Graphical Mapping issue

[Former Member]

Hi gurus,

we have a following challenge to solve;

INPUT

<root>
<header>
	<stats>
		<ID>AAA</ID>
		<color>green</color>
    	</stats>	
    	<stats>
		<ID>BBB</ID>
		<color>blue</color>
    	</stats>
    	<stats>
		<ID>CCC</ID>
		<color>red</color>
    	</stats>
</header>
<item>
	<order>
		<ID>AAA</ID>
		<value>1001</ID>
   	</order>
   	<order>
		<ID>AAA</ID>
		<value>1002</ID>
   	</order>
   	<order>
		<ID>BBB</ID>
		<value>1003</ID>
   	</order>
   	<order>
		<ID>CCC</ID>
		<value>1004</ID>
   	</order>
<item>
</root>

NEEDED OUTPUT

<root>
<header>
	<stats>
		<ID>AAA</ID>
		<color>green</color>
	<order>
		<ID>AAA</ID>
		<value>1001</ID>
   	</order>
   	<order>
		<ID>AAA</ID>
		<value>1002</ID>
   	</order>
    	</stats>	
    	<stats>
		<ID>BBB</ID>
		<color>blue</color>
	<order>
		<ID>BBB</ID>
		<value>1003</ID>
   	</order>
    	</stats>
    	<stats>
		<ID>CCC</ID>
		<color>red</color>
	<order>
		<ID>CCC</ID>
		<value>1004</ID>
   	</order>
    	</stats>
</header>
 

Who can help me?

Thanx in advance for any hints!

Regards,

PM

[PriyankaAnagani · Accepted solution]

Hi Peter,

Try with the below logic....

header--->header

id>removeContext>sort>SplitByValueChange>CollapseContext--->Status

id(change context to ItemLevel)->sort>SplitByvalueChange-->Order

id & color are one to one mapping...For VAlue use the below logic

formatByExample--->value

formatByExample:

input1:resultofSortByKey

input2:id(change context to ItemLevel)>sort>splitByEachValue

SortByKey:

input1:id-->removeContext

input2:value-->removeContext

Regards,

Priyanka

[Former Member]

Hi,

Use the mapping like below

Status to Status node mapping

Status-ID to Status-ID field mapping

Status-Color to Status-color field mapping

Under status we have Order node.

create queue type UDF

Inputs are Header-ID, Item-ID, Item-value

String flag='A';

for(int i =1; i<Header-ID.length; i++)

{

for(int j=0; j<Item-ID.length; j++)

if ((Header-ID<i>).equals(Item-ID[j]))

{

result.addValue(Item-value[j]);

flag='B';

}

}

if (flag.equals('B')) result.addValue("");

Map this UDF output to Order node

Map this UDF output to SplitByValue node function then Value field in Order

For ID field:

take the ID field from Order node then pass to OneAsMany node function as first parameter, 2nd and 3rd inputs are above UDF (After udf use SplitByValue then map) then output map to status node ID field.

Regards,

Venkata Ramesh

[former_member193306]

Hi, Peter!

try this mapping:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
    <xsl:output method="xml" indent="yes" version="1.0" encoding="UTF-8"/>
    <xsl:key name="dep" match="root/header/stats" use="concat(ID, '|', color)"/>   
    <xsl:key name="dep1" match="root/item/order" use="concat(ID, '|', value)"/>   
    <xsl:template match="root">
        <root>
            <header>
                <xsl:for-each select = "header/stats[generate-id(.)=generate-id(key('dep', concat(ID, '|', color)))]">
                    <stats>
                        <ID>
                            <xsl:value-of select="ID"/>
                        </ID>   
                        <color>
                              <xsl:value-of select="color"/>
                        </color>
                        <xsl:for-each select = "../../item/order[generate-id(.)=generate-id(key('dep1', concat(ID, '|', value)))][ID = current()/ID]">
                            <order>
                                <ID>
                                    <xsl:value-of select="ID"/>
                                </ID>
                                <value>
                                     <xsl:value-of select="value"/>
                                </value>
                            </order>
                        </xsl:for-each>
                    </stats>
                </xsl:for-each>
            </header>		
        </root> 
    </xsl:template>
</xsl:stylesheet>

Thanks & Regards

Stanislav

Edited by: Stanyslav Bobrovskiy on Dec 6, 2011 2:22 PM

[Former Member]

Hi Peter,

UDF is required to populate order record values in status record, also we need to populate the target side nodes in a order (not all status records in a sequence and not all the order record values in a sequence).