Subject Re: [Firebird-Architect] Counter proposal to Temporary tables
Author Ann W. Harrison
Martijn,

>No-one used COALESCE before it existed.

OK, but in that case we implemented something as
described in the standard and it solves a real
problem.

>No-one used UDFs before they existed.

Wouldn't know. They've always existed.


>No-one used Stored Procedures before they existed.

Actually, people used non-updateable views with triggers
as if they were stored procedures.

>They all could be "worked around". How is this an argument
>against functionality that with which we could "tick of" another
>"SQL standard" item?

The short answer is that the feature, temporary tables as
defined by the SQL standard, doesn't meet the needs of the
sponsor.

>How is this an argument against easier transaction from other
>engines?

Transaction? Transition? It won't be easier transition from
MSSQL because their implementation is grossly non-standard.
Most requests for temporary tables come from MSSQL users.

>Not having to run the clean-up procedure yourself is exactly
>one of the reasons why temp-tables are, well, not "needed",
>but surely very convinient.

A problem is that the SQL-03 standard makes temporary tables
nearly incompatible with the <as subquery> clause in table
definitions. Adding temporary tables won't solve the migration
from MSSQL problems. Adding a non-standard temporary table
implementation will make future standards a choice between
doing the right thing - maintaining backward compatibility
and doing the right thing - and doing the right thing -
following the standard.

Regards,


ann