Search found 53125 matches

by ray.wurlod
Wed May 30, 2007 11:18 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Partitioning with Transformer stage
Replies: 8
Views: 2718

Report a bug in the version 8 client to your support provider. It works fine in version 7.5.2.
by ray.wurlod
Wed May 30, 2007 11:16 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: reject link in Oracle stage
Replies: 4
Views: 1049

NULL will be dropped because you can not transfer NULL into a not null field in DataStage. And primary key columns are ALWAYS not null. You need to intercept the null farther upstream in your job and direct it into an ODBC table. However, null primary key will not be allowed there either.
by ray.wurlod
Wed May 30, 2007 11:12 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: using SIEBEL PACK
Replies: 8
Views: 1612

Did you reset the job, as I asked? How many messages were logged by the reset run?
:roll:
by ray.wurlod
Wed May 30, 2007 11:09 pm
Forum: IBM<sup>®</sup> SOA Editions (Formerly RTI Services)
Topic: General SOAP Quesion
Replies: 5
Views: 2803

OK, scrub it.
by ray.wurlod
Wed May 30, 2007 7:06 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Partitioning with Transformer stage
Replies: 8
Views: 2718

Yes it's possible. Open the stage properties, select the Inputs tab, select the Partitioning tab. Choose the partitioning algorithm from the drop-down list. On the Outputs tab you can set the Preserve Partitioning flag.
by ray.wurlod
Wed May 30, 2007 7:03 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Message Digest function in Datastage
Replies: 5
Views: 1791

I got the feeling that what was required was a function with a repeatable algorithm. Yes this could be done with a string key and generated sequence column.
by ray.wurlod
Wed May 30, 2007 7:00 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Paremetrize the columns in a DB stage.
Replies: 7
Views: 1764

Try relying upon Runtime Column Propagation. You will, of course, need to specify the join key, so that may limit your flexibility somewhat.
by ray.wurlod
Wed May 30, 2007 6:09 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Message Digest function in Datastage
Replies: 5
Views: 1791

What you are seeking to write is the first half of a hashing algorithm. There's nothing out of the box in DataStage. But you could write your own. CRC32() is dangerous with large volumes, as it carries approximately a one in four million chance of generating the same value for two different inputs. ...
by ray.wurlod
Wed May 30, 2007 6:06 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Code of different Stages in PX
Replies: 2
Views: 818

Source code for the Stages is IBM proprietary information (intellectual property). Similarly source code for Orchestrate operators and APT classes is IBM intellectual property. None of it is in the public domain. If you look in $PXHOME/include you will encounter header files that will get you most o...
by ray.wurlod
Wed May 30, 2007 4:31 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Routine Source Code
Replies: 4
Views: 839

Go back to the machine where the routines were created and export them again, this time checking the Source Code box. Without that the source code is not exported, so you import only executable code.
by ray.wurlod
Wed May 30, 2007 4:23 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Concatenate duplicate values..
Replies: 2
Views: 825

Lookup against the hashed file and construct D then write back to the hashed file.
Disable both read cache and write cache, and lock for update when performing the lookup.
by ray.wurlod
Wed May 30, 2007 4:17 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Help Required
Replies: 4
Views: 998

What is your jo design intended to do? Are there any business rules that have been implemented and which therefore need to be tested? Create a test plan, including test data with which to test the efficacy or otherwise of your implementation of those business rules.
by ray.wurlod
Wed May 30, 2007 2:40 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: error Loading from Sequential File
Replies: 2
Views: 810

Get the fixed widths RIGHT.
by ray.wurlod
Wed May 30, 2007 2:38 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Read-only jobs
Replies: 7
Views: 1482

Curiously I answered the wrong question, yet the issue is now marked as resolved. What I advised was how to unprotect a project. Re-reading the original question, what I should have answered was how to change a job's status from readonly. That can be determined via a Search on READONLY. However, the...
by ray.wurlod
Wed May 30, 2007 2:34 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: using SIEBEL PACK
Replies: 8
Views: 1612

That's not the same warning as your original post. Is it the same job?

Did you reset the job, as I asked? How many messages were logged by the reset run?