dynamic Header handling in File(Fixed length) to Proxy scenario

[Former Member]

Dear all,

My souce file may get n number of headers and trailers.

I need to group as the below file structure.

My file structure is

H(Header Detail record),D(Detail Record)

<ROOT>

H1

D1,D2,D3.......

H2

D1,D2,D3,...

...

...

...

...

HN

D1,D2...

</ROOT>

my question is can this structure handled in SAP PI?

if yes give me pointers on how to handle .

Thanks,

Srinivas

[Former Member · Accepted solution]

Hi Sree, you didnot specify about the trailer in the file structure...and also elaborate more on the file strcuture and file coming in...

[Former Member · Accepted solution]

There is no Trailer

That Structure has multiple headers and detail records which is a .txt file .

I need to segregate header with its details the second header with its details and third header with its details.

Please let me know how to handle this (mapping and FCC) .

Eg:

Header1 x w e c n
a d e f g
Header2 d
w d
Header3 g
z d g h

..

...

Header n a

there will be 1 key value to differentiate bet header and detail

THANKS,

SREE

[baskar_gopalakrishnan2 · Accepted solution]

You might want to check this blog that gives idea about pattern recordset structure.

http://scn.sap.com/people/shabarish.vijayakumar/blog/2005/06/08/content-conversion-patternrandom-con...

[former_member184681 · Accepted solution]

Hi,

Use the following set of parameters for FCC:

Recordset Name = Document
Recordset Structure = Header,*,Item,*
Recordset Sequence = Ascending (this is very important for further mapping)
Key Field Name = LineType

Header.fieldNames	LineType,Field1,Field2
Header.fieldSeparator	;
Header.endSeparator	'nl'
Header.keyFieldValue	H
Item.fieldNames	LineType,Field3,Field4
Item.fieldSeparator	;
Item.endSeparator	'nl'
Item.keyFieldValue	D

As a result, you will get the following source structure for each document that occurs
<Document>
<Header>
<LineType>H</LineType>
<Field1>content</Field1>
<Field2>content</Field2>
</Header>
<Item>
<LineType>D</LineType>
<Field3>content</Field3>
<Field4>content</Field4>
</Item>
...
</Document>

Some more important design points:

• You need a field that will let PI distinct between Header and Item. In my example, it is LineType, that equals H for header lines and D for details, but of course this can be any value anywhere in the structure. It doesn't even have to be at the beginning of the line.
• Thanks to Recordset Sequence = Ascending, you will get a new Root each time a new Header occurs. This will allow you to do the mapping properly.
• After you perform a source conversion like this, you can go on with the scenario as usually, building graphical mapping or anything else you need.

Hope this helps,
Greg

[Former Member · Accepted solution]

In the FCC parameters u need to use the below paramters, if u have an unique identifier for header and detail records..something like the below structure..

H,1

D,1

D,2

H,2

D,3

D,3

so, H and D will be the unique identifiers for header and detail respectively.

Parameters

header.keyFieldValue

header.fieldFixedLengths or header.fieldSeparator (it depends on file)

header.endSeparator

header.fieldNames

header.processFieldNames (value should be "fromConfiguration")

detail.keyFieldValue

detail.fieldFixedLengths or header.fieldSeparator (it depends on file)

detail.endSeparator

detail.fieldNames

detail.processFieldNames (value should be "fromConfiguration")

hope this helps...

[Former Member · Accepted solution]

I am having multiple headers as mentioned.

if the input is h1,h2....
                   d1,d2....D may belong to h1 or h2.I could able to design it.

if the input is h1->d1,d2,dn.
                    h2->d1,d2,dn

How to design the same.

[Former Member · Accepted solution]

Reopening the same in other thread with exact requirement.

[Former Member · Accepted solution]

Hi Shree,

In order to establish relationship between header and detail records, you need to have a common field between header and detail structure that will hold disnict values for each header and same values for all detail records belonging to a particular header.

Please refer to the below example.

<Document>

<Header>

<LineType>H</LineType>

<HeadNo>123</HeadNo>

<Field1></Field1>

</Header>

<LineType>H</LineType>

<HeadNo>124</HeadNo>

<Field1></Field1>

</Header>

<Detail>

<LineType>D</LineType>

<HeadNo>123</HeadNo>

<Field2></Field2>

</Detail>

<Detail>

<LineType>D</LineType>

<HeadNo>123</HeadNo>

<Field2></Field2>

</Detail>

<Detail>

<LineType>D</LineType>

<HeadNo>124</HeadNo>

<Field2></Field2>

</Detail>

</Document>

In the above structure, we have 2 headers. The 1st header has 2 details and 2nd header has 1 detail records. Here the node <HeadNo> is used to group detail records based on headers.

In configuration, Greg has written all you need.

Hope this helps.

Thanks & Regards,

Samir

[baskar_gopalakrishnan2]

Reopening the two threads for the similar requirements does not help author and experts. This is mere waste of time. Close one thread if you create an another one.