DSXchange

It's not, but... thanks!

Ah... sorry, more of an indirect method, it seems. Should have looked more closer at the example.

For Oracle those two terms are synonymous and the log file should be in your "scratch" area, from what I recall.

Nothing posted but did find this in my archives, hopefully Ken doesn't mind it gettting posted here. Be careful with this as it could destroy more than it fixes if you get something wrong. Also note it was create for versions prior to 8.x so no guarantees with the latest and greatest. You'll need to...

In your version, project level Auto-Purge settings are only for new (or newly imported) jobs, it does not apply to all existing jobs immediately. I believe that Ken posted some job code that will take values and retroactively apply them across all existing jobs, I'll search for that.

Sorry... "extra" columns?

Agreed. I usually create small jobs that do nothing but work fine, finish with warnings or abort on command for this. Once I'm sure the Sequence structure is working, I sub in the 'real' jobs.

Your solution has already been posted here - twice now - first by myself and then again by arunpramanik. Read and apply that advice. The only thing we've had to guess on are your actual key fields but I think it's a pretty good guess.

If by "job control" you mean in some hand-coded BASIC, you can just do the assignment without any "eval" needed: varC = varB, for example. If that's not working for you, we would need more details as to why not.

Typically that would be a trimming problem, one or both sides have extra spaces so "XXX" doesn't match to "XXX ", for example. Triple-check that that's not the case.

What part are you having a problem with? :? Create two job parameters and include them in your SQL with pound/hash signs like normal. You'll need to create a mechanism to store the last value then retrieve and increment it properly, of course. That could be as simple as a flat file that you read and...

One solution: use DCount() to get the number of fields then use that count in the Field() function to get the last field:

Nothing really special about them so use the Convert() function to remove them:
Off the top of my head.

It does but not one they could actually use, seeing as how that one column essentially needs to be both a key and a non-key.

The dsjob Command Line Interface is fully documented, plus has full 'usage' statements if you run it with missing arguments like any good UNIX command. So you need to do some research, as noted check on the following options: dsjob -ljobs dsjob -jobinfo Between those two you'll get everything you ne...