Help! Executing SQL from ASP

This is my Code..
Set OraSession = CreateObject("OracleInProcServer.XOraSession")
Set OraDatabase = OraSession.DbOpenDatabase(strDB, strID & "/" &
strPASS, 0)
OraDatabase.DbExecuteSQL("DELETE tmp_clean;")
It then runs a SQL statement which populates the table
tmp_clean. I don't really want to have to write a procedure to
do this. It seems not to like the semi colon, but with out it
it doesn't execute the SQL. Please Help.

It's okay, I worked it out. I hadn't done a COMMIT from my SQL
Plus session, so the table was still empty, and it musn't like
that.

Similar Messages

Executing sql from table with in plsql procedure

I wonder if anyone can help.
I have a table that stores sql statements. They are actually a set of rules that I want to apply to my application. My intention is to execute the statement in plsql.
EG a sql statment held in the table might be:
'select count(id) from item where item_id = ' || constants_pkg.c_test_item || ' and client_id = ' || client_id_in
I am trying to call this from a plsql procedure, Ive CUT OUT some of the procedure task_rec.rule holds the statement above. The procedure is failing invalid SQL statement, I think the client_id_in may not be evaluating as the variable, passed into the procedure and constants_pkg.c_test_item not being taken from the constants pkg.
EG.
PROCEDURE create_additional_info_list(
client_id_in IN client.id%TYPE,
IS
result VARCHAR2(100);
BEGIN
EXECUTE IMMEDIATE
task_rec.rule
INTO result;
END;
If I was to do below the code would work ok. However I require the rule to come from the table.
EXECUTE IMMEDIATE
'select count(id) from item where item_id = ' || constants_pkg.c_test_item || ' and client_id = ' || client_id_in
INTO result;
Is there anyway constants_pkg.c_test_item and client_id_in can be evaluated as variables?

Not that I've tried this but, if possible, change the SQL in the table to have bind variable; e.g.,
'select count(id) from item where item_id = :1 and client_id = :2 ';
Then in the procedure
execute immediate task
INTO result
USING constants_pkg.c_test_item ,
client_id_in ;

Solution to execute sql from mysql application logs

We have some custom mysql application, which generate logs that are stored on common location on open linux box which has below format
$ cat minor20100809.log
20100809001^Aselect count(prod_id) from product_logs;
20100809002^Aselect count(order_id) from order_logs;
ID and the SQL statement with control A as seperater and daily 1000 logs files are generated
Now we have one oracle DWH, where we import all the mysql data and we need to test these counts on daily basis.
Could you please suggest a solution in oracle for this?
at high level we need to follow below steps
a) Import these logs in a table --> which is better way sqlldr, impdp, or plsql read file ?
b) execute these sql statement one by one
c) store the counts in log table on daily basis

Hi,
It is a simple two step process.
Step 1. Convert your log files statements to meaningful oracle statements. like removing prefixes 20100809001^A from each line.
This can be done in two ways.
(a). Using OS editors OR
(b). Following PL/SQL code...
-- Connect as sysdba
create or replace directory dir_tmp as '/tmp';                        -- If oracle server is on Unix
create or replace directory dir_tmp as 'C:\your_folder';            -- If oracle server is on Windows
-- Grant permission to your_schema
grant read, write on directory dir_tmp to your_schema;    Removing prefixes using:
       SELECT regexp_replace (vNewLine, '([[:alnum:]]{1,})\^A', '\2')
         INTO vStmt
         FROM DUAL;
CREATE OR REPLACE PROCEDURE convert_to_sql_stmts
( input_file     in varchar2,
output_file   in varchar2
AUTHID CURRENT_USER
IS
    InFile   utl_file.file_type;
    OutFile utl_file.file_type;
    vNewLine VARCHAR2(4000);
    vStmt    VARCHAR2(4000);
    i        PLS_INTEGER;
    j        PLS_INTEGER := 0;
    SeekFlag BOOLEAN := TRUE;
BEGIN
-- open a file to read
InFile := utl_file.fopen(dir_tmp, input_file,'r');
-- open a file to write
OutFile := utl_file.fopen(dir_tmp, output_file, 'w');
-- if the file to read was successfully opened
IF utl_file.is_open(InFile) THEN
    -- loop through each line in the file
    LOOP
      BEGIN
        utl_file.get_line(InFile, vNewLine);
        i := utl_file.fgetpos(InFile);
        dbms_output.put_line(TO_CHAR(i));
        SELECT regexp_replace (vNewLine, '([[:alnum:]]{1,})\^A', '\2')
          INTO vStmt
          FROM DUAL;
        utl_file.put_line(OutFile, vStmt, FALSE);
        utl_file.fflush(OutFile);
        IF SeekFlag = TRUE THEN
          utl_file.fseek(InFile, NULL, -30);
          SeekFlag := FALSE;
        END IF;
      EXCEPTION
        WHEN NO_DATA_FOUND THEN
          EXIT;
      END;
    END LOOP;
    COMMIT;
END IF;
utl_file.fclose(InFile);
utl_file.fclose(OutFile);
EXCEPTION
WHEN OTHERS THEN
    RAISE_APPLICATION_ERROR (-20099, 'Unknown UTL_FILE Error');
END convert_to_sql_stmts;
Step 2. Directly executing above output log file
Use spool to generate the count....
    set termout off
    set feedback off
    set trimspool on
    set pagesize 0
    set timing OFF
    spool   count.log
    @new_file.log
    spool off;Hope this helps...

Executing SQL From A Forms Procedure

I am looking to execute a composed SQL command (ALTER USER ***** IDENTIFIED BY ****). The old execute immediate command in forms 5 would be the one I would have used, does any know the Forms 6i equivalent.
Regards
Stephen

Hi John,
What I was having a problem with is that the following code from a Forms 5 procedure isn't working with Forms 6i :
Procedure dropSynonyms ()
sql_stmt carchar2(2000);
BEGIN
sql_stmt := 'drop synonym RT_'| |rec.runtime_table_name;
execute immediate sql_stmt;
END;
The method of executing the sql_stmt sees to have changed with Forms 6i but the help documentation isn't clear on the new syntax.
Any help appreciated.
Regards
Stephen

Urg:Executing SQL From Java App using Jdbc Driver

Hi
I am using JDBC Driver version 9.0.1.1.0.
I am running this in Thin Client Mode
My Code Snippet Looks Like This:=
==========================================================
String url = "jdbc:oracle:thin:@localhost:1521:VpDb";
String userName = "scott";
String password = "tiger";
Connection conn = null ;
Statement st = null ;
ResultSet rs = null ;
String categoryCode="ABC";
try
conn = DriverManager.getConnection (url, userName, password);
st = conn.createStatement ();
String sqlStatement = "Select Count(*) From News Where CategoryCode=" +"\'" + categoryCode + "\'" + ";";
System.out.println(sqlStatement);
rs = st.executeQuery ( sqlStatement );
if( rs.next() )
System.out.println("Headline Exists");
else
System.out.println("No Headline Exists");
catch (SQLException e )
System.out.println();
System.out.println (" SQL ERROR IN IS_NEWS_EMPTY: " + e) ;
=========================================================
I have added the classes12.zip and nls_charset12.zip in the classpath.
Now when i run this it gives me an error saying the following error message:-
SQL ERROR IN IS_NEWS_EMPTY: java.sql.SQLException: ORA-00911: invalid character
Can anyone help me with this as to whats going wrong because the exact equivalent of my sqlStamenet runs on SQL command line but not from java code.Why??
Any Help appreciated
Thanks
Mum

I think it is complaining about the ";" that you add at the end of your string. You don't need to add that to the query.

How to execute sql from table

For a migration I have created a sql script that puts the full statement to be executed in a specific table column
I'm now looking for a way how to execute those statements.
Table structure like: oldName, newName, sqlStatement, id, otherField, otherField2
Instead of the full statement(sqlStatement) it would also be possible to use the oldName, newName and id field if that simplifies the problem.
The reason for building the statement and running that one is a) to validate upfront b) a self join is applied to the query result table to prevent some items for update.
The initial idea was to take the sql statements and execute them as a sql script but that process is pretty slow.

user2833655 wrote:
For a migration I have created a sql script that puts the full statement to be executed in a specific table column
I'm now looking for a way how to execute those statements.That is wrong, tables are for data not executable code
Re: ref cursor - result set not known
http://en.wikipedia.org/wiki/Data_%28computing%29
>
Data is often distinguished from programs. A program is a sequence of instructions that detail a task for the computer to perform. In this sense, data is thus everything that is not program code.
>
user2833655 wrote:
The initial idea was to take the sql statements and execute them as a sql script but that process is pretty slow.This alternative will just make it difficult, complicated and unreliable as well.

After executing sql from batch, its disconnecting

Hello All,
I have written a batch program to execute a pl/sql procedure. But after execution, it disconnects.
But i want to continue in the same session
echo exec tracer.trace_pck.start_('alter session set sql_trace=true') | sqlplus hr/hr
Output:
Press any key to Enable Tracing
SQL*Plus: Release 11.2.0.2.0 Production on Thu Oct 10 15:59:56 2013
Copyright (c) 1982, 2010, Oracle. All rights reserved.
Connected to:
Oracle Database 11g Express Edition Release 11.2.0.2.0 - Production
SQL>
PL/SQL procedure successfully completed.
SQL> Disconnected from Oracle Database 11g Express Edition Release 11.2.0.2.0 -
Production
Trace Enabled Successfully. Press any key to input you sql

If you are using a batch file I assume you are using the windows OS.
Essentially what you need to do is to call a sql file from the batch file and then get the sql file to call all the other files.
Batchfile test.bat calls sql file test.sql
sqlplus <username>/password @test.sql
test.sql calls test1.sql and test2.sql and then exits
@test1.sql;
@test2.sql;
exit;
test1.sql shows the current sid, and serial# fro our session and our filename
select 'test1',sid,serial# from v$session
where audsid=userenv('sessionid');
test2.sql shows the current sid, and serial# for our session and our filename
select 'test2',sid,serial# from v$session
where audsid=userenv('sessionid');
When you run this you get t output similar to the following: (Your SID and Serial# will be different )
'TEST        SID    SERIAL#
test1        771         97
'TEST        SID    SERIAL#
test2        771         97
This shows that two separate sql files were called but are still in the same session

Need Help Duplicating SQL from "Administer Concurrent Managers" Form

I am trying to duplication the "Administer Concurrent Managers" form. This is what I have but it does it is not correct. I am having problems with the "Processes Actual" values. In my SQL I was thinking it was the MAX_PROCESSES but that does not seem to be it.
select
q.CONCURRENT_QUEUE_ID
,q.CONCURRENT_QUEUE_NAME
,q.USER_CONCURRENT_QUEUE_NAME
,q.MAX_PROCESSES
,q.RUNNING_PROCESSES
,running.total running
from (select /*+ ORDERED */
count(*) as total
,prc.CONCURRENT_QUEUE_ID
from apps.fnd_concurrent_processes prc
,apps.FND_CONCURRENT_REQUESTS req
where req.phase_code='R'
and req.controlling_manager = prc.concurrent_process_id
group by prc.CONCURRENT_QUEUE_ID) running
,apps.fnd_concurrent_queues_vl q
where q.CONCURRENT_QUEUE_ID = running.CONCURRENT_QUEUE_ID(+)
and q.MAX_PROCESSES > 0
order by
DECODE(q.application_id,0,DECODE(q.CONCURRENT_QUEUE_ID,1,1,4,2))
,sign(q.max_processes) desc
,q.CONCURRENT_QUEUE_NAME
,q.application_id

Thanks Hussein;
I am on 11.5.10.2
I have looked at the table FND_CONCURRENT_QUEUES and it seems to show the same problem as my SQL. What I have happening right now is my CCM's have crashed in a test environment. The "Administer Concurrent Mnagers" form show these values:
Name = Standard Manager
Node = MyNode
Processes Actual = 0
Processes Actual = 5 (The correct number of defined managers)
When I look at the columns in my SQL and the table you provided me I see these values:
CONCURRENT_QUEUE_NAME = STANDARD
TARGET_NODE = MyNode
MAX_PROCESSES = 5
RUNNING_PROCESSES =5
I do not see where the "Processes Actual = 0" is coming from.

Need help in Report From SQL Query

Hi All,
I am facing a problem with a report. I need your help.
I am creating a Report From SQL Query (Portal) with some arguments passed at runtime. I am able to view the output, if the query returns few rows ( arount 1000 rows). But for some inputs it needs to generate >15000 records, at this point the page is getting time out (i think!) and showing error page. I am able to execute query from the SQL Plus console ot using TOAD editor. Here the query is not taking more that 2 mins time to show the result.
If i am executing from Portal i observed that, once i give the appropriate input and hit submit button a new oracle process is getting created for the query on UNIX (I am usign "TOP" command to check processes). The browser page will be shown error page after 5 minutes (i am assuming session time out!) , but on the backend the process will be executed for more than 30 mins.
I tried also increase the page time out in httpd.conf, but no use.
The data returned as a result of the query is sized more than 10 MB. Is caching this much data is possible by the browser page? is the returned data is creating any problem here.
Please help me to find appropriate reasone for the failure?

user602513 wrote:
Hi All,
I am facing a problem with a report. I need your help.
I am creating a Report From SQL Query (Portal) with some arguments passed at runtime. I am able to view the output, if the query returns few rows ( arount 1000 rows). But for some inputs it needs to generate >15000 records, at this point the page is getting time out (i think!) and showing error page. I am able to execute query from the SQL Plus console ot using TOAD editor. Here the query is not taking more that 2 mins time to show the result.
If i am executing from Portal i observed that, once i give the appropriate input and hit submit button a new oracle process is getting created for the query on UNIX (I am usign "TOP" command to check processes). The browser page will be shown error page after 5 minutes (i am assuming session time out!) , but on the backend the process will be executed for more than 30 mins.
I tried also increase the page time out in httpd.conf, but no use.
The data returned as a result of the query is sized more than 10 MB. Is caching this much data is possible by the browser page? is the returned data is creating any problem here.
Please help me to find appropriate reasone for the failure?Do you get any errors or warnings or it is just the slow speed which is the issue?
There could be a variety of reasons for the delayed processing of this report. That includes parameter settings for that page, cache settings, network configurations, etc.
- explore best optimization for your query;
- evaluate portal for best performance configuration; you may follow this note (Doc ID: *438794.1* ) for ideas;
- third: for that particular page carrying that report, you can use caching wisely. browser cache is neither decent for large files, nor practical. instead, explore the page cache settings that portal provides.
- also look for various log files (application.log and apache logs) if you are getting any warnings reflecting on some kind of processing halt.
- and last but not the least: if you happen to bring up a portal report with more than 10000 rows for display then think about the usage of the report. Evaluate whether that report is good/useful for anything?
HTH
AMN

HELP: SSIS 2012 - Execute SQL Task with resultset to populate user variables produces DBNull error

I am experiencing an unexplainable behavior with the Execute SQL Task control in SSIS 2012. The following is a description of how to simulate
the issue I will attempt to describe:
1. Create a package and add two variables User::varTest1 and User::varTest2 both of String type with default value of "Select GetDate()"
for each, have shown this to not matter as the same behavior occurs when using a fixed string value or empty string value.
2. Add a new Execute SQL Task to the control flow.
3. Configure an OLE DB connection.
4. Set SQLSourceType = "Direct Input"
5. Set the ResultSet property of the task to "Single Row"
6. In the ResultSet tab add two results as follows:
Result Name: returnvalue1, Variable Name: User::varTest1
Result Name: returnvalue2, Variable Name: User::varTest2
7. Set an expression for the SqlStatementSource property with a string value of "Select 'Test' returnvalue1, 'Testing' returnvalue2'"
The idea is that the source would be dynamically set in order to run a t-sql statement which would have dynamic values for database name
or object that would be created at runtime and then executed to set the user variable values from its resultset. Instead what occurs is that a DBNull error occurs.
I am not sure if anyone else has experienced this behavior performing similiar actions with the Execute SQL Task or not. Any help would be
appreciated. The exact message is as follows:
[Execute SQL Task] Error: An error occurred while assigning a value to variable "varRestoreScript": "The type of the value
(DBNull) being assigned to variable "User::varRestoreScript" differs from the current variable type (String). Variables may not change type during execution. Variable types are strict, except for variables of type Object.
User::varRestoreScript is the first return value. And even with the a dummy select the same result occurs. I have narrowed the issue down
to the T-SQL Statement structure itself.
The following works just fine within the execute sql task control as long as no resultset is configured for return:
"Declare @dynamicSQL nvarchar(max)
Select @dynamicSQL = name
From sys.databases
Where name = 'master'
Select atest_var1=@dynamicSQL, atest_var2='static'"
I have tried various iterations of the script above with no success. However, if I use the following derivative of it the task completes
successfully and variables are set as expected. This will not work for my scenario however as I need to dynamically build a string spanning multiple resultsets to build one of the variable values.
"Select atest_var1=name, atest_var2='static'
From sys.databases
Where name = 'master'
I have a sample package which can reproduce this issue using the above code. You can get to that through the post on www.sqlservercentral.com/Forums/Topic1582670-364-1.aspx
Scott

Arthur, the query when executed doesn't return a null value for the @dynamicSQL variable. It returns "master" as a string value. Even the following fails which implements that suggestion:
Declare @dynamicSQL as nvarchar(max)
Select @dynamicSQL = name
From master.sys.databases
Where name = 'master' -- (or any other DB name)
I believe I have found the cause of the issue. It is datatype and size related. The above script will properly set the variables in the resultset as long as you don't use nvarchar(max) or varchar(max). This makes the maximum data size for
a String variable 4000 characters whether unicode or non-unicode typed.

How to export the result from executing sql statement to excel file ?

HI all,
Great with Oracle SQL Developer, but I have have a trouble as follwing :
I want to export the result from executing sql statement to excel file . I do easily like that in TOAD ,
anyone can help me to do that ? Thanks so much
Sigmasvn

Hello Sue,
I just tried to export to excel with the esdev extension and got java.lang.NumberFormatException. I found the workaround at Re: Windows Multi-language env, - how do I set English for application lang?
open the file sqldeveloper\jdev\bin\sqldeveloper.conf and add the following two lines:
AddVMOption -Duser.language=en
AddVMOption -Duser.country=USyet now my date formats in excel are 'american-style' instead of german. For example 01-DEC-01 so excel does not recognize it as date and therefore I can not simply change the format.
When export to excel will be native to 1.1 perhaps someone can have a look at this 'feature'
Regards
Marcus

Executing sql query(not through sqlfile) from batchfile

Hi,
I have written a batchfile below which should execute the sql query. Can someone please help me in modifying the below code. I know we can make use of sql file ,but wanted it to be more robust and hence everything in a single batch file instead of having combination of sqlfile and batchfile.
@echo off
set ORACLE_HOME=%RAMBOH%
set ORACLE_SID=%RAMBSID%
set path=%RAMBOH%\bin;%RAMBOH%\admin\Ramb_Scripts;%path%
sqlplus sys/XXX as sysdba
@echo SET SERVEROUTPUT ON;
@echo DECLARE
@echo     v1   DATE;
@echo BEGIN
@echo     SELECT SYSDATE INTO V1 FROM DUAL;
@echo     DBMS_OUTPUT.PUT_LINE(v1);
@echo END;
@echo /

I used the below syntax to execute a procedure
set ORACLE_SID=orcl
echo exec MyProcedure;
echo exit
)|sqlplus system/manager >MyProcedure.log
Try making modifications to this and see if this helps you in executing the SQL from the batch file

How do I return two values from a stored procedure into an "Execute SQL Task" within SQL Server 2008 R2

Hi,
How do I return two values from a
stored procedure into an "Execute SQL Task" please? Each of these two values need to be populated into an SSIS variable for later processing, e.g. StartDate and EndDate.
Thinking about stored procedure output parameters for example. Is there anything special I need to bear in mind to ensure that the SSIS variables are populated with the updated stored procedure output parameter values?
Something like ?
CREATE PROCEDURE [etl].[ConvertPeriodToStartAndEndDate]
@intPeriod INT,
@strPeriod_Length NVARCHAR(1),
@dtStart NVARCHAR(8) OUTPUT,
@dtEnd NVARCHAR(8) OUTPUT
AS
then within the SSIS component; -
Kind Regards,
Kieran.
Kieran Patrick Wood http://www.innovativebusinessintelligence.com http://uk.linkedin.com/in/kieranpatrickwood http://kieranwood.wordpress.com/

Below execute statement should work along the parameter mapping which you have provided. Also try specifying the parameter size property as default.
Exec [etl].[ConvertPeriodToStartAndEndDate] ?,?,? output, ? output
Add a script task to check ssis variables values using,
Msgbox(Dts.Variables("User::strExtractStartDate").Value)
Do not forget to add the property "readOnlyVariables" as strExtractStartDate variable to check for only one variable.
Regards, RSingh

Why is my Package just stopping when going from one Execute SQL Task to another Execute SQL Task

I have one flow constraint going for my one Execute SQL Task to another Execute SQL Task. It's executing #1 and successfully and just stops there rather than continuing on to Execute SQL Task #2. Why would it just stop? My constraint is NOT conditional...just
a straight ole flow constraint.
Any help is greatly appreciated and Thanks in Advance for your help.
PSULionRP

Hi PSULionRP,
First make sure the color of the Precedence Constraint is green (Success) or blue (Completion) other than red (Failure). Then, make sure the Execute SQL Task is not disabled (right click the executable, and “Disable” option is available in the right-click
menu).
Regards,
Mike Yin
TechNet Community Support

Execute SQL Task does not Update from a Date Variable Reliably

I'm using a DateTime variable in SSIS 2008 that is used to set the SQLStatement property of an Execute SQL Task.
"DELETE FROM Labor WHERE Week = '" + (DT_WSTR, 100) @[User::Week] + "'"
Week is the next Sunday:
DATEADD( "day", @[User::DaysTillSunday] , @[User::TheDayThatIsTwentyMinutesPrior] )
DaysTillSunday:
DATEPART( "dw", @[User::TheDayThatIsTwentyMinutesPrior] ) == 1 ? 0 : 8 - DATEPART( "dw", @[User::TheDayThatIsTwentyMinutesPrior] )
TheDayThatIsTwentyMinutesPrior:
(DT_DATE)(DT_DBDATE)DATEADD("minute",-20,GETDATE())
The SSIS Package deletes the current week's data, reloads it with fresh data, then calculates the difference between the current week and last week.
The problem is that randomly, instead of deleting the current week, it will delete the previous week. This happens maybe 5-10% of the time. At least it does until I rebuild the package and import it into SQL Server again.
I'm guessing that the Execute SQL Task is not updating the value of the Week variable before it executes. I started with the source type being a variable. Then I decided to try Direct input and pass in the Week as a parameter (OLE DB Connection
Type). That didn't work either.
Most recently I tried writing the Week variable to a table first, then having a sequence container with all the tasks second. Slightly better but I still saw the date was wrong 2 times in about 90 executions. I was hoping that writing the Week
variable out to the database would force an update of any associated connections to it, but that didn't seem to work.
Any ideas? Is this a known issue, am I missing a setting?
thanks,
John

John, computers either work all the time or have a bug. I suspect it is the latter.
To find it [faster] you need to log what the resulting expression was used in the package.
I am baffled how rebuilding a package would fix anything like setting a date.
It might be even dependant on when you run the package.
Why
DATEADD("minute",-20,GETDATE())
DATEADD( "day", -8 , GETDATE() )
It must be enough to set the week (that appears to be a date) as above.
Arthur
MyBlog
Twitter

Help! Executing SQL from ASP

Similar Messages

Maybe you are looking for