DSXchange

Hi All,

We have created a huge xml file(8 million recs) by using xmloutput stage.The job aborts in the middle of xml extraction.

Kindly let me know how to split the same into 2 or more xml files.





Raj.

Is the job aborting when you are trying to read the xml file or while you are creating it?

Tim

Assuming the problem is creating it, look into the use of the Trigger Column option, value changes there will trigger a switch to a new output filename.

chulett wrote:Assuming the problem is creating it, look into the use of the Trigger Column option, value changes there will trigger a switch to a new output filename. ...

He said there are 8 million records. By enabling this, wont it create as many files?

It depends entirely on the logic of the job. The "trigger" is just a column on the input link.....what you set it to is up to you.....a series of counters that (for example) lead to a column that is always "1" for the first 500,000 rows, and then "2" for the next 500,000....etc. would be one way to use it.

Ernie

Exactly - you control when the values changes and it doesn't matter what it is. We typically used a Mod() function to increment a counter every X output records.