Search found 53125 matches

by ray.wurlod
Thu Feb 12, 2009 1:54 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Grouping the records of DS_AUDIT
Replies: 15
Views: 5584

What do you mean by "date format"? Give an example of your desired output.

Also, since you don't have premium membership yet, you can't read all of my previous answer.
by ray.wurlod
Thu Feb 12, 2009 1:51 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Transformer BlockSize and node_node1: Player 1 terminated
Replies: 2
Views: 4549

Can you explain more completely what SEF is, and how it extracts data from UniData? Are you using a state file for the surrogate key storage? What syntax are you using for the function that delivers the next surrogate key value in the Transformer stage?
by ray.wurlod
Thu Feb 12, 2009 1:45 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Convert datetime to gmt datetime
Replies: 3
Views: 1481

Did you search for "GMT" and "daylight saving"?

Have you looked at the time-handling transforms and routines in the SDK?
by ray.wurlod
Thu Feb 12, 2009 1:41 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Job is not in the right state
Replies: 7
Views: 2592

First, I would rewrite it so it doesn't abort. Abort is anathema to unattended running.

I have no idea about whether your wait time is adequate, because you have not provided enough information about the expected arrival mechanism of file X.
by ray.wurlod
Thu Feb 12, 2009 1:38 pm
Forum: General
Topic: Snowflaking
Replies: 2
Views: 935

There are reasons, but they should be rare. Allow me to quote from The Data Warehouse Lifecycle Toolkit by Kimball et al: Generally snowflaking is not recommended in a data warehouse environment, Snowflaking almost always makes the user presentation more complex and more intricate. Database designer...
by ray.wurlod
Thu Feb 12, 2009 1:31 pm
Forum: General
Topic: Traceback error
Replies: 5
Views: 3200

Transformer stage functions are intolerant of NULL. You must handle NULL explicitly.

Code: Select all

 If IsNull(InLink.FieldName) Then SetNull() Else Trim(InLink.FieldName)
Of course you can generate a default value, say "", instead of using the SetNull() function.
by ray.wurlod
Thu Feb 12, 2009 1:28 pm
Forum: General
Topic: Appending leading Zeros and writting to excel file
Replies: 9
Views: 2346

Can you prefix the result with a single quote character, so that Excel treats the value as literal text?
by ray.wurlod
Thu Feb 12, 2009 1:25 pm
Forum: Information Analyzer (formerly ProfileStage)
Topic: Datastage Export from Commandline
Replies: 8
Views: 9591

In version 8 exports are done from the Designer, since there is no longer any Manager client. What Craig asserts is true for exports done from Designer.
by ray.wurlod
Thu Feb 12, 2009 1:41 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: DSGetLogSummary - having to offset start time by one hour
Replies: 5
Views: 1763

Add it to the project in the Adminstrator client. Then, perhaps when you use it as a job parameter, set up the default value as $ENV (which means "pick up the value from the underlying UNIX environment).
by ray.wurlod
Thu Feb 12, 2009 12:46 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: DSGetLogSummary - having to offset start time by one hour
Replies: 5
Views: 1763

In the log early in the job run is an entry listing environment variable values. What is the value of the TZ environment variable?
by ray.wurlod
Thu Feb 12, 2009 12:40 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Grouping the records of DS_AUDIT
Replies: 15
Views: 5584

1. DTM is VarChar, so you can use LIKE operator. WHERE DTM LIKE '2009-02-11%' 2. It's the MAX(DTM) in your query that is eliminating some of the User2 modifications. Lose the MAX() function. 3. You don't need IN if the sub-query is SELECT MAX(DTM)... because the sub-query can only ever return one ro...
by ray.wurlod
Thu Feb 12, 2009 12:36 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: reading the file blockwise
Replies: 5
Views: 1320

Specify precisely what you want to do.
Read the file one line at a time, with the whole line as a single VarChar (or Char) field.
Parse into fields in the Transformer stage.

Another approach would be to use the Complex Flat File stage rather than the Sequential File stage.
by ray.wurlod
Thu Feb 12, 2009 12:34 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: String to Date conversion error
Replies: 5
Views: 5992

When your job is compiled the stages you use are converted into Orchstrate operators, and the functions that you use to convert data are converted to those used by the modify operator (which is also used for the Modify stage). As a general rule the "To" function names are for the Transform...
by ray.wurlod
Wed Feb 11, 2009 4:35 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Migrating jobs from v 7.5.1 to IIS v8.1
Replies: 20
Views: 6277

Export/import is the approved technique.