OLEDB Command
Parameterized queries are a powerful means to extract just the information needed
to satisfy a given criteria. You would use a parametric query when you try finding
patient information, given the patient-ID from a patient database. This transformation
requires a data flow task, and an OLEDB data source or a flat file source.
Term Lookup
This transformation and term extraction transformation that follows provide support
for word search in text documents. The frequency of occurrence of words is captured
using a reference containing words to be matched. This is an excellent tool for
preparing custom lists of words in a text with statistics.
Percentage Sampling
This transformation allows you to sample out a percentage of the data (Percentage of
total rows) which may be useful in testing a package with a smaller set of data. For
example, instead of using the full thousand rows of data, the package may be tested
with a smaller number of the rows, which is representative of the entire set. You will
find an example in Chapter 9.
Pivot
While normalized data are a standard for OLTP applications, de-normalized data are
common in OLAP. A list of rows containing data does not reveal hidden information
in a simple listing, but by carefully grouping data, hidden features can be revealed.
Pivot transformation does it automatically if it is properly configured.
Row Count
As the name implies this transformation counts the number of rows that go through
the data flow.
Pages:
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38