Re: Apparent anomaly with views and unions

Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:

> I see nothing at all in the spec that justifies the idea of "keep around
> the source text and reinterpret it". They don't think that way; they
> think in terms of thoroughly-processed "descriptors" stored in the
> system catalogs.

Huh. The thing I find most jarring about this way of thinking is that it means
I can have objects in my database that don't correspond to any source code I
have saved.

Say I want to add an expression to a view, I can't safely take the source as I
created it in the past, add the column, and recreate it. The actual view in
the database may no longer correspond to the create statement I have saved.

I suppose you could say that's no different than the create table statements
which could be obsoleted by a series of alter tables.

--
greg