Hi,
While loading data from Sybase to Oracle/Sql last byte is truncating for the fields having Korean characters what is the problem how to resolve this problem.
Help to resolve this problem...
While loading data from Sybase to Oracle/Sql last byte is tr
Moderators: chulett, rschirm, roy
-
rverharen
- Premium Member
- Posts: 34
- Joined: Fri Jan 19, 2007 9:09 am
- Location: Breda, The Netherlands
Oracle type Varchar2(CHAR)
make sure the oracle column type = varchar2 (char) instead of varchar2 (byte)
import the table definition again and see that the length in datastage is 4 times the lenght of the column in oracle (use this new length in your stages)
import the table definition again and see that the length in datastage is 4 times the lenght of the column in oracle (use this new length in your stages)
i have 190 jobs for all i have to modify or any other way ....
i heared that one korean characters will stored in two bytes, i our case the call centre people they are copy pasting the descrtiopns in the front ent applications at that time some korean characters are storing half byte only i want to remove that particular broken character how to remove the broken character.
Help me
i heared that one korean characters will stored in two bytes, i our case the call centre people they are copy pasting the descrtiopns in the front ent applications at that time some korean characters are storing half byte only i want to remove that particular broken character how to remove the broken character.
Help me
Vijay
-
rverharen
- Premium Member
- Posts: 34
- Joined: Fri Jan 19, 2007 9:09 am
- Location: Breda, The Netherlands
use the basic function 'Change' in your derivation in the transformer if you want to remove a specific charactervijaydev wrote:i have 190 jobs for all i have to modify or any other way ....
i heared that one korean characters will stored in two bytes, i our case the call centre people they are copy pasting the descrtiopns in the front ent applications at that time some korean characters are storing half byte only i want to remove that particular broken character how to remove the broken character.
Help me
do question yourself if that is what you want because next time you get another character and you will be getting the same problem
i know that , becuase i dont know the korean language to under stand also that way i removed ? using ereplace function its given good results for some rows for some rows the last byte is truncated. i am bit confused with this how do and its very urgent for me alsorverharen wrote:use the basic function 'Change' in your derivation in the transformer if you want to remove a specific charactervijaydev wrote:i have 190 jobs for all i have to modify or any other way ....
i heared that one korean characters will stored in two bytes, i our case the call centre people they are copy pasting the descrtiopns in the front ent applications at that time some korean characters are storing half byte only i want to remove that particular broken character how to remove the broken character.
Help me
do question yourself if that is what you want because next time you get another character and you will be getting the same problem
Vijay
Based on your description so far, it appears that your data is "truncated"/"lost" when it is stored in the target "Oracle" database, correct?
The front-end apps are storing the data into the Sybase database, correct?
Do you have any viewing application to check this at the target end?
Did you try to "peek" (debug) at the different stages to check where the bytes are getting lost?
It looks like you are definitely getting Unicode data feeds. Is the target server NLS enabled?
I haven't worked on Unicode data handling in Datastage, but will be doing soon. But based on past experience in Unicode data on Sybase, NVarchar is the type that will handle multi-byte characters without you having to allocate the extra bytes.
My thoughts...
The front-end apps are storing the data into the Sybase database, correct?
Do you have any viewing application to check this at the target end?
Did you try to "peek" (debug) at the different stages to check where the bytes are getting lost?
It looks like you are definitely getting Unicode data feeds. Is the target server NLS enabled?
I haven't worked on Unicode data handling in Datastage, but will be doing soon. But based on past experience in Unicode data on Sybase, NVarchar is the type that will handle multi-byte characters without you having to allocate the extra bytes.
My thoughts...
-V