DSXchange

New problem = new post. This has nothing to do with grouping by a date column so shouldn't be discussed here. Otherwise we'd just allow one post per user (probably with a subject line like 'Hi' or 'Urgent') that just gets longer and longer. As noted earlier, did you try sending the internal formatt...

You are at the wrong 'command line' - dsjob is an operating system command. So no 'uvsh' or 'dssh' before-hand, just run it from DOS.

Sure there is - just 'Add to schedule' over and over at 15 minute intervals.

Ah... missed that fact, first time through, that you are storing the string of unprintable characters in common so they are only built once. That will certainly help.

How is this different from your other post?

Ok, I can keep that in my back pocket. I was worried that with a solution of that nature that there would be a performance impact, as we could be applying it to several million records in a run.

Guess I could run some timing tests, with and without...

I'm just wondering how people are handling invalid characters in XML they are generating? Specifically, we build a string element with a user-supplied 'Business Description' and today one had a ^C in the middle of it. Things like "&" get properly translated but it just passed the ^C thru in the ...

You were asked "what has changed" not "what did you change", quite a difference there.

As noted, search the forums for JobMonApp - your issue with it has been asked and answered several times already.

Seems to depend on the stage author, what exactly gets captured where. Suggest you create a test and include all of the 'Link Variables' in the reject record, see which has what you are looking for. REJECTEDCODE? SQLERR? Don't recall what all is there.

All true, but it has long been acknowledged that the DataStage implementation (Xerces, I believe) of the XML stages isn't exactly the fastest solution out there. Pretty sure even Ernie has posted to that effect. So, when people look for more speed generating their XML, they tend to either: A) Use th...

[sigh] Simple as that.

Thank Gawd I didn't post my big goofy Swiss Army Knife version. At least now I know what the trimmed down version will look like. With the appropriate copyrights to Wurlod Industries included, of course.

Start here: Read about multiple instances here and in the documentation. We cover it extensively. The technique is to divide (partition) your source data into sets and use multiple job instances to process the different subsets (partitions) of the source data simultaneously, thus, you utilize more c...

select 'What do you mean?' from dual where stage_name in ('FUNNEL','AGGREGATOR','MERGE','PIVOT') Sorry, but that's all that came to mind when I saw your post. Perhaps someone else has something more constructive, but I honestly have no idea what you are trying to find out here. There are no...

OK, one hurdle down. Next problem - do you have a command line utility that will encode a string down to its binary equivalent? That's what I assume you mean by 'to 0s and 1s'. Recognize the fact that the stage is looking for a 'command line', meaning there is no DataStage utility at work here but r...