Wido,
We actually have plenty of places that uses enums. I think you're specifically
talking about saving the ordinal values of an enum into the database. You can
do that easily by putting this annotation on the field.
@Enumerated(value=EnumType.ORDINAL)
The DB layer will take care of saving the ordinal value and retrieving and
matching that value back to the enum.
With that said, I do want to point out that saving ordinal values have
operational issues. The biggest is people forget what they are and have to
keep referring back to documents/code to figure out what it means. The gain
from saving a ordinal value vs the string value in indexing is fairly minimal
due to the range of values being fairly limited in most enum definitions.
Therefore, my preference is to save the String value instead.
To do that you can use this annotation.
@Enumerated(value=EnumType.STRING)
There's an example of it in HostVO.java with the Type enum.
--Alex
> -----Original Message-----
> From: Wido den Hollander [mailto:w...@widodh.nl]
> Sent: Thursday, August 1, 2013 11:12 AM
> To: dev@cloudstack.apache.org
> Subject: Do we allow ENUM database types?
>
> Hi,
>
> I'm working on a disk cache setting for per disk offering which allows users
> to
> set it to:
> - none (current behaviour and default)
> - writeback
> - writethrough
>
> I've done some work in the disk-cache branch [0] where I created a new
> column called "cache_mode" in the disk_offering table.
>
> It's currently a varchar(32), but only 3 strings are allowed in the column, as
> listed above.
>
> We don't use enums anywhere in the database, so I was wondering if there
> is a good reason for that?
>
> Otherwise I think using a enum here would be a good use-case.
>
> Wido
>
>
>
>
> [0]:
> https://git-wip-
> us.apache.org/repos/asf?p=cloudstack.git;a=shortlog;h=refs/heads/disk-
> cache
