How to avoid first five rows from the source in Importing

[subbuvp]

Hi,
I have 200 rows and i need to import the 95 rows to my source (all are in text files).
So how can i avoid first five rows to sourch while importing...

Can you please help me to solve this

thanks

Subbu

[Sreenivasulu]

Hi

The is a DSLinkRowCount function by which you can ignore the first 5 rows. Set DSLinkRowCount > 5.

Regards
Sreeni

[thebird]

Hi,
So how can i avoid first five rows to sourch while importing...

Use the Filter option in the Sequential file stage. If you know the number of rows in your file (200 in this case) you can give Tail -195 in the filter and this will read only the last 195 records from the file.

But is the number of records in your file going to be a constant?

[balajisr]

You can parameterize the rows to be read in the sequential file filter as

tail -#rowcount#
where rowcount is the job parameter. You need to pass the appropriate value to rowcount job parameter.

[kumar_s]

So how can i avoid first five rows to sourch while importing

You can use

Code: Select all

tail +6 filename

. Also note that it does not works on all flavors of unix.

[narasimha]

So how can i avoid first five rows to sourch while importing...

Subbu,
This can be your other option,
If you have a transformer in between your two files, add the below constraint

Code: Select all

@INROWNUM > 5

[thebird]

This can be your other option,
If you have a transformer in between your two files, add the below constraint

Code: Select all

@INROWNUM > 5

This wouldn't work as there would be Fan out happening from the file and the data would get partitioned when it reaches the transformer - and the ordering would be lost.
Moreover as the transformer is working in parallel, each node would try to execute that constraint - which would not meet the requirement.

Aneesh