On Jul 12, 2007, at 1:23 PM, Tom Lane wrote:
> chester c young <chestercyoung@yahoo.com> writes:
>>> SELECT field1 / 2 AS foo,
>>> field2 * 2 AS bar,
>>> foo + bar AS total
>>> WHERE foo < 12;
>
>> First, I think it would be great if this worked - like the alias
>> to an
>> update table added in 8.2 - saves a lot of typing and makes queries
>> much more readable.
>
> This is not an "extension", it is *directly* contrary to both the
> letter
> and spirit of the SQL standard. I can hardly believe that M$ did that
> ... oh, actually, I can entirely believe it. The OP has a serious
> problem of vendor lockin now, and that's exactly what M$ wants.
>
> regards, tom lane
Hear hear!
What's really screwy is what I found when I hooked access into my
PostgreSQL database using pgsqlODBC (I know, it's an abomination) and
I logged the statements that PostgreSQL was processing. In MS Access
this query:
SELECT foo AS bar, bar * 2 AS gleep FROM table;
became this in PostgreSQL's logs
SELECT foo FROM table;
Vewwy Intewesting! I think Microsoft and ODBC might be making extra
work for themselves (obviously they are since they are allowing these
aliases) Or maybe not. Maybe I can glean something from this. Who knows.
--Joel