TRIM - DataStage 4.2

[admin]

Now im working with DataStage 4.2.1r2 (64 bits) on a server HP-UX 11.00 PA 2.0 Risc. Many jobs, that we have are export from a version of Datastage 32 bits.

I have a job, where i do a TRIM of a char field, and the return, is an unexplain group of characters ( 1d!@07@ ). But it only occours in one field,but not in all records.


I can understand what happens.
Has anyone faced this situation? Any help is appreciated.

Regards
Rui Soares


Rui Soares
Tlm : +351 . 96 347 0840

[admin]

Theres not enough information in the error messages. You need to use the Debugger or a stage trace on the stage where the Trim() function is failing, and capture the value before and after. Then send the trace file (just the section for a row that fails will suffice) to support, who should be able to analyse it for you. Or you may like to examine the trace file yourself using an method that can display non-printing characters (for example the od -x command or the UniVerse editor in up-arrow mode). Trim() is an innocuous function; you might like to experiment with some of its options to be more specific about which characters you want to trim. If youre using NLS, you might also like to check the TRIMMABLE characters in your CTYPE locale category.

-----Original Message-----
From: Rui Miguel de Sousa Soares (Novabase - IGestao [mailto:rumsoares@bes.pt]
Sent: Monday, 02 October 2000 09:44
To: Datastage-Users (E-mail)
Subject: TRIM - DataStage 4.2



Now im working with DataStage 4.2.1r2 (64 bits) on a server HP-UX 11.00 PA 2.0 Risc. Many jobs, that we have are export from a version of Datastage 32 bits.

I have a job, where i do a TRIM of a char field, and the return, is an unexplain group of characters ( 1d!@07@ ). But it only occours in one field,but not in all records.


I can understand what happens.
Has anyone faced this situation? Any help is appreciated.

Regards
Rui Soares


Rui Soares
Tlm : +351 . 96 347 0840