Name of the PK, unique constraints and to support the index

Similar Questions

• Index naming standards does not and unique constraints

  I found yet another question of naming standards

  I have tried to create a unique constraint as a normal index via the unique constraints and indexes in Properties of Table with the assumption that it will use the design properties of models.  This isn't.

  

  As noted in the image of the model above, my Unique constraint model is {table} _ {columns} lancement_remarques

  and my model of Index idx _ {columns} {table}.  Yet, when I try to create the Unique constraint, it creates the naming convention in names like CUSTOMER_ADDRESS__UK (2 lines of underscore?) as shown below.

  

  My assumption is that he needed a column, so I added the column and also applied the rules naming - nothing.

  

  Is my wrong assumption on these constraints being appointed through the models?  If this is not the case, what naming rules is in this context?

  The same behavior is evident when you try to create index.

  I get not all comments on my posts.  Are these bugs or just user error?

  Hello

  Thank you to report the problem, I logged a bug. You can get 'Rules of naming' works properly if you press the button 'Apply' before that.

  I get not all comments on my posts.

  I don't think that anyone (or any job) has been ignored on this forum. Depending on the complexity of the problem and our current commitment (other urgent tasks, holidays, time zone), you can get answer almost immediately or after a certain time.

  It is useful for us, if you mark your messages as replied if you get the right answer.

  Philippe

• Bad alphabetically in the index

  Dear community,

  I use ID CS6.

  I have discovered a reproducible error, given the order of the names in the index sheets.

  When I put marks index to names

  • Anton huh
  • Martin huh
  • Johannes Heindl

  in this way they will be sorted:

  1. Huh, Anton
  2. Heindl, Johannes
  3. Huh, Martin

  which is obviously incorrect. The order should rather be 1,3,2.

  I think the problem is that the spaces between the first and last name are ignored:

  The names "Smith, John" is treated as "Smithjohn".

  Corsica, I can force ID to sort correctly by using the "sort by"-field. But why is that need?

  Names are generally classified by NAME and if they are the same, you look on the first NAME. Word by word instead of the letter by letter.

  Are there really no feature that provides "the word of the order of the words" without interfering manually?

  Thank you in advance!

  I have the honour, of course, you put these scripts to use, but they serve a particular purpose and they mess around with the document considerably. Because everything what needs index of martin is setting/updating the fields from the sort order, it is easier to target the index panel immediately: it is faster and safer.

  You are partially right that remove the space after that a comma sorts the sorting problem, but using the sign $ is delicate because you might want to do things later, and symbols that have a special meaning in regular expressions are to be avoided if possible, just to make life easier. So, I use the number 5 instead. The advantage is that I can use different numbers to refine the sort order. Here's the script:

  (function () {
    if (app.documents[0].indexes.length > 0) {
      var order;
      var topics = app.documents[0].indexes[0].allTopics;
      for (var i = topics.length-1; i >= 0; i--) {
        order = topics[i].sortOrder;
        // If the topic already has a sort order defined, change it
        if (order !== '' && order.indexOf(', ') > -1) {
          topics[i].sortOrder = order.replace (', ', ',5');
        // Otherwise use the topic's name
        } else if (topics[i].name.indexOf(', ') > -1) {
          topics[i].sortOrder = topics[i].name.replace (', ', ',5');
        }
      }
    }
  }());
  

  Peter

• removal of the index

  Hello
  
  I want to delete the index of a column, but I don't know the name of the index of this column, I only know the table name and the column name that is as shown below:
  
  C2M_CLIENT_H (RATING_ID)
  
  How can I write the query?

  SELECT POSITION_COLONNE,
  COLUMN_NAME,
  ALL_IND_COLUMNS A.* HAS
  WHERE TABLE_NAME LIKE '% C2M_CLIENT_H '.
  ORDER BY 3,4,1;

• composite unique constraint on the values of parent and child?

  Is it possible to have a composite unique constraint that contains the values of the child elements? The example below has the "child" elements are offline, but it's preferred, but optional, I know that you can have a unique constraint in the set of tables without using a reference table that contains the constraint and the two columns. How xdb manages this requirement?
  
  permit:

  <parent ID="1">
     <child><name>test1</name></child>
     <child><name>test2</name></child>
  </parent>
  <parent ID="2">
     <child><name>test1</name></child>
     <child><name>test2</name></child>
  </parent>

  not allowed:

  <parent ID="1">
     <child><name>test1</name></child>
     <child><name>test1</name></child>
  </parent>

  <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" 
             xmlns:xdb="http://xmlns.oracle.com/xdb"
             xdb:storeVarrayAsTable="true"
             elementFormDefault="qualified">
      
      <xs:element name="parent" type="Parent_T"
          xdb:columnProps="CONSTRAINT parent_pkey PRIMARY KEY (XMLDATA.ID)"
          xdb:defaultTable="PARENT"/>
  
      <xs:complexType name="Parent_T" xdb:SQLType="PARENT_T" xdb:maintainDOM="false">
          <xs:sequence>
              <xs:element name="child" type="Child_T" minOccurs="1" maxOccurs="unbounded" xdb:SQLName="CHILD"
                        xdb:SQLInline="false" xdb:defaultTable="CHILD" "/>
          </xs:sequence>
          <xs:attribute name="ID" xdb:SQLName="ID" use="required" />
      </xs:complexType>
      
      <xs:complexType name="Child_T" xdb:SQLType="CHILD_T">
         <xs:sequence>
           <xs:element name="name" type="xs:string" xdb:SQLName="NAME"/>
         </xs:sequence>
       </xs:complexType>     
  </xs:schema>

  xdb:columnProps = "CONSTRAINT parent_pkey PRIMARY KEY (XMLDATA.ID), * UNIQUE (XMLDATA.» "Child.Name) *" triggers the non-existent attribute
  
  
  A possible solution would be to copy the value of the primary key parent of the child element, then I could create a composite unique constraint using only the values of the child. However, I have this same requirement elsewhere in my lowest nested schema, and it can become messy / bad design with cascading of all primary keys on the schema. For example, I have a recursive element in which two attributes must be unique only within the parent company:

  <parent id="1">
     <child a="1" b="1">
        <child a="1" b="2">
           <child a="1" b="1" /> *not allowed
        </child>
     </child>
     <child a="1" b="2" /> *not allowed
  </parent>

  Possible solution:

  <child a="1" b="2" parent_id="1" />
  <xs:complexType name="Child_T>
     <xs:sequence>
        <xs:element name="child" xsd:SQLInline="false" xsd:columnProps="UNIQUE(XMLDATA.a,XMLDATA.b,XMLDATA.parent_id)" minOccurs="0" maxOccurs="unbounded" type="Child_T">
     </xs:sequence>
     </xs:element
  </xs:complexType>

  Is there a better design?
  
  

  Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - 64bi 
  PL/SQL Release 10.2.0.4.0 - Production                           
  CORE     10.2.0.4.0     Production                                       
  TNS for Linux: Version 10.2.0.4.0 - Production                   
  NLSRTL Version 10.2.0.4.0 - Production 

  You can do something like this:

  SQL> DECLARE
    2
    3    xsd_doc xmltype := xmltype('
    4  
    5    
    6      
    7        
    8          
    9        
   10      
   11    
   12    
   13      
   14        
   15          
   16        
   17      
   18    
   19    
   20      
   21        
   22          
   23        
   24        
   25      
   26    
   27    
   28      
   29        
   30          
   31        
   32      
   33    
   34  ');
   35
   36  BEGIN
   37
   38    dbms_xmlschema.registerSchema(
   39      schemaURL => 'test_parent.xsd',
   40      schemaDoc => xsd_doc,
   41      local => true,
   42      genTypes => true,
   43      genbean => false,
   44      genTables => false,
   45      enableHierarchy => dbms_xmlschema.ENABLE_HIERARCHY_NONE
   46    );
   47
   48  END;
   49  /
  
  PL/SQL procedure successfully completed
  
  SQL> CREATE TABLE my_xml_table OF XMLTYPE
    2  XMLTYPE STORE AS OBJECT RELATIONAL
    3  XMLSCHEMA "test_parent.xsd"
    4  ELEMENT "root"
    5  VARRAY xmldata."parent" STORE AS TABLE my_parent_tab
    6  (
    7    VARRAY "child" STORE AS TABLE my_child_tab
    8  )
    9  ;
  
  Table created
  
  SQL> ALTER TABLE my_parent_tab ADD CONSTRAINT parent_uk UNIQUE (nested_table_id, "ID");
  
  Table altered
  
  SQL> ALTER TABLE my_child_tab ADD CONSTRAINT child_uk UNIQUE (nested_table_id, "name");
  
  Table altered
   
  

  Then:

  SQL> insert into my_xml_table values (
    2  xmltype('
    3     test1
    4     test2
    5  
    6  
    7     test1
    8     test2
    9  ')
   10  );
  insert into my_xml_table values (
  *
  ERREUR à la ligne 1 :
  ORA-00001: violation de contrainte unique (DEV.PARENT_UK)
  
  SQL> insert into my_xml_table values (
    2  xmltype('
    3     test1
    4     test1
    5  
    6  
    7     test1
    8     test2
    9  ')
   10  );
  insert into my_xml_table values (
  *
  ERREUR à la ligne 1 :
  ORA-00001: violation de contrainte unique (DEV.CHILD_UK)
  
  SQL> insert into my_xml_table values (
    2  xmltype('
    3     test1
    4     test2
    5  
    6  
    7     test1
    8     test2
    9  ')
   10  );
  
  1 ligne créée.
  

  http://download.Oracle.com/docs/CD/B19306_01/AppDev.102/b14259/xdb06stt.htm#sthref987

• XControls: Options for change "Unique identifier" and the "localized name Long' properties has disappeared

  Is it only me or is a new bug in LabView? Miss me the options to change the "Unique identifier" and the "Long name" real estate XControl. In the past they where in the red zone, now they're gone!

  I use LV2009SP1.

  Any ideas?

  Ingo Bartsch

  

  Thanks for your efforts, Duffy2007!

  Problem was a corrupted in some sort of XControl or project file. After re - import the XControl in a new project, all was well again.

  Kind regards

  Ingo

• After scanning in a file, I can't save the PDF file with a unique file name, it will only back up in its original name; Otherwise, if I save the name of the file under a new name, save it box goes outside and the file is renamed and saved. Help!

  If I analyze the file and its original name is:

  Ssrv395alm76439... and I want to rename it "background."

  If I have the rename, the file won't save. If I leave the file as it will save and I have to go to an additional folder and rename it.

  Help, it's too long. I just click on "Save as" and rename and save the document in the necessary folder at a time!

  Hi Allisonchristinele ,

  Please provide the exact point of the software & OS installed on your system. Also check if it is there any update available for the software after crossing "" help > check updates "»

  Follow this thread to reset the Acrobat preferences:- How to reset preference settings in format Acrobat.

  Like 'save' will save the changes with the file & "save under" will give the window using which can save the file to any location on the system.

  I would like to know is what is happening with some or all the files.

  Also, please try to check with the different user account to reproduce the problem.

  Kind regards

  Christian

• Find the name and the table constraint type based on the name of the field

  I have a field name or a column with me... I want to find the name of the table is a primary key... Any request to make this work so that the running costs are low...

  Currently, I mix dba_constraints and ALL_TAB_COLUMNS to get this... is possible to extract the data of a process cost effective single table/any other is appreciated

  You must include the USER/ALL_CONSTRAINTS with USER/ALL_CONS_COLUMNS to achieve your desired results. For your specific condition, you can try this query.

  Select constraint_name

  column_name

  de)

  Select c.constraint_name

  cc.column_name

  count on column_count (c.constraint_name score) (cc.column_name)

  from user_constraints c

  Join user_cons_columns cc

  on c.constraint_name = cc.constraint_name

  where constraint_type = 'P '.

  )

  where column_name = 'EMPNO '.

  and column_count = 1;

  I used USER_CONSTRAINTS and USER_CONS_COLUMNS. If you use ALL or DBA dictionary table add join OWNER condition as well.

• ORA-01031: insufficient privileges during the activation of a unique constraint

  Hello
  DB is Oracle 10 g.
  I have a problem, which allows a unique constraint on a table using a stored procedure.
  The procedure travels a few queried from the view all_constraints primary key constraints allowing them without any problem.
  
  Then, on the first unique constraints, it throws the exception
  ORA-01031: insufficient privileges
  
  The sql line built at the source of the problem is
  run immediately ' alter table 'My_Schema_Name '. "' My_TableName ' enable the 'UNIQUE_Myconstraint_UQ' CONSTRAINT '.
  
  What confuses me, is that the same line (without "immediate execution") run from the console window of the sql developers, which was used to make the appeal process that has failed, will succeed.
  
  The user used to make the call is not owner of the scheme, but has Alter_Any_Table system permissions (and, as mentioned above, the call will succeed if entered directly in the sql developer worksheet).
  
  Stored procedures running in a different (limited) security context (sorry if the setting isn't right) than that of the calling user?
  Or what is the problem here?
  
  Any help is appreciated.
  Markus.
  
  Note:
  Before, I don't have the name of constraint in quotes and allowing the primary key failed with error ORA-02430: impossible to enable the constraint (PK_ConstraintName) - no such constraint 02430. 00000 - "could not activate constraint (%s): no such constraint ', which is an interesting problem in itself, but now fixed. In any case, thoughtI would mention it

  Hello

  This means that alter table system privilege is given to the user through certain roles. If it's true that the roles has no effect when executing stored procedures. You must give the direct privilege to this user and not by role.

  With regard to your note:
  This means that you created the constraint in lower case or mix of lower and upper case in quotes. Therefore you should now always return in the same case in quotes each time you reference any object. So it is always advisable to create objects on top of case (if the use of the quotes) or do not use quotation marks.

  Concerning
  Anurag

• Problem on the constraints and validation

  Hello
  I want to add a new unique constraint in a table. But the current data do not match the coercion.
  My constraint is like this:
  
  ALTER TABLE T_STUDENT ADD
  CONSTRAINT UNIQ_STUDENT
  UNIQUE (STUDENT_NUMBER, NAME, FIRST NAME)
  DISABLE
  VALIDATE;
  
  I want to add the constraint, and I don't want to update the data. So, how can I add constraint without validating the current data?

  user4030266 wrote:
  Hello

  Hello

  I want to add a new unique constraint in a table. But the current data do not match the coercion.
  I want to add the constraint, and I don't want to update the data. So, how can I add constraint without validating the current data?

  Try like this:

  ALTER TABLE t_student ADD
   CONSTRAINT uniq_student
    UNIQUE (student_number, NAME, surname)
  DEFERRABLE INITIALLY DEFERRED
    NOVALIDATE;
  

  +
  It might be interesting for you
  http://www.Oracle.com/technology/oramag/Oracle/03-Nov/o63asktom.html

• Deployment of ADFLibrary: provide a unique name for the resources in this project

  JDeveloper 12 c.

  I'm deploying an application as a library ADF - it will be used my multiple applications.  So it's a bit generic. Sometimes it is OK and other times it gets stuck... See the ChangeLog

  What do mean exactly by ' resourceprovide a unique name of the indicated "? 

  What should I name the files?  xyz_abc_appNotices.jsff or the directory structure has to change?

  Thank you!

  Stuart

  [16: 38:44]-deployment began.  ----

  [16: 38:44] the target platform is Java EE Standard.

  [16: 38:44] analysis of the dependence running...

  [16: 38:44] building...

  [16: 38:48] deployment of profile...

  [16: 38:49] WARNING: the commonly named resources below could cause errors of reusability to your consumers.

  Provide a unique name for the resources in this project, or start with a different project.

  appNotices.jsff

  appNoticesMain.jsf

  [16: 38:49] Mobile WEB-INF/adfc - config.xml for META-INF/adfc-config. XML

  Well, Yes, it would do. The warning say you that your library resources should use unique names, that I hope are not used by other libraries. From the resources app Strip is common chance is that it will be a duplicate.

  Timo

• Unique constraint on the values in both directions

  I'm looking to create a unique constraint that works in two ways. Say I got a constraint unique in columns 1 and 2. I want it to be impossible for the two lines below the two exist at the same time. Is there a way to do this? I googled around for a while now and I found nothing that works so far.

  Header 1	Header 2
  Dog	CAT
  CAT	Dog

  Hello

  You can create an index based on a single function:, like this:

  CREATE UNIQUE INDEX table_x_header_1_header_2

  ON table_x (LESS (header_1, header_2)

  More LARGE (header_1, header_2)

  );

  How will you use these values?  You might be better to simply create a regular old unique constraint, but also have a CHECK constraint to ensure that header_1<= header_2. ="" that="" way,="" when="" you="" want="" to="" search="" for="" the="" combination="" ('cat',="" 'dog')="" you="" won't="" have="" to="" search="" for="" ('dog',="" 'cat')="">

• What is the order to turn off the constraints and index

  Hi, I need to make some out of constraints and indexes on tables to load quite a date in them. I know that I can just drop them and recreate later. But is it feasible to figure who forced and in what order must be do disable or unusable before a load and after the load that is to say. without dropping to preserve? R before P data loading, and then P before R like I think that controls can jump in evertytime and are quite safe to deal with. But I did not carefully think about more, maybe you have some scripts that say exactly what should be the order to turn off the coast and later on this subject.

  Thank you!

  Hello

  946279 wrote:

  Hi, I need to make some out of constraints and indexes on tables to load quite a date in them. I know that I can just drop them and recreate later. But is it feasible to figure who forced and in what order must be do disable or unusable before a load and after the load that is to say. without dropping to preserve? R before P data loading, and then P before R like I think that controls can jump in evertytime and are quite safe to deal with. But I did not carefully think about more, maybe you have some scripts that say exactly what should be the order to turn off the coast and later on this subject.

  Thank you!

  That's right: disable foreign key constraints before disabling the primary (or Unique) key constraint correspondent.  When you reactivate them, re - enable the constraint of primary (or Unique) key before the corresponding foreign key constraints.

  Aside from that, the order should not issue.

• total suite of employees in the Department and all sections of it, but display only the name of the Department

  Hi all

  We use the Oracle 10 g R2 64 bits on 64-bit windows server 2008.

  I have the following tables

  CREATE TABLE dept
     ( DEPARTMENT_ID NUMBER primary key, 
    DEPARTMENT_NAME  VARCHAR2(100) NOT NULL, 
  DEPARTMENT_TYPE CHAR(1) NOT NULL, /* S for section, D for DEPARTMENT*/ 
    REFERENCE_DEPARTMENT NUMBER/* Self reference to department_id */ 
  );
  alter table dept add constraint fk_dept_id_reference_dept foreign key REFERENCE_DEPARTMENT references dept(department_id);
  
  insert into dept values (1,'ETC','D',Null);
  insert into dept values (2,'CSS','S',1);
  insert into dept values (3,'ESS','S',1);
  commit;
  
  create table emp (
  empno number primary key,
  ename varchar2(1000),
  department_id number references dept(department_id)
  );
  insert into emp values (1,'Jones',1);
  insert into emp values (2,'Smith',2);
  insert into emp values (3,'Scott',3);
  
  commit;
  

  Now we need a query that selects all employees, in each Department and the name of the service

  We use the following query

  select count(e.empno) Total_Employees,d.department_name from emp e, dept d
  where e.department_id=d.department_id;
  

  The result shows as well the section names, while we want to display a result as below

  Total_Employees         Department_Name
  3                                   ETC
  

  We do not want to display the names of section, of course we want to count employees in all sections of a Ministry, or we only want to display the name of the Department.

  I hope I'm clear in the description of my problem.

  Thank you.

  Hello

  Maahjoor wrote:

  I have 14 departments and nearly 25 sections.

  I can't just include every department as HE said in HIS application

  1. Select department_id 1, "ETC." DEPARTMENT_NAME 'd' DEPARTMENT_TYPE, Null...

  
  

  I have many ministries and numerous sections.

  If I include all services as IT were included in the above sentence, my request will become 2-3 hundreds of lines.

  
  

  IT should use only the names of actual columns in your query, not the service name, for example ETC.

  
  

  concerning
  

  No, the length of the query is the same, regardless of the number of rows in the tables.  As Etbin explained in response to #8, emp and dept subqueries are not part of the solution; they are just average Etbin of testing the solution without creating any actual table.  Since you have real tables, you can start with the following subquery solution, i.e.:

  WITH the_data as

  (select d.department_id, d.department_name, d.department_type, d.reference_department, c.emp_count

  by d Dept.

  ...

• Unique constraint in the form of master detail error

  Hi all

  I need help, the following requirement.

  I have a master detail form developed on master-child table. the tables have the composite key.

  Old masters has a composite key on columns (A, B)

  Children table has a composite key on columns (A, B, C)

  Child block look something like below

  C       A          B

  10 AAA 1000

  20 1300 BBB

  30 CCC 1400

  40 DDD 1200

  Increments of column C with 10 for each record, and if a new record is insert in intermediaries the records it is incremented to 5.

  My requirement is when an end user attempts to insert record between 20 and 30 or 30 and 40 and clicks on save, the value of the C column must regenerate as shown below

  C       A          B

  10 AAA 1000

  20 1300 BBB

  XXX 30 900

  40 CCC 1400

  50 DDD 1200

  Button Save I wrote the following code

  Declare
    ln_Count NUMBER :=0;
    ln_c number :=0
    cursor c1 is
    select c
    from child
    where a=:child.a
    and b=:child.b
  Begin
    Go_Block('Child');
    First_Record;
    LOOP
    ln_Count:= ln_Count+10;
    :child.C := ln_Count;
    EXIT When :System.Last_Record = 'TRUE';
    Next_Record;
    END LOOP;
  
  
    For c_cur in c1 Loop
    update child
    set c:=ln_c+10
    where a=:child.a
    and b=:child.b 
    and c=c_cur.c
    end loop;
    Forms_DDL('commit');
       
       commit_prc('Commit');  -- We have our own program unit to call commit_form
  
  END;
  

  I tried above in a way because, before approving the changes to the table, I update the existing values in the table of the C column so I would not get unique constraint error.

  When you click the button Save, I get a constraint exception. Hope I made my requirement clear.

  Can someone give me a clue to this implementation.

  Thank you

  malandain

  With this update, all your C-columnvalue became negative. When you post the form thereafter, forms update agaion records one by one the new positive figures. Because the numbers of 'old' in the db are now negative, there will be no violation-UK.