DSXchange

The trim() function removes just one character, not a series of characters. You should use the ereplace function to do this, i.e. EREPLACE('This is a long string','long','very long') . In your case are you sure that "nt" is the only text that can show up? Perhaps you should approach the cleansing wi...

You are using DataStage PX, so you will need to maintain correct data types in all stages. If your incoming Date and Time columns are strings you can concatenate them together, but then you need to call a datatype conversion routine to convert them to the target datatype. Note that date and timestam...

In addition to pure volume and performance you could ask yourself: 1. Is there complex transform work to be done, particularly with strings 2. Is there a project-level or system-level restriction on server or px jobs? 3. Is there going to be a skill-set for maintaining the type of job you write in t...

Vijay, first off, take this code into your manager and FORMAT it; in it's present state it is very difficult to read. I am assuming that you did not write this routine but got it from somewhere. After formatting you really need to debug this yourself instead of getting some answer from this forum. I...

Kris, both errors are caused by data and should be version independant. The first error should be tracked using a reject link to catch the bad data in the datetime field. The second error is clear, your 2nd column in k_exp_CUST_DMT_DIM is declared as "not nullable" in the database yet you are trying...

Just look at the sample and additional information on these pages to get information about the examination. I think that is all the material there is to date.

Sunshine,

this should be in the PX forum, right? The limit is not a physical one, but a logical one - I wouldn't ever go above defining 3 nodes per physical CPU and the easiest and most efficient value would be 1. Perhaps a value up to 2 per CPU might be efficient in some PX installations.

There is no built-in RTF file processor.The RTF 1.5 specs are here and you might be able to write some code that will parse such a file, but the easiest way is to use some program (i.e. Word for Windows) to read the file, then write a text file which can be read easily.

The value of the text is too long

I think you were using a builting null handling function, which gets expanded inline before compilation.

No, you can't. When the after-stage is executed the information that you are looking for has not been updated in the log files. The After-job subroutine is called when this information has been updated.

I missed the 8Kb in the message. Still, that long expression used a longish link name, chopping that down to 1 could save enough bytes to compile, though.

Please don't double-post in different forums; the only thing you will reap from doing this is responses like mine.

Moderator - please delete duplicate post

I can't think of a way to do that directly. I might approach this in a multi-pass system - the first pass gets all the possible file names into some list, then one more pass for each filename to filter the XML. Not necessarily efficient, but also not terribly complex to code.

Naveen,
no, you can't