PI Mapping to Idoc Creation - Something is missing

[elizabeth_stapleton]

Hello there

can anybody help with this please?

I have created a non-standard inbound Idoc interface.  Which is fine, as far as I've got.

But I have had to increase the length of some of the fields in the Idoc segment after it was released.

This is now released from ECC client 100 to 110, and the idoc structure has been reimported from ECC back into PI Netweaver.  I can see as part of the WSDL that the field lengths have increased as required.

When I run a file through, looking at the PI mapping, everything looks fine   All the data is in the right place.

But when I come to look at the Idoc itself via BD87 I can see that the data has moved up.

By this I mean the first longer field has brought in the first few characters of the following field.

The same has happened with each of the 3 fields I have extended....so ultimately the data appears to have squashed into the top part of the Idoc structure, leaving the last fields empty.  e.g

Where before I had 3 fields

ABC, DEFG, HIJ

I now have

ABCDEFG, HIJ,

So what ever is reading in the data to create the Idoc, doesn't recognise the new field lengths.

I am not sure what creates the actual Idoc in the EDID* tables

Help would be most gratefully received

Many thanks

Elizabeth

[former_member186851 · Accepted solution]

Hello Elizabeth,

Did you referesh the Idoc metadata in PI?

[elizabeth_stapleton]

Many thanks for the assistance.

I looked in the RFCs for the Unicode Mark.   In comparison with others we have set up, they looked ok, though admittedly I don't know an awful lot about that side of things.   All good learning though.

Deleting the Idoc via IDX2 worked, so excellent thank you for that

Regards

Elizabeth

[Former Member]

Delete the metadata in PI in IDX2(IDOC adapter) or in NWA(IDOC_AAE) adapter and try.

[markangelo_dihiansan]

Hi Elizabeth,

Can you try checking your PI+ECC unicode settings? The same thing can also be observed for RFC sender.

Regards,

Mark