DSXchange

Use spaces as a delimiter, then trim out duplicates and leading/trailing, then change to dashes. Try:

But thanks for taking a shot at helping out

I think all utilities which runs on mainframes work the same way. even when datastage runs in unix. it translates the code into a language (orchestrate) unix understands. which other way you expecting datastage to run:-P lol Not with regards to mainframes. DS/390 is totally separate framework, just...

In the reference link, you need a key expression to match to the primary key. Create three links: link_A, link_B, link_C. I'm so confused so here's two different answers: Map column A as the key expression for link_A, B for B, C for C. If the incoming value matches, you'll get your value. Or Put lit...

Build another hashed file with col_code as the primary key and ID as the attribute.

Your original question stated you needed to align a source column named A to a specific lookup for 3 known values. I now do not understand your followup question because it doesn't correlate to your original.

Sounds like you need multiple reference links to the same hashed file, each one with different literal values in the key expression.

If you don't require the full table as a reference source and only need a small portion, scan your source data and spool the unique identifiers (natural or primary keys) to a text file, load that text file into a work table in your target database, inner join that work table to your large table and ...

Do you mean "start" a PX job running via a call from a mainframe? The answer is that the command line Unix program dsjob doesn't care who runs that program. If you mean, can you run PX or Server jobs on a mainframe? The answer is NO. That's why DS/390 exists. Do you mean can you run PX or Server job...

One example is Samb, a package that lets Windoze machiens write to Unix filesystems as if they are mapped drives.

You can still use FTP stages, as well as command line scp and ssh. Since your example is doing a directory listing, ssh is a great option.

Welcome aboard! If you paste "string too long, limit 8192" into the Search facility and choose Exact Match, you will have your answer in about 3 seconds.

I did a project two years ago with MySQL, it was the same situation. What we ultimately did for the customer was use the 3rd party ODBC driver simply for metadata gathering, but did all DB processing thru scripts. We found that the command line was suitable for the types of processing you do in ETL....

You can specify how to pad, either globally or on an individual field level. Some you may want space filled, others zero filled. It's all in there...

You can deadlock yourself if you have parallel streams loading different table partitions with a global index. This is why locally managed indexes should be highly considered, because you can load a partition without conflict to other partitions. When two partitions are contending to update the same...

Missed it by: that much!