#/null character in CHAR infoobject from flat file

[former_member221140]

hello,

I am loading an InfoCube from a flat file and directly mapping to the flat file datasource up to the Infocube. One of my fields, zsite, which is of type CHAR infobject does not have a blank value (#) in the file, yet when I load the file there is a # value in the SID table, how does this get there and how do I remove it? I ask because I am using the zsite infoobject as a prompt via Query Browser in Dashboards and it is displaying the '#' character.

Thanks,

[Former Member · Accepted solution]

Hi Gary,

The system always generates one entry in the SID table for a characteristic to be able to represent the "Not Assigned" value.

There is no way to get rid of the value in the SID table. However, you can set the info object property - Filter Value Selection During Query Execution - so that during query exection it only displays "posted values", e.g. values that are actually in the cube.

Cheers

Henry

[sat_p2 · Accepted solution]

Please try option suggested by Henry Janz.

and also try to restrict with exclude option in query level.

[former_member221140 · Accepted solution]

That is what I thought... (the generated not assigned entry)

I did as you suggested and changed the InfoObject property in BEx under the Advanced tab to both

Access Types for Result Values: Set to "posted values" (unchecked use characteristic setting)

Filter Value Selection During Query Execution: Only Values in InfoProvider (unchecked use characteristic setting) and also tried Only Posted Values for navigation.

The # sign still shows up in the variable as a possible value when I run the query. I have also set an exclusion filter for # - no dice

appreciate the help, thanks!

[sat_p2 · Accepted solution]

try to put in Default values section( excluded with Not assigned values).

and properties set as only selected from info provider and save the query.

[colm_boyle]

Hi Gary,

the # symbol doesn't only represent null values, it represents all disallowed charachters in the SAP application. My advise would be to trace the value in the PSA and try to see the value in the source flatfile afterwards.

You can enter these symbols/values/charachters in transaction RSKC to add them to the list of permitted characters, see the following note...

1373402 - Issues with invalid characters or SID Generation

Best of luck,

Colm

[former_member221140]

Hi Colm,

Thanks for the response.

That's interesting, I didn't know # represented all disallowed characters.

In the PSA there is no blank records or records with special characters, as far as I can tell. Are you suggesting I add the value of each field in the PSA to RSKC to figure out where the special character may be if there is one?

I'm not getting any errors upon activation or loads as this note suggests.

Even if I look at something like the master data table of 0MATERIAL there seems to be a blank record.