Re: identity columns

Thank you for this extensive testing. I will work on getting the bugs
fixed. Just a couple of comments on some of your points:

On 9/9/16 11:45 PM, Vitaly Burovoy wrote:
> It compiles and passes "make check" tests, but fails with "make check-world" at:
> test foreign_data ... FAILED

I do not see that. You can you show the diffs?

> 1. The standard requires "... ALTER COLUMN ... SET GENERATED { ALWAYS
> | BY DEFAULT }" (9075-2:2011 subcl 11.20), but the patch implements it
> as "... ALTER COLUMN ... ADD GENERATED { ALWAYS | BY DEFAULT } AS
> IDENTITY"

SET and ADD are two different things. The SET command just changes the
parameters of the underlying sequence. This can be implemented later
and doesn't seem so important now. The ADD command is not in the
standard, but I needed it for pg_dump, mainly. I will need to document
this.

> 14. It would be fine if psql has support of new clauses.

What do you mean by that? Tab completion?

> 16. I think it is a good idea to not raise exceptions for "SET
> GENERATED/DROP IDENTITY" if a column has the same type of identity/not
> an identity. To be consistent with "SET/DROP NOT NULL".

These behaviors are per SQL standard.

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services