Converting Mainframe Files with EBCDIC into ASCII
Posted: Tue Nov 11, 2008 2:48 pm
We have DataStage 7.5.2 in production and want to upgrade to version 8.0.
Here is how we handle mainframe flat files in 7.5.2 using 2 jobs where we have to convert any EBCDIC characters into ASCII equivalent.
(1) Job 1 : FTP plugin > TRN>Sequential File (FTP type is binary)
(2) Job 2 : CFF Stage > Any Target
Example:
Job 1 :
If we have a mainframe flat file with 6 columns totalling 130 characeters we simply download the complete row in binary format. We don't define each column in FTP stage in STEP 1. That we way we download the mainframe file ASIS from mainframe to ETL Server in binary format.
Job 2 :
Then in second job we use the sequential file as source and using CFF Stage we import the CFD and apply against the the sequential file that we had FTP'd in Job 1. Target can be anything, another file or table.This process work flawlessly in parallel job and we don't have to do any manual conversion column by column.
Question
(1)
The above is a 2 step process in 7.5.2. Has anyone been able to process mainframe flat files in ONE STEP only in VERSION 8.0?
(2)
Can we even process a mainframe file in just 1 step in VERSION 8.0 and successfully decode all EBCDIC characeters? If yes, please let me know what were the STEPS involved?
Here is how we handle mainframe flat files in 7.5.2 using 2 jobs where we have to convert any EBCDIC characters into ASCII equivalent.
(1) Job 1 : FTP plugin > TRN>Sequential File (FTP type is binary)
(2) Job 2 : CFF Stage > Any Target
Example:
Job 1 :
If we have a mainframe flat file with 6 columns totalling 130 characeters we simply download the complete row in binary format. We don't define each column in FTP stage in STEP 1. That we way we download the mainframe file ASIS from mainframe to ETL Server in binary format.
Job 2 :
Then in second job we use the sequential file as source and using CFF Stage we import the CFD and apply against the the sequential file that we had FTP'd in Job 1. Target can be anything, another file or table.This process work flawlessly in parallel job and we don't have to do any manual conversion column by column.
Question
(1)
The above is a 2 step process in 7.5.2. Has anyone been able to process mainframe flat files in ONE STEP only in VERSION 8.0?
(2)
Can we even process a mainframe file in just 1 step in VERSION 8.0 and successfully decode all EBCDIC characeters? If yes, please let me know what were the STEPS involved?