Search found 53125 matches

by ray.wurlod
Thu Sep 27, 2007 4:39 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: FUNCTION: @INROWNUM
Replies: 17
Views: 5382

The row counts can be had using DSGetStageInfo() or DSGetLinkInfo() functions. You can get total row count (across all nodes) or per-node row counts. It's all described in the help. You ought not to need to generate the row count within the job. However, if you use @INROWNUM, you will get per-node r...
by ray.wurlod
Thu Sep 27, 2007 4:35 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: How do placeholders work with lookup stage?
Replies: 3
Views: 1613

Only if you mark cusrut_comp_customer_number and cusrut_comp_customer_number, and no other column, as Key.
by ray.wurlod
Thu Sep 27, 2007 4:33 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Job design for n-way outer join??
Replies: 7
Views: 2319

The Join stage does support more than two inputs. However, internally it still produces pairwise joins (as would a database server) joining the next-rightmost input to the intermediate result of joining the previous pair. I think your design makes it clearer what's going on, and will probably be eas...
by ray.wurlod
Thu Sep 27, 2007 4:30 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Difference between ODBC and Plug-In
Replies: 2
Views: 1184

By far the most accurate metadata import is through the Orchestrate orchdbutil facility.
by ray.wurlod
Thu Sep 27, 2007 4:28 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: 'SIGINT' while reading a big file in PX
Replies: 1
Views: 755

If anyone has faced this problem earlier, they probably posted on the forum. A Search would let you know without needing to await anyone's answer to this thread.
by ray.wurlod
Thu Sep 27, 2007 4:26 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Automizing parameter values with a concurrent architecture
Replies: 2
Views: 736

You should create your job sequence such that operations can NOT occur in the wrong order. You could also use any of the public semaphores (task synchronization locks) within your job design. These are described in the DataStage BASIC manual. Neither your question nor this answer appear to have any ...
by ray.wurlod
Thu Sep 27, 2007 4:20 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Using Windows System Variables inside Before/After routine
Replies: 2
Views: 872

The non-viewable character is the line terminator, converted to a field mark.

Code: Select all

Call DSExecute('DOS', "echo %WIN_VARIABLE%", return_value, return_code) 
Result = Convert(@FM, "", return_value)
by ray.wurlod
Thu Sep 27, 2007 4:16 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: COBOL FILE DEFINITION
Replies: 1
Views: 725

Not something I've seen. You could perhaps work around it by changing the FD. 10 XF-PRIM-REG-ID-FLD. 15 XF-PRIM-REG-ID-SIGN PIC X. 15 XF_PRIM-REG-ID PIC 9(02)V9(06). 10 XF-PRIM-REM-ID-FLD. 15 XF-PRIM-REM-ID-SIGN PIC X. 15 XF-PRIM-REM-ID PIC 9(03)V9(06).
by ray.wurlod
Thu Sep 27, 2007 4:12 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: KeyMgtGetNextValue & KeyMgtGetNextValueConcurrent
Replies: 9
Views: 2959

See, one CAN hijack one's own thread!
by ray.wurlod
Thu Sep 27, 2007 4:10 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Dealing with Occurs in .cfd files using CFF stage
Replies: 14
Views: 3913

Just have the CFF stage "normalize" the repeating group. The CFF stage itself can generate the multiple rows. Then run the single output through a Transformer stage to filter out those occurrences that have the "empty" value - did you say it was all zeroes?
by ray.wurlod
Thu Sep 27, 2007 4:06 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Complex Flat Files adding extra row
Replies: 4
Views: 1872

Check whether there's an extra line terminator at the end of the file. If so, pre-process the file (for example with sed) to remove it. Examples have been posted in the past, for which you can search.
by ray.wurlod
Thu Sep 27, 2007 4:03 pm
Forum: IBM QualityStage
Topic: Implementing Scenario in DS with QS
Replies: 4
Views: 2523

Can you be more specific about what you mean by "an address is valid or not"? For example, do you need to go to the full extent of checking against a postal certification database (which is obtainable for some countries, including the USA, with QualityStage)? Or are you only seeking to verify that t...
by ray.wurlod
Thu Sep 27, 2007 3:28 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: How to write the job log to a text file
Replies: 2
Views: 1047

The easiest way to get the job log into a text file is to choose Print from the Director's Project menu, and check the Print to File check box.
by ray.wurlod
Thu Sep 27, 2007 3:25 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Datetime field overflow
Replies: 3
Views: 1641

What is the highest legal date in the database? Have you supplied a date later than this?
by ray.wurlod
Thu Sep 27, 2007 1:56 am
Forum: Site/Forum
Topic: Timeout when Posting
Replies: 2
Views: 2270

It means I'm now having to spend an unconscionable amount of time on DSXchange.