DSXchange

I have a job which tries to update rows on a table. If these rows don't exist on the table, then under some conditions I get warnings and under other conditions I get no warnings. Does anyone know why, or what the rules are for this behaviour? It seems that if I use 'Generated SQL' (update existing ...

Still wet and windy here in Yorkshire - but I'm off to sunny Liverpool-on-sea for the weekend. :D I was hoping that I could write a generic job that would work out the column names itself from the information that the db2 stage gets from db2 - and that way I wouldn't have to worry about passing in t...

I'd like to know if anyone has a good way of getting the column name so that it can be used as data? What I'm trying to do is... Read from a database table (using rcp for the column definitions), then lookup to a reference table which has a keyfields of column_name and value. So, for each field, I w...

Hi Arnd! I don't think Aakash is around today.

I get...

: Executed command: echo "\000" | tr '\000' '?' | wc
*** Output from command was: ***
1 1 2

Does this (setting ur) have to be done as 'user defined' sql then - or can it be done on any of the other 'Read methods'?

Or should you 'set change isolation to ur' as an 'Open Command' (would that work?)

Forgive my ignorance but I'm trying to learn about the restructure stages so that I'll be able to spot when they'd be useful. Can anyone give me an explanation of a subrecord and a vector. Does column import/exports work on subrecords or vectors - what's the difference. Are vectors always within sub...

Thanks all.

I think that most of what I was trying to get the modify stage to do is done better in a column generator then.

Thanks all, I'll start using those stages with confidence now.

Just for information, the other thread with ray.wurlod refers to is...
viewtopic.php?t=114683&highlight=

ok - thanks. dsusr, I'd already done what you've suggested (and another couple of different ways of doing it) - and it gives the right output without the warning - but its quite a lot slower (job start up time is about twice as long, and job run time is about 3 times as long), and this isn't as quit...

The way I found recommended in the documentation was to use a Surrogate Key Generator stage, with 'Execution Mode' set to sequential
and 'Collector Type' to round robin.

Any use?

Why don;t you turn it around? If you remove the duplicates first and after that you sort it the way you want the result to be sorted, it shouldn't be a problem. JeroenDmt, the problem is, I don't really want to remove all duplicate rows (based on the Key1 and Value), I only want to remove a duplica...

dsusr - here's that code...
(what I refer to as Key1 in the example is actually called ARRG_ID in the code, and what I refer to as DateKey in the example is actually called REC_ACDT in the code)

I've removed the code from here - it's dull

I'd been advised not to use a column generator because 'if its in the "Development/Debug" section of the palette its only to be used in Development or debugging - and not in live jobs." What's your opinion, is there anything wrong with using "Development/Debug" stages in liv...

I'd been told not to use a column generator because 'if its in the "Development/Debug" section of the palette its only to be used in Development or debugging - and not in live jobs." What's your opinion, is there anything wrong with using "Development/Debug" stages in live j...

Thanks for that. I've found a way I can add a field onto a row and set it to '1' It relies on 1 of the original fields being not nullable. I've used the following... MyNewIndicatorField:Int8=notnull(OriginalNotNullableField) If I just put this one line in, I get the new field set to 1 for every row,...