Database Reference
In-Depth Information
and, depending on your definition, at least 20 possible transformations including these
10 most common ones:
1. SELECt statement if a SQL data Load rule
2. Field moves, splits, joins, creates with joins, creates with text
3. Case changes
4. Prefixes
5. Suffixes
6. Scaling
7. AnD/or operators at a field and global rule level
8. reject/Select
9. Sign flip
10. Sorting
I will illustrate the danger of a Load rule by showing Sample.Basic's Data Load rule
in Figure 2.1. Can you see the nine field edits? no? me neither.
A few, very simple transformations within a Load rule are perfectly acceptable. Indeed,
there are some actions that are best done in a Load rule or at least when done in a Load
rule provide enormous flexibility. For instance, later in this chapter, I will illustrate how
to dynamically define a dimension via a SQL-driven header; this simply cannot be done
elsewhere. nevertheless, performing extensive transformations in a Load rule is a main-
tenance nightmare because the changes can be in so many different places and because
the Load rule interface does not expose any transformations except through examina-
tion of all of the transformation options both at a global rule level and in each column.
2.3.2 The Data Challenge Remains, Even If Essbase Cannot Resolve It
Essbase excels in data analysis and has an excellent (no, amazing) calculation engine.
Essbase is not a data manipulation tool, despite the pretensions of foolish young consul-
tants to the contrary. And, yet, somehow data that is complete and accurate has to get
Figure 2.1
Sample.Basic's Data Load Rule.
