
      </linerec>
         <polinenum>3</polinenum>
      </linerec>
   </t_record>
</ns0:targetrecord>

As you see the fields coming in a node with the same ponum should be collected in ansubnode to for the ponum and sorted. The sort are no problem but the big poblem I have are to get all the rigth fields in the same node.

I hope somebody can help me.

BR

Andreas

[justin_santhanam · Accepted solution]

Andreas,

Please see the below snapshot and let me know if it helps!

http://www.flickr.com/photos/23855877@N07/2551290946/sizes/o/

http://www.flickr.com/photos/23855877@N07/2551290948/sizes/o/

http://www.flickr.com/photos/23855877@N07/2551290950/sizes/o/

http://www.flickr.com/photos/23855877@N07/2551290952/sizes/o/

Results-http://www.flickr.com/photos/23855877@N07/2551290954/sizes/o/

Thanks,

raj.

[Former Member · Accepted solution]

That was exactly the solution I was looking for.

Thanks Raj.

Points on the way.

Will also award points for the xslt solution.

[Former Member]

Hi Andreas,

here you are tring to group the records by ponum, so you nedd to use the below udf in your graphical mapping.

Give the name of the udf as "groupingByPonum" with the cache type queue and two agguments ponum,polinenum:

	public groupingByPonum(String[] ponum,String[] polinenum,ResultList result,Container container) {
		int i,j=0,k=0,i1,j1;
		Vector V[]=new Vector[ponum.length];
		String[] pos= new String[ponum.length];	String t=new String();		
		for(i=0;i<ponum.length;i++) {
			for(j=0;j<k;j++) {
				if(pos[j].equals(ponum<i>))	break;
			}
			if(j==k) {
				pos[k]=new String(ponum<i>);	k++;
				V[j]=new Vector();				
			}
			V[j].addElement(polinenum<i>);
		}
		Vector temp=new Vector();
		for(i=0;i<=j;i++) {
			for(i1=1;i1<V<i>.size();i1++) {
				for(j1=0;j1<i1;j1++) {
					if(Integer.parseInt(V<i>.elementAt(j1).toString())>Integer.parseInt(V<i>.elementAt(i1).toString())) {
						t=(String)V<i>.elementAt(i1);	V<i>.removeElementAt(i1);	V<i>.insertElementAt(V<i>.elementAt(j1), i1);	V<i>.removeElementAt(j1);	V<i>.insertElementAt(t, j1);
					}
				}
			}
		}
		for(i=1;i<=j;i++) {
			for(k=0;k<i;k++) {
				if(Integer.parseInt(pos[k])>Integer.parseInt(pos<i>)) {
					temp=V<i>;	V<i>=V[k];	V[k]=temp;
				}
			}
		}
		for(i=0;i<=j;i++) 
			for(k=0;k<V<i>.size();k++) 
				result.addValue(V<i>.elementAt(k));

	}	

Now do this graphical mapping:

ponum>removeContexts>sort>splitByValue\[value changed\]>collapseContexts-->t_record

ponum>removeContexts>sort>splitByValue\[value changed\]>collapseContexts>splitByValue\[each value\]>ponum

ponum>removeContexts>sort>splitByValue\[value changed\]>linerec

ponum>removeContexts>groupingByPonum>splitByValue\[each value\]>polinenum

polinenum>removeContexts

(in the udf give two inputs

1: ponum-->removeContexts

2: polinenum-->removeContexts)

that's all......

you can use this udf for any kind of grouping operation in message mapping.

--Sankar Choudhury

[Former Member]

Hi,

if you want you can try this XSLT mapping:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"  version="1.0">
  <xsl:output method="xml"/>
  <xsl:key name="po" match="s_record" use="ponum"/>
  <xsl:template match="/">
    <ns0:targetrecord xmlns:ns0="http://test.it">
      <xsl:for-each select="//s_record[generate-id(.)=generate-id(key('po',ponum))]">
        <xsl:sort select="ponum" order="ascending"/>
        <t_records>
          <ponum>
            <xsl:value-of select="ponum"/>
          </ponum>
          <xsl:for-each select="key('po',ponum)">
            <xsl:sort select="polinenum"/>
            <linerec>
              <polineum>
                <xsl:value-of select="polinenum"/>
              </polineum>
            </linerec>
          </xsl:for-each>
        </t_records>
      </xsl:for-each>
    </ns0:targetrecord>
  </xsl:template>
</xsl:stylesheet>

Copy it in a file with ex. xslt and import it in mapping.

Be careful to put a valid namespace in the source xml.I used for istance xmlns:ns0="http://test.it", otherwise you cannot test it.

Regards,

Emiliano

[Former Member]

Ok, with the UDF I mentioned above instead it should be mapped as:

ponum -> ponum

ponum, ponum, polinenum -> doMap -> linerec

(change the context of the second "ponum" and "polinenum" to "sourcerecord" - right-click->context->...)

ponum, ponum, polinenum -> doMap -> splitByValue -> polinenum

(change the context of the second "ponum" and "polinenum" to "sourcerecord" - right-click->context->...)

Regards,

Yaghya

[Former Member]

How do you map t_record?

What occurences do you use for the nodes and fields?

//Andreas

[Former Member]

ponum -> removeContexts -> t_record

Can you please clarify your question regarding occurences of fields and nodes? Do you still need to define the data types?

Edited by: Yaghya Nana on Jun 4, 2008 2:35 PM

[Former Member]

No I have the data types but I think you are not using the same occurenses as I do.

[Former Member]

The occurences are correct and are handled by context changes.

[Former Member]

hi

check with standard function FORMAT BY EXAMPLE

check the below blog which illustrates the above function

/people/stefan.grube/blog/2005/12/29/new-functions-in-the-graphical-mapping-tool-xi-30-sp13

regards

chandrakanth

[Former Member]

Create a User Defined function "doMap" with type "context" that takes three parameters a, b, and c.

String currPonum = a[0];
for (int i=0; i< b.length; i++){
if (b<i> == currPonum)
	result.addValue(c<i>);
}

Now map as follows:

ponum -> ponum

ponum, ponum, polinenum -> doMap -> polinenum

(change the context of the second "ponum" and "polinenum" to "sourcerecord" - right-click->context->...)

Regards,

Yaghya

Edited by: Yaghya Nana on Jun 4, 2008 2:04 PM

[Former Member]

Did you read my question before I edited the reply i wanted in the system?

Just to be sure because about 2 min after I posted I changed the spec on what the solution needed to cover.

BR

Andreas