EBCDIC ! and ] not translated properly

horserider · Post by **horserider** » Mon Apr 14, 2008 8:49 am

STEP (1) Downloading the Mainframe File as Binary.
STEP (2) Reading downloaded file through CFF Stage.

This is working perfectly fine for all chanracters except ! and ] characters in mainframe. When I compare the data between mainframe data and final processed data in UNIX

Character ! in mainframe is converted to ] in ASCII (UNIX Env)
Character ] in mainframe is converted to ' ' (blank) in ASCII (UNIX Env)

I know that we can use some forced conversion like reading the SEQ() for ! and ] when reading data in Step 2 and then changing back to ! and ] in ASCII, but question is

"When all other characters are converted proplerly using the same 2 steps, wny not EBCDIC ! and ] characters? Is there any documentation for this issue?

ray.wurlod · Post by **ray.wurlod** » Mon Apr 14, 2008 4:16 pm

Look in the DataStage BASIC manual for an Appendix containing EBCDIC and ACCII codes. As to why they're not "translating" properly that will depend largely on how you are getting them out of the mainframe. Examine the received file with a hex editor and see what's actually arrived on the UNIX machine.

horserider · Post by **horserider** » Tue Apr 15, 2008 9:22 am

(1) Character ] on mainframe (with a HEX VALUE BB ) is being converted to ' ' (blank) (SEQ of 1A) in Data Stage Server.

(2) Character ! on mainframe (with a HEX VALUE 5A ) is being converted to ] (SEQ of 5D) in Data Stage Server.

bcarlson · Post by **bcarlson** » Tue Apr 15, 2008 10:31 am

I am not entirely sure of this and am dong some checking on my own. However, as I recall there are multiple code pages that represent EBCDIC aand ASCII and the conversions between then, and there are subtle variations in them.

I was given a printout that referenced this website:

Conversion character differences between CCSID 037 and CCSID 500

I'll see if I can find more info about this from my past email chain.

Brad.