DSXchange

The plugins are documented, including the configuration requirements for each. Check the documentation pdf files that were installed with the client. From the DataStage Documentation pdf you can get to the DataStage Plug-In Installation and Configuration Guide pdf, which lists what is required for e...

No, you can copy across multiple Designer sessions. Just did.

Nevermind, don't leave it blank. It compiles but won't run. Silly, but I ended up putting /dev/null for the Filename in my little test job and that worked.

Hmm... that's not the behaviour I recall. Sorry. Try leaving the filename blank.

It doesn't need to know the directory any longer, as it takes its input from Standard Out when you use the Filter option. You do, however, need to specify the full path in the Filter command box.

Ray, do I still specify a file name in the "General" tab? I am not familiar with the "Cat *" command. I would surmise that it concatenates the files in a folder but do I specify a file or just the directory? You have to put something in the Filename, otherwise it complains. But, from what few times...

Yes, it is possible. One of the very nice things they finally added in the 7.x versions was copy and paste, which means you can copy a stage in one job and paste it into another. I'm running 7.0.1 right now and do it all the time. The other ong-awaited big winner was allowing 'dangling links'. :D

Well, to me the easiest way is to leverage your Operating System. Use a simple 'cat' command to create a third file from two originals: cat fileA fileB > fileC Use it in a Before Job 'ExecSH' command to prep the file your Sequential File stage is expecting. You can also use the 'cat' command without...

What else are you doing in your job? Is your query a simple select or are you doing group bys, subqueries, what? Are you taking advantage of indexes? As a test, if you cut your job back to the bone - just OCI, Transformer and Sequential file stages - how fast can it spool data from your database to ...

No, it doesn't. As Ken notes, you can't simply concatenate the two fields together and get a meaningful result. You'd need to keep them as two seperate fields and sort on both... or better yet, stick with the ISO solution presented in the other thread. My bad for simply answering the question asked,...

Try using "D/MDY[2,2,4]" instead as that actually matches how your date is coming in.

kcbland wrote:The fact that the entire world has failed to comprehend this elegance mystifies me.

Well... not quite the entire world.

Yes, you can - if you format them as suggested first.

Flip it around so that it's in a sortable format and then... sort on it. Get them into YYYY-MM-DD HH24:MI:SS format and then they should sort just fine. That is, if you want to do it inside your job, rather than outside as suggested by Ken. If you have a large amount of data, his ways will be faster.

DataStage controls access to jobs by locking them while they are being edited, thus only one person can be 'working' on (or even looking at) a job at any given time. There is a product called Version Control that ships with the client software, which is the official product for doing versioning in D...

2005?