Default Columns with Surrogate Key (SK) as 0 and -1 will be inserted into
destination (DIM) table, only if data is present in the table
DIM_SCD_SEEDED.
PRTY_LOOKUP_REQD_FLG should always be set to ‘N’.
3. The data type of SK column in destination (DIM) table should always be
INT/BIGINT and it will be generated using the following logic:
MAX_SKEY + row_number(n) where (n) is rowid.
Query to fetch Maximum SKEY value will give performance improvement, if
indexing is done on DIM table.
Stage Column where Column Type = ‘ED’ should be updated with Date in Hive
Format – ‘yyyy-mm-dd’.
Apart from this only ‘dd-Mon-yyyy’ format is supported to keep the current
seeding intact. Final data in Date column will always be inserted in
‘yyyy-mm-dd’ format.
Columns which are not part of STG and DIM mapping will be passed as ‘’ (empty
strings).
Columns with column type STRING/VARCHAR/CHAR will be inserted as empty strings
and all other column types will be inserted as NULL.
Stage table should not contain duplicate records for the same MISDATE.
Two or more SCDs executing in parallel should not update the same Dimension
table. In such cases, ensure the processing is sequential. Similar limitation is
applicable for the option Map Ref No: =-1.