Encoding for German Character set

[Felix-Braun]

Hello,

I noticed that the DB2Developerextension doesn't support custom encoding and no JDBC-Parameter exist for encoding.

Problem

Our DB2 subsystems zparm are set up with CCSID 037 (USA) and Tables are usually set up with EBCDIC encoding.
Our 3270 Emulators are set up with CCSID 1141 (German Euro). So when we query a Table we get the correct conversion of Character.
e.g.

Select 'ä'
,SUBSTR(CMCACATEG,17,1)
,HEX(SUBSTR(CMCACATEG,17,1))
from
DB2B.CMTTCATG
where CMCAID = 5
with UR;

db2developerextension:

mainframe:

Normal Insert using JDBC
SPECIAL CHARACTER CONVERSION FROM JDBC TO MAINFRAME
**Encoded using encode("cp1141").decode("cp037")**
value from select JDBC: äöüÄÖÜ}{[]§&ß
value from select z/OS: {¦}[\]üä¬|@&~
decoded value         : {¦}¢\!üäÄÜ@&~
expected Output       : {¦}¢\!üäÄÜ@&~

Normal Insert on Mainframe
SPECIAL CHARACTER CONVERSION FROM MAINFRAME TO JDBC
**Encoded using encode("cp037").decode("cp1141")**
value from select JDBC: {¦}¢\!üäÄÜ@&~
value from select z/OS: äöüÄÖÜ}{[]§&ß
decoded value         : äöüÄÖÜ}{[]§&ß
expected Output       : äöüÄÖÜ}{[]§&ß


**encoded_value = 'äöüÄÖÜ}{[]§&ß'.encode("cp1141").decode("cp037")
sql = f"""INSERT INTO DB2B.TESTTEST (TSTTVALUE, TSTTCOMMENT)
VALUES ('{encoded_value}', 'SPECIAL CHARACTER CONVERSION FROM JDBC TO MAINFRAME WITH ENCODING CP1141  AND CP037')
"""**
SPECIAL CHARACTER CONVERSION FROM JDBC TO MAINFRAME WITH ENCODING CP1141  AND CP037
value from select JDBC: {¦}¢\!üäÄÜ@&~
value from select z/OS: äöüÄÖÜ}{[]§&ß
decoded value         : äöüÄÖÜ}{[]§&ß
expected Output       : äöüÄÖÜ}{[]§&ß

Solution

If possible, add a "db2forzosdeveloperextension.encoding" setting for encoding using the extensions logic.
It could encode the values to "cp037" (EBCDIC) and decode them using "cp1141" (IBM-1141)
or viseversa for insert using JDBC to mainframe
in java

byte[] ebcdicBytes = ebcdicString.getBytes(Charset.forName("Cp037"));
return new String(ebcdicBytes, Charset.forName("Cp1141"));

in python

        decoded_text = records.encode("cp037").decode("cp1141")
# or
        decoded_text = tst_value.encode("cp1141").decode("cp037")

[VijayaKatikireddy]

Hi Felix, we just published "IBM Db2 Developer Extension" yesterday, which caters Db2 Linux, Unix and Windows Users. For this extension, this issue is unrelated.

From reading above, this issue seems to be related to "IBM Db2 for z/OS Developer Extension" in which case, you need to open the issue here:
https://github.com/IBM/db2forzosdeveloperextension-about/issues

I will also make the Z tools team aware this issue. Thanks.

[wolfgang0101]

Hi Felix, we just published "IBM Db2 Developer Extension" yesterday, which caters Db2 Linux, Unix and Windows Users. For this extension, this issue is unrelated.

From reading above, this issue seems to be related to "IBM Db2 for z/OS Developer Extension" in which case, you need to open the issue here: https://github.com/IBM/db2forzosdeveloperextension-about/issues

I will also make the Z tools team aware this issue. Thanks.

Hi,

we have this issue too. Since I wasn't able to find it in the link provided, @VijayaKatikireddy do we have to open the issue by ourselfs? Did you already pass the information to the Z tools team?

[VijayaKatikireddy]

I prefer to transfer the issue, to the Z. so that it is properly prioritized and addressed.
@IBM/db2forzosdevextension

[VijayaKatikireddy]

Transferred to the right repository, it will be reviewed by Db2 z/OS team.

[manjula-1312]

Thanks @VijayaKatikireddy @Felix-Braun for bringing this up. We are looking into the issue.