PowerDesigner doesn't respect dependencies between objects when generating script

Good afternoon,
I'm getting very frustrated trying to do something that should be 'by default' or easy to configure within PowerDesigner 16.5 working with a SQL Server 2008 R2 database.
I reverse-engineered a database into a physical model for modifications and I want to generate DDL from the physical model. I've worked around some issues including how to support SQL Server filtered indexes and a few others things.
However a really stupid think that I have not been able to figure out is how to have the script respect dependencies between objects in the model. When looking at the model at the Dependencies table the dependencies seem to be in place correctly. But when I choose the 'Generate Database...' menu item to generate a single script to create all of the objects within a database it does create them all but in alphabetical order which does not work because there are dependencies between objects (views, in particular) that requires specific views to be created in a different order depending on their dependencies.
Most tools understand the dependencies and create a script that will work.
How can I generate a database script from PowerDesigner that respects the dependencies between the objects (in this case between different databse views) so they get scripted out in the correct order.
Thanks much,
Don

I figured this out via the use of 'Traceability Links' within the View Properties.
Let me tell you... the only way to set these is to create a diagram and visually link things together because working through the grid on the property pages to select additional Linked Objects is just incredibly slow and painful.   You should be able to just free-text enter Linked Object names (with the model validating them) without having to traverse the model to individually select each object.   I had a lot of dependencies and it took several hours to incorporate them all.
I'm also disappointed that the reverse-engineering of a SQL Server database into a physical model doesn't use the sysdepends data to create 'Traceability Links' to associate dependencies between views, stored procedures, etc.   That would have saved me a lot of work.
I know that PowerDesigner is a 'generic' product that works with several databases... but being a person that only works with a particular database (in my case SQL Server) the lack of out of the box support (filtered index reverse engineering is an example that quickly comes to mind) is frustrating.

Similar Messages

ActiveX can't create object when VB Script called from Labview

I have an interesting issue that I can't find a solution for. I am using the DIAdem Run Script.VI in Labview to call a script that opens an Outlook object and sends an email. When the script is called via LabView I get this error:
However, when I manually run the script from the DIAdem script tab it works as expected with no errors.
This is the code:
'Begin email send function
Dim oOutlookApp
Dim oOutlookMail
Dim cnByValue : cnByValue = 1
Dim cnMailItem : cnMailItem = 0
' Get Outlook Application Object
Set oOutlookApp = CreateObject("Outlook.Application")
' Create Mail Item
Set oOutlookMail = oOutlookApp.CreateItem(cnMailItem)
' Set Mail Values
With oOutlookMail
.To = "[email protected]"
.Subject = "Report: " & Data.Root.ActiveChannelGroup.Name & " for " & CurrDate
.Body = "test automatic report emailing with VB Script."
' Add Attachement
Call .Attachments.Add(strLocFileName, cnByValue, 1 )
' Send Mail
Call .Send()
End With
(Original code includes Option Explicit and all variables are properly included/declared, I just took the snippet of what's causing the error).
I have looked at the following threads for info already:
http://forums.ni.com/t5/DIAdem/Some-errors-when-calling-LabVIEW-VIs-Interactively-from-DIAdem/td-p/2...
http://forums.ni.com/t5/DIAdem/Active-X-component-cannot-create-object-Diadem-8-1/m-p/71212/highligh...
-I tried running the script via Windows explorer (per Brad's suggestion) by itself without the DIAdem specific functions and it runs fine.
http://forums.ni.com/t5/DIAdem/Error-while-runing-diadem-asynchronous-script-from-labview-on/m-p/111...
-I am not running the scripts asynchronously
Using Windows 7 (64bit), DIAdem 11.2 and LabView 7.1.1
Thank you.

Hey techerdone -
I'm afraid I personally can't be of much help - I tested your code both from DIAdem and from LabVIEW and each worked without issues in both cases (Outlook closed, Outlook open). I'm using DIAdem 2011 SP1, LabVIEW 2011, and Outlook 2007...
Derrick S.
Product Manager
NI DIAdem
National Instruments

Dependencies between Objects

Hi Folks,
My client has asked me to move from one db to another. (I don't know why, may be they are restructuring). Now the problem is that there are many stored procedures, functions and tables which I have created and also using somewhere. Now for moving these objects
I have to do manual checks on each of my procedure to find out dependencies.
First I used SP_DEPENDS but that didn't show correct results. Then I did use
SYS.DM_SQL_REFERENCING_ENTITIES, but a weird thing I noticed. Some of my procedures are not listed in this table.
Basically I need the output in below format:
NAME
TYPE
DEPENDED UPON
TYPE
COLUMN NAME
RealTimeLeads
Stored Proc
INSTrigger
Trigger
OfferDateTime
RealTimeLeads
Stored Proc
INSTrigger
Trigger
EnquiryDateTime
RealTimeLeads
Stored Proc
cc_lead__c_upd
User Table
ID__c
RealTimeLeads
Stored Proc
cc_lead__c_upd
User Table
CustomerName__c
RealTimeLeads
Stored Proc
cc_lead__c_upd
User Table
Product__c
SP_CampaignDashboard
Stored Proc
CampaignDashboard
User Table
CD_ID
SP_CampaignDashboard
Stored Proc
EmailCampaign_Attempts
User Table
AttemptCount
SP_CampaignDashboard
Stored Proc
OfferProduct_Master
User Table
Product__c
SP_CampaignDashboard
Stored Proc
Portal_Campaign_Attempts
User Table
AttemptCount
SP_CampaignDashboard
Stored Proc
Responses
User Table
ResponseDate
SP_CampaignDashboard
Stored Proc
SMS_Campaign_Attempts
User Table
AttemptCount
Chaos isn’t a pit. Chaos is a ladder. Many who try to climb it fail and never get to try again. The fall breaks them. And some are given a chance to climb, but they refuse. They cling to the realm, or the gods, or love. Illusions. Only the ladder is real.
The climb is all there is.

I did some tests. If you do not have sys.dm_sql_referencing_entities, you probably run SQL Server 2005 or lower. Alas, SQL_Dependencies seems to have the same shortcommings as the old fashioned dbo.sysdepends view. When you create an object referencing an
object that doesn't exist yet, this reference is not captured. Changing the order of creation can help you out. However, circular references will never be captured. Also, dropping and re-creating an object destroys the information in this view. Proof:
The below creation order causes the link between TestProc and TestView to be missed.
CREATE PROC TestProc AS SELECT * FROM TestView
GO
CREATE TABLE TestTab (i int not null)
GO
CREATE VIEW TestView AS SELECT * FROM TestTab
GO
-- Transact-SQL Statement to list all objects and their dependencies (SQL Server 2005).
SELECT SCH.name + '.' + OBJ.name AS ObjectName
,OBJ.type_desc AS ObjectType
,REFSCH.name + '.' + REFOBJ.name AS ReferencedObjectName
,REFOBJ.type_desc AS ReferencedObjectType
,REFCOL.name AS ReferencedColumnName
,DEP.class_desc AS ReferenceClass
,DEP.is_selected AS IsSelected
,DEP.is_select_all AS IsSelectAll
,DEP.is_updated AS IsUpdated
FROM sys.sql_dependencies AS DEP
INNER JOIN
sys.objects AS OBJ
ON DEP.object_id = OBJ.object_id
INNER JOIN
sys.schemas AS SCH
ON OBJ.schema_id = SCH.schema_id
INNER JOIN sys.objects AS REFOBJ
ON DEP.referenced_major_id = REFOBJ.object_id
INNER JOIN sys.schemas AS REFSCH
ON REFOBJ.schema_id = REFSCH.schema_id
LEFT JOIN sys.columns AS REFCOL
ON DEP.class IN (0, 1)
AND DEP.referenced_minor_id = REFCOL.column_id
AND DEP.referenced_major_id = REFCOL.object_id
ORDER BY ObjectName
,ReferencedObjectName
,REFCOL.column_id
GO
DROP TABLE TestTab
GO
DROP VIEW TestView
GO
DROP PROC TestProc
GO
Changing the order helps:
CREATE TABLE TestTab (i int not null)
GO
CREATE VIEW TestView AS SELECT * FROM TestTab
GO
CREATE PROC TestProc AS SELECT * FROM TestView
GO
-- Transact-SQL Statement to list all objects and their dependencies (SQL Server 2005).
SELECT SCH.name + '.' + OBJ.name AS ObjectName
,OBJ.type_desc AS ObjectType
,REFSCH.name + '.' + REFOBJ.name AS ReferencedObjectName
,REFOBJ.type_desc AS ReferencedObjectType
,REFCOL.name AS ReferencedColumnName
,DEP.class_desc AS ReferenceClass
,DEP.is_selected AS IsSelected
,DEP.is_select_all AS IsSelectAll
,DEP.is_updated AS IsUpdated
FROM sys.sql_dependencies AS DEP
INNER JOIN
sys.objects AS OBJ
ON DEP.object_id = OBJ.object_id
INNER JOIN
sys.schemas AS SCH
ON OBJ.schema_id = SCH.schema_id
INNER JOIN sys.objects AS REFOBJ
ON DEP.referenced_major_id = REFOBJ.object_id
INNER JOIN sys.schemas AS REFSCH
ON REFOBJ.schema_id = REFSCH.schema_id
LEFT JOIN sys.columns AS REFCOL
ON DEP.class IN (0, 1)
AND DEP.referenced_minor_id = REFCOL.column_id
AND DEP.referenced_major_id = REFCOL.object_id
ORDER BY ObjectName
,ReferencedObjectName
,REFCOL.column_id
GO
DROP TABLE TestTab
GO
DROP VIEW TestView
GO
DROP PROC TestProc
GO

Mapping dependencies between objects

I'm querying a database to obtain information from 10 tables. The goal is to display the information on a JSF based portlet using a tree-view, where an user can use the tree interface to drill to desired level of detailed information. The interface may need to show 3/4 different tree-views of the same data set.
To begin with, I'm doing a join across those tables and load a hash (representing the first hierarchy in the tree) with multiple levels to represent the tree-view or the dependency between various parameters. If there are 3 views required, I'm thinking about building 3 such hashes - where the elements in the hash will probabaly be derived from a TreeNode class to build the gui. Since all the views refer to the same underlying data model, each node of the Hash will actually refer to some object that stores the model data.
I have two questions:
* Currently I'm building the hash by parsing the ResultSet obtained from query; Is there a better way to do this? I thought about using O/R mappers; but I couldn't think of an elegant solution to load the collections from a query to represent the dependencies dictated by the view.
* Is there a better way to design the query?
I would really appreciate any comments.

Thanks for the suggestions. I have a generic class for building a tree-view - and that uses recursion. However, that doesn't need to know about the parent-child relationship. I was thinking of parsing the db query result and store the parent child relationship in a nested hash. My generic Tree class implementation would know to take this nested hash and build the tree. Essentially an instantiation of the generic class or it's extended version would act as a backing bean for a JSF component.
Anyway, I'll try to focus more on a Hibernate based approach as you guys hinted. I'll wait for Dave's feedback. Actually, I have his book on the portlet API.

Import Oracle Designer model dependencies between objects

Hello,
I am using the Oracle Designer model import to fill Data Modeler. There are some views which have dependencies on each other, resulting that views which depend on other views are incomplete.
Is there a way to rerun the import again in the same relational model, since the import utility now opens a new relational model again.
Is there a kind of logging where you can find these kind of errors?
Best regards,
Joop

Thanks Phillip, but I was actually referring to the version level of the Designer repository itself, not the database instance in which it resides. I seem to remember that you could not upgrade your Designer repository without also upgrading the Oracle Designer client as well; hence my question about compatibility between SQL Developer Data Modeler and various Repository versions.

Schema/Owner prefix when generating scripts

I'm trying to create a foreign key between two schema's in designer. I have already added an external reference. However when I look at the code generated then no schema prefix is added and therefore the foreign key is not created.
Does anyone know how I can make sure the code is generated with the prefix?

Generating from the DB Admin tab does not prefix the objects I am generating with the schema name. There isn't a generator preference (that I've found at least) to accomplish this. Has anyone been able to generate database objects prefixed with the schema name?
Thanks,
Wayne

Why doesn't the rotate gesture on my trackpad behave as it used to for switching between tabs when it works the same with all other programs besides Firefox?

Why doesn't the rotate gesture on my trackpad behave as it used to for switching between tabs when it works the same with all other programs besides Firefox?
I changed the about:config to make the rotate gesture move between tabs, and it worked great for well over a year. Now it behaves erratically as described here:
https://bugzilla.mozilla.org/show_bug.cgi?id=877598
It used to be very controllable to move one tab over, now it is erratic and moves quickly and to unexpected tabs regardless of how slow I do the rotate gesture. This is the main reason I use Firefox over Chrome and I would like to continue to do so unless there is no fix to this.

Just set the browser.gesture.twist.threshold to something around 15-25.

Differences Between Object And System Privileges

Hi,
Whats the difference between object and system privileges in oracle?
Cheers
Paul

System Privileges
A system privilege is the right to perform a particular action, or to perform an action on any schema objects of a particular type. For example, the privileges to create tablespaces and to delete the rows of any table in a database are system privileges.
Schema Object Privileges
A schema object privilege is a privilege or right to perform a particular action on a specific schema object:
For example, the privilege to delete rows from the departments table is an object privilege.
Some schema objects, such as clusters, indexes, triggers, and database links, do not have associated object privileges. Their use is controlled with system privileges. For example, to alter a cluster, a user must own the cluster or have the ALTER ANY CLUSTER system privilege.
A schema object and its synonym are equivalent with respect to privileges. That is, the object privileges granted for a table, view, sequence, procedure, function, or package apply whether referencing the base object by name or using a synonym.
Granting object privileges on a table, view, sequence, procedure, function, or package to a synonym for the object has the same effect as if no synonym were used. When a synonym is dropped, all grants for the underlying schema object remain in effect, even if the privileges were granted by specifying the dropped synonym.

Account Requires a CO Object when DN PGI

Dear all,
I got an error message "Account 510102 requires an assignment to a CO Object" when DN Post Goods Issue.
The material has a profit center assigned to it, and it was bought forward to SO, and also to the DN line.
I know that this problem can be solved if I set a cost center at the "Default acct. assign." tab page of the account's cost element. However, the problem is the cost center structure is material dependent, so a default cost center is not a proper solution.
Could anyone please help me? Thanks in advance.
Best regards,
Chris

Hi,
OKB9 is used purely for integration purpose between FI and CO. You must maintain the cost element for the purpose of flow from FI to CO. Cost ELement plays an important role for integration between FI and CO. So the transaction releted to FI and CO will take place via OKB9 in which you assign the cost element. Account assignment to CO object plays an important role in CO. You should assign that cost element to Cost center or Profitability Segment or Business Area according to scenario of your Business.
If you activated CO-PA, you can assign relevant profit segment instead of dummy cost center for the cost element in OKB9. As you know, all costs should finally be charged to customers or services, so any cost can be recognized to respective profit segment, where you make profit analysis based on profit segments.
As for substitution, in this case we only use it to make CO object assignment similar to the function of OKB9 but it has higher priority than OKB9. The limitation could be only dummy cost centers can be substituted
Thanks & Regards,
Kiran

Gap between objects

I know there are several threads about this but I haven't found a decent answer. I have a logo I'm working on with overlapping shapes, strokes, etc. Everything looks fine in Illustrator but when I export the file to show the client, whether it be a .png, .pdf, .jpg, there is a hairline gap between objects. Is this only a screen issue or will the hairline gap print?

That's right, it's a screen issue. So printing the vector file will be fine.
You have a problem is when you want to show a pixel version to a client.
Place the AI file as a smart object in Photoshop
See my answer here for a solution:
Illustrator divide problem
You can also skip the Photoshop route by using the Rasterise effect in Illustrator and export as .png, jpegs etc.

Capture now doesn´t respect tc

Why when I capture through of capture now with firewire doesn´t respect the tc of my camera?
Thanks a lot.

Well, if you have device control set to UNCONTROLLED DEVICE, that ignores timecode. Device Control must be set to Firewire NTSC (or Firewire PAL for Europe), or Firewire BAsic.
Shane

Relationship between Objects Not Found in PA

Dear all,
I have uploaded objects (Org Unit, Job, and Position) and the relationships between objects separately using LSMW.
When i checked the update of the relationship between objects in PA30, but the relationship between objects (Org Unit, Job, and Position) did not exist, yet exists in PP02.
I tried to upload the relationships between objects using holder (A008) in LSMW again, still I could not find the relationships in PA30, but exist in PP02.
Then I tried to run rhinte00 and rhinte30 to link the relationship between objects. I managed to get through rhinte00, but failed to run rhinte30.
Below is the report log when I tried to run rhinte30.
Report log:
Processing mode:   Processing successful
Number of personnel nos. selected :   0
Personnel numbers with errors:

Check the following.
1. Check if integration is active in T77S0 table PLOGI PLOGI = Your active plan version & PLOGI ORGA = X
2. Check if all your objects are created with this active plan version
3. Check the feature PLOGI to see if your PA PSA etc for the employee are integrated.
cheers
Ajay

Creating custom mappings between objects

I currently have a db schmea that supports "soft" deletes. Example:
Client table has fields:
id - primary key (number)
name - varchar
active ("Y" or "N" where "Y" means active and "N" means "INACTIVE")
We also have a table Users:
id - primary key (number)
name - varchar
active ("Y" or "N" where "Y" means active and "N" means "INACTIVE")
When I delete a user I want to set its active field to "N" and then when
someone reads users from the client (after I have saved my changes) I want
only users whoses active field is "Y" to be returns, i.e.
Client clientA = <read client A>
Collection users = clientA.getUsers();
Is there a way to configure/customize the mapping/relationship between
objects?
Thanks!
Bruce

All of the emails are really my way of getting a soft delete to work. Does
KODO support this feature or is their a standard work around to support
this feature?
Thanks!
Bruce
Alex Roytman wrote:
You could break you collection into multiple mutually exclusive
subcollections (like active/inactive) and have your object model to map them
separately (via views with mutually exclusive criterion) When changing a
collection member's status you will need to move it from active collection
to inactive. Complete collection can be dode as dybamic (delegating)
composition of the subcollections (but it will cost twice as much to fetch
it)
Try to resolve your issues on object model level. I believe altering sql
mappings to make them very fancy will cause you lots of grief in the long
run
"Bruce" <[email protected]> wrote in message
news:[email protected]...
I also want to do something similiar to support object history. I could
have a Client object which gets modified and instead of editing this
object I would copy it, inactivate the copy and then edit the original.
This allows for full history. I can always select by active (or a more
complicated status) to get the rest of the history instead of the "top".
Views here seems like a hack. I have activew fields in all of my tables so
I would nee views for each table which is a lot to manage if fields are
changing (need to change them in both places). Also this creates issues
since views are really read-only.
Isn't there a way to change the SQL used to read a relationship?
Bruce
Alex Roytman wrote:
map your classes against views with (where active = 'Y') to make sure
"soft
deleted" records do not get read and have your object model to handlesoft
deletes (like removing from collections etc) use helper methods to do"soft
deletes" masking real deletes is not good idea - you will need them atsome
point
"Stephen Kim" <[email protected]> wrote in message
news:[email protected]...
While there is nothing default that does particularly what you desire,
but from my perspective, you should not have Kodo do anything by
default
as you will probably want to delete the inactive instances at somelater
point.
Instead you should look into Query.setCandidates:
http://www.solarmetric.com/Software/Documentation/2.4.3/docs/jdo-javadoc/jav
ax/jdo/Query.html
i.e. public Collection getActiveUsers ()
PersistenceManager pm = JDOHelper.getPersistenceManager
(this);
Query q = pm.newQuery (User.class, true);
q.setFilter ("active == "N"");
q.setCandidates (users);
return q.execute ();
I haven't tested the above code and there are a lot of optimizations
you
could do such as caching the query in a transient collection but Ithink
you get the idea.
You may want to post your thoughts on our newsgroups
(news://news.solarmetric.com) and see how other people are tackling
similar probles..
Bruce wrote:
I currently have a db schmea that supports "soft" deletes. Example:
Client table has fields:
id - primary key (number)
name - varchar
active ("Y" or "N" where "Y" means active and "N" means "INACTIVE")
We also have a table Users:
id - primary key (number)
name - varchar
active ("Y" or "N" where "Y" means active and "N" means "INACTIVE")
When I delete a user I want to set its active field to "N" and then
when
someone reads users from the client (after I have saved my changes)I
want
only users whoses active field is "Y" to be returns, i.e.
Client clientA = <read client A>
Collection users = clientA.getUsers();
Is there a way to configure/customize the mapping/relationship
between
objects?
Thanks!
Bruce
Stephen Kim
[email protected]
SolarMetric, Inc.
http://www.solarmetric.com

Diff between object based and ob. oriented

diff between object based and ob. oriented

Object-oriented requires at least three factors:
1. Encapsulation: Combining data and the methods that manipulate the data into a self-contained unit called an "object" in which the data is hidden from the user and only the method interface is visible.
2. Inheritance: Ability of a class to extend the features of another class by obtaining all the public features of the other class while adding its own specialized features (Example: Great Dane inherits from Dog)
3. Polymorphism: The ability to perform "late binding" by invoke methods of a subclass on behalf of its superclass. (Polymorphism includes other things and it difficult to explain.
If a language does not include all three, it is called "object-based." Visual Basic 6 is object-based, because it allows you to define classes that encapsulate, but doesn't support an inheritance hierarchy nor polymorphism.

Maximizing communication between objects.

Hi,
I have a question concerning communication between objects. Particularly I want to learn to maximize communication between objects. Obviously there are several ways of increasing communication between objects, unfortunately I only really understand one methodology, which is as follows:
For example:
If I were to create an object called GUI of class UserInterFace, like so
class UserInterface extends JFrame
private variable1;
private variable2;
UserInterface()
    setBounds(x,y,width,height);
    setVisible(true);
}I could then create an object called fileOps of class FileOperations and pass the entire GUI object into the constructor of the fileOps class, as follows:
class FileOperations
private FileReader input;
private FileWriter output;
private UserInterface GUI;
FileOperations(UserInterface GUI) //GUI object introduced.
    this.GUI = GUI;
}By following the code fragments above it becomes obvious that I can establish communications between the fileOps object and the GUI object. With code similar to the above; fileOps will be able to access all of the accessor and mutator methods within object GUI. Using the above methodology always requires making all variables private and accompanying those variables with public methods (thus preserving encapsulation).
If anybody can suggest any other methodologies for increasing communication between objects they will be greatly appreicated. Of interest to me is how one can go about ensuring complete two way communication between objects, my example above is rather one sided (fileOps can see GUI, but not vice versa). Obviously the instantiation of objects must occur in some procedural sequence, I want to know how to capitilize on this. Suggestions to read a particular book will be appreciated, however suggestions born from peoples experiences are most desired. Thanks.
Regards
David

Rommie and Rajesh,
Thank you both kindly for sharing your time and valuable knowledge. As I have only been programming with Java for a year part time (as one section of a Computer Systems Diploma) I consider myself to be a beginner in programming terms. I want to do really well with Java as quickly as possible and you have both helped me tremendously.
Rommie the methodology that you have passed onto me is great, thankyou so much, I feel like a new section of Java has completely opened to me. I will be using the third "reference" class methodology from everyday now on.
Rajesh your words are encouraging, and I will be reading up on Hierarchy and Aggregation immediately. When you said "Before using either, decide on the existence of the object, its intended functionalities e.t.c. ...", could you please explain with an example as to where Hierarchy is desirable and where Aggregation is desirable.
Please feel free to e-mail me with further suggestions if you come up with something else that aids in the communication between objects.
E-mail: [email protected]
Thanks again.
Regards
David

PowerDesigner doesn't respect dependencies between objects when generating script

Similar Messages

Maybe you are looking for