implicit conversion to sql

Hi guys,.

I have a question on how the oracle is posting implicit conversion in the following cases.

I have a table: table1 with two columns: a number (10), b varchar2 (10). Both are indexed unique.

If I have a SQL like this:

select * from table1 where a = '99'

the implicit conversion is performed on '99' and the final SQL will be something like:

select * from table1 where a = to_number ('99')

the same thing will happen with 'aaa' and I get an error. Please note that represents the column and the index will be used.

If I have a sql like:

select * from table1 where b = 99

Note that b is the varchar2 column.
the implicit conversion will be done on the column and the index will not be used

select * from table1 where to_number (b) = 99

THEN, forgetting for a moment that it is not OK to allow an implicit conversion, what are the rules that apply here? If the number, the conversion will be made on the value and varchar conversion will be performed on the column? What else?

Thank you

http://docs.Oracle.com/CD/E11882_01/server.112/e26088/sql_elements002.htm#SQLRF51047

Tags: Database

Similar Questions

If an implicit conversion takes place
Hi all

I have two queries tell

SQL > SELECT * FROM EMP WHERE HIREDATE BETWEEN February 20, 81 'AND February 20, 82';

EMPNO, ENAME, JOB HIREDATE DEPTNO COMM SAL MGR
----- ---------- --------- ---------- --------- ---------- ---------- ----------
7499 ALLEN 7698 1600 20 FEBRUARY SALESMAN 81 300 30
7521 WARD 7698 1250 22 FEBRUARY SALESMAN 81 500 30
7566 JONES MANAGER 7839 2975 2 APRIL 81 20
7654 MARTIN 7698 28 - SEP - 81 1250 1400 30 SALESMAN
7698 BLAKE MANAGER 7839 2850 MAY 1, 81 30
7782 10 removed MANAGER 7839 2450 June 9 81 10
7839 10 deleted PRESIDENT November 17, 81 5000 10
7844 TURNER 7698 08 - SEP - 81 1500 0 30 SALESMAN
7900 7698 DECEMBER 3, CLERK JAMES 81 950 30
7902 7566 3000 3 DECEMBER ANALYST FORD 81 20
7934 10 deleted CLERK 7782 1300 January 23 82 10

AND

SQL > select' 17-SEP-2010 + 1 double;

ERROR on line 1:
ORA-01722: invalid number

now in the first case, I spent a string and that turned into a date
but in the second case, I try to add a day to this date it is not be converted in the second case.

When in fact the implicit conversion is done.

Thank you and best regards,
Sri ram.
Probably below may be the reason.
In the first query you asked oracle to match the hiring date between two dates. Oracle implicitley equal strings depending on the hiring date data type.
But in the 2nd query you specify a string in your oracle query does not include what format to convert the string.

CVI2013: WARNING: implicit conversion changes signedness: "long" to "int".

What is Clang was trying to say with this warning?
```
int f( void )
{
  int i1 = 4;

  int i2 = 1l << ( i1 - 1 );
  //       ^^  warning: implicit conversion changes signedness: 'long' to 'int' [-Wsign-conversion]

  return i2;
}
```
Hello CVI - User!

The explanation is quite technical...

The result type of the expression of SHIFT is "signed for a long time." The target type of the initialization is "signed int". Clang injects an implicit "signed long" conversion to "signed int" and check the different properties of the conversion.

In particular, clang tests if the sign of the value can change during the conversion. For this test, clang calculates the range of 'values' of the source 1 L< (i1="" -="" 1)'="" and="" the="" "value="" range"="" of="" the="" target="" 'int="" i2'.="" the="" warning="" indicates="" that="" the="" signs="" of="" the="" source="" and="" target="" ranges="" do="" not="">

The range of values of the source depends on the expression 1 L< (i1="" -="" 1)'="" and="" here="" clang="" is="" applying="" a="" special="">
```
    [excerpt from GetExprRange in the clang 2.9 source code]

    // Left shift gets black-listed based on a judgement call.
    // ...except that we want to treat '1 << (blah)' as logically
    // positive.  It's an important idiom.
```
Therefore the range of values of source is: wide, 32-bit positive.

The range of values of the target depends on the type of "i2" and that's the range of default values for "signed int": 32 - bit wide, positive or negative.

In other words: 1 L< (i1="" -="" 1)'="" is="" considered="" positive="" (unsigned)="" and="" it="" is="" converted="" to="" a="" signed="">

Note that this is really a special case and that you do not get the warning for other values:
```
  int i2 = 2l << ( i1 - 1 ); //no warning
```
Unfortunately, I can not explain why clang didn't just compare the source and target types, it calculates the ranges of values and why 1< (blah)'="" is="" treated="" specially.="" i="" don't="">

Peter

TopLink grid 11g: conversion of SQL queries to the filters of consistency
I would like to know how efficient the conversion of SQL queries for filters of coherence is wen search an object in the cache coherence. TopLink does something as instruction cache cache filters consistency as well? In this regard, you have number of performance?

Thank you very much!

Ming-Wen
TopLink grid currently has a rudimentary filter cache that allows TopLink to only translate NamedQuery once. This allows TopLink grid to reuse the same filter on subsequent runs of this NamedQuery.
-Gordon

Conversion of SQL Query Help
Hello, I'm doing a migration of SQL Server T - SQL to oracle 11 g PL/SQL, and I don't want to use a global temporary table; Instead, consider using sliders and pl/sql table.

I got hit in a scenario where I need to convert the following in a single sql statement.

Stage of T - SQL:
============

Insert (#ActCde)
InvntrySbLdgrChngeID
DlHdrID
DlDtlID
ChemID
StrtgyID
Quantity
Value
Modified)
Select InvntrySbLdgrChngeID
DlHdrID
DlDtlID
ChildPrdctID
StrtgyID
VolumeChanged
ValueChanged
, 'N'
of InventorySubLedgerChange
where LastInChain = 'Y '.

-updated column "Changed" of particular documents to 'Y' based on conditions

Update #ActCde
Changed the value = 'Y '.
of #ActCde (nolock)
Left Outer Join InventorySubledgerDetail ISD (nolock)
on #ActCde.DlHdrID = DSI. DlDtlChmclDlDtlDlHdrID
and #ActCde.DlDtlID = ISD. DlDtlChmclDlDtlID
and #ActCde.ChemID = DSI. DlDtlChmclChmclChldPrdctID
and #ActCde.StrtgyID = DSI. StrtgyID
and the DSI. AccntngPrdID = @i_AccntngPrdID - variable parameter
Where abs (#ActCde.Qty - isnull (ISD.) BeginningVolume, 0)) >.01
or abs (#ActCde.Value - isnull (ISD.) BeginningValue, 0)) >.01
or (abs(#ActCde.Qty) = 0 and)
ABS(#ActCde.value) = 0 and
ABS (isnull (ISD. BeginningVolume, 0)) = 0 and
ABS (isnull (ISD. BeginningValue, 0)) = 0
)

Convertion of SQL:
===============
I would like to do something like

Select A.InvntrySbLdgrChngeID,
A.DlHdrID,
A.DlDtlID,
A.ChemID,
A.StrtgyID,
A.Qty,
A.Value,
A.Changed
Of
(select InvntrySbLdgrChngeID
DlHdrID
DlDtlID
ChildPrdctID AS ChemID
StrtgyID
VolumeChanged AS Qty,
ValueChanged value have
, 'N'
of InventorySubLedgerChange
where LastInChain = 'Y') A LEFT OUTER JOIN InventorySubledgerDetail B
ON A.DlHdrID = B.DlDtlChmclDlDtlDlHdrID
and A.DlDtlID = B.DlDtlChmclDlDtlID
and A.ChemID = B.DlDtlChmclChmclChldPrdctID
and A.StrtgyID = B.StrtgyID
and B.AccntngPrdID = i_AccntngPrdID - parameter varibale
Where abs (A.Qty - isnull(B.BeginningVolume,0)) >.01
or abs (A.Value - isnull(B.BeginningValue,0)) >.01
or (abs (A.Qty) = 0 and)
ABS (A.value) = 0 and
ABS (IsNull(B.BeginningVolume,0)) = 0 and
ABS (IsNull(B.BeginningValue,0)) = 0
)

And I need to change the 'A.CHANGED' column value 'Y', based on the outer join, as is above T - SQL.

IF this wont work so pls help me to another way to do this.

in this regard, your help is appreciated.

Thank you!

SS
Try this
```
select #ActCde.InvntrySbLdgrChngeID
       , #ActCde.DlHdrID
       , #ActCde.DlDtlID
       , #ActCde.ChildPrdctID
       , #ActCde.StrtgyID
       , #ActCde.VolumeChanged
       , #ActCde.ValueChanged
       , case when abs(#ActCde.Qty - isnull(ISD.BeginningVolume,0)) > .01  or
                   abs(#ActCde.Value - isnull(ISD.BeginningValue,0)) > .01 or
                   (
                         abs(#ActCde.Qty)                   = 0 and
                         abs(#ActCde.Value)                 = 0 and
                         abs(isnull(ISD.BeginningVolume,0)) = 0 and
                         abs(isnull(ISD.BeginningValue,0))  = 0
                   )
              then 'Y'
              else 'N'
         end changed
  from InventorySubLedgerChange #ActCde
  Left Outer Join InventorySubledgerDetail ISD (nolock)
    on #ActCde.DlHdrID = ISD.DlDtlChmclDlDtlDlHdrID
   and #ActCde.DlDtlID= ISD.DlDtlChmclDlDtlID
   and #ActCde.ChemID = ISD.DlDtlChmclChmclChldPrdctID
   and #ActCde.StrtgyID = ISD.StrtgyID
   and ISD.AccntngPrdID = @i_AccntngPrdID - paramenter variable
 where LastInChain = "Y"
```

Conversion of SQL worksheet.

Hello

If I sent someone a microsoft excel spreadsheet, the chances that you would be able to convert in SQL if I named a few tables and columns that were to be used are there?

Please let me know,

Thank you SM.

If your excel workbook is formatted so that you have a 'table' by the worksheet contained in it and the first row on the worksheet defines column names and the following lines are the data, then you can treat the workbook as an external database using the ODBC Excel driver.

Set up:

1- Go to Control Panel>Administrative Tools>Data Sources (ODBC)>System DSN and create a data source with appropriate driver. Name it EXCL. 

2- In %ORACLE_HOME%\Network\Admin\Tnsnames.ora fie add entry:

EXCL =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = TCP)(HOST = 10.12.0.24)(PORT = 1521))
)
(CONNECT_DATA =
(SID = EXCL)
)
(HS = OK)
)

Here SID is the name of data source that you have just created.

3- In %ORACLE_HOME%\Network\Admin\Listener.ora file add:

(SID_DESC =
(PROGRAM = hsodbc)
(SID_NAME = )
(ORACLE_HOME = )
)

under SID_LIST_LISTENER like:

SID_LIST_LISTENER =
(SID_LIST =
(SID_DESC =
(SID_NAME = PLSExtProc)
(ORACLE_HOME = d:\ORA9DB)
(PROGRAM = extproc)
)
(SID_DESC =
(GLOBAL_DBNAME = ORA9DB)
(ORACLE_HOME = d:\ORA9DB)
(SID_NAME = ORA9DB)
)
(SID_DESC =
(PROGRAM = hsodbc)
(SID_NAME = EXCL)
(ORACLE_HOME = D:\ora9db)
) ) 

Dont forget to reload the listener 

c:\> lsnrctl reload

4- In %ORACLE_HOME%\hs\admin create init.ora. For our sid EXCL we create file initexcl.ora.

In this file set following two parameters:

HS_FDS_CONNECT_INFO = excl
HS_FDS_TRACE_LEVEL = 0

5- Now connect to Oracle database and create database link with following command:

SQL> CREATE DATABASE LINK excl
2 USING 'excl'
3 /

Database link created.

Now you can perform query against this database like you would for any remote database.
SQL> SELECT table_name FROM all_tables@excl;

TABLE_NAME
------------------------------
DEPT
EMP

Or refer to this Article...
http://asktom.Oracle.com/pls/asktom/f?p=100:11:0:P11_QUESTION_ID:4406709207206

need of a conversion in sql criteria

Hi all

I have a situation where I need help from you guyz. Kindly help, if you could. I have an output of the query to an array of data as follows:

SELECT OBJECT_ID PO_num

, DECODE (SEQUENCE_NUM, 1,' 1 approver: ', 2,' 2nd approver: ', 3, '3rd approver', 4,' 4th approver: ', 5,' 5th approver:', Null) APPROVER_Num

EMPLOYEE_NAME | ' @ ' || to_char(ACTION_DATE, 'DD-mon-YY hh12:mi:SS AM') APPROVER_ACTION

OF PO_ACTION_HISTORY_V

WHERE OBJECT_TYPE_CODE = 'PO '.

AND SEQUENCE_NUM <>0

and object_id = 129302 - in. header id

ORDER BY SEQUENCE_NUM

PO_NUM	APPROVER_NUM	APPROVER_ACTION
129302	1 approver:	Anwar, M. Muhammad Naveed
129302	2nd approver:	Iqbal, M. Zafar
129302	3rd approver	Azeem, Mr. Muhammad Ali
129302	4th approver:	Mr. Faisal Masood Ali Khan
129302	5th approver:	Tijuana, Mr. Rizwan Ali

But I need a query to get something like the following:

PO_NUM	1 approver:	2nd approver:	3rd approver	4th approver:	5th approver:
129302	Anwar, M. Muhammad Naveed	Iqbal, M. Zafar	Azeem, Mr. Muhammad Ali	Mr. Faisal Masood Ali Khan	Tijuana, Mr. Rizwan Ali

Kindly help!

Concerning

Shabbir

so, something like this:

SELECT OBJECT_ID PO_num

, MAX (DECODE (SEQUENCE_NUM, 1,' 1 approver:' | employee_name)) AS '1 approver.

, MAX (DECODE (SEQUENCE_NUM, 2,' 2nd approver:' | employee_name)) AS '2nd approver.

, MAX (DECODE (SEQUENCE_NUM, 3,' 3rd approver:' | employee_name)) AS "3rd approver.

, MAX (DECODE (SEQUENCE_NUM, 4,' 4th approver:' | employee_name)) AS '4th approver.

, MAX (DECODE (SEQUENCE_NUM, 5,' 5th approver:' | employee_name)) AS '4th approver.

OF PO_ACTION_HISTORY_V

WHERE OBJECT_TYPE_CODE = 'PO '.

AND SEQUENCE_NUM <> 0

AND object_id = 129302 - in. header id

Group by object_id, object_type_code

HTH

The implicit conversion of "char to number" or "number of char".
CREATE TABLE test_table
(
NUM NUMBER,
VCH VARCHAR2 (10)
);

INSERT INTO table_test VALUES
(1, '1') ;

INSERT INTO table_test VALUES
(2, '2') ;

SELECT * FROM test_table WHERE vch = 1;

When comparing, what type of data is converted into what - all vch values are converted in number or 1 is converted to char '1'?
Hello

Welcome to the forum!

The answer to your question is in the predicate explain plan section:

filter predicates: to_number ("VCH") = 1

Best regards
Nikolai

Implicit conversion
Hi gurus,

IM test cast of oracle database and describe this Hovanessian:

create table girot (a varchar2 (30));
insert into values girot ('9857");
insert into values girot ('9855');

Select * from gggg where a = 9857 - OK

but another table in my database is error:

Select usur_nmlogin in the seguranca.usuario where usur_id = 99014528
9855

Select * from seguranca.usuario
where usur_nmlogin = 9855;

ORA-01722: invalid number
01722 00000 - "invalid number."
* Cause:
* Action:

table describe:
CREATE TABLE "SEGURANÇA. "" USER ".
(
ACTIVATE THE 'USUR_ID' NUMBER(*,0) NOT NULL,
ACTIVATE THE 'UTIP_ID' NUMBER(*,0) NOT NULL,
NUMBER(*,0) "FUNC_ID"
ACTIVATE THE "USUR_NMLOGIN" VARCHAR2 (11 BYTE) NOT NULL,
VARCHAR2 (40 BYTE) "USUR_NMSENHA");

I do not understand this!
Well, since your column is a varchar2, use a varchar2 thus setting. Place the value in quotes should work.

Nicolas.

Why sys_extract_utc (sysdate) training never block pl/sql error?
Oracle 11.2
--
I can't use
```
SELECT sys_extract_utc(sysdate) FROM DUAL
```
But I can't use
```
v := sys_extract_utc(sysdate)
```
in a procedure or a trigger without causing an error in compilation or execution.

In my view, the function sysdate return value does not contain any information of time zone, so he shouldn't be able to be used as a parameter to sys_extract_utc function, because she needs zone information to perform the conversion.

Clues?

Thanks in advance.
Just pl/sql has an implicit conversion. where sql can not for function return types. run in sql, it also works...

Select sys_extract_utc (to_timestamp (sysdate)) of double;

Error when the concatenation of CLOB using conversion implicit to_char
Hello

Please look at the code below (Oracle 11.1.0.7.0):
```
DECLARE
  c CLOB;
BEGIN
  FOR i IN 1..10
  LOOP
    -- Please uncomment ONE of the lines below
    c := c ||  LPAD('x', 8000, 'z') || i; -- raises error ora-06502
    --c := c || (LPAD('x', 8000, 'z') || i);  -- works
    --c := c ||  LPAD('x', 8000, 'z') || to_char(i);  -- works
  END LOOP;
END;
```
OK, making use of implicit type conversion may not be the best thing to do, but just out of curiosity, is there any explanation for this error to occur?

It seems that, when an implicit conversion is involved without the help of hooks, Oracle tries to convert all in varchar2 before concatenate in the clob, which causes the error.

Thank you
Luis
Hello
>
# 6.1, we have a clob and a varchar2, as in #2; Why Oracle has decided to convert the clob in varchar2 as you say, and not vice versa as in #3?
>

# 5, we have clob and number at the beginning. In order to evaluate the expression oracle requires that two variables was of the same type.
As the implicit conversion there is no number-> clob
but there are several-> varchar2 and clob-> varchar2 oracle choose this one.

SQL Server-> conversion Oracle database FDM
Anyone has any experience in the conversion from SQL Server to Oracle for FDM? (11.1 oracle now supports).

My thought would be to re - treat all files of importation rather than try a conversion of database type.

Thoughts?

Thank you

-John
John,

FDM has always support Oracle DB as a repository.

To convert the SQL Server DB to an Oracle DB server, you have to re-create the metadata for the application at hand. Currently, the feature export to XML of FDM does not take into account differences in filing structures and cannot be used.

You must then re-import all files or continue on creating this day with impatience.

Thank you.

SQL output varies depending on the value with or without single quotes

Hello
My version of the database is the database Oracle 11 g Enterprise Edition Release 11.2.0.1.0 - Production PL/SQL Release 11.2.0.1.0 - Production.

(1) when I use the query select * from hr.employee where id > = '82210' I get the below of lines

13205 selected lines.

2.) when I use the query select * from hr.employee where id > = 82210 I get the below of lines

11871 selected lines.

The ID table in hr.employee is ID NOT NULL VARCHAR2 (80).

I don't have much knowledge in SQL so please can someone explain why this difference in output?

Kind regards
007

Look at the example

create the table test16

(

Identification number,

SAL varchar2 (10),

name varchar2 (20)

);

insert into test16 values (1, "2005", "ABC");

insert into test16 values (2, '2045', 'ABC');

insert into test16 values (3, '2065', 'ABC');

insert into test16 values (4, '2075', 'ABC');

insert into test16 values (5, '1705', 'ABC');

insert into test16 values (6, '1805', 'ABC');

insert into test16 values (7, '1920', 'ABC');

insert into test16 values (8, '2010', 'ABC');

insert into test16 values (9, '0999', 'ABC');

insert into test16 values (10, '0789', 'ABC');

insert into test16 values (11, '510', 'ABC');

insert into test16 values (12, '455', 'ABC');

commit;

Select * from test16 where sal > = 200;

All 12 rows selected because there is an implicit conversion from column sal in numbers

Select * from test16 where sal > = "200";

Only 7 rows selected but waited 12

2005

2045

2065

2075

2010

510

455

Only the sal of lines that begin with a 2 character or more, those who are selected.

Another 1805,1920 sal... .are not selected

Note: Don't store the numbers in a column varchar2.

Getting Ora-06502: PL/SQL of Numeric or value error.

Hello
I m getting the following error when a calculation for form fields.
""Server Ajax Call returned error Ora-06502: PL/SQL digital error or value ".
I m using the dynamic action to calculate my value fields display. Here is the code of my DA.
(Required value) event: change
Selection type (required): product (s)
Article (s) (required): P141_PROP_COST, P141_DOWNPAY_AMT, P141_PREM_PERC, P141_DOWNPAY_PERC, P141_PREM_AMT

Begin
: P141_TOTAL_SELL_PRICE: =: P141_DOWNPAY_AMT +: P141_PREM_AMT;
: P141_COMM_AMT: =: P141_TOTAL_SELL_PRICE*.02;
: P141_TOTAL_SELL_PRICE_FEE: =: P141_TOTAL_SELL_PRICE +: P141_TRANSFER_COST +: P141_COMM_AMt;
: P141_INST_REM_AMT: =: P141_PROP_COST-: P141_DOWNPAY_AMT;
End;
Page point to presented: P141_PROP_COST, P141_DOWNPAY_AMT, P141_PREM_AMT, P141_TRANSFER_COST
Return point page: P141_TOTAL_SELL_PRICE, P141_COMM_AMT, P141_TOTAL_SELL_PRICE_FEE, P141_INST_REM_AMT
I m using another DA for the calculation and validation of my form fields rest. Here is the code:
{$("#P141_DOWNPAY_PERC").change (function ()}
var x = ($("#P141_DOWNPAY_PERC").val () / 100) * $("#P141_PROP_COST").val ();
$("#P141_DOWNPAY_AMT").val (x);
});
{$("#P141_DOWNPAY_AMT").change (function ()}
If ($("#P141_DOWNPAY_AMT").val () > $("#P141_PROP_COST").val ())
{
Alert ("section may be no greater than the basic price");
}
on the other
{var y = ($("#P141_DOWNPAY_AMT").val () / ($("#P141_PROP_COST").val ())) * 100;}
y = Math.Round (y);
$("#P141_DOWNPAY_PERC").val (y) ;}
});
{$("#P141_PREM_PERC").change (function ()}
var x = ($("#P141_PREM_PERC").val () / 100) * $("#P141_PROP_COST").val ();
$("#P141_PREM_AMT").val (x);
});
{$("#P141_PREM_AMT").change (function ()}
If ($("#P141_PREM_AMT").val () > $("#P141_PROP_COST").val ())
{
Alert ("section may be no greater than the basic price");
}
on the other
{var z = ($("#P141_PREM_AMT").val () / ($("#P141_PROP_COST").val ())) * 100;}
z = Math.Round (z);
$("#P141_PREM_PERC").val (z) ;}
});
Any help.
Concerning
Pa

'Problems of implicit conversion' comes to mind when I see this:

: P141_TOTAL_SELL_PRICE: =: P141_DOWNPAY_AMT +: P141_PREM_AMT;

This is similar to the problems I've seen when using the default formats for the Date Picker Item Types.

I will add in some APEX_DEBUG. MESSAGE() lines to check what are the string values.

(remember, all bind variables are of type VARCHAR2)

You should maybe explicitly to convert strings to numbers using TO_NUMBER() and include the appropriate format.

(a bit like you have to do with the conversion of strings to dates.)

My $0.02 worth

MK

Problem with Sql Loader

Hi s/n,.
I am facing problems when loading the dates in the database by using sql loader. My data file can have several date formats, so I have a function that interprets date and removes the time settings and the date of return.
Something like that.
For example:
I have a table-
CREATE TABLE TEMP1234
(
IDENTIFICATION NUMBER,
DATE OF ASOF_DATE
);
Data file
10001172 | 09/12/1945
Control file:
OPTIONS (DIRECT = TRUE, SILENT (FEEDBACK) =, skip = 0)
DATA RELATING TO SUNK COSTS
REPLACE
in the temp1234 table
' fields completed by "|" possibly provided by ' "'
TRAILING NULLCOLS
(ID,
ASOF_DATE "decode (: ASOF_DATE, null,:ASOF_DATE,conv_date1(:ASOF_DATE))).
)
Function CONV_DATE1:
FUNCTION to CREATE or REPLACE conv_date1 (p_str IN VARCHAR2)
DATE OF RETURN
IS
RETURN_VALUE DATE;
p_str1 VARCHAR2 (15): = NULL;
FmtArray TYPE IS a TABLE OF VARCHAR2 (30);
g_fmts fmtArray
: = fmtArray ("yyyy-mm-dd",
"yyyy/mm/dd"
mm/dd/yyyy"."
"dd-mm-yyyy",
"dd/mm/yyyy",
"mm-dd-yyyy");
BEGIN
p_str1: = SUBSTR (p_str, 1, 10);
BECAUSE me in 1... g_fmts. COUNTY
LOOP
BEGIN
return_value: = TO_DATE (p_str1, g_fmts (i));
EXIT;
EXCEPTION
WHILE OTHERS
THEN
NULL;
END;
END LOOP;
IF (return_value IS NULL)
THEN
RAISE PROGRAM_ERROR;
END IF;
RETURN Return_value;
END;
/
In this case, if the year in the data file shown in 1945, date which is load in the poster 2045 database.
But when I run this function by putting a sql editor - it returns the correct value.
Select double conv_date1('12/09/1945');
Please help me understand what is causing the problem.

I think there may be an implicit conversion going on with the combination of decode and checking nvl and function within SQL * Loader. I put the nvl checking in function and simply use the function in the control file, as shown below.

OPTIONS (DIRECT = TRUE, SILENT (FEEDBACK) =, skip = 0)

DATA RELATING TO SUNK COSTS

REPLACE

in the temp1234 table

' fields completed by "|" possibly provided by ' "'

TRAILING NULLCOLS

(ID,

asof_date ' conv_date1 (: asof_date).

)

FUNCTION to CREATE or REPLACE conv_date1 (p_str IN VARCHAR2)

DATE OF RETURN

IS

RETURN_VALUE DATE;

p_str1 VARCHAR2 (15): = NULL;

FmtArray TYPE IS a TABLE OF VARCHAR2 (30);

g_fmts fmtArray

: = fmtArray ("yyyy-mm-dd",

"yyyy/mm/dd"

mm/dd/yyyy"."

"dd-mm-yyyy",

"dd/mm/yyyy",

"mm-dd-yyyy");

BEGIN

IF p_str IS NULL

THEN RETURN NULL;

ON THE OTHER

p_str1: = SUBSTR (p_str, 1, 10);

BECAUSE me in 1... g_fmts. COUNTY

LOOP

BEGIN

return_value: = TO_DATE (p_str1, g_fmts (i));

EXIT;

EXCEPTION

WHILE OTHERS

THEN

NULL;

END;

END LOOP;

IF (return_value IS NULL)

THEN

RAISE PROGRAM_ERROR;

END IF;

RETURN Return_value;

END IF;

END;

/

implicit conversion to sql

Similar Questions

Maybe you are looking for