Thread: guids / bytea and index use ?

guids / bytea and index use ?

From
rahul143
Date:
Hi,

We are currently using a 32byte varchar for our primary keys. We tried to
reduce this down to 16 bytes but varchar didn’t seem to store this
correctly. I’d like to use bytea instead so we could use 16bytes, but are
indexes used properly ? Does anyone have any other suggestions on how to
store guids ? We don’t have any experience in recompiling the postgres
source code etc.



-----




--
View this message in context: http://postgresql.1045698.n5.nabble.com/GENERAL-guids-bytea-and-index-use-tp5734650.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.


Re: guids / bytea and index use ?

From
"Albe Laurenz"
Date:
rahul143 wrote:
> We are currently using a 32byte varchar for our primary keys. We tried to
> reduce this down to 16 bytes but varchar didn’t seem to store this
> correctly. I’d like to use bytea instead so we could use 16bytes, but are
> indexes used properly ? Does anyone have any other suggestions on how to
> store guids ? We don’t have any experience in recompiling the postgres
> source code etc.

http://www.postgresql.org/docs/9.2/static/datatype-uuid.html

You should use this data type.
In old versions of PostgreSQL, a bytea should do.

I am not sure what you mean with "used properly".
As long as you search for equality on the field,
the index would be eligible.

Yours,
Laurenz Albe