Similar Questions

• Dynamic problem with lookup-query for the purpose of resource request

  Hi all
  I need to set up several IOM user exchange mailboxes, I can set up AD account and account of the mailbox without any problem, but only for the first ad and Exchange account. For the second and third, etc. I get the error: "Invalid login name" during Exchange of account provisioning. I discovered that this problem exists with Exchange Connector - it is not able to collect correct GUID. So in my xml Dataset I use dynamic query Lookup to select manually correct Alias, the login name and GUID. The query for the GUID is the sequel (I cloned RO for AD and Exchnage):
  
  AttributeReference available-in-bulk = "true" length = "32" widget = 'search query' type = 'String' attr-ref = "Object GUID" name = "Object GUID" >
  * < lookupQuery search-query = "select distinct UD_KFUSER_OBJECTGUID GUID, ud_KFUSER_uid like UD_KFUSER UD_KFUSER, orc orc Login, sta sta where UD_KFUSER.orc_key = orc.orc_key and orc.usr_key = ' $Form data." Take ' and UD_KFUSER. "UD_KFUSER_AD = 27 and orc.orc_status = STA.sta_status AND STA.sta_bucket! = 'Cancelled'" display-field = "GUID" save-field = "Object GUID" / > *.
  * < / AttributeReference > *.
  
  
  My questions are:
  
  1. I have to type * to run the query in the user interface, without * I got error:
  
  + < 17 February 2012 11:12:22 THIS > < error > < oracle.adfinternal.view.faces.config.ric +.
  h.RegistrationConfigurator > < BEA-000000 > < ADF_FACES - 60096:Server Exception durin
  PPR, #10 g
  oracle.iam.platform.canonic.base.NoteException: an error occurred during executin
  g the search query.
  to oracle.iam.platform.canonic.agentry.GenericEntityLookupActor.perform)
  GenericEntityLookupActor.java:337)
  Is this right?
  
  2. when I got correct values (from the search query) - they are missing on the details of the application and form of RO - what Miss me?
  
  I use OIM 11.1.1.5, in my xml dataset I use correct attr-Ref (labels), when I type the values manually, they are propagated to form RO and Exchange mailbox is created.
  Best
  MP

  I not had no problem when writing search query.
  This works very well for me.
  The request will be filled for the field, so why choose *?
  I used as
  

• What is the problem with this query in the source

  A blank page is region 2. Region1: enter rec_no (-rec_no is varchar2) region2:hide and show that displays one record based on rec_ no region1 entered.
  The problem is that I get the following error
  
  ORA-06550: line 1, column 7: PLS-00428: an INTO clause in this SELECT statement
  
  Here's my simple query in a process of pl/sql
  Start
  Select rec_no, rec_name from mytable where rec_no = upper(:P16_rec_no);
  end;
  
  Why? The same query works under the sql command.
  
  what I'm trying to accompolish here, it is because he has given huge if I want to display hide also multiples, regions and allow users to update each region as a result. Help, please. It kills me that I just can't understand what is causing the problem to the query. Thank you.

  Hello

  The list is set to the field EMPLOYEE_ID itself? What is the primary key of the table? In my example, DEPTNO is the primary key and I used the element generated by the wizard page to display the list. The parameters of the Source of the item itself or the column DEPTNO and database so that the process "process line...". "to identify the updated row.

  Andy

• Confusion between the performance problem with a query

  I have a query that takes 3 minutes (according to another developer, but when I run it not even returned after 10 minutes) to return lines nil. But the interesting part is that instead of select * if I give a count (*) select, he returned in less than a second. And even the subquery returns no rows in less than one second. The query is as follows. I see no option to answer this confusing behavior. Please suggest me an option I can verify, what explains this scenario. I am interested to know the root cause of it. Please let me know if you need any details.
  

  select * from order_lines_flmnt_stg_in ol
   WHERE order_lines_flmnt_stg_in_id IN (
            SELECT DISTINCT order_lines_flmnt_stg_in_id
                       FROM order_lines_flmnt_stg_in ol,
                            order_header_flmnt_stg_in oh,
                            order_lines oln
                      WHERE ol.TYPE = 'Original'
                        AND ol.item_code NOT IN (
                               SELECT oln.item_code
                                 FROM order_lines oln,
                                      order_lines_flmnt_stg_in ol,
                                      order_header_flmnt_stg_in oh
                                WHERE oh.po_id = ol.po_id
                                  AND ol.item_code = oln.item_code)
                        AND (    ol.error_message IS NULL
                             AND oh.error_message IS NULL
                             AND oh.activate_flag = 1
                            ));

  SQL*Plus: Release 10.2.0.1.0 - Production on Wed Sep 30 16:41:53 2009
  
  Copyright (c) 1982, 2005, Oracle.  All rights reserved.
  
  
  Connected to:
  Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - 64bit Production
  With the Partitioning, Real Application Clusters, OLAP and Data Mining options
  
  SQL> set lines 10000
  SQL> alter session set current_schema = epix;
  
  Session altered.
  
  SQL> set timing on
  SQL>           SELECT DISTINCT order_lines_flmnt_stg_in_id
    2                       FROM order_lines_flmnt_stg_in ol,
    3                            order_header_flmnt_stg_in oh,
    4                            order_lines oln
    5                      WHERE ol.TYPE = 'Original'
    6                        AND ol.item_code NOT IN (
    7                               SELECT oln.item_code
    8                                 FROM order_lines oln,
    9                                      order_lines_flmnt_stg_in ol,
   10                                      order_header_flmnt_stg_in oh
   11                                WHERE oh.po_id = ol.po_id
   12                                  AND ol.item_code = oln.item_code)
   13                        AND (    ol.error_message IS NULL
   14                             AND oh.error_message IS NULL
   15                             AND oh.activate_flag = 1
   16                            );
  
  no rows selected
  
  Elapsed: 00:00:00.11
  SQL> select count(*) from order_lines_flmnt_stg_in ol
    2   WHERE order_lines_flmnt_stg_in_id IN (
    3            SELECT DISTINCT order_lines_flmnt_stg_in_id
    4                       FROM order_lines_flmnt_stg_in ol,
    5                            order_header_flmnt_stg_in oh,
    6                            order_lines oln
    7                      WHERE ol.TYPE = 'Original'
    8                        AND ol.item_code NOT IN (
    9                               SELECT oln.item_code
   10                                 FROM order_lines oln,
   11                                      order_lines_flmnt_stg_in ol,
   12                                      order_header_flmnt_stg_in oh
   13                                WHERE oh.po_id = ol.po_id
   14                                  AND ol.item_code = oln.item_code)
   15                        AND (    ol.error_message IS NULL
   16                             AND oh.error_message IS NULL
   17                             AND oh.activate_flag = 1
   18                            ));
  
    COUNT(*)
  ----------
           0
  
  Elapsed: 00:00:00.18
  SQL> select count(*) from order_lines_flmnt_stg_in;
  
    COUNT(*)
  ----------
        3198
  
  Elapsed: 00:00:00.18
  SQL> 

  I don't know if it's relevant, but there are no locks on the tables involved.
  
  Published by: release 6' - 8 "September 30, 2009 16:49

  count (*) SELECT can use the index only without having to consult the table at all...
  Amiel

• PROBLEM WITH SQL QUERY.

  Hi all
  
  I am executing the following query, which I use to retrieve unique records.
  
  
  SELECT OH. ORDER_ID, OH. ORDER_DATE, OH. GRAND_TOTAL, OSI. DESCRIPTION ACE ORDER_STATUS,
  PMT. DESCRIPTION AS PAYMENT_METHOD, TR. AS PAYMENT_STATUS DESCRIPTION
  OF ORDER_HEADER OH INNER JOIN ORDER_PAYMENT_PREFERENCE OPP ON OH. ORDER_ID = OPP. ORDER_ID and
  OPP. CREATED_DATE = (SELECT MAX (CREATED_DATE) OF ORDER_PAYMENT_PREFERENCE WHERE THE ORDER_ID = OH. ORDER_ID)
  PAYMENT_METHOD_TYPE PMT, STATUS_ITEM SO, STATUS_ITEM OSI
  WHERE OPP. PAYMENT_METHOD_TYPE_ID = PMT. PAYMENT_METHOD_TYPE_ID AND
  OPP. BATCH = IF. BATCH AND OH. BATCH = OSI. BATCH AND
  (OH. BATCH = "ORDER_CREATED" | OH. BATCH = "ORDER_HOLD") AND
  OH. "CREATED_STAMP > = ' 2011-12-03 11:41:04.
  
  Problem I am facing is, for some a few order_ids we have several entries in ORDER_PAYMENT_PREFERENCE with CREATED_DATE even where query return me of duplicate records.
  I applied OPP. "CREATED_DATE = (SELECT MAX (CREATED_DATE) OF ORDER_PAYMENT_PREFERENCE WHERE THE ORDER_ID = OH. ORDER_ID)' forced to get the record, unique, but as I said to few recordings, we have several lines in ORDER_PAYMENT_PREFERENCE corr. ORDER_ID single and multiple records have same CREATED_DATE.
  
  Can you suggest a few changes, the query to retrieve unique records.
  
  I know I'm a very abstract information, shortly I'll provide table structure and little info I want in this query.
  It's because of the tight work schedule.
  
  
  Thank you
  Jagdeep Singh

  Hello

  Basically, you need to convert a number in his order inside each created_date order_payment_preference. However, if you have more than one created_date which is him even you'll actually choose one of them at random unless you specify an order more squeezed by the clause. For example, you might have a numeric value from a sequence that could be added to determine which input was really the most recent. In any case the following should give you an idea what I mean...

  SELECT
        order_id,
        order_date,
        grand_total,
        order_status,
        payment_method,
        payment_status
  FROM
      (
          SELECT oh.order_id,
                 oh.order_date,
                 oh.grand_total,
                 osi.description AS order_status,
                 pmt.description AS payment_method,
                 si.description AS payment_status
                 ROW_NUMBER() OVER(PARTITION BY opp.order_id ORDER BY opp.created_date DESC) rn
          FROM   order_header oh
                 JOIN order_payment_preference opp ON oh.order_id = opp.order_id
                 JOIN payment_method_type pmt ON opp.payment_method_type_id = pmt.payment_method_type_id
                 JOIN status_item si ON opp.status_id = si.status_id
                 JOIN status_item osi ON oh.status_id = osi.status_id
          WHERE
          AND    oh.status_id IN('ORDER_CREATED','ORDER_HOLD')
          AND    oh.created_stamp >= '2011-12-03 11:41:04'
      )
  WHERE
       rn = 1    
  

  Some notes if - it is not a good idea to mix the styles of the junction between ansi and oracle style, it makes it difficult to follow. Just choose one or the other.

  OH.STATUS_ID='ORDER_CREATED' || OH.STATUS_ID='ORDER_HOLD'
  

  Makes no sense. I took a guess that you meant, OR then I converted to an IN clause that will do the same job.

  HTH

  David

• problem with a query

  Hello experts,
  
  I know that is not the exactly the right forum but I'm having some trouble in a query
  

  select retailer_num, retailer_name 
  from lotto_retailers 
  where agent_status_id = 2
  and salesrepid= :criteria.rep_id
  UNION
  SELECT '0', '**ALL**'
  FROM lotto_retailers 
  where agent_status_id = 2
  and salesrepid= :criteria.rep_id
  ORDER BY 1

  I created a text field in the forms of oracle with a list of values
  
  I want to have an option "ALL" where it will select all retailer_num based on my salesrep_id condition.this does not seem to work.
  can someone help please
  
  Thank you

  You look at something like this:

  select retailer_num, retailer_name
  from lotto_retailers
  where agent_status_id = 2
  and salesrepid= (CASE WHEN upper(:criteria.rep_id) = 'ALL' then salesrepid ELSE :criteria.rep_id END)
  ORDER BY 1
  

• Problem with a query of xmlt

  Hello
  
  I need to query this xml file:
  
  < a > < action id = "1" > < alarm id = "1000" > 3 < / alarm > < alarm id = '1001' > 2 < / alarm > < alarm id = "31" > 2 < / alarm > < / action > < Code action = "3" > < alarm id = "16" > 5 < / alarm > < / has >
  
  I need to add the action/@id attribute to this query:
  

  select aa.hour,
         to_number(extract(value(xa),'alarm/@id')) alm_,
         to_number(extract(value(xa),'alarm/text()')) freq_
    from t_alarm_stats aa,
         xmltable('$XML/a/action/alarm' passing aa.x_alarm AS "XML") xa
   where aa.id_device = 3194
     and aa.hour > trunc(sysdate-1);

  I have now the recordset is fine enough:
  
  time, id_alarm and frequency of the alarm:
  28/03/2011 01:00 1000 3
  28/03/2011 01:00 1001 2
  28/03/2011 01:00 31 2
  28/03/2011 01:00 16 5
  
  I need to also add the corresponding action/@id, but I don't know how to navigate to a higher level of my xml so that the Recordset:
  
  time, action id, id_alarm and frequency of the alarm:
  28/03/2011 01:00 1 1000 3
  28/03/2011 01:00 1 1001 2
  28/03/2011 01:00 1 31 2
  28/03/2011 01:00 3 16 5
  
  Y at - it the possibility to do this with 1 query?
  
  Thanks in advance,
  Samuel

  Hello

  This should do it:

  select aa.hour
       , xa1.action_id
       , xa2.alarm_id
       , xa2.alarm_freq
  from t_alarm_stats aa,
       xmltable('$XML/a/action'
         passing aa.x_alarm AS "XML"
         columns action_id number  path '@id',
                 alarms    xmltype path 'alarm'
       ) xa1,
       xmltable('/alarm'
         passing xa1.alarms
         columns alarm_id   number path '@id',
                 alarm_freq number path '.'
       ) xa2
  where aa.id_device = 3194
  and aa.hour > trunc(sysdate-1)
  ;
  

• Problem with select query that includes a date

  I fought with it for about 4 hours - Yes, quite literally... I was never good with stuff to date. I have a brain block.

  In any case, here's the situation:

  I have an Access database that has a date/time field. There is no time, just a date in the format dd/mm/yyyy.

  This table is a list of classes held for three years. I want to do a query that displays only the classes that have completed after January 1st of this year. I don't want to put the 01/01/2010 in the query itself because I want it to work again next year.

  It's the current query:

  < name cfquery = "classes" datasource = "" #DSN # "dbtype ="ODBC">"

  SELECT classesTitles.ClassTitle, classesTitles.ClassDesc, classCatagories.catagoryName, classes.classStart, classes.classEnd, classes.classNotes, owners.fName, owners.lName, classTime.timeSlot, classes.classID, classes.classLimit

  FROM classCatagories INNER JOIN (((classes INNER JOIN classesTitles ON classes.classTitle = classesTitles.classTitleID) INNER JOIN classTime WE classes.classTime = classTime.timeID) INNER JOIN owners WE classes.classInstructor = owners.ownerID) ON classCatagories.catagoryID = classesTitles.ClassType

  WHERE the classes.classEnd < 01/01 / #DatePart ("yyyy", now())) #.

  ORDER BY classes.classStart, classCatagories.catagoryName, classTime.timeSlot, classesTitles.ClassTitle

  < / cfquery >

  I don't get any results, but I should get a list of each class that has an end date before 1 January of this year - about 200 cases.

  When I swap around the place where clause:

  < name cfquery = "classes" datasource = "" #DSN # "dbtype ="ODBC">"

  SELECT classesTitles.ClassTitle, classesTitles.ClassDesc, classCatagories.catagoryName, classes.classStart, classes.classEnd, classes.classNotes, owners.fName, owners.lName, classTime.timeSlot, classes.classID, classes.classLimit

  FROM classCatagories INNER JOIN (((classes INNER JOIN classesTitles ON classes.classTitle = classesTitles.classTitleID) INNER JOIN classTime WE classes.classTime = classTime.timeID) INNER JOIN owners WE classes.classInstructor = owners.ownerID) ON classCatagories.catagoryID = classesTitles.ClassType

  WHERE classes.classEnd > 01/01 / #DatePart ("yyyy", now())) #.

  ORDER BY classes.classStart, classCatagories.catagoryName, classTime.timeSlot, classesTitles.ClassTitle

  < / cfquery >

  I get EACH class in the database, and I should get only about 50

  The classes are long about 8 weeks, the dates of beginning and end are in the database in the form of date/time fields, so I need to show everything that ends in 2010, regardless of when it began.

  I'm sure it's something pressing, but he's avoiding me.

  Thank you

  Michelle

  Your field format date in access is irrelevent to the situation at hand.  Use objects appropriate date instead of trying to build an appropriate string.  On 1 January of the current year, the createdate() of coldfusion functions and year() now() are your friends.  Then, a message from Adam was not clear, use cfqueryparam with the resulting variable.

• What is the problem with this query? My head is numb!

  Can someone tell me why this request;

  < cfquery datasource = "manna_premier" name = "qGetstats" >
  SELECT the user name,
  UserFirstName,
  UserLastName,
  LastLogin,
  TotalLogins,
  UserZone,
  TM_Name
  USERS
  WHERE UserID = ' #. TM_log UserID # '.
  < / cfquery >

  generates this error message;

  Run database query error.

  [Macromedia] [SequeLink JDBC Driver] [ODBC Socket] [Microsoft] [ODBC Microsoft Access driver] Too few parameters. 1 expected.

  The error occurred in D:\Inetpub\mannapremier\TM_log_report2.cfm: line 28

  26 :         TM_Name 27 : FROM Users 28 : WHERE UserID = "#TM_log.UserID#" 29 : </cfquery> 30 :

  SQLSTATE 07002 SQL SELECT the user ID, UserFirstName UserLastName, LastLogin, TotalLogins, UserZone, TM_Name FROM USERS WHERE UserID = '10120. VENDORERRORCODE -3010 DATASOURCE manna_premier

  I see that my variable WHERE comes through but the domain name is not working. I'm sure that the answer is obvious, but it drives me crazy!

  Help, please!

  I tried and got this result;

  No, you now use different data. Now, the user ID is empty, while she was 10120 first.

  The only problem seems to be double quotes. To check this, test with

  WHERE UserID = 10120
  

  I'm assuming that the userID field to be an integer. If it is text, then use single quotes (not!), as in

  WHERE UserID = '#TM_log.UserID#'
  

• What is the problem with this query?

  Hello
  
  I need some advice on how to change the following query (with perhaps some analytical function) to speed it up. Currently, it takes 6 + minutes. This query is executed in response to a request from the front-end application and 6 + min is certainly unacceptable.
  
  I am trying to provide as much information I can think, but if more information is needed, please let me know.
  
  I have a table called "wave_result". It contains millions of rows. PK is Wave_Id, Version_nbr, node_nbr and prod_nbr. For each 'wave_id + node_NBR + prod_nbr' there are several versions (version_nbr). In the following query, I try to extract a line with MAX version_nbr for combination of ' wave_id + node_NBR + prod_nbr.
  
  H3. Request:

  SELECT ip1.fnln_cat,
              ip1.sub_cat,
              ip1.bus_cat,
              NVL (SUM (ip1.lsu), 0) val
      FROM ideal_prod ip1, ideal_store s, wave_result wr
     WHERE  wr.wave_id = 51
           AND wr.prod_nbr = ip1.prod_nbr
           AND wr.wave_id = ip1.wave_id
           AND wr.version_nbr =
                  (SELECT MAX (wr1.version_nbr)
                     FROM wave_result wr1
                    WHERE   wr1.wave_id = wr.wave_id
                          AND wr1.node_nbr = wr.node_nbr
                          AND wr1.prod_nbr = wr.prod_nbr)
           AND NVL (wr.ovrd_dcsn_nm, wr.dcsn_nm) = 'Add'
           AND s.wave_id = wr.wave_id
           AND s.node_nbr = wr.node_nbr 
  GROUP BY ip1.fnln_cat, ip1.sub_cat, ip1.bus_cat

  H3. Rank of charges:
  ========
  "wave_result" is the largest table with millions of rows.

  Table                Total Rows       Rows for wave_id = 51
  Ideal_prod           188K             38K
  Ideal_store          3K               574
  Wave_result          90M              19M

  H3. Stats
  are updated almost daily by the DBA (not sure if that's a good or bad).
  
  H3. Explain the Plan:
  =========

  SELECT STATEMENT ALL_ROWS Cost: 330,737 Bytes: 401,787 Cardinality: 14,881 
       12 HASH GROUP BY Cost: 330,737 Bytes: 401,787 Cardinality: 14,881 
            11 VIEW VIEW SYS.VM_NWVW_2 Cost: 330,737 Bytes: 401,787 Cardinality: 14,881 
                 10 FILTER 
                      9 HASH GROUP BY Cost: 330,737 Bytes: 2,425,603 Cardinality: 14,881 
                           8 HASH JOIN Cost: 327,183 Bytes: 41,233,784 Cardinality: 252,968 
                                1 INDEX RANGE SCAN INDEX (UNIQUE) APAPOPR.XPKIDEAL_STORE Cost: 4 Bytes: 13,202 Cardinality: 574 
                                7 HASH JOIN Cost: 327,178 Bytes: 35,415,520 Cardinality: 252,968 
                                     5 HASH JOIN Cost: 198,619 Bytes: 18,764,328 Cardinality: 183,964 
                                          3 TABLE ACCESS BY INDEX ROWID TABLE APAPOPR.IDEAL_PROD Cost: 939 Bytes: 2,272,380 Cardinality: 37,873 
                                          4 TABLE ACCESS FULL TABLE APAPOPR.WAVE_RESULT Cost: 197,063 Bytes: 7,974,414 Cardinality: 189,867 
                                     6 INDEX RANGE SCAN INDEX (UNIQUE) APAPOPR.XPKWAVE_RESULT Cost: 82,467 Bytes: 721,495,854 Cardinality: 18,986,733 

  H3. Sample data:
  ===========

     For wave_id = 51
                   there are 28466854 rows in "wave_result" table
  
     For wave_id = 51 and node_nbr = '0201' and  prod_nbr = '0226960'
                   there are 3 rows in "wave_result" table

  H3. Version of database information are below to:
  =========================

  Oracle Database 11g Enterprise Edition Release 11.1.0.7.0 - 64bit Production
  PL/SQL Release 11.1.0.7.0 - Production
  CORE            11.1.0.7.0                        Production
  TNS for IBM/AIX RISC System/6000: Version 11.1.0.7.0 - Production
  NLSRTL Version 11.1.0.7.0 - Production

  Thank you very much!!
  
  Concerning

  For wave_id = 51
                   there are 28466854 rows in "wave_result" table
  
     For wave_id = 51 and node_nbr = '0201' and  prod_nbr = '0226960'
                   there are 3 rows in "wave_result" table
  

  As you mentioned that this particular request is called by a front screen, how different search criteria you get it to show on the front end?
  Assuming that if the user is searching just of wave_id = 51 which lines returns to 28466854, it is certainly not a good approach to go look up many records to put end Front end is very light and can not hold many records and will be finally the web server will throw off out of memory exception. Always a limit on the number of records that you retrieve from the database and display it on the front plane.
  OR
  The second case where the user makes a search by wave_id, node_nbr and prod_nbr, even if she returns to 3 lines, behind the scens, he still made a full scan on the wave_result table. If we go with analytical function, we can reduce a sweep of extra table on the wave_result table. Also the table has millions of lines and the max (version_nbr) seems to be a frequently used on this table sub query. While inserting a record into the table wave_result, you must probably have a calculated value that will tell you its value max and you will choose just the rank of this value, instead of at each time max(). something like below...

  SELECT ip1.fnln_cat,
              ip1.sub_cat,
              ip1.bus_cat,
              NVL (SUM (ip1.lsu), 0) val
      FROM ideal_prod ip1, ideal_store s, wave_result wr
     WHERE ipl.wave_id = 51
           AND wr.prod_nbr = ip1.prod_nbr
           AND wr.wave_id = ip1.wave_id
           AND wr.wave_id = s.wave_id
           AND wr.node_nbr = s.node_nbr
           AND wr.precomputed_max_value = 'MAX'
           AND NVL (wr.ovrd_dcsn_nm, wr.dcsn_nm) = 'Add'
  GROUP BY ip1.fnln_cat, ip1.sub_cat, ip1.bus_cat;
  

• Creating a column blank for the query.

  Hi all
  
  I have a query which translates like this
  
  A (col1) B (Col2) C (Col3)
  1 90 x
  
  But I want to add an extra empty column at the end as D (Col4). My query is joined by tables 6 and 7. So please can someone help with that.
  
  Thank you
  Sri

  Sri,

  Do you mean something like this:

  select emp,deptno,name,NULL other from employee;
  

  Other this is a column_name that does not really exist in the framework, but can be used in queries.

  See you soon,.
  Manik.

• Problem with SQL query in the view of a line

  Hello
  
  I have a table TABLEA it has 3 columns:
  
  Structure of TABLEA
  
  Code Tax_Code_1 Tax_Code_2
  
  1A
  1 B
  
  
  I want that the output voltage:
  
  Code Tax_Code_1 Tax_Code_2
  1 A AND B
  
  Any help will be necessary for me
  
  Thanks and greetings

  ..

• What is the problem with this query? exception does not work.

  DECLARE
  v_employee_id EMPLOYEES. EMPLOYEE_ID % TYPE;
  v_last_name EMPLOYEES. LAST_NAME % TYPE;
  v_salary EMPLOYEES. % SALARY TYPE.
  e_invalid_emp EXCEPTION;
  BEGIN
  SELECT last_name, salary v_employee_id, v_last_name, v_salary, employe_id
  EMPLOYEES where employee_id = & employee_id;
  
  IF SQL % NOTFOUND THEN
  RAISE e_invalid_emp;
  END IF;
  
  DBMS_OUTPUT. Put_line(v_employee_id||) e '|| v_last_name | » '|| v_salary);
  
  EXCEPTION
  WHEN e_invalid_emp THEN
  DBMS_OUTPUT. Put_line ("' employee not found...");
  END;
  /
  

• problem with a query with clause

  Hello
  
  I am facing a problem, so I describe a little:

  create table schoolboys (
  id integer,
  name varchar2(20)
  /
  
  create table marks (
  schoolboy_id integer,
  mark integer,
  mark_type char(1))
  /
  
  create table avg_marks (
  schoolboy_id integer,
  avg_mark number (5,2)
  )
  /
  
  
  insert into A values (50, 9, 'N');
  insert into A values (50, 7, 'N');
  insert into A values (50, 6, 'T');
  insert into A values (88, 9, 'N');
  insert into A values (88, 7, 'N');
  insert into A values (88, 10, 'T');
  insert into A values (20, 4, 'N');
  insert into A values (20, 5, 'N');
  insert into A values (20, 3, 'N');
  insert into A values (20, 5, 'T');

  and I make this request for several insertion in another table

  INSERT INTO avg_marks 
             (schoolboy_id, 
              avg_mark) 
  (SELECT   schoolboy_id, 
            (Avg(Decode(mark_type,'T',NULL,
                                  mark)) * 3 + Sum(Decode(mark_type,'T',mark, 
                                                                    0))) / 4
   FROM     marks 
   GROUP BY schoolboy_id);

  But what I want is to insert in avg_marks that records where the

  ((Avg(Decode(mark_type,'T',NULL,
                                  mark)) * 3 + Sum(Decode(mark_type,'T',mark, 
                                                                    0))) / 4) > 5

  So, I want to limit inclusion in avg_marks.
  If there are results with (avg (decode...)) < 5 these files, I want to be inserted into another table, with the same columns as avg_marks
  How should I code this? What should I use?
  I tried with
  

  with t as(
  (SELECT   schoolboy_id,
            ((Avg(Decode(mark_type,'T',NULL,mark)) * 3 
            + Sum(Decode(mark_type,'T',mark,0)) / 4))>5) MARK
  from MARKS
  )
  INSERT INTO avg_marks
             (schoolboy_id,
              avg_mark) values (t.schoolboy_id, t.MARK)
   
  ORA-00923: FROM keyword not found where expected

  Best regards
  
  Edited by: Roger22 the 18.06.2009 11:36
  
  fix

  No, not at all sure you insert records into a table, then another! ;) Do you mean an multi-table insert, i.e. two INSERTs in the transaction?

  INSERT FIRST
    WHEN avg_mark >= 5 THEN
      INTO avg_marks (schoolboy_id,  avg_mark)
      VALUES (schoolboy_id, avg_mark)
    WHEN avg_mark < 5 THEN
      INTO avg_marks2 (schoolboy_id,  avg_mark)
      VALUES (schoolboy_id, avg_mark)
    SELECT   schoolboy_id,
            (Avg(Decode(mark_type,'T',NULL,
                                  mark)) * 3 + Sum(Decode(mark_type,'T',mark,
                                                                    0))) / 4 avg_mark
      FROM     marks
   GROUP BY schoolboy_id;
  

• Problems with Sub query

  Hello, I'm looking to create a sub query to display certain information. On the form that the information is entered, the user has the option to enter a phone number extension, as appropriate. If this is the case, I want the phone number to show as 555-867-5309 (x 1234). However, without the problem of controls, the (x) will be displayed without worrying and seems sloppy. Here is the code I have, please notify.
  
  There will be two places, it will be necessary - for the telephone number of the applicant and the app (SME) contact information manager.
  
  Thank you.
  
  

  select
      pm.pk_proj_master_id "Project Number",
      pm.trackit_work_order "TrackIt Work Order",
      pm.name "Project Name",
      pm.status "Project Status",
      req.last_name||', '||req.first_name||', '||req.middle_initial||'.' "Requestor Name",
      rde.department_group_descr "Requestor Department Name",
      req.department_descr "Requestor Division Name",
      
      pm.requester_ext,
      case
          when pm.requester_ext in 
              (
              select 
                  pm.requester_phone|| '(x'||pm.requester_ext ||')' "Requesters Number"
              from 
                  protrac_master pm
              where
                  pm.requester_ext is not null
              )
          
          else
              (
              select
                  pm.requester_phone "Requesters Number"
              from 
                  protrac_master pm
              )   
      end as "Requester Number",
       
      man.last_name||', '||man.first_name||', '||man.middle_initial||'.' "SME Name",
      mdg.department_group_descr "SME Department Name",
      man.department_descr "SME Division Name",
      
      pm.app_manager_ext,
      case
          when pm.app_manager_ext in 
              (
              select 
                  pm.app_manager_phone|| '(x'||pm.app_manager_ext ||')' "SME Number"
              from 
                  protrac_master pm
              where
                  pm.app_manager_phone is not null
              )
          
          else
              (
              select
                  pm.app_manager_phone "SME Number"
              from 
                  protrac_master pm
              )   
      end as "Requester Number",
  
      pm.createby_date "Date Entered",
      pm.date_begin "Date Began",
      pm.date_completed "Date Completed",
      pm.estimated_date "Estimated Completion Date"
  
  from
      protrac_master pm,
      cobr.vw_pps_payroll req, cobr.department_group rde,
      cobr.vw_pps_payroll man, cobr.department_group mdg
      
  where
      pm.requester_id         = req.emple_no and
      pm.requester_dept_id    = rde.pk_department_group_id and
      pm.app_manager_id       = man.emple_no and 
      pm.app_manager_dept_id  = mdg.pk_department_group_id
      
  order by 
      pm.pk_proj_master_id

  I think you can avoid the sub query thing simply by using the NVL2 function:

  select ...
       , NVL2( pm.requester_ext
             , pm.requester_phone|| '(x'||pm.requester_ext ||')'
             , pm.requester_phone) "Requesters Number"
         ...
       , NVL2( pm.app_manager_ext
             , pm.app_manager_phone|| '(x'||pm.app_manager_ext ||')'
             , pm.app_manager_phone) "Manager Number"
         ...
    from protrac_master pm
       , ...