When EGT was developing its database structure it considered many models and was searching for a database model that could promote quick changes and easy development. This way if someone thought to reindex a table it would be an easy undertaking, as the same if someone thought a table name had to change to support a new standard.
EGT was also looking for a database structure that would support quick searches based on an ID or based on some quick information. What EGT came up with, is used in a similar context in many data applications. Its commonly known as a Unique ID, and EGT’s personal version is called an EGT Universal Reference File key (eURF).
The structure is simple, EGT records the creation of a new Unique ID in a Universal Reference Table with enough information to identify the table and supply a “common language description.” Something that developers could look at and understand what file created this URF key.
The requested Unique ID is then added to a table as the primary key, and that record receives its data information. For instance, if EGT was going to add a comment to a comment table. EGT would request a Unique ID and add it to the Universal Reference Table and to the comment table.
Thus if a user remembers the id of that comment, the user could perform an ID search and it could be found from the Universal Reference Table that the ID goes to the specific comment.
Generating the EGT Unique ID:
When EGT began implementing this Unique ID, it needed an ID guaranteed to be unique through the entire existence of an application. They only guaranteed way to perform a Unique ID is through a numeric odometer that has practically infinite possibilities.
It contains the first 4 characters of “EGT-” and then a series of 17 alhpa and numeric characters. Of each next request the odometer function call gathers the last generated eURF, converts the alpha-numeric characters to numbers, advances the number sequence 1, then converts the results back into an alpha numeric sequence.
Applications that use this will use it to a much smaller amount of characters but the principle is sound, generating the next ID only requires the previous ID, and the number of possible Unique IDs are endless.
Implementation:
Implementation happens to be very simple, on creating a new record you can attach a database trigger that will perform the creation of the next Unique ID automatically or you can have a programmatic function that will be called on each insert statement. EGT choose to create a function call that would return the next eURF in sequence.
EGT’s reasoning is, the applications used to create the eURF have more processing power than the database itself. EGT leaves the database for its data storage capabilities and puts the logic in the hands of a software function call.
Can I use something similar?
Yes, you can implement this global database structure model without any need for our permission. It’s not something we own, its a globally recognized method to keep track of your database and for simple data control.
You’ll always be guaranteed that your Unique ID will not conflict with another until the sequence resets itself. If that happens, simple change the preceding 4 characters to another sequence and you have just made a global application change from one function call. In fact in all EGT Applications and Web services its a matter of changing a simple settings file.
By the way, we can sell you the odometer code for a very small amount, to inquire simply post a comment, include your e-mail and we will send you a quote.