Cardinality in FK relations of relational model

I'm using the latest Data Modeler and tryint to set the cardinality and mandatory property of an FK relation. At first I thought its impossible to do in the Relational model and only possible in Logical, but I can see this being done in PDF diagrams from "\datamodeler\reports\Reporting Schema diagrams" folder. How can I do that? Thanks.

I'm using the latest Data Modeler you can download 3.0 EA2 it was released few hours ago.
You can set mandatory(optional) in foreign key dialog - double click on FK line or use "Properties" in pop-up menu for FK.
There are no other cardinality settings for foreign keys. However in EA2 they will be shown without crow's foot if you set unique/primary constraint over FK columns only.
Philip

Similar Messages

Changing relationship cardinality in relational model

Using SDDM 3.3.0.747.
When a foreign key relationship is defined between two tables in the relational model, the cardinality symbol displays as one-to-many by default.
Is there a way to specify that the relationship, and therefore the cardinality symbol, should be something different, such as one-to-one?

Kent and Philip, thank you. That's exactly what I was looking for. "Correct" answer to Philip because it worked and he replied first. "Helpful" points to Kent.
And Kent, that's all I needed. Thank you for going further and asking.

Select schemas from relational model on import from data dictionary option

Hi All,
I have one relational model with 3 diferent schemas,
I want to compare one of my schemas with the data dictionary I have in a database,
I select the import option in the general file menu, select from data dictionary option,
select the connection from my database,swap target model checked,select a physical model and select the objects i want to compare from the database,
My problem is that the result is the comparison between all the objects in my model and the objects in the database that I have selected,
what I really want is to compare a list of objects in my model to a list of objects in my database,
this could be possible? or always need to compare all the objects of the model?
Thanks in advance

Hi jbellver,
there is no any development in DM 3.1.0.691 on that problem. In production release you'll be able to compare objects in subview with database or just to select several objects and compare them with database. And of course "generate in DDl" filtering still can be used - it works at "Compare dialog" level.
Philip

How can i import tables from a different schema into the existing relational model... to add these tables in the existing model? plss help

how can i import tables from a different schema into the existing relational model... to add these tables in the existing relational/logical model? plss help
note; I already have the relational/logical model ready from one schema... and I need to add few more tables to this relational/logical model
can I import the same way as I did previously??
but even if I do the same how can I add it in the model?? as the logical model has already been engineered..
please help ...
thanks

Hi,
Before you start, you should probably take a backup copy of your design (the .dmd file and associated folder), in case the update does not work out as you had hoped.
You need to use Import > Data Dictionary again, to start the Data Dictionary Import Wizard.
In step 1 use a suitable database connection that can access the relevant table definitions.
In step 2 select the schema (or schemas) to import. The "Import to" field in the lower left part of the main panel allows you to select which existing Relational Model to import into (or to specify that a new Relational Model is to be created).
In step 3 select the tables to import. (Note that if there are an Foreign Key constraints between the new tables and any tables you had previously imported, you should also include the previous tables, otherwise the Foreign Key constraints will not be imported.)
After the import itself has completed, the "Compare Models" dialog is displayed. This shows the differences between the model being imported and the previous state of the model, and allows you to select which changes are to be applied.
Just selecting the Merge button should apply all the additions and changes in the new import.
Having updated your Relational Model, you can then update your Logical Model. To do this you repeat the "Engineer to Logical Model". This displays the "Engineer to Logical Model" dialog, which shows the changes which will be applied to the Logical Model, and allows you to select which changes are to be applied.
Just selecting the Engineer button should apply all the additions and changes.
I hope this helps you achieve what you want.
David

How to find the level of each child table in a relational model?

Earthlings,
I need your help and I know that, 'yes, we can change'. Change this thread to a answered question.
So: How to find the level of each child table in a relational model?
I have a relacional database (9.2), all right?!
     O /* This is a child who makes N references to each of the follow N parent tables (here: three), and so on. */
    /↑\ Fks
   O"O O" <-- level 2 for first table (circle)
/↑\ Fks
"o"o"o" <-- level 1 for middle table (circle)
   ↑ Fk
"º"Tips:
- each circle represents a table;
- red tables no have foreign key
- the table in first line of tree, for example, has level 3, but when 3 becomes N? How much is N? This's the question.
I started thinking about the following:
First I have to know how to take the children:
select distinct child.table_name child
from all_cons_columns father
join all_cons_columns child
using (owner, position)
join (select child.owner,
               child.constraint_name fk,
               child.table_name child,
               child.r_constraint_name pk,
               father.table_name father
          from all_constraints father, all_constraints child
         where child.r_owner = father.owner
           and child.r_constraint_name = father.constraint_name
           and father.constraint_type in ('P', 'U')
           and child.constraint_type = 'R'
           and child.owner = 'OWNER') aux
using (owner)
where child.constraint_name = aux.fk
   and child.table_name = aux.child
   and father.constraint_name = aux.pk
   and father.table_name = aux.father;Thinking...
Let's Share!
My thanks in advance,
Philips
Edited by: BluShadow on 01-Apr-2011 15:08
formatted the code and the hierarchy for readbility

Justin,
Understood.
Nocycle not work in 9.2 and, even that would work, would not be appropriate.
With your help, I decided a much simpler way (but there is still a small problem, <font color=red>IN RED</font>):
-- 1
declare
type udt_roles is table of varchar2(30) index by pls_integer;
cRoles udt_roles;
begin
execute immediate 'create user philips
    identified by philips';
select granted_role bulk collect
    into cRoles
    from user_role_privs
   where username = user;
for i in cRoles.first .. cRoles.count loop
    execute immediate 'grant ' || cRoles(i) || ' to philips';
end loop;
end;
-- 2
create table philips.root1(root1_id number,
                           constraint root1_id_pk primary key(root1_id)
                           enable);
grant all on philips.root1 to philips;
create or replace trigger philips.tgr_root1
   before delete or insert or update on philips.root1
   begin
     null;
   end;
create table philips.root2(root2_id number,
                           constraint root2_id_pk primary key(root2_id)
                           enable);
grant all on philips.root2 to philips;
create or replace trigger philips.tgr_root2
   before delete or insert or update on philips.root2
   begin
     null;
   end;
create table philips.node1(node1_id number,
                           root1_id number,
                           node2_id number,
                           node4_id number,
                           constraint node1_id_pk primary key(node1_id)
                           enable,
                           constraint n1_r1_id_fk foreign key(root1_id)
                           references philips.root1(root1_id) enable,
                           constraint n1_n2_id_fk foreign key(node2_id)
                           references philips.node2(node2_id) enable,
                           constraint n1_n4_id_fk foreign key(node4_id)
                           references philips.node4(node4_id) enable);
grant all on philips.node1 to philips;
create or replace trigger philips.tgr_node1
   before delete or insert or update on philips.node1
   begin
     null;
   end;
create table philips.node2(node2_id number,
                           root1_id number,
                           node3_id number,
                           constraint node2_id_pk primary key(node2_id)
                           enable,
                           constraint n2_r1_id_fk foreign key(root1_id)
                           references philips.root1(root1_id) enable,
                           constraint n2_n3_id_fk foreign key(node3_id)
                           references philips.node3(node3_id) enable);
grant all on philips.node2 to philips;
create or replace trigger philips.tgr_node2
   before delete or insert or update on philips.node2
   begin
     null;
   end;
create table philips.node3(node3_id number,
                           root2_id number,
                           constraint node3_id_pk primary key(node3_id)
                           enable,
                           constraint n3_r2_id_fk foreign key(root2_id)
                           references philips.root2(root2_id) enable);
grant all on philips.node3 to philips;
create or replace trigger philips.tgr_node3
   before delete or insert or update on philips.node3
   begin
     null;
   end;
create table philips.node4(node4_id number,
                           node2_id number,
                           constraint node4_id_pk primary key(node4_id)
                           enable,
                           constraint n4_n2_id_fk foreign key(node2_id)
                           references philips.node2(node2_id) enable);
grant all on philips.node4 to philips;
create or replace trigger philips.tgr_node4
   before delete or insert or update on philips.node4
   begin
     null;
   end;
-- out of the relational model
create table philips.node5(node5_id number,
                           constraint node5_id_pk primary key(node5_id)
                           enable);
grant all on philips.node5 to philips;
create or replace trigger philips.tgr_node5
   before delete or insert or update on philips.node5
   begin
     null;
   end;
-- 3
create table philips.dictionary(table_name varchar2(30));
insert into philips.dictionary values ('ROOT1');
insert into philips.dictionary values ('ROOT2');
insert into philips.dictionary values ('NODE1');
insert into philips.dictionary values ('NODE2');
insert into philips.dictionary values ('NODE3');
insert into philips.dictionary values ('NODE4');
insert into philips.dictionary values ('NODE5');
--4
create or replace package body philips.pck_restore_philips as
procedure sp_select_tables is
    aExportTablesPhilips     utl_file.file_type := null; -- file to write DDL of tables
    aExportReferencesPhilips utl_file.file_type := null; -- file to write DDL of references
    aExportIndexesPhilips    utl_file.file_type := null; -- file to write DDL of indexes
    aExportGrantsPhilips     utl_file.file_type := null; -- file to write DDL of grants
    aExportTriggersPhilips   utl_file.file_type := null; -- file to write DDL of triggers
    sDirectory               varchar2(100) := '/app/oracle/admin/tace/utlfile'; -- directory \\bmduhom01or02
    cTables                  udt_tables; -- collection to store table names for the relational depth
begin
    -- omits all referential constraints:
    dbms_metadata.set_transform_param(dbms_metadata.session_transform, 'REF_CONSTRAINTS', false);
    -- omits segment attributes (physical attributes, storage attributes, tablespace, logging):
    dbms_metadata.set_transform_param(dbms_metadata.session_transform, 'SEGMENT_ATTRIBUTES', false);
    -- append a SQL terminator (; or /) to each DDL statement:
    dbms_metadata.set_transform_param(dbms_metadata.session_transform, 'SQLTERMINATOR', true);
    -- create/open files for export DDL:
    aExportTablesPhilips := utl_file.fopen(sDirectory, 'DDLTablesPhilips.pdc', 'w', 32767);
    aExportReferencesPhilips := utl_file.fopen(sDirectory, 'DDLReferencesPhilips.pdc', 'w', 32767);
    aExportIndexesPhilips := utl_file.fopen(sDirectory, 'DDLIndexesPhilips.pdc', 'w', 32767);
    aExportGrantsPhilips := utl_file.fopen(sDirectory, 'DDLGrantsPhilips.pdc', 'w', 32767);
    aExportTriggersPhilips := utl_file.fopen(sDirectory, 'DDLTriggersPhilips.pdc', 'w', 32767);
    select d.table_name bulk collect
      into cTables -- collection with the names of tables in the schema philips
      from all_tables t, philips.dictionary d
     where owner = 'PHILIPS'
       and t.table_name = d.table_name;
    -- execution
    sp_seeks_ddl(aExportTablesPhilips,
                 aExportReferencesPhilips,
                 aExportIndexesPhilips,
                 aExportGrantsPhilips,
                 aExportTriggersPhilips,
                 cTables);
    -- closes all files
    utl_file.fclose_all;
end sp_select_tables;
procedure sp_seeks_ddl(aExportTablesPhilips     in utl_file.file_type,
                         aExportReferencesPhilips in utl_file.file_type,
                         aExportIndexesPhilips    in utl_file.file_type,
                         aExportGrantsPhilips     in utl_file.file_type,
                         aExportTriggersPhilips   in utl_file.file_type,
                         cTables                  in out nocopy udt_tables) is
    cDDL       clob := null; -- colletion to save DDL
    plIndex    pls_integer := null;
    sTableName varchar(30) := null;
begin
    for i in cTables.first .. cTables.count loop
      plIndex    := i;
      sTableName := cTables(plIndex);
       * Retrieves the DDL and the dependent DDL into cDDL clob       *
      * for the selected table in the collection, and writes to file.*
      begin
        cDDL := dbms_metadata.get_ddl('TABLE', sTableName, 'PHILIPS');
        sp_writes_ddl(aExportTablesPHILIPS, cDDL);
      exception
        when dbms_metadata.object_not_found then
          null;
      end;
      begin
        cDDL := dbms_metadata.get_dependent_ddl('REF_CONSTRAINT', sTableName, 'PHILIPS');
        sp_writes_ddl(aExportReferencesPhilips, cDDL);
      exception
        when dbms_metadata.object_not_found2 then
          null;
      end;
      begin
        cDDL := dbms_metadata.get_dependent_ddl('INDEX', sTableName, 'PHILIPS');
        sp_writes_ddl(aExportIndexesPhilips, cDDL);
      exception
        when dbms_metadata.object_not_found2 then
          null;
      end;
      begin
        cDDL := dbms_metadata.get_dependent_ddl('OBJECT_GRANT', sTableName, 'PHILIPS');
        sp_writes_ddl(aExportGrantsPhilips, cDDL);
      exception
        when dbms_metadata.object_not_found2 then
          null;
      end;
      begin
        cDDL := dbms_metadata.get_dependent_ddl('TRIGGER', sTableName, 'PHILIPS');
        sp_writes_ddl(aExportTriggersPhilips, cDDL);
      exception
        when dbms_metadata.object_not_found2 then
          null;
      end;
    end loop;
end sp_seeks_ddl;
procedure sp_writes_ddl(aExport in utl_file.file_type,
                          cDDL    in out nocopy clob) is
    pLengthDDL pls_integer := length(cDDL);
    plQuotient pls_integer := null;
    plRemainder pls_integer := null;
begin
      * Register variables to control the amount of lines needed   *
     * for each DDL and the remaining characters to the last row. *
    select trunc(pLengthDDL / 32766), mod(pLengthDDL, 32766)
      into plQuotient, plRemainder
      from dual;
      * Join DDL in the export file.                            *
     * ps. 32766 characters + 1 character for each line break. *
    -- if the size of the DDL is greater than or equal to limit the line ...
    if plQuotient >= 1 then
      -- loops for substring (lines of 32766 characters + 1 break character):
      for i in 1 .. plQuotient loop
        utl_file.put_line(aExport, substr(cDDL, 1, 32766));
        -- removes the last line, of clob, recorded in the buffer:
        cDDL := substr(cDDL, 32767, length(cDDL) - 32766);
      end loop;
    end if;
      * If any remains or the number of characters is less than the threshold (quotient = 0), *
     * no need to substring.                                                                 *
    if plRemainder > 0 then
      utl_file.put_line(aExport, cDDL);
    end if;
    -- record DDL buffered in the export file:
    utl_file.fflush(aExport);
end sp_writes_ddl;
begin
-- executes main procedure:
sp_select_tables;
end pck_restore_philips;<font color="red">The problem is that I still have ...
When creating the primary key index is created and this is repeated in the file indexes.
How to avoid?</font>

Sql datamodeler: Engineer to relational model: General Options

Hello
When you engineer your logical model to a relational model you get a wizard. In lower part of this wizard byou have several tabs, one of the tabs is called general.
On this general tab you have check box: Apply Name Translation
The corresponding help text:
Apply Name Translation: Controls whether formal names are translated to abbreviated names when the logical model is forward engineered to a relational model, and whether abbreviated names are translated to format names when a relational model is reverse engineered to the logical model. Name translation is applied only for valid names. In addition, translations between the words entity/attribute/key and table/column/index are performed.
Does anybody have a clue what is meant here?
Regards Erik

Erik,
you have to create glossary (tools>Glossary editor) with words permitted for usage in names in logical model. For each word you can define abbreviation that will be used in transformation process.
Example:
Employee - EMP
Salary - SAL
Attribute "Employee Salary" will be transformed to column "EMP_SAL".
You can define preferred abbreviation for entities and attributes and it can be used instead of glossary definitions.
You also have to set your glossary in "Tools>General Options>Naming standards>Glossary"
Best regards,
Philip

Logical model fwd engineer to relational model. Missing FK relationships

We are experiencing missing relationships in the relational model when engineering from logical model to the relational model. The logical model objects have properly defined primary keys defined. We are using SQL Data Modeler v3.1.700. Any suggestions on how to engineering these relationships without deleting the objects from the relational model.
Thanks,
Ed

Hi Ed,
I presume you had some relationships in the Logical Model that you were expecting to be engineered to Foreign Key relationships in the Relational Model.
Please can you provide more detail to help resolve your problem. For example:
Was the problem on an initial engineer from the logical model? Or on a re-engineer?
Are you using Entity subtyping in your Logical Model?
Are there any relevant error messages in the log file? (This is normally file datamodeler.log in folder datamodeler\datamodeler\log.)
David

Relational Models Sequence Order

Is there any way to amend the display order of relational models that are listed under the 'Relational Models' node? - a sequence attribute would be nice.

a sequence attribute would be niceMay be order by name will be enough.
Philip

SDDM EA 3.1: Engineer to Relational Model (subtypes)

It seems that mandatory attributes of entity subtypes become mandatory columns on the corresponding tables even when you engineer to "one table" (the super type).
In that case I would expect that the column would become optional.
Even better: a check constraint would verify that the column was not null for records belonging to the originating subtype.
/ Marc de Oliveira

1) Funny, when I dropped the column, and had it recreated by the engineering process, it worked correctly. But re-engineering the old column kept setting it to mandatory even when I manually set it to optional. I have a sense that some of my objects are ignored by the tool... Only when I drop them and recreate them they seem to be recognized. I had the same issue with some entities that consistenetly did not got engineered into my relational model. Only when I created a new ER-diagram and included the same entities they would get engineered into the relational model...
2) Designer did this. When a supertype table was generated in the "one table" mode there would always be a new column called SUBTYPE with a constraint allowing the short names of each subtype as values (ex: for the table PARTIES the SUBTYPE column could only contain the values PERS and ORG). With this in place you could create constraints such as "SUBTYPE = 'PERS' and FIRST_NAME is not null or SUBTYPE = 'ORG' and FIRST_NAME is null".
Regards,
Marc de Oliveira

SQL Datamodeler: engineer to relational model: column names

Hello
Suppose I have an attribute called DAY OF BIRTH. Is there a way I can forward engineer this attribute to a column name DAY_OF_BIRTH?
(I know you could have a coulmn name "day of birth" but thats not what I want)
Regards Erik

Hi Erik,
1) you should check separator settings in General options>Naming Standard -it should be "space" for Logical model and "underscore" for Relational model;
2) "Apply name translation" should be selected in engineering dialog
Best regards,
Philip

It's adding same column to relational model over and over

I'm using 3.3.0.747 now, and I'm pretty sure that it wasn't doing this when I was using 3.1.4. I go into the logical model and add an attribute (say "My notes") to an existing entity (say "Specifications"), adding a preferred abbreviation of "notes" to it. I engineer to the logical model, and it dutifully says that it's going to add a column to the corresponding table ("specs"). And it does.
Now, later, I add some attributes to some other table. When I do the engineering, it says that "notes" doesn't exist in "specs" and it's going to add it. If I let the engineering go through, I find that the relational model for "specs" now has two columns on it, "notes" and "notes1". If I do it again, I get "notes2". If I delete "notes1" and "notes2" and engineer again, "notes1" comes back.
I'm not finding that it does it this every time, but it's doing it for maybe half the columns I add to existing tables. (I haven't figured out a pattern.) If it happens to be a foreign key relationship, it creates an additional foreign key.
So far I've just added a step to my process to hunt down the new columns and delete them (or flag the logical model to not engineer the table at all if I know I haven't changed anything), but there must be something obvious I'm missing..
Edited by: dbomp on Apr 17, 2013 10:53 AM
I'll take back one thing I said: If the once-new column is related to a foreign key, it does keep re-creating the column, but it doesn't create new foreign key relationships to go with it.

I figured it out so I thought I'd put this here in case others bump into it.
What's going on is that the logical-to-relational engineering comparison tool doesn't recognize relational columns that the engineering tool didn't create. Sometimes I would add columns to both the logical and relational models manually rather than engineering. Say all I wanted to do was to add an active_flag to a table. I'd add it to the relational model and then add it to the logical model's corresponding entity with exactly the same values. Later, when I had a bunch of logical model changes to engineer, it wouldn't recognize the active_flag ("preferred abbreviation") in the entity as being the same thing as the active_flag in the relational table, and it'd create an active_flag2 column.
Lesson: only create new relational table columns by engineering them from the logical model.

Fwd engineering entity name changes to relational model?

I changed an entity name in my logical model after generating a relational model. But, I cannot figure out how to have SDDM (3.0.0.665) propagate this change to the relational model.
In the Engineer To Relational Model screen, if I select the changed entity from the tree view, in the Details pane there is a "Selected" check box next to the Name property (this line shows the discrepancy between logical name and relational name) but I can't check it. Under the Compare/Copy Options tab in that same dialog, the check box next to Name is checked. But, when I forward engineer the name is not changed.

BTW I just noticed that if I enable "Name Translation" then it seems to work. This is not at all intuitive or obvious, nor does it seem to be explained in the documentation.
According to the help:
Apply Name Translation: Controls whether formal names are translated to abbreviated names when the logical model is forward engineered to a relational model, and whether abbreviated names are translated to format names when a relational model is reverse engineered to the logical model. Name translation is applied only for valid names. In addition, translations between the words entity/attribute/key and table/column/index are performed.

DM3.0 EA2: Arcs do not engineer correctly to relational model

It seems to me that an arc between two mandatory relations in the logical model are engineered into two mandatory foreign keys in the relational model. This is clearly not the purpose of an arc.
What is even worse is that it seems that it is not possible to change the foreign keys into being optional. Both the foreign key columns and the foreign keys themselves are mandatory with a read only property that cannot be changed.
- Marc de Oliveira

I logged bug for that.
Philip

Engineering logical subtypes to Relational model

I have a logical model with subtype (box in box). When I forward engineer to a relational model there is nothing in the model to represent the subtypes. I was expecting (as in old Designer) to see a differentiator "type" column to be generated with a column check constraint based on my subtypes. In the logical model I selected "single table" as the implementation strategy.
Note I also have these entities tagged as Dimensions and did notice the engineering dialog showed the substypes as hierarchies too.
Am I missing a a preference setting somewhere?

Hi Kent,
I was expecting (as in old Designer) to see a differentiator "type" column to be generated support for that will be added later.
Note I also have these entities tagged as Dimensionswhat do you expect - if you change implementation from "single table" to something else you'll get tables also classified as dimension, no other magic will be involved
and did notice the engineering dialog showed the substypes as hierarchies tooI don't see your point here.
Philip

Engineering from relational model to logical model duplicated views

Hi all,
Several times, also in different projects, when I engineer from relational model to the logical model, the comparator tool show me de differences.
In these differences there are no view differences, but when I finish the process everything goes right except that in the logical model under views, I see all the views duplicated with [view_name]v1.
I've tried not selecting the view checkbox, in the compare dialog,but I get duplicated views also.
I don't know fir sue if this happens with table also.
What can I do to solve this problem? Is this a bug?
Edited by: morfeo8marc on 04-abr-2012 2:02

The problem is fixed in DM 3.1.1.703 and it's published. You can run following script and save design in order to get mappings for views recovered.
Philip
lmodel = model.getDesign().getLogicalDesign();
rmodels = model.getDesign().getRelationalDesigns().toArray();
for (var i = 0; i < rmodels.length; i++) {
rm = rmodels;
eviews = lmodel.getEntityViewSet().toArray();
for (var k = 0; k < eviews.length; k++) {
ev = eviews[k];
tv = rm.getTableViewSet().getObjectByID(ev.getGeneratorID());
if(tv == null){
tv = rm.getTableViewSet().getObjectGeneratedBy(ev);
if(tv != null){
rm.getRMExtendedMap().createMapping(ev,tv);

Cardinality in FK relations of relational model

Similar Messages

Maybe you are looking for