Search found 53125 matches
- Tue Feb 17, 2004 3:04 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Creation of file with a name taken from a sequential file
- Replies: 3
- Views: 904
To paraphrase what Ken posted, your job control routine reads the sequential file, constructs the file name based on what it has read, and passes this to the job as a parameter value using DSSetParam before requesting execution of the job. Within the job, the parameter reference is used to provide t...
- Tue Feb 17, 2004 3:00 pm
- Forum:
- Topic: What is metastage /QualityStage
- Replies: 2
- Views: 4745
MetaStage has been around for years. It's a mechanism for storing metadata in a generic fashion so that various applications (such as CASE tools, ETL tools, BI tools) can get at it. They do this through specialized filters called Metabrokers. It's also possible for MetaStage automatically to capture...
- Tue Feb 17, 2004 12:04 am
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Lookup behavior in PX
- Replies: 7
- Views: 1815
Both parallel and server jobs are behaving as documented. Sure they're mutually inconsistent but, then again, they were developed separately. Changing one (which one?) could negatively impact existing jobs that rely on the particular functionality. Since there's no such thing as a hybrid Transformer...
- Mon Feb 16, 2004 8:03 pm
- Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
- Topic: How to get current timestamp in the px
- Replies: 4
- Views: 3358
- Mon Feb 16, 2004 8:00 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Poor extraction performance
- Replies: 11
- Views: 3530
Set the constraint on your bulk load stage to "1<>1" which ensures no rows go down this path but the control file gets created when the job starts. Save some machine cycles! :D Don't evaluate 1<>1. Just use 0 (or the system variable @FALSE, which the compiler replaces with 0) as the constraint expr...
- Mon Feb 16, 2004 2:55 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Call Transforms in a controlling job
- Replies: 3
- Views: 1199
Not every Transform invokes a Routine. Certainly you can call Routines from other Routines (probably needing a DEFFUN declaration). However, if the Transform is simply an encapsulated expression, the answer is definitely no. And using the Transform's defining expression in your code means that there...
- Mon Feb 16, 2004 2:37 am
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Record schema of DataSets
- Replies: 8
- Views: 2755
- Mon Feb 16, 2004 2:35 am
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Call Transforms in a controlling job
- Replies: 3
- Views: 1199
It is not possible to call a Transform from any kind of routine, including a job control routine. If you use a Transform in a server job (Transformer stage) and examine the code generated by compiling that job, you will see that the Transform is replaced by its defining expression. In effect, at com...
- Sun Feb 15, 2004 3:11 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: DS 7.0 install issue
- Replies: 6
- Views: 1673
If they are UniVerse services they are: UniVerse Resource Service UniVerse Telnet Service UniVerse RPC Service (perhaps called UniRPC Service) UniVerse REXEC Service If they are DataStage UniVerse services they may have the above names or they may have DataStage instead of UniVerse in the name, and ...
- Sat Feb 14, 2004 4:48 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Hash Files & Slowly changing dimension
- Replies: 7
- Views: 6849
I've done these too. Even though not part of an SCD solution I've used an alternate technique for a larger number (when LOCATE started to degrade) implementing a "bit map" technique - actually a character string of "1" and "0" values and substring. It was to determine whether a particular STD (telep...
- Sat Feb 14, 2004 4:39 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Record schema of DataSets
- Replies: 8
- Views: 2755
- Sat Feb 14, 2004 4:34 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Partition while using lookup
- Replies: 5
- Views: 1642
Data will still be partitioned. However, you have no way of knowing in advance in which partition your particular key will occur, so you have to use the Entire partitioning method so that the lookup can "see" the entire data set. PX will automatically look after ensuring that the retrieved row ends ...
- Sat Feb 14, 2004 4:29 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Zero Fill
- Replies: 5
- Views: 2561
Technically it can be done. Legally it's a bit of a grey area; location of the column definitions within the metadata is not publicly documented. As you may have gleaned, the objects in a job are stored in DS_JOBOBJECTS as far as their design time properties are concerned, but it may be a more accur...
- Sat Feb 14, 2004 4:17 pm
- Forum: Site/Forum
- Topic: Appropriate forum for Version Control?
- Replies: 3
- Views: 3090
Since VC was written by and, by arrangement with Ascential, is maintained by, the good folks who bring you this forum, there's a good chance that ideas you post here will get back to those who need to know. On the other hand, requests for enhancements should really go through "the proper channels". ...
- Thu Feb 12, 2004 8:00 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Read Hash File in Routine
- Replies: 3
- Views: 1050
Re: Read Hash File in Routine
I was trying to look for a document about how to read hash file using basic language, but could not get any. Does anyone know whether there is a reference book available? The manual you need is the DataStage BASIC manual, which is part of the documentation set. OPEN, READ, WRITE, CLOSE and their va...