Below is my code snippet from a Oracle stored procedure which inserts into a table "A" from table "B", I have created a nested block (as I would have more insert statements later, so wanted to maintain some modularity). I have exception block with the generic handler "others" for catching the errors, and within that I have an autonomous procedure log_errors_SP which I am using to log errors. I have used raise_application_error within the exception block to notify the invoker of this procedure(.NET code) about exceptions else update the log table with completion status (the last update statement in the code). I need to understand:-
- if this the correct way to program as for some reason when the procedure is executed from .NET code the log_errors_sp is not executed.
- log_errors_sp is executed fine when I run my parent stored procedure manually from Oracle developer (so no issues with log_errors_sp).
- Is it because the error is being trapped on .NET side by it's try catch block rather than the stored procedure's exception block ?.
- if point 2 is valid how can I change the stored procedure to have the log_errors_SP invoked ?.
create or replace procedure some_sp (P_JSON CLOB) is
BEGIN
do the basic IN parameter check here..
--nested block to insert data
BEGIN
insert into A
(
CREATED_DT,
CREATEDBY,
LASTUPDATED_DT,
LASTUPDATEDBY
)
select sysdate,'SMICLOAD',sysdate,' '
from b where <conditions>;
EXCEPTION
WHEN OTHERS THEN
ov_err_code := 'Table Fail: '||SQLERRM;
log_errors_SP (p_id,ov_err_code);
raise_application_error(-20101, 'A:Unable to Insert into the table');
END;
UPDATE p_prcs_log
set
COMPLETION_TS=sysdate,
STATUS='C',
ERR_MSG='NONE'
where prcs_id=p_id;
END
question from:
https://stackoverflow.com/questions/65835917/oracle-exception-handling-for-net-application 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…