Difference between revisions of "Column Names"

From CaseTalk Wiki
Jump to: navigation, search
m
m
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
Fact Oriented Modeling requires all fact types to be uniquely named. When a FOM Model is transformed towards a logical model, or implementation, the column names may become too long and elaborate. Therefor CaseTalk and the Viewer allow column names to be formatted using naming conventions.
Fact Oriented Modeling requires all fact types to be uniquely named. When a FOM Model is transformed towards a logical model, or implementation, the column names may become too long and elaborate. Therefor CaseTalk and the Viewer allow column names to be formatted using naming conventions.


A generated employee table may contain columns for first name and last name, originating from a person fact type.  
A generated ''Project'' table may contain columns for project, manager and client. In the
diagram it shows all resulting column names, depending on the algorithm preference.


= Full Path =
[[File:Column naming Relational.png]]
The full path will name the full paths from the very first role to the final label type.
* Employee_Person_PersonID
* Employee_Person_PersonFirstName_FirstName
* Employee_Person_PersonLastName_LastName
 
= Relevant Path =
The relevant path setting will skip intermediate fact types names. Naming the first and complimenting that with a unique tail.
* Employee_PersonID
* Employee_PersonFirstName_FirstName
* Employee_PersonLastName_LastName
 
= Compressed Path =
This compressed path setting starts with the full paths, and then removes repetitive redundant names from it.
* Employee_PersonID
* Employee_PersonFirstName
* Employee_PersonFirstName
 
= Unique Tail =
The unique tail setting starts from the label type and work it's way to the table name. Inserting parts until all column names are unique.
* PersonID
* FirstName
* LastName
 
= Unique Path =
When applied to the Employee Table, the relevant path would show that table. This variant will attempt to skip the table name, resulting in this list of columns. If this fails for some reason, the table name still may be inserted at the start.
* PersonID
* PersonFirstName_FirstName
* PersonLastName_LastName

Latest revision as of 05:35, 5 February 2024

Fact Oriented Modeling requires all fact types to be uniquely named. When a FOM Model is transformed towards a logical model, or implementation, the column names may become too long and elaborate. Therefor CaseTalk and the Viewer allow column names to be formatted using naming conventions.

A generated Project table may contain columns for project, manager and client. In the diagram it shows all resulting column names, depending on the algorithm preference.

Column naming Relational.png

Retrieved from "https://www.casetalk.com/wiki/index.php?title=Column_Names&oldid=1975"