DSXchange

There are a few date/timestamp manipulation function in UDB. Your best bet would be google.
Check this site out for a start. Also the IBM's public library.

Are you using a Basic Transformer. If yes then Craigs solution would work. Actually it would work even with the Parallel Transformer because all that needs to be done is the change detention and restricting the output to 11.

Or maybe he meant DataStage Control Job as in job control, Ray, wheres your 'ReadMyMind Stage'. Really need it now

Craig is right. IBM, Acsential tailored a product that attends to the general needs, and to some extent, specific needs of an ETL process. Restartability is the outer layer. Its one of the key elements of a process. It needs to be designed/tailor coded at times. No doubt DataStage Sequence jobs have...

Thanks guys. The DSX community has always been very kind and supportive.
And Craig, yea, its one of those

Thanks buddy. And as i said then, still a long way to go

I, personally agree with you Craig. Such a need arises if 'SHGVVW'

Abracadabra hocus focus select * from books where book in (select t.book countversion from books t where t.version between #start_version# and #end_version# group by t.book having count(distinct t.version) = #end_version# - #start_version# + 1) assuming that table name is books and ...

how to get prevBook and preversion. i have shown it in my stage variables. when i say in.book, that means your column 'book' and 'in' is for input, so in other words, your input column 'book', similarly for your version. From your input just drag and drop the book and version to these two stage var...

These jobs functioned before, only that I migrei they stops another environment, and started to give these errors. I think you answered your own question. Its the environment. See if ORACLE_HOME is set up correctly and is pointing to a valid ORACLE_HOME. Look into your dsenv file in the DSEngine. D...

Ok for that you cannot do it in a single transformer. You need to determine what books are in order, flag them and then in the second transformer only let them pass through. Use the conditional checks in the first transformer to build a hashed file with only one column 'book'. So basically in the fi...

I dont know where does your db2profile sit.
For me its

Ray,
The link that I had posted is not for the website, but a unix directory path where the OP can find some examples to see from.

So let me make sure i got this straight!
for 1-100 you are missing version 1 and 3 and for book 1-200 you are not missing any version. So you have to output
1-100 0
1-100 3
for the data sample that you gave. Correct?
Or you just have to raise a red flag if they are out of sequence.

Yes. I think thats what it is. You are not sourceing the db2 profile in your dsenv file. And yes you have to manually put that in there. Also, try building an odbc connection to your db2 and then try to import the db2 table meta data from odbc import and see if you can do that. Then try using the DB...