We have created a BLOB type column to store 6000 to 7000 ASCII characters of payload.
I am inserting the payload using following code.
PreparedStatement stmt=conn.prepareStatement(insertQuery);
String record="My ASCII Payload";
stmt.setBytes(1, record.getBytes());
stmt.execute();
When i run below query i can see the length of chars in the output.
select dbms_lob.getLength(MY_PAYLOAD)
from RAW_DATA;
6085
I wanted to perform text search in this BLOB column , i tried following options nothing works. What is the correct way to perform text search in a BLOB column ?
SELECT * FROM RAW_DATA t
WHERE dbms_lob.instr(t.MY_PAYLOAD,utl_raw.cast_to_raw(‘1234’))>0
select *
from RAW_DATA
where dbms_lob.instr (MY_PAYLOAD, -- the blob
utl_raw.cast_to_raw ('1234'),
1, -- where to start. i.e. offset
1 -- Which occurrance i.e. 1=first
) > 0
You can create text indexes on blobs as long as they contain text (naturally)
SQL> create table t ( x int, b blob);
Table created.
SQL>
SQL> insert into t values ( 1, utl_raw.cast_to_Raw('Hello there'));
1 row created.
SQL> insert into t values ( 2, utl_raw.cast_to_Raw('Goodbye tomorrow'));
1 row created.
SQL>
SQL> create index t_ix on t ( b )
2 indextype is ctxsys.context;
Index created.
SQL>
SQL> select x
2 from t where contains(b,'Hello') > 0;
X
----------
1
SQL>
SQL> select utl_raw.cast_to_varchar2(b)
2 from t
3 where contains(b,'Hello') > 0;
UTL_RAW.CAST_TO_VARCHAR2(B)
---------------------------------------------------------------------------------
Hello there
Related
I have procedure like this...
declare
v_psg varchar2(10);
id_no number;
begin
select value into v_psg from settings_am where key = 'PSG';
select id into id_no from product where to_char(psg) in (v_psg);
end;`
The value returned from select value into v_psg from settings_am where key = 'PSG'; would be
'1','2','3'
when i run this procedure i am returned with ora error - ORA-01403.
please advise how i should pass the v_psg value to psg column of product table?
EDIT - Tried with test case suggested
If you got ORA-01403, you were kind of lucky. It is the NO_DATA_FOUND error, which means that one (probably the first) query didn't return anything.
Those two statements could be combined into
select id
from product
where to_char(psg) in (select value
from settings_am
where key = 'PSG'
);
Why would you select value first, and then use it in another query? Besides, it just wouldn't work. v_psg is declared as VARCHAR2 variable. The way you described it, it contains the following string: '1','2','3', as if this is what you have:
SQL> create table settings_am (key varchar2(10),
2 value varchar2(20)); --> note size here
Table created.
SQL> insert into settings_am (key, value)
2 values ('PSG', q'['1','2','3']');
1 row created.
SQL> select * From settings_am;
KEY VALUE
---------- --------------------
PSG '1','2','3'
SQL>
As you can see, I enlarged the value column size, although variable you declared says 10. Why? Because of
SQL> select length(value) from settings_am where key = 'PSG';
LENGTH(VALUE)
-------------
11
i.e. you can't put something that is long 11 into something that accepts length 10.
Or, if your data actually contains 3 rows for the PSG key, are those values already enclosed into single quotes? If so, that's strange; people usually don't do that. Anyway, suppose that you managed to get string '1,2,3' (which is what I presume you actually have) into a VARCHAR2 variable, then you have to split it into rows in order to be able to use it in the IN clause:
SQL> create table product (id number, psg varchar2(10));
Table created.
SQL> insert into product (id, psg) values (100, '1');
1 row created.
SQL> insert into product (id, psg) values (200, '2');
1 row created.
SQL>
Query is then (where lines #3 - 5 represent splitting a string into rows):
SQL> select p.id
2 from product p
3 where p.psg in (select regexp_substr('&&v_psg', '[^,]+', 1, level)
4 from dual
5 connect by level <= regexp_count('&&v_psg', ',') + 1
6 );
Enter value for v_psg: 1,2,3
ID
----------
100
200
So, wouldn't it be simpler to use
SQL> select id
2 from product
3 where to_char(psg) in (select value
4 from settings_am
5 where key = 'PSG'
6 );
ID
----------
100
200
SQL>
Note that both options also show why your query is wrong: you can't put two values (rows) into a variable declared as id_no number; as you'd get TOO_MANY_ROWS error.
Finally, what is it that you'd want to do? What problem are you trying to solve? Apparently, except for special cases (only one row for each value) your query can't work. If you could provide test case (create table & insert into sample data), as well as expected output, it would be easier to help you.
I have data in the below format and I want to update a destination table column of type varchar2 with below values. But the problem is it updates as .462 instead of 0.462 by using trim with leading '0'.
source destination column
----------------- ------------------
0000004.304300000 4.3043
0000005.504500000 5.5045
0000141.400000000 141.4
0000138.900000000 138.9
0000000.462000000 0.462
0000000.000297000 0.000297
A little bit of TO_CHARing and TO_NUMBERing with appropriate format mask might do the job. Have a look at the example:
SQL> create table test (source varchar2 (20), destination varchar2(20));
Table created.
SQL> insert into test (source)
2 select '0000004.304300000' from dual union all
3 select '0000000.462000000' from dual union all
4 select '0000141.400000000' from dual union all
5 select '0000033.000000000' from dual;
4 rows created.
SQL> alter session set nls_numeric_characters = '.,';
Session altered.
SQL> update test set
2 destination = rtrim(to_char(to_number(source), 'fm999990D99999999'), '.');
4 rows updated.
SQL> select * From test;
SOURCE DESTINATION
-------------------- --------------------
0000004.304300000 4.3043
0000000.462000000 0.462
0000141.400000000 141.4
0000033.000000000 33
SQL>
How do I return the value of an identity column (id) in Oracle 12c after insertion? Seems like most of the approaches out there uses sequence to get back the id of the inserted item.
Simply use the RETURNING clause.
For example -
RETURNING identity_id INTO variable_id;
Test case -
SQL> set serveroutput on
SQL> CREATE TABLE t
2 (ID NUMBER GENERATED ALWAYS AS IDENTITY, text VARCHAR2(50)
3 );
Table created.
SQL>
SQL> DECLARE
2 var_id NUMBER;
3 BEGIN
4 INSERT INTO t
5 (text
6 ) VALUES
7 ('test'
8 ) RETURNING ID INTO var_id;
9 DBMS_OUTPUT.PUT_LINE('ID returned is = '||var_id);
10 END;
11 /
ID returned is = 1
PL/SQL procedure successfully completed.
SQL>
SQL> select * from t;
ID TEXT
---------- --------------------------------------------
1 test
SQL>
As we all know LONG is deprecated in Oracle a long back but Oracle itself is still using this datatype in their views.
So if I have to change LONG into some kind of text datatype how can I achieve that.
I am trying to query this and getting error.
ORA-00932: inconsistent datatypes: expected - got LONG
Query -
SELECT NVL(ie.column_expression, ic.column_name)
from user_ind_columns ic left join user_ind_expressions ie on ic.index_name = ie.index_name and ic.table_name = ie.table_name
where ic.table_name = 'Some Table'
There are several methods, one such is create table using TO_LOB. It is designed to convert a LONG or LONG RAW column to a CLOB or BLOB, respectively. Other methods are using PL/SQL, DBMS_XMLGEN. You can also use TO_LOB in insert statements.
Let's see how to convert LONG into CLOB-
SQL> CREATE TABLE t (x INT, y LONG);
Table created.
SQL>
SQL> INSERT INTO t VALUES (1, RPAD('*',9,'!'));
1 row created.
SQL> INSERT INTO t VALUES (2, RPAD('*',9,'#'));
1 row created.
SQL> INSERT INTO t VALUES (3, RPAD('*',9,'#'));
1 row created.
SQL> COMMIT;
Commit complete.
SQL>
So, we have table t with column y s LONG data type.
SQL> CREATE TABLE t1
2 AS
3 SELECT * FROM t
4 /
SELECT * FROM t
*
ERROR at line 3:
ORA-00997: illegal use of LONG datatype
SQL>
We can see the LONG restriction.
Let's use TO_LOB to convert it into CLOB.
SQL> CREATE TABLE t1
2 AS
3 SELECT x,
4 to_lob(y) as y
5 FROM t
6 /
Table created.
SQL> desc t1;
Name Null? Type
----------------------------------------------------- -------- ------------------------------------
X NUMBER(38)
Y CLOB
SQL>
Now you have the same table with the LONG column converted to CLOB.
this is stupid (as in probably not efficient) but it works for samll lengths of y (ie < 2000 characters)..
CREATE TABLE t (x INT, y LONG);
INSERT INTO t VALUES (1, RPAD('*',9,'!'));
CREATE TABLE t1
AS
SELECT x,
regexp_substr(SYS.DBMS_XMLGEN.GETXML('select y from t where rowid = '''||rowid||''''),'<Y>(.*)</Y>',1,1,'in',1) y
FROM t
/
it works by using dbms_xmlgen to generate a clob based on the LONG column.. then substr-ing the value back out.
this only works for small contents of the LONG column. but that is all i had and this worked for me.
I had a similar need, to list the objects and their sizes (including info on columns used in indexes), and came with this solution:
select idx1.table_owner owner, idx1.table_name, idx1.index_name, listagg(nvl(idx1.column_expression,idx1.column_name),',') within group (order by idx1.column_position) column_name
from xmltable(
'/ROWSET/ROW'
passing (select dbms_xmlgen.getxmltype('select ic.table_owner, ic.table_name, ic.index_name, ic.column_position, ic.column_name, ie.column_expression
from all_ind_columns ic
left outer join dba_ind_expressions ie on ie.table_owner=ic.table_owner and ie.table_name=ic.table_name and ie.index_name=ic.index_name and ie.column_position=ic.column_position') from dual)
columns index_name varchar2(30) path 'INDEX_NAME'
, table_owner varchar2(30) path 'TABLE_OWNER'
, table_name varchar2(30) path 'TABLE_NAME'
, column_position number path 'COLUMN_POSITION'
, column_name varchar2(30) path 'COLUMN_NAME'
, column_expression varchar2(4000) path 'COLUMN_EXPRESSION') idx1
group by idx1.table_owner, idx1.table_name, idx1.index_name
How can we set default value for a number typed column has '00'?I tried this but it still saved it has '0',I need to do this.
//alter table table_name add column column1 default '00';
Please suggest a way for me.
You can store as number and when you retrieve do as
select column1,to_char(column2,'00') from mytable
SQL Fiddle Demo
It is not possible to store numbers like that - In a format you have described ('00'). You can store numbers as numbers (as values of numeric data type of course) and use to_char function or to_char function combined with lpad function to represent numbers in a format you like. Here is an example:
SQL> create table TB_SingleNumberColumn(
2 col number
3 )
4 /
Table created
SQL> insert into TB_SingleNumberColumn(Col) values(1);
1 row inserted
SQL> insert into TB_SingleNumberColumn(Col) values(5);
1 row inserted
SQL> insert into TB_SingleNumberColumn(Col) values(11);
1 row inserted
SQL> insert into TB_SingleNumberColumn(Col) values(111);
1 row inserted
SQL> commit;
Commit complete
-- The values as they are
SQL> select * from TB_SingleNumberColumn;
COL
----------
1
5
11
111
-- Values padded with zeros.
SQL> select to_char(col, '000') res
2 from TB_SingleNumberColumn;
RES
----
001
005
011
111
SQL> select lpad(to_char(col), 3, '0')
2 from TB_SingleNumberColumn
3 ;
LPAD(TO_CHAR(COL),3,'0')
------------------------
001
005
011
111