conditional select into statement

Similar Questions

• Select Into statement in the function db - query issued schema table

  problem with "select into" in the function db to point 10.2
  
  
  There are two schemas. "mdbdev" is the master database, and 'devusr' is granted SELECT access table to run queries in the mdbdev schema.
  
  
  with devusr, in SQL, I am able to run the following query
  
  Select wm_concat (strConcatedCountryList)
  from (select country_name as mdbdev.country_master mdbcm strConcatedCountryList
  where mdbcm.country_ship_status = < param? >
  order of country_name)
  
  
  
  but when I use the same procedure/function with 'select into' motion, failed to compile with error * "table or view does not exist."
  
  
  FUNCTION GETCOUNTRYLISTTOSHIP (SHIP_STATUS NUMBER)
  RETURN VARCHAR2
  IS
  var2CountryList VARCHAR2 (1000);
  BEGIN
  Select wm_concat INTO var2CountryList (strConcatedCountryList)
  from (select country_name as mdbdev.country_master mdbcm strConcatedCountryList
  where mdbcm.country_ship_status = value < SHIP_STATUS >
  order of country_name);
  Return var2CountryList;
  END;
  
  
  Please advise/help/suggestion :)

  You don't have to ask for the DBA, you can run a query to find out:

  select *
  from  all_tab_privs_recd
  where owner = 'MDBDEV'
  and table_name = 'COUNTRY_MASTER'
  

  What give you?

• How to modify a statement "select into" how to use a cursor

  The following code fails with an exception too many lines. How can I modify statement Select Into the procedure to use a cursor?
  

  CREATE OR REPLACE PROCEDURE Track_Asset(
     business_date IN NUMBER DEFAULT NULL,
     missing_table_name  OUT VARCHAR2) 
  IS
     ln_business_date NUMBER;
      incorrectdateformat EXCEPTION;
  BEGIN
     IF business_date < 0 
     THEN
        RAISE incorrectdateformat;
     ELSE
        DECLARE
          ln_business_date NUMBER;
        BEGIN
           SELECT MAX(business_date) 
           INTO ln_business_date
           FROM sproof ;
        EXCEPTION
          WHEN NO_DATA_FOUND THEN
           dbms_output.put_line('NO MATCH FOUND'); 
          WHEN OTHERS THEN
          dbms_output.put_line('ORACLE ERROR :' || SQLERRM);        
        END;
        
        DECLARE
          missedfeedfnd EXCEPTION;
        BEGIN
           SELECT 'Missing Value : ' || table_name 
           INTO missing_table_name 
           FROM ( 
              SELECT UPPER(table_name) table_name 
              FROM filespec
              WHERE data_table_name IN ('TABLE1','TABLE2','TABLE3') 
              MINUS ( 
              SELECT DISTINCT UPPER(first_table_name) 
              FROM dpca
              WHERE business_date = ln_business_date 
              AND first_table_name IN ('TABLE1','TABLE2','TABLE3') 
              GROUP BY UPPER(first_table_name) UNION 
              SELECT UPPER(first_table_name) 
              FROM dpca
              WHERE business_dt_num = TO_NUMBER( SUBSTR('201111', 1, 6) || '01' )
              AND first_table_name = 'TABLE4' 
              GROUP BY UPPER(first_table_name) ));
              
              IF missing_table_name  IS NOT NULL THEN
                 dbms_output.put_line('Missing Value : '|| missing_table_name);
                 RAISE missedfeedfnd;
              ELSE
                NULL;
              END IF;
        EXCEPTION
           WHEN TOO_MANY_ROWS THEN
     DBMS_OUTPUT.PUT_LINE (' SELECT INTO statement retrieved multiple rows');
            WHEN missedfeedfnd THEN 
            raise_application_error ( - 20003, 'Missed Feed');
        END;
      END IF;
        EXCEPTION
     WHEN incorrectdatevalue 
     THEN
        raise_application_error ( - 20001, 'Incorrect/Bad Date Entered');
  END;

  OK, try this - OUT param will be filled with the table names comma-separated list:

  PROCEDURE Track_Asset(
     business_date IN NUMBER DEFAULT NULL,
     missing_table_name  OUT VARCHAR2)
  ...
  ...
  
  cursor c_table_names is
  select datatablename
  from   ( select upper(datatablename) datatablename
           from   filespec
           where  data_table_name in ('TABLE1','TABLE2','TABLE3'                                 )
          MINUS
          ( select upper(first_table_name)
            from   dpca
            where  business_dt_num = [-- this date is retrieved by getting the MAX(business_date) from sproof table]
                   and fus_data_table_name in ('TABLE1','TABLE2','TABLE3'
                                              )
            group  by
                   upper(first_table_name)
           UNION
            select upper(first_table_name)
            from   dpca
            where  business_dt_num = to_number( substr('201111',1,6) || '01' )
                   and first_table_name = 'TABLE4'
            group  by
                   upper(first_table_name)
          ));
  
  ...
  ...
  begin
     ...
     for rec in c_table_names
     loop
         missing_table_name  := missing_table_name  || rec.datatablename ||',';
     end loop;
     missing_table_name  := rtim(missing_table_name , ',');
  ...
  ...
  end ;
  

  HTH

  Published by: user130038 on December 28, 2011 08:46

• SELECT INTO does not return no loop lines and stops

  I scoured the internet looking for an appropriate way to handle this, but have developed in white. I'm not a pro so please excuse me if I asked a basic question stupidly. Firstly I have search the Forum but no joy (or rather too much results, none seemed relevant to my case).
  
  In any case, I have the following:
  
  DECLARE
  v_period_start DATE: = TO_DATE ('2010-02-21 00:00:00 ',' YYYY-MM-DD HH24:MI:SS');
  v_period_end DATE: = TO_DATE (' 2010-03-20 23:59:59 ',' YYYY-MM-DD HH24:MI:SS' ");
  
  v_last_active DATE;
  
  CURSOR cust_list
  IS
  SELECT a.cust_no,
  a.service_id,
  b.Name,
  c.account_no,
  c.agreement_no,
  c.agreement_part_no,
  c.MSISDN,
  c.equipment_no,
  a.ordered_from,
  a.cancelled_to,
  a.contract_period_until,
  c.valid_from,
  c.valid_to,
  a.Status,
  d.campaign_service_no,
  d.Name LIKE 'CAMPAIGN '.
  Order_rows a,
  b services
  mobilebb_equips_p_agree_part c,
  campaign_services d
  WHERE < where the big clause >
  
  FC cust_list_futur % ROWTYPE;
  
  BEGIN
  FOR cf IN cust_list
  LOOP
  
  -Search for a command line with the same equipment which ended within 24 hours of the new order from. This should identify up/downgrades
  
  SELECT o.cancelled_to
  IN v_last_active
  Of order_rows o
  WHERE o.agreement_no = cf.agreement_no
  AND o.agreement_part_no = cf.agreement_part_no
  AND o.equipment_no = cf.equipment_no
  AND o.service_id <>cf.service_id
  AND o.cancelled_to BETWEEN cf.ordered_from - 1 AND cf.ordered_from
  AND rownum < 2;
  
  INSERT INTO miken_mobint_cust
  VALUES (cf.cust_no, cf.service_id, cf.name, cf.account_no, cf.msisdn, cf.equipment_no, cf.ordered_from, cf.cancelled_to,
  CF.contract_period_until, cf.valid_from, cf.valid_to, cf.status, cf.campaign_service_no, cf.campaign, cf.agreement_no, cf.agreement_part_no, v_last_active);
  
  END LOOP;
  EXCEPTION
  WHEN NO_DATA_FOUND THEN v_last_active: = NULL;
  END;
  
  The problem seems to be that the SELECT INTO retuens that all lines and my main INSERT never never seems to run. If I comment the SELECT INTO and set the field last slot insert on a NULL value of the table Gets the appropriate records.
  
  I do not expect a return value very often of the SELECT INTO statement, so I need to know how it default null and continue the treatment when nothing is returned.

  Try this.

   DECLARE
      v_period_start DATE := TO_DATE('2010-02-21 00:00:00','YYYY-MM-DD HH24:MI:SS');
      v_period_end   DATE := TO_DATE('2010-03-20 23:59:59','YYYY-MM-DD HH24:MI:SS');
  
      v_last_active  DATE;
  
      CURSOR cust_list
      IS
      SELECT a.cust_no,
             a.service_id,
             b.name,
             c.account_no,
             c.agreement_no,
             c.agreement_part_no,
             c.msisdn,
             c.equipment_no,
             a.ordered_from,
             a.cancelled_to,
             a.contract_period_until,
             c.valid_from,
             c.valid_to,
             a.status,
             d.campaign_service_no,
             d.name AS "CAMPAIGN"
        FROM order_rows a,
             services b,
             mobilebb_equips_p_agree_part c,
             campaign_services d
       WHERE 
  
      cf cust_list_futur%ROWTYPE;
  
   BEGIN
      FOR cf IN cust_list
      LOOP
  
   -- Look for an order row with the same equipment that finished within 24 hours of the new order starting. This should identify up/downgrades
  
      BEGIN
            SELECT o.cancelled_to
              INTO v_last_active
              FROM order_rows o
             WHERE o.agreement_no = cf.agreement_no
                  AND o.agreement_part_no = cf.agreement_part_no
                  AND o.equipment_no = cf.equipment_no
                  AND o.service_id <> cf.service_id
                  AND o.cancelled_to BETWEEN cf.ordered_from - 1 AND cf.ordered_from
                  AND rownum < 2;
  
      EXCEPTION
          WHEN NO_DATA_FOUND THEN
              v_last_active := null;
      END;
  
         INSERT INTO miken_mobint_cust
         VALUES (cf.cust_no, cf.service_id, cf.name, cf.account_no, cf.msisdn, cf.equipment_no, cf.ordered_from, cf.cancelled_to,
                 cf.contract_period_until, cf.valid_from, cf.valid_to, cf.status, cf.campaign_service_no, cf.campaign, cf.agreement_no, cf.agreement_part_no, v_last_active);
  
      END LOOP;
   EXCEPTION
      WHEN NO_DATA_FOUND THEN v_last_active := NULL;
   END;
  

• How to add a condition to this statement.

  Hi friends,
  
  I have a requirement of the customer.
  
  Let me explain the scenario.
  
  I have a table, which contains some employees say 1, 2,3,4.
  
  
  So I wrote a query so that it...
  
  Select * from where a trunc (sysdate) between effective_start_date and effective_end_date person_id =: p_person_id;
  
  in this statement, I would like to add a condition more say
  
  If person_id in (1,2,3,4) then
  p_person_id: = null;
  on the other
  p_person_id = person_id;
  
  the addition of this condition in mysql statement...
  
  can someone help me with this...
  
  Thanks in advance.

  Hello

  Looks like that (no data, not yet tested)

  select * from A
   where trunc(sysdate) between effective_start_date and effective_end_date
     and person_id = CASE
                        WHEN NOT :p_person_id in (1,2,3,4)
                        THEN :p_person_id
                        ELSE person_id
                     END;
  

  Concerning
  Peter

• No data found in SELECT INTO

  Hi all
  
  I try to choose in a variable in a loop of cursor but I get NO DATA FOUND when
  Select does not return a value.
  I want to set the variable to 0 when the selection fails.
  
  I tried to add an exception, but I get the following:
  
  ORA-00900: invalid SQL statement
  
  
  Here is the code I am trying to run (the exception is commented):
  
  Line: -.
  PROCEDURE MyProc IS
  v_op NUMBER;
  BEGIN
  
  FOR cursor1 IN (select * from myTable)
  LOOP
  Operation of SELECTION IN v_op from operationsTable where id = cursor1.ID;
  
  
  -EXCEPTION *.
  -WHEN NO_DATA_FOUND THEN
  -v_op: = 0;
  
  -END; *
  
  END LOOP;
  MyProc END;
  Line: -.
  
  Does anyone know how to set a variable when the selection fails and NO DATA FOUND?
  Thank you
  dyahav

  You need to hold your exception by BEGIN and END

  BEGIN
  
  select into...
  
  EXCEPTION WHEN NO_DATA_FOUND
  do_something
  END;
  

• Comment of "Syntax error in INSERT INTO statement." validation error.

  I had this working perfectly on MySQL. Then I was told that I had to do with a mdb database in Access, now the errors pop up like crazy. I fixed all of them except this one, I get a "syntax error in INSERT INTO statement." When I try to post a comment on a post in my mini-blog site.

  The thing is, I use almost the same code for the comments I make blog themselves. The positions work fine, the comments are not.

  He said that the error is on my page addcomment_process as 119. It is the following (line 119 is underlined, bold line is also bolded in the error message):

  < name cfquery = "input_comments."

  DataSource = "" #DSN # ""

  username = "" #DSNUSER # ""

  password = "#DSNPASS #" >

  INSERT INTO (comments)

  speaker,

  Comment,

  DateTime,

  post_id

  )

  VALUES)

  "#form.commenter #"

  "#form.comment #"

  "#form.datetime #"

  "#form.post_id #".

  )

  < / cfquery >

  It is all fine data, by simply inserting not, because it says:

  "INSERT INTO comments (speaker, comment, datetime, post_id) VALUES ('Anonymous', 'test commentary.', 25 April 12 22:55 ', ' 2').

  It worked perfectly, when it was a MySQL database, now I keep getting errors present a part of the site now that it is MDB. Anyone know what is happening?

  DateTime is a reserved word in Access.

  --

  Adam

• Select Into when no line is detected

  In my select Into, it is possible that we will not find the lines and it causes an error. How to define v_3Month_lb null so that when no line is detected, that an exception is not thrown?
  
  
  Select Pa_Entered_Value
  IN v_3Month_lb
  of wr_measure_value
  where wr_measure_oid = 2
  and project_associate_oid = p_ProjectAssociateOID
  and period_oid = (p_PERIOD_OID)
  and wr_item_oid = p_ItemOID;

  How to define v_3Month_lb null so that when no line is detected, that an exception is not thrown?

  You do not have.
  Exception handler is not required.

• Select the statement to insert into a table by using a loop

  Hello
  

  create table uploadtab(
  itema varchar2(3),
  xtype varchar2(1),
  salesa number,
  margina number,
  salesb number,
  marginb number,
  salesc number,
  marginc number);
  
  insert into uploadtab
    (itema, xtype, salesa, margina, salesb, marginb, salesc, marginc)
  values
    ('abc', 'a', 100, .40, 300, .10, 450, .25);
  
  create table testinsert(itema varchar2(3),
  xtype varchar2(1),
  sales number,
  margin number);

  what I want to do is create 3 records based on this one in a loop
  so here's my desired for testinsert output
  

  abc  a 100  .40
  abc  a 300  .10
  abc  a 450  .25

  I don't want to use 3 insert tables if possible
  
  any help would be greatly appreciated
  Thanks in advance
  
  Published by: DM on July 7, 2010 14:22

  This gives a shot:

  INSERT INTO testinsert
  ( itema
  , xtype
  , sales
  , margin
  )
  SELECT  itema
  ,       xtype
  ,       DECODE
          ( RN
          , 1,salesa
          , 2,salesb
          , 3,salesc
          )
  ,       DECODE
          ( RN
          , 1,margina
          , 2,marginb
          , 3,marginc
          )
  FROM            uploadtab
  CROSS JOIN      (
                          SELECT ROWNUM RN
                          FROM   dual
                          CONNECT BY LEVEL <= 3
                  )
  ;
  

• SELECT INTO VARIABLE base

  I'm trying to make a
  
  SELECT EMP_TOTAL FROM EMPLOYEES WHERE ADD_DATE = SYSDATE
  
  In the statement above "WHERE ADD_DATE = SYSDATE' this condition eliminates several values since in my table there is only a 1 value for each sysdate
  
  And this unique value, I want to choose in a variable coz
  
  If it is not equal to zero
  
  I want to run a bunch of SQL
  ON THE OTHER
  
  SOME other SQL.
  
  The problem I face is when selecting in the variable.
  
  
  I get error when I try
  
  DECLARE a number
  SELECT EMP_TOTAL FROM EMPLOYEES WHERE ADD_DATE = SYSDATE;
  IF a! = 0 THEN
  
  -------SQL----------
  
  ON THE OTHER
  
  --------SQL-------
  
  
  I also tried
  
  DECLARE a NUMBER: SELECTION EMP_TOTAL OF EMPLOYEES WHERE ADD_DATE = SYSDATE
  
  Help, please.
  
  Newbie in PLSQL

  I get error when I try

  Let me guess, the error is right no data found? He should know that sysdate (in fact, all dates in Oracle) always has a component "time" to the second, so it is unlikely that you would have a file meets the criteria. Depending on whether you actually store times in your area of add_date that you must do something more like:

  If there is no time in your area

  DECLARE
     a number;
  BEGIN
     SELECT emp_total INTO a
     FROM employees
     WHERE add_date = TRUNC(sysdate) ;
     IF a != 0 THEN
        -------SQL----------
     ELSE
        --------SQL-------
     END IF;
  END;
  

  If you are actually storing time:

  DECLARE
     a number;
  BEGIN
     SELECT emp_total INTO a
     FROM employees
     WHERE add_date >= TRUNC(sysdate) and
           add_date < TRUNC(sysdate + 1);
     IF a != 0 THEN
        -------SQL----------
     ELSE
        --------SQL-------
     END IF;
  END;
  

  John

• How to test different Select into a PL/SQL block?

  Hello
  
  I'm relatively new to PL/SQL and I'm doing several int selects only one PL/SQL block. I am faced with the fact that if a single select statement returns no data, I have to go to the when exception DATA_NOT_FOUND.
  
  Or, I would test selects different.
  
  In an authentication script, I'm looking for a table for a USER ID (USERID) and an ID application, to check if a user is registered under this username for this APPLICATION.
  
  There are different possibilities: 4 possibilities:
  -Existing USER name or do not exist and
  -Aplication ID found or not found for this particular USER ID.
  
  I would test teas 4 possibilities to obtain the status of this user do regardin this application.
  
  The problem is that if a select returns no rows, I'll not found exception data.
  In the example below you can see that if no line is returned, with the exception

  DECLARE
  P_USERID VARCHAR2(400) DEFAULT NULL;
  P_APPLICATION_ID NUMBER DEFAULT NULL;
  P_REGISTERED VARCHAR2(400) DEFAULT NULL;
  BEGIN
  SELECT DISTINCT(USERID) INTO P_USERID FROM ACL_EMPLOYEES
  WHERE  USERID = :P39_USERID AND APPLICATION_ID = :APP_ID ;
  :P39_TYPE_UTILISATEUR := 'USER_REGISTERED';
  EXCEPTION
  WHEN NO_DATA_FOUND THEN
  :P39_TYPE_UTILISATEUR := 'USER_NOT_FOUND';
  END;

  I would like to first of all make this statement:

  SELECT DISTINCT(USERID) INTO P_USERID FROM ACL_EMPLOYEES
  WHERE  USERID = :P39_USERID 

  Do this if the user is found:

  SELECT DISTINCT(USERID) INTO P_USERID FROM ACL_EMPLOYEES
  WHERE  USERID = :P39_USERID AND APPLICATION_ID = :APP_ID ;

  etc...
  
  Basically, I don't want to go to the not found exception before you have tested the 4 possibilities.
  
  Do you have a suggestion?
  
  Thank you very much for your help!
  
  Christian

  If you I had to check several conditions, this is how I would do:

  DECLARE
    P_USERID         VARCHAR2(400) DEFAULT NULL;
    P_APPLICATION_ID NUMBER DEFAULT NULL;
    P_REGISTERED     VARCHAR2(400) DEFAULT NULL;
  BEGIN
   SELECT USERID
       , MAX(DECODE(application_id, :APP_ID, :APP_ID)) app_id_valid
    INTO P_USERID
       , P_APPLICATION_ID
    FROM ACL_EMPLOYEES
   WHERE USERID = :P39_USERID
   GROUP BY USERID
  
   IF P_APPLICATION_ID IS NULL
   THEN
     :P39_TYPE_UTILISATEUR := 'NOT REGISTERED TO APPLICATION_ID';
   ELSE
     :P39_TYPE_UTILISATEUR := 'USER_REGISTERED';
   END IF;
  EXCEPTION
    WHEN NO_DATA_FOUND
    THEN
      :P39_TYPE_UTILISATEUR := 'USER_NOT_FOUND';
  END;
  /
  

  not tested

  C.

• Insert records into statement

  Refresh us the data from the production environment by using the Task Scheduler tool (channels, programs, programs, etc.).  It is probably not the best way, but I inherited structure, is what it is.  Seen retrieve data on a database link, apply logic to the data (collection ontime, ontime delivery) and fill the tables.

  Procedures to insert in temporary tables, which carries data over a database link.  Then records are deleted and inserted into the actual table based on records from the temporary table using the IN clause on the primary key.  Some tables have several columns of hundreds (new legacy, not preferred) using an UPDATE is not feasible.  Here is an example of one of the procedures.

  

  DATE_CUTOFF is the create date or the date of update of the document.  NN_SHIPMENT_INFO_XA is the actual table.  NN_SHIPMENT_INFO_XA_TEMP is the temporary table.

  NN_SHIPMENT_INFO_XA has 292 columns, so deletion and insertion of 10 k lines takes nearly two hours.  I read what one of the most inefficient operators.  I tried to gather statistics on the temporary table every time, but that has not helped.  Break the view into pieces and reducing the number of columns is difficult because there are several layers of logic and tables between the database link and the layer of the end user.  Because of its complexity, I am committed to the design of my predecessor in the foreseeable future.  Is there a better way to build this procedure to make it run faster?

  Any suggestions are greatly appreciated.

  In the code you posted, you do:

  Select max (date_cutoff_nnsi_xa) - interval time '12'

  in last_cutoff_nnsi_xa

  of nn_shipment_info_xa

  Keep little and use last_cutoff_nnsi_xa as a predicate on the view using a part of the merge as you statement used to fill the temporary table.  So, the merger would look more like:

  merge into nn_shipment_info_xa targ

  using (select * from nn_shipment_info_xa_temp)

  where date_cutoff_nnsi_xa > last_cutoff_nnsi_xa) CBC

  on (targ.shipment_id = src.shipment_id)

  When matched then

  Update

  Set targ.col1 = src.col1,

  Targ.col2 = src.col2,

  ...

  -possibly

  where targ.col1 <> src.col1 and

  Targ.col2 <> src.col2

  ...

  When not matched then

  Insert (shipment_id, col1, col2,...)

  values (CBC, col1, src.shipment_id, src.col2,...)

  John

• After a SELECT INTO, the use of the result as a parameter to another SQL variable?

  Hello world

  My apologies for these basic questions, but I need your help in order to run the report in STATSPACK (spreport.sql) in a NON-interactive manner.

  To do this, I need 3 parameters:

  1. begin_snap
  2. end_snap
  3. report_name
     

  I have a version of the report named spreportBH.sql. This sql receives 3 parameters and generates the report success.

  SQL > @spreportBH.sql 923 1111 /u01/BHB/test4.txt

  For this example begin_snap=923, end_snap=1111 and report_name is /u01/BHB/test4.txt

  My problem is that the snapid must be calculated based on the date.

  To achieve this goal, I can query the table STATS$ SNAPSHOT.

  For example, this query gives me the 923:

  Select min (snap_id)

  de PERFSTAT . STATS$ SNAPSHOT

  où snap_time >= TO_DATE ('12-06-2014','dd-mm-YYYY');

  
  

  I know that I can put the value in a variable. For this, I'm doing:

  
  

  sqlplus "virtue sysdba" s < <!   1 > > ${fichier_de_sortie} 2 > > ${error_file}

  DECLARE

  firstSnapId NUMBER;

  BEGIN

  Select min (snap_id) INTO firstSnapId

  of PERFSTAT. STATS\$ SNAPSHOT

  where snap_time > = to_date ('12-06-2014', ' dd-mm-YYYY');

  END;

  /

  !

  
  

  How can I use firstSnapId to call @spreportBH.sql?

  I'd appreciate any help that can give me.

  Thank you d ' attribute

  Bety

  
  

  
  

  Hi, Bety,

  So, you want to call a script in SQL * more.  I'm not suggesting using PL/SQL use SQL * Plus Substitution Variable instead.

  -The following command of the COLUMN said that everything which appears in a column called

  -firstSnapIdCol will also go in a variable of substtitution called firstSnapId

  COLUMN firstSnapIdCol new_value by firstSnapId

  Select min (snap_id) AS firstSnapIdCol  

  of PERFSTAT. STATS$ SNAPSHOT

  where snap_time > = to_date (6 December 2014 ", 'dd-mm-YYYY');

  -Now that firstSnapId has a value, we can use it like this:

  @spreportBH.sql & firstSnapID 1111 /u01/BHB/test14.txt

• SELECT INTO make case-insensitive query?

  Hi all
  
  I'm having a problem with a query as if it is case-insensitive when I do a «SELECT table.x variable INTO...» ». If I do the same query, without the "variable INTO", I get a result as expected.
  
  The query is similar to the following:
  SELECT table_id variable FROM table WHERE table.varchar2column = 'name' AND table.integercolumn = int_variable;
  
  There is a unique constraint, forcing the varchar2column / integercolumn of the unique values.
  
  If the varchar2column has two entries that differ only in this case, a regular at the statement select returns a result. However, the procedure stored with the, I get the following error message:
  
  ORA-01422: exact fetch returns more than number of lines
  
  He gets not only if I remove the entry that only differs in the case of the letters. It is 10.2.0.1.0 Oracle, if that matters.
  
  I need this application to be case-sensitive. I'm not an expert of the DB, I'm just a developer trying to solve a problem. I tried to run "ALTER session SET nls_sort = binary" in the stored procedure, but I guess I can't do that...
  
  Any suggestion would be appreciated.

  As mentioned before... name your best-er settings.

  Your setting is the same as the name of the column on the table that you query... bad move.

  Here I change the username to work things p_username and 'magically '.

  CREATE OR REPLACE FUNCTION get_cached_user_id(p_username user_security_table.username%type, authserver user_security_table.authentication_server_id%type)
    RETURN user_security_table.user_security_id%type IS
  
    cacheduserid user_security_table.user_security_id%type;
  BEGIN
    6
    SELECT uat.user_security_id
    INTO cacheduserid
    FROM user_security_table uat
    WHERE uat.username = p_username AND uat.authentication_server_id = authserver;
  
    IF cacheduserid IS NULL THEN
       SELECT uat.user_security_id INTO cacheduserid FROM user_security_table uat WHERE uat.username = 'anonymous' AND uat.authentication_server_id IS NULL;
    END IF;
    RETURN cacheduserid;
    EXCEPTION
       WHEN NO_DATA_FOUND THEN
         SELECT uat.user_security_id INTO cacheduserid FROM user_security_table uat WHERE uat.username = 'anonymous' AND uat.authentication_server_id IS NULL;
         RETURN cacheduserid;
  END;
   21   /
  
  Function created.
  
  TUBBY_TUBBZ?SELECT get_cached_user_id('sample',20) from dual;
  
  GET_CACHED_USER_ID('SAMPLE',20)
  -------------------------------
                       2
  
  1 row selected.
  
  TUBBY_TUBBZ?
  

  Simple fix... development of a standard in the future, as for all procedure using p_ settings can save you a lot of grief.

• Query internal within a select if statement

  I would like to know how to write an if condition inside a select statement of the present
  

  b.inventory_item_id in(
              select distinct inventoryid
              from lcm.sis_inventario_new
              where fec_envio > to_date('28/01/2010','DD/MM/YYYY'))

  I mean something like

  select b.inventory_item_id as id, 
         b.segment1 as c,
         b.segment2 as "Código",
         b.description as "Descripción", 
         tl.description as "Descripción Larga", 
         b.attribute1 as "Descripción Corta",
         c.segment1 as "Departamento",c.segment2 as "Familia", c.segment3 as "Subfamilia",
         b.creation_date as "Creado",
         b.INVENTORY_ITEM_STATUS_CODE as "Status",
         case b.attribute5 when '4' then 'Si' else 'No' end as "Pesado",
         case b.inventory_item_id when in  select distinct inventoryid
              from lcm.sis_inventario_new
              where fec_envio > to_date('28/01/2010','DD/MM/YYYY')) then 'Si' else 'No' end as "Existe"

  There are a number of ways, and the best solution for you will depend on several factors, including the index, the number of rows in the tables, etc..

  Go using analytical functions;

  with b as (
     select rownum inventory_item_id from dual
     connect by rownum <=2),
  lcm as (
     select 1 inventoryid, to_date('28/01/2000', 'DD/MM/YYYY') fec_envio from dual union all
     select 1, to_date('31/12/2010', 'DD/MM/YYYY') from dual union all
     select 2, to_date('28/01/2000', 'DD/MM/YYYY') from dual union all
     select 2, to_date('28/01/2008', 'DD/MM/YYYY') from dual)
  select b.inventory_item_id id,
         case
            when max(fec_envio) over(partition by b.inventory_item_id) >
                 to_date('28/01/2010', 'DD/MM/YYYY') then
             'Si'
            else
             'No'
          end as existe
  from b, lcm
  where b.inventory_item_id = lcm.inventoryid;
  
          ID EXISTE
  ---------- ------
           1 Si
           1 Si
           2 No
           2 No