DSXchange

The corrupted hashed file RT_LOG35 corrupts its index when the point of corruption (address 0x944) is accessed or referenced. Could this log have reached 2GB in size at any point (which is suggested by the 0x80000000 address)? Clearing RT_LOG1188 won't fix the problem in RT_LOG35, but clearing RT_LO...

Use a sequence with two Job activities. The first job determines whether there are any rows in the table and parks that result, say in its user status area. This activity has a custom trigger that fires only if that count is greater than zero. The second Job activity performs the truncate.

And, if that assumption holds, you only need one Convert() function.

So what you really want is a counter? Sort the data by the grouping column then use stage variables in a Transformer stage to generate/reset the counter. The Sort stage gives you a Key Change column, so the ability to detect change by that means allows you to use just one stage variable. svCounter <...

If your data are likely to contain single "<" or ">" characters prefer the Ereplace() function. This is not supplied by IBM for parallel jobs, but one of our posters has posted the code here on DSXchange.

Simply refer to it in your pattern action language. And make sure that the reference table is in the same data folder as the rest of the rule set, so that it will be provisioned when you use "Provision All".

This is best achieved using a server job and the dynamic array handling abilities of the Hashed File or UniVerse stage types.

If you must use a parallel job, then it looks like a job for a Column Import stage to parse the nested information, and a Pivot stage to generate the multiple rows.

The "report on job" is a compilation option in the sequence's Job Properties.

By getting whoever supplies the file to give you a legal CSV format - fields that contain delimiter characters must be quoted.

Frighten your client with a stern lecture about data governance.

I reckon someone may have attempted to use Derivation Substitution and got it wrong.

Like Sainath I would advocate storing hashed files at the project level, because they are meant to be shared between jobs in the project.

Your model does not actually provide for different projects.

Look at the various XML stage types.

Open up the \Stage Types\Parallel\Real Time branch of your repository. Open the properties of each stage type in turn noting in particular what's documented in the Long Description field. This shoul ... thank you Ray, where can I find the stage types.....can u tell me the path in the designer. than...