Search found 53125 matches

by ray.wurlod
Wed Nov 29, 2006 5:10 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Loading Huge Data
Replies: 11
Views: 2710

If you generate YEAR upstream in the job sequence (maybe as a sequence parameter) you can use a Routine activity to generate the four date ranges, perhaps pass them back as delimited strings, and unpack them. For example the routine could return a dynamic array containing four fields each with three...
by ray.wurlod
Wed Nov 29, 2006 4:59 pm
Forum: General
Topic: Upgrade from DS7.5.1A to DS 7.5.2
Replies: 5
Views: 2703

From memory that's PS1 values that contain a newline character. Is that so? What response did you get when you reported this previously?
by ray.wurlod
Wed Nov 29, 2006 2:30 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DB2/UDB Load Question
Replies: 10
Views: 5119

Get your DBA to monitor what DB2 is receiving.
by ray.wurlod
Wed Nov 29, 2006 2:26 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Row level restart
Replies: 4
Views: 2031

Welcome aboard. :D There is nothing "out of the box". There are several techniques you can use; all of them require some additional storage. Safest of all is to stage the results of your transformation in a load-ready area (Data Sets) so that your loads can be effected in the shortest possible perio...
by ray.wurlod
Wed Nov 29, 2006 2:21 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Age of a person
Replies: 10
Views: 4362

Age
Use a stage variable svCurrentDate initialized to current_date().
Use DaysSinceFromDate() to get age in years. Divide by 365.249 to get age in years, then apply the floor() function to get age in completed years.

Date Difference
Use subtraction of Julian dates, as posted earlier.
by ray.wurlod
Wed Nov 29, 2006 2:16 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: APT configuration file
Replies: 4
Views: 1437

SMP is everything in one machine - the fastname will be the same for all nodes in the configuration file. MPP is a cluster or grid - the fastname will have varying values in different nodes in the configuration file. For a grid, a dynamic configuration file is needed. That is outside the scope of th...
by ray.wurlod
Wed Nov 29, 2006 2:12 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: where do i find Basic transformer
Replies: 5
Views: 1412

Now that you know where to find it, don't use it.

If you do, your complaints about performance will fall on deaf ears.
by ray.wurlod
Wed Nov 29, 2006 2:11 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: I want to merge 2 inputfiles into one
Replies: 11
Views: 2457

The easiest way is to preprocess the files so that they each have two additional columns, the file indicator and the row number (both of these can be generated by the Sequential file stage using optional properties).
Sort on these.
You can drop those columns prior to your final write.
by ray.wurlod
Wed Nov 29, 2006 2:08 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Error while writing null value to date column in seq. file
Replies: 6
Views: 2984

You need to set the Null Field Value property (from the Edit Row dialog) even though you've handled nulls upstream. The value that you specify is what is written into the file if a null arrives on the input link. Alternatively, change the Nullable property to "No".
by ray.wurlod
Wed Nov 29, 2006 2:04 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Trim Function in Modify Stage
Replies: 21
Views: 12414

Sometimes it does - I've successfully combined string_trim and substring for example.
by ray.wurlod
Wed Nov 29, 2006 2:02 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Loading Huge Data
Replies: 11
Views: 2710

You have it exactly.

Use the job parameter references to replace constants in the WHERE clause of the extraction SQL. For example:

Code: Select all

WHERE sel_date >= '#StartDate#' AND sel_date < '#EndDate#'

You can do this on the Selection tab - you do not need to resort to user-defined SQL.
by ray.wurlod
Wed Nov 29, 2006 1:59 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Transformer satge output links
Replies: 3
Views: 977

If you want to implement something like the C break construct, you can use link variables from prior output links in constraint expressions. The REJECTED variable will be false if the row has been processed by any prior link. Or you can AND the current link's condition with the logical inverse of pr...
by ray.wurlod
Wed Nov 29, 2006 1:57 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: how to configure an odbc connection on unix
Replies: 19
Views: 7239

If it's any consolation, on older versions of Windows (we're talking 3.1 and earlier here) you had to edit an ODBC.INI file. That's why the Registry entry is still called ODBC.INI.
by ray.wurlod
Wed Nov 29, 2006 1:55 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Migration from Windows 2000 to Windows 2003
Replies: 4
Views: 872

I don't believe that Windows 2003 is a supported platform for DataStage version 5.1.
by ray.wurlod
Wed Nov 29, 2006 1:54 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DSP.Open Error writing to Oracle - again.
Replies: 15
Views: 5935

Welcome aboard, and thank you for making an extremely valuable contribution.

We sometimes assume that posters in the US don't have NLS installed, but that's not always the case. And posters elsewhere in the world often do have NLS installed, of course.