It can
bring in data from files together with associated resources such as graphics, binary
documents, etc. into database columns. Again there are restrictions on the output
data type being limited to DT_Text, DT_NText, and DT_Image.
Lookup
This transformation looks for data in a reference database. The package that uses
this transformation requires a dataflow task. It is carried out via EQUIJOINS with the
reference database.
Merge
Merge transformation is similar to "union all" transformation. It simply merges
two sorted datasets into a single dataset. Merging is carried out using key column
values. Merging data from two data sources, such as tables and files, nested merge
transformation, etc. are possible.
Merge Join
Merge join by definition, requires at least one data flow task and two data flow
components to participate in the join. Just like merge transformation, merge join
transformation also requires sorted inputs. Several different kinds of joins, such as
LEFT JOIN or LEFT OUTER JOIN, RIGHT JOIN or RIGHT OUTER JOIN, FULL JOIN or FULL
OUTER JOIN, and CROSS JOINS are possible using this transformation.
Chapter 1
[ 19 ]
Multicast
As the name implies, the data is copied to multiple outputs (every row is copied).
This is different from a conditional split; it can be used in cases where you may want
to apply different transformations to the same data set getting sorted, while, the
other undergoing some kind of aggregation.
Pages:
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37