DSXchange

do you have two columns checked as a "key"....? each link can only have one column marked as a key...it should be one of the columns "in" the repeating element.

Spend some time on the forum looking thru XML posts...in particular, make sure you download and do a thorough read of the XML Best Practices (it's on Kim Duke's web site...you'll find that url in lots of places). You have two repeating units (so far)......number and attributes. They have to be built...

It's non-negotiable decisions like that which ultimately lead to higher costs of ownership.....there are pros and cons to everything. This same site will potentially be in a situation 3 years from now when they have serious DataStage experience but their long time DBAs have all left the company....a...

It works. I just did a test with your snippet of xml (btw...it is very invalid xml... where did you get this snippet from? I'll assume for the moment that you typed it in? It's missing proper close tags and the spaces in the element names are a problem too...perhaps they are underscores in your actu...

Well....you're on the right track if you have been able to retrieve the parent info ok. Ultimately you will need a minimum of two links to retrieve this data, because there are at least two independent repeating nodes....Item Desc and Item Grade Name. I'm a little concerned about the lack of a conta...

I'm not an expert on the "details" of it, but many customers I speak to on this topic have implemented an LDAP solution....and then the mainstream products such as Information Server and Business Objects, etc., will typically support LDAP as an authentication mechanism. This way the user n...

...and forget about Data Lineage with a 100% SP solution.... Data Lineage can be done with SP's but not without a whole lot more effort than just doing things in DataStage either via auto generation of SQL or even user-defined. I talk to sites who are in the process of moving away from unmanageable ...

....and for now there is no migration....choose the new XML Pack as you move forward, are looking for better performance, or need new functionality. As with anything, there are always decisions to be made....if I have a Job that reads a very small XML document from an MQ message queue, and I know it...

Excellent question. Right now, no changes. They exist "as is" and continue to function "as is".

Ernie

What tooling are they using to dump the files? There are lots of tools out there that will put the column headers into the first line.....perhaps they are using a method that supports this..... ultimately that's all you need and will be the simplest for everyone if the files are pure ascii delimited...

Multiple instances and "starting every two minutes" are two different things....... The only reason it would "start another instance" every two minutes would be because you still are not at the minimum, and ISD needs to start additional ones. Either that or you have your minimum ...

Look at the output link of your XML Input Stage.....go to the Descriptions of each column..... chances are, the elements you are retrieving are missing the final /text() from the xpath syntax. Without the "text()" it means "get the whole element"....... ....so it should be .../.....

Yes...but that's not new.....it is just like the existing pack has always allowed you to read 100's of documents that use the same schema, in one shot, at one time, with all their rows in a single load.

Ernie

Great pointer, Vincent, and even greater post in your blog! The new XML Pack also specifically helps with a lot of things we've seen in threads here on dsXchange --- such as reading extremely large documents --- it streams them so it reads them just fine, and it reads them much faster. It also optio...

No "direct" way that I know of to import a SQL create stmt "as is". I've seen a whole lot of solutions over the years, such as creating a .dsx (look at one for a table, it's a very simple format), or using the manual shared table creation method that I describe below at my blog. ...