DSXchange

Daunting, tedious, but not difficult. Suggestion, build a few at a time in stage variables: there is an upper limit on the length of an expression.

And beware of nulls.

I don't believe so. Certainly the posted quote is truncated. I have no convenient way to test your theory.

Clicking on Help would have answered this question. A protected project is protected mainly from, you guessed it, developers!

There's another DataStage role (Production Manager) which is the only one that can load objects into a protected project.

tunning: (pp)
filling a tun

tun: (n)
1. a large cask for holding liquids, esp. wine, ale or beer
2. a measure of liquid capacity, usually equivalent to 252 wine gallons

No wonder there's a need to want to save space!

You use it anywhere an expression can be used. In a Transformer stage this includes stage variable initialization, stage variable derivation, output link constraint expressions and output column derivation expressions. You can use IsNull() in expressions in job sequences also. And you can use the fu...

The outputs are different also. And the Compare stage gives yet another variant.

OSH code is not executable, except in the appropriate virtual machine. It's a similar model to Java or Perl (or even DataStage BASIC) in this respect. Orchestrate shell scripts are not executable as operating system executables.

Are they truly fixed width? Certainly the examples you posted are not. You probably can use schema files, but only downstream of reading the single field first. Perhaps you can parse in a Transformer stage, perhaps you can parse in a Column Import stage. Perhaps you can get more coherent requirements.

What precisely do you mean by "not accepting"?

You get the "OEM 7.5 manuals" from IBM through your support provider. They should be at no charge if delivered as PDF documents.

This is really the same question as the one in this post which was also posted by you today.

Setting Length to a positive value should be all that's required, unless you're using a schema file in which case you need to get it right (for example string[max=20]) in the record schema described therein. Enable the environment variable that causes player processes to log the actual schema each i...

Of course you don't need to partition. Just don't expect correct results if you don't. Even in an SMP environment, you need Entire partitioning to make use of shared memory. If you choose some other partitioning algorithm (hopefully a key-based algorithm such as Hash or Modulus) then any given key v...

Welcome aboard. All the documentation I've read suggests that binary data types are not supported at all in server jobs. So the fact that you're reading them at all is a bonus. It is possible to set NLS maps on a per-column basis, though it will slow processing marginally. I can only suggest that yo...