DSXchange

ScreenOutput will be a dynamic array, where every line in the output is a line in the array. Read up on DSExecute in your DS BASIC manual.

uvconfig is in the DSEngine directory on Windoze as well. Check it out. T30FILE is not a true table, but a shared memory thingy. There are no old and obsolete entries. A reboot once a month never hurts. Your issue is too many simultaneous dynamic hashed files being used. Just up the number, you'll b...

Why not set COMMIT to 0 on the output link, on the reject link set Abort after N rows constraint to 1. The abort will cause a rollback if 1 row rejects.

You can use the LENGTH function to get the number of characters in a string. INDEX finds the character location of a substring within a string. COUNT counts the number of characters. Here's some code off the top of my head: If LENGTH(yourstring) <= 255 Then FirstLine = yourstring SecondLine ...

Thanks, but that isn't ISQL for Sybase or SQL-Server. My findings so far says that it can't be done the way I'd like, but I was hoping for some solution better than the one I'm facing.

This is not a DS question, but I thought someone out there may have an answer. Google is not working well for me today, and Sybase and Microsoft can't seem to get me to the answer. I need to run a SQL script from a DS job. The script has a few variables in it that need resolution at runtime (from da...

Supposed to be key3, I fixed the original post.

Thanks again for your help with the UtilityHashLookup. This is my company's first project with DataStage so I didn't really have an opportunity to join (much) sooner. Only 3 days I knew that blank defaulted to the current project, but I thought I'd be sure and obvious by typing in the current proje...

Ken, So if a hashed file has two keys defined, how can we specify that in the utility then? You have to pass that to the function concatenated together using an @TM delimiter. A multi-part key would look like: UtilityHashLookup("",key1:@TM:key2:@TM:key3, "")

You should have joined the forum sooner... Leaving the "account" (project) blank just defaults the hashed file to the current project.

The comments inside the utility function are: * Executes a lookup against a hashed file using a key * * Input Parameters : Arg1 = Hash Table Name * Arg2 = Hash Key Value * Arg3 = Column Position to return. * If empty, return entire row in a dynamic array. Arg3, if left blank, returns the entire row....

Your key is position 0, the attribute you need is position 1. Try passing a "1" in the third argument.

By the way, welcome aboard.

Take this opportunity to switch to using externally pathed hashed files to a specified directory, you won't have this problem anymore. If you have persistent large hashed files, that another issue. You could write a job to transfer the hashed file from the project to a directory using the opportunit...

Just create a new project, export/import the jobs, and drop the old project

Have you considered writing a Server job to do some things that you wish, then use multiple job instances to handle partitioned parallelism "the old fashioned way"? If it's PX or bust, then forget I mentioned it. You may find a mixed-bag approach could be feasible and a little more pleasant.