Database best practices - Status

问题

I'm creating an application in which objects have a status lookup. To give some context, let's use the following example.

A helpdesk application where jobs are created and move through the following workflow:

New - Job created but unassigned
In Progress - Job assigned to a worker and in progress
Complete - Job complete ready for invoicing
Closed - Job invoiced

So, I create a status table with the following details:

int ID
string Name

and a lookup column on the jobs table

int ID
string Name
int CustomerID
int StatusID -> looks up the status

So in the real world, let's say we have the following requirements.

Users need to get a report to show all incomplete jobs (jobs which are either New or InProgress)
Down the line, someone will want to add a new status that sits in the middle of completed and closed for example.

So with this in mind, my initial thoughts are to create a new column on the Status table called SortOrder or similar and assign numbers to it such as

New - 10
In Progress - 20
Completed - 30
Closed - 40

This would mean that for Case #1 above, I could simply query the database for all jobs whose status is greater than or equal to 30. This would also be great for case #2 because it means that if I introduced a new status in between completed and closed it would not break this report.

I can see that it would come up often in different applications. Has anyone implemented a solution like this or come across this problem before?

回答1:

What we do is have a single Status table, and a companion Status Group table.

create table status_group (
    id integer primary key not null,
    alias varchar(20) not null,
    descr varchar(128)
)

create table status (
    id integer primary key not null,
    status_group_id integer,
    alias varchar(20) not null,
    descr varchar(128)
)

Then all of the statuses live in a single spot, but are grouped together vs having a zillion individual ones.

来源：https://stackoverflow.com/questions/9919146/database-best-practices-status

标签

database

database-design