Skip to Main Content

Berkeley DB Family

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

what is the current Berkeley DB Java Edition release? is there a public source code repository

AndyGlickDec 7 2020

The standard Oracle Berkeley DB Java Edition pages give the current release as 7.5.11. And offers what looks to be a complete distribution. This is some years old.
Maven Central offers a release of com.sleepycat:je:18.3.12 which offers a more recent jar file, a source jarfile and a javadoc jar file, but it does not offer the test code or the examples.
Given that this product is released under an open source license it would be helpful if a source code repository were available. In addition, as I understand it the Affero GPL requires that OSS licensees are obligated to contribute their enhancements and bug fixes back to Oracle, but there seems to be no way of doing that, or rather there does not appear to be anyway that OSS participants can share changes among each other.
If the current release is in fact 18.3.12 then would you please create such a full release? If you aren't going to treat it as a release, then what are your intentions going forward for this product and how can OSS licensees see each other's work?

Comments

21205
show it
SQL> declare
  2     ignore_column_name varchar2 (100) := 'ENAME';
  3     CURSOR C1 is
  4     SELECT column_name
  5     FROM user_tab_columns
  6     WHERE table_name = 'EMP'
  7  --   AND column_name NOT IN (ignore_column_name)
  8  ;
  9  begin
 10     for r in c1
 11     loop
 12        dbms_output.put_line ('result: '||r.column_name);
 13     end loop;
 14  end;
 15  /
result: EMPNO
result: ENAME
result: JOB
result: MGR
result: HIREDATE
result: SAL
result: COMM
result: DEPTNO

PL/SQL procedure successfully completed.

SQL> 
SQL> 
SQL> declare
  2     ignore_column_name varchar2 (100) := 'ENAME';
  3     CURSOR C1 is
  4     SELECT column_name
  5     FROM user_tab_columns
  6     WHERE table_name = 'EMP'
  7     AND column_name NOT IN (ignore_column_name)
  8  ;
  9  begin
 10     for r in c1
 11     loop
 12        dbms_output.put_line ('result: '||r.column_name);
 13     end loop;
 14  end;
 15  /
result: EMPNO
result: JOB
result: MGR
result: HIREDATE
result: SAL
result: COMM
result: DEPTNO

PL/SQL procedure successfully completed.

SQL> 
646894
Hello,

the only difference between your program and mine is that..

ignore_column_name variable is declared but not initialized before the cursor. I mean..

declare
ignore_columns_name varchar2(100);
cursor c1 is
select column_name
from user_tab_columns
where column_name not in (ignore_column_name);
begin
ignore_column_name := 'LNAME';
for r in c1
loop
column := column||','||r.column_name;
end loop;
##REST OF THE CODE##
end;
/

the thing is .. ignore_column_name is not hardcoded, instead it is generated from a different part of the code. so I cannot initialise it like u have shown it. Can u help me? is it possible to achieve this?

Thanks
BluShadow
Azeroth wrote:
Hello,

the only difference between your program and mine is that..

ignore_column_name variable is declared but not initialized before the cursor. I mean..

the thing is .. ignore_column_name is not hardcoded, instead it is generated from a different part of the code. so I cannot initialise it like u have shown it. Can u help me? is it possible to achieve this?

Thanks
It doesn't make a difference if it's initialized before the cursor declaration or not, just as long as it's initialized before the cursor is opened.
SQL> declare
  2    ignore_column_name varchar2 (100);
  3    CURSOR C1 is
  4    SELECT column_name
  5    FROM user_tab_columns
  6    WHERE table_name = 'EMP'
  7    AND column_name NOT IN (ignore_column_name);
  8  begin
  9    ignore_column_name := 'ENAME';
 10    for r in c1
 11    loop
 12       dbms_output.put_line ('result: '||r.column_name);
 13    end loop;
 14  end;
 15  /
result: EMPNO
result: JOB
result: MGR
result: HIREDATE
result: SAL
result: COMM
result: DEPTNO

PL/SQL procedure successfully completed.

SQL>
646894
Hello,

This was a really helpful reply.. it cleared my only suspicion but not my problem. Here I am pasting the entire code.. I will highlight the part thats not working.. Please let me know what you think the problem is.

CREATE OR REPLACE PROCEDURE
test_etl1
(
table1 IN user_tab_columns.table_name%TYPE,
table2 IN user_tab_columns.table_name%TYPE
) IS
col_list VARCHAR2(3000);
new_col_list VARCHAR2(3000);
column_name VARCHAR2(100);
num number;
table3 VARCHAR2(100);
column_list VARCHAR2(3000);
stmt VARCHAR2(3000);
CURSOR C1 IS
SELECT column_list
FROM ignore_columns
WHERE table_name = table1;
CURSOR C2 IS
SELECT column_name
FROM user_tab_columns
WHERE table_name = table3
AND column_name NOT IN (new_col_list);
BEGIN
col_list := null;
FOR R1 in C1
LOOP
col_list := R1.column_list;
END LOOP;
num := length(col_list);

WHILE num <> 1
LOOP
column_name := SUBSTR(col_list, 0,INSTR(col_list,' '));

IF (NVL(SUBSTR(col_list, INSTR(col_list, ' '), LENGTH(col_list)),'1') != '1')
THEN
col_list := SUBSTR(col_list, INSTR(col_list, ' '), LENGTH(col_list));
col_list := LTRIM(col_list);
new_col_list := new_col_list||''''||RTRIM(column_name)||''''||',';
ELSE
num := 1;
END IF;
END LOOP;

new_col_list := RTRIM(new_col_list,',');

table3 := upper(table1);
FOR R2 IN C2
LOOP
* column_list := column_list||''''||R2.column_name||''''||',';*
END LOOP;
column_list := RTRIM(column_list,',');
stmt := 'CREATE TABLE error_table AS SELECT '||column_list||' FROM '||table1||' MINUS SELECT '||column_list||'FROM '||table2;

END;

Table ignore_columns contains a single row.
table_name = patient_v
column_list = 'LAST_NAME'; with the quotes

Thanks
1 - 4

Post Details

Added on Dec 7 2020
11 comments
1,177 views