I can't figure out the correct syntax for the following pseudo-sql:
INSERT INTO some_table
(column1,
column2)
SELECT col1_value,
col2_value
FROM other_table
WHERE ...
RETURNING id
INTO local_var;
I would like to insert something with the values of a subquery.
After inserting I need the new generated id.
Heres what oracle doc says:
Insert Statement
Returning Into
OK i think it is not possible only with the values clause...
Is there an alternative?
You cannot use the RETURNING BULK COLLECT from an INSERT.
This methodology can work with updates and deletes howeveer:
create table test2(aa number)
/
insert into test2(aa)
select level
from dual
connect by level<100
/
set serveroutput on
declare
TYPE t_Numbers IS TABLE OF test2.aa%TYPE
INDEX BY BINARY_INTEGER;
v_Numbers t_Numbers;
v_count number;
begin
update test2
set aa = aa+1
returning aa bulk collect into v_Numbers;
for v_count in 1..v_Numbers.count loop
dbms_output.put_line('v_Numbers := ' || v_Numbers(v_count));
end loop;
end;
You can get it to work with a few extra steps (doing a FORALL INSERT utilizing TREAT)
as described in this article:
returning with insert..select
T
to utilize the example they create and apply it to test2 test table
CREATE or replace TYPE ot AS OBJECT
( aa number);
/
CREATE TYPE ntt AS TABLE OF ot;
/
set serveroutput on
DECLARE
nt_passed_in ntt;
nt_to_return ntt;
FUNCTION pretend_parameter RETURN ntt IS
nt ntt;
BEGIN
SELECT ot(level) BULK COLLECT INTO nt
FROM dual
CONNECT BY level <= 5;
RETURN nt;
END pretend_parameter;
BEGIN
nt_passed_in := pretend_parameter();
FORALL i IN 1 .. nt_passed_in.COUNT
INSERT INTO test2(aa)
VALUES
( TREAT(nt_passed_in(i) AS ot).aa
)
RETURNING ot(aa)
BULK COLLECT INTO nt_to_return;
FOR i IN 1 .. nt_to_return.COUNT LOOP
DBMS_OUTPUT.PUT_LINE(
'Sequence value = [' || TO_CHAR(nt_to_return(i).aa) || ']'
);
END LOOP;
END;
/
Unfortunately that's not possible. RETURNING is only available for INSERT...VALUES statements. See this Oracle forum thread for a discussion of this subject.
You can't, BUT at least in Oracle 19c, you can specify a SELECT subquery inside the VALUES clause and so use RETURNING! This can be a good workaround, even if you may have to repeat the WHERE clause for every field:
INSERT INTO some_table
(column1,
column2)
VALUES((SELECT col1_value FROM other_table WHERE ...),
(SELECT col2_value FROM other_table WHERE ...))
RETURNING id
INTO local_var;
Because the insert is based on a select, Oracle is assuming that you are permitting a multiple-row insert with that syntax. In that case, look at the multiple row version of the returning clause document as it demonstrates that you need to use BULK COLLECT to retrieve the value from all inserted rows into a collection of results.
After all, if your insert query creates two rows - which returned value would it put into an single variable?
EDIT - Turns out this doesn't work as I had thought.... darn it!
This isn't as easy as you may think, and certainly not as easy as it is using MySQL. Oracle doesn't keep track of the last inserts, in a way that you can ping back the result.
You will need to work out some other way of doing this, you can do it using ROWID - but this has its pitfalls.
This link discussed the issue: http://forums.oracle.com/forums/thread.jspa?threadID=352627
Related
I have a trigger whose purpose is to fire whenever there is a DELETE on a particular table and insert the deleted data into another table in json format.
The trigger works fine if I am specifying each column explicitly. Is there any way to access the entire table row?
This is my code.
TRIGGER1
AFTER DELETE
ON QUESTION
FOR EACH ROW
DECLARE
json_doc CLOB;
BEGIN
select json_arrayagg (
json_object ('code' VALUE :old.id,
'name' VALUE :old.text,
'description' VALUE :old.text) returning clob
) into json_doc
from dual;
PROCEDURE1(json_doc);
END;
This works fine. However, what I want is something like this. Instead of explicity specifying each column, I want to convert the entire :OLD data
TRIGGER1
AFTER DELETE
ON QUESTION
FOR EACH ROW
DECLARE
json_doc CLOB;
BEGIN
select json_arrayagg (
json_object (:old) returning clob
) into json_doc
from dual;
PROCEDURE1(json_doc);
END;
Any suggestion please.
The short and correct answer is you can't. We have a few tables in our application where we do this and the developer is responsible for updating the trigger when they add a column: this is enforced with code reviews and is probably the cleanest solution for this scenario.
The long answer is you can get close, but I wouldn't do this in production for several reasons:
Triggers are terrible for performance
Triggers are terrible for code clarity
This requires reading the row again using flashback query so
You aren't getting the values of this row from inside your current transaction: if you update the row in your transaction and then delete it the JSON will show what the values were BEFORE your update
There is a performance penalty for reading from UNDO
There is potential that UNDO won't be available and your trigger will fail
Your user needs permission to execute flashback queries
Your database needs to meet all the perquisites to support flashback queries
Deleting a lot of rows will cause the ROWID collection to get large and consume PGA
There are probably more reasons, but in the interest of "can it be done" here you go...
DROP TABLE t1;
DROP TABLE t2;
DROP TRIGGER t1_ad;
CREATE TABLE t1 (
id NUMBER,
name VARCHAR2(100),
description VARCHAR2(100)
);
CREATE TABLE t2 (
dt TIMESTAMP(9),
json_data CLOB
);
INSERT INTO t1 VALUES (1, 'A','aaaa');
INSERT INTO t1 VALUES (2, 'B','bbbb');
INSERT INTO t1 VALUES (3, 'C','cccc');
INSERT INTO t1 VALUES (4, 'D','dddd');
CREATE OR REPLACE TRIGGER t1_ad
FOR DELETE ON t1
COMPOUND TRIGGER
TYPE t_rowid_tab IS TABLE OF ROWID;
v_rowid_tab t_rowid_tab := t_rowid_tab();
AFTER EACH ROW IS
BEGIN
v_rowid_tab.extend;
v_rowid_tab(v_rowid_tab.last) := :old.rowid;
END AFTER EACH ROW;
AFTER STATEMENT IS
v_scn v$database.current_scn := dbms_flashback.get_system_change_number;
v_json_data CLOB;
v_sql CLOB;
BEGIN
FOR i IN 1 .. v_rowid_tab.count
LOOP
SELECT 'SELECT json_arrayagg(json_object(' ||
listagg('''' || lower(t.column_name) || ''' VALUE ' ||
lower(t.column_name),
', ') within GROUP(ORDER BY t.column_id) || ') RETURNING CLOB) FROM t1 AS OF SCN :scn WHERE rowid = :r'
INTO v_sql
FROM user_tab_columns t
WHERE t.table_name = 'T1';
EXECUTE IMMEDIATE v_sql
INTO v_json_data
USING v_scn, v_rowid_tab(i);
INSERT INTO t2
VALUES
(current_timestamp,
v_json_data);
END LOOP;
END AFTER STATEMENT;
END t1_ad;
/
UPDATE t1
SET NAME = 'zzzz' -- not captured
WHERE id = 2;
DELETE FROM t1 WHERE id < 3;
SELECT *
FROM t2;
-- 13-NOV-20 01.08.15.955426000 PM [{"id":1,"name":"A","description":"aaaa"}]
-- 13-NOV-20 01.08.15.969755000 PM [{"id":2,"name":"B","description":"bbbb"}]
I have a created a procedure as
create or replace procedure availability(num in number) as
begin
delete from vehicle_count;
insert into vehicle_count from select engine_no,count(engine_no)
from vehicle
where engine_no = num
group by engine_no;
end;
/
The procedure was created successfully but now i have to write a separate query to view the contents of vehicle_count as
select * from vehicle_count;
I tried inserting the select statement into the procedure after insertion but it showed a error stating "an INTO clause is expected in the select statement".
How can I create procedure to select the required contents and display it in a single execute statement?
Table schema
vehicle(vehicle_no,engine_no,offence_count,license_status,owner_id);
vehicle_count(engine_no,engine_count);
Check this (MS SQL SERVER)-
create or alter procedure availability(#num as int) as
begin
delete from vehicle_count;
insert into vehicle_count
output inserted.engine_no,inserted.count_engine_no
select engine_no,count(engine_no) as count_engine_no
from vehicle
where engine_no=#num
group by engine_no;
end;
If you want to use a SELECT into a PL/SQL block you should use either a SELECT INTO or a loop (if you want to print more rows).
You could use something like this:
BEGIN
SELECT engine_no, engine_count
INTO v_engine, v_count
FROM vehicle_count
WHERE engine_no = num;
EXCEPTION
WHEN NO_DATA_FOUND THEN
v_engine := NULL;
v_count := NULL;
END;
v_engine and v_count are two variables. You can declare them in your procedure, and they will contain the values you want to print.
You said that the procedure you wrote (actually, you posted here) compiled successfully. Well, sorry to inform you - that's not true. This is not a valid syntax:
insert into vehicle_count from select engine_no,count(engine_no)
----
from? Here?
Consider posting true information.
As of your question (if we suppose that that INSERT actually inserted something into a table):
at the beginning, you delete everything from the table
as SELECT counts number of rows that share the same ENGINE_NO (which is equal to the parameter NUM value), INSERT inserts none (if there's no such NUM value in the table) or maximum 1 row (because of aggregation)
therefore, if you want to display what's in the table, all you need is a single SELECT ... INTO statement whose result is displayed with a simple DBMS_OUTPUT.PUT_LINE which will be OK if you're doing it interactively (in SQL*Plus, SQL Developer, TOAD and smilar tools). Regarding table description, I'd say that ENGINE_NO should be a primary key (i.e. that not more than a single row with that ENGINE_NO value can exist in a table).
create or replace procedure availability (num in number) as
l_engine_no vehicle_count.engine_no%type;
l_engine_count vehicle_count.engine_count%type;
begin
delete from vehicle_count;
insert into vehicle_count (engine_no, engine_count)
select engine_no, count(engine_no)
from vehicle
where engine_no = num
group by engine_no;
-- This query shouldn't return TOO-MANY-ROWS if ENGINE_NO is a primary key.
-- However, it might return NO-DATA-FOUND if there's no such NUM there, so you should handle it
select engine_no, engine_count
into l_engine_no, l_engine_count
from vehicle_count
where engine_no = num;
dbms_output.put_line(l_engine_no ||': '|| l_engine_count);
exception
when no_data_found then
dbms_output.put_line('Nothing found for ENGINE_NO = ' || num);
end;
/
There are numerous alternatives to that (people who posted their answers/comments before this one mentioned some of those), and the final result you'd be satisfied with depends on where you want to display that information.
I want to create a procedure in which I use a cursor to select certain lines and then insert them in another table. I wonder if there is a notation to write it faster.
For instance here is the complete procedure
create or replace procedure myProc as
Cursor lines is
select * from table1 where c = '2';
begin
for line in lines loop
insert into table2 values(line.a, line.b, line.c, line.d ....);
end loop;
end;
/
I want to know if I can replace the 'insert into' line by something like
insert into table2 values(line.something);
or
insert into tables2 values(something(line));
(I think a view could be more effective but it's not the question here.)
Absolutely:
create or replace procedure myProc as
begin
insert into table2( . . .)
select a, b, c, d, . .
from table1
where c = '2';
end;
/
You should list the columns in table2 as well. That is what the table2( . . . ) means.
Despite the following code should work in case structure of both tables were identical...
INSERT INTO target_table
SELECT * FROM source_table;
... you should avoid this way of programming because any column addition to source or target table will end in SQL became invalid.
Looks like you're looking for a short-cut in order to avoid qualifying column names in an insert statement.
Well, although I do not recommend this, it can be done bug it's a little tedious. Here is an example:
if you have a table employee (employee_id, employee_name, designation). You'll need to create two types in the database-
a. An object type which is similar to employee-
create type employee_obj as object (employee_id number(28,0),
employee_name varchar2(100),
designation varchar2(30));
b. Create the set type of this record-
create type employee_obj_set is table of employee_obj;
c. In your PL/SQL procedure you can use something like:
DECLARE
empset employee_obj_set;
-- More variables here
BEGIN
-- other operations
-- populate your empset
-- other operations
INSERT INTO employee
SELECT * FROM table(empset);
-- other operations
END;
/
For faster inserts use bulk collect/FORALL insert instead of singular inserts.Find below sample...
DECLARE
CURSOR lines is select * from table1 where c = '2';
type lines_ty is table of table2%rowtype;
l_lines_t2 lines_ty;
BEGIN
OPEN lines;
LOOP
FETCH lines BULK COLLECT INTO l_lines_t2 LIMIT 500;
FORALL i IN 1..l_lines_t2.COUNT
INSERT INTO table2 VALUES l_lines_t2(i);
EXIT WHEN lines%NOTFOUND;
END LOOP;
CLOSE lines;
END;
Hi i am using the below PLSQL script to insert rows in new table new_table.
set serveroutput on SIZE 1000000;
DECLARE
CURSOR get_record IS
SELECT * from cycle_table ;
BEGIN
FOR rec IN get_record
LOOP
DBMS_OUTPUT.put_line('Inserting Record into new_table..');
EXECUTE IMMEDIATE('insert into new_table
select cycle_code,cycle_instance,cycle_start_date,cycle_end_date
from cycle_table');
END LOOP;
COMMIT;
END;
/
Now the table cycle_table consist only 4 rows. The loop runs only four times beacuse its printing 'Inserting Record into new_table..' 4 times only.
But when i see the new_table it consist 16 rows. Which means everytime the loop iterates it insert all the 4 rows and thus total 16 rows.
What i want is that it insert single row at a time.
So that i can perform other actions on that row also. Like if the row already exist, insert in some other table or anything.
Please suggest what can I do here? I am using SQL developer on oracle 10g
Thanks in advance
It is very simple:
set serveroutput on SIZE 1000000;
DECLARE
BEGIN
FOR rec in (select * from cycle_table)
LOOP
DBMS_OUTPUT.put_line('Inserting Record into new_table..');
insert into new_table (cycle_code,
cycle_instance,
cycle_start_date,
cycle_end_date)
values (rec.cycle_code,
rec.cycle_instance,
rec.cycle_start_date,
rec.cycle_end_date);
END LOOP;
COMMIT;
END;
/
I would discourage this approach, though, as you could run into a performance issue if there is a large number of records. You have only four, so it's fine.
The reason I'm against this is that there is context switching involved between Oracle's PL/SQL engine and SQL engine. I'd suggest you do an insert into .... select... or use a forall instead, as these are the least resource-consuming approaches.
A more efficient way is to eliminate all of the looping, and allow the SQL to handle everything. Here is my suggestion:
BEGIN
-- Handle matches first, because after you handle non-matches, everything matches
INSERT INTO match_table (cycle_code, cycle_instance, cycle_start_date
, cycle_end_date)
SELECT cycle_table.cycle_code, cycle_table.cycle_instance, cycle_table.cycle_start_date
, cycle_table.cycle_end_date
FROM cycle_table INNER JOIN new_table ON (new_table.cycle_code = cycle_table.cycle_code);
-- Single insert to insert all non matching records
INSERT INTO new_table (cycle_code, cycle_instance, cycle_start_date
, cycle_end_date)
SELECT cycle_code, cycle_instance, cycle_start_date
, cycle_end_date
FROM cycle_table
WHERE NOT EXISTS
(SELECT NULL
FROM new_table
WHERE new_table.cycle_code = cycle_table.cycle_code);
COMMIT;
END;
In an Oracle table (e.g. MYTABLE, with a numeric sequenced field as primary key), I have to insert several thousand of rows, but some of them are supposed to already exist in the table.
Naturally, I should try to use MERGE but I need, as well, to retrieve all created (when inserting) and existing (when updating) primary keys.
As well, it should be as fast as possible.
Is the following attempt (pseudo code) the only way to go? Thanks.
keys_list = empty array
for each row to merge
do query 'SELECT PK_MYTABLE FROM MYTABLE WHERE PK_MYTABLE = '+row.pk_mytable
==> retrieve key
if found then:
add key to keys_list
else:
do query 'INSERT INTO MYTABLE (PK_MYTABLE, ...) VALUES (SEQ_MYTABLE.NEXTVAL, ...)'
do query 'SELECT SEQ_MYTABLE.CURRVAL FROM DUAL' ==> retrieve key
add key to keys_list
Add a MODIFICATION_DATE column to the table
Grab and save the sysdate.
When you merge update/insert the value of the sysdate as well.
When the merge is complete, select the rows where the MODIFICATION_DATE = SYSDATE and you
have the set you are interested in.
Why can't you use a MERGE statement for this? This is exactly what a MERGE is for. Here is a rough idea of how it would look...
merge into mytable mt
using
(
select key_field, value_field from sourcetable
) st
on
( mt.key_field = st.key_field )
when matched then update
set mt.value_field = st.value_field
when not matched then insert
( key_field, value_field )
values
( st.key_field, st.value_field )
;
Using a MERGE statement is fast because it is a single statement and the Oracle optimizer can utilize indexes and choose a better explain path than iterating through a cursor using PL/SQL.
If the keys are being generated from a sequence, then the normal way to get the key generated by that insert is to use the returning clause:
declare
v_insert_seq integer;
begin
insert into t1 (pk, c1)
values (myseq.nextval, 'value') returning pk into v_insert_seq;
end;
/
However, as best as I can tell, the merge statement doesn't support that returning feature.
Depending on the source of your new rows, there are different ways you could do this. If you are inserting one row at a time, then the approach above will work pretty well.
To detect the duplicate records, just catch the exceptions when you are inserting (when dup_val_on_index) and then handle them with updates.
If your source of rows is another table, you probably want to look at bulk inserts, and allowing Oracle to return you an array of new PK values. I tried this, but couldn't get it working, so perhaps it's not supported (or I'm missing something today - it gives a syntax error):
declare
type t_type is table of t1.pk%type;
v_insert_seqs t_type;
begin
insert into t1 (pk, c1)
select level newpk, 'value' c1value
from dual
connect by level <= 10 returning pk bulk collect into v_insert_seqs;
exception
when dup_val_on_index then
raise;
end;
/
The next best thing is to select the rows into arrays and then use bulk binds with the returning clause to capture the new PK IDs and also use Save Exceptions to catch all the rows that failed to inserted. Then you can process any of the failed inserted afterwards:
set serveroutput on
declare
type t_pk is table of t1.pk%type;
type t_c1 is table of t1.c1%type;
v_pks t_pk;
v_c1s t_c1;
v_new_pks t_pk;
ex_dml_errors EXCEPTION;
PRAGMA EXCEPTION_INIT(ex_dml_errors, -24381);
begin
-- get the batch of rows you want to insert
select level newpk, 'value' c1
bulk collect into v_pks, v_c1s
from dual connect by level <= 10;
-- bulk bind insert, saving exceptions and capturing the newly inserted
-- records
forall i in v_pks.first .. v_pks.last save exceptions
insert into t1 (pk, c1)
values (v_pks(i), v_c1s(i)) returning pk bulk collect into v_new_pks;
exception
-- Process the exceptions
when ex_dml_errors then
for i in 1..SQL%BULK_EXCEPTIONS.count loop
DBMS_OUTPUT.put_line('Error: ' || i ||
' Array Index: ' || SQL%BULK_EXCEPTIONS(i).error_index ||
' Message: ' || SQLERRM(-SQL%BULK_EXCEPTIONS(i).ERROR_CODE));
end loop;
end;
/
If you are running Oracle 10 or better, you might be able to do much the same thing, for nearly free by issuing a commit before the merge to update the SCN, then after the merge,
use the ORA_ROWSCN to detect which rows have changed.