Re: The plan for UTF-8

"David E. Wheeler" <[email protected]> Wed, 9 Feb 2011 09:52:15 -0800
Newsgroups gmane.comp.db.postgresql.dbdpg
Message-ID <[email protected]>
On Feb 8, 2011, at 6:36 PM, Greg Sabino Mullane wrote:

>> Just to clarify, when you say "everything" here, I assume you mean 
>> everything except data in a BYTEA column, right?
> 
> Correct - the bytea data comes back in an escaped form, and then 
> we run dequote on it, but we explicitly do NOT set the resulting 
> string as utf8. All the other dequote functions *will* set it 
> as needed, however (either by looking at the global var or 
> seeing if the string is already set that way, I haven't got 
> to that part yet).

Sound good.

I assume that large objects should be treated the same? Or is that interface separate enough that it doesn't apply?

Best,

David