Interested to Buy Any Domain ? << Click Here >> for more details...
Source (Flatfile) - Target (1-1 mapping). How to load first or
last 20 record to target?
- 5 Answers
- 11514 Views
- Flextronics, I also Faced
- E-Mail Answers
Answers were Sorted based on User's Feedback
Answer / joe
We cannot use a sorter because it will shuffle the data.
For the last 20 rows. In a seperate pipeline use aggregator
and do a Count(key) and Max(key) to get the total number of
rows.
use an output port o_count and o_max to populate this count
to an expression transformation before the Filter.
o_count_last_20 = o_count - 20
i.e. if there are 35 records we need records from 15 to 35
In the filter check nextval between o_count_last_20 and
o_Max
| Is This Answer Correct ? | 3 Yes | 0 No |
Answer / sunny
Hello joe, your solution is partially correct - what if the
records are not in sequence to do MAX minus 20 ? I guess
this logic can be applied only after passing through
sequence generator, by getting a sequence number.
| Is This Answer Correct ? | 2 Yes | 0 No |
Answer / joe
1. Use a sequence generator to associate a NEXTVAL to the
row.
2. Filter condition : NextVal <=20
For source- dbase: Do a filter at SQ by including the
following condition . WHERE Rownum <=20
| Is This Answer Correct ? | 1 Yes | 0 No |
Answer / sri
For this you can write SQL query by TOP clause in expression
transformation.
Or else You can use Rank Transformation.
| Is This Answer Correct ? | 0 Yes | 3 No |
Answer / raju
You can also achieve using following flow in the mapping.
1)Create an out put port SNO in expression transformation
and connect it to NEXTVAL port in sequence generator
transformation.
2)Use sorter transformation.In sorter ,use SNO port as key
column and Direction is Descending.
3)Use filter transformation with filter condition as
SNO<=20.
| Is This Answer Correct ? | 0 Yes | 3 No |
if we have a delimiters at unwanted places in a flat file how can we over come those.
what is the logic will you implement to load data into a fact table from n dimension tables?
can any one give some examples for pre sql and post sql(Except dropping and creating index).
How can we use batches?
Without using any transformations how u can load the data into target?
what is the need of session partitioning... how it is useful in real time
how to create a sample mapping for dynamic lookup transformation
Workflow is long running due to long running sql query so when we refer the query plan it tells the issue is due to partition of the db table. How to handle this?
What is informatica metadata and where is it stored?
How to enter same record twice in the target table,explain?
Create a mapping which contains 2 target tables. When the session runs for the first time it shud load Target table 1 and when it runs for second time it shud load Target table 2.
without matching columns in two tables. how can you join
Teradata 
