Mapping E1EDPT1 and E1EDPT2 from 2 source nodes on the same level

[Former Member]

Hi Gurus,

I am currently mapping E1EDPT1 and E1EDPT2 segments, as we know the nesting of these segments are as follows:

-IDOC

-->E1EDP01

-

---

>E1EDPT1

-

---

>E1EDPt2

my source structure is:

-Recordset

-->Header (1..1)

-->HeaderText (1..unbounded)

-->ItemDetail (1..unbounded)

-->ItemDetailText (0..unbounded)

the mapping I did is

ItemDetail --> SplitByValue(ValueChange) --> E1EDP01

create E1EDPT1 only if ItemDetailText node exists so:

ItemDetailText (exists) --> IfWithoutElse --> E1EDPT1 (where value of "then" == ItemDetail --> removeContexts -->mapWithDefault)

ItemDetailText --> SplitByValue(ValueChange) --> E1EDPT2

The problem is that if I have 4 Recordsets then I have 4 IDocs, but the E1EDPT1 is being generated only on the 1st occurence of the IDoc, on the 2nd, 3rd, and 4th occurence it doesn't get generated.

Output below:

IDOC

-E1EDK***

-E1EDP01

---E1EDPT1

-

---

E1EDPT2

-

---

E1EDPT2

-E1EDP01

-

---

No E1EDPT1 and 2 generated

the source has 2 ItemDetail and ItemDetailText nodes.

The desired output is that for every ONE E1EDP01 generated there should be ONE E1EDPT1 segment only having ANY number of E1EDPT2 below it.

Is this possible using standard functions only in PI 7.1?

I have seen this blog [url] http://www.riyaz.net/blog/xipi-convert-flat-file-to-deeply-nested-xml-structures-using-only-graphica... [url]

but I wanted to verify first from our forum if my requirement is not achievable using standard functions only and therefore will follow the blog...

my scenario by the way is FCC (File) to IDoc

Hoping for your kind response

Thanks

[peter_wallner2 · Accepted solution]

Hello HJ Dlec,

ItemDetailText (exists) --> IfWithoutElse --> E1EDPT1 (where value of "then" == ItemDetail --> removeContexts -->mapWithDefault)

Why are you doing "mapWithDefault" here? After the "IfWithoutElse" I would do "removeContexts"--> "SplitByValue".

Is your source a hierarchical or a flat structure?

Yes, you can definitely achieve this with graphical Mapping!

Best regards,

Peter

PS: I think you could even do it easier:

ItemDetailText --> E1EDPT1

will create E1EDPT1 as often as there is ItemDetailText

Constant (Empty) --> E1EDPT2

ItemDetailText --> TDLINE

Then you should get an E1EDPT1 per ItemDetailText with a corresponding E1EDPT2

and then you fill the TDLINE

Edited by: Peter Wallner on Aug 26, 2010 4:36 PM

[Former Member · Accepted solution]

Hi Peter,

Thanks for your answer, but I tried it and output is still not correct.

The E1EDPT1 & E1EDPT2 segments are produced only on the first occurence of the IDoc (1st occurence of E1EDP01 segment)

my desired output is:

Souce = 4 Item Detail each having "1" Item Detail Text for each

Recordset

-

---

Header

-

---

HeaderText

-

---

ItemDetail

-

---

ItemDetailText

-

---

ItemDetail

-

---

ItemDetailText

Recordset

-

---

Header

-

---

HeaderText

-

---

ItemDetail

-

---

ItemDetailText

-

---

ItemDetail

-

---

ItemDetailText

-

---

ItemDetail

-

---

ItemDetailText

-

---

ItemDetail

-

---

ItemDetailText

Output =

IDoc

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

IDoc

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

P.S. my question is on the Item Level only, The Header level desired output is OK

right now the current output is

IDoc

-

---

E1EDP01

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDPT1

-

---

E1EDPT2

-

---

E1EDP01

-

---

No E1EDPT1 segment

IDoc

-

---

E1EDP01

-

---

No E1EDPT1 segment

as you can see my 1st recordset has 2 item detail & item text, 2nd recordset has 4 item detail & item text

but in output the 1st occurence of the IDoc AND 1st occurrence of E1EDP01 this is where the 2 item texts are produced, which should not be the case because cardinality of E1EDP01 - E1EDPT1 - E1EDPT2 = 1:1

On the second occurence of the IDoc there is totally no E1EDPT1 produced

[peter_wallner2 · Accepted solution]

Hello HJ Dlec,

I set up your scenario here and tried around.

So I am mapping:

Recordset --> IDOC

That causes one iDoc per Recordset (IDoc is 0....unbounded)

ItemDetail --> E1EDP01

That causes one E1EDP01 per ItemDetail

Queue of ItemDetail:

ItemDetail

ItemDetail

CONTEXT CHANGE

ItemDetail

ItemDetail

ItemDetail

ItemDetail

ItemDetailText --> SplitByValue --> E1EDPT1

That causes one E1EDPT1 per ItemDetailText

Queue of ItemDetailText

ItemDetailText

ItemDetailText

CONTEXT CHANGE

ItemDetailText

ItemDetailText

ItemDetailText

ItemDetailText

Constant --> E1EDPT2

ItemDetailText --> SplitByValue -->TDLINE

That causes each text put into the TDLINEs

Queue of ItemDetailText:

ItemDetailText

ItemDetailText

CONTEXT CHANGE

ItemDetailText

ItemDetailText

ItemDetailText

ItemDetailText

Please try again!

Best regards,

Peter

PS: Is Header, HeaderText, ItemDetail, ItemDetailText really one one level, no sublevels here?

[Former Member · Accepted solution]

Hi Peter,

Thanks again for the effort in helping me out here.

Btw, Header, HeaderText, ItemDetail, ItemDetailText are not on one level only.

my Source structure is Fixed Field Length & using FCC:

Header (1..1)

HeaderText (1..n)

ItemDetail (1..n)

ItemDetailText (0..1)

so for example: (**pls disregard the spaces below, for illustration purposes only, structure is fixed field length)

ItemDetail-KEYFIELD -field1 -field2 -field3 field4 -NewLine

ItemDetailText-KEYFIELD -field1 -field2 -field3 -field4 -field5 -NewLine

ItemDetail-KEYFIELD -field1 -field2 -field3 field4 -NewLine

ItemDetailText-field3 is what I need to map to E1EDPT2-TDLINE

[peter_wallner2 · Accepted solution]

Hello HJ Dlec,

Oh, I see. It is a text file and you use the content conversion of the file adapter. But the conversion to XML works right?

By "on the same level" I meant are there sublevels in the structure?

Please look at:

http://public.fotki.com/pezibaer420/pi-hj-dlec/text-test-mapping-struc.html

and look on the left you see the structure, I hope you can read it. Everything is below "RecordSet" on the same level and "RecordSet" is unbounded.

Does your structure look like that? If yes, then my answer from before should work!

Beste regards,

Peter

[Former Member]

Hi Gurus,

I am also facing little bit similar problem for text segment. I am using one of the user exit for passing the text segment data. But the standard program is not sending this segment (E1EDPT1 & E1EDPT2) to my user exit. i mean is there any configuration setting which we need to do to pass text segment in IDOC type (ORDER05). Here i am using an extension IDOC which has the basic type as ORDER05.

Please help.

Thanks.

Sourabh Batwara

[peter_wallner2]

Hello Sourabh Batwara,

Please open a new thread for this - we cannot start mixing threads! Your question is different from the original question here!

Thank you!

Best regards,

Peter

[Former Member]

HJ Dlec

.

.....iis being generated only on the 1st occurence of the IDoc, on the 2nd, 3rd, and 4th occurence it doesn't get generated.

I have a question How to populate the IDOC ?.

Did you change the ocurrence to accept multiple IDOCs .

Check /people/michal.krawczyk2/blog/2005/12/04/xi-idoc-bundling--the-trick-with-the-occurance-change

and the another is are HeaderText ,ItemDetail, ItemDetail Text on the same level?