2011-06-23 20:44 keltezéssel, Gurjeet Singh írta:
> On Wed, Jun 22, 2011 at 3:37 PM, Joshua D. Drake <jd(at)commandprompt(dot)com
> <mailto:jd(at)commandprompt(dot)com>> wrote:
> It seems we did come up with a use case in the procpid discussion. The ability to
> change the names of columns/databases etc, to handle the fixing of bad decision
> decisions during development over time.
> Instead of just synonyms of columns, why don't we think about implementing virtual
> columns (feature as named in other RDBMS). This is the ability to define a column in a
> table which is derived using an expression around other non-virtual columns. I agree it
> would be much more difficult and some may even argue it is pointless in the presence of
> views and expression indexes, but I leave that as an exercise for others.
A few years ago I wrote a patch for (I think) 8.2 that implemented
IDENTITY and GENERATED columns. Look it up in the archives.
The "virtual" column is called GENERATED in the SQL standard
and is part of the table and maintained as a function/expression
of other fields in the same record.
Cybertec Schönig & Schönig GmbH
A-2700 Wiener Neustadt, Austria
In response to
pgsql-hackers by date
|Next:||From: Yeb Havinga||Date: 2011-08-04 09:12:24|
|Subject: Re: Transient plans versus the SPI API|
|Previous:||From: daveg||Date: 2011-08-04 09:04:42|
|Subject: Re: error: could not find pg_class tuple for index 2662|