Null file Handling

Ambuj743 · Post by **Ambuj743** » Sat Sep 24, 2005 7:57 pm

I want to pass a null sequential file in the datastage job and want to generate a new sequential file with one row. The same file will be used to compare another sequential file.

ray.wurlod · Post by **ray.wurlod** » Sat Sep 24, 2005 11:50 pm

What exactly do you mean by "a null file"?

The design is easy enough, but needs a specification of what must be in the one row of your output file. For example you could use a Row Generator stage to produce that. But what should be done with the "null file", whatever that is?

Ambuj743 · Post by **Ambuj743** » Sun Sep 25, 2005 7:51 am

I am passing a file which contains zero record (blank file) and trying to insert one row.Please suggest me the steps.

ray.wurlod wrote:What exactly do you mean by "a null file"?

The design is easy enough, but needs a specification of what must be in the one row of your output file. For example you could use a Row Generator stage to produce that. But what should be done with the "null file", whatever that is?

roy · Post by **roy** » Sun Sep 25, 2005 8:25 am

Hi,
So you want to read an empty file and produce 1 row that contains what?

IHTH,

Ambuj743 · Post by **Ambuj743** » Sun Sep 25, 2005 8:42 am

That contains 1 record with all the values zero.

roy wrote:Hi,
So you want to read an empty file and produce 1 row that contains what?

IHTH,

Ambuj743 · Post by **Ambuj743** » Sun Sep 25, 2005 8:44 am

That contains 1 record with all the values zero.

roy wrote:Hi,
So you want to read an empty file and produce 1 row that contains what?

IHTH,

chulett · Post by **chulett** » Sun Sep 25, 2005 9:58 am

Reading an 'empty' file will produce exactly zero rows of output.

If you want to create a job that writes out one row, then build a Server job that starts with a Transformer and outputs to your desired 'all values zero' file. Add a stage variable to the Transformer so the compiler doesn't bark. Add a constraint to the Transformer of @OUTROWNUM=1 so it stops after writing one row. Hard code the values in all derivations to zero. Run. Done.

ray.wurlod · Post by **ray.wurlod** » Sun Sep 25, 2005 2:30 pm

A Row Generator stage would be more efficient. Again specify to generate one row. Put zero into each column. You don't need an input file, null or otherwise.

Why not just

Code: Select all

echo '0,0,0,0,0,0,0' > filepath

as a command?

chulett · Post by **chulett** » Sun Sep 25, 2005 3:14 pm

ray.wurlod wrote:A Row Generator stage would be more efficient.

Really? More efficient? Or just more PX'ian?

ray.wurlod · Post by **ray.wurlod** » Sun Sep 25, 2005 10:27 pm

More efficient. No overhead in calling the library created by compiling the Transformer stage, which would be significant in a one-row job.

I think the echo approach would be the most efficient of all.

dsxdev · Post by **dsxdev** » Wed Sep 28, 2005 5:23 am

Hi Ambuj,
I hope your case is to use a job which will read an empty file and generate a new file with a record containing default values.

Since you are using this job to compare with another file, this job would be run conditionally. More precise run this job only when the input file for this job is empty.

Any way if you want to do it you can do this in an After job routine which would be a simple way to do it.

Check the link statistics or line count of the file and just write a new record.
This would work fine.

DSXchange

Null file Handling

Null file Handling

File which contains zero record.

Generate a dummy record reading by reading an empty file.