Implementation
The customer defined fields that are stored in the CUGEX1 table are implemented in such a way that it is already included in hierarchy in Source and Dimension spaces. There is no need to manually setup scripted sources and hierarchies into a separate model space as they are already integrated into the dimensional hierarchies.
In the Source space, a Birst Connect query defined in Modeler Enterprise imports CUGEX1 records from the Data Lake. This data is processed and stored in the CUGEX1 lookup, which is a persistent storage of up-to-date extension field values. This lookup is shared through packages to the Dimension and Fact spaces. CUGEX1 extension scripts in these spaces are used to affix and update the extension fields in the hierarchies.
Scripted sources in the Dimension space follow a standard naming convention and are always included in a specific publishing group. The publishing group that contains scripted sources that update these fields is not yet included in the main workflow as this feature is optional. This feature needs to be manually added in the workflow steps. See KB2114690.
| Column | Data type | Comment |
|---|---|---|
| FILE | Varchar(6) | This column contains the exact source table name for which customer defined fields are added in M3. |
| CONO | Integer | Company value. This is used in conjunction with the PK Columns to identify a specific dimension record. |
| PK01 to PK08 | Varchar(30) | These columns contain the exact matching values of the dimension source key fields. Eight key columns are provided for futureproofing. |
| A030 to A930 | Varchar(30) | The ten string attributes with length of 30 that you can fill for a table. |
| A121 to A122 | Varchar(120) | The two string attributes with length of 120 that you can fill for a table. |
| A256 | Varchar(256) | A string attribute with length of 256 that you can fill for an M3 table. |
| N096 to N996 | Number | The ten number attributes that you can fill for an M3 table. |
| CHB1 to CHB9 | Integer | Used for Yes or No flag attributes. |
| DAT1 to DAT9 | Integer | This column is for date fields that used M3 date integer format. |
| variationNumber | Integer | This column is used for incremental loading the custom defined fields. If the dimension record already contains the additional fields from the CUGEX1 table, it is only updated if the variationNumber in the CUGEX1 table is higher than the variationNumber in the dimension. |