SQL & PL/SQL :: Reducing Function Computation When Called From Statement?
Nov 3, 2011
In the below SQL, I am calling a function "listagg_nm(deptno)", and function is to get all the employee name for a particular department with comma separated
output is:
-------------------------------------------------
DEPTNO SAL LISTAGG_NM(DEPTNO)
10 5000 KING,CLARK,MILLER
10 1300 KING,CLARK,MILLER
10 2450 KING,CLARK,MILLER
20 2975 JONES,SCOTT,FORD,SMITH,ADAMS
[code]....
below are the code
select deptno,sal,listagg_nm(deptno) from emp order by deptno;
FUNCTION listagg_nm (p_deptno NUMBER)
RETURN CLOB
IS
v_str CLOB;
[code]....
My doubt/question is for every row in the SQL the function is being called. t means that for 1st row deptno=10, the function is called and return all concatenated name for department 10, then again for 2nd row deptno=10, it again do the same, similarly for all other department.
Can we avoid this, I mean, if function is already returned a concatenated string of ename for deptno=10, then in the next function call it will checks if it is for same department (i.e. deptno=10) then the function will not go into for loop and it will directly return the concatenated string of ename for deptno=10 which we already done for the first row and so on for other department number ( something like storing the value in variable. Is it possible to achive this and does this is good for performance as we are not going into loop inside the function to return the concatenated string of ename for the same deptno everytime ( I have to return more than 10000 char., so i used clob in function and similarly the numbers of row return by SQL is more than 1000)and also how to code this, I tried but unable to achive.
is it possible to obtain a called procedure/function name within package?
For a measuring and tracing purpose, I would like to store an info at the beginning of each procedure/function in package with timestamp + additional details if needed.
For example:
CREATE OR REPLACE PACKAGE BODY "TEST_PACKAGE" IS PROCEDURE proc_1 IS BEGIN
[Code]....
I would like to replace "???????" with a function which would return a name of called procedure, so result of trace data after calling TEST_PACKAGE.proc_2 would be:
After much tedious mucking about in hyperspace, I'm back to tweaking my dataguard environment. I'm generating a lot of stored archive logs, this is causing problems with the backups, (since they live on the same drive, /u01 and also contain backups of the logs) I could move the logs onto /u02 but that's not dealing with the problem IMO.
This is the init script for the primary, (prod) CODE archive_lag_target=0 audit_file_dest='/u02/oradata/prod/dump/audit' compatible='10.2.0'
we have a tablespace of size 900 GB where 90% of space is occupied by two tables having BLOB data and now i need to drop these two tables and then to recover the space, i need to resize the tablespace (datafiles).
I have used computation in form .i have take 7 check box item with value
MON =1 TUES =2 WED =3 THU =5 FRI =6 SAT =6 SUN =7
these item are not database column.
Using computation after submit i have concatinate it and insert in to OPERATING_DAYS Column into table. example :if i check MON AND WED AND FRI THEN concatenate value insert into table 136
:P6_MON||:P6_TUS||:P6_WED||:P6_THR||:P6_FRI||:P6_SAT||:P6_SUNPloblem is when i edit record but this 136 value of operating days does not come into MON,WED and FRI.
How to extract these days value from OPERATING_DAYS Column into respective Days Item.
The data in one of the temporary table has been compiled as below. Number of Days elapsed between two transaction dates is required to be computed and multiplied with the balance. There can be multiple number of same transaction dates. When previous date and the current transaction dates are same the resulting number of day difference should be 0. But when they are different the difference between them is to be computed. On the last day of the Financial Year i.e. '31/03' of any year the difference day should be shown as 1 so as to make 365 or 366 days in a year. Simply deducting d1 from d2 on 31st will not be suffice as the difference is one day less.
I am trying to output some html in an Apex (v4.1.0.00.32) interactive report in a computation field but it just outputs the html..How can I put something like a br tag or a non breaking space {noformat} {noformat}
eg 'AllTo: ' || J || '<br>Dev: ' || KOk, code I want is in the image below
CREATE OR REPLACE FUNCTION is_overdue (due_date IN DATE,paid_date IN DATE) RETURN BOOLEAN IS days_between NUMBER (2) := due_date - paid_date; BEGIN [code]......
When i try using this compilation, i am getting error for this program:
SQL> begin 2 dbms_output.put_line(is_overdue('07-nov-1987','01-aug-2012')); 3 end; 4 / dbms_output.put_line(is_overdue('07-nov-1987','01-aug-2012')); * ERROR at line 2: ORA-06550: line 2, column 7: PLS-00306: wrong number or types of arguments in call to 'PUT_LINE' ORA-06550: line 2, column 7: PL/SQL: Statement ignored
To get output for this program? Show me how to compile.
When we create sql and some pieces of sql are implemented as oracle function. and we run this sql only once every day. I read article that function after first run located in cache. This part of cache ( with function ), is it really consume one oracle resources? Or it will be erased after while.
Can we call a function within decode statement. I am able to do the same for simple example function . But In my actual procedure it's giving the error message . Are there any restrictions to call function with in decode statement?
I am trying to use decode function in sql and inside decode function can I use select statement ?
here is my sql
select we.wf_entity_id, decode(object_type_id, 1, select audit_number from ea_audit_general where sys_audit_id=object_id 2,'test', object_type_id ) from wf_entity we where
[code]....
see this
decode(object_type_id, 1, select audit_number from ea_audit_general where sys_audit_id=object_id 2,'test', object_type_id )
I am trying to search a way to get the SQL statement that caused an exception withing an oracle function. I tried:
SELECT sql_text from v$session ses, v$sql sql where sql.sql_id = ses.prev_sql_id and ses.sid = sys_context('userenv','SID') AND ROWNUM = 1;
but this doesn't always return the last statement that the function has executed. if needed i can send the complete script for the function and its tables and stored procedures for testing.
I am getting a (PL/SQL: ORA-00903: invalid table name) compile error in a procedure using a merge statement. I have seen many examples using this technique and am at a loss as to why I can't compile.
the pipelined function is:
FUNCTION f_crcli_pipe(pi_source_data IN sys_refcursor, pi_limit_size IN PLS_INTEGER DEFAULT pkg_crcli_variables.c_cursor_limit_def) RETURN CRCLI_AA PIPELINED PARALLEL_ENABLE(PARTITION pi_source_data BY ANY) IS
[code].....
the error is pointing to the TABLE function in the USING clause of the merge statement.
I have an issue with rather complicated function.Basically it is using DBMS_SQL to execute a very long statement with many parameters (~6000 of them) and binding them with DBMS_SQL.BIND_VARIABLE. Variables are called :1,:2,...,:6000.
When this arguments set is too large - I am receiving error "ORA-00939: too many arguments for function".
Currently I am thinking about dividing the query into subqueries and executing them all with performance decrease.
A computation after submit pl/sql function process to trim off the first part of the string (CQ..) within the list manager values. Support for example the list manager contains values such as
The computation process should trim off the first part(CQ..) and should return the list manager value as SAMPLE1..TEST1SAMPLE2..TEST2SAMPLE1..TEST2 Oracle APEX 4.0.2 is the version and Oracle 10g r2 is the database.
WHILE EXISTS ( SELECT * FROM tblOrgChart WHERE fxOrgID = v_chrTempKeyDept ) LOOP v_intDept := CAST(v_chrTempKeyDept AS NUMBER) + 1 ; v_chrTempKeyDept := LPAD('',3 - LENGTH(CAST(v_intDept AS VARCHAR2)),'0') || CAST(v_intDept AS VARCHAR2) ; END LOOP;
Error: PLS-00204: function or pseudo-column 'EXISTS' may be used inside a SQL statement only
How to call a function with a row type return in an Oracle select statement.
For e.g. :
If I had this function with a rowtype return: ------------------------------ create function abc return xyz%rowtype is rec xyz%rowtype; begin select * into rec from xyz where col1 = n; return rec; end; -------------------------------- How could I use this in a select clause, as there is a multi column return by the function ?
I have a table called "Subjects" which lists subjects to match with notations in another table I have created a simple sequence (CREATE sequence subjectid) to created the subject id for the table. But I notice that if there is a skip in the date, the sequence increments automatically when I am not even using it. It even appears to be incrementing even when I am not doing any database activity.
This is not an issue of data integrity, because the values in the subject_id column do not need to be sequential, they just need to be unique. But it really has me curious. I created another table called "keep_track" to keep track of what is happening:
I have a procedure in the live environment , which operate in 2 mode.
1 regular load at 10 min 2 nightly
The code is same except they set the last loaded time for both in 2 dfferent rows, so that next time run will get incremental data.
I can trace this procedure on mode 1 that is is scheduled but unable to find how the 2nd one is getting trigger every night. there is no jobs corresponding to that. Is there any way i can find which method is trigger the procedure during night.
I am trying to select the owner of a certain object, only knowing the name of the object and the user calling it.
Problem is this object might exist in more than one schema with the same name, and I only need the one that is called by the active user, which himself/herself might have access to other schemas containing their own copy of the object.
Example:
SELECT OWNER INTO v_schema_name FROM ALL_OBJECTS WHERE OBJECT_NAME = p_object_name;
This works fine, until the object exists in more than one schema to which the current user has access.
A Java program would call a sequence of 5 PLSQL procedures, and they all need to participates the Tranasction defined in Java.
What do I do for the transactions inside PLSQL to ensure all 5 pariticpate in the Java transaction? I suppose I just do not specify any COMMIT in the PLSQL, but that probably would not work as PLSQL would COMMIT by default after exiting the proc?
Wanted to know whether it is possible to know a trigger is getting fired due to which all packages/procedures. If it has got fired for a particular transaction say multiple times, then wanted to get details of the packages/procedures which caused an event so that trigger got fired?
I am facing below Oracle error when ever I am changing the contents of the package within it. %ORA-06508: PL/SQL: could not find program unit being called:
The packages and the dependents are recopmiled successfully and there are no invalid objects.The problem is resolved only if the database is reastarted.
What might be the problem?I am using a PL/SQL collection of record datatype in this package.
