Difference between revisions of "Modeler:Tutorial/Deriving a Database Schema"

From CaseTalk Wiki
Jump to: navigation, search
 
(19 intermediate revisions by the same user not shown)
Line 3: Line 3:
= Checking the integrity of an information grammar =
= Checking the integrity of an information grammar =


'''Step 19'''
== Preparation ==


1. Open the Repository menu and select Validate:
1. Open the ''Run'' menu and select ''Settings''.
 
[[File:Integritycheck.GIF]]
2. Check ''Copy Active Diagrams'' and ''Step by Step'' and make sure the other settings are as shown below:
 
[[File:PrepareGLRSettings.png]]
 
3. Click OK.
 
Note: Some CaseTalk versions didn't pick up the settings until restarted and the project is reopened.
 
== Verify well-formedness ==
 
1. Click menu Run, Verify well-formednes..
Casetalk automatically performs a model Integrity Check. If the casetool encounters any errors the Integrity Check window will be opened so you can examine the error. If Integrity Check encounters only warnings, the Integrity Check window will not be opened.
 
2. After the Integrity Check, the result is visible below the diagram. You can change the window-size for a better view.
 
[[File:Integritycheck.png]]
   
   
A detailed description of the checks can be found in appendix B.
A detailed description of the checks can be found in appendix B.
2. Click Check.
3. Close the Integrity Check window.


= Deriving a grouping proposal =
= Step-by-step grouping =
 
Step by step grouping is hardly used in daily practice because of the large number of roles that are present in most models. The following steps are only for the purpose of education and understanding the grouping process using a small model.
 
1. Open the Run menu and select Transform
 
2. Casetalk automatically transforms the model. After a while, this window appears:
 
[[File:ConfirmGlrSettings.png]]
 
3. Click yes.
 
4. CaseTalk now shows this window:
 
[[File:StepByStepStudent.png]]
 
Since there are two separate UC’s on role 14 and 15, CaseTalk has to decide how to group these roles, and it proposes to group role 14 into the fact type ''Student''.
 
5. Click Yes.
 
Casetalk shows the next proposal. You can click Yes for each separate proposals, or click Yes to all to approve all following proposals.
 
6. Casetalk now shows a window to save your transformed model:
 
[[File:SaveIGG.png]]
 
7. Save the transformed diagram as '''Apprenticeship.igd'''.
 
The project manager on the right now shows the grouped grammar and diagram:
 
[[File:ProjectWithIGG.png]]


'''Step 19'''
For the purpose of this tutorial, we’re going to rerun the transformation with other settings, so delete the two files that you’ve just created:


1. Open the Repository menu and select GLR:
8. Rightclick on '''lesson_glr.igg''' and choose ''Remove from project''. Confirm with Yes and notice that the grouped diagram is also deleted.  
[[File:Groupproposal.GIF]]
Make sure all options are set as displayed above.
2. Click 'Start processing'.
First the casetool will perform an Integrity Check. If the casetool encounters any errors the Integrity Check window will be opened so you can examine the error. If Integrity Check encounters only warnings, the Integrity Check window will not be opened.
[[File:Glrwarning.GIF]]
3. Click Ok.
The casetool will now commit all changes to the repository. Copies are made of both the IG and the openend IGD's. The copies are automatically added to the Project Manager. The casetool will open the copies and continue the GLR algoritm on these copies.
Roles which can be grouped will be marked with a 'G' in the diagram:
[[File:Igdgroupproposal.GIF]]
Note: A datatype proposal has now been performed. Display the label types and associated datatypes by selecting Show Data Type Info from the IGD menu if necessary.
4. Double click 'lesson.ig' in the Project Manager. Do not save 'lesson.igg' and 'IGD1>'.


= Step-by-step grouping =
Note: if the repository-window on the left has suddenly disappeared, double click on the green IG-icon on the projectmanager:


'''Step 19'''
[[File:ReselectIg.png]]


1. Open the diagram 'lesson.igd' from the Project Manager.
9. Save the project by selecting Save All from the File menu.
2. Open the Tasks menu and select GLR:
[[File:Glrgroupstepbystep.GIF]]
Make sure all options are set as displayed above.
3. Click Start Processing: (Overwrite 'lesson.igg')
[[File:Glrgroupstepbystep2.GIF]]
4. Click Yes to group role 7 into 'Apprenticeship'.
5. Repeat step 4 for every role which can be grouped.
[[File:Glrgrouped.GIF]]
6. Double click 'lesson.ig' in the Project Manager. Do not save 'lesson.igg' and 'IGD1>'.


= Setting grouping marks before grouping =
= Setting grouping marks before grouping =


'''Step 19'''
1. Open the diagram.


1. Open the diagram.
2. Right click on the fact type Apprenticeship Assignment.
 
3. Choose Task, Edit from the pop up menu
 
4. Click on the Role button in the window that appears.
 
5. Select role 14 and set the grouping to ''Don’t group'':
   
   
2. Click on 'Assigned Apprenticeship' in the Repository window.
[[File:Glrgroupoverride.png]]
 
3. Select Edit Roles from it's popup-menu:
6. Repeat this for role 15.
[[File:Glrgroupoverride.GIF]]
4. Select role 14 and set the grouping to 'Do not group'.
5. Repeat this for role 15.
   
   
By unchecking roles 14 and 15 you have indicated you do not want to group these roles.
By unchecking roles 14 and 15 you have indicated you do not want to group these roles.
   
   
6. Open the Tasks menu and select GLR:
7. Open the ''Run'' menu and select ''Settings''. Make sure all options are set as displayed on the right.
 
[[File:GlrNonStepped.png]]
 
Note: Do not check Step-by-step grouping. The Step-by-step grouping algorithm ignores the User Override property.
   
   
[[File:Glrgroupsubtypes.GIF]]
9. Open the ''Run'' menu and select ''Transform''.
 
10. If this window appears, select ''YES'':
 
[[File:ConfirmGlrSettings.png]]
   
   
Make sure all options are set as displayed above.
11. Save the new diagram as '''Apprenticeship.igd'''.
 
The generated information grammar is not in Optimal Normal Form (ONF). This structure might be choosen if ''Assigned Apprenticeship'' is unknown for most of the students. (Grouping ''Assigned Apprenticeship'' then results in a ''Student'' table with too many NULL values in the ''Assigned Apprenticeship'' column).
   
   
''Note: Do not check Step-by-step grouping. The Step-by-step grouping algorithm ignores the User Override property.''
[[File:Glrgrouped2.png]]
 
12. Rightclick on the '''lesson_glr.igg''' in the project manager and choose ''Remove from project''. Confirm with Yes and save the project by choosing Save All from the File menu.
 
= Deriving a proposal =
   
   
7. Click Start processing.
1. Open the ''Run'' menu and check ''Proposal''.
 
The elements in the diagram will show with ''G'', ''L'' and ''R'' indicators.
   
   
The generated information grammar is not in Optimal Normal Form (ONF). This structure might be choosen if 'Assigned Apprenticeship' is unknown for most of the students. (Grouping 'Assigned Apprenticeship' then results in a Student table with too many NULL values in the Assigned Apprenticeship column).
[[File:Glrlexicalizeproposal.png]]
   
   
[[File:Glrgrouped2.GIF]]
3. Uncheck the menu item ''Proposal'' to hide the indicators again.


= Deriving a lexicalizing proposal =
= Choosing a Primary Key =
 
1. Select the GLR window:
When ‘Automatic primary key selection’ is active and no primary UC is marked, Casetalk chooses a primary key based on the role with the lowest number. For instance, the fact type ‘Apprenticeship Assignment’ has two separate UC’s on role 14 and 15. So Casetalk chooses role 14. When this option is disabled, the user kan choose the primary key.
 
And select only the lexicalize and proposal option.
1. Reactivate the ''IG Model'' and select the ''Run >> Settings'' menu.
2. Click Start...
All roles which should be lexicalized will be marked with a 'L' in the diagram:
[[File:Glrlexicalizeproposal.GIF]]
3. Close the project by selecting Close Project from the File menu. Do not save the project.


= Lexicalizing =
2. Disable the option ''Automatic Primary Key Selection'', and click OK.
To show all options of the lexicalizing and reducing algorithm we will use a slightly different information grammar in the following exercises.
A fact type expression has been added to 'City'. Without this fact type expression City would have been removed during the lexicalizing process. Now it will be reduced during the reducing process.
A new fact type (Native City) has been added to 'Student'. By adding this fact type we have created a column naming conflict in the 'Student' table.
[[File:Exam_glr.GIF]]
1. Open the File menu and select Open Project. Open the 'EXAM_GLR' project from the '...\LESSONS\EXAM_GLR' directory.
2. Open the Tasks menu and select GLR:
[[File:Glrgrouplex.GIF]]
Make sure all options are set as displayed above.
3. Click Start...
[[File:Glrgrouplexfix.GIF]]
Roles 4 and 18 are both played by 'city name'. To prevent a column naming conflict this Column Fix Editor is shown.
By clicking the Generate button the casetool will automatically add numbers behind the column names.
4. Select role 4 and enter the prefix 'Residence'. (Don't forget to set the Column Fix Type.)
5. Select role 18 and enter the prefix 'Native', then click Ok.
[[File:Glrpkedit.GIF]]
In the appearing Primary Key Editor you can select which unicity constraint should be taken as primary keys. (If there are fact types with multiple unicity constraints).
In the left section a list of fact types with multiple unicity constraints is displayed. Below this list you will find a the list with the unicity constaints of the selected fact type. Each of the unicity constraints in this list can be checked. In the right section the selected fact type is displayed.
6. Choose UC14 for 'Assigned Apprenticeship'.
7. Select 'Apprenticeship Preference'.
8. Choose UC12 for 'Apprenticeship Preference'.


9. After you've completed this lesson, please select the menu 'File\Close All' and than select 'File\Reopen\AppReg\Lesson.prj'.
3. Open the menu ''Run'' and select ''Transform''.


= Deriving a reduction proposal =
4. In the appearing ''Primary Key Editor'' you can select which uniqueness constraint should be taken as primary key.


[[File:Glrpkedit.png]]
   
   
1.Select the GLR window:
5. In the left section a list of fact types with multiple uniqueness constraints is displayed. Below this list you will find a the list with the uniqueness constraints of the selected fact type. Each of the uniqueness constraints in this list can be checked. In the right section the selected fact type is displayed .
 
[[File:Glrreduceproposal.GIF]]
6. Choose UC 13 for ''Apprenticeship Preference'', so that the primary key is created using roles 11.1, 11.2 and 12.
Make sure the options are set as displayed above.
2.Click Start...
The object type 'City' will be marked for reduction with an (R) behind the object type name.


= Reducing =
= Reducing =
   
   
1. Select the GLR window:
1. Select Settings from the Run menu.
 
2. Select only the option ''Step by Step'' as seen on the below:
 
[[File:Glrreducestepbystep.png]]
 
3. Click OK.
 
4. Open the ''Run'' menu en select ''Transform''.
 
5. The ''step-by-step'' window appears:
 
[[File:Glrreducestepbystep2.png]]
   
   
[[File:Glrreducestepbystep.GIF]]
6. Click Yes to all to reduce everything.  
 
Make sure the options are set as displayed above.
7. Save the new diagram as '''lesson_r.igd'''.
 
2. Click Start...
8. Rightclick on '''lesson_r.igg''' at the Project Manager on the right, remove it from the project. Confrm with Yes.
 
[[File:Glrreducestepbystep2.GIF]]
9. Open the File menu and select Save All to save the project.
3. Click Yes to reduce everything.
4. Open the File menu and select Close Project. Do not save the project.


= Generating Table Documentation =
= Generating Table Documentation =


'''Step 19'''
1. Open the Run menu and select Settings.


1. Open the Options menu and select Project, than activate the GLR tab:
2. Derive data types:
If this option is selected the casetool will derive a datatype proposal during the GLR process. CaseTalk will generate Char and Numeric datatypes. If you have entered other datatypes they will not be affected by this option. If you have used Char and Numeric yourself and the casetool has detected labels exceeding the total and/or decimal length you have specified these attributes will be updated.
If the option ''Derive data types'' is not checked, check it.
 
3. Clear Population:
If this option is selected, all population will be removed during the GLR process.
Leave this option unchecked for now.
 
4. Make sure the settings in this window are like this:
   
   
[[File:Projectoptionsglr.GIF]]
[[File:Projectoptionsglr.png]]
   
   
2. Check the options Automatic Column Fix Generation and Automatic Primary Key Generation, then click Ok. Read the information box below for more information about the GLR options.
5. Click OK.
 
6. Select Transform from the Run menu.
The information grammar will now be grouped, lexicalized and reduced.
 
7. Save the new diagram as ‘Apprenticeship.igd’
 
8. Select ‘Apprenticeship Preference from the repository.
 
9. Open the View menu and select Table Report.
 
10. A new tab will appear containing the table documentation for 'Apprenticeship Preference':
   
   
GLR Options:
[[File:Tabledocumentation.png]]
   
   
Derive data types:
11. Keep the Table Documentation tab open and select another fact type from the Repository window. The contents of the Table Documentation tab might not change automatically. If so, adjust the tab window a bit.
If this options is selected the casetool will derive a datatype proposal during the GLR process. The casetool will generate Char and Numeric datatypes. If you have entered other datatypes they will not be affected by this option. If you have used Char and Numeric yourself and the casetool has detected labels exceeding the total and/or decimal length you have specified these attributes will be updated.
 
12. Close the Table Documentation tab.  
Clear Population:
 
If this option is selected, all population will be removed during the GLR process.
13. Open the File menu and select Save All.  
 
Automatic Column Fix Generation:
You can print the table documentation by selection Print Table Documentation from the file menu.
If this option is selected and column naming conflicts are discovered, conflicting columns will be made unique by adding a number as postfix.
 
----
Automatic Primary Key Generation:
 
If this option is selected and fact types with multiple primary keys are discovered, this algorithm will choose the first unicity constraint not covering optional roles. If all unicity constraints cover optional roles (only for generalisations) then all unicity constraints will be made primary.
Back: [[:Modeler:Tutorial/Constraints|Constraints]]
 
3. Open the Tasks menu and select GLR:
Next: [[:Modeler:Tutorial/Using the plugin modules|Using the plugin modules]]
 
[[File:Glrall.GIF]]
[[Category:Modeler]]
Make sure all options are set as displayed above.
4. Click Start...
The information grammar will now be grouped, lexicalized and reduced.
5. Click on 'Apprenticeship Preference' in the Repository.
6. Open the View menu and select Table Documentation:
A window will appear containing the table documentation for 'Apprenticeship Preference':
[[File:Tabledocumentation.GIF]]
7. Keep the Table Documentation window open and select another fact type from the Repository window.
Note: The contents of the Table Documentation window will change automatically.
8. Close the Table Documentation window.
9. Open the File menu and select Save Project. If necessary save the diagram as 'lesson2.igd'.
''You can print the table documentation by selection Print or Print All from it's the popup-menu.''

Latest revision as of 05:14, 24 March 2021

In this lesson you will learn how to check the integrity of an information grammar and how to derive a database schema.

Checking the integrity of an information grammar

Preparation

1. Open the Run menu and select Settings.

2. Check Copy Active Diagrams and Step by Step and make sure the other settings are as shown below:

PrepareGLRSettings.png

3. Click OK.

Note: Some CaseTalk versions didn't pick up the settings until restarted and the project is reopened.

Verify well-formedness

1. Click menu Run, Verify well-formednes.. Casetalk automatically performs a model Integrity Check. If the casetool encounters any errors the Integrity Check window will be opened so you can examine the error. If Integrity Check encounters only warnings, the Integrity Check window will not be opened.

2. After the Integrity Check, the result is visible below the diagram. You can change the window-size for a better view.

Integritycheck.png

A detailed description of the checks can be found in appendix B.

Step-by-step grouping

Step by step grouping is hardly used in daily practice because of the large number of roles that are present in most models. The following steps are only for the purpose of education and understanding the grouping process using a small model.

1. Open the Run menu and select Transform

2. Casetalk automatically transforms the model. After a while, this window appears:

ConfirmGlrSettings.png

3. Click yes.

4. CaseTalk now shows this window:

StepByStepStudent.png

Since there are two separate UC’s on role 14 and 15, CaseTalk has to decide how to group these roles, and it proposes to group role 14 into the fact type Student.

5. Click Yes.

Casetalk shows the next proposal. You can click Yes for each separate proposals, or click Yes to all to approve all following proposals.   6. Casetalk now shows a window to save your transformed model:

SaveIGG.png

7. Save the transformed diagram as Apprenticeship.igd.

The project manager on the right now shows the grouped grammar and diagram:

ProjectWithIGG.png

For the purpose of this tutorial, we’re going to rerun the transformation with other settings, so delete the two files that you’ve just created:

8. Rightclick on lesson_glr.igg and choose Remove from project. Confirm with Yes and notice that the grouped diagram is also deleted.

Note: if the repository-window on the left has suddenly disappeared, double click on the green IG-icon on the projectmanager:

ReselectIg.png

9. Save the project by selecting Save All from the File menu.

Setting grouping marks before grouping

1. Open the diagram.

2. Right click on the fact type Apprenticeship Assignment.

3. Choose Task, Edit from the pop up menu

4. Click on the Role button in the window that appears.

5. Select role 14 and set the grouping to Don’t group:

Glrgroupoverride.png

6. Repeat this for role 15.

By unchecking roles 14 and 15 you have indicated you do not want to group these roles.

7. Open the Run menu and select Settings. Make sure all options are set as displayed on the right.

GlrNonStepped.png

Note: Do not check Step-by-step grouping. The Step-by-step grouping algorithm ignores the User Override property.

9. Open the Run menu and select Transform.

10. If this window appears, select YES:

ConfirmGlrSettings.png

11. Save the new diagram as Apprenticeship.igd.

The generated information grammar is not in Optimal Normal Form (ONF). This structure might be choosen if Assigned Apprenticeship is unknown for most of the students. (Grouping Assigned Apprenticeship then results in a Student table with too many NULL values in the Assigned Apprenticeship column).

Glrgrouped2.png

12. Rightclick on the lesson_glr.igg in the project manager and choose Remove from project. Confirm with Yes and save the project by choosing Save All from the File menu.

Deriving a proposal

1. Open the Run menu and check Proposal.

The elements in the diagram will show with G, L and R indicators.

Glrlexicalizeproposal.png

3. Uncheck the menu item Proposal to hide the indicators again.

Choosing a Primary Key

When ‘Automatic primary key selection’ is active and no primary UC is marked, Casetalk chooses a primary key based on the role with the lowest number. For instance, the fact type ‘Apprenticeship Assignment’ has two separate UC’s on role 14 and 15. So Casetalk chooses role 14. When this option is disabled, the user kan choose the primary key.

1. Reactivate the IG Model and select the Run >> Settings menu.

2. Disable the option Automatic Primary Key Selection, and click OK.

3. Open the menu Run and select Transform.

4. In the appearing Primary Key Editor you can select which uniqueness constraint should be taken as primary key.

Glrpkedit.png

5. In the left section a list of fact types with multiple uniqueness constraints is displayed. Below this list you will find a the list with the uniqueness constraints of the selected fact type. Each of the uniqueness constraints in this list can be checked. In the right section the selected fact type is displayed .

6. Choose UC 13 for Apprenticeship Preference, so that the primary key is created using roles 11.1, 11.2 and 12.

Reducing

1. Select Settings from the Run menu.

2. Select only the option Step by Step as seen on the below:

Glrreducestepbystep.png

3. Click OK.

4. Open the Run menu en select Transform.

5. The step-by-step window appears:

Glrreducestepbystep2.png

6. Click Yes to all to reduce everything.

7. Save the new diagram as lesson_r.igd.

8. Rightclick on lesson_r.igg at the Project Manager on the right, remove it from the project. Confrm with Yes.

9. Open the File menu and select Save All to save the project.

Generating Table Documentation

1. Open the Run menu and select Settings.

2. Derive data types: If this option is selected the casetool will derive a datatype proposal during the GLR process. CaseTalk will generate Char and Numeric datatypes. If you have entered other datatypes they will not be affected by this option. If you have used Char and Numeric yourself and the casetool has detected labels exceeding the total and/or decimal length you have specified these attributes will be updated. If the option Derive data types is not checked, check it.

3. Clear Population: If this option is selected, all population will be removed during the GLR process. Leave this option unchecked for now.

4. Make sure the settings in this window are like this:

Projectoptionsglr.png

5. Click OK.

6. Select Transform from the Run menu. The information grammar will now be grouped, lexicalized and reduced.

7. Save the new diagram as ‘Apprenticeship.igd’

8. Select ‘Apprenticeship Preference from the repository.

9. Open the View menu and select Table Report.

10. A new tab will appear containing the table documentation for 'Apprenticeship Preference':

Tabledocumentation.png

11. Keep the Table Documentation tab open and select another fact type from the Repository window. The contents of the Table Documentation tab might not change automatically. If so, adjust the tab window a bit.

12. Close the Table Documentation tab.

13. Open the File menu and select Save All.

You can print the table documentation by selection Print Table Documentation from the file menu.

Back: Constraints

Next: Using the plugin modules

Retrieved from "https://www.casetalk.com/wiki/index.php?title=Modeler:Tutorial/Deriving_a_Database_Schema&oldid=1516"