|
View previous topic :: View next topic
|
| Author |
Message |
hsinaz50
New User
Joined: 09 Apr 2014
Posts: 5
Location: United States
|
|
|
|
I have a name/address text field in DB2 that has characters such as À ñ Ö ë etc. When I retrieve them with SPUFI/other query software they look exactly like that. When I debug using IBM Problem Determination tools, they look correct (same as above), when I look at their movement through various in-memory locations, again, they look the same. e.g. ÉXPAÑSION
1) When I write them to disk/a flat file on the mainframe through a COBOL program, they change to completely different characters such as ^s or blanks or underlined capital Zs or Us. e..g the above changes to ^XPAZSION
2) Then if I browse at the disk/flat file using IBM file manager, they look yet again different. Most of the ^ or blanks or underlined letter in #1 above look to be dots (period signs). e.g. the above yet again looks like .XPA.SION
All throughout, the receiving fields are defined as CHAR.
Q1. Is there a way I can get COBOL to output the same characters that are in my source (DB2)?
Q2. If Q1 can be fixed, how can I send the exact same characters via NDM to a windows server (i.e. is there a NDM/connect direct option to be specified)
Thank you. |
|
| Back to top |
|
 |
Robert Sample
Global Moderator
Joined: 06 Jun 2008
Posts: 8696
Location: Dubuque, Iowa, USA
|
|
|
|
| You need to research ccsid. Probably your DB2 ccsid isn't the COBOL default (which can be changed with a compiler option. |
|
| Back to top |
|
|
hsinaz50
New User
Joined: 09 Apr 2014
Posts: 5
Location: United States
|
|
|
|
| Robert Sample wrote: |
| You need to research ccsid. Probably your DB2 ccsid isn't the COBOL default (which can be changed with a compiler option. |
Researched this today. DB2 database level, and even at source column level CCSID is 37. COBOL compiler is CODEPAGE(1140). Per IBM they both are equivalent  |
|
| Back to top |
|
|
hsinaz50
New User
Joined: 09 Apr 2014
Posts: 5
Location: United States
|
|
|
|
| hsinaz50 wrote: |
| Robert Sample wrote: |
| You need to research ccsid. Probably your DB2 ccsid isn't the COBOL default (which can be changed with a compiler option. |
Researched this today. DB2 database level, and even at source column level CCSID is 37. COBOL compiler is CODEPAGE(1140). Per IBM they both are equivalent |
Any further ideas on this please. I checked more today and the DB2 hex values of these characters and the on-disk hex values when written match, but the on-disk character look different in ISPF edit that DB2 as I mentioned above. |
|
| Back to top |
|
|
|
|
