ORA-00904 after creating a relationship with object_ref

Similar Questions

• ORA-00904 on CREATE TABLE with a virtual column based on the XMLTYPE content

  Hello
  
  This is another one for the gurus of the syntax...
  
  Try the following, fails with ORA-00904: "MESSAGE". "' GETROOTELEMENT": invalid identifier

  CREATE TABLE XML_TEST_VIRT
    (
     MSG_TYPE         GENERATED ALWAYS AS (MESSAGE.GETROOTELEMENT()) VIRTUAL,
     MESSAGE  XMLTYPE             NOT     NULL,
     IE906    XMLTYPE             DEFAULT NULL
    )
     XMLTYPE COLUMN MESSAGE STORE AS SECUREFILE BINARY XML
     XMLTYPE COLUMN IE906   STORE AS SECUREFILE BINARY XML
  /

  While it succeeds

  CREATE TABLE XML_TEST_VIRT
    (
     MSG_TYPE         GENERATED ALWAYS AS (EXTRACT(MESSAGE, '/*').GETROOTELEMENT()) VIRTUAL,
     MESSAGE  XMLTYPE             NOT     NULL,
     IE906    XMLTYPE             DEFAULT NULL
    )
     XMLTYPE COLUMN MESSAGE STORE AS SECUREFILE BINARY XML
     XMLTYPE COLUMN IE906   STORE AS SECUREFILE BINARY XML
  /

  The GETROOTELEMENT from SYS member function. XMLTYPE is stated as "PARALLEL_ENABLE DETERMINISTIC" the method called is not the problem, as evidenced by the 2nd case.
  Using the MESSAGE column that is of type XMLTYPE directly seems to be the problem. But the question is "why." The result of the EXTRACT function is of type XMLTYPE and call his works of members, the column is also of type XMLTYPE still call its members fails...
  
  Thanks in advance for any ideas on that.
  
  Best regards
  Philippe

  Going on the means to go far, far back.

  2003 re: function getRootElement ORA-00904

• How to format the column in order to create a relationship with the other field of the table

  I have two tables. Table A, table B
  In table A, I have records with over no matter what hyphen e.g. 124545, domain name is late but in table B I have the files archived with a hyphen after the second digit as 12-4545.
  
  I am trying to create a relationship between two tables.
  
  Select a.TaxId, B.TaxId
  Table A, table B
  where a.TaxId = late b.
  
  How can I remove the hyphen to create a relationship between two tables

  Hello

  REPLACE is good for this:

  SELECT     *     -- or whatever
  FROM     table_a      a
  JOIN     table_b  b  ON  a.taxid     = REPLACE (b.taxid, '-')
  ;
  

  REPLACE (b.taxid, '-') returns a copy of b.taxid with any '-is removed, it will return "12345" If b.tax_id is one of the following
  12345
  12-345
  1-2345
  03/12/45
  -12345
  -1 - 2 - 3 - 4 - 5-

• "Getting Started" VM said "ORA-01471: cannot create a synonym.

  People,

  I downloaded the VM with demo integration OGG + ODI for 12.1.3, as well as the PDF "Getting Started with Oracle Data Integrator 12 c". Try step 10.3.2 on pg 92: "Initial running load staged Mapping", but it throws the error message:

  ODI-1228: task Create synonym on target-LKM Oracle for Oracle (DBLINK) - fails when connecting the TRG_ODIDEMO target.

  Caused by: java.sql.SQLException: ORA-01471: cannot create a synonym with the same name as the object

  He told the truth, I tried the code in sqlplus with the same error:

  SQL > conn odi_staging/oracle;

  Connected.

  SQL > create synonym ODI_STAGING. C$ _0OGG_CUSTOMER1 for ODI_STAGING. C$_0OGG_CUSTOMER1@ORCL;

  ERROR on line 1:

  ORA-01471: cannot create a synonym with the same name as the object

  Initially, the definition of database server did not have the Instance = orcl inside, I added and got the same error with and without "orcl". Why is this happening? Another odd piece that may be associated with: I tried this mapping even earlier, and it got stuck on the front line, trying to create a view, but was insufficient privs for ODI_STAGING select the ogg_customer table. So I just did "grant dba to odi_staging;" and that was not enough! Tried stop/start ODI, then the DB without change. Hmmm, s/n role not enough? I've also had "grant select any table of odi_staging;", then this line worked.

  Back to the current problem: I can get the other procs to run, as the Clean Up target tables works very well, and the demo Client works very well.

  (1) anyone has experience this problem of synonym?

  (2) no matter who else get it to work?

  -Thanks, Steve F.

  I got the fix in the product manager. There are two very similar LKMs:

  • LKM Oracle for Oracle (DBLINK) (must be imported)
  • LKM Oracle for Oracle shoot (DB Link). GLOBAL (integrated)

  Of course, they look the same to me except that the second works and one doesn't. In all honesty, the book on pg 89 says 'Pull' in the name. Chalk it up to user error. Detective Conan.

• DROP table and CREATE with column name and ORA: 00904

  Hello
  I created a table with an attribute date CRTD_ON and two other attributes. Successfully created. I can see the data tab and the window empty data with column names in the header...
  Now, I had to change the name of field attribute date on OPNG_DATE so I don't bother to change, instead, I deleted and created with the name of the column has changed. Then I restart the ddl. Once again managed to create. But this time I am not able to see the data tab or the column headings. And I see the error of the log of the data Publisher ORA-00904: "CRTD_ON": invalid identifier
  Is this a bug?

  There is a current bug where you will get the data if you have a public synonym with the same table name pointing to another table with another structure. Can you confirm that in your environment?

  K.

• ORA-00904 with nested tables

  Hello. I have a problem of creating tables with nested tables.
  After you have created a few types, I would create a table based on these, but I get
  
  SQL error: ORA-00904: invalid identifier
  + 00904. 00000 - '% s: invalid identifier ' +.
  
  over and over again. This is my SQL code:
  
  1 create or replace TYPE t_indirizzo () AS OBJECT
  via VARCHAR (45), 2
  3 NUMBER,
  4 cap INTEGER (5).
  5 Città varchar (30),
  6 provincia varchar (30),
  region 7 varchar (30)
  )
  
  8 create or replace TYPE t_telefono () AS OBJECT
  9 num_tel Number (15)
  )
  
  10 create or replace TYPE t_listaTelefono AS TABLE OF THE t_telefono
  
  11 create or replace TYPE t_cliente () AS OBJECT
  12 cod_cliente NUMBER (8).
  13 address t_indirizzo,
  14 email varchar (30),
  Telefono 15 t_listaTelefono
  (16) NOT FINAL
  
  17 CREATE TABLE client (t_cliente)
  18 cod_cliente NOT NULL,
  19 address NOT NULL,
  email 20 NOT NULL,
  21 PRIMARY KEY (cod_cliente)
  (22) TABLE IMBRIQUE 'telefono' STORE AS 'numTelCli_tab '.
  23 RETURN VALUE
  
  My version of Oracle is 10.2.0. Thank you in advance.
  
  Published by: user11231503 on 2.35 giu-8-2009

  so, just to be sure, run this:

  create or replace TYPE t_indirizzo AS OBJECT (
  via VARCHAR(45),
  numero NUMBER,
  cap INTEGER(5),
  citta VARCHAR(30),
  provincia VARCHAR(30),
  regione VARCHAR(30)
  );
  / 
  
  create or replace TYPE t_telefono AS OBJECT (
  num_tel NUMBER(15)
  );
  / 
  
  create or replace TYPE t_listaTelefono AS TABLE OF t_telefono
  / 
  
  create or replace TYPE t_cliente AS OBJECT (
  cod_cliente NUMBER(8),
  indirizzo t_indirizzo,
  email VARCHAR(30),
  telefono t_listaTelefono
  ) NOT FINAL;
  / 
  
  CREATE TABLE cliente OF t_cliente(
  cod_cliente NOT NULL,
  indirizzo NOT NULL,
  email NOT NULL,
  PRIMARY KEY (cod_cliente)
  ) nested table telefono store as numTelCli_tab
  return as value
  /
   
  

• SDO ORA-00904 identifier valid after patch 20544696 and 20225988 (CPU Apr 15)

  The seller of our web mapping system uses a syntax that contains the identifier SDO_GEOM. SDO_GEOMETRY and similar. An example simplified using the point constructor ODD as in...
  Select sdo_geom.sdo_geometry (2001, 81989, sdo_geom.sdo_point_type (531599, 184530, null), null, null) double;

  This syntax has stopped working in the TEST database, after I applied the patches April 2015 CPU. It generates an ORA-00904: "SDO_GEOM. "" SDO_GEOMETRY ': identifier not valid, which in fact I was expecting. The package SDO_GEOM in schema MDSYS, often referenced without the option scheme by the public synonym, do not contain a type or function SDO_GEOMETRY. I've never used this syntax and wrote instead...

  Select mdsys.sdo_geometry (2001, 81989, mdsys.sdo_point_type (531599, 184530, null), null, null) double;

  .. of course, who continues to work. However, to my surprise, I found that the syntax of the provider, which generates the ORA-00904 in all databases of fix post, works perfectly well in all our pre patch databases.

  Is the same for all other types in the MDSYS (SDO_ELEM_INFO_ARRAY, SDO_ORDINATE_ARRAY...) schema.

  I'm dealing with the strange situation where I not only understand why something has stopped working; first of all I do not understand why the syntax of the seller has already worked.

  We are using Oracle 11.2.0.4 EE with Oracle Spatial on Windows Server 2008 R2 64-bit. Patches that have been requested are

  1) patch 20544696: WINDOWS DB BUNDLE PATCH 11.2. 0.4.15
  2) patch 20225988: WINDOWS ORACLE JAVAVM COMPONENT BUNDLE PATCH 11.2. 0.4.3

  Kind regards

  Torsten

  Hi Torsten,

  Well, here you have a kind of strange thing Oracle Spatial.  If you told me that

  SELECT SDO_GEOM.SDO_GEOMETRY(2001, 81989, SDO_GEOM.SDO_POINT_TYPE(531599, 184530, NULL), NULL, NULL) FROM dual;
  

  is a valid statement, I'd say of course not. But on my production server which is the power supply 11.2.0.4.5, it works!

  However, on a 12 c test server it does not work.

  

  All I can say is that you are right in everything you say.  Oracle has removed everything that made this weird syntax once the work and it seems that the PSU April stops for good.  Your provider needs to fix things and perhaps address why they used such a weird thing in the first place.

  Like you I don't see what makes this function on 11g.  There is no synonym that I see.  And if wrap you it in an anonymous block fails

  DECLARE
     foo SDO_GEOMETRY;
  BEGIN
     SELECT SDO_GEOM.SDO_GEOMETRY(2001,81989,SDO_GEOM.SDO_POINT_TYPE(531599,184530,NULL),NULL,NULL)
     INTO foo
     FROM dual;
  END;
  /
  

  Strange thing indeed.  I guess there could be a wider discussion here on what patches are in turn brought in 11g and if the side effects are always worth.

  See you soon,.

  Paul

• With the help of the ORA-00904 granted by ROLE

  Hi all

  I have an error using the grant through ROLE (Oracle 11.2.0.1 on redhat Linux)

  Here is the explanation of my problem:

  I create a package to wear patterns

  create or replace 
  PACKAGE                     usera.pkg_utils AS 
   function f_my_function return RAW;
  END pkg_utils;
  

  I create the package body:

  create or replace 
  PACKAGE BODY usera.pkg_utils  AS 
   function f_my_function return RAW  is
    v_uuid RAW(16);
  begin
    v_uuid := sys.dbms_crypto.randombytes(16);
    return (utl_raw.overlay(utl_raw.bit_or(utl_raw.bit_and(utl_raw.substr(v_uuid, 7, 1), '0F'), '40'), v_uuid, 7));
  end f_random_uuid;
  END pkg_utils;
  

  I create a 'R_ROLE1' role to which I admit "EXECUTE ON USERA. PKG_UTILS TO R_ROLE1.

  I give you that R_ROLE1 to the PUBLIC so that each user can access the package USERA. PKG_UTIS

  If I call another user f_my_function, it works well.

  If I call another user from a procedure f_my_function I get "ORA-00904".

  ex: to UserB

  create or replace 
  PROCEDURE P_PROC
  AS
  num int;
  uuid_ RAW(16);
  BEGIN
      select usera.pkg_utils.f_my_function() into uuid_ from dual;
  END P_PROC;
  /
  

  If I give you that directly "GRANT EXECUTE ON USERA. PKG_UTILS TO PUBLIC' so it works well.

  Thus, the error comes from the fact that I use a role, but I can't fix it. I don't understand why USERB does not use its grant to the role of r_role1 granted to PUBLIC.

  Is there something as USERB is unusable cascade role during the call to a procedure? is there something to "activate"? ".

  Thank you all

  See the doc database security for your answer and the explanation of what others have said.

  http://docs.Oracle.com/CD/B28359_01/network.111/B28531/authorization.htm#i1007304

  Operation of the roles in PL/SQL blocks

  The use of roles in a PL/SQL block depends on if it is an anonymous block, or a block named (stored procedure, function, or trigger), and if it runs with the rights of the author or the rights of the applicant.

  Roles used in the appointed with rights to define blocks

  All roles are disabled in any PL/SQL block named (stored procedure, function, or trigger) that runs with the rights of the author. Roles are not used to privilege control, and you cannot define roles within the procedure of a DEFINER rights.

  The SESSION_ROLES view shows all roles that are currently enabled. If a named PL/SQL block that runs with Definer Rights queries SESSION_ROLES , then the query will return no rows.

• What's * NOT * less out error with ORA-00904?

  SQL> select * from v$version;
  
  BANNER
  --------------------------------------------------------------------------------
  Oracle Database 11g Express Edition Release 11.2.0.2.0 - Beta
  PL/SQL Release 11.2.0.2.0 - Beta
  CORE    11.2.0.2.0      Production
  TNS for 32-bit Windows: Version 11.2.0.2.0 - Beta
  NLSRTL Version 11.2.0.2.0 - Production
  
  
  
  SQL> drop table t1;
  
  Table dropped.
  
  SQL> drop table t2;
  
  Table dropped.
  
  SQL> create table t1 (some_order_line_id number);
  
  Table created.
  
  SQL> create table t2 (order_line_id number);
  
  Table created.
  
  SQL> SELECT order_line_id FROM t1
    2  /
  SELECT order_line_id FROM t1
         *
  ERROR at line 1:
  ORA-00904: "ORDER_LINE_ID": invalid identifier
  
  
  SQL> SELECT * FROM t2 WHERE order_line_id IN (SELECT order_line_id FROM t1);
  
  no rows selected
  
  SQL>

  If it's in the manual, please point to me. I fully expect it give me the error the second time around.
  
  PS:
  It was the same in 11 GR 2 Enterprise Edition too.
  
  Published by: RPuttagunta on August 16, 2011 12:53

  No, you cannot rewrite the query it.

  SELECT * FROM t2 WHERE t2.order_line_id IN (SELECT t2.order_line_id FROM t1);
  

  would be equivalent to

  SELECT *
    FROM t2
   WHERE t2.order_line_id IS NOT NULL
     AND EXISTS (
      SELECT 1
        FROM t1
      )
  

  If you assume that T1 has at least one line and that T2. ORDER_LINE_ID is NOT NULL, it is equivalent to

  SELECT *
    FROM t2
  

  For each row in T2, Oracle would choose the T2. ORDER_LINE_ID of T1. Assuming that T1 has at least one line and that ORDER_LINE_ID is NOT NULL, it will come back N rows with values not equal NULL (1 row for each row of T1). The IN, by definition, could be set to TRUE.

  Thus, while it is a valid syntax, it is very unlikely that the syntax that you'd really like.

  Justin

• How to convert a 2D Image into 3D to create a video with still photograph old 3D Photo converted video 2d style 3D (Photoshop and After Effects)

  Hello

  How to convert a 2D to 3D Image

  Create a video with always shoot in 3d

  Old Photo converted to 3D as video 2d

  (Photoshop and After Effects)

  Thank you

  It's called a cinemagraphs.com. You slice the image in the foreground and background elements, plug the holes and then place them in a 3D space - in after effects or a 3D program, no PS

  Mylenium

• I am trying to create a slideshow with Adobe Photoshop 12.  After a few episodes to work on it, some of the slide pictures disappear.  Very frustrating.  Why?  I'm to access photos from the catalogue and folders on my computer.  E

  I am trying to create a slideshow with Adobe Photoshop 12.  After a few episodes to work on it, some of the slide pictures disappear.  Very frustrating.  Why?  I'm to access photos from the catalogue and folders on my computer.  Disappearances seem random.

  Elements Photoshop Elements

• Update fails with ORA-00904 invalid identifier

  Hi all
  I have problems of construction of an ' update '. Execute the following select statement,
  

  SELECT 
   XREF.REGN_ID,
         XREF.COUNTY_NM,
         OWNER.CASE_OWNER_ID,
         OWNER.FORMER_AREA,
         OWNER.AREA,
         OWNER.COUNTY_NUMBER,
         OWNER.OFFICE
    FROM PS2_CASE_OWNER_SEYED OWNER,
         DWFSSD.TAFS_REGN_AREA_XREF_MART XREF
            WHERE    (OWNER.AREA = XREF.AREA_ID(+))
               AND (OWNER.OFFICE = XREF.OFFICE_CD(+) )
               AND (OWNER.COUNTY_NUMBER = XREF.COUNTY_NUM(+))

  Returns

  REGN_ID   COUNTY_NM                        CASE_OWNER_ID   FORMER_AREA   AREA   COUNTY_NUMBER   OFFICE   
  "1"       "Alfalfa"                        "5008756"       "1"           "1"    "02"            "C"      
  "1"       "Alfalfa"                        "5008954"       "1"           "1"    "02"            "C"      
  "1"       "Beckham"                        "5008803"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008222"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008223"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008424"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008442"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008780"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008787"       "1"           "1"    "05"            "C"      
  "1"       "Beckham"                        "5008788"       "1"           "1"    "05"            "C" 

  I am trying to execute the following statement, "Update", update the PS2_CASE_OWNER_SEYED table.

  UPDATE (
    SELECT OWNER.AREA, XREF.REGN_ID, XREF.COUNTY_NM
    FROM PS2_CASE_OWNER_SEYED OWNER, DWFSSD.TAFS_REGN_AREA_XREF_MART XREF
     WHERE    (OWNER.AREA = XREF.AREA_ID(+))
       AND (OWNER.OFFICE = XREF.OFFICE_CD(+) )
       AND (OWNER.COUNTY_NUMBER = XREF.COUNTY_NUM(+))
  )
  SET OWNER.FORMER_AREA = OWNER.AREA,
      OWNER.AREA = XREF.REGN_ID,
      OWNER.COUNTY_NAME = XREF.COUNTY_NM;

  The "Update" statement returns the following error:
  

  ORA-00904: "XREF"."COUNTY_NM": invalid identifier

  Any ideas what's wrong with the update statement?
  
  Thank you
  
  Seyed

  Hi, Seyed,

  Here's a way to do this UPDATE:

  UPDATE     emp
  SET     sal     = sal * 1.05
  ,     comm     = NVL (comm, 0) + 500
  WHERE     deptno     IN (
                 SELECT     deptno
                 FROM     dept
                 WHERE     loc     = 'DALLAS'
               )
  ;
  

  If you want to use an inline view, you can do this:

  UPDATE     (
            SELECT     e.sal
            ,     e.comm
            FROM     emp     e
            JOIN     dept     d  ON     d.deptno  = e.deptno
            WHERE     d.loc     = 'DALLAS'
       )
  SET     sal     = sal * 1.05
  ,     comm     = NVL (comm, 0) + 500
  ;
  

  Whatever it is, the emp table is left as you wish:

  EMPNO ENAME  JOB         MGR HIREDATE          SAL  COMM DEPTNO
  ----- ------ --------- ----- ---------- ---------- ----- ------
   7782 CLARK  MANAGER    7839 06/09/1981       2450           10
   7839 KING   PRESIDENT       11/17/1981       5000           10
   7934 MILLER CLERK      7782 01/23/1982       1300           10
   7876 ADAMS  CLERK      7788 05/23/1987       1155   500     20
   7902 FORD   ANALYST    7566 12/03/1981       3150   500     20
   7566 JONES  MANAGER    7839 04/02/1981    3123.75   500     20
   7788 SCOTT  ANALYST    7566 04/19/1987       3150   500     20
   7369 SMITH  CLERK      7902 12/17/1980        840   500     20
   7499 ALLEN  SALESMAN   7698 02/20/1981       1600   300     30
   7698 BLAKE  MANAGER    7839 05/01/1981       2850           30
   7900 JAMES  CLERK      7698 12/03/1981        950           30
   7654 MARTIN SALESMAN   7698 09/28/1981       1250  1400     30
   7844 TURNER SALESMAN   7698 09/08/1981       1500     0     30
   7521 WARD   SALESMAN   7698 02/22/1981       1250   500     30
  

• ORA-00904 ROW_NUMBER, problem with ORDER BY id

  Hello world
  
  I use Oracle 9i, I use ROW_NUMBER and I got the ORA-00904 because of the ORDER BY identifier.
  

  SELECT      * 
  FROM      (SELECT c.id_comp, 
            c.nom, 
            c.id_type_comp, 
            c.libelle, 
            c.standard, 
            c.id_comp_std, 
            c.id_comp_grp, 
            (SELECT      t.nom 
            FROM      type_composant t 
            WHERE      t.id_type_comp=c.id_type_comp ) 
            AS      nom_type_comp , 
            (SELECT t.libelle 
            FROM      type_composant t 
            WHERE      t.id_type_comp=c.id_type_comp ) 
            AS      libelletype, 
            (SELECT cc.nom 
            FROM      composant cc 
            WHERE      cc.id_comp=c.id_comp_std ) 
            AS      nom_comp_std, 
            (SELECT ccc.nom 
            FROM      composant ccc 
            WHERE      ccc.id_comp=c.id_comp_grp ) 
            AS      nom_comp_grp, 
            ROW_NUMBER() OVER (ORDER BY nom_type_comp) R 
       FROM composant c) 
  WHERE      nom LIKE '%' 
  AND      R BETWEEN 0 and 200 
  ORDER BY nom_type_comp;

  And here is the error:
  

   ROW_NUMBER() OVER (ORDER BY nom_type_comp) R FROM composant c)
                                      *
  ERROR at line 25:
  ORA-00904: "NOM_TYPE_COMP": invalid identifier

  Apparently, it takes the pseudonym I use. I tried with the query instead of the alias, and it worked. Is there a way to do it with the alias because I would use a variable with my ORDER BY.
  
  Any help would be nice!
  
  Andalusians
  
  Published by: Guich may 12, 2011 10:40

  You may not use an alias defined on the same query plan.
  The alias must be from a subquery.
  Such as:

  SELECT      *
  FROM      (select my_table.*, ROW_NUMBER() OVER (ORDER BY nom_type_comp) R
           from  (SELECT c.id_comp,
                   c.nom,
                   c.id_type_comp,
                   c.libelle,
                   c.standard,
                   c.id_comp_std,
                   c.id_comp_grp,
                   (SELECT      t.nom
                   FROM      type_composant t
                   WHERE      t.id_type_comp=c.id_type_comp ) AS      nom_type_comp ,
                   (SELECT t.libelle
                   FROM      type_composant t
                   WHERE      t.id_type_comp=c.id_type_comp ) AS      libelletype,
                   (SELECT cc.nom
                   FROM      composant cc
                   WHERE      cc.id_comp=c.id_comp_std ) AS      nom_comp_std,
                   (SELECT ccc.nom
                   FROM      composant ccc
                   WHERE      ccc.id_comp=c.id_comp_grp ) AS      nom_comp_grp
              FROM composant c) my_table)
  WHERE      nom LIKE '%'
  AND      R BETWEEN 0 and 200
  ORDER BY nom_type_comp;
  

  Nicolas.

• Create MVIEW fails with ORA-30036

  Hello
  
  I get the error ORA-30036 when you try to create a MVIEW:
  
  ORA-30036: unable to extend segment by 4 in undo tablespace "UNDOTBS1.
  
  I would like to understand why I get this error? is the creation of MVIEW uses the undo tablespace? What other info I can give you if it helps?
  

  CREATE MATERIALIZED VIEW MY_MVIEW
  AS
  SELECT col1,col2,col3,col4,col5,col6,col7
    FROM Table
   WHERE col1 = 'A'

  Thank you.

  Yes, creating a view Matt will use cancellation space. A materialized view actually creates a table with the results of the view, so he uses undo to save inserting rows, and to tracj, updates the allocation data dictionary and the space.

  John

• ORA 00904: invalid identifier "JAN".

  Hello

  I have the rest of the table.

  ===========================================

  create table ATT_ATTENDANCESHEET as

  (

  Select 1 empid, to_date('21/01/2014','dd/mm/yyyy') prdate, 240 reg, 0 unpaid all double union

  Select 2, to_date (January 21, 2014 ', ' dd/mm/yyyy'), 200 reg, 0 unpaid all double union

  Select 3, to_date (January 21, 2014 ', ' dd/mm/yyyy'), 240 reg, 0 unpaid all double union

  Select option 4, to_date (January 21, 2014 ', ' dd/mm/yyyy'), 480 reg, 0 unpaid all double union

  Select 5, to_date (January 21, 2014 ', ' dd/mm/yyyy'), 240 reg, unpaid double 0

  );

  =================================================

  I want to remove rows from table ATT_ATTENDANCESHEET.

  If the parameter passed empids, lines for employees would be eliminated.

  If empids is null, all records between the given period will be deleted.

  I created after the procedure to that effect.

  ------------------------------------------------------

  CREATE or replace FUNCTION deleteAttendanceDetails (startdate DATE, enddate DATE, empids varchar2) RETURN NUMBER as

  
  

  n number;

  condition varchar2 (200);

  vsql varchar2 (2000);

  BEGIN

  IF empids is not null

  THEN

  condition: condition = |' and empid in ('| empids |') ' ;

  END IF;

  vsql: =' delete from ATT_ATTENDANCESHEET where prdate between ' | StartDate | 'and' | EndDate | condition;

  EXECUTE IMMEDIATE (vsql);

  return n;

  END;

  /

  The empids parameter contains the employee IDS separated by commas.

  for example

  "1,2,3,4"

  -----------------------------------------------------------

  But when I call this function using

  Select deleteAttendanceDetails (to_date('25/01/2014','dd/mm/yyyy'), to_date (' 01/31/2014 ',' dd/mm/yyyy'), ") double;

  I get the following error

  ORA 00904: invalid identifier "JAN".

  ORA-06512: at.. .line 21

  I use oracle 10g.

  Help, please

  Do not use dynamic sql. It's more trouble that it's worth.

  Your error comes from the fact that you convert a date to a string, implicitly and then come back.

  In addition, your example doesn't have a field of prddate... so I added that, in my test table so that it works.

  Try something like that, entirely avoid dynamic sql:

  CREATE or replace FUNCTION deleteAttendanceDetails (
                                startdate DATE,
                                enddate DATE,
                                empids varchar2
                             )
     RETURN NUMBER
  as
     n number;
  BEGIN
     delete from ATT_ATTENDANCESHEET
        where ( empid IS NULL
              OR empid in ( select regexp_substr ( empids, '[^,]+', 1, level) empid
                           from dual connect by level <= (LENGTH(empids) - LENGTH(REPLACE(empids, ',')) + 1)
                        )
              )
          and prdate between nvl(startdate, to_date('01-jan-1900','dd-mon-yyyy'))
                         and nvl(enddate  , to_date('01-jan-5000','dd-mon-yyyy'));
  
    return n;  -- what is "n" ?
  END;
  /
  

  Not really sure what you want to do with "n"... you have nothing in your code... so I did the same