Search found 581 matches

by Kryt0n
Wed Jul 25, 2012 4:40 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Not In clause in Filter stage...
Replies: 8
Views: 15160

The solution given is identical to your second sample that you said works. But if it works, why a problem, just keep using it.

Your first sample is wrong because Field1 will never equal all three values at the same time therefore two of the three checks will always be true
by Kryt0n
Tue Jul 24, 2012 4:27 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Loading Decimal field data from fixed width Flat Files
Replies: 8
Views: 11017

Read it as 8,0 and then divide by 100 in a transformer to your output of 8,2
by Kryt0n
Tue Jul 24, 2012 1:00 am
Forum: General
Topic: Job is not in the right state (compiled and not running)
Replies: 60
Views: 23581

You will also need to make sure every caller uses a different Instance ID when calling the job
by Kryt0n
Mon Jul 23, 2012 4:29 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Scenerio with datastage limitations
Replies: 9
Views: 9024

An issue I foresee with the external target is that it executes against every line so your redirect should really use ">>" and you need to ensure the file doesn't exist/is empty prior to running. Alternatively, run the command supplied via a sequence command stage if source is already in t...
by Kryt0n
Thu Jul 19, 2012 11:53 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: APT_ORACLE_PRESERVE_BLANKS
Replies: 12
Views: 6968

Are you bulk loading or inserting?

Push your data to a sequential file instead of your oracle stage so you can see if any modification has happened prior to oracle
by Kryt0n
Wed Jul 18, 2012 9:57 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: how i can achieve this result in datastage 7.5 px
Replies: 11
Views: 4926

Admittedly their request was as vague as can be managed but think they are trying to do a lookup, not merge columns
by Kryt0n
Wed Jul 18, 2012 4:46 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: how i can achieve this result in datastage 7.5 px
Replies: 11
Views: 4926

nani1974 wrote:I think you can do this by using Column_Export stage.
I'm intrigued... how?

As for requirement, create a date range out of your lookup.

Or depending on data sizes and db abilities, do a sparse lookup getting rows with date <= current date (as per row) order by date desc and limit to 1
by Kryt0n
Wed Jul 18, 2012 4:37 pm
Forum: General
Topic: Datastage import using command line
Replies: 9
Views: 6517

Are you trying to do an import or export?

Not sure dscmdimport/dscmdexport can do a folder at a time, think it is whole projects. Look at "istool" as an alternative although think you have to accept you won't get a "dsx" from it.
by Kryt0n
Wed Jul 18, 2012 4:31 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: ETL logic using server jobs
Replies: 4
Views: 3743

You can do it the way you specified but would need to add a "key" which could be the DIV by two.

Alternatively, you keep keep every odd row in a Stage Variable and output on every even row
by Kryt0n
Wed Jul 18, 2012 4:29 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: date format
Replies: 1
Views: 1037

There is an admin level setting for the century cutoff, alternatively you can specify year cutoff in your format, I think it is %[2000]yy or of such ilk, help files should... help
by Kryt0n
Mon Jul 16, 2012 4:50 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Index Position of Last Occurrence of String
Replies: 13
Views: 10616

I actually meant count rather than dcount (had dcount on my mind due to other thoughts). As for efficiency, I see minimal difference unless the odds on having 2 or more @ symbols in the string is high, since you are always searching for 2 in your index query, it is likely to scan the full string whe...
by Kryt0n
Mon Jul 16, 2012 4:41 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Table definitions at runtime
Replies: 5
Views: 2494

Might be possible through RCP
by Kryt0n
Mon Jul 16, 2012 4:37 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: datastage
Replies: 4
Views: 2139

Can only see that happening if RCP is enabled otherwise it wouldn't compile. And if RCP is enabled and this is a runtime error, it's more because of a reference to SRC_SYS_CD when that field isn't in the source stream.
by Kryt0n
Mon Jul 16, 2012 4:26 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Unable to read a Seq File
Replies: 4
Views: 2936

Agree with the above although wouldn't %dd expect a two digit day? Can't say I've ever tried against a 1 digit day as I've autmatically used %(d,s)
by Kryt0n
Sun Jul 15, 2012 4:46 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Index Position of Last Occurrence of String
Replies: 13
Views: 10616

Or that DCount(field, '@') = 1...