count rows in a sequential file

[PeterPol]

Hello,

I want to calculate the number of rows in a sequential file by using the Count rows option in the aggregator stage.
Unfortunately, I get no rowcount when the sequential file is empty.

Any idea how to solve this?

Peter

[ray.wurlod]

Execute Command activity in job sequence to dir the file and get the size from that, on the basis of which you can decide to execute the job or not.

Clearly, if DataStage processes 0 rows, there's nothing there to count.

[Maveric]

You can use an 'External Source' stage. give the command to be executed as 'wc -l'. this returns the value as an integer. output will be a single column, single row (count). this sometimes might return two rows of same data. just change the spaces before and after '-' and try. it is only one space but not sure if it is before or after the '-'. Not sure about Aggregator stage though. this is not exactly wat you wanted but just another way of getting the source count. hope it helps.

[ray.wurlod]

Didn't realize that Windows has a wc command, or that server jobs have an External Source stage. I suspect that neither is true.

[Maveric]

my mistake. will never again post without reading the question properly. Anyway the command is wc -l Filepath\filename. this is for parallel job and Unix.

[PeterPol]

Hello,

This one works for my situation:

- Use Sybase OC stage to create a single row (SELECT 1)
- Use the Link Collector stage to append the source rows of the sequential file with this extra row.
- Use the Aggregator stage to count the number of rows
- Subtract one of this count to get the number of rows in the sequential file.

Peter

[ray.wurlod]

Clever. Resolved?

[coolstar4en]

Thansk PeterPol and everyone who replyed this topic，I got same problem today and resolved after reading this topic.

thank you all!