cancel
Showing results for 
Search instead for 
Did you mean: 

How to verify data from function or data from input

former_member920516
Participant
0 Kudos

Dear Experts,

I have a question for SEM-BPS for Excel input.

Our SEM version is v3.5

Now for SEM data in our project, it contains two parts.

One is from user key in, another is from sem function (formula, copy function, etc)

Now we want to only check the data from user key in. (we check record master data)

But for function data, we do not want to check it.

Could you kindly provide some idea to solve this requirement

Thanks and best regards

Alex yang

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi.

I'm not sure that I have understood your issue ...

What do you mean by "we check record master data" ?

When you enter new data do you perform any validations ?

Could you plz ellaborate ...

Regards.

former_member920516
Participant
0 Kudos

Dear Andrew,

I mean that we have master data consolidation check for BPS.

It will check the master data.

Now we want this function only check user key in data but for function data, we want to ingore it.

If you have any problem, please inform us.

Thanks and best regards

Alex yang

former_member920516
Participant
0 Kudos

Dear Experts,

Did you have some idea on this change?

For master data check for SEM record, we use character relationship in Planning area.

But we only want to check the user key in data not function data.

Thanks and best regards

Alex yang

Answers (5)

Answers (5)

former_member920516
Participant
0 Kudos

This issue has several solutions:

1. Add one InfoObject for user id. Then in Charac Relationship code, set the condition for user id (This may impact performance)

2. Copy data into another cube

former_member5472
Active Contributor
0 Kudos

Hi,

Use versioning approach.

This will avoid any change in cube structure.

Also keep in mind if you are not using any set of data which are entered through layout or through planning function it is advisable to delete the redundant set.

Thanks

Pratyush

former_member920516
Participant
0 Kudos

Dear Sir,

Many thanks for your idea.

But for new version, we already consider it.

But due to some business reason, we can not use this way.

Thanks and best regards

Alex yang

Former Member
0 Kudos

Hi Alex,

actually I don't understand what you need.

Master data can be checked with char. relationship thats ok so far.

If you like to verify data enterd by a user you need to create an exitfunction with a function module which verfies the data in the planning buffer. If you need some information about that, you should investigate in this forum after the topic XTH_DATA.

Good luck!

Clemens

former_member920516
Participant
0 Kudos

Dear Clemens,

Many thanks for your informaiton.

Actually we create some check function and set it in planning area for characteritic relationship.

But this function will not only check user key in data but also check function data.

For second object, we do not need the function to check.

This is why I want to know how we can verify different data.

If you have any problem, please feel free to let me know.

Thanks and best regards

Alex yang

Former Member
0 Kudos

Hi

You can take one character and restrict it with a character value,for example ZTEST character with value 'E',maintain it's master data with 'E' and 'C'.Now when you load the cube you can restrict it with 'C' and restrict it in filter with the value 'E' for input ready layout.Now when you will run your function you can pickup the records where ZTEST value 'E' as they are been entered through layout.

You can also derive it through characteristic relationship for the changed record as specified in how to:-

http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/a05666e5-91bf-2a10-7285-a80b7f5f7...

former_member5472
Active Contributor
0 Kudos

Hi,

If you want to differentiate the data set, either of the approach is correct.

If data volume is less you can use versioning ( 2nd) or else you can go for the first one..

Thanks

Pratyush

former_member920516
Participant
0 Kudos

Dear Pratyush,

The data volumn is same for these two methods.

Could you kindly have some ideas to solve this?

thanks and best regards

Alex yang

former_member5472
Active Contributor
0 Kudos

Hi,

In SEM, master data validations occurs for all data going into the cube; As you mentioned you have CR maintained.

CR is maintained at cube level so it will perform its functions for all data set that is being entered into the cube irrespective of the way ( through function or through manual input).

SEM dosen't distinguish between the two while performing the checks..

If you can eleborate your requirement then we can think of some alternatives..

Thanks

Pratyush

former_member920516
Participant
0 Kudos

Dear Pratyush,

Many thanks for your information.

Currently we have two solutions for this.

1. Add new infoobject to verify different data.

After user key in data in SEM layout, after they press save, there is one copy function to copy all key in data into object value = 1.

After this, characteritic check function only check value = 1 data.

We just worried about the performance issue.

2. Use new version.

This is almost same as above solution, only difference is we copy layout data into new version.

Function only check this verison.

The issue is also for performance.

Do you have any new idea?

Thanks and best regards

Alex yang

former_member920516
Participant
0 Kudos

Dear Experts,

I have one point to update.

For user key in data.

It has two part:

1. user create new record in excel directly

2. user change the function result

Thanks and best regards

Alex yang

former_member920516
Participant
0 Kudos

Dear Experts,

Could you kindly provide some advice for this?

Thanks and best regards

Alex yang