UPDATE SAP ANALYTICS CLOUD IMPORT XLSX

[f_allocca]

Hi Experts,

How it's possible to define key field during an update process of a new xlsx file?

in the mapping area i don't see any option to define it.

In my case, I need to refresh my data model with modified rows. After the updating, the result in SAC is a duplication of the rows modified in the new xlsx file. I want to underline that i haven't modified the key fields in the xlsx file, but only the values of other fields.

Can someone help me to understand better how the UPDATE works in SAC?

[JensBraun · Accepted solution]

Hello Francesco,

for update mode all dimensions are considered key fields, so the import method update is for changing measure value and adding new rows.

If you want to change dimension values I suggest that you use the import method "Clean and update selected version data" or "Clean and replace subset of data", both are based on the version.

If you want to schedule regular updates from files I also recommend that you load the data from a file server using the SCP Cloud Connector (see here for more info https://www.sapanalytics.cloud/guided_playlists/connect-file-server/). That way you can load new data from the file without having to reapply your transformation logic in the wrangler.

Regards,
Jens

[SayedZubair · Accepted solution]

Hi jensbraun, All

I have a similar question. I have an import model connected to a file server and created a job to import the data.

Used all the 4 import methods and in all its observed that any change in the Dimension value gets added as a new row rather than actual change in the dimension. Is this right?

Ex:

In file:

Name: ABC, PQR, XYZ

Updated Data in file:

Name: ABC, PQR, HIJ

Output in SAC post import schedule

Name: ABC, PQR, XYZ, HIJ

How to overcome this issue?

Regards,

Sayed Zubair

[JensBraun · Accepted solution]

Hello Sayed,

this is the typical challenge with data import: how to handle deletions.

In your case you provide a full update of data and I suggest that you try the reset model option that is available in the import settings

This is of course only applicable if this your only source and if you really import all the data.

Otherwise I think that the clean and replace subset option should work:

Best regards,

Jens

[SayedZubair · Accepted solution]

Hi jensbraun

Yes, I am following the "Clean and Replace" and it kind of helps. Hopefully this doesn't create data duplicity later.

Regards,

Sayed Zubair