Search found 459 matches

by T42
Tue Mar 01, 2005 3:50 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Failed to connect JobMonApp on port
Replies: 11
Views: 14949

Yeah. That's why it's called yousucks.h, Ray. :wink:
by T42
Tue Mar 01, 2005 3:49 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Material Configuration for a Server
Replies: 6
Views: 2537

parsimonious
Trust Ray to teach you a new word a day!

Heh. Where the heck did you come up with that word, Ray?
by T42
Tue Mar 01, 2005 3:45 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Write data to Seq. File - Remove Additional Delimeter in it
Replies: 5
Views: 2790

Trim() can do it. Convert() can do it. Each works in a different way -- just depend on what you want to do. You will need to pass a hex value of the newline character in order for that to work with newline character.

("0xblah")
by T42
Tue Mar 01, 2005 3:26 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: How to split data into files?
Replies: 15
Views: 5866

You will need to create a buildop or custom stage to do this type of task.

Just pre-sort the data, and throw it in a buildop which will open a file, watch the data flows through, and when the data changes, close the file, and open a new one.

Buildop definitely will serve you well on this one.
by T42
Tue Mar 01, 2005 3:19 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Join Stage
Replies: 4
Views: 2437

Ray is correct. In fact, you probably are better off with a Lookup stage. Heavy volume? Hah. Maybe for output, but not for a lookup. 300k of reference file that does not have to be sorted? Do a lookup. Join stage sorts within the framework (unless you already sort the data beforehand.) That adds time.
by T42
Tue Mar 01, 2005 3:15 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Failed to connect JobMonApp on port
Replies: 11
Views: 14949

I just discovered that there was a nasty bug within JobMonApp. If your job crash hard (i.e. "Error on line 10234 of file yousucks.h"), your JobMonApp may go down with it.

Fortunately, this bug was fixed for 7.5.
by T42
Tue Mar 01, 2005 3:11 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Warning Message
Replies: 3
Views: 2432

Modify Operator (Stage) used to be the primary method for conversion of nullability type. Now you can do it within Transformer Stage. Do an "If IsNull(Field) Then Value Else Value" Modify Operator use OSH functions, not DataStage functions. So IsNull() will not work. is_null() will work.
by T42
Tue Mar 01, 2005 3:02 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Retruning strings (char pointers) from C routines
Replies: 10
Views: 5818

Re: Retruning strings (char pointers) from C routines

I would like to create a PX routine that returns a (modified) substring of an input string, much like this one posted by another user of this forum Modified substring? That can easily be done in Transformer stage. Give us your rules, and I can whip something up without you having to build a BuildOP...
by T42
Tue Mar 01, 2005 2:59 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: which stage is more efficient: Lookup, Join or Marge
Replies: 5
Views: 2712

As a rule of thumb if I have under 50,000 reference rows I try to use the Lookup stage. It has the best interface of the three and has good reject and conditional lookup capabilities. Only 50k? Geez, a bit stingy with your Lookup use. I have seen excellent (read: better than Join) performance upwar...
by T42
Tue Mar 01, 2005 2:57 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Material Configuration for a Server
Replies: 6
Views: 2537

Read the Install and Upgrade Guide. I'm sure they have included a minimal set of requirements for you.

But again, as Ray says: More is BETTER. What is the most you can afford? That's the one you're going to get.
by T42
Tue Mar 01, 2005 2:51 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Hex to integer in transfrmer stage
Replies: 10
Views: 5343

Unfortunately, I am not able to test my solution on 7.5. I did test it on 7.1r2, and it worked. I did not do anything special (AsInteger() or anything like that) -- just plain straight varchar to integer conversion, doing the conversion with the values you provided. It produced the expected results....
by T42
Tue Mar 01, 2005 2:46 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Lookup Range
Replies: 12
Views: 5498

Ranged Lookup is not yet supported in EE. It is a feature coming very soon now (or probably already in 7.5.1, need to double check that.)

You will need to use a buildop if you're using a EE solution. Routines are also a good alternative if you're using Server.
by T42
Tue Mar 01, 2005 2:42 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: date column value
Replies: 19
Views: 4320

Here's a true PX solution: * * * Presuming that input field is entitled "DSLink3.currentDate". Have 3 stage variables in a Transform stage: A - JulianDayFromDate(DSLink3.currentDate) B - WeekdayFromDate(DSLink3.currentDate) C - If (b >= 5) Then (a - (b - 5)) Else (a - ((b + 7) - 5)) In the...
by T42
Tue Mar 01, 2005 2:26 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: calling a url in datastage job
Replies: 7
Views: 3457

That is a question only Ascential can properly give you. They have specific rates for specific regions that your sales group will need to negotiate on.

Contact your Ascential provider for further details.
by T42
Tue Feb 22, 2005 5:30 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Oracle 10G with DataStage 6.0r3
Replies: 4
Views: 2205

You may try it, but you will NOT be supported by Ascential.

Good luck.