Search found 53125 matches

by ray.wurlod
Mon Dec 20, 2004 7:27 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Can't find path for shared library
Replies: 6
Views: 2680

Depending on what kind of UNIX you're on this environment variable may be called LD_LIBRARY_PATH, SHLIB_PATH, LIBPATH, or even something else! :roll:
But it WILL be there.
by ray.wurlod
Mon Dec 20, 2004 7:24 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Calling a stored procedure from a Routine
Replies: 2
Views: 1064

Welcome aboard! :D This really should have been posted on the DataStage forum. The FAQ discussion forum is for talking about what is, and should be, on the FAQ forum. How you call a stored procedure from DataStage BASIC depends on a number of things, primarily whether your ODBC driver is licensed fo...
by ray.wurlod
Mon Dec 20, 2004 3:20 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: how to convert julian date & time to TIMESTAMP
Replies: 4
Views: 3407

The DRS stage seems to be a bit more forgiving than the OCI stage on this one. :D
by ray.wurlod
Mon Dec 20, 2004 3:18 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Read XML from Oracle BLOB
Replies: 8
Views: 3541

In the Derivations part of the Columns grid, change the derivation to

Code: Select all

CAST(tablename.colname AS VARCHAR2(4000))
and change the SQL type to VarChar and precision to 4000.

DataStage can not handle BLOB, CLOB, and so on, and isn't all that happy with Long VarChar either.
by ray.wurlod
Mon Dec 20, 2004 3:14 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: data type overflow
Replies: 4
Views: 2324

SQL Server uses milliseconds in the timestamp format.

There are 23 characters in "YYYY-MM-DD HH:MM:SS.sss".

That's where the (23,3) comes from.
by ray.wurlod
Mon Dec 20, 2004 3:11 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Last Change Date and User
Replies: 11
Views: 4126

No. If the user is still logged in, you can determine the identity of that user's client machine (from LIST_READU, or from a query on DS_LICENSE such as SELECT @ID FMT '56L' FROM DS_LICENSE; ), but which of those common users is the culprit? It's not recorded in DS_AUDIT. Best practice is NOT to use...
by ray.wurlod
Mon Dec 20, 2004 3:07 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Can a DataStage Basic Program be run from Unix shell
Replies: 10
Views: 3593

DataStage BASIC source can not be compiled if stored in a hashed file. You need to create a directory (in UniVerse-speak a "Type 19 file") for source code. Certainly you can store it in a hashed file, but you have to copy it into the Type 19 file to compile it. That's what the DSU_BP file is used fo...
by ray.wurlod
Mon Dec 20, 2004 3:03 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Not able to import a routine
Replies: 6
Views: 1358

There are a few possibilities here. You have an empty export file. The routine already exists in the protected project. The routine in the export file is the same as the one in the project. Try doing a partial import, so that you get a list of what's in the export file without actually importing any...
by ray.wurlod
Mon Dec 20, 2004 3:00 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: SubRoutine returning error values
Replies: 4
Views: 1441

The DSGetJobInfo function is not Cataloged with a "DSJ." prefix. In fact, it is Cataloged with no prefix. So either lose the CALLING clause from your DEFFUN statement, or lose the "DSJ." prefix from within the CALLING clause. DEFFUN DSGetJobInfo(JobName,Info) DEFFUN DSGetJobInfo(JobName,...
by ray.wurlod
Mon Dec 20, 2004 2:57 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: calling user defined subroutines
Replies: 6
Views: 2182

Or maybe you're asking about transform functions, another kind of Routine. A user-written transform function is, as noted earlier, implemented as a subroutine with the first argument position reserved for the result. Therefore, the indirect CALL mechanism and the DEFFUN mechanism and the SUBR() func...
by ray.wurlod
Mon Dec 20, 2004 2:54 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: calling user defined subroutines
Replies: 6
Views: 2182

See answer A3 above.

There is only one kind of Transform. It is a simple (one-line) expression, stored as a re-usable component in the Repository.
by ray.wurlod
Mon Dec 20, 2004 2:52 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to pass values between jobs?
Replies: 12
Views: 4464

In a Command activity you could use cat or head to read the file. In a Routine activity you could invoke a routine (which you'd have to write) that uses OpenSeq , ReadSeq and CloseSeq statements to read the file. You could customise the Job Sequence code by turning it into Job Control code in a serv...
by ray.wurlod
Mon Dec 20, 2004 2:49 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to get job name for list of hash file
Replies: 22
Views: 6642

Default line length is 80. You have specified 25, 25 and 45, which total to 95. If you're working interactively, you can use the TERM or SET.TERM.TYPE command to change the line length. (Your telnet terminal emulator may yet wrap lines; you need to change its properties in this case.) TERM 132 SET.T...
by ray.wurlod
Mon Dec 20, 2004 2:39 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: how to convert julian date & time to TIMESTAMP
Replies: 4
Views: 3407

Welcome aboard! :D This really should have been in the DataStage forum, not in the FAQ Discussion forum. DataStage has many powerful date and time manipulation possibilities, so anything to anything is possible. However you must be aware of the format you have and the format you need. For example, i...
by ray.wurlod
Mon Dec 20, 2004 3:43 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to pass values between jobs?
Replies: 12
Views: 4464

Your basic choice is how you want to write it to disk; in a text file or in a hashed file or in a database table. Separate jobs run in separate processes. In theory you could have the one job write to a named pipe and a second job (job sequence) reading from that named pipe, but the coordination of ...